I was called to a shop with a complaint of a check engine light on a 2004 Jeep Wrangler with a 4.0L having about 61,000 miles on it (Figure 1). This vehicle belonged to one of the mechanics at the shop, and he had recently experienced an engine shutdown with a no-start condition. A well-known problem on this Jeep engine is the crankshaft sensor fails, resulting in an engine stall. The shop mechanic found this to be the problem on his engine as well and replaced it. About a week later he experienced a check engine light on with a Code P0340. This code directed him to a possible failure of the cam sensor. He then purchased a new camshaft sensor from the dealer, along with a new camshaft trigger wheel and shaft assembly (Figure 2). He made sure to index the shaft housing and position the trigger wheel in the same position that the old one was in. It will be important to properly align the camshaft sensor to the crankshaft sensor. If the camshaft trigger wheel is not properly orientated, it could result in a cam sensor code being set.

The shop mechanic took the Jeep for a test ride; however, the check engine light came back on. It's frustrating enough when you’re working on a customer's car, but when it's your own vehicle it always becomes a personal challenge. When the mechanic returned to the shop, he proceeded to check the wiring from the cam sensor to the ECM. He did not see any issues such as a short or open circuit in the cam sensor harness. Upon not finding a wiring problem with the camshaft sensor circuit, the only alternative was the ECU, but this is expensive so he was reluctant to throw this in the mix. At this point, he decided to call me in to get a second opinion.

Figure 1

Figure 2

2, 4 or more?

When I arrived at the shop I hooked up a generic scan tool to verify the trouble code the Jeep had set. The ECM had stored a DTC P0340, thus indicating that the ECM was not happy with the cam signal (Figure 3). This code has many different scenarios that can set the fault. The sensor could be inoperative, the signal could be corrupt by noise, or the crankshaft to camshaft timing could be out of synchronization. The only way to accurately tell what is occurring is to display all the signals involved on a multi-trace scope.

The shop did not have a scope, but the mechanic was willing to look over my shoulder to get a crash course on how to set up a scope and put together a game plan of what to view. About 20 years ago I started out with a 2-Trace Fluke 98 scope. This scope was complicated and had a long uphill learning curve. I learned how to use the scope by connecting to know good components. This allowed me to become familiar with the scope settings such as time, voltage and triggers. I then learned how to stabilize the voltage waveform on the scope display with triggers. I also learned the hard way that using triggers can hide an intermittent problem. When using the trigger, the last trigger event will be displayed on the scope. If the trigger is not there and the display shows a waveform, it looks like the waveform is still present even though it is not. In order to become better at using a scope, I took a few scope classes. Once I felt comfortable using a two-trace scope, I needed to move to the next level. There were many four-trace scopes hitting the market back then, and I needed to view more signals at one time on the same screen to enhance my diagnostics.

Figure 3

I acquired a few different four-trace scopes and mastered them along the way, but as time moved forward there was an increasing need to watch more than four signals at once. Vehicles were getting more complex and engine management systems were moving to multiple ignition coils and multiple cam sensors. I needed to go to the next level and purchased an eight-trace scope. It seems like the levels of moving forward never end, but it is what you need to do in order to keep on the cutting edge of the automotive technology. You don't want to stand still as the advances in the industry surpass you. There are many technicians that are happy where they are but you need to embrace the technology and the equipment that handles it better.

There is a need for an eight-trace scope, especially if you didn't want to waste time constantly moving your leads around. And once you move your leads around, you always wonder what you may have missed. You don't want to have limitations to what you can view on a screen and an eight-trace scope will help you to build visual associations with many computer signals within the controller you are testing. This in turn will visually teach you how the ECU driver circuits react to input sensors. This is how you learn system strategies to help build your diagnostic routines. You want to have the ability to look at eight coil primary voltage patterns or even eight injector voltage patterns parading on a screen. This cannot be accomplished with a four-trace scope.

Get Help with Drivability Issues
	Need help with drivability issues? The no-cost courses at NACE Automechanika Chicago are designed to sharpen your critical thinking skills and equip you with real diagnostic strategies. Use this link and choose “Yes, I have a coupon code” to register for FREE!

Time to test

The cam sensor on this jeep is a three-wire Hall Effect sensor that has a power reference, a ground reference, and a 5 volt signal line that's toggled to ground. The configuration and speed of the trigger wheel will determine the pattern and frequency of the signal you view on the screen. As part of my game plan I decided to use five traces of my eight-trace scope to view the cam and crank sensors and the three ignition coil drivers. This would allow me to check the integrity of the sensors and make sure the coils were being triggered correctly. It is always best to test the circuits directly at the ECM to ensure that you are testing the complete circuits involved. Testing any circuit at the component does not always guarantee that the circuit is not open from the ECM to the component.

Figure 4

I went to my information system and pulled up a wiring diagram of the vehicle. I then located the circuits I wanted to test and found their pin locations at the ECM to attach my test leads (Figure 4). I hooked up my yellow lead to the crank sensor, red lead to the cam sensor and my green/blue/white leads to the coil drivers. When I started the engine I could see six primary patterns within one cam signal as it transitioned from high to low state (Figure 5). It is important to check the frequency of the coil primary to determine how many coil firings occur in one camshaft revolution. If all coils are fired correctly it is a direct result of whether the vehicle has proper cam sensor to crank sensor correlation. There are some systems that may drop out coil primary and fuel injector commands if cam-to-crank correlation deviates more than a few degrees and in other cases there could be a hard or no start condition.

Figure 5

Figure 6

The coil primary saturation times were about 5.4 mS. and the primary patterns seemed very consistent. What I had to do now was to home in on the cam/crank patterns. I deselected the coil primary patterns and focused on the cam and crank signal waveforms (Figure 6). The crank sensor patterns were in packs of four paraded transitions from high to low and three sets of the packs were equidistant within one transition of the cam sensor. The patterns were in proper orientation with one another based on my experience of viewing many good patterns on these 4.0 liter Jeep engines. The one thing that did catch my eye was that the cam signal was not meeting its low to high transition properly. The cam sensor was being feed a 5 volt reference voltage so the signal line should have been toggling from 0 to 5 volts keeping within its 10 percent threshold. If this signal never went above 4.5 volts the ECM would still see this as a low transition. The cam signal was not going above 2.1 volts but yet it was producing a frequency pattern. This was an indication that there was a current path to ground. When the camshaft sensor is at 5 volts it is in an open circuit condition. Source voltage is always present to the point of the open circuit, which in this case is the camshaft sensor. You need to first understand that when a trigger wheel travels through the Hall Effect sensor and blocks the magnet from acting on the Hall Effect sensing device the signal line will be high or in an open circuit condition. When the trigger wheel travels out of the Hall Effect sensor and exposes the magnet to act on the Hall Effect sensor the signal line will be low or pulled to ground which produces a current path so a voltage drop can occur. I simply created this effect by removing the cam sensor and passing a razor blade through the Hall Effect assembly (Figure 7). I set my scope up to only view the cam sensor waveform pattern and watched the scope pattern as the cam sensor signal line transitioned from 0 volts to 2.1 volts. I then left the razor blade in place to block the magnet from acting on the Hall Effect sensing device. While the razor blade was in place I proceeded to unplug the cam sensor connector. The scope pattern of the cam signal transitioned from 2.1 Volts to 5 volts (Figure 8). This vehicle had a defective brand new cam sensor from the dealer. The sensor was not allowing the signal to transition above 2.1 volts due to a current path to ground, but yet did allow it to transition low near 0.1 volts when the resistance of the circuit changed.

Figure 7

Figure 8

New doesn’t always mean good

The shop at this point ran to get another cam sensor from the dealer up the street. To my surprise the dealer took the old sensor back and gave him another sensor without any questions. There has always been a policy with most parts counters to not accept returns on electrical parts but I guess this depends on the relationship you have with the guy at the parts counter. When the mechanic returned he installed the second new sensor. Luckily it was easy to install and I proceeded to take another look at the scope patterns of the cam and crank sensors (Figure 9). The signal patterns were now running true, the cam sensor was toggling between 0 and 5 volts, the check engine light was off and there was no pending code P0340 in memory.

Figure 9

The garage mechanic was in total denial and disbelief. He could not accept the fact that he went the sure route to buy a genuine Jeep part only to have it create another problem that would need to be solved. This is not uncommon in the field and I see brand new defective parts all the time. It's hard enough to go through the diagnostic process just to pinpoint a problem, but it is worse when you purchase a brand new defective part that has you second guess yourself. When this occurs you can end up with the same or similar problem that may take you down another path. This can be quite confusing because you feel that you have still not resolved the initial problem. It is important to have the equipment necessary to properly diagnose your customer’s vehicles and possess an understanding of how to use this equipment. A picture is worth a thousand words but a scope displaying a waveform to properly diagnose the vehicle in your service bay is priceless.  I service over a thousand shops and I recommend to each and every one of them in order to repair modern vehicles you will need a scope.

It is so important that you test with accurate methods so you know that your test results are valid. This will allow you to never second guess yourself. If you know how to properly test a component and understand its strategy of operation then it will allow you to validate whether your findings were correct or whether you’re dealing with a defective or incorrect part for the vehicle. We now live in a world where the repair technician is not only responsible for accurately diagnosing the problem, but is also responsible for the quality of the parts they install. There is no guarantee that a manufacturer has quality control of parts they manufacture The true technician is the one who will take nothing for granted and is keen enough to weed out the bad items that he purchases to do a job with. My only hopes that this story hits home with many of my fellow technicians and you take the proper steps not to fall in the trap of "Diagnostic Diversion."

Article Details

Mon, 05/01/2017 (All day)

John Anello

<p>A well-known problem on this Jeep engine is the crankshaft sensor fails, resulting in an engine stall. The shop mechanic found this to be the problem on his engine as well and replaced it.</p>

<p>cam sensor, scope, Jeep Wrangler</p>

Whenever we get a vehicle in for one simple service and find a lot of other stuff that needs attention, any well-trained, reliable technician will make a list of the needed repairs for the customer, putting the safety-related ones at the top — loose front end parts, failing brakes, expired or worn out or expired tires, and so on.  The caveat is that if a customer is shocked by a large estimate of needed repairs they didn’t expect, they’ll tell all their friends your shop tried to sell them the moon. And today, it doesn’t take many needed repairs to produce an estimate that climbs off the chart above what some customers can afford to have done. Even if they can afford the repairs, some savvy customers will opt to get a second opinion, so honesty is always key when making a list like that.

Show and tell is the best way to handle those situations. And your communication skills must peak in situations like this. Someone has quoted Einstein as having said, “If you can't explain it to a six-year-old, you don't understand it yourself.” And we all know some customers are sharper than others when it comes to absorbing what you’re telling them.

This canteen green Pathfinder hadn’t darkened our door before, but we had done numerous jobs for these folks on their other vehicles.

The other way the laundry list estimate goes is when they bring one with them when they come, and in my department, we get that regularly. These folks are typically the busy drivers who have been putting off first one repair and then another one for quite a few thousands of miles and then they’ll decide they want all those problems handled all at once. And some of their repairs aren’t quick and easy, either.

One of the recent ones we got was a 2005 F-150 with an inoperative moon roof that was stuck in the open position, no taillights, inoperative outside rearview mirrors and an erratic gas gauge. That same day we got a 2009 Chevy C2500 with a “fix whatever you find wrong” order, and there was quite a lot we had to do to that one. Then there was the 2005 Nissan Pathfinder with a laundry list that was a knuckle-busting adventure from beginning to end.

Happy customers

This family loves the work we do, and they tend to bring us most of it, but this was the first time we had ever seen the Pathfinder, which had 185,654 miles on the odometer. On the phone, the owner told me the instrument cluster was acting crazy, and I figured that’s all she wanted done initially, but then by the time her husband got there with it, she also wanted the heater core replaced – what an afterthought that was! It had long ago been bypassed.

As for the instrument cluster, it was doing wacky things. The temp gauge, the tach and the speedometer would come and go, and the brake, ABS and VDC warning lights would flash on and off just as randomly. The scan revealed a network code or two, but not much else. One thing we did notice is that the cluster couldn’t communicate during the dead-needle times. Filing that away mentally, I had Thomas launch into the heater core job.

In the meantime, the two other laundry-list vehicles rolled in. That 2009 2500 series Silverado mentioned earlier had been neglected for many a mile and year, with StabiliTrak and Tire Pressure Monitor messages, a gaggle of inoperative and busted lights and inoperative door locks. The 2005 F-150 was one a police officer brought in with an inoperative moon roof, tail lamps that didn’t work and a squirrely gas gauge.

The Silverado wasn’t all that interesting, except for the StabiliTrak message displayed on the cluster. The DTC and the troubleshooting led to the replacement of the steering angle sensor, which was fairly involved because of the rusty, dusty fasteners. Robert jerked the steering column out, put it in a vise, and did the surgery – that took care of the StabiliTrak. The rest of the repairs were fairly straightforward, but we did need to mount a couple of universal tag lights in the rear bumper – you can get a traffic ticket in these parts if your tag lights are out. We also replaced the busted CHMSL/Cargo lamp assembly. We replaced the driver-side power door lock switch for corrosion, but then found two of the four door-lock actuators were dead, along with two of the tire pressure monitor sensors.

The 2005 F-150’s moon roof was open and wouldn’t close (not good on rainy days), and so when we ran through the process of checking switches and wires we found a bad moon roof motor. We left the permanent magnet casing off the motor, remounted it and turned the armature with fingers to close the moon roof, because he didn’t want to spend the $300 on a motor. The issue with the gas gauge and the taillights had its roots in an oddly melted connector shell just outside the frame rail on the left side. The wires leading into the front side of that connector looked like a flame had been held under them — the tape and insulation was melted, and that side of the connector was, too. We could twist and wiggle the connector and get taillight and gas gauge normalization, and so we opted to clip that connector out and bypass every wire with solder and heat shrink. It was a good repair, because even if we were to find new replacement connector shells for this, they’d be too expensive.

On the 2005 F-150, this oddball melting almost looked like somebody had built a fire under it at some point – when we wiggled it, the tail lights and the gas gauge would go nuts, so we removed the connector and made the harnesses one at that point.

A patch job and a no-fueler

One of our directors owns a fairly decent little 2001 Tacoma he uses for deer hunting, and he came to me one day because he was having to add a gallon of water a week to keep the cooling system filled. It turned out that the coolant was making its way into one of the cylinders and out the tailpipe — one of the spark plugs was ultra-rusted. He made it plain that he didn’t want to start with a head job on that deer hunting truck, and so he asked if I had any other ideas. For his purposes, we decided to run some head gasket sealer through it, carefully following the instructions on the bottle for time, then we refilled it with coolant mix. About a month later he came by and told me that he hadn’t had to add any more water. Take that for what it’s worth.  When somebody’s in a bind, we do what they ask if it’s not dangerous.

This was the rusty plug from the Toyota Tacoma that fingered the cylinder head gasket as the cause of coolant loss. The liquid head gasket sealer paid off on this one. We’ll see how long it lasts.

About that time a 1999 Lexus rolled in that wouldn’t take gas at the pump, which can be one of the most frustrating issues known to man, and we found a plugged vent hose. Some insect lost his homestead and that customer was a lot less frustrated the next time he pulled up to the pump.

Even in the south, we get seized parts, and they’re always fun to deal with. One thing we do get a lot of down here are dirt-dabber nests in annoying places – the Lexus wouldn’t take gas with this clog.

Back to the Pathfinder

With the heater core in place on the Pathfinder, Thomas came to inform me that the brittle heater pipe manifold under the hood had broken when he was reattaching the hoses to the heater core, and this wasn’t something we could fix, so we ordered the $220+ manifold with its built-in plastic water pump and did that job up right. Filling the cooling system was challenging, but with the front jacked up, we managed to make it happen.

Before we re-attacked the cluster issue, we figured we’d do the alignment, and Thomas started out with the rear wheels because we always align those first if there are adjustments. The problem was that the adjustment bolts were rusted to the bushing sleeves on one side and the first bolt he fought with popped off right below the nut, which had become an irremovable part of the bolt. This was becoming difficult and irritating beyond words.

The rear control arm adventure was quite the knucklebuster. We attempted to drill this (didn’t have a Rescue Bit® on hand), but it was pointless. We opted to engage the high speed cutter, get some replacement cam bolts, and put a new control arm on it.

I called the owner to enlighten her, and she told me the Pathfinder had found most of its early paths at the beach, because that’s where it lived for the first five years of its life. Yeah, I know you Northern wrench guys see this every day, but we aren’t used to it down here in the south, although we do see some rides from up your way now and then. We ordered replacement cam bolts from Nissan and a lower control arm from the parts store, but to get the old control arm out of there we had to use the high-speed cutter’s 4-inch wheel to clip the adjustment bolts just inside the flanges.

Got that part of the job done, finished the alignment, and then we went after the cluster. Checking the network with the Pico, we found a pattern that was somewhat noisy, but after eliminating first one module and the other to no avail, we decided the cluster itself must be at fault because sometimes it’d talk and sometimes it wouldn’t.

The Pico pattern always looked pretty much the same, so after disconnecting every module on the network (one at a time, we decided to replace the cluster and that’s all it needed to normalize the needles.

This cluster is a plug-and-play unit, and when we told the customer what we had decided, the owner found a used one for $75, and when we popped it in there everything was peachy keen.

The 2009 F-150 transmission problem

In and among all these jobs, we had a 2009 F-150 with intermittent 6R80 transmission problems. The symptom was that the truck would have spells where it wouldn’t back up and during those times it would also stick in third gear until you cleared the codes. We were told that a transmission shop had pulled the pan and found good fluid and no debris, and they were kind of stymied as to what needed to be done next, so they sewed it up and the owner brought the truck to us.

We got a Transmission Range sensor code, but that was pretty much it. In the years that I’ve done this, it’s a pretty good bet that the transmission control module (or PCM) is suspect if the transmission starts acting strange and wiping the codes clears it up for a while. This is obviously not always the case — sometimes the transmission controller will go into limp-in mode for other reasons. With zero experience on this 6R80 gearbox, I called one of my guys who does them all the time — only he’s accustomed to the newer ones. He told me we’d need the “leadframe,” because he has to change them regularly for this kind of problem. That device looks like a big hard-wire harness with the speed sensors built in, but it’s actually the Transmission Control Module. Why they call it the “leadframe” is beyond me.

This is the “leadframe” as Ford calls it that actually turns out to be the TCM.  We wound up having to replace the whole valve body on this 2009 F150 – when we first removed the valve body we found this broken adapter and replaced it. And every time, we were careful to use a torque wrench when reinstalling the valve body.

My guy decided to help out and called the parts department to ask if they had one, and then I called, gave them a purchase order, and they billed it out at $125. The way this went down was a perfect-storm situation, because the year model was lost somewhere in the process of passing information from pillar to post, and it cost us some work.

When we pulled the valve body to replace the leadframe, we saw that the plastic-and-rubber adapter between the valve body and the pump was cracked, and so I got another one of those from my guy at the Ford place. The only problem was that when we put everything back together put the fluid in, we found that the transmission wouldn’t engage at all and the TCM (leadframe) wouldn’t talk to the IDS either. But we could plug the old leadframe into the wires and let it swing and it’d talk to the tool just fine.  What was going on here?

This was strange to me — for years Ford told us that electronics couldn’t cause a no-engagement issue, but here it was. Things have obviously changed. With the absence of electronics, this one dumps the pressure instead of raising it.

It was then we discovered you can’t buy a leadframe for a 2009 model — you have to buy the whole valve body, leadframe and all ($1,000). And even though the later leadframes look identical and are replaceable separately, they won’t talk to the IDS and they won’t function on a 2009 model. So we got a whole valve body, installed and torqued it, did the fluid fill, and fixed the truck. It was messy but fun pumping transmission oil into that one through the hole where the dipstick tube used to go and checking it with that tiny plastic dipstick right next to the catalyst with the engine running and hot. That was a knuckle-BURNER. One way or another, we won that fight and all the rest of them on this round, with busted and burned knuckles galore. Who knows what we’ll see next week?

Article Details

Fri, 06/02/2017 (All day)

Richard McCuistian

<p>Whenever we get a vehicle in for one simple service and find a lot of other stuff that needs attention, any well-trained, reliable technician will make a list of the needed repairs for the customer, putting the safety-related ones at the top.</p>

<p>F-150, customers, estimate</p>

I always expect the unexpected, the unusual or the bizarre solution will be what it takes to fix the car when the owner of Ken Davis Auto Repair calls me in to look at a customer’s vehicle in his shop. I was pleasantly disappointed this time!

I have a great deal of respect for Ken, and all others in automotive repair, who attend training on a regular basis. At an early point in my career, I recognized that in order to make my life easier during my work day and prove my claim of wanting to work smarter and not harder, I had to accept the fact my job isn’t just 9 to 5. It’s one that requires frequent afterhours (and before-hours, weekends, etc.) investments in training.

I mean after all, when you think about it, where is the MOST expensive training facility?  It’s that service bay where a car sits, and you don’t yet know how it’s supposed to operate (and therefore, don’t yet know how to fix)! I’d much rather the cars come in, get fixed, leave and it’s on to the next one than to be stuck learning a new system, holding my bay hostage and impacting an entire shop’s scheduled work load. Ken is an owner/tech and is very much like me in his desire for peak efficiency. So when he calls on my diagnostic services, I kind of expect there’s something REALLY weird going on. In the case of this 2014 Chevrolet Impala LTZ, this was so true!

GDS2 report screen

During his initial phone call he mentioned a few of the strange things the car started doing a couple of months ago that have happened on a more and more frequent basis.  Ken told me, “The vehicle had been trouble free in the 20,000 miles since the current owners had purchased this used car. Recently, an intermittent no-crank, no-start condition has worsened and now numerous systems not performing as designed.”

More specifically, the problems identified by the vehicle owner are that the starter would not make any sound when the push-button to start the engine was depressed; while driving, the instrument cluster would suddenly and without any common reason, just go dark and all gauges would seem to lose power; the headlights illuminated normally but at times would go out and then come back on. There were more symptoms, but it isn’t necessary to list them here. This description of symptoms sounds fairly complicated, doesn’t it? What are your initial thoughts?

Then he went on to say the owners had first taken the car to the Chevrolet dealership where it was noted that there had been some body repairs made to the front of the vehicle, which the owners were unaware of at the time they purchased it. In part, due to the repairs including aftermarket parts, the dealership politely refused to accept the vehicle into their service bays and suggested Ken’s shop might be better suited to “handle it.”

I’d made some suggestions that he and his tech could check, things that are typical causes of the intermittent phenomena. The next day, in his second call to me, he elaborated on the test results for the suggestions I had made. I’d asked them if a complete electrical system test had been performed. I had also asked them to perform a vehicle-wide Diagnostic Trouble Code (DTC) check and to record the findings. In addition, I was curious if any aftermarket electrical devices were installed on or in the vehicle.

Their aftermarket scanner was able to access a majority of the modules on the network in this well-equipped Impala. Almost every module had stored codes, some relating to “Low Voltage” and most had DTCs that I like to call “Tattle-Tale” codes. Those are codes about module “State Of Health” messages.

Enhance Your Electrical Skills
	Ready to enhance your electrical skills? Today's top trainers tackle network diagnostics, ignitions systems, waveform analysis, voltage drop testing and more at NACE Automechanika Chicago this July. Use this link and choose “Yes, I have a coupon code” to register for FREE!

Each module is supposed to identify itself on the network at some point. All modules are told to “keep a look out” for the other modules’ report that they are on the network and that they are working properly. When a module does not receive an expected notification that another module was supposed to send, then this module stores a DTC that implies the other module is not healthy. It matters not for whatever reason that message didn’t arrive; when a module expects the announcement on the network that the other module is “healthy” and it is not received, the code is stored.

When Ken received the test results from his tech Jerry and saw how many DTCs were stored in every module, he asked Jerry to hold off on doing any more testing and made an executive decision. Jerry was well-versed in network diagnostics, but felt a bit intimidated by the number of codes set and also had some big jobs in the shop that needed to get done that week. So Ken asked me to take the job. It didn’t hurt Jerry’s feelings in the least when I showed up to relieve him!

General Motors’ Global Diagnostic System version Two (GDS2) is the diagnostic software used on this car by the dealership technicians and is also available to the aftermarket repair shops. It is the diagnostic tool I chose to use in this case. I have numerous aftermarket scanners and PC-based diagnostic tools I could have selected from but had I done so, I may not have gotten all the information the car had to offer.

I started my diagnosis by reviewing with Ken and Jerry what I thought I’d heard them say. We all have been guilty of multitasking while conversing and not hearing everything correctly. I was driving at the time Ken made his first call to me, so distraction from my call could have occurred. Our chat confirmed we were all on the same page, so I proceeded to perform my diagnosis.

I began working on the car with a verification of the electrical system integrity. This is where every (electrical fault-related) diagnostic sequence should begin. Since the battery is the starting and ending point of every circuit in the car, shouldn’t we make sure there is nothing wrong with it? I tested the vehicle’s battery with a conductance tester and a carbon pile load tester, and then I tested the alternator and the starter. All passed with flying colors! Jerry’s test results had been the same.

GDS2 low voltage

Then I performed a vehicle-wide DTC & ID Information scan, which indicated numerous instances of "Low Voltage" problems and network-wide communication faults. During this time I was closely monitoring the Data Link Communication (DLC) voltage displayed on the GDS2 screen. I saw it drop a few times to below 10 volts as modules were queried for their information one at a time.

It was during a visual inspection that ground wire terminals were found to be attached to painted surfaces rear of the left headlight and on the left strut tower. Here is where I say something’s wrong. Remember what I said about the car’s battery being the beginning and the ending point for all circuits? I haven’t had the privilege (in very many years) of inspecting brand-new Impalas as they were prepared for delivery, but it seems as if attaching ground wires to painted surfaces could potentially cause problems right from the start. So, I don’t know if that’s how they come from the factory — but I do know that once I cleaned the paint off one of the studs, I recorded a reduced amount of voltage drop — by a large amount! Looking closer I found evidence indicating this vehicle had been repaired after a frontal collision.

Ground on a painted surface

The voltage drop tests performed while cranking the engine showed no less than 216 mVDC (0.216 VDC) differences on any of the first four ground terminal eyelets tested (those were the easiest locations to test). Could this amount of ground circuit “loss” cause all the modules to act as if they’ve lost their voltage supply (lost their minds)? Maybe.

You might ask, “what is this voltage drop test and why is it important?” Well, imagine a wire that has voltage running through it. The amount of voltage measured at one end should be very close to the amount measured at the other. This would be true under most circumstances. Now, if there were to be something wrong with that wire, something that caused some sort of resistance, then the amount of voltage read at either end of that wire would be different from the other. That difference is what is called the “voltage drop.”

Resistance in a circuit can be in the form of corrosion (such as when the insulation on a wire is violated and the wire eventually turns green), looseness (such as when wire terminals aren’t tight or two aren’t mating correctly) and many other ways including the introduction of something between connections (like paint, which doesn’t conduct electricity very well). Yes, paint can interfere with the return of electrons to the battery, the amount which can be measured using the various techniques of performing voltage drop testing.

Voltage drop connection points	Some of the nine grounds

The Mitchel1 wiring diagram indicated there were (only) nine ground points under the hood of this car. Some were grouped closely together and others were at each of the four corners in the engine compartment. I decided to measure the ground circuits beginning with the battery ground cable. Measuring from the battery post to the eyelet at the other end of the battery ground cable my DVOM (Digital Volt/Ohm Meter) displayed the drop as 0.212 Volts DC. I use this measurement as a standard to go by because every test afterwards will include this drop. It’s important to note that a voltage drop test should only be performed while loading the circuit. I did this by engaging the starter at the same time as recording it using the “Min/Max” function of my DVOM.

Battery ground cable voltage drop	Largest voltage drop recorded

Once I have established the standard, then I can subtract that amount from future tests if I need to. I now know that amount of difference in voltage to the battery post is caused by how much resistance the battery ground cable has. If I test another ground circuit without attaching to the battery negative post, I can add the drop in the battery ground cable to get the TOTAL voltage drop in that circuit.

Say I test a ground in the trunk by attaching my DVOM leads to a ground eyelet (for whatever system) and to a ground stud in the trunk. I will read the drop found in that area alone. But, to know the total voltage drop that that system is subject to, I have to add the drop of the battery’s negative cable as well since that cable attaches the battery negative post to the rest of the car.

Once I had tested a few of the easy-to-get-to ground eyelets, I then moved my lead to the bell housing, near where another ground cable attaches. The readings changed every time I tried to record the maximum voltage drop! When the starter did engage, I read as low as 00.40 (400 mVDC) but sometimes, when the starter wouldn’t respond to the push-button command I was giving it, I read over 10 VDC! Ten volts drop on the ground circuit is close enough for me to consider it an open circuit! It was a simple process of elimination at this point to locate the culprit. I worked my way back towards the battery.

Arcing evidence

Under the battery tray on this car are located numerous studs that are welded to the inner fender well. To each of these studs is attached a ground wire terminal (eyelet). I was able to reach each one and gave them a little “tug” to see if they were tight or not. Sure enough, the eyelet of the cable that attaches to the transmission bell housing was loose and easily moved to the slightest pressure. I removed the nut and while closely inspecting the terminal, found evidence of arcing. A lot of current was finding it difficult to complete its path to the battery through this loose connection!

Can you guess which nut was left loose?

The repair authorized by the vehicle owner was to combine the transmission bell housing ground cable on the same stud as the battery ground cable. I’d cleaned up the corrosion from the eyelets, removed the paint from the stud and nut then I coated all components with a liberal amount of dielectric silicone grease prior to reassembly. I cleaned and protected all other eyelets and studs under that hood just to make sure there would be no potential for failure in the future. I cleared all DTCs from every module, tested and re-tested many times for the customer complaint to reoccur, without incident.

It should be mentioned that all other aspects of the collision repair appeared to be done professionally. The owner had received the car unaware that any body damage had ever been repaired and had driven it for years before any repair-related incident ever showed up. I’m hoping the loose cable attachment was simply an oversight on the technician’s part, one any of us could have made, but also hope collision repair specialists reading this would be more careful with grounds in their future repairs.

Please don’t attach them to painted surfaces and expect them to be perfectly fine. No matter what you do, please perform voltage drop tests on them before returning the car to its rightful owner. This is a simple test and provides proof the connection made is what it should be (or not).

Article Details

Mon, 06/05/2017 (All day)

Jaime Lazarus

<p>A minor collision repair misstep causes a multitude of system DTCs</p>

<p>ground, voltage, DTC</p>

Recently I was asked to meet a kind, older gentleman in our parking lot at the shop. This patient, old man wanted to speak with me regarding his 2008 Chrysler Town and Country. It seems he's been disappointed with the vehicle, as it has left him stranded several times here of late. I waited patiently as he told me quite a lengthy story regarding the entire history of this quirky failure.

What I managed to extract from him is that, very intermittently, the vehicle fails to start. I asked him what he meant when he says "won’t start." He made it perfectly clear that when the key was turned to the "START" position, the starter would engage momentarily and quit shortly thereafter. On many occasions, the battle to start the van would be on for well over an hour before the starter would remain engaged long enough to get the engine idling.  He summoned me to the parking lot to demonstrate the issue for me. Unfortunately, the issue could not be reproduced, and this was the root cause of his frustration. This van has suffered from this erratic malfunction for well over a year. He informed me that the starter has been replaced twice as well as six ignition keys, two ignition switches and even two batteries! In fact, this malfunction occurred so erratically, any previous repair attempt would seem to fix the vehicle, but a few weeks later, the strange symptom reoccurred to his dismay. I assured him that if I could reproduce the concern while performing a series of tests, I should be able to isolate the malfunction and rectify the issue. Little did I realize how elusive this gremlin was going to be.

I always want to know exactly how a system is supposed to work before I try to figure out why it isn’t working. Here, I learned that there were 2 modules in charge of the relay.

Start with a full system scan

I began my investigation with a DTC scan of the entire vehicle. To my surprise, there were absolutely none stored. After many attempts to reproduce the issue, I was disappointed to find that it cranked, started and ran each and every time I turned the ignition key to the "START" position. I inspected the vehicle for aftermarket components with no suspects to condemn. I then chose to research TSBs and search Identifix for any similar failures (I like to go this route, especially when I have issues duplicating a complaint). Unfortunately, there were no offerings for me in either inquiry. What to do next? I never feel comfortable pursuing an electrical issue unless I can reproduce a symptom of some sort, but basic tests like load testing of the battery and starting system are a must, especially in this case. Ruling out what isn’t wrong with the vehicle — early on in the diagnostic process — is just as valuable as finding the cause of the symptom the customer is experiencing. Both the battery and starter circuitry showed no signs of failure or of any contribution to the peculiar malfunction this kind fellow was experiencing, for such a long period of time. As the afternoon came to a close, the gentleman assured me I could have the vehicle as long as I needed, so long as he could have it back with a confirmed fix. Tomorrow would be another opportunity.

I arrived at the shop Wednesday morning before 7 a.m. Typically, vehicles with issues like the one will be evaluated after the morning rush. I turned the key to remove the vehicle from the workshop and to my surprise, the vehicle cranked for about 500ms before the starter ceased! I returned the key to the “Off” position and reattempted to start. The failure finally showed its ugly face.

Because the issue was so elusive, I thought the best course of action may be to capture the actual fault right at the source (the starter circuit, because it quits working) and work my way out of the hole. Looking at the wiring diagram, it’s easy to see that the starter relay is controlled by both the Totally Integrated Power Module (TIPM) and the PCM. It’s time to do some research. I have to educate myself on the functionality of this starting system as to avoid yet another misdiagnosis. To start troubleshooting at this point may prove to be a premature, costly mistake. It’s easy to feel unproductive if we aren’t hands-on working on the vehicle. Experience will show you that taking the time to devise a game plan and executing it will get you to the cause a lot more efficiently than rushing in without a plan of attack.

The TIPM energizes the starter relay coil by applying voltage to terminal No. 85. The PCM creates a ground path for the starter relay coil by driving the low side at terminal No. 86. Makes perfect sense. Two computers control the starter relay, one at either end, but how do the respective ECUs know what to do and when to do it? Surely, we aren’t done researching yet. We still haven’t enough information to determine the total system functionality. This required research of each individual component. Unfortunately, the description/ operation of the starting system as a whole was non-existent. Careful research was critical to solve this issue once and for all and without unnecessary part replacement.

Know what’s supposed to happen first

Not being familiar with this ‘08 Town and Country's total starting system functionality, I had to start researching at the beginning — at the turn of the key. The ignition switch is where I went. After all, I still couldn’t be sure if I was dealing with normal system operation during a security breach. Perhaps it was a "normal characteristic" to cancel starter operation?

As it turns out, the ignition switch is a lot more than just a switch. It's contained in another node known as a WIN module (or Wireless Ignition Node). This unit not only contains the ignition switch, but also stores info for the tire pressure monitoring system and queries the ignition key for the security system. The WIN has a self-contained ignition switch which is simply an input to the WIN. When the WIN sees the ignition switch turned to the "Start" position and the key is valid, it will communicate the start “request” on the CAN bus to the TIPM. The TIPM will then output a voltage signal to the starter relay and simultaneously send a start request on the same CAN bus to the PCM. The PCM will respond and supply a ground path for the starter relay! WOW — that is one busy system!

Figure 1

So now it’s time to build a game plan. I still want to begin my analysis at the high current side of the system (starter relay) where the work is performed. Although some may not agree with my initial test location, I know I experienced the starter energizing momentarily. The method to my madness is to capture what is "disappearing" when the starter ceases its operation. I will, at that time, determine which direction to head in from there. I want simple access to the circuit so I can gain a bunch of info for a small investment of time.  

I removed the relay and wrapped the individual terminals with very thin wire. The idea is to allow me to test the relay in its normal location. By leaving the relay in this position, I've eliminated the chance of poor terminal contact in the fuse box eluding my capture of the failure.

 Now, take glance at Figure 1. The red trace is located at terminal No. 85 of the relay (the TIPM side) and the yellow trace is located at terminal No. 86 of the relay (the PCM side). As you can see, there are three distinct events on either trace. The first displays a proper crank and start of the engine. I then shut the key off and a few seconds later, I cranked and the engine starts once again, without a hitch. If you look at the traces and imagine what is occurring, keeping in mind our test location and how the system operates, you can clearly see that the TIPM is supplying the voltage (red trace) and the PCM is supplying the ground (yellow trace). The third attempt looks a bit different. The yellow trace and the red trace have both been interrupted; the ground side of the circuit (yellow trace) has clearly risen/no longer being pulled low, and the voltage supply-side (red trace) drops off almost simultaneously.

Of course a failure on either side wouldn’t promote current flow through the relay coil windings and as a result, the starter would cease to operate. Could this be it? Could the failure be due to a problem on the PCM side of the system? Perhaps, but remember, the PCM only does what it is told to do. It may have a failed driver or perhaps a poor B+/Ign/Ground feed. If that were the case, how do we explain a loss of supply voltage from the TIPM? There is also a possibility that the TIPM is telling the PCM to stop doing its job, too.    

Now I have some ideas on where to look     

Now, I have witnessed the failure through the eyes of a 2-trace labscope and have determined that the both sides of the circuit seem to be experiencing a failure. It’s time to take a step back and do what is easiest. Because we've performed our due diligence and we are aware that this system functions via the CAN bus, I've interfaced my scan tool and selected to watch the TIPM for the request for "Start" from the WIN. During the time the failure was exhibited, the TIPM showed the crank request from the WIN the entire time I held the key in the "Start" position. Logic is telling me that the WIN is not the cause of the erratic starter operation, because the TIPM never lost the command from the WIN. The PCM, however, may not be receiving the request from the TIPM during the failure. Unfortunately, after many attempts to monitor the PCM PID for crank-request, the failure just wouldn’t occur while I was watching (just my luck). I wasn’t ready to quit yet, though! 

Figure 2

I chose to pinpoint my testing further than just at the relay terminals. I’ve elected to monitor the entire starter-circuit at multiple points utilizing my 8-trace labscope from Automotive Test Solutions. Some call it overkill but I think John Anello said it best: "It’s like fishing with a net instead of a hook." I have a lot of time into this diagnosis because the erratic failure is so infrequent. I would hate to have to test different sections of these circuits individually and take the chance of having the failure not show its ugly face, when I needed it to most!  I want to show you a zoom of the failure-event. In Figure 2, and you’ll see the brown trace (PCM) is being pulled low, shortly after-key-up, providing the ground path for the starter relay coil. About 250ms later, the purple trace is being driven high (TIPM) supplying the voltage to the relay coil. At that moment, you can see that the starter began to operate by the (CEMF) counter electromotive-force acting upon the circuit.  Voltage is hovering around 10 volts (fairly normal to what type of load a typical operating starter places on a battery). As you can see by the time of the scope-sweep, this starter operation lasted approximately 100 ms before the TIPM ceased to energize the relay. What's more is, during this failure event, the PCM remained "low" and continued to do its Job. So, in summary... the symptom of the failure remains the same (starter interruption) but the way the circuit failed is different from what we experienced earlier in Figure 1. It certainly appears now that the TIPM may be the cause of the issue.

How can I attribute the TIPM being at fault? A combination of my actual test results and my newly attained knowledge of the system's configuration (the "players" involved in making the starter function). In Figure 1, both the TIPM and the PCM quit simultaneously. In Figure 2, the PCM remained functional. We know from the research we performed earlier, that the TIPM instructs the PCM to function. Knowing that information, it's logical that the PCM quit because the TIPM told it too. On the subsequent failure the TIPM was the only guy to throw in the towel. In both cases, the TIPM is the culprit.

After I replaced the TIPM, I ran some basic tests again to make sure all worked as designed.

Module Or “PIG?”

Now, before I condemn any computer, I must first check on the "PIGs" (Power/Ignition/ground feeds). That is an abbreviation I used in the past to remind myself that a computer needs all three of these to function, just as I need water, food and oxygen to survive. Figure 3 shows a cranking event during failure, while monitoring the PIGs at the PCM. The PCM has everything it needs to perform its duties. I've just ruled out a supply issue of any kind for the PCM. The TIPM was much easier to test. Because the logic/computer is located inside the TIPM (internal to the fuse box), voltage supply is sourced internally. Ground supply to the TIPM was sourced from five different terminals on three of the connectors, none of which lifted when the engine was cranking. A new TIPM was installed and the customer’s concern was corrected.

Figure 3

Not all diagnoses are a walk in the park. Some take lots of research and much patience. Most can be deciphered fairly quickly with a combination of proper education, tooling, logic, information and a well carried out interrogation process. If all of these key ingredients are utilized, there are very few malfunctions that will eat your lunch. I take tough finds like this one with me forever because these are the building blocks to proficiency and efficiency. Jobs like this one remind me why I love what I do so much and why my employer keeps me in the payroll.

Article Details

Sat, 07/01/2017 (All day)

Brandon Steckler

<p>&nbsp;&ldquo;No starts&rdquo; are usually easy to isolate and repair &ndash; but not in this case!</p>

<p>no-start, Town and Country, TIPM</p>

Those of us who have been in this industry a long time can remember when a vehicle was pretty much used up at the 100k mark. Odometers “rolled over” after 99,999. I read somewhere that in the 1930s, most engines needed rebuilding every 10-20K miles. Technology has certainly improved, and there are more than a few brands out there that can rack up some stratospheric odometer numbers with very few debilitating problems along the way.

When I worked at the dealer, we saw more than a few Ford pickups and Jeep Cherokees with 300-400K miles. The folks with Jeep Cherokees would keep the one they had driven hundreds of thousands of miles and buy another one. A few years back, my department was given a 1997 Pontiac Grand Prix with 250k on the clock and it still runs like a new car, even though the interior trim has aged to the point of coming apart in places.

This is one of those stealth problems — one of my guys had replaced the radiator in this high mileage Ranger, and since the lower radiator hose clamp had seemed okay, he re-used it. What he couldn’t see was that this clamp wasn’t quite long enough — the screw was only holding a couple of slots, and they gave way one day about six weeks after the radiator was replaced and dumped the coolant.

Some of the customers’ vehicles we work on in my department are low-mileage cars only a couple of years old, but we spend a fair amount of time on older ones. This time around there are several jobs to discuss; the most recent one we did yesterday. It was a high-mileage 2010 Ford Edge that had developed a serious transmission cooler leak after hitting a dog, and she had driven it until it was six quarts low on fluid before realizing that there was some serious dripping going on. As a side note, while it was on the lift, the lift itself breached a hydraulic hose and started leaking, and we had to fix that too – my contention is that that the transmission leak on the Edge was contagious. That one got a transmission cooler and the six quarts of replacement juice and the lift got $200 worth of new hydraulic hoses. Two crises dealt with in tandem.

Another recent one was a 2004 Trailblazer that rolled in with really high miles, an engine skip and a $450 estimate another shop had given her to fix that skip. They had proposed those $13 apiece Iridium spark plugs and a whole set of new coils. Well, this gal is a single mother and rejected that estimate out of hand. We pulled the codes, found a misfire on the No. 1 hole, did a compression test for good measure, then put a set of platinum plugs in there and a single coil.  For grins, we also polished the headlights, which had taken on the color and cloudy opacity that might be compared to dirty lemon juice. And yeah, we don’t charge labor, but why does a vehicle that old need the most expensive plugs and a whole set of coils? There is such a thing as pricing yourself out of a repair.

The Avalanche

The Avalanche

A lady called me one week while I was off during a break and asked if I’d have a look at a 2003 Mazda B3000 she had sitting on the curb in front of her house. The story on that one was that it was her son’s truck and that it had failed to start one night in a parking lot and they had tried to jump it off with no results. Figuring it was a bad starter, they simply parked it (strange, I know, but that’s what happened). It had been sitting there for three months when I opened the hood, noticed that the battery had been removed, connected my 30-lb. jumper cables, and fired it up. Faux jumper cable connections on crummy battery cables can de-rail a DIY diagnosis of a no-crank in short order. The B3000 ran like brand new and even had cold air, so she washed it and got it ready for a quick sale. A couple of weeks later it failed to start at the car wash, but that turned out to be a tripped inertia switch – somebody must have slammed the door or kicked it or something. But during the three months the Mazda was down, she had sold the boy her 2004 Chevy Avalanche and had bought herself a newer truck.

Now her son reported that the Avalanche, which boasted 268,587 miles, was leaking power steering fluid, and she wanted to know if we could check that out. I agreed, and when the truck arrived, we discovered that it had a dreadful engine oil leak that made the power steering leak look like a minor drip. It was odd that he was more concerned with having to add a half a pint of power steering fluid once a week than he was that the engine was bleeding oil to the point of what could be an early death. When I asked him how much engine oil he was having to add, it turned out to be a quart every two or three days. Yet the first thing on his mind was the power steering leak, probably because it’d whine and get his attention and he was tired of that. Squeaky wheel gets the grease, I suppose.

Well, we went after the oil leak first – it was dripping off the bell housing, but since that’s the lowest place, the leak might be coming from the pan gasket, the oil filter adapter or the intake. The bell housing was dry on the outside leading up to the intake, and it didn’t look like the oil pan was leaking (which these love to do). We looked closely at the oil filter housing before popping the small round cover off the underside of the bell housing, and through that hole, we found engine oil puddled in there, pointing to a rear main seal. We would attack that first, proposing the rear main, a torque converter seal and an oil pan gasket just for grins, since GM was kind enough to put a crossmember under the oil pan and make that part of it an easy fix.

My guys plowed into that one, and we were extremely happy it wasn’t one of those later model GM platforms with the stainless steel exhaust fasteners. Whoever came up with that idea should be chastised harshly. You can’t cut those stainless nuts with a torch and heating them doesn’t help either. But I digress.

The Explorer

This 2008 Explorer had been in who knows how many times for oil changes. The instructor who drives it makes a 120-mile round trip to work and back, and once a couple of years ago she came in with a bad engine vibration due to a busted cooling fan and two full-grown dead cats lying in the bottom of the fan shroud.  On another trip, the pulley ring of her harmonic balancer had slipped back toward the engine so that the belt was riding on the naked rubber part of the balancer, and the crank sensor was being machined away by the misplaced pulley. We saw two of those slipped 4.0L balancer failures that same week and haven’t seen another one since.

	This 2008 Explorer had been the equivalent of 10 trips around the world before the spark plugs were finally replaced, but it still ran like a champ.

On a humorous note, we decided on one trip to check the fuel filter on the Explorer, which was almost completely blocked. I used that for an object lesson as to why it’s always a good idea to check the fuel filter on a high-mileage vehicle. There was a time when Ford required the filter replaced every 15K on trucks. Later when I was changing the oil on my 2007 Taurus I decided to check the fuel filter and it was just as bad as hers was.

On the last oil change, I suggested we have a look at her spark plugs and it turned out that they were the originals – with 238,000 accumulated miles, and this one was still running great with not so much as a flicker from the MIL. The plugs had the paint spot on the tip and on the way out they did that heavy-duty squeaking ancient spark plugs do when they’ve been in there forever. Furthermore, the business end of those plugs was textbook worthy.

The Suburban

About the time we got the transmission out of the Avalanche, a high-mileage 2004 Chevy Suburban came rolling in with an engine skip that turned out to be on cylinder 4. This one was blessed with the trusty old 5.3L, which I like because of the camshaft-in-the block, but even without the overhead cams and all those nylon sliders and tensioners, this engine isn’t without its problems. Camshaft lobes wear down and head gaskets blow. When we pulled the plug out of the misfiring cylinder, there was a piece of ceramic that had been cracked and shucked off the center electrode sheath by some catastrophic mechanical event (that according to the Denso chart anyway), and that cylinder had no compression. A cylinder leakage test fingered the exhaust valve, and I wondered if that chipped-off piece of ceramic might be stuck in some valve carbon holding the valve open, but it would seem to have been hammered to bits and spit out the back, because that’s what usually happens.

We won’t know until later this summer what happened on the 2004 Suburban to cause this, but the cylinder it came from has almost no compression at all, and since the mileage is so high we’ll probably stuff an engine in it.

The prevailing question I had was what had caused the spark plug’s ceramic to fail in the first place. Were there detonation or preignition events that cracked and damaged it or what? They had to go on a trip and opted to drive it that way, but in the coming month we’ll stuff an engine in that one – the cool thing is that we can upgrade to a 6.0L if we want to because the 4.8L, 5.3L, and 6.0L are plug-and-play engines.

The ’98 F-150

One of my colleagues owns an F-150 he inherited from a relative, and we got one of those laundry list requests – the fuel economy had dropped off, the passenger side power window wouldn’t work, the transmission needed servicing, the intermittent wipers were intermittent, and, of all things, the Check Engine light didn’t work – and he wanted all of it fixed. A lot of people “fix” the MIL by covering it with a picture of somebody or by installing a piece of tape blocking the view of it, but this one had a breach in the wire between pin 2 on the PCM and pin 13 on the bulkhead connector, so we did an overlay on that one and brought the MIL back to life. Several of my people worked on that problem because it was a great troubleshooting and repair exercise in electrical systems, and it was a bug I didn’t plant.

This ’98 F150 had a laundry list of issues, the least likely of which was the inoperative check engine light.  We ran an overlay between the PCM connector and this bulkhead connector and got the light back online.

For a truck this old, that re-operational MIL might be problematic, because now, if the truck had issues of which he had previously been oblivious, he’d be swinging by regularly to have those issues handled. After we replaced the spark plugs, the passenger-side window regulator, the PRNDL indicator and the Combination Switch, the truck had no starter operation, and we tracked that to the big C172 connector near the battery – one of the students had begun disconnecting that connector, gotten side tracked, and left it that way. In the end, that F-150 rolled out with no codes and no illuminated MIL, which was something of a surprise on such a high miler.

The hunting truck and a Nissan

The Toyota pickup on which we had used head gasket sealer came back in for a timing belt and a water pump – it still wasn’t leaking coolant any more, not even from the water pump, but the bearings in the pump were rattling, and so we stripped it down and did the kit thing – belt, idler, tensioner, water pump, etc. After we filled it with coolant and fired it up to do the final burp-out, I saw coolant leaking from the rear of the engine and discovered a head gasket breach that was trickling coolant down the bell housing. Whether he’ll want to redo the head gasket sealer or replace the head gasket remains out with the jury, but he opted to take the truck and use it for a while, keeping a check on the coolant level. His prerogative, I suppose.

The Toyota hunting truck revealed this leak after we did a water pump and a timing belt.  It was a tough shot to get, but in the real world you can see coolant trickling from under the right cylinder head. This stain told the tale.

Then there was the 2000 Nissan Frontier with an A/C belt squeak after a few minutes of at-idle A/C operation. This was condenser airflow-related because the head pressure started out normal and slowly climbed until the compressor had to struggle. Checking for radiator and condenser fin blockage, we rinsed them out with soap for good measure but to no avail. When we put a fan in front of the condenser blowing through it, the pressures normalized, and when we tested the fan clutch by heating the bimetal spring, it never got any stiffer, so we fixed that one with a fan clutch.

Finishing up the Avalanche

The power steering leak on the Avalanche was the last thing we tackled on that one. It wasn’t leaking from the pressure hose as we had figured. It was dripping fluid from the hydroboost unit, and so it’d need one of those babies to close out that job. We got one from the parts store, swapped it out, refilled everything, used the vacuum bleed cap I built to purge the air, I charged out the parts, and we put that one on the yard. Oh, yeah, we polished the headlights on that one too, and the whole truck looked better. “It’s the little things,” as one customer told us. We do what we can to breathe new life into those high milers, and it felt good to be done with another one.

Article Details

Tue, 08/01/2017 (All day)

Richard McCuistian

<p>The ones that just keep on givin&rsquo; sometimes need help.</p>

<p>high-mileage, vehicles, leaks</p>

Over the years, I have learned some very good electrical diagnostic routines that have made finding a variety of electrical problems both fun and very fulfilling. I have heard it said that when hunting for electrical problems, there is no right and no wrong way to do the job. I would question, and most likely argue, that point.

The process of electrical problem analysis for me started many years ago with some training, and has continued with much practice and thought about how this process can be made better. Of all the classes I have taken on electrical problem analysis, two classes stick in my mind as great classes. One was a two-hour class done by our local Interstate battery distributor and the white-haired old guy talked about the simplicity of using volt-drop testing when looking for unwanted resistance problems. That was a simple class, but the content of the class has stuck with me over the years. The other class was done by a great electrical trainer, Vince Fischelli. The class was about critical thinking and using the built-in loads on the vehicle (starter, headlights etc.) to load the circuits and actively test for proper circuit operation.

These two trainers captured my attention on electrical testing and from the things I learned from them, I was able to turn “the finding of electrical problems” into a game. So, how do you play the game; what are the rules? 

Rules of the game
Many years ago, Mr. Ohm and Mr. Kirchhoff wrote down some really cool stuff about electricity. These laws deal with voltage, resistance and current flow and how they all play together. Once you gain an understanding of these rules (electrical principles), playing the game can be fun.

When I got into this business in 1992, I was told to throw away my conventional test light, since using it on the electrical systems of “today” would kill computer sensors and drivers. Here we are 25 years later, and I have a very good selection of test lights using a variety of bulbs. Do I use them for testing today? You bet I do, although I have learned there is a time and a place for every tool and if these test lights are used in the proper places, they can be great tools.

Admittedly, one of the tools that I have replaced test lights with is the Power Probe. This is a wonderful tool but it, too, can be used in the wrong places. If it is used improperly, computer drivers magically go away. Every tool has its place and knowing the place to use the tools can be a mark of a true craftsman.

Equip yourself

When it comes to finding electrical problems, the first piece of information you need is a wiring diagram. We have all heard wiring diagrams called “electrical road maps” and this seems like a good description, since not only does the diagram show you where the electricity flows, but also information like how the circuit is designed and what components are in the circuit are also included in the diagram. Without a wiring diagram, you have no way of knowing which wires power the component or which wires are ground or control wires. These few pieces of information are very crucial to the operation of any electrical component.

If you want to test the powers and grounds in a circuit, you can use a volt meter, a bulb test light, or an LED test light, but stop and consider what you are testing. It doesn’t do much good to use a volt meter to test the voltage on a circuit that is not loaded, since the voltage of a non-grounded circuit will always be system voltage, regardless of the resistance in the circuit. The circuit needs to have a load put on it for the test to be accurate. This will require you to either turn the circuit on, or simulate a load with a bulb test light, or even a variable load tool I built that uses 1157 bulbs. With this tool, I can regulate a load from 2.5 amps through 15 amps. It is sort of like a test light on steroids. With a circuit properly loaded, you can get an accurate voltage and volt drop measurements.

Another valuable piece of equipment needed to make testing of electrical circuits easy is a scan tool that will communicate with the modules on the vehicles and have the capability of bidirectional control of the components. Being able to turn components on and off without having the engine running or the vehicle in operation can be very valuable. Let’s say you are working on an A/C compressor clutch that will not lock up — being able to have the engine off and using the scan tool to activate the clutch while testing the power and ground can be a real time saver.

Last but not least on the list of time savers is having such a simple thing as a package of colored highlighters to identify the different parts of the circuit. Being able to identifying the different parts of the circuit with different colors can make it simple to just glance at a wiring diagram and see which part of the circuit is powered, grounded or switched, or even if there is a different voltage other than vehicle system voltage. All this information can be found in a wiring diagram, if you spend the time to print it out and study it. 

Let’s play!

In the shop is a nice-looking 2001 Mazda 626 with an overheating problem. The vehicle is powered with a 2.5 V6 engine with automatic transmission. The odometer has recorded 205,000 miles and the vehicle is nice and clean. This is a vehicle I have serviced in my shop for a few years.

2001 Mazda 626 — Odometer 200,500 miles. Powered with a 2.5 V6 engine with an automatic transmission.

The vehicle came to the shop with an overheating complaint. When interviewing the person who drives the vehicle, I found the temperature gauge would climb when the vehicle was climbing steep grades. More questioning found the problem to also be after the vehicle was driven for an extended period of time.

Checking the basics, I found the cooling system low on coolant. Any time I see this condition, my thoughts turn to combustion chamber or head gasket leaks or a plugged up radiator. Before I can condemn the engine, I first need to check for any external coolant leaks. If there are external leaks, they need to be fixed and then the engine tested for combustion leaks.

Pressure testing the cooling system revealed a heater hose with a pinhole leak that was loosing coolant. This repair was simple, with just a piece of heater hose. The other coolant hoses were inspected with no problems found.

Being thorough with testing problems like this is very important, since there are several things that can cause an overheat problem. The low coolant is a sure bet to cause an overheating problem, but is there more? Is the leaking hose a symptom of another problem, or is it the cause of the problem?

With the cooling system holding pressure and full of coolant, the engine was ran until it reached operating temperature. The cooling system was tested for combustion gasses with none found. The next place to test is to verify proper cooling fan operation. This is easily done by running the engine until it reaches the proper temperature to turn the cooling fans on. To know this information, a scan tool must be used to verify engine coolant temperature and the cooling fan operation.

Cooling fan wiring diagram of the 2001 Mazda 626. This simple diagram shows not only the circuit routing to the fans, but it also shows what controls the fans, and what makes them work. I have added colors to each part of the circuit; Red = B+ power, Green = ground, yellow is switched ground and orange is switched power. With this color coding, I can look at the circuit at any place and know what the proper voltage should be.

Before I get ahead of myself, I need some information on how the cooling fans on this vehicle work. The quickest place to find this information is with a wiring diagram. The wiring diagram shows two electric fans on the vehicle. One is labeled cooling fan and the other is the condenser fan. If needed, more information on the fan operation could be found in the service information, although, at this time the wiring diagram information told me all I wanted to know. The wiring diagram shows both the cooling fan and the condenser fan are operated by separate relays, which are controlled by the PCM.

This is great information. Since the PCM is controlling the fan relays, there should be some scan data PIDs and possibly some bidirectional controls for these relays. By using the scan tool data, I found the right side fan (condenser fan) would come on when the ECT (Engine Coolant Temperature) reached 245-248º. When the A/C is turned on, only the condenser fan runs. By watching the scan tool, I can see the relays for both the cooling fan and condenser fan are being commanded on.

Electrical load and current limiting tool. This is made from a steel electrical box, six switches & six light sockets for 1157 bulbs. This can be set to draw between 2.5 amps up through 15 amps. Makes for a wonderful test light when wanting to either limit the current in a circuit, or apply a load to a circuit for volt drop testing.

Since the scan tool tells me both the condenser fan and the radiator cooling fan relays are being commanded on by the PCM, I need to know one more thing before I do any testing. I need to know where the relays are located. Service information says the relays are behind the battery and in front of the fuse box. Looking at the vehicle there are three relays in that location and they all look the same. These relays are plain mechanical/electric relays. Any time these relays are turned on or off, they make a “click” noise, and will also make a small vibration. All that is needed to determine which relay is which is to use the scan tool and cycle the relays. By feeling the relays with my hand while they are activated, it is an easy, quick and accurate way to find which relay is the cooling fan relay.

With the relay located and the relay removed from its socket I find the relay is a four pin relay. By using the wiring diagram, I find two of the terminals should be powered when the ignition key is turned on. Testing the terminals in the relay socket, I found power at these two terminals. All that is left to test is to verify the ability of the PCM to ground the relay, and the ability of the circuit to flow current to the cooling fan motor. By using the scan tool to command the cooling fan relay on, I found a path to ground for the relay coil, but there is no path to ground on the terminal that powers the cooling fan motor.

The problem is now narrowed down to either an open circuit in the cooling fan motor, or an open in the power supply circuit between the cooling fan motor and the relay terminal. The problem was found at the harness plug that connected to the cooling fan motor. The positive wire terminal had overheated and had to be replaced.

By using a scan tool, a wiring diagram and a logical diagnostic process, this problem was quickly and easily found.

The cause of the inoperative cooling fan was a poor connection at the motor plug. The use of the electrical diagram and a logical diagnostic approach led me to the problem without having to chase any wild rabbits.

With the electrical problem fixed and the cooling fans operating as designed, the vehicle was test driven up a mountain pass with a 6 percent grade. The engine performed well, and the cooling system cooled the engine properly.

When problems like this come in, it is too easy to jump to conclusions at the first problem found, which in this case the cooling system was low on coolant. Be through with the diagnostic process and test the complete system and let the system test its self, by making it work as it was designed to work.

Article Details

Fri, 09/01/2017 (All day)

Albin Moore

<p>When it comes to finding electrical problems, the first piece of information you need is a wiring diagram.</p>

<p>electrical diagnosis, service repair, wiring diagram</p>

We use wiring diagrams in many of our diagnostics, but if we are not careful, they can sometimes lead us to make decisions that are not accurate, which can lead to wasted diagnostic time, unnecessary parts costs for the replacing parts that are not defective, and sometimes even missing a simple repair.

One area where I have noticed a wide skills gap when helping other technicians diagnose a problem is in the use of wiring diagrams — not reading them, but more importantly interpreting them. While there have been several very informative articles and training classes on the subject, the one that has had the greatest impact on improving my circuit diagnosis was a technique invented by Jorge Menchu of AESwave called Color Coding. His technique uses various colors to represent what types of signals to expect at certain points in a circuit and help narrow down where the problem is by seeing what is and isn’t working as it is designed to. I point this out since the colors I used to highlight the circuits in this article are based off of this technique, and I also use this information to design/change my diagnostic plan. A color coding kit (AES# 02-WDCC) is available from AESwave.com.

But alas, even when using circuit wiring diagrams and having proper techniques, there are times when the provided information does not show the whole picture, which can cause inaccurate diagnostic summaries and wasted replacement of components.

When it isn’t the bulb

How often when a vehicle comes in with a complaint of a bulb not working do we or the customer automatically just install a new one? In 95 percent of the vehicles that have this concern, bulb replacement fixes it, so for the most part it could be a valid first step. However, if it doesn’t work, it can turn out to be a problem vehicle especially if the wiring diagrams get a little complicated. This is what happened on a 2008 GMC Acadia SLT that had 82,439 miles with the complaint of an inoperative RF turn signal. The technician who was originally assigned the repair order started with a replacement bulb, but found that this repair was not going be that easy. Apparently, the bulb has already been replaced by either the customer or another shop so their next step was to determine if there was correct voltage and ground supplied to the bulb; a quick check with a Digital Multimeter (DMM) showed no voltage. Looking at the wiring diagram for the exterior lighting, they determined that the Body Control Module (BCM) was at fault because in their thought process that is what supplies the voltage to the turn signal and since the Right Rear Turn Signal was working; it must be getting the request from the Multifunction Switch. The tech checked the powers and grounds to the BCM and they were OK, so a new BCM was installed and set up. Obviously if I’m writing about this vehicle, that didn’t fix the concern. 

Figure 1 - The only code that appeared in the BCM was a B2615 for Courtesy Lamp Control, but since the circuit description did not have anything to do with the exterior lighting I kept my focus on the turn signal concern.

Like most diagnostics, if I’m not sure how a system is designed to work I do some research before testing. This is also the point at which I print out a wiring diagram and highlight what a correctly working circuit should look like. I find the same circuit also includes the turn signal on the RF side-view mirror, and I noticed that it is not working either; however, the Right Rear Turn Signal is on a completely different circuit and is working as designed. I also gathered from the wiring diagram that the BCM (Connector 4 Pin 5 DK BLU/WHT wire) is what controls the circuit once the input from the turn signal switch is received (Connector 1 Pin 16 DK BLU/WHT wire). Since the BCM controls the turn signal circuit, it’s a good idea to check for codes and when I did I found a B2516 Passenger Compartment Dimming 2 Circuit (Figure 1). A quick look up of the code with a description of the circuit shows that this is related to the courtesy lighting circuit, which I notice is not working. This does not seem to have any effect on the exterior lighting circuit so I decided to stay focused on the turn signal problem and keep that info in the back of my mind. Now I remove the RF turn signal bulb socket to start my voltage tests. I can see from the wiring diagram that the ground for the RF turn signal — in this case G102 — is a constant; it is the first signal I check with my LOADpro voltmeter Leads to test the circuit under a load. Next we move on to the supplied voltage side of the circuit. Since the BCM is easily accessible by the driver-side kick panel, I perform my testing there.

Known good – known bad?

When using a scope to diagnose a problem, it is a good idea to have a known good signal to compare a possibly defective signal to, so I also monitored the Left Front Turn Signal input and output (Connector 1 Pin 16 LT BLU/WHT and Connector 5 Pin 4 LT BLU/WHT respectively), since I know this side is working as designed. As you can see (Figure 2), both inputs are working correctly but only the LF turn signal output is being generated by the BCM; nothing is happening on the RF turn signal output circuit. I also turn on the hazard flashers as another input source to the BCM and have the same result with an inoperative RF turn signal output. Next I use a Power Probe to apply battery voltage to the RF turn signal circuit at the BCM harness with the connector unplugged and the directional bulb at the RF corner illuminates. This tells me that the circuit is intact and can handle the load when applied. Now I am starting to see why the previous tech suspected the BCM.

Figure 2 - This is the scope capture from the BCM input and output controls of the turn signals.  Notice that the input is received, but there is no output for the RF turn signal.

While looking at the wiring diagram for the exterior lighting circuit, I notice a few pins that are B+ supplies to the BCM, and one of the fuses is even labeled as the Right Turn Signal. Something important to remember when testing voltage supplies and grounds to a module is to look at the actual module wiring diagram. While the exterior lighting wiring diagram shows some power supplies, it does not show the whole picture of the module itself (Figures 3, 4).  I start with the grounds first; it looks like Pins 1 and 5 (Connector 3, both BLK/WHT) and Pin 9 (Connector 4 BLK) are the grounds, and all three test fine. Next I move on to verifying the voltage supply pins. I find that there are four pins, all RED/WHT wires numbers 1-4 that are supposed to have B+, but find that Pin 2 does not; it is an open circuit. Guess where the voltage supply comes from?  Remember the code in the BCM for the courtesy circuit? The fuse that supplies B+ to this pin was open. After replacing the fuse, the RF turn signal worked.

Figure 3 - (Diagram courtesy of Mitchell Pro Demand) The wiring diagram for the Exterior Lighting shows only 3 B+ inputs for the BCM, which all tested fine.

Figure 4 - (Diagram courtesy of Mitchell Pro Demand) The wiring diagram for the BCM itself shows another B+ input at Pin 2, note that this does not show on the Exterior Lighting wiring diagram and was not tested by the original tech that diagnosed the vehicle.

I asked the other tech again to verify that he tested all the B+ supply circuits; he said yes and showed me the exterior lighting diagram and found it only lists Pins 1, 3 and 4 as B+; however, Pin 2 is not shown on the exterior lighting circuit. That is why it is important to use the actual module wiring diagram to check for B+ and ground. I do not understand why this power supply would affect only the RF turn signal, especially since there appeared to be a dedicated fuse for the right-side turn signal, but it does go to show that we must not get tunnel vision when performing something as simple as a lighting circuit diagnostic, as there may be a bigger picture.

Not quite done

So the vehicle is fixed right? Well, sort of. The RF turn signal is working (Figure 5), but the directional on right side-view mirror is still inoperative. As stated before, the wiring diagram shows both the RF turn signal and Passenger Outside mirror are on the same circuit, in fact the mirror is spliced to the same wire from the BCM before going through the Underhood Fuse Block so it eliminates that part of the wiring automatically. Well, it looks like the best place to test is at the connector for the mirror itself, so we can see if the voltage signal is present and test the ground. After removing the door panel the problem was pretty easy to see: the mirror that was on the vehicle was incorrect for the application, the connector pins for the mirror side of the harness did not align to the pins in the original door harness, and there was a second mirror harness connector on the door that did not have anything plugged into it.

Figure 5 - The scope capture of the repaired circuit showing all inputs and outputs are working as designed.

Someone had just attached a side-view mirror that looked correct (on the outside) from a GM vehicle with different options. In hindsight I could have saved myself the trouble of removing the door panel by trying to move the mirror glass with the controls as none of the functions of the mirror worked. Repairing the circuit for the RF turn signal restored the double-time flashing of the right turn signal indicator on the instrument cluster; the inoperative side-view mirror directional did not affect the rate of the flasher. I did not find out what caused the courtesy fuse to blow, but I don’t know what happened to the original side-view mirror on the vehicle, either.

Another bulb in question

The next vehicle that was given to me was a 2008 Dodge Avenger with 112,976 miles and a 2.4L engine for a complaint of an inoperative right front low beam headlamp. A little background about this vehicle before it ended up in my bay: The customer has already tried replacing the bulb themself, however when the vehicle arrived there was no bulb to be found at the RF headlamp connector, in fact there was a new connector already spliced in with butt connectors (Figure 6). The technician who got to look at the vehicle first also knew the customer had tried replacing the bulb, so they installed a voltmeter across the bulb connector and turned the headlamps on - 12V! They assumed that maybe the customer had purchased a defective bulb, but it was not found in the vehicle, and we didn’t have another in stock to test. Since the bulbs are very easy to replace on this vehicle, he swapped the left front low beam bulb to the right side instead of ordering a new one, and he knew the left headlamp worked fine. Same problem: the bulb did not illuminate on the right side. He swapped it back to the left side and it again worked perfectly.

Figure 6 - The customer has already attempted to perform their own wiring repairs to the vehicle along with replacing the bulb, which was not in the vehicle when we received it for diagnostics.  Not sure why they used so many butt connectors but we had to make sure his attempted repairs were not negatively affecting the circuit.

I can understand the confusion and frustration of the technician since he verified he had voltage and ground at the connector with the headlamp switched on, so why was the bulb not working? He pulled a wiring diagram for the headlamp circuit and saw that the RF low beam headlamp is a fairly simple circuit that has a constant ground and voltage is supplied by the Totally Integrated Power Module (TIPM). So he asked me for a second opinion before recommending a new module.

When looking at the wiring diagram, I like to start with the ground side of the circuit and highlight it. I noticed that ground is constant as he stated, but it is also shared by the right front high beam bulb and the right front fog lamp bulb, both of which are working normally, so it doesn’t look like we have a problem with high resistance on the ground portion of the circuit. Another item I noticed is the customer replaced the connector to the RF low beam headlamp, with multiple butt connectors. Fortunately they were not affecting the operation of the circuit.

Figure 7 - A halogen headlamp is substituted for the missing bulb by wiring it into the connector.  This is also a great way to load test a system.

Figure 8 - A capture from the graphing multimeter shows that when the bulb is connected the supplied voltage drops to 0V, but when it is disconnected again the voltage returns. This is why the technician found battery voltage on their DMM when testing the circuit, it was not under load.

Next I move on to the voltage supply side of the system. Again as the technician stated, voltage is supplied to the RF low beam headlamp from the TIPM. So to verify my understanding of the circuit I used a graphing multimeter and back probe Pins 1 and 2 of the RF low beam connector and wired in a headlamp bulb (Figure 7) that I also use to load test the circuit. When the headlamps switch was turned on my GMM showed no voltage; unplugging my wired-in headlamp from connector I had battery voltage again. Reconnecting the headlamp to the circuit I experienced the voltage dropping back to 0V (Figure 8) .   

It appears to be a defective driver in the Totally Integrated Power Module (TIPM), but let’s not jump the gun until we verify the voltage and ground supplied to it first; we’ve already experienced that in our last case study. The TIPM is easy to access and had several connectors attached to the underside of the module. Using the actual TIPM wiring diagram and not the wiring diagram for the headlamp circuit, we find there is a larger B+ wire directly from the battery, which supplies voltage to the module and multiple grounds to check, again testing them under load since simple checking voltage it not going to reveal a problem as we just witnessed with the headlamp circuit. All the voltage and ground circuits to the TIPM are fine, but just to be on the safe side, I simulate the work of the TIPM and supply voltage to Connector 5 Pin 3 WHT/TAN wire to verify the integrity of the rest of the circuit to the headlamp and my wired-in headlamp bulb illuminates brightly, proving the problem is with the RF low beam driver inside of the TIPM.

The customer approved the replacement of the TIPM but did not want to pay us to replace the headlamp bulb; they had the bulb at home and would install it themselves. With the new TIPM installed I still wanted to verify my repair so I attached my tester headlamp in place of the missing headlamp bulb and it worked great, at least by doing this I can be confident when the customer installs their new bulb it will work.

As I stated in the beginning, the repairs themselves were simple but using the correct wiring diagrams and techniques to understand how the circuits that operated them are designed to work is the key. Not doing so makes it easy to get misled if a solid understanding of the wiring diagrams is not in place.

Article Details

Mon, 10/02/2017 (All day)

Michael Miller

<p>We use wiring diagrams in many of our diagnostics, but if we are not careful, they can sometimes lead us to make decisions that are not accurate, which can lead to wasted diagnostic time, unnecessary parts costs for the replacing parts that are not defective, and sometimes even missing a simple repair.</p>

<p>wiring diagrams, automotive repair, techniques</p>

I was called to a shop for a simple task of reprogramming a new ECM on a 2012 Ford Focus with a 2.0L engine (Figure 1). There are a lot of shops in the industry that do not want the responsibility of programming a new ECM due to the liability and costs involved in doing the job. Some shops may not be properly equipped with the proper interface or laptop and may not want to go out to the Internet to fill out an application to sign up with an individual manufacturer and purchase a daily subscription. This may be coupled with a need to be registered through NASTF as a Security Professional to access theft procedures that may be needed during the post programming procedures. There is also the need to have a proper battery charger in the shop that can provide a "Programming" mode that will maintain a specific voltage range while a current surge occurs such a cooling fan coming on during the programming process. Some manufacturers will terminate programming if voltage falls below 13.1 volts or if voltage exceeds 14.0 volts. An older battery charger with only Low, Medium and High settings is not recommended and may pose a problem if too much AC ripple is introduced.

Figure 1

Is the ECM to blame?
When I arrived at the repair shop I proceeded to question the shop technician to see why he had made the decision to replace the ECM. It is not uncommon for a shop to jump the gun to condemn a control module without probable cause. I'm not in the business to just go ahead and reprogram a Control module at will without making sure it will resolve their issue with the vehicle. The last thing I need is to charge a shop to program a new or used control module only to find out that it was wasted revenue without a cure for their problem. I will usually question the shop to see if they followed the proper procedures to condemn the module. This would include checking power and ground feeds, shorted reference voltages, communication lines or even an issue with their testing equipment.

Figure 2

The shop told me that they just replaced the transmission (Figure 2) and when they got done the vehicle would no longer start. They checked the vehicle and found that the ECM was not responsive so they figured it might have been damaged during the repair procedure. This is a bad situation of "Drive them in / Push them out". We have all been there before at one time or another in the repair business. It becomes an unwanted marriage between you and the vehicle and your only way out is to resolve the issue at your own cost because you can't expect the customer to pay it. In the end it all becomes a learning experience but you’re under the gun to get things resolved ASAP before the customer is aware of his or her new dilemma.

When I hooked up to the new ECM to program it I was unable to communicate with it. The shop was under the impression that the new ECM would not communicate because it needed to be programmed. This is misinformation that I see a lot of shops seem to believe in and I had to school them on this belief and to educate them on how to evaluate no communication issues with a vehicle. They have to always make sure that prior to condemning a controller they check EVERY power and ground feed at the ECM and scan the entire vehicle to make sure it is not just a single control module issue without focusing too much of their attention only on one single controller. By scanning the entire vehicle they will get a better evaluation of all the operating systems that are responsible in starting the vehicle. Sometimes the clues to their problems may be resonated in other modules on board that may put them on a better path in their diagnostic process.

Start at the beginning
I placed my scan tool on the vehicle and did a full scan of the entire vehicle. I discovered that only five control modules were present on the multiple networks this vehicle had on board (Figure 3). These were the Audio, Body, GPS, Instrument and Tire Pressure Monitor control modules. This vehicle had three separate CAN networks on board: Medium Speed, High Speed and Entertainment CAN networks. The High Speed CAN network was inoperative. The controllers on this network included the Transmission, Engine, ABS, Power Steering, Steering Angle, Air Bag and Occupant Classification control modules. It was highly unlikely that all these control modules were bad or that they all had a common power or ground feed failure. What was common to all of them was a network circuit that was either open or shorted.

Figure 3

It is very important to understand how a CAN network is structured prior to testing it. Keep in mind that all the controllers wired on a single CAN network are all considered nodes and they are all wired in parallel. The CAN network consists of a twisted pair of communication lines to eliminate electromagnetic interference. There has to be a way to stabilize waveform reflection within the network and this is done by putting a terminating resister at each end of the network. A 120 ohm resistor is commonly used at both ends of the network and may be internal to a module or externally mounted on the network wiring. The 120 ohm resister internal to a control module will usually be the control modules furthest at the ends of the CAN network. If we use Ohms Law we can mathematically figure out the total resistance of the network. The total resistance of the network will always be the sum of the inverse of each individual resistance. Total resistance will also always be lower than the smallest resistance. In this case:

1/Rt = 1/R1 + 1/R2

1/120 = 0.0083

0.0083 + 0.0083 = 0.0166

1/0.0166 = 60.24

With this equation we should expect about 60 ohms across each network.

The battery would have to be disconnected prior to doing an ohm check because the system cannot be alive while using an ohm meter. I used an OBD interface box to make things easier while working under the dash. As you get older you want to work smarter without having to put too much pressure on the lower part of your vertebrae while hanging upside down under the dash. The interface has labeled ports that reflect the 16 pins of the OBD II connector and each port is coupled with LED's that alert your attention to presence of power, ground or data activity. I used my Fluke meter and tested the Medium Speed CAN network first at pins #3 and #11 and acquired a reading of 62.8 ohms (Figure 4). Then I preceded to the Entertainment CAN at pins #1 and #8 and acquired a reading of 62.3 ohms (Figure 5). These readings were well within the tolerances of a healthy CAN network.

Figure 4	Figure 5

Module or wiring?
I now proceeded to check the High Speed CAN network pins #6 and #14 and acquired a reading of 2.5 ohms (Figure 6). This network was shorted together. This would make sense why seven controllers fell off the radar. The hard part now was to put together a game plan of attack to locate the source of the problem. The possibilities that could be laid out were that any one of these seven controllers could be shorting out the network or the network harness could be shorted somewhere in the vehicle. I printed out a diagram and circled the seven controllers involved and checked off the ones that were the easiest to access (Figure 7). The eighth controller was an optional Parking Aid Control Module that was not fitted on the vehicle. My plan was to unplug each controller one at a time to see if the short would be eliminated. If the problem remained then I knew I had a harness problem.

Figure 6

Figure 7

I wanted to start with the easiest controllers to gain access to without too much work involved so I went to the engine compartment first. I disconnected the engine and transmission control modules that were already exposed near the base of the transmission and there was no change. I next decided to go after the ABS control module that was located at the driver side firewall. In order to access this module I had to pull the battery out of the way (Figure 8). As I was reaching for the ABS connector I was looking down at something that caught my eye. There was a harness crushed under the battery tray (Figure 9). This was not good. I proceeded to remove the crushed harness and exposed what turned out to be damaged CAN network wires (Figure 10).

Figure 8	Figure 9

Problem found — on to the fix
The two wires that were crushed in the network connector were for the High Speed CAN network. I simply repaired the wiring and put everything back together and the vehicle started right up with no problems. It was amazing how these two wires took down the vehicle and put it out of commission during a routine R&R of a transmission assembly. These seven control modules were cut off from the rest of the vehicle as if they were stranded on Gilligan's Island away from all interaction with the rest of the crew. The seven modules were are still responsive and in good working order with good powers and grounds but were no longer able to communicate among each other or with other controllers on the network. This led to the other controllers setting "U" codes for loss of communication with modules within the networks.

Figure 10

Repairing vehicles today are much more difficult because you’re now working in tighter quarters and components are not always easy to get to or even remove and replace. It is so important to watch your every step to make sure things are put back the way you find them and to be very careful not to put components in harm’s way. When you experience a problem you always need to go back to where you were last. Don't be so homed in to one module problem but branch out you diagnostic routine to include a full scan of the entire vehicle to enable you to dot the I's and cross the T's.

Article Details

Wed, 11/01/2017 (All day)

John Anello

<p>I was called to a shop for a simple task of reprogramming a new ECM on a 2012 Ford Focus with a 2.0L engine</p>

<p>ECM, reprogramming</p>

In this month’s Tech Corner, I would like to share an experience I had when still full-time as a technician. It’s an experience I’m sure most of you have also enjoyed, or not, depending on your outlook on life. You know, the old “glass half full” kind of thing.

The car in question is an older Ford Mustang with the 3.8 liter V-6, with a hard misfire on cylinder #1. Follow along and see how you would have tackled this one!

The First Mistake

The customer had brought the Mustang in for a complaint of a rough idle and stumble on acceleration. After a short test drive, it was easy enough to tell that there was a serious misfire going on. I hooked up my scan tool and found code P0301 (cylinder #1 misfire) and P0316 (misfire detected on startup) stored in the Engine Control Module (ECM). 

Figure 1 - The answer to the Ford misfire is in this picture. Do you see it?

This vehicle uses a Direct Ignition System (DIS) that fires two plugs simultaneously. Opening the hood, I could hear the distinctive "tick" of a spark jumping to ground outside the cylinder. Looking a little more closely, I could see the spark jumping to the valve cover from the #1 wire. The wires looked like original equipment, and a closer inspection revealed signs of leakage in the others. 

On this type of coil, one plug is "positive," and one is "negative." When the coil discharges, current first travels to ground thru the negative plug, then back to the coil through the positive plug. When the coil is stressed, the internal insulation can fail, reducing total coil output. In this low state, there is just not enough voltage left to jump the gap on the second plug, even though the first plug continues to run just fine. That's why it's possible to have a DIS coil with one dead plug. 

Thinking I had this one nailed, I ordered a replacement coil and ignition wires and moved on to the next car on my list. Time is money when you’re working flatrate! 

The Second Mistake

When the parts arrived later in the day, I pulled the Mustang back in to the bay. It is a simple installation and took no time at all. I cleared the codes and went to verify the repair. Have you guessed yet? The miss was still there, and the MIL light was back on. 

You would think that after all the time I've had in this business I would remember my personal rules regarding diagnostics - Never take a shortcut, especially on a misfire code. 

A misfire code can be set by any condition that doesn't allow for complete combustion in the cylinder. My normal procedure is to first do a relative compression test to ensure the engine is mechanically sound. Doing that test now indicated that the #1 cylinder had a problem.

If I see a low cylinder indication on this quick and dirty test, I follow up with a normal compression test. 60 psi was all I got on the misfiring cylinder. What I found had me muttering a few words under my breath. I was kicking myself for breaking the rules, and now I had a major engine fault to explain to my customer.

Was the original repair necessary? Replacing the ignition wires may have been; however, the coil was a rushed diagnosis. The low firing line I had seen on my scope was a result of low compression – not low spark energy. Remember, the firing line is typically affected by pressure, gaps in the system and the amount of hydrocarbons available for conduction. The scope was trying to tell me something. I just wasn't listening, instead choosing to see what I wanted to see based on an assumption. 

What’s The Fix?

The next step I took was to perform a cylinder leak-down test. This test uses a tool called a differential cylinder pressure tester and has two gauges on it. One indicates line pressure (supplied by shop air), and the other is the pressure being contained in the cylinder. When connected, and with the cylinder to be checked at TDC of its compression stroke, the tool pressurizes the cylinder and you compare the two pressure readings on the gauges.

When connected to the 3.8’s #1 hole, the left side gauge displayed the line pressure of 90 psi. The right side gauge reads the pressure in the cylinder, showing 70 psi. That's a 20 psi difference, or a little more than 20 percent of line pressure.  Not a lot, but standard specification is no more than 10 percent difference.

With the line still connected, I removed the oil fill cap, radiator cap and air filter housing. That 20 percent of air pressure is going somewhere, and you can actually hear it escaping. That's the nice thing about this tool. It allows you to hear if the loss of compression is from the valves (air escaping from the throttle body or exhaust pipe), the rings (air escaping from the oil fill) or from the head gasket (air escaping from the radiator).

This one was a no-brainer. Air was rushing out of the throttle body with no evidence of air flowing through any of my other checkpoints. OK, now I've got it. The intake valve is leaking. I got authorization to remove the head, confident that this was the problem. 

With the head removed, I verified the valve was leaking by pouring solvent into the intake port and looking to see if any leaked past the valve on the combustion chamber side. It began to pour out as soon as the solvent got to the valve face. But because I had been burned on my first diagnosis, I needed to be extra thorough. I also checked the installed valve height to see if there might be a problem with bent valves or recessed valve seats and found no problems there. I inspected the push rods for damage, and the cam lobes for wear. While the head was off, I rotated all the cylinders to the bottom of their travel to look for damage to the cylinder walls.  

Everything looked good.

I got the head back a few days later and reinstalled it on the Mustang. I turned the key, and you should have heard the expletives that followed!

What Had I Missed??

You've got to know that I am really upset by now. I felt I had done a thorough diagnosis, and I had definitely found a major flaw in the leaking intake valve. Thinking that maybe the machine shop had done something wrong, I checked cylinder leakage with my tester. This time, the results showed no leakage.

But what about compression? Again, I got a low reading on the #1 cylinder. What is going to make a tight cylinder low on compression? The only answer I could come up with is that the cylinder couldn't breathe. However, I had checked the valve train and had found no problem. 

I pulled off the valve cover and rechecked the valve operation, measuring opening and closing heights of the valves on #1 and comparing them to #2 and #3. I could not find the problem. 

There was only one answer left. It had to be in the bottom end.

Again, I removed the head on my way to the piston, and here I'll tease you with the photo in Figure 1. Do you see what I should have noticed the first time? 

Figure 2 - I still don’t know how the #1 rod was bent, but it caused the piston to fall short of TDC and lowered the effective compression in that cylinder.

With the piston removed, the problem was obvious. Looking at Figure 2, do you see what should have caught my eye? The interesting thing about this failure is that the rod bent almost perfectly along its axis, effectively shortening its length. Other than that, there were no other symptoms — no noise, no vibration and no bearing damage.

Look closely at the stain on the cylinder wall where the ring travel ends near the top. The cylinder in the foreground is #1; #2 is behind it. Notice how the stain is thicker on #1, showing that the piston wasn't reaching TDC. I should have caught this when I had the head off the first time. Would you have caught it? 

What would bend the rod? Perhaps it was hydraulic lock from a leaking head gasket and the failed rod went undetected, or ignored, during that repair. When I first performed my visual inspection, the oil level was correct with no sign of intermix. Coolant levels in both the reservoir and radiator were correct as well, and there was no air escaping thru the radiator during the first leak-down test.

Learn from My Mistakes

Hindsight is always 20/20, they say. My first mistake was not performing my normal diagnostic routine and checking engine integrity from the start. My second mistake was not considering that a 20% leakage rate (a nominal amount) would result in a compression reading of only 60 psi. The two did not agree with one another, what the ECM would flag as a “rationality” error. Now, in my defense, who among you would have expected to find a bent rod or would have caught the visual signs that were present especially considering that there was no other evidence present? Even so, the visual cues were there when I had the head off the first time, and I missed it. When troubleshooting any kind of problem, remember what Spock told Captain Kirk – “We must fall back upon the old axiom that when all other contingencies fail, whatever remains, however improbable, must be the truth.”

Article Details

Fri, 12/01/2017 (All day)

Pete Meier

<p>Every technician has been there &mdash; deep in the diagnosis, tunnel vision sets in, and often memorable mistakes are made. But it&#39;s only a failure of you don&#39;t learn from the experience!</p>

<p>misfire, diagnostics, auto repair</p>

One day I got a very terse call from the vice president of the company where I was responsible for fleet maintenance back in the late ‘70s. It seemed that an almost new (1978) Dodge one-ton we had was pointed at the gate with a gooseneck trailer behind it and that truck and trailer needed to arrive at our offshore diving and salvaging dock within the next 30 minutes – 25 miles away. I had no idea why that trip to that dock was so urgent, but someone had misplaced the key to the Dodge.

“Get that truck started and on the road within the next 10 minutes,” he told me with his gravelly voice, “and I don’t care what it takes. Just make it happen.” 

I must admit that I was in my element under pressure in those days, so I hung up the phone and grabbed a jumper wire with a couple of ‘gator clips on each end out of my toolbox. I opened the hood on the Dodge and made a connection from the positive battery terminal to the ballast resistor to feed current to the ignition coil. Making sure the tranny was in neutral, I “pocket screwdrivered” the starter to fire the engine up. Ninety seconds had expired and the steering wheel was still locked, but I knew I could defeat the pewter collar around that silly spring-loaded steering wheel lock peg, and I slid into the seat and muscled the wheel hard to the right, and broke the lock. Mission accomplished in less than three minutes and the truck was headed out the gate.

Then there was the time at that same job where I had to drive down Highway 87 toward Galveston and take a steamy ride on one marsh buggy through a swarm of mosquitoes and dragonflies to another marsh buggy that had jumped time, stranding a different vice president and his passengers a couple of miles off the road. Putting a timing belt on while standing in snake and alligator-infested water and swatting away mosquitoes wasn’t my idea of a good time, but I was motivated enough that I got that job done in record time, too.

This is my 2007 F150 that was victim of a surgical strike by some toothy critter that was copper-hungry

The point is that every job isn’t interesting, but in our line of work, challenges are the spice of life, and it feels good to be a problem-solver. It feels even better to be appreciated, and usually we are, but that isn’t always the case.

Critters

Dogs and squirrels chew wires, as do rats. Rats and squirrels build nests in engine compartments, and cats looking for a warm place to sleep can die under the hood and under the car in very gruesome ways sometimes. I’ve had to kill spiders and roaches, wasps, dirt daubers and all manner of other wildlife in my under-the-hood and under-the-vehicle odysseys. One morning I did a classroom presentation on critter damage, and a day or so later I walked out to where I park my own F-150, slid in behind the wheel, and thought I was going somewhere in my truck, but it wasn’t to be. The battery was good and hot, but I had no starter operation and no scan tool communication. The red theft light was blinking, which can point to a few different problems, but it usually means a module (usually the PCM) isn’t talking. With the key on, I checked for voltage at the EGR assembly and found 9 volts on the gray-red signal return wire – which should have been grounded through the PCM. What that meant to me was that the PCM had lost its own ground reference somehow.

Having Alldata available on the smartphone is pretty handy when you’re under the gun to find out what’s wrong and you’re somewhere else besides the shop

Next it was time to bust out my smart phone and dig into ALLDATA, where I found that PCM G103 is located behind the battery on the bulkhead. With my flashlight, I peered down there and saw that about eight inches of that wire had been removed by some sharp little teeth and my much larger main power feed cable to the inside fuse panel had been just as viciously attacked, but it had survived without being severed. Some chew-happy squirrel must have a nice piece of wire lining its nest and a belly full of copper as I type these words.

Is Your Routine Holding You Back?
	If you are not applying your training to every single customer you assist, you are doing them and yourself a disservice. Read how in Pete Meierís newest story and learn about a free trial of the best video training on the market. Button text: Read Peteís Story

There was another ground wire in that same area that was compromised as well. While removing the battery and doing some solder and heat shrink work was almost enjoyable that Saturday morning, I found myself wondering if I was going to have this problem again. No other wires under the hood had been attacked.  It was almost like the critter had pulled up a wiring schematic and did a surgical strike to prevent my truck from going anywhere. And it worked.

I suppose I should have been thankful that these wires were the only ones the critter chewed – he could have done a lot more damage than he did – fixing this took about thirty minutes.

I prevailed in that fix and placed some rat poison in the general area. We’ll see how that works out.

The 2004 Suburban

In a previous article, I mentioned a 2004 Suburban with a 5.3L that was misfiring on cylinder 4 with low compression and, during the cylinder leakage test air was escaping into the exhaust, but the owner chose to drive it skipping for a while before having it fixed. Finally, the Suburban returned and we hashed out what needed doing.

This was one of those high-milers, and so I talked them into a reman engine because of the better warranty, which we managed to stuff in there in pretty good time. 

We were going back into the Suburban with the reman engine and had just put it in place when this photo was taken. You can see the burned valve in this photo of the old engine’s head if you know what to look for.

After the swap, I had Robbie jerk the head off so we could inspect the valves and the head of the piston on the offending cylinder and we found a valve that had become mis-matched with its seat and was leaking compression. With the new engine in place, the MIL was off, the monitors all cleared, O2 sensors were switching handily, and fuel trims were bouncing around the zero line, so we put that one back on the road. There was a strange caveat though. For some reason, the transmission wouldn’t go into park well enough to not roll away on a slope.

This was a deal-breaker, to be sure. The shift cable was adjusted as far as it could go. I could disconnect the cable and put the transmission fully in park, so there was nothing wrong inside the case. Eventually I decided to try the shift lever off another transmission I had on the shelf and with that lever installed, it would go completely into park just fine even though it looked the same. I have yet to figure that one out, but it was safe when it left.

The Silverado, the Fusion, and the MKZ

While all this was going on, another instructor who drives a 2003 Silverado 2500 Duramax asked if we could replace his master cylinder. He’s ordinarily pretty savvy, and since he brought us the master cylinder I had a guy pop it on there and begin the bleeding process. Well, the pedal felt like you were stepping on a plum, and there was fluid dripping from underneath the truck and we found that classic rusty brake line situation a lot of you guys have to fix every day. He didn’t look under the truck, I don’t guess. A careful exam of the whole system revealed that this line was a lot worse than any of the other lines, all of which looked pretty good, and so we got a roll of that dandy nickel-alloy rust-free stuff and built a replacement line from stem to stern (complete with new double flare fittings), and after the bleeding procedure, we got that one rolling again with a good firm pedal and a master cylinder he didn’t need. I gave him the rest of that $60 roll of brake line just in case something else would be needed later.

This rusty brake line syndrome is fairly common everywhere on trucks of all kinds, especially on trucks that do a lot of mudding, but salt did this one in. This Silverado had spent its early life in Panama City Florida, where the salt air took its toll.

The 2010 Fusion that came in around this time was making a bump noise underneath on the left side during parking lot maneuvers, and it was one of those cranky situations where you can’t see anything but you know something is wrong. And every bolt was tightened to no avail. This one has that odd double-ball joint design with two lower control arms, and when we applied the Chassis Ear^® we found that one of the control arms was the source of the bump, and when we got it out of there you could see the problem. The hidden rubber that is couched in the frame area had died, and that was allowing the sudden pressure of certain braking and steering maneuvers to give a metal-to-metal contact sound. The fix was easy enough.

This inner control arm bushing isn’t visible on the Fusion until you remove the control arm. This was the rear of two control arms that car is blessed with on each side.

That Fusion reminded me of another vehicle, a 2012 MKZ that came in with an alternator you could hear whining from 100 feet away. She had been to a tire shop complaining about a noise, and the first guy who rode with her at that shop said he thought the noise was a hub bearing, but the more experienced mechanic said, “no, that’s the alternator,” because it was making the noise when the car was sitting still and the pitch of it matched engine speed. When I heard it, I agreed with the older guy’s prognosis. That alternator was making a LOT of noise that changed with throttle.

Getting the alternator off a 2012 MKZ isn’t for wimps — the refrigerant has to be recovered and the A/C compressor has to be removed, and the alternator comes out the bottom. There’s nothing easy about any of that job, but my guy got it done. I knew this 17-year-old could handle it — he had just finished replacing the heater core in a 2010 Wrangler, and after that extremely difficult job, this one was a cake walk.

The new alternator didn’t whine — the car sounded normal under the hood now, but it did have what sounded like a noisy hub bearing on the right front at road speed. It was one of those situations where the customer didn’t believe she had needed the alternator to begin with, because she was still hearing a noise on the road and one noise was masking the other. She chose to drive the car for a few days but came back and claimed the car had “put her down” and implied that it was our fault for replacing the alternator.

Here was another needful repair. The customer on this one was complaining of a vibration with the blower on high – easy to figure out and easy to fix, but needful all the same.

“I left it running when I got here,” she told me, and then said, “I had to jump it off this morning and then I drove it here (15 miles). It wasn’t giving that problem before you replaced my alternator.”

I had her pull it into the shop. I carefully explained that if the alternator wasn’t charging, the engine would have died as soon as the jumper cables were removed. Then I switched the car off and tried to restart it, but the battery was too weak. When I jumped it off and connected the Snap-on tester I showed her that the alternator was indeed charging and suggested that she find a cool place to rest while I did some more troubleshooting to figure out what was going on, but I told her I’d need the rest of the morning to be sure of what was going on.

“There was nothing wrong with my battery,” she snipped. “We’ll get to the bottom of this,” I told her.

After she walked away, I put a good stiff charge on the battery with a heavy-duty charger, then I got out the $1,700 Midtronics unit I bought from Joey Henrichs and ran through the entire routine, which records everything, including battery health and alternator ripple, printing it out for the customer. The MKZ showed a clean bill of health all the way around except for the battery.

Taking it a step further, I did a parasitic drain test, connecting a meter in series with the battery and waiting until all the modules finished charging their stuff. End of story — there was no drain, only a weak battery.

When she came back I showed her the results and told her she’d need to get the hub bearing noise handled at the tire shop. Sometimes it’s best to send some customers down the road, so that’s what I did.

Another Silverado

This 2003 1500 5.3L came to us with an overheating complaint — the guy said another shop claimed it must be a blown head gasket, but I explained that we wanted to diagnose it ourselves before we did any unnecessary surgery. Sure enough, it was overheating, but it was happening slowly, and there was no quick pressure buildup in the cooling system when the engine was started. The fan kicked on at 228 but the engine kept getting hotter until the fans kicked on high, and all that took a while, but I noticed that the radiator was still cool.

Here’s the overheating Silverado. Even with the radiator removed and bypassed and the thermostat gutted there was no flow through the hose.  Presumably this was a water pump problem?

“Let’s try a thermostat,” I told my guy. Cheap and easy comes first. We put one of those in there and burped it out, but nothing changed. The radiator was cool, but the engine was getting hot.  So I had him pull the water pump, and the borescope didn’t show anything wrong down in the pump, so we reinstalled it. With the radiator removed (no external clogging seen) I bypassed the radiator using the long hose, and we also looped out the transmission cooler lines and took the guts out of the old thermostat to allow free flow. With the engine running we had to squeeze the hose in the middle to neutralize the natural kink so as to facilitate flow but even with that hose in place of the radiator, there was still no flow through the hose, which was only warm on the ends — not in the middle. And the engine continued to try and run hot. What madness was this? If coolant had been flowing, the hose would have been hot its entire length.

Here’s the Silverado’s water pump with the rear cover removed, but I couldn’t see a problem, nor could I feel one manipulating the pulley and holding the impeller.

This was one of those exhaust bolts that had rust-melted from a 15mm down to something just a little larger than a 9/16, and it wasn’t in an easy place to access. We air-hammered this wiggler onto the bolt to get it out. It was a needful thing, but getting the bolt separated from the socket was tough.

To make a long story short, a radiator and a water pump fixed that one. Mission accomplished, but I couldn’t figure out what was at the root of this problem — I thought that plastic impeller might have been spinning on the shaft, but water pump forensics didn’t show that to be the case. One way or another, the truck never runs over 210 degrees now. Happy customer.

Article Details

Fri, 12/01/2017 (All day)

Richard McCuistian

<p>Every job isn&rsquo;t interesting, but in our line of work, challenges are the spice of life, and it feels good to be a problem-solver. It feels even better to be appreciated, and usually we are, but that isn&rsquo;t always the case.</p>

<p>auto repair, Silverado, Suburban</p>

Years ago we troubleshot a Grand Prix that had run just fine until the owner’s cousin had changed the intake manifold gasket and afterwards it was skipping dead on cylinder 2, so she asked if we could have a look at it. This was an engine skip – how hard could it be? First, we checked the obvious stuff (spark plug, compression, etc.) and came up short. But what we did find was that the number 2 injector didn’t sound right with the stethoscope, so we replaced that injector with a known good one, but to no avail. We then checked the entire injector circuit for shorts of any kind and excessive resistance, pin fit at the ECM and the injector, current flow through that circuit with the injector artificially energized (0.8 amps) and ran a temporary circuit overlay. Nothing changed.

When I finally scoped the injector pulse and compared it to the others, the pulse was strangely narrow, so I called a local salvage yard and obtained a replacement ECM. No cigar. Not even close. I replaced that ECM with a second one, because the salvage yard had a bunch of them on hand and they were only $20 each. I double checked everything. This made no sense at all. Finally, I Scotch-locked that injector’s trigger wire to the adjacent injector’s trigger and the car ran great from then on with no more problems. Remember, this was an OBDI system.

I didn’t like that temporary fix, but one of the GM engineers who was as stumped as I was told me those early GM ECM injector drivers can each handle 4 amps, and it’d be just fine carrying two 0.8 amp nozzles. Even if it had burned out a driver and I needed to keep digging, I still had two other good ECMs on hand. One way or another, that Grand Prix holds the distinction of being a grueling fueling enigma that still has me wondering to this day.

Burning in bad info

In the world of politics, news media and other sensitive areas, some have discovered that you can repeat some supposed fact enough that most of the hearers begin to believe it, regardless of its veracity. Our customers – some of them anyway – can also convince themselves that they know what’s wrong when they have little or no useful data except the symptom. Then there are those who have a vehicle concern and somebody they know who seems to have a bit of automotive knowledge makes a superficial jackrabbit diagnosis, hopping quickly across the high points without doing much else. And don’t you love those customers who bring you some parts they want installed based on an offhand diagnosis made by somebody who either doesn’t know how to do the work or “doesn’t have time?”

Even when we begin to gather data scientifically, we can still misfire on our diagnosis, and anybody who claims they haven’t been there isn’t being truthful. For just one example among many, I would have sworn in a court of law that the left rear axle bearing was ruined on my aunt’s ‘92 Crown Vic – after all, that’s where the noise seemed to be coming from, and it changed for the worse with a swerve to the right – as it turned out, she had a noisy left front tire and for some reason the noise was telegraphing to the left rear.

Back in the early ’80s a guy wanted me to replace his carburetor because two different shops using offhand diagnostics told him they didn’t do carburetor work, but that it needed replacing. One of them even claimed to have used an ignition scope and was a tune-up shop. It was a small carburetor on an inline six, so first, I bought a $6 Delco carb kit before I did anything else. Afterwards, I did a mild throttle snap and found it dropping a cylinder under load. I identified the cylinder, replaced a bad brand-new spark plug, and fixed that one.

And then there are quite a lot of people who will play the blown head gasket card without having seen anything other than an overheating issue. I don’t know how many times I’ve heard that, and I experienced it once myself on a 1993 Camry I checked for a friend beside the road. That one had split its radiator, overheated and was puking hot, sweet-smelling geysers out of the filler neck when we refilled it and fired it up. After it came to the shop on the hook, I wanted to show the class how that kind of head gasket failure looks and smells, but all those symptoms were gone – all it needed was a radiator. Go figure.

I have a 1989 Ford Bronco that was donated because the owner believed the head gasket was blown, but it was running crappy and puking coolant out the neck because it was a 5.8L, and he had wired it up using an old 5.0L firing order. When we wired it up with the right firing order, all the filler neck geysers went away.

A few years ago, we checked a 2.4L Dodge Stratus with a horrible oil leak that a shop had pegged as a rear main seal (how many offhand rear main seal diagnoses have we seen?) and used dye to find it coming from under the corner of the head.

Those of us who teach for a living know from experience that people who already believe they have all the facts are kind of difficult to convince otherwise.

2011 Chevy HHR 2.2L Ecotec

Bearing bad news

The owner of a 2011 Chevy HHR, 22L Ecotec with 123,598 miles had spent some time and money doing his own offhand diagnosis trying to get it started – he had checked the fuel pressure with a rented gauge, replaced the fuel pump and fiddled around some with a scan tool before realizing that he was in over his head. The HHR been sitting for a few months when it came in on a trailer. They had determined that it had to be something simple and were hoping we could get it going for just a few bucks. Somebody had postulated that it might have a bad crank sensor, and they brought it to us with the hope that we’d find out it was something simple. Well, it wasn’t. This one spun with very uneven compression, and when we removed the valve cover, we found some broken roller rockers, which typically means valves had contacted pistons, usually the result of timing component failure. But the timing chain was nice and tight, and looking down into the chain area I didn’t see any looseness or shattered sliders. Could it have been over-revved enough to float a valve? We didn’t do exploratory surgery, but we sold them on the idea of replacing the bad engine with a good used one.

When we pulled the valve cover on the HHR, we found several broken roller rockers. Something catastrophic happened here, so we decided to stuff a used engine in it.

The salvage yard sent an engine for that one with a few minor differences – the fuel rail had a different shape, along with a couple milder changes. When we were done, that one ran like a top and when we fixed an A/C leak and juiced up the icebox, it even had cold air.

The Expedition and the Crown Vic

A very regular customer brought her 2001 Expedition to us with a nasty coolant leak – this one’s a Triton and they tend to protect themselves from engine damage, but she just kept driving it. The water pipe that travels through the valley under the intake had rusted through and was dumping water almost as fast as you could pour it in. Initially we just removed the intake manifold, cut the rusted-through portion of that pipe out and replaced it with a hose and some clamps along with a new intake, but when we filled it with coolant, put a new thermostat in it and started warming it up, the warming never stopped – pressure was building very rapidly throughout the system and it was evident that this one had indeed blown a gasket.        

She’s a hands-on shopper, so she did her own search for a replacement engine at a price she liked, found one somewhere in Florida, and had it delivered to the shop. It had the heat pucks in some of the expansion plugs and so I knew it came from a reasonably savvy salvage yard. I crossed my fingers, hoping they didn’t turn the engine backwards while removing the torque converter bolts! Sometimes that flips one out of time.

Since I had people doing transmissions this time around, we went ahead and jerked the transmission out first, then I had another guy remove the original engine, and we carried it on the hoist over to the area where we do component swaps.  One of the first things we noticed was the narrow pulleys on the replacement engine – apparently this one had come from a Crown Victoria or a Town Car, but I couldn’t be sure. Oddly enough, a power steering pump came with the replacement engine, and so we took that narrow pulley and put it on the Expedition’s PS pump. We also replaced the idler and the belt tensioner, because those wide ones wouldn’t work on the replacement engine’s timing cover – and we weren’t about to swap out the timing cover if we could get out of it.

Initially, the guy who put the engine in the Expedition had put the generator wire on the top post at the solenoid, and that kept the starter energized when the generator was trying to work. The starter was a casualty in this case, but it’s an easy mistake for a beginner to make. The bottom photo shows the naked grooves in the generator pulley – the A/C compressor had the same issue, but we swapped out the power steering pump pulley to have the right one.

There were some other minor differences, but at the end, that engine was sitting in the frame with a new belt and everything plugged in, and the transmission was reinstalled – having drained the transmission and replaced the filter, we needed to start it up to get all the fluid back in the gearbox. We started with five quarts and hit the key.

When we started the engine to finish refilling the trans, we noticed that it had a large vacuum leak, and we also heard strange noises and smelled something burning – never a good sign, and it wasn’t the oil smoke from exhaust manifold handprints, either. As it turned out, the guy who replaced the engine did everything right except that he made one very easy mistake. He connected the alternator charge wire at the starter relay to the wrong post, which delivered alternator output current to the starter solenoid circuit while the engine was running, and that kept the starter energized, which destroyed the starter. Thankfully, it didn’t destroy anything else. But with that heavy rubber sleeve on the wires leaving the solenoid, it was easy to make that mistake if you weren’t ultra-familiar with the wiring.

This was another easy mistake to make – put a bolt that’s just a little too long in one of these and you’ve ruined a gas tank. We used the bolts and the gasket that came with the new tank, and so this leak really surprised us – even more so when we found out the new pump was faulty.

The 2003 Crown Vic ran out of gas while the gauge was reading a half tank. We replaced the sending unit with a new one from Carquest, and the guy who did the job used one bolt that was a bit too long when installing the pump, and punctured the gas tank. We got a new replacement tank, but after it was filled with gas, the leak was worse than ever. But that leak wasn’t around the gasket, it was around the plastic grommet where the wires pass through the mounting plate – I have not seen that before. Anyway, all’s well that ends well, and there was no fire. Only a bit of wasted gas.

The Xterra

The 2001 Xterra came to us with the concern that it had quit in a parking lot and failed to start, and somebody’s offhand diagnosis was that it had jumped time. This is a dicey situation, because that one isn’t a free-spinner, but even on interference engines, a timing belt can slip enough to stop the engine without valves kissing pistons. Had that happened on this one? I asked her if she had tried to re-start it (of course she had), but she told me she had only tried once and was hoping there was no damage. We didn’t want to do a lot of engine spinning on this one in the bay for fear of possibly making a simple no-start into something worse, so we checked the timing marks first.

On this one you can pull the upper part of the timing cover, slowly turn the engine with a breaker bar (feeling for interference) until the cam gear marks line up, and then check the crank pulley for zero alignment. Well, when we did that, we found that the Xterra had NOT jumped time. We did decide to do a timing belt and a water pump while we were there, so we bought the kit, and when we got the bottom part of the timing cover off, we found that the front crank seal was leaking – no surprise on a high miler like this one.

Putting the new crank seal in was something of a demanding process – we tried a few tricks, all of which unseated the garter spring and tried to roll the lip. I kept thinking of transmission seal protectors and how I could fabricate one for this job. Finally, I fetched a soft red plastic hole plug that had been protecting one of the ports on the 2011 HHR’s replacement engine and modified the plug with my pocketknife, making a seal protector for the Xterra front crank seal that worked so well I should have patented it.

The actual cause for the customer’s concern was deep in the distributor – it’d spark and then it wouldn’t and vice versa. We didn’t want to take a chance on that kind of “maybe,” so this one also got a brand new one.

At the end of that job, we found the real reason for that no-start. The spark coming out of that distributor was a come-and-go event. We got no spark from the towers, and so, with the cap off, we checked it at the coil. On the first spin, there was no spark – on the second spin, spark was popping there, and so we reinstalled the cap and the engine fired up and ran like new.

Unwilling to trust that come-and-go spark, we replaced the distributor with a reman unit. Now she has a new timing belt, crank seal, water pump and distributor.  Maybe that Xterra will be good for a while.

Article Details

Mon, 01/01/2018 (All day)

Richard McCuistian

<p>Some people will almost always believe a half-baked diagnosis, and some mistakes are very easily made.</p>

<p>auto repair, diagnosis, Xterra</p>

I was called to a shop with a complaint of an ABS light on. The vehicle was a 2017 Subaru Impreza with a 2.0L engine (Figure 1). It was involved in an extensive front-end collision and experienced severe damage to the right front suspension. The shop had replaced a lot of suspension parts including a right front wheel speed sensor. They cleared all onboard control modules of error codes in memory after all the repairs were completed, but the ABS codes still remained in memory.  

Figure 1

Figure 2

When I arrived at the shop I noticed that the vehicle had multiple warning lights illuminated on the instrument dash panel (Figure 2). There was a light for the antilock brakes, stability control, collision avoidance, lane keep safe and the eye sight control systems. It is almost hard to imagine that all of these operating systems were experiencing problems at the same time. They all shared the same network so it was possible that one problem could be resonating a fault in one controller that had an adverse effect on the operation of the other controllers within the network. Many of these controllers today share input sensors among each other on a CAN bus network rather than wiring one sensor input to various controllers.  

Time to dive in 

It is always best to go into every controller and record codes stored as ”current” or ”past” codes. Then do an overview to see if these codes all point to one area of concern. If there is a problem in one controller that is on a shared network, you can almost be guaranteed that other controllers will record complaint codes redirecting your attention to the specific controller having an issue. Do not get in the habit of doing a vehicle scan of all controllers at once because you may not be guaranteed to pull ALL codes from every control module. Some scan tools may pull “present” codes but may not pull “past” codes and this could defeat the purpose of you putting together a complete game plan of attack. It is vital to compile as much information as needed to narrow down the suspect in your diagnostic routine. 

Figure 3

In this particular case, I performed a quick vehicle scan just get a basic overview of what was going on but knowing that if needed I would have to go to individual modules for a deeper dig. The engine and transmission modules on the network were pointing fingers towards the ABS control module for a fault in vehicle speed error by setting a code P0500. The ABS module stored codes C0022 for front right ABS sensor signal fault and a code C1424 for ECM abnormal (Figure 3). The right front wheel speed sensor had some kind of operational issues. At this point I had to verify the problem and decide whether it was s mechanical issue or an electrical issue. 

Most ABS systems will perform a static and dynamic test of the operating system to alert the driver of any issues. On startup there will be an integrity check just like there would be in the engine management system for component tests. This could require either one or even two-three key cycles. So I simply cleared all the codes and turned off the ignition switch for one minute. I then proceeded to start the vehicle and let it run for one minute. I did this three times to insure a three-cycle event and the warning lights did not come on at all. This is a quick procedure that can be done from the driver seat and holds great value to limit a lot of wasted time by guaranteeing that your problem is not electrical but more mechanical. Understand that the ABS module is sending reference voltage/reference ground into all of the ABS Wheel Speed sensors and checking all solenoid/relay/ lamp circuits for a driver threshold status while it validates the system during key on operation. If there were any electrical issues a light would have been on. Now it was time to go ahead with a dynamic test. 

Narrowing down the possibilities 

I went to the ABS data PIDs and selected the front and rear Wheel Speed Sensors and proceeded to drive the vehicle about 10 MPH (Figure 4). The right front wheel did not show any wheel speed signal at all and as I continued to drive the warning lights all came back on. This was definitely a mechanical condition. The culprits here on my list as I was driving back to the shop were the ABS sensor was not seated in the mounting hole properly, the tone ring was bad or the sensing device within the new sensor was bad. 

Figure 4

This vehicle uses a Magnetic Resistive type of wheel speed sensor and is not your normal AC output type. On a typical magnetic sensor, you will find a field coil wrapped around a magnet. This sensor field coil could commonly have a resistance value of about 500-1200 OHMs depending on its design and reads off a metallic trigger wheel. It will output an AC signal whose amplitude depends on the strength of the sensor magnet, resistance of the field coil and the air gap between the magnet and trigger wheel. The only downfall with these types of sensors are that they are more prone to pick up unwanted electrical noise and cannot produce enough AC voltage below 3 MPH to detect vehicle creep. 

The Magnetic Resistive type sensor looks similar to a magnetic sensor but its circuitry is different. This sensor also uses two wires, but it is dependent on a reference supply voltage and a reference ground feed to the sensing device that will in turn act on the reference voltage line to toggle it and produce a digital signal. The sensor itself will need to be triggered by magnetic bar segments and the air gap between the sensing device and magnets is crucial for proper operation. This type of sensor is more dependable because it can actually measure vehicle creep because its amplitude is not dependent on vehicle speed and it is less susceptible to electrical interference. 

The front hub bearing has a seal on each side of the bearing. Only one side of the bearing has segmented magnets imbedded into the seal along its circumference for the ABS wheel speed sensor to read. It is not uncommon for any shop to put this bearing in backwards because it can be installed in either direction. The segmented magnetic ring side must face the sensor. There is an installation tool (ATE #760130) that you can purchase off of Amazon to insure proper installation that houses a very fine metal powder in an enclosed plastic film that when placed over the proper side of the bearing will show the magnetic segments. My guess was that the bearing could be in backwards. 

I had the shop pull apart the right front suspension so I could take a close look at the repairs. This was a point where I had to perform “Sesame Street” tactics that I learned many years ago when I was but a six-year old kid watching my favorite show and singing “One of these things was not like the other.” It is so vital to perform visual inspections as part of your diagnostic routine. I had one vehicle a few months back where a shop installed a used spindle on a vehicle and the wheel speed sensor hole was offset by a half inch because it was the wrong year spindle for the car so at this point I had to keep an open mind and play outside the sand box. 

Figure 5

When I started to inspect the right-side hub bearing I could see it was installed properly with the magnetic ring facing toward the sensor (Figure 5) and everything seemed fine. At this point I was scratching my head and decided to take the other side apart to start comparing side to side measurements. As I was looking closely I sat there in awe with what I discovered. If you looked at the Wheel Speed Sensor hole on the left side of the vehicle the hub bearing was protruding three-fourths into the sensor hole diameter (Figure 6). The right-side bearing was only about one-fourth into the hole diameter. I could not believe what I was seeing and I never knew this could even happen. 

Figure 6

Figure 7

After taking out the new hub assembly and comparing it to the old hub assembly you could see that the inner part of the new hub bearing was shorter in length (Figure 7). This goes back to the old saying that the parts guy can be your best friend or your worst enemy. It’s so hard to believe that a parts guy could hand you a wrong part and at the same time an installer does not take the time to match a part up. We are in such a rush in today’s automotive world that we just don’t take the time to check everything in our routine repairs. There are no longer guarantees that you are getting a correct part or even a working part. It just ends up putting you down a different path of “denial diagnostics” that can ruin your day and create many unwanted hours of wasted time and not to mention your loss in confidence in your work. My only hopes is that this story hits home with many of you techs out there and that you keep on the watch for this not to happen to you.

Article Details

Thu, 02/01/2018 (All day)

John Anello

<p>When I arrived at the shop I noticed that the vehicle had multiple warning lights illuminated on the instrument dash panel. There was a light for the antilock brakes, stability control, collision avoidance, lane keep safe and the eye sight control systems.</p>

<p>Subaru Impreza, ABS light, service repair</p>

I’ve been working on my own vehicles since I first started turning a wrench. Not because I wanted to — it was a simple matter of economics. I couldn’t afford to pay someone else to do it for me. For the last 20-odd years, I’ve had to work in the dirt driveway of my current place of residence but was blessed a year or so ago with the chance to build a real garage next to my house, complete with two-post lift (a gift from a very good friend)!

I refer to my shop as the “Motor Age Garage” in honor of the column that has been a staple of this magazine for as long as I can remember. Writing for that column was my first role as a new contributor to the magazine, and it’s where we start our newbies to this day. By the way, we’ve got some great young talent to introduce you to!

The birth of the shop — where all the magic happens!

In addition to making life easier maintaining my own cars, it also provides me with the means to produce the videos you see on our YouTube channel and has opened up a variety of opportunities to share topics that I couldn’t before. On the other side of the coin, it opens me up to a lot of requests for help from my youngest son and his friends. I’ve done everything from oil changes to evaporator core replacements (Ford F-150, full dash removal) and soon, I’ll have his Chevy truck in for a valve “tick” that I suspect may be a wiped camshaft. Oh, the joy…

I’ll admit, there are times when I don’t feel like working on a car but I then find myself enjoying the work, getting greasy and beat up again. And I’m keenly aware that, in order for me to do my job here, I have to stay “in the trenches” as much as I can. After all, I can’t tell you guys to continue your education and skill development if I don’t!

Anyway, I recently had a project in the shop that I thought might be of interest to you. It starts with my very favorite customer – my wife.

The shuddering Scion

My better half owns a 2014 Scion tC that is her pride and joy. She’s gone full tilt with it; installing a TRD (Toyota Racing Development) exhaust, leather interior, custom lighting and more. When she cleans it up, she’ll spend a solid 3-4 hours making it look as good as it did when it first rolled off of the showroom floor. And she is just as picky when it comes to maintaining it mechanically.

My wife’s 2014 Scion tC, with a little TRD added

The car has just over 37,000 miles on it and recently displayed a new behavior that my beloved finds truly annoying. She works at the local hospital and leaves the house when it’s still dark. One recent (and also our first “cold”) morning, she headed off to work as she normally does but on the way there, experienced a severe “shudder” when she came to a full stop. The vibration was enough to cause a loud “tapping” noise that could be heard in the cabin and was strongly felt in the steering wheel. Whatever was causing the condition, it wasn’t something the ECM was concerned with – the MIL light remained off.

When she returned home later that day, I met her in the garage to see what the car was doing. Of course, I couldn’t duplicate the problem and after reviewing the scan tool data, could find no clues as to what the problem might be. So I told my spouse I would check it the next morning and she could take my vehicle to work instead.

The following day I went out to the garage and started the car up. Following the same route she takes to work, I tried to duplicate the problem but again, without success. The only difference between my drive to work and hers was the temperature. It doesn’t stay cold long in central Florida!

Waiting for the chill

A few days later, the weather forecast called for another cold night. Here was my chance! Or so I hoped.

I have an exterior/interior thermometer I use for A/C work and I first recorded the air temperature in the shop. It was reading a chilly 41°F, which matched up with the ECT (Engine Coolant Temperature) and IAT (Intake Air Temperature) readings on my scan tool. I started the car and it dutifully went into its cold start mode, keeping the idle high as the engine warmed up. I decided to monitor the ECT, IAT and engine RPM along with STFT (Short Term Fuel Trim) and LTFT (Long Term Fuel Trim) as I retraced my wife’s commute.

A few minutes into the drive, the ECT had reached normal operating temperature and the idle speed had settled down. As soon as the car came to its next full stop, I could feel the vibration and hear the tapping noise my wife was complaining of. The engine rpm appeared to be lower than it should be, hovering around 620 rpm.

It was great having a concrete floor to work on. Adding a lift, though, makes life so much easier!

Specification in Neutral is 680-780 for this car so I shifted into Park to check. Yep, that was OK and at the higher idle, the vibration and noise were gone and the engine was running smoothly. Stepping on the brake and putting the car back into Drive immediately brought the rpm down to 620, and adding the headlights dropped it even further to 580-590. That had to be too low! A quick look at the trims, though, showed perfect fuel control with both numbers staying under +/- 4%.

Next, I applied the parking brake and released the pressure on the brake pedal. The rpm increased to 670-680 and the engine condition was gone. That seemed like a more normal loaded idle speed to me and I was surprised that applying the brake pedal would result in that much of a change. After all, the transmission was already loaded. What possible difference could the brake application make? And why only under “cold” weather conditions?

I think I got it!

I admit, I was at a momentary block in my thinking. I turned to some talented techs that hang out on Facebook and got some ideas from them, and I also talked to the shop lead at my local Toyota dealer. He wasn’t too optimistic about my finding a solution, though, sharing that this was a common complaint that, to date, they have been unable to solve!

One great idea that came out of the discussions was to consider the impact the brake booster may have on the engine. The vacuum booster could be leaking and doing so only when the brake pedal was in a certain position, the shop foreman shared. He told me that he had found several similar issues when troubleshooting P0171 (System Lean) problems for his customers.

It made sense, so I thought I’d try it. I disconnected the vacuum feed to the booster and closed off the line at both ends and went back out for another test drive. And I’d like to report that the complaint had been resolved.

I’d like to, but I can’t…

The problem was still there; the vibration, the accompanying noise, and the unusual drop in rpm with the brake applied. I did identify one additional factor though. The rpm would not drop any lower than 580. I mean, I could get the rpm to drop lower by adding load but the ECM would recover and bring it back to that minimum. Yet I’m convinced that the ECM’s “minimum” was too low for the car. What else could it be?

A coked throttle body can cause idle issues, but this one was not that bad. I cleaned it anyway.

I should add, before your emails start, that I did clean the throttle body (it was a little dirty, but not badly coked) and performed an induction cleaning (setting a P0304 in the process – a possible clue?). I also performed the idle relearn procedure using the Toyota method and the aftermarket method many of my FB friends suggested. Problem is still not resolved. And, of course, I checked for any related TSBs (Technical Service Bulletins) but found nothing that helped.

But, faithful readers, I’m not done yet. My next step is to research every system that uses the brake pedal position sensor as an input. I’m also going to do some more in-depth engine inspections to see if the problem may be related to any weakness in an individual cylinder or in the VVT (Variable Valve Timing) system. I’m not ready to give up quite yet. I — and I can’t believe I’m saying this — only hope the cold weather remains around long enough for me to verify any fix I make! And I’ll be sure to let you know what I find out in a future Tech Corner column.

Article Details

Mon, 02/26/2018 (All day)

Pete Meier

<p>One recent (and also our first &ldquo;cold&rdquo;) morning, my wife headed off to work as she normally does but on the way there, experienced a severe &ldquo;shudder&rdquo; when she came to a full stop in her 2014 Scion tC.</p>

<p>Scion tC, brake booster, P0171</p>

When a vehicle shows up with a misfire, the first though that comes to mind is something ignition related like a spark plug or an ignition coil because they are such common failures. If that is not the cause, then it’s on to something in the fuel system like a clogged fuel injector or a fuel supply problem. Even though this is not the best approach, it is the one most technicians follow due to previous repairs with the usual suspects. While a good percentage of the time they are able to repair a vehicle with this course of action, when a problem occurs outside of the realm of these common items, the next diagnostic steps become unclear. Even when performing the correct steps, interpretation of the results can be misleading. This vehicle was one of those cases.

This is a 2007 Lincoln Navigator with a 5.4L V8 with 3 valves per cylinder

The patient history
The vehicle is a 2007 Lincoln Navigator with a 5.4L 3-valve Triton engine and 82,810 miles. This SUV was in approximately a week ago for a misfire concern under acceleration during the time the transmission shifts between third and fourth gear. The technician determined that the misfire was due to failing ignition coils and eight new Motorcraft coils were installed. The technician test drove the vehicle under the same conditions that revealed the problem and found that no further misfires were present. The customer was relieved to find that the vehicle did not have a transmission problem, since it was easy to see why they thought that because the misfire occurred just as the transmission was performing an upshift. The vehicle was returned to the customer and all seemed fine until eight days later. 

This is a common problem that occurs when trying to remove spark plugs on this type of engine.  While the threaded portion separates from the cylinder head, the shell of the spark plug remains seized in the combustion chamber and special tooling is required to remove it.

The vehicle returned with an almost constant misfire that was present at idle and while driving it into the shop. After a visual inspection for a coil connector that may have fallen off, the technician grabbed a scan tool and retrieved a code P0305 (Misfire Cylinder 5). He again inspected the connector by pulling it off, checking the pins on both the connector and coil and reinstalling it but to no avail. He then swapped coils between cylinder #5 and #6 and found that the misfire remained on cylinder #5. Staying with the original plan of “Misfire = Ignition Problem,” he attempted to remove the spark plug from cylinder #5. Anyone who has replaced spark plugs on one of these engines knows what I mean when I said “attempted.” The threaded portion of the spark plug and the body came out, but the rest of the metal shell remained frozen inside the combustion chamber. The spark plug was extracted without any incident and a new Motorcraft spark plug was installed. I inspected the spark plug and could not find any signs of carbon tracking or damage to the nose of the plug other than what was caused by the extraction tool during removal. 

Misfire equals something else
The misfire remained without any change in intensity and I did suggest to the technician to verify the spark plug was actually firing before reinstalling it into the cylinder head. So, if it’s not ignition, it has to be fuel right? (If Misfire ≠ Ignition, then Misfire = Fuel Problem). The fuel injectors on these engines are fairly easy to swap with only a couple of bolts holding down each side of the fuel rail, so the injectors between cylinders #5 and #6 were swapped and the fuel rail tightened back down. Of course, the engine still had the misfire when restarted, but the technician was confident that it had moved to cylinder #6. So he cleared the code and was going to test drive it to let the code prove that the misfire followed the fuel injector. Because the misfire was pretty severe, I offered to let him use my Ford IDS and its power balance test to show a graph of each cylinder and its misfire history. Much to his amazement, cylinder #5 was still the only cylinder misfiring. So he has ruled out spark and fuel, only thing left is compression, right?

An easy first test to perform with the Ford IDS scan tool is a relative compression test. While cylinder 5 is slightly lower than the other cylinders, it did not provide conclusive evidence that an internal engine problem existed.

Unfortunately, Ford does not list an actual specification for compression but rather a Min/Max range between cylinders. After the dilemma with removing the first spark plug, he was not going to remove the other seven to perform a traditional compression test. But he did test cranking compression on cylinder #5. He stated the result was 190 psi so compression was good. At this point, he threw his hands up and asked the service writer to send this one to me.

It’s got to be something!
One of the great benefits of the Ford IDS scan tool is the power balance test; but caution must be used since there has been a time or two when the wrong cylinder has been identified as the one misfiring. The misfire monitoring is a strategy that is learned, unlike a circuit code for a problem with an injector or ignition coil. There can be variances in the tooth spacing of the crankshaft reluctor, so these characteristics must be learned to enable the misfire monitor. This is usually done after the Keep Alive Memory (KAM) has been cleared and is accomplished by performing a few decelerations from 60 mph to 40 mph without braking. The fuel is cut to the engine so no combustion takes place during the learning process eliminating that variable from the calculation. There should also be a code P0315 (Unable to Learn Profile) if the vehicle was not able to learn the profile correction. That code was not present, but that does not guarantee that cylinder #5 was the one misfiring.

I have chased an incorrectly identified misfiring cylinder before, and it was a frustrating learning experience. A simple way to verify that the correct cylinder is being identified is to create another misfire or two on different cylinders from the one identified and verify that those cylinders are correctly identified also. I perform this by watching the power balance test and unplugging a coil on a different number cylinder and verifying it is correctly shown on the test. An injector would have the same effect if it is easier to access than the ignition coil. Needless to say, cylinder #5 was the one with the misfire so that eliminated that possibility.

Another test that should be performed on any vehicle with a misfire before delving too deep is a relative compression test. While I usually perform this test with an oscilloscope, a current clamp, and sync it with the cylinder #1 ignition coil, the Ford IDS has a built in test that can be performed by depressing the throttle fully to the floor (to prevent fueling during the test) and cranking the engine for 10 seconds.

What I did find when performing this test initially was that cylinder #5 had 4 percent lower compression than the other cylinders, indicating the possibility of a mechanical engine problem, but it seemed too small of a difference for this scenario. However, since the cylinder was misfiring, not to mention a fuel injector swap between cylinders, there was also the possibility of a cylinder wall that has partially washed down from excessive fueling.

While the other technician already swapped the ignition coil and fuel injector from cylinder 5 without any change in the misfire for that cylinder, I needed to verify that the actual signals which activated the ignition coil and fuel injector were present, since even a known good component will not work correctly without the proper signals.

Even though the previous technician swapped the ignition coil and fuel injector from cylinder #6, that didn’t mean that there wasn’t an issue with the signals from the Powertrain Control Module (PCM) or wiring to the fuel injector or ignition coil for that cylinder. I attached a scope to both the fuel injector and ignition coil of cylinder #5 and found that both were receiving the correct voltage signal and were operating as designed. At this point I am satisfied that there is some type of mechanical problem affecting only cylinder #5. 

Cranking compression isn’t the whole story
An easy test to perform without removing anything other than the vacuum hose is to perform a cranking vacuum test with a First Look Sensor (FLS). I installed the FLS into the hose going to the vacuum brake booster and synced to the Cylinder 1 ignition coil. The FLS measures changes in pressure, so this means that each time an intake valve opens a downward hump or “pull” is indicated in the waveform.

Here is the capture using the First Look Sensor in the intake while synced to the cylinder #1 ignition coil. Note that the intake pull for cylinder #1 starts approximately 360 degrees from when the ignition coil fires.

An engine in good condition will produce a consistent and relatively even pull for each cylinder in the firing order, since each cylinder should be drawing the same amount of air and each intake valve should be open for the same amount of time, but when an engine has a problem the steady repeatable pattern will become erratic. This is where the sync comes in. By triggering off a known good cylinder and using the rotational rulers of the scope, it is possible to divide the period between ignition firings into eight evenly spaced divisions with each representing an individual cylinder. Keep in mind that the intake stroke for the cylinder that has the ignition sync occurs about 360° before the intake stroke. The firing order for this engine is 1-3-7-2-6-5-4-8. Looking at the capture, there is definitely a problem, but it is not something that most can pick out unless they are very familiar with FLS pattern recognition.

Here is the same capture with the relative piston stroke position of cylinder 5. Notice how the symptom of a leaking intake valve does not show up during the intake pull of cylinder 5 but the effect it has on the other cylinder’s intake pulls depending on where the cylinder 5 piston stroke position is during their intake valve opening.

The pull for cylinder #5 looks fine, however where the intake pulls for cylinders #3 and #7 should be, there are actually pushes or upward pulses. One of the reasons behind this type of a pattern is that if the intake side of cylinder #5 is not sealing, it can leak cylinder pressure back into the intake manifold (where the FLS is connected to) and disrupt the normal balance between cylinders. However, I am not confident enough in my abilities to make the call just yet so a little more testing needs to be performed first.

The cranking compression results found while using the Pico WPS-500X pressure transducer matched what the previous technician found using a mechanical gauge.

While there are no definite specifications listed for compression of this vehicle, 190psi would normally be a very good number on most vehicles.  However after a few more revolutions, the number started dropping.

Since the other technician had already broken the original spark plug while trying to remove it and installed a replacement, I was not hesitant to remove the new plug and install a compression hose attached to a pressure transducer for my next test. If that was not the case on this type of engine, I would have opted for different testing methods rather than risking removing and possibly breaking the spark plug unless absolutely necessary. With the pressure transducer installed and the injectors disabled, a cranking compression test is performed, and I was a bit confused by the results. I know the previous technician also performed a test with a mechanical gauge, and I also came up with the same results – 190 psi of cranking compression and did not notice any discernable indications of a mechanical failure.

However, is 190 psi of compression good? Is it high? Is it low? Remember, Ford does not list a min/max specification, only a range between what the highest and lowest cylinders can vary by. Based on what happened with the first plug that was removed, I am definitely not going to remove more spark plugs to perform comparison compression tests. Something interesting to point out is that as the cranking compression test continued, a decrease in pressure occurred by almost 14 psi as shown in the capture. The next test to perform, especially since the pressure transducer and compression hose are already installed, is a running compression test. 

The running compression test is where the problem is revealed.  Normallthe compression would be between 1/3 and 1/2 of the cranking compression reading, but here it does not even reach 16psi.

A look inside
For the previous technician and several others, this is an overlooked test that reveals a lot of information. This is a dynamic test compared to the cranking compression test and some performance and tuning shops use this test to determine how well a particular cylinder is contributing to the engine. The test is performed with the fuel disabled to the cylinder being tested, which on this vehicle was as simple as unplugging the injector and running the engine at idle. The pressure of a good cylinder should be approximately one-third to one-half of the cranking compression test result. A throttle snap is also recommended during a running compression test, which should be approximately 80 percent of cranking compression results for that cylinder; however, when using an in-cylinder pressure transducer instead of a conventional compression gauge, it is not uncommon to see a higher reading than on the original cranking compression test. So, starting the engine and reading the test results showed a startling discovery — the running compression for cylinder #5 was only around 16 psi, a lot less than the 63-95 psi that would be expected and a snap throttle only increased to 70 psi.

While the vacuum achieved on the expansion stroke appears good at 21inHg, the vacuum reached during the intake stroke is around 15inHg, indicating an intake valve is not sealing.

On a healthy cylinder, this would have been 150 psi or higher, but was under half of that number. A good amount of the time, when snap throttle results are lower than expected but the running compression is normal, the problem lies in the intake side of the system, and if they are higher it points to the exhaust side, sometimes a restriction. This, however, showed very low running and snap throttle results. So now we know we definitely have an internal engine mechanical problem on cylinder #5 which is going to be intake related. This is also gathered from looking at the vacuum achieved on the expansion stroke and intake stroke. Usually both vacuum pockets are in the vicinity of a normal engine vacuum and are even with each other. When the expansion stroke vacuum is relatively good and the intake stroke vacuum is low, it generally points to an intake valve not sealing. The reason the expansion stroke vacuum is relatively good is because when the intake valve is leaking, it is open to the intake manifold where a vacuum already exists. An important point to remember that I was taught in pressure transducer training class is that the camshaft lobe opens the valve, but the valve spring closes the valve and keeps it closed.

With the driver side valve cover removed, a broken spring for one of the two intake valves of cylinder 5 is found.

Going by these observations, a phone call was made to the customer and approval was given to remove the Bank #2 valve cover for further inspection. With the valve cover on the driver side removed the test results were confirmed. One of the two intake valve springs for cylinder #5 was broken which was the cause of the misfire. 

An old-school style technique was used to hold the valve in place while changing the spring. Some small nylon rope was fed into the spark plug hole while the piston was at the bottom of the cylinder and then the crankshaft was rotated to compress the rope against the face of the valve which worked perfectly. The rocker arm pops out and back in without any bolts but a specific type of valve spring compressor was needed due to space constraints.

The cause of the misfire, a broken intake valve spring on cylinder 5.

Once everything was reassembled the Navigator idled and drove smoothly without any further misfire complaints. With ignition being such a common cause of misfires it can sometimes be hard to step back and consider basic mechanical failures, especially with modern engine reliability. However, with the increased amount of labor required to remove engine covers and components for inspection, old style tests can be performed with new tools and techniques to help isolate problems before teardown.

Article Details

Thu, 03/01/2018 (All day)

Michael Miller

<p>This &rsquo;07 Navigator had a misfire. Follow along and see if you would have taken these same diagnostic steps.</p>

<p>Lincoln Navigator, misfire, diagnostics</p>

I started out my diagnosis on a 2003 Ford Escape with an automatic 3.0 VIN 1 24 Valve V-6 just like I would any other vehicle that I bring in the bay. First, the customer interview to gather the information and the complaint — with just shy of 200K on the odometer, the MIL light is on — then gather some codes and data, come up with a plan, repair and verify and then move onto the next one. Little did I know my plans were about to be derailed! 

Figure 1

Figure 2

The customer had no complaints other than a glowing MIL light. A quick scan revealed a P0452 "EVAP System Pressure Low" both pending and current (Figure 1). Nothing new for us in the rust belt to see EVAP codes on a day to day basis, that's for sure! While sitting in the driver's seat, I decided to have a quick look at the Fuel Tank Pressure (FTP) data PID and quickly discovered it was at 2.6v KOER, which is normal for a Ford EVAP pressure sensor sitting at atmospheric pressure. Wanting to see if the FTP sensor would respond to vacuum, I opened the vapor management valve (purge solenoid) with the scan tool and saw the voltage drop to around 2.19v (Figure 2). That is about what I would expect to see considering the canister vent was still open. OK, so what is going on here? I don't recall at the time what possessed me to increase the rpms, but I did. Perhaps it was in an effort to gain a bit more vacuum on the decel while watching the FTP data PID with the vapor management valve still open. Whatever the thought was, it turned out well because it revealed the problem! At EXACTLY 3000 rpms, the FTP data PID would drop straight to zero volts (Figure 3)! OK, you have my attention, I thought.

Figure 3

Figure 4

I now knew what the ECM was looking at and why it flagged the code. After all, it met the code setting criteria. It was time to get to know my enemy. I pulled up a wiring diagram to see what I could gather from it and verify what I already knew (Figure 4). It is a standard three-wire 5v pressure transducer that was easily accessible under the driver side rear seat. A ground wire, signal wire and a 5v reference. It also shared the same 5v reference as the Differential Pressure Feedback EGR (DPFE) and the Throttle Position Sensor (TPS). Thinking to myself that if the sensor shares the 5v reference and it was losing it, and it happened to be at or before that splice S105 that I saw in the diagram, certainly we could see that in scan data because it should affect the other sensors sharing the same 5v reference, right? 

Time to test a theory

After a moment of recreating the conditions and monitoring the TPS and DPFE PIDs, they appeared to be unaffected. I knew it could not be losing the ground side because if that were the case, the voltage should go high. I verified this by simply unplugging the sensor and indeed it did go high (Figure 5). While I still had the sensor unplugged, I recreated the conditions of the fault and the signal stayed nice and steady. However, after plugging it back in the condition was still present (Figure 6). That being said, we did not correct the condition by unplugging it or messing with the harness. You know as well as I do how frustrating that can be.

Figure 5

Figure 6

Now that I had grabbed what data I could, I decided to grab a graphing DVOM and have a look at the wires right at the FTP sensor in hopes to get an idea as to what was going on. It is a two-channel meter, so I monitored the 5v ref, signal wire and ground all at the same time and it revealed exactly what the scan tool was showing. When the signal wire dropped to 0v at 3000 rpm, the ground and 5v ref stayed perfect. Or so they seemed. Not really knowing where to go next, I decided to just double check the circuit integrity of the signal wire from the ECM to the FTP sensor. I unplugged the FTP sensor and the ECM, supplied battery voltage on the signal wire and with a 780mA test light on the other end and it seems to carry current just fine (Figure 7).

Figure 7

Figure 8

At this point, I really just don't know. With a shop parking lot full of other work, it was getting hard to concentrate. It is a terrible feeling as a mechanic and shop owner, as many of us know. Part of me is thinking the ECM is gone wonky but the fact that it is so predictable and it only does it at 3000 rpm is still weighing in my mind. I decided to play some "swaptronics" and substitute a known good (used) FTP sensor (Figure 8). A quick trip to see my friends at the local salvage yard and we were back at the shop. Used sensor is now installed and low and behold it does the same exact thing! Needing to think for a bit I decided to gather a bit more data on a test drive and it revealed that driving it made no difference and it still drops out at 3000 rpm. Engine torque, bumps, forward, reverse, hot or cold, nothing made a difference.

Not beaten yet!

Feeling like I am at the end of my rope and out of ideas for the moment, I decided to use my "phone a friend” life line. I called my good friend and mobile tech Keith DeFazio from New Level Auto Diagnostics in Staten Island, NY. We discussed what tests had been performed and what the possible causes were, and tried several things such as unplugging DPFE, TPS and retesting. 

Low and behold, unplugging either the DPFE or the TPS sensors caused the problem to disappear! WHY!? Was something happening on the 5v reference? It was time to have a closer look, using the scope feature on my DVOM. With the scope hooked back up, there it was — staring me in the face. At 3000 rpm, the 5v reference would go into a high frequency hysteria, for lack of a better term (Figure 9). Seeing this only brought about more questions though. The main one being, where is all this noise coming from and why is it effecting the signal wire and making it drop right to zero volts? 

Figure 9

While I was still contemplating ideas in my head knowing I was one step closer, I needed to know if the signal was being "sent" by the ECM or "received." The easiest way to do that was to just cut the wire near the ECM and scope both ends of it and I discovered it was coming from the FTP sensor to the ECM. I still had no idea why though. However, it did lead me away from thinking it was the ECM at the time. All I knew was something was changing at 3000 rpms and was effecting it. Of course, at this point I have tried unplugging the coils, the alternator and various other components, just to try and get an idea where this noise is coming from and was not making any progress. I decided to study scan data some more to try to gather some clues as to what could be changing. I thought to myself, is the alternator duty cycle changing, is there a device or output turning on at the same time, ANYTHING! Still nothing. 

Deciding to call it a night after a busy day at the shop and working on this in between other cars, I figured it is best to go home and sleep on it hoping it comes to me in a dream. Laugh if you want, but I am certain many of you have been there before! I woke up in a panic at 3 a.m. and rushed back down to the shop to fix this mind-bender. On my way out the door, it dawned on me I never saw a Power Steering Pressure (PSP) switch input on the scan tool. Could that be changing? I clearly remember thinking to myself, "that is a 5v switch, right?" I walked back in the shop, fired the scan tool back up and unplugged the PSP switch. I GOT IT!!! The problem was gone! But why!? 

At this point, part of me did not care, but the rest of me wanted to know what was going on. I grabbed the scope, back probed the connector and there it was (Figure 10). The PSP switch at 3000 rpm was opening and closing like a mad man. I also discovered I was wrong; it is not a 5v switch — it is indeed a 12v switch. Good thing I was tired and did not give it a second thought. Had I known it ran off a 12 circuit would I have gone back in and checked it? Hard to say at this point. As they say, hind sight is always 20/20.

Figure 10

A new PSP switch was fitted the following morning. The EVAP code was repaired, drive cycle was completed and the symptoms were corrected. As a mechanic who grew up in this field and learns from self-study and by getting my butt kicked from time to time, this one humbled me and taught me a few things. The first thing it taught me was to use my scope feature from the get go even if it seems unnecessary. You never know what it might show you. It also reminds us sometimes we do need to walk away, clear our heads and not give up! I also learned the value of having a great friend in the industry who shares the same passion for repairing vehicles as I have and is always willing to bounce ideas around.

I tend to chuckle every time I have a vehicle come in now with an EVAP code tripping the MIL because you just never know. It may just need a new power steering pressure switch! Who would have thought.

Article Details

Sun, 04/01/2018 (All day)

Eric Obrochta

<p>I started out my diagnosis on a 2003 Ford Escape with an automatic 3.0 VIN 1 24 Valve V-6 with the customer interview to gather the information and the complaint &mdash; with just shy of 200K on the odometer, the MIL light is on. Then I gather some codes and data, come up with a plan, repair and verify. Little did I know my plans were about to be derailed!</p>

<p>EVAP Code P0452, auto repair, South Main Auto</p>

We use wiring diagrams in many of our diagnostics, but if we are not careful, they can sometimes lead us to make decisions that are not accurate, which can lead to wasted diagnostic time, unnecessary parts costs for the replacing parts that are not defective, and sometimes even missing a simple repair.

One area where I have noticed a wide skills gap when helping other technicians diagnose a problem is in the use of wiring diagrams — not reading them, but more importantly interpreting them. While there have been several very informative articles and training classes on the subject, the one that has had the greatest impact on improving my circuit diagnosis was a technique invented by Jorge Menchu of AESwave called Color Coding. His technique uses various colors to represent what types of signals to expect at certain points in a circuit and help narrow down where the problem is by seeing what is and isn’t working as it is designed to. I point this out since the colors I used to highlight the circuits in this article are based off of this technique, and I also use this information to design/change my diagnostic plan. A color coding kit (AES# 02-WDCC) is available from AESwave.com.

But alas, even when using circuit wiring diagrams and having proper techniques, there are times when the provided information does not show the whole picture, which can cause inaccurate diagnostic summaries and wasted replacement of components.

When it isn’t the bulb

How often when a vehicle comes in with a complaint of a bulb not working do we or the customer automatically just install a new one? In 95 percent of the vehicles that have this concern, bulb replacement fixes it, so for the most part it could be a valid first step. However, if it doesn’t work, it can turn out to be a problem vehicle especially if the wiring diagrams get a little complicated. This is what happened on a 2008 GMC Acadia SLT that had 82,439 miles with the complaint of an inoperative RF turn signal. The technician who was originally assigned the repair order started with a replacement bulb, but found that this repair was not going be that easy. Apparently, the bulb has already been replaced by either the customer or another shop so their next step was to determine if there was correct voltage and ground supplied to the bulb; a quick check with a Digital Multimeter (DMM) showed no voltage. Looking at the wiring diagram for the exterior lighting, they determined that the Body Control Module (BCM) was at fault because in their thought process that is what supplies the voltage to the turn signal and since the Right Rear Turn Signal was working; it must be getting the request from the Multifunction Switch. The tech checked the powers and grounds to the BCM and they were OK, so a new BCM was installed and set up. Obviously if I’m writing about this vehicle, that didn’t fix the concern. 

Figure 1 - The only code that appeared in the BCM was a B2615 for Courtesy Lamp Control, but since the circuit description did not have anything to do with the exterior lighting I kept my focus on the turn signal concern.

Like most diagnostics, if I’m not sure how a system is designed to work I do some research before testing. This is also the point at which I print out a wiring diagram and highlight what a correctly working circuit should look like. I find the same circuit also includes the turn signal on the RF side-view mirror, and I noticed that it is not working either; however, the Right Rear Turn Signal is on a completely different circuit and is working as designed. I also gathered from the wiring diagram that the BCM (Connector 4 Pin 5 DK BLU/WHT wire) is what controls the circuit once the input from the turn signal switch is received (Connector 1 Pin 16 DK BLU/WHT wire). Since the BCM controls the turn signal circuit, it’s a good idea to check for codes and when I did I found a B2516 Passenger Compartment Dimming 2 Circuit (Figure 1). A quick look up of the code with a description of the circuit shows that this is related to the courtesy lighting circuit, which I notice is not working. This does not seem to have any effect on the exterior lighting circuit so I decided to stay focused on the turn signal problem and keep that info in the back of my mind. Now I remove the RF turn signal bulb socket to start my voltage tests. I can see from the wiring diagram that the ground for the RF turn signal — in this case G102 — is a constant; it is the first signal I check with my LOADpro voltmeter Leads to test the circuit under a load. Next we move on to the supplied voltage side of the circuit. Since the BCM is easily accessible by the driver-side kick panel, I perform my testing there.

Known good – known bad?

When using a scope to diagnose a problem, it is a good idea to have a known good signal to compare a possibly defective signal to, so I also monitored the Left Front Turn Signal input and output (Connector 1 Pin 16 LT BLU/WHT and Connector 5 Pin 4 LT BLU/WHT respectively), since I know this side is working as designed. As you can see (Figure 2), both inputs are working correctly but only the LF turn signal output is being generated by the BCM; nothing is happening on the RF turn signal output circuit. I also turn on the hazard flashers as another input source to the BCM and have the same result with an inoperative RF turn signal output. Next I use a Power Probe to apply battery voltage to the RF turn signal circuit at the BCM harness with the connector unplugged and the directional bulb at the RF corner illuminates. This tells me that the circuit is intact and can handle the load when applied. Now I am starting to see why the previous tech suspected the BCM.

Figure 2 - This is the scope capture from the BCM input and output controls of the turn signals.  Notice that the input is received, but there is no output for the RF turn signal.

While looking at the wiring diagram for the exterior lighting circuit, I notice a few pins that are B+ supplies to the BCM, and one of the fuses is even labeled as the Right Turn Signal. Something important to remember when testing voltage supplies and grounds to a module is to look at the actual module wiring diagram. While the exterior lighting wiring diagram shows some power supplies, it does not show the whole picture of the module itself (Figures 3, 4).  I start with the grounds first; it looks like Pins 1 and 5 (Connector 3, both BLK/WHT) and Pin 9 (Connector 4 BLK) are the grounds, and all three test fine. Next I move on to verifying the voltage supply pins. I find that there are four pins, all RED/WHT wires numbers 1-4 that are supposed to have B+, but find that Pin 2 does not; it is an open circuit. Guess where the voltage supply comes from?  Remember the code in the BCM for the courtesy circuit? The fuse that supplies B+ to this pin was open. After replacing the fuse, the RF turn signal worked.

Figure 3 - (Diagram courtesy of Mitchell Pro Demand) The wiring diagram for the Exterior Lighting shows only 3 B+ inputs for the BCM, which all tested fine.

Figure 4 - (Diagram courtesy of Mitchell Pro Demand) The wiring diagram for the BCM itself shows another B+ input at Pin 2, note that this does not show on the Exterior Lighting wiring diagram and was not tested by the original tech that diagnosed the vehicle.

I asked the other tech again to verify that he tested all the B+ supply circuits; he said yes and showed me the exterior lighting diagram and found it only lists Pins 1, 3 and 4 as B+; however, Pin 2 is not shown on the exterior lighting circuit. That is why it is important to use the actual module wiring diagram to check for B+ and ground. I do not understand why this power supply would affect only the RF turn signal, especially since there appeared to be a dedicated fuse for the right-side turn signal, but it does go to show that we must not get tunnel vision when performing something as simple as a lighting circuit diagnostic, as there may be a bigger picture.

Not quite done

So the vehicle is fixed right? Well, sort of. The RF turn signal is working (Figure 5), but the directional on right side-view mirror is still inoperative. As stated before, the wiring diagram shows both the RF turn signal and Passenger Outside mirror are on the same circuit, in fact the mirror is spliced to the same wire from the BCM before going through the Underhood Fuse Block so it eliminates that part of the wiring automatically. Well, it looks like the best place to test is at the connector for the mirror itself, so we can see if the voltage signal is present and test the ground. After removing the door panel the problem was pretty easy to see: the mirror that was on the vehicle was incorrect for the application, the connector pins for the mirror side of the harness did not align to the pins in the original door harness, and there was a second mirror harness connector on the door that did not have anything plugged into it.

Figure 5 - The scope capture of the repaired circuit showing all inputs and outputs are working as designed.

Someone had just attached a side-view mirror that looked correct (on the outside) from a GM vehicle with different options. In hindsight I could have saved myself the trouble of removing the door panel by trying to move the mirror glass with the controls as none of the functions of the mirror worked. Repairing the circuit for the RF turn signal restored the double-time flashing of the right turn signal indicator on the instrument cluster; the inoperative side-view mirror directional did not affect the rate of the flasher. I did not find out what caused the courtesy fuse to blow, but I don’t know what happened to the original side-view mirror on the vehicle, either.

Another bulb in question

The next vehicle that was given to me was a 2008 Dodge Avenger with 112,976 miles and a 2.4L engine for a complaint of an inoperative right front low beam headlamp. A little background about this vehicle before it ended up in my bay: The customer has already tried replacing the bulb themself, however when the vehicle arrived there was no bulb to be found at the RF headlamp connector, in fact there was a new connector already spliced in with butt connectors (Figure 6). The technician who got to look at the vehicle first also knew the customer had tried replacing the bulb, so they installed a voltmeter across the bulb connector and turned the headlamps on - 12V! They assumed that maybe the customer had purchased a defective bulb, but it was not found in the vehicle, and we didn’t have another in stock to test. Since the bulbs are very easy to replace on this vehicle, he swapped the left front low beam bulb to the right side instead of ordering a new one, and he knew the left headlamp worked fine. Same problem: the bulb did not illuminate on the right side. He swapped it back to the left side and it again worked perfectly.

Figure 6 - The customer has already attempted to perform their own wiring repairs to the vehicle along with replacing the bulb, which was not in the vehicle when we received it for diagnostics.  Not sure why they used so many butt connectors but we had to make sure his attempted repairs were not negatively affecting the circuit.

I can understand the confusion and frustration of the technician since he verified he had voltage and ground at the connector with the headlamp switched on, so why was the bulb not working? He pulled a wiring diagram for the headlamp circuit and saw that the RF low beam headlamp is a fairly simple circuit that has a constant ground and voltage is supplied by the Totally Integrated Power Module (TIPM). So he asked me for a second opinion before recommending a new module.

When looking at the wiring diagram, I like to start with the ground side of the circuit and highlight it. I noticed that ground is constant as he stated, but it is also shared by the right front high beam bulb and the right front fog lamp bulb, both of which are working normally, so it doesn’t look like we have a problem with high resistance on the ground portion of the circuit. Another item I noticed is the customer replaced the connector to the RF low beam headlamp, with multiple butt connectors. Fortunately they were not affecting the operation of the circuit.

Figure 7 - A halogen headlamp is substituted for the missing bulb by wiring it into the connector.  This is also a great way to load test a system.

Figure 8 - A capture from the graphing multimeter shows that when the bulb is connected the supplied voltage drops to 0V, but when it is disconnected again the voltage returns. This is why the technician found battery voltage on their DMM when testing the circuit, it was not under load.

Next I move on to the voltage supply side of the system. Again as the technician stated, voltage is supplied to the RF low beam headlamp from the TIPM. So to verify my understanding of the circuit I used a graphing multimeter and back probe Pins 1 and 2 of the RF low beam connector and wired in a headlamp bulb (Figure 7) that I also use to load test the circuit. When the headlamps switch was turned on my GMM showed no voltage; unplugging my wired-in headlamp from connector I had battery voltage again. Reconnecting the headlamp to the circuit I experienced the voltage dropping back to 0V (Figure 8) .   

It appears to be a defective driver in the Totally Integrated Power Module (TIPM), but let’s not jump the gun until we verify the voltage and ground supplied to it first; we’ve already experienced that in our last case study. The TIPM is easy to access and had several connectors attached to the underside of the module. Using the actual TIPM wiring diagram and not the wiring diagram for the headlamp circuit, we find there is a larger B+ wire directly from the battery, which supplies voltage to the module and multiple grounds to check, again testing them under load since simple checking voltage it not going to reveal a problem as we just witnessed with the headlamp circuit. All the voltage and ground circuits to the TIPM are fine, but just to be on the safe side, I simulate the work of the TIPM and supply voltage to Connector 5 Pin 3 WHT/TAN wire to verify the integrity of the rest of the circuit to the headlamp and my wired-in headlamp bulb illuminates brightly, proving the problem is with the RF low beam driver inside of the TIPM.

The customer approved the replacement of the TIPM but did not want to pay us to replace the headlamp bulb; they had the bulb at home and would install it themselves. With the new TIPM installed I still wanted to verify my repair so I attached my tester headlamp in place of the missing headlamp bulb and it worked great, at least by doing this I can be confident when the customer installs their new bulb it will work.

As I stated in the beginning, the repairs themselves were simple but using the correct wiring diagrams and techniques to understand how the circuits that operated them are designed to work is the key. Not doing so makes it easy to get misled if a solid understanding of the wiring diagrams is not in place.

Article Details

Mon, 10/02/2017 (All day)

Michael Miller

<p>We use wiring diagrams in many of our diagnostics, but if we are not careful, they can sometimes lead us to make decisions that are not accurate, which can lead to wasted diagnostic time, unnecessary parts costs for the replacing parts that are not defective, and sometimes even missing a simple repair.</p>

<p>wiring diagrams, automotive repair, techniques</p>

I was called to a shop for a simple task of reprogramming a new ECM on a 2012 Ford Focus with a 2.0L engine (Figure 1). There are a lot of shops in the industry that do not want the responsibility of programming a new ECM due to the liability and costs involved in doing the job. Some shops may not be properly equipped with the proper interface or laptop and may not want to go out to the Internet to fill out an application to sign up with an individual manufacturer and purchase a daily subscription. This may be coupled with a need to be registered through NASTF as a Security Professional to access theft procedures that may be needed during the post programming procedures. There is also the need to have a proper battery charger in the shop that can provide a "Programming" mode that will maintain a specific voltage range while a current surge occurs such a cooling fan coming on during the programming process. Some manufacturers will terminate programming if voltage falls below 13.1 volts or if voltage exceeds 14.0 volts. An older battery charger with only Low, Medium and High settings is not recommended and may pose a problem if too much AC ripple is introduced.

Figure 1

Is the ECM to blame?
When I arrived at the repair shop I proceeded to question the shop technician to see why he had made the decision to replace the ECM. It is not uncommon for a shop to jump the gun to condemn a control module without probable cause. I'm not in the business to just go ahead and reprogram a Control module at will without making sure it will resolve their issue with the vehicle. The last thing I need is to charge a shop to program a new or used control module only to find out that it was wasted revenue without a cure for their problem. I will usually question the shop to see if they followed the proper procedures to condemn the module. This would include checking power and ground feeds, shorted reference voltages, communication lines or even an issue with their testing equipment.

Figure 2

The shop told me that they just replaced the transmission (Figure 2) and when they got done the vehicle would no longer start. They checked the vehicle and found that the ECM was not responsive so they figured it might have been damaged during the repair procedure. This is a bad situation of "Drive them in / Push them out". We have all been there before at one time or another in the repair business. It becomes an unwanted marriage between you and the vehicle and your only way out is to resolve the issue at your own cost because you can't expect the customer to pay it. In the end it all becomes a learning experience but you’re under the gun to get things resolved ASAP before the customer is aware of his or her new dilemma.

When I hooked up to the new ECM to program it I was unable to communicate with it. The shop was under the impression that the new ECM would not communicate because it needed to be programmed. This is misinformation that I see a lot of shops seem to believe in and I had to school them on this belief and to educate them on how to evaluate no communication issues with a vehicle. They have to always make sure that prior to condemning a controller they check EVERY power and ground feed at the ECM and scan the entire vehicle to make sure it is not just a single control module issue without focusing too much of their attention only on one single controller. By scanning the entire vehicle they will get a better evaluation of all the operating systems that are responsible in starting the vehicle. Sometimes the clues to their problems may be resonated in other modules on board that may put them on a better path in their diagnostic process.

Start at the beginning
I placed my scan tool on the vehicle and did a full scan of the entire vehicle. I discovered that only five control modules were present on the multiple networks this vehicle had on board (Figure 3). These were the Audio, Body, GPS, Instrument and Tire Pressure Monitor control modules. This vehicle had three separate CAN networks on board: Medium Speed, High Speed and Entertainment CAN networks. The High Speed CAN network was inoperative. The controllers on this network included the Transmission, Engine, ABS, Power Steering, Steering Angle, Air Bag and Occupant Classification control modules. It was highly unlikely that all these control modules were bad or that they all had a common power or ground feed failure. What was common to all of them was a network circuit that was either open or shorted.

Figure 3

It is very important to understand how a CAN network is structured prior to testing it. Keep in mind that all the controllers wired on a single CAN network are all considered nodes and they are all wired in parallel. The CAN network consists of a twisted pair of communication lines to eliminate electromagnetic interference. There has to be a way to stabilize waveform reflection within the network and this is done by putting a terminating resister at each end of the network. A 120 ohm resistor is commonly used at both ends of the network and may be internal to a module or externally mounted on the network wiring. The 120 ohm resister internal to a control module will usually be the control modules furthest at the ends of the CAN network. If we use Ohms Law we can mathematically figure out the total resistance of the network. The total resistance of the network will always be the sum of the inverse of each individual resistance. Total resistance will also always be lower than the smallest resistance. In this case:

1/Rt = 1/R1 + 1/R2

1/120 = 0.0083

0.0083 + 0.0083 = 0.0166

1/0.0166 = 60.24

With this equation we should expect about 60 ohms across each network.

The battery would have to be disconnected prior to doing an ohm check because the system cannot be alive while using an ohm meter. I used an OBD interface box to make things easier while working under the dash. As you get older you want to work smarter without having to put too much pressure on the lower part of your vertebrae while hanging upside down under the dash. The interface has labeled ports that reflect the 16 pins of the OBD II connector and each port is coupled with LED's that alert your attention to presence of power, ground or data activity. I used my Fluke meter and tested the Medium Speed CAN network first at pins #3 and #11 and acquired a reading of 62.8 ohms (Figure 4). Then I preceded to the Entertainment CAN at pins #1 and #8 and acquired a reading of 62.3 ohms (Figure 5). These readings were well within the tolerances of a healthy CAN network.

Figure 4	Figure 5

Module or wiring?
I now proceeded to check the High Speed CAN network pins #6 and #14 and acquired a reading of 2.5 ohms (Figure 6). This network was shorted together. This would make sense why seven controllers fell off the radar. The hard part now was to put together a game plan of attack to locate the source of the problem. The possibilities that could be laid out were that any one of these seven controllers could be shorting out the network or the network harness could be shorted somewhere in the vehicle. I printed out a diagram and circled the seven controllers involved and checked off the ones that were the easiest to access (Figure 7). The eighth controller was an optional Parking Aid Control Module that was not fitted on the vehicle. My plan was to unplug each controller one at a time to see if the short would be eliminated. If the problem remained then I knew I had a harness problem.

Figure 6

Figure 7

I wanted to start with the easiest controllers to gain access to without too much work involved so I went to the engine compartment first. I disconnected the engine and transmission control modules that were already exposed near the base of the transmission and there was no change. I next decided to go after the ABS control module that was located at the driver side firewall. In order to access this module I had to pull the battery out of the way (Figure 8). As I was reaching for the ABS connector I was looking down at something that caught my eye. There was a harness crushed under the battery tray (Figure 9). This was not good. I proceeded to remove the crushed harness and exposed what turned out to be damaged CAN network wires (Figure 10).

Figure 8	Figure 9

Problem found — on to the fix
The two wires that were crushed in the network connector were for the High Speed CAN network. I simply repaired the wiring and put everything back together and the vehicle started right up with no problems. It was amazing how these two wires took down the vehicle and put it out of commission during a routine R&R of a transmission assembly. These seven control modules were cut off from the rest of the vehicle as if they were stranded on Gilligan's Island away from all interaction with the rest of the crew. The seven modules were are still responsive and in good working order with good powers and grounds but were no longer able to communicate among each other or with other controllers on the network. This led to the other controllers setting "U" codes for loss of communication with modules within the networks.

Figure 10

Repairing vehicles today are much more difficult because you’re now working in tighter quarters and components are not always easy to get to or even remove and replace. It is so important to watch your every step to make sure things are put back the way you find them and to be very careful not to put components in harm’s way. When you experience a problem you always need to go back to where you were last. Don't be so homed in to one module problem but branch out you diagnostic routine to include a full scan of the entire vehicle to enable you to dot the I's and cross the T's.

Article Details

Wed, 11/01/2017 (All day)

John Anello

<p>I was called to a shop for a simple task of reprogramming a new ECM on a 2012 Ford Focus with a 2.0L engine</p>

<p>ECM, reprogramming</p>

In this month’s Tech Corner, I would like to share an experience I had when still full-time as a technician. It’s an experience I’m sure most of you have also enjoyed, or not, depending on your outlook on life. You know, the old “glass half full” kind of thing.

The car in question is an older Ford Mustang with the 3.8 liter V-6, with a hard misfire on cylinder #1. Follow along and see how you would have tackled this one!

The First Mistake

The customer had brought the Mustang in for a complaint of a rough idle and stumble on acceleration. After a short test drive, it was easy enough to tell that there was a serious misfire going on. I hooked up my scan tool and found code P0301 (cylinder #1 misfire) and P0316 (misfire detected on startup) stored in the Engine Control Module (ECM). 

Figure 1 - The answer to the Ford misfire is in this picture. Do you see it?

This vehicle uses a Direct Ignition System (DIS) that fires two plugs simultaneously. Opening the hood, I could hear the distinctive "tick" of a spark jumping to ground outside the cylinder. Looking a little more closely, I could see the spark jumping to the valve cover from the #1 wire. The wires looked like original equipment, and a closer inspection revealed signs of leakage in the others. 

On this type of coil, one plug is "positive," and one is "negative." When the coil discharges, current first travels to ground thru the negative plug, then back to the coil through the positive plug. When the coil is stressed, the internal insulation can fail, reducing total coil output. In this low state, there is just not enough voltage left to jump the gap on the second plug, even though the first plug continues to run just fine. That's why it's possible to have a DIS coil with one dead plug. 

Thinking I had this one nailed, I ordered a replacement coil and ignition wires and moved on to the next car on my list. Time is money when you’re working flatrate! 

The Second Mistake

When the parts arrived later in the day, I pulled the Mustang back in to the bay. It is a simple installation and took no time at all. I cleared the codes and went to verify the repair. Have you guessed yet? The miss was still there, and the MIL light was back on. 

You would think that after all the time I've had in this business I would remember my personal rules regarding diagnostics - Never take a shortcut, especially on a misfire code. 

A misfire code can be set by any condition that doesn't allow for complete combustion in the cylinder. My normal procedure is to first do a relative compression test to ensure the engine is mechanically sound. Doing that test now indicated that the #1 cylinder had a problem.

If I see a low cylinder indication on this quick and dirty test, I follow up with a normal compression test. 60 psi was all I got on the misfiring cylinder. What I found had me muttering a few words under my breath. I was kicking myself for breaking the rules, and now I had a major engine fault to explain to my customer.

Was the original repair necessary? Replacing the ignition wires may have been; however, the coil was a rushed diagnosis. The low firing line I had seen on my scope was a result of low compression – not low spark energy. Remember, the firing line is typically affected by pressure, gaps in the system and the amount of hydrocarbons available for conduction. The scope was trying to tell me something. I just wasn't listening, instead choosing to see what I wanted to see based on an assumption. 

What’s The Fix?

The next step I took was to perform a cylinder leak-down test. This test uses a tool called a differential cylinder pressure tester and has two gauges on it. One indicates line pressure (supplied by shop air), and the other is the pressure being contained in the cylinder. When connected, and with the cylinder to be checked at TDC of its compression stroke, the tool pressurizes the cylinder and you compare the two pressure readings on the gauges.

When connected to the 3.8’s #1 hole, the left side gauge displayed the line pressure of 90 psi. The right side gauge reads the pressure in the cylinder, showing 70 psi. That's a 20 psi difference, or a little more than 20 percent of line pressure.  Not a lot, but standard specification is no more than 10 percent difference.

With the line still connected, I removed the oil fill cap, radiator cap and air filter housing. That 20 percent of air pressure is going somewhere, and you can actually hear it escaping. That's the nice thing about this tool. It allows you to hear if the loss of compression is from the valves (air escaping from the throttle body or exhaust pipe), the rings (air escaping from the oil fill) or from the head gasket (air escaping from the radiator).

This one was a no-brainer. Air was rushing out of the throttle body with no evidence of air flowing through any of my other checkpoints. OK, now I've got it. The intake valve is leaking. I got authorization to remove the head, confident that this was the problem. 

With the head removed, I verified the valve was leaking by pouring solvent into the intake port and looking to see if any leaked past the valve on the combustion chamber side. It began to pour out as soon as the solvent got to the valve face. But because I had been burned on my first diagnosis, I needed to be extra thorough. I also checked the installed valve height to see if there might be a problem with bent valves or recessed valve seats and found no problems there. I inspected the push rods for damage, and the cam lobes for wear. While the head was off, I rotated all the cylinders to the bottom of their travel to look for damage to the cylinder walls.  

Everything looked good.

I got the head back a few days later and reinstalled it on the Mustang. I turned the key, and you should have heard the expletives that followed!

What Had I Missed??

You've got to know that I am really upset by now. I felt I had done a thorough diagnosis, and I had definitely found a major flaw in the leaking intake valve. Thinking that maybe the machine shop had done something wrong, I checked cylinder leakage with my tester. This time, the results showed no leakage.

But what about compression? Again, I got a low reading on the #1 cylinder. What is going to make a tight cylinder low on compression? The only answer I could come up with is that the cylinder couldn't breathe. However, I had checked the valve train and had found no problem. 

I pulled off the valve cover and rechecked the valve operation, measuring opening and closing heights of the valves on #1 and comparing them to #2 and #3. I could not find the problem. 

There was only one answer left. It had to be in the bottom end.

Again, I removed the head on my way to the piston, and here I'll tease you with the photo in Figure 1. Do you see what I should have noticed the first time? 

Figure 2 - I still don’t know how the #1 rod was bent, but it caused the piston to fall short of TDC and lowered the effective compression in that cylinder.

With the piston removed, the problem was obvious. Looking at Figure 2, do you see what should have caught my eye? The interesting thing about this failure is that the rod bent almost perfectly along its axis, effectively shortening its length. Other than that, there were no other symptoms — no noise, no vibration and no bearing damage.

Look closely at the stain on the cylinder wall where the ring travel ends near the top. The cylinder in the foreground is #1; #2 is behind it. Notice how the stain is thicker on #1, showing that the piston wasn't reaching TDC. I should have caught this when I had the head off the first time. Would you have caught it? 

What would bend the rod? Perhaps it was hydraulic lock from a leaking head gasket and the failed rod went undetected, or ignored, during that repair. When I first performed my visual inspection, the oil level was correct with no sign of intermix. Coolant levels in both the reservoir and radiator were correct as well, and there was no air escaping thru the radiator during the first leak-down test.

Learn from My Mistakes

Hindsight is always 20/20, they say. My first mistake was not performing my normal diagnostic routine and checking engine integrity from the start. My second mistake was not considering that a 20% leakage rate (a nominal amount) would result in a compression reading of only 60 psi. The two did not agree with one another, what the ECM would flag as a “rationality” error. Now, in my defense, who among you would have expected to find a bent rod or would have caught the visual signs that were present especially considering that there was no other evidence present? Even so, the visual cues were there when I had the head off the first time, and I missed it. When troubleshooting any kind of problem, remember what Spock told Captain Kirk – “We must fall back upon the old axiom that when all other contingencies fail, whatever remains, however improbable, must be the truth.”

Article Details

Fri, 12/01/2017 (All day)

Pete Meier

<p>Every technician has been there &mdash; deep in the diagnosis, tunnel vision sets in, and often memorable mistakes are made. But it&#39;s only a failure of you don&#39;t learn from the experience!</p>

<p>misfire, diagnostics, auto repair</p>

One day I got a very terse call from the vice president of the company where I was responsible for fleet maintenance back in the late ‘70s. It seemed that an almost new (1978) Dodge one-ton we had was pointed at the gate with a gooseneck trailer behind it and that truck and trailer needed to arrive at our offshore diving and salvaging dock within the next 30 minutes – 25 miles away. I had no idea why that trip to that dock was so urgent, but someone had misplaced the key to the Dodge.

“Get that truck started and on the road within the next 10 minutes,” he told me with his gravelly voice, “and I don’t care what it takes. Just make it happen.” 

I must admit that I was in my element under pressure in those days, so I hung up the phone and grabbed a jumper wire with a couple of ‘gator clips on each end out of my toolbox. I opened the hood on the Dodge and made a connection from the positive battery terminal to the ballast resistor to feed current to the ignition coil. Making sure the tranny was in neutral, I “pocket screwdrivered” the starter to fire the engine up. Ninety seconds had expired and the steering wheel was still locked, but I knew I could defeat the pewter collar around that silly spring-loaded steering wheel lock peg, and I slid into the seat and muscled the wheel hard to the right, and broke the lock. Mission accomplished in less than three minutes and the truck was headed out the gate.

Then there was the time at that same job where I had to drive down Highway 87 toward Galveston and take a steamy ride on one marsh buggy through a swarm of mosquitoes and dragonflies to another marsh buggy that had jumped time, stranding a different vice president and his passengers a couple of miles off the road. Putting a timing belt on while standing in snake and alligator-infested water and swatting away mosquitoes wasn’t my idea of a good time, but I was motivated enough that I got that job done in record time, too.

This is my 2007 F150 that was victim of a surgical strike by some toothy critter that was copper-hungry

The point is that every job isn’t interesting, but in our line of work, challenges are the spice of life, and it feels good to be a problem-solver. It feels even better to be appreciated, and usually we are, but that isn’t always the case.

Critters

Dogs and squirrels chew wires, as do rats. Rats and squirrels build nests in engine compartments, and cats looking for a warm place to sleep can die under the hood and under the car in very gruesome ways sometimes. I’ve had to kill spiders and roaches, wasps, dirt daubers and all manner of other wildlife in my under-the-hood and under-the-vehicle odysseys. One morning I did a classroom presentation on critter damage, and a day or so later I walked out to where I park my own F-150, slid in behind the wheel, and thought I was going somewhere in my truck, but it wasn’t to be. The battery was good and hot, but I had no starter operation and no scan tool communication. The red theft light was blinking, which can point to a few different problems, but it usually means a module (usually the PCM) isn’t talking. With the key on, I checked for voltage at the EGR assembly and found 9 volts on the gray-red signal return wire – which should have been grounded through the PCM. What that meant to me was that the PCM had lost its own ground reference somehow.

Having Alldata available on the smartphone is pretty handy when you’re under the gun to find out what’s wrong and you’re somewhere else besides the shop

Next it was time to bust out my smart phone and dig into ALLDATA, where I found that PCM G103 is located behind the battery on the bulkhead. With my flashlight, I peered down there and saw that about eight inches of that wire had been removed by some sharp little teeth and my much larger main power feed cable to the inside fuse panel had been just as viciously attacked, but it had survived without being severed. Some chew-happy squirrel must have a nice piece of wire lining its nest and a belly full of copper as I type these words.

Is Your Routine Holding You Back?
	If you are not applying your training to every single customer you assist, you are doing them and yourself a disservice. Read how in Pete Meierís newest story and learn about a free trial of the best video training on the market. Button text: Read Peteís Story

There was another ground wire in that same area that was compromised as well. While removing the battery and doing some solder and heat shrink work was almost enjoyable that Saturday morning, I found myself wondering if I was going to have this problem again. No other wires under the hood had been attacked.  It was almost like the critter had pulled up a wiring schematic and did a surgical strike to prevent my truck from going anywhere. And it worked.

I suppose I should have been thankful that these wires were the only ones the critter chewed – he could have done a lot more damage than he did – fixing this took about thirty minutes.

I prevailed in that fix and placed some rat poison in the general area. We’ll see how that works out.

The 2004 Suburban

In a previous article, I mentioned a 2004 Suburban with a 5.3L that was misfiring on cylinder 4 with low compression and, during the cylinder leakage test air was escaping into the exhaust, but the owner chose to drive it skipping for a while before having it fixed. Finally, the Suburban returned and we hashed out what needed doing.

This was one of those high-milers, and so I talked them into a reman engine because of the better warranty, which we managed to stuff in there in pretty good time. 

We were going back into the Suburban with the reman engine and had just put it in place when this photo was taken. You can see the burned valve in this photo of the old engine’s head if you know what to look for.

After the swap, I had Robbie jerk the head off so we could inspect the valves and the head of the piston on the offending cylinder and we found a valve that had become mis-matched with its seat and was leaking compression. With the new engine in place, the MIL was off, the monitors all cleared, O2 sensors were switching handily, and fuel trims were bouncing around the zero line, so we put that one back on the road. There was a strange caveat though. For some reason, the transmission wouldn’t go into park well enough to not roll away on a slope.

This was a deal-breaker, to be sure. The shift cable was adjusted as far as it could go. I could disconnect the cable and put the transmission fully in park, so there was nothing wrong inside the case. Eventually I decided to try the shift lever off another transmission I had on the shelf and with that lever installed, it would go completely into park just fine even though it looked the same. I have yet to figure that one out, but it was safe when it left.

The Silverado, the Fusion, and the MKZ

While all this was going on, another instructor who drives a 2003 Silverado 2500 Duramax asked if we could replace his master cylinder. He’s ordinarily pretty savvy, and since he brought us the master cylinder I had a guy pop it on there and begin the bleeding process. Well, the pedal felt like you were stepping on a plum, and there was fluid dripping from underneath the truck and we found that classic rusty brake line situation a lot of you guys have to fix every day. He didn’t look under the truck, I don’t guess. A careful exam of the whole system revealed that this line was a lot worse than any of the other lines, all of which looked pretty good, and so we got a roll of that dandy nickel-alloy rust-free stuff and built a replacement line from stem to stern (complete with new double flare fittings), and after the bleeding procedure, we got that one rolling again with a good firm pedal and a master cylinder he didn’t need. I gave him the rest of that $60 roll of brake line just in case something else would be needed later.

This rusty brake line syndrome is fairly common everywhere on trucks of all kinds, especially on trucks that do a lot of mudding, but salt did this one in. This Silverado had spent its early life in Panama City Florida, where the salt air took its toll.

The 2010 Fusion that came in around this time was making a bump noise underneath on the left side during parking lot maneuvers, and it was one of those cranky situations where you can’t see anything but you know something is wrong. And every bolt was tightened to no avail. This one has that odd double-ball joint design with two lower control arms, and when we applied the Chassis Ear^® we found that one of the control arms was the source of the bump, and when we got it out of there you could see the problem. The hidden rubber that is couched in the frame area had died, and that was allowing the sudden pressure of certain braking and steering maneuvers to give a metal-to-metal contact sound. The fix was easy enough.

This inner control arm bushing isn’t visible on the Fusion until you remove the control arm. This was the rear of two control arms that car is blessed with on each side.

That Fusion reminded me of another vehicle, a 2012 MKZ that came in with an alternator you could hear whining from 100 feet away. She had been to a tire shop complaining about a noise, and the first guy who rode with her at that shop said he thought the noise was a hub bearing, but the more experienced mechanic said, “no, that’s the alternator,” because it was making the noise when the car was sitting still and the pitch of it matched engine speed. When I heard it, I agreed with the older guy’s prognosis. That alternator was making a LOT of noise that changed with throttle.

Getting the alternator off a 2012 MKZ isn’t for wimps — the refrigerant has to be recovered and the A/C compressor has to be removed, and the alternator comes out the bottom. There’s nothing easy about any of that job, but my guy got it done. I knew this 17-year-old could handle it — he had just finished replacing the heater core in a 2010 Wrangler, and after that extremely difficult job, this one was a cake walk.

The new alternator didn’t whine — the car sounded normal under the hood now, but it did have what sounded like a noisy hub bearing on the right front at road speed. It was one of those situations where the customer didn’t believe she had needed the alternator to begin with, because she was still hearing a noise on the road and one noise was masking the other. She chose to drive the car for a few days but came back and claimed the car had “put her down” and implied that it was our fault for replacing the alternator.

Here was another needful repair. The customer on this one was complaining of a vibration with the blower on high – easy to figure out and easy to fix, but needful all the same.

“I left it running when I got here,” she told me, and then said, “I had to jump it off this morning and then I drove it here (15 miles). It wasn’t giving that problem before you replaced my alternator.”

I had her pull it into the shop. I carefully explained that if the alternator wasn’t charging, the engine would have died as soon as the jumper cables were removed. Then I switched the car off and tried to restart it, but the battery was too weak. When I jumped it off and connected the Snap-on tester I showed her that the alternator was indeed charging and suggested that she find a cool place to rest while I did some more troubleshooting to figure out what was going on, but I told her I’d need the rest of the morning to be sure of what was going on.

“There was nothing wrong with my battery,” she snipped. “We’ll get to the bottom of this,” I told her.

After she walked away, I put a good stiff charge on the battery with a heavy-duty charger, then I got out the $1,700 Midtronics unit I bought from Joey Henrichs and ran through the entire routine, which records everything, including battery health and alternator ripple, printing it out for the customer. The MKZ showed a clean bill of health all the way around except for the battery.

Taking it a step further, I did a parasitic drain test, connecting a meter in series with the battery and waiting until all the modules finished charging their stuff. End of story — there was no drain, only a weak battery.

When she came back I showed her the results and told her she’d need to get the hub bearing noise handled at the tire shop. Sometimes it’s best to send some customers down the road, so that’s what I did.

Another Silverado

This 2003 1500 5.3L came to us with an overheating complaint — the guy said another shop claimed it must be a blown head gasket, but I explained that we wanted to diagnose it ourselves before we did any unnecessary surgery. Sure enough, it was overheating, but it was happening slowly, and there was no quick pressure buildup in the cooling system when the engine was started. The fan kicked on at 228 but the engine kept getting hotter until the fans kicked on high, and all that took a while, but I noticed that the radiator was still cool.

Here’s the overheating Silverado. Even with the radiator removed and bypassed and the thermostat gutted there was no flow through the hose.  Presumably this was a water pump problem?

“Let’s try a thermostat,” I told my guy. Cheap and easy comes first. We put one of those in there and burped it out, but nothing changed. The radiator was cool, but the engine was getting hot.  So I had him pull the water pump, and the borescope didn’t show anything wrong down in the pump, so we reinstalled it. With the radiator removed (no external clogging seen) I bypassed the radiator using the long hose, and we also looped out the transmission cooler lines and took the guts out of the old thermostat to allow free flow. With the engine running we had to squeeze the hose in the middle to neutralize the natural kink so as to facilitate flow but even with that hose in place of the radiator, there was still no flow through the hose, which was only warm on the ends — not in the middle. And the engine continued to try and run hot. What madness was this? If coolant had been flowing, the hose would have been hot its entire length.

Here’s the Silverado’s water pump with the rear cover removed, but I couldn’t see a problem, nor could I feel one manipulating the pulley and holding the impeller.

This was one of those exhaust bolts that had rust-melted from a 15mm down to something just a little larger than a 9/16, and it wasn’t in an easy place to access. We air-hammered this wiggler onto the bolt to get it out. It was a needful thing, but getting the bolt separated from the socket was tough.

To make a long story short, a radiator and a water pump fixed that one. Mission accomplished, but I couldn’t figure out what was at the root of this problem — I thought that plastic impeller might have been spinning on the shaft, but water pump forensics didn’t show that to be the case. One way or another, the truck never runs over 210 degrees now. Happy customer.

Article Details

Fri, 12/01/2017 (All day)

Richard McCuistian

<p>Every job isn&rsquo;t interesting, but in our line of work, challenges are the spice of life, and it feels good to be a problem-solver. It feels even better to be appreciated, and usually we are, but that isn&rsquo;t always the case.</p>

<p>auto repair, Silverado, Suburban</p>

Years ago we troubleshot a Grand Prix that had run just fine until the owner’s cousin had changed the intake manifold gasket and afterwards it was skipping dead on cylinder 2, so she asked if we could have a look at it. This was an engine skip – how hard could it be? First, we checked the obvious stuff (spark plug, compression, etc.) and came up short. But what we did find was that the number 2 injector didn’t sound right with the stethoscope, so we replaced that injector with a known good one, but to no avail. We then checked the entire injector circuit for shorts of any kind and excessive resistance, pin fit at the ECM and the injector, current flow through that circuit with the injector artificially energized (0.8 amps) and ran a temporary circuit overlay. Nothing changed.

When I finally scoped the injector pulse and compared it to the others, the pulse was strangely narrow, so I called a local salvage yard and obtained a replacement ECM. No cigar. Not even close. I replaced that ECM with a second one, because the salvage yard had a bunch of them on hand and they were only $20 each. I double checked everything. This made no sense at all. Finally, I Scotch-locked that injector’s trigger wire to the adjacent injector’s trigger and the car ran great from then on with no more problems. Remember, this was an OBDI system.

I didn’t like that temporary fix, but one of the GM engineers who was as stumped as I was told me those early GM ECM injector drivers can each handle 4 amps, and it’d be just fine carrying two 0.8 amp nozzles. Even if it had burned out a driver and I needed to keep digging, I still had two other good ECMs on hand. One way or another, that Grand Prix holds the distinction of being a grueling fueling enigma that still has me wondering to this day.

Burning in bad info

In the world of politics, news media and other sensitive areas, some have discovered that you can repeat some supposed fact enough that most of the hearers begin to believe it, regardless of its veracity. Our customers – some of them anyway – can also convince themselves that they know what’s wrong when they have little or no useful data except the symptom. Then there are those who have a vehicle concern and somebody they know who seems to have a bit of automotive knowledge makes a superficial jackrabbit diagnosis, hopping quickly across the high points without doing much else. And don’t you love those customers who bring you some parts they want installed based on an offhand diagnosis made by somebody who either doesn’t know how to do the work or “doesn’t have time?”

Even when we begin to gather data scientifically, we can still misfire on our diagnosis, and anybody who claims they haven’t been there isn’t being truthful. For just one example among many, I would have sworn in a court of law that the left rear axle bearing was ruined on my aunt’s ‘92 Crown Vic – after all, that’s where the noise seemed to be coming from, and it changed for the worse with a swerve to the right – as it turned out, she had a noisy left front tire and for some reason the noise was telegraphing to the left rear.

Back in the early ’80s a guy wanted me to replace his carburetor because two different shops using offhand diagnostics told him they didn’t do carburetor work, but that it needed replacing. One of them even claimed to have used an ignition scope and was a tune-up shop. It was a small carburetor on an inline six, so first, I bought a $6 Delco carb kit before I did anything else. Afterwards, I did a mild throttle snap and found it dropping a cylinder under load. I identified the cylinder, replaced a bad brand-new spark plug, and fixed that one.

And then there are quite a lot of people who will play the blown head gasket card without having seen anything other than an overheating issue. I don’t know how many times I’ve heard that, and I experienced it once myself on a 1993 Camry I checked for a friend beside the road. That one had split its radiator, overheated and was puking hot, sweet-smelling geysers out of the filler neck when we refilled it and fired it up. After it came to the shop on the hook, I wanted to show the class how that kind of head gasket failure looks and smells, but all those symptoms were gone – all it needed was a radiator. Go figure.

I have a 1989 Ford Bronco that was donated because the owner believed the head gasket was blown, but it was running crappy and puking coolant out the neck because it was a 5.8L, and he had wired it up using an old 5.0L firing order. When we wired it up with the right firing order, all the filler neck geysers went away.

A few years ago, we checked a 2.4L Dodge Stratus with a horrible oil leak that a shop had pegged as a rear main seal (how many offhand rear main seal diagnoses have we seen?) and used dye to find it coming from under the corner of the head.

Those of us who teach for a living know from experience that people who already believe they have all the facts are kind of difficult to convince otherwise.

2011 Chevy HHR 2.2L Ecotec

Bearing bad news

The owner of a 2011 Chevy HHR, 22L Ecotec with 123,598 miles had spent some time and money doing his own offhand diagnosis trying to get it started – he had checked the fuel pressure with a rented gauge, replaced the fuel pump and fiddled around some with a scan tool before realizing that he was in over his head. The HHR been sitting for a few months when it came in on a trailer. They had determined that it had to be something simple and were hoping we could get it going for just a few bucks. Somebody had postulated that it might have a bad crank sensor, and they brought it to us with the hope that we’d find out it was something simple. Well, it wasn’t. This one spun with very uneven compression, and when we removed the valve cover, we found some broken roller rockers, which typically means valves had contacted pistons, usually the result of timing component failure. But the timing chain was nice and tight, and looking down into the chain area I didn’t see any looseness or shattered sliders. Could it have been over-revved enough to float a valve? We didn’t do exploratory surgery, but we sold them on the idea of replacing the bad engine with a good used one.

When we pulled the valve cover on the HHR, we found several broken roller rockers. Something catastrophic happened here, so we decided to stuff a used engine in it.

The salvage yard sent an engine for that one with a few minor differences – the fuel rail had a different shape, along with a couple milder changes. When we were done, that one ran like a top and when we fixed an A/C leak and juiced up the icebox, it even had cold air.

The Expedition and the Crown Vic

A very regular customer brought her 2001 Expedition to us with a nasty coolant leak – this one’s a Triton and they tend to protect themselves from engine damage, but she just kept driving it. The water pipe that travels through the valley under the intake had rusted through and was dumping water almost as fast as you could pour it in. Initially we just removed the intake manifold, cut the rusted-through portion of that pipe out and replaced it with a hose and some clamps along with a new intake, but when we filled it with coolant, put a new thermostat in it and started warming it up, the warming never stopped – pressure was building very rapidly throughout the system and it was evident that this one had indeed blown a gasket.        

She’s a hands-on shopper, so she did her own search for a replacement engine at a price she liked, found one somewhere in Florida, and had it delivered to the shop. It had the heat pucks in some of the expansion plugs and so I knew it came from a reasonably savvy salvage yard. I crossed my fingers, hoping they didn’t turn the engine backwards while removing the torque converter bolts! Sometimes that flips one out of time.

Since I had people doing transmissions this time around, we went ahead and jerked the transmission out first, then I had another guy remove the original engine, and we carried it on the hoist over to the area where we do component swaps.  One of the first things we noticed was the narrow pulleys on the replacement engine – apparently this one had come from a Crown Victoria or a Town Car, but I couldn’t be sure. Oddly enough, a power steering pump came with the replacement engine, and so we took that narrow pulley and put it on the Expedition’s PS pump. We also replaced the idler and the belt tensioner, because those wide ones wouldn’t work on the replacement engine’s timing cover – and we weren’t about to swap out the timing cover if we could get out of it.

Initially, the guy who put the engine in the Expedition had put the generator wire on the top post at the solenoid, and that kept the starter energized when the generator was trying to work. The starter was a casualty in this case, but it’s an easy mistake for a beginner to make. The bottom photo shows the naked grooves in the generator pulley – the A/C compressor had the same issue, but we swapped out the power steering pump pulley to have the right one.

There were some other minor differences, but at the end, that engine was sitting in the frame with a new belt and everything plugged in, and the transmission was reinstalled – having drained the transmission and replaced the filter, we needed to start it up to get all the fluid back in the gearbox. We started with five quarts and hit the key.

When we started the engine to finish refilling the trans, we noticed that it had a large vacuum leak, and we also heard strange noises and smelled something burning – never a good sign, and it wasn’t the oil smoke from exhaust manifold handprints, either. As it turned out, the guy who replaced the engine did everything right except that he made one very easy mistake. He connected the alternator charge wire at the starter relay to the wrong post, which delivered alternator output current to the starter solenoid circuit while the engine was running, and that kept the starter energized, which destroyed the starter. Thankfully, it didn’t destroy anything else. But with that heavy rubber sleeve on the wires leaving the solenoid, it was easy to make that mistake if you weren’t ultra-familiar with the wiring.

This was another easy mistake to make – put a bolt that’s just a little too long in one of these and you’ve ruined a gas tank. We used the bolts and the gasket that came with the new tank, and so this leak really surprised us – even more so when we found out the new pump was faulty.

The 2003 Crown Vic ran out of gas while the gauge was reading a half tank. We replaced the sending unit with a new one from Carquest, and the guy who did the job used one bolt that was a bit too long when installing the pump, and punctured the gas tank. We got a new replacement tank, but after it was filled with gas, the leak was worse than ever. But that leak wasn’t around the gasket, it was around the plastic grommet where the wires pass through the mounting plate – I have not seen that before. Anyway, all’s well that ends well, and there was no fire. Only a bit of wasted gas.

The Xterra

The 2001 Xterra came to us with the concern that it had quit in a parking lot and failed to start, and somebody’s offhand diagnosis was that it had jumped time. This is a dicey situation, because that one isn’t a free-spinner, but even on interference engines, a timing belt can slip enough to stop the engine without valves kissing pistons. Had that happened on this one? I asked her if she had tried to re-start it (of course she had), but she told me she had only tried once and was hoping there was no damage. We didn’t want to do a lot of engine spinning on this one in the bay for fear of possibly making a simple no-start into something worse, so we checked the timing marks first.

On this one you can pull the upper part of the timing cover, slowly turn the engine with a breaker bar (feeling for interference) until the cam gear marks line up, and then check the crank pulley for zero alignment. Well, when we did that, we found that the Xterra had NOT jumped time. We did decide to do a timing belt and a water pump while we were there, so we bought the kit, and when we got the bottom part of the timing cover off, we found that the front crank seal was leaking – no surprise on a high miler like this one.

Putting the new crank seal in was something of a demanding process – we tried a few tricks, all of which unseated the garter spring and tried to roll the lip. I kept thinking of transmission seal protectors and how I could fabricate one for this job. Finally, I fetched a soft red plastic hole plug that had been protecting one of the ports on the 2011 HHR’s replacement engine and modified the plug with my pocketknife, making a seal protector for the Xterra front crank seal that worked so well I should have patented it.

The actual cause for the customer’s concern was deep in the distributor – it’d spark and then it wouldn’t and vice versa. We didn’t want to take a chance on that kind of “maybe,” so this one also got a brand new one.

At the end of that job, we found the real reason for that no-start. The spark coming out of that distributor was a come-and-go event. We got no spark from the towers, and so, with the cap off, we checked it at the coil. On the first spin, there was no spark – on the second spin, spark was popping there, and so we reinstalled the cap and the engine fired up and ran like new.

Unwilling to trust that come-and-go spark, we replaced the distributor with a reman unit. Now she has a new timing belt, crank seal, water pump and distributor.  Maybe that Xterra will be good for a while.

Article Details

Mon, 01/01/2018 (All day)

Richard McCuistian

<p>Some people will almost always believe a half-baked diagnosis, and some mistakes are very easily made.</p>

<p>auto repair, diagnosis, Xterra</p>