Don't Ignore The Obvious Issues

Dan Frazier - Diagnostician

Diagnostician - Dan Frazier Dan has been in the automotive industry for over thirty years and is an ASE Certified Master Technician. Dan has a college background in electronics engineering and specializes in diagnostics and computer controls for Certified Transmission.

I've always loved cars and knew from an early age, that I wanted to be a mechanic when I grew up. Well, I haven't grown up, I'm not a mechanic - I'm an automotive technician, and I've been fortunate to have able to see the progress of automotive technology over many years. The evolution and integration of computer controlled components and their speed and accuracy has and will continue to change the challenges of diagnosing these systems.

How we go about diagnosing electrical and computer problems all depends on what rolls in the door, but the same basic strategies can be applied to just about anything with 4 wheels and a battery. That being said, you have to take into consideration your resources, capabilities, tooling and experience to know what you can make a profit on.

The 1st step is just doing an initial evaluation on the vehicle. Does it start, run, or move? Everybody does this differently but needs to include at a minimum a fluid level check, a visual inspection, and a test drive with a scan tool. I like to take movies of the scan data in case I need to go back and look at anything - volumetric efficiency come to mind.

OK, you got paid for your first .5 for your professional opinion of what is wrong with this transmission. So what else do we need to know? Won't move, full of burnt fluid at 175k, needs a unit, I'm OK and move on. Need to do pressure tests or electrical evaluations? That's different. Got a 08 BMW 550i that intermittently has the transmission go into limp mode and has low voltage and communication codes in just about every module? To the dealer it goes - with the aftermarket wet-cell battery that wasn't registered to the vehicle. Got a 2000 Dodge Ram 5.9 gas with a P0753 - 3-4 shift solenoid circuit? I'm all in. How about a 2006 Cobalt with TCM communication issues intermittently that comes from another very good shop that wants a second opinion? I'll swing a bat at that because I've got plenty of resources to help me deal with issues like that.

So what are your resources? O.E. and aftermarket scan tools, service information from various providers, information from professional trade groups, and personal experience in the field to name a few. I have access to some O.E. Scan tools and reprogramming, but BMW isn't one of them. I also have your standard service information providers, and it seems a lot of information is very limited on Euro stuff. I have a lot of experience in electrical diagnostics, but again, very weak on Euro and very strong on domestic car lines. That's why the BMW left and the old Dodge truck and Cobalt stayed. Everyone has strong and weak points - take advantage of your strengths and try to learn more on your weaknesses.

So let's talk about this fine 2000 Dodge Ram with the P0753 - 3-4 shift solenoid circuit, after seeing this on my initial inspection.

Article Figure 1

Figure 1

I recommended 2 hours of diagnostic time. I'm pretty used to seeing bungled up wiring, usually caused by rodent damage or bad installs, but wasn't sure where this diagnosis would lead me. I knew how to fix the duct tape on the TV cable though. The code was set for a specific circuit and I had a good idea where I would need to go with this. One of the 1st things I considered was this was a circuit code as opposed to a performance code.

If I have a circuit code, the controlling device has detected an electrical issue, like a short or open, rather than a failure to respond to a command. Failing to respond to a command would usually result in a performance code. The module in control doesn't see the expected results from the command but sees no issues electrically. The computer can be applying a clutch pack, VVT solenoid, fuel injector, or whatever - and it knows what changes it wants to see under certain conditions. This will help to lead you in what direction to go. If I have a circuit code, I'm busting out the electrical diagnostic stuff. If I have a performance code, I may be checking pressures or other data on the movie I took while on my initial test drive.

OK, I'll spill the beans early; it's another Chrysler with a bad PCM. After seeing the multiple butt connectors on the injectors, I was really thinking wiring damage was part of the problem, but I went after what the PCM was seeing to set this code. The PCM is seeing the wrong voltage on the 3-4 shift solenoid circuit when it's commanding the solenoid on or off. On this particular transmission, electrical diagnosis is fairly easy, as there only a few electronically controlled devices - the pressure control solenoid, TCC solenoid and the 3-4 shift solenoid.

Article Figure 2

Figure 2

We can see B+ supplied to the transmission from the Transmission control relay to the solenoid pack at pin 1. If I had an issue with the power supply to the solenoid pack, I likely would have seen codes for the other solenoid circuits or transmission relay stuck on or off. Ground for each circuit is provided by the PCM. So I can easily check total circuit resistance by removing the transmission control relay, and measuring the resistance between pin 87 of the relay and the control pin at the PCM. I had 28 ohms; spec for this circuit is 20 to 40 ohms, so I felt I was good there. Easily from there, I can check for a short to ground (which there wasn't), or to power. Now, with the transmission being in limp mode, it shuts off power to the transmission control relay, so you will have to supply power to that circuit by activating the relay with a scan tool, or by using a fused jumper wire or relay bypass and check voltage at the PCM control wire. With power supplied to the circuit, I had close to no voltage at the PCM control wire when the solenoid was commanded off, where I should have close to B+. Unplugging the connector to the PCM gave me a reading of battery voltage at the PCM connector, proving the PCM had an internal short on the 3-4 shift solenoid circuit. A used PCM and a TV cable (remember the duct tape?) and this one is out the door.

BTW, the story behind this particular vehicle is a comedy of errors so to speak. A young kid bought this truck knowing it had a transmission problem - stuck in limp mode. Took it to 2 different shops and was told it needed a solenoid, or some other misinformation that I don't know. Bought a new transmission, installed it and had the same problem as before. It took less than a half hour to reach the correct conclusion as to what was wrong with it.

Ok, we made a little bit of money on that diagnosis. What about this 06 Cobalt with intermittent TCM communication issues? We're going to fix this one by using a kind of different approach. Resources and experience is going to be the key on this one.

We had a very good wholesale customer wanting us to take a look at this Cobalt that intermittently (several times a week) the transmission would go into limp mode and then be OK after the key was turned off and restarted. The shop had already replaced the TCM and flashed it to the latest calibration.

I'm a big fan of the IATN website, (International Automotive Technicians Network), and have used it's database, waveform library, and forums to expand my knowledge and help me gather information about a lot of issues I can use in my everyday routine. If you have Identifix, there's a link to IATN on the home page. Some time ago, I had come across a discussion in the Technical Discussion Forum about the use and issues that aftermarket devices such as insurance dongles, remote start systems, etc., can have an effect on communications, driveability, and transmission operation on GM vehicles. It gives a pretty detailed list of codes and symptoms that can be caused by such devices. If you want to view it, it's in GM Tech Connect from Feb 11, 2013.

Anytime I get a vehicle in with communication issues relating to the transmission, one of the 1st things I look for are any added switches, LED's, or anything that would indicate a non-O.E. device being installed. Another good thing is to take a quick peek under the trim panel below the steering column. If you find something like this, you might want to remove it and see if your issue goes away.

Article Figure 3

Figure 3

On this Cobalt, I found a wire tapped into the hi-speed CAN circuit that was intermittently causing a communication error. This picture isn't from the Cobalt (clutch pedal?) but gives you an idea of what to look for. And while you're looking, be especially wary of those dang Scotch-locks that seem to be so popular. Once you cut into a wire, the damage is done and it can be very difficult to trace a wiring issue down after the fact.

Expanding your knowledge base and resources is often crucial to your success and sometimes doesn't cost very little if anything at all. Talking with other shops, technicians, networking with other professionals through electronic media, and sharing with others is almost necessary today to keep at the top of the pack.