By Jeff Bach
I recently received a Saturn SC2 in my shop to test the transmission electrical system for the cause of multiple transmission codes. Along with the codes that were setting, the transmission would slam so hard into gear that little 1.9 would bark the tires when it shifted from first to second and chirp them on the 2/3 shift.
The car was sent to us by a transmission shop and had already been scanned and diagnosed as having an electrical problem. I am seeing more and more cars with this type of problem. It has what I call a "technical problem," and a technical problem needs a "technical solution." Technical solutions come at a cost of technical knowledge and technical tools, neither of which are cheap.
This technical problem - like so many I see these days - reminds me of the "Highlights" magazines I read as a child. I remember trying to find hidden objects in the middle of a picture, objects that were hidden in plain view by putting them in with similar-looking shapes. I'm probably in left field but this is how I see technical problems addressed by today's manufacturers. The answers are there if you use your imagination, but obscured by intrusive, irrelevant, misleading technical information. I started seeing a tendency toward this trend (and for lack of a better description I'll call it "masking") in the early OBD-II cars.
"Masking" is what I see as designing a diagnostic strategy around a failure that will deliberately mislead a technician by having a code stored in the power control module (PCM), which has in its mentions in the diagnostic trouble tree the specific names of parts involved in the monitoring and setting of codes. However, none of these codes are related to the actual failed part. This strategy seemed to work well in OBD-II cars in conjunction with the actual physical lockout of much of the pertinent available diagnostic data in the PCM to all but the manufacturer-specific scan tools. I say it worked for them in the earlier OBD-II cars because it continues to become much more prevalent in the '98 and newer cars.
Along with this trend, I have noticed a distinct defeatist attitude among many of my more caring colleagues in the independent service industry, not because the information is technically "not there" but that the "masking" is done so well that it requires an inordinate amount of time to get to the bottom of the actual problem, especially when you're led via the diagnostic "scenic route." It seems to help me, when dealing with a lot of the problems I encounter on the '98 and newer cars, if I approach them with an understanding of "the game." It's also helped me to realize the need to keep abreast of the current technology. "Adapt and overcome," I say.
According to the description of diagnostic operation for the transmission control system on this Saturn, there are five solenoids used to control the four gears, the torque converter clutch (TCC) and the pressure control solenoid. They are the second and reverse clutch solenoid, third clutch solenoid, fourth clutch solenoid, a TCC solenoid and a trans fluid press (TFP) or pressure control solenoid. They are all the same part numbers and are all capable of being in a "fully on," "fully off" or "pulse width modulated (PWM) state" somewhere in between. They operate with no current supplied and the control pressure port is open, allowing fluid flow. The PCM turns the solenoid on to stop the fluid flow, so the longer the pulse width, the less the line pressure. While driving, the PCM monitors the shift times in mS and varies the pressure control solenoid pulse width to achieve the desired shift.
After recording some pertinent freeze frame failures records, I determined that the PO748 code, which returned first, was the main code I needed to troubleshoot. One of the other codes that kept returning was a code for the trans temp sensor circuit operation. I bypassed the temp sensor in the warm mode with a 1/4-watt resistor to be sure it wasn't interfering with the PCM's shift strategy.
When a PO748 trouble code sets, it means the line pressure has been commanded low (long pulse width) and the shift times are still too short. In the trouble code chart test description to check for solenoid operation, it states to command solenoid quick test using a scan tool. This will cycle the solenoids at a set rate. Then you should listen to the solenoid and see if it sounds "similar" to the other solenoids. While this might work OK as a "go-no-go" test, I would be embarrassed to call it "the solenoid test" without any pinpoint test to follow. While running this test I had a thought that if someone saw me with my "listening stick," I would say something like: "Two riders approaching, in a hurry, one with a rifle, the other one is left-handed, horses are tired."
I decided to scope the TFP solenoid circuit with my current probe and test drive it to see if I could make sense of the logic involved while I watched what the TCM did with the current. I recorded the freeze frame failure records, which showed that the PCM was commanding minus Kilopascals (KPA) for the shift times.
The waveform in Figure 1 is from the pressure control solenoid cold start. It looked fairly decent although the "sea gull" seemed a little high on the ramp. I checked the voltage supply with the second channel of the scope to be sure I wasn't dealing with a low voltage or high resistance condition. I got Figure 2 during a test drive, which shows a full voltage pattern and good grounding - allowing me to trust the current waveforms.
The test drive had me thinking this was an intermittent problem due to the way it was shifting now, which was perfectly normal. The shift adapt values, according to the chart, were well within the normal range. The PWM waveform for the electronic pressure control (EPC) sol made sense also. The "on" signal was nearly at maximum, thereby holding the solenoid closed against spring pressure to deliver low hydraulic pressure during idle to drive. You could watch the solenoid "on current" and read the pulse width just like a transmission pressure gage needle.
By the time the trans heated up to the point where the solenoid was sticking pretty good, it was taking twice as long for the solenoid plunger to move. Note the sea gull in Figure 3 has done a "Ronald Reagan" (shift to the right).
Seeing this waveform left me with no doubt what was happening with the shift adapts, and I knew for certain what was causing the problem. I had seen a lot of sticking solenoids before, mostly TCC, but never have had the experience of watching one move this far out on the timeline before it stuck completely.
Figure 4 shows the progress of the sticking solenoid as the engine goes from cold-start to warm in about a 20-mile trip, with me taking waveforms at about 1.5 mile intervals.
Notice the PCM was progressively commanding a longer pulse width to get the desired shift times, thereby causing the shift integrator to go into negative numbers. The code didn't set during this whole test drive and the harsh shifting didn't occur until the solenoid plunger stopped moving (uppermost waveform), but you can see what was happening with the scope and current probe.
Confident in my findings, I called the local Saturn dealer to order a new pressure control solenoid. The gentleman that took the call asked me if I meant a solenoid contact plate? I said, "No, I mean a pressure control solenoid." He asked me to hold on a minute, and calls to someone who then picks up the phone and asks, "What is it that you think you need?" The gentleman I'm talking to at this point is the "transmission guy," and I tell him I want the transmission pressure control solenoid. He then asks me if I mean the "solenoid contact plate," and I repeat my request of a "pressure solenoid." He then recommends that I check for a lumpy green spot in the harness.
I know the guy is just trying to be helpful, because I've read about the harness problem on the iATN Web site, along with the contact plate problem. I explain to him that I know it's the solenoid sticking because I can see it on the scope, but by now I can see I'm speaking "Greek" to him. He again tries to convince me that it has to be a bad harness, contact plate or the PCM - that the solenoids rarely go bad. I finally talk him into letting me buy a solenoid, so I quickly send a driver over before they change their mind.
After changing the solenoid, I cross my fingers because I know I'm going against the advice of the "experts." Then I take the waveforms again with the min/ max running and the thing doesn't vary on the whole 20-mile trip (Figure 5).
One thing I have to say that has me excited about the "new game" is that my favorite "game pieces" are becoming more relevant - now more than ever - and in conjunction with the "right" scan tools I can still see my future in this business and it looks to be more fun as time goes on. I can't wait until I get to start scoping 42-volt electrical systems and fuel-cell-driven devices.
 |
Jeff Bach is the owner of CRT Auto Electronics, an ASA-member shop in Batavia, Ohio. For more information on this topic, contact Bach at (515) 732-3965. His e-mail address is northstarguy@zoomtown.com
|
;){kind=link}
;){kind=link}
;){kind=link}
;){kind=link}
;){kind=link}
Copyright (c) 1996-2008. Automotive Service Association. All rights reserved.
