We can be misled if we look too deep.
When I first specialized in drivability and electronics in January 1986, I was working at a Ford dealership. Older techs who are familiar with that nameplate know that Ford didn't have datastream capability until 1988, and then it was only on Lincoln Continental. I had gone to Ford training and had been handed the huge book we called an H manual then (later referred to as a Powertrain and Emissions Control Diagnosis or PCED manual). For about two weeks, I tried to let that stupid book guide me, and finally I threw it on the shelf and started measuring things on my own.
I learned what the voltages and values were supposed to be by measuring live data on good cars (by gently back probing or building homemade breakout harnesses). When I became familiar enough with what the numbers were supposed to look like, I could flash out codes and then use a meter and a test light to troubleshoot just about anything on an EEC IV. The point is that I gathered good numbers on good cars and used those numbers as my own private frame of reference. I found out in the early stages of this operation that voltage is more reliable than ohms when troubleshooting sensor values, but it took a few years for Ford service engineers to recognize that.Scan tools and the datastream, however, give us information we can't measure directly with our meters and scopes. For example, we have no idea what the fuel trim numbers are unless the PCM reveals that information to us. Pre-J1930 regulation, engineers called long fuel trim numbers "Block Learn" and short fuel trim "Integrator," and they were based on the digital 0-255 eight-bit binary scale with 128 as the zero point.
As for closed loop fuel control, we usually can watch the way a healthy set of OWhen the datastream doesn't capture things rapidly enough, or there are some pieces of data that the scan tool can't provide, a multi-trace oscilloscope is the order of the day. But there are a lot of guys who don't own a scope or won't use the one they do own.
The PT Cruiser was a prime candidate for scoping.Maintenance Gone Bad
We put a timing belt on this PT Cruiser in 2006 when it had about 120,000 miles on the clock. With so many miles since we had replaced the belt, the owner decided it was time to have it done again. She took the curvy little ride to a local garage. When the job was done, the PT ran just fine for a few miles, but then began bucking and jerking to beat the band.
The problem was worse when the vehicle was hot. She returned the Cruiser to the shop, where the tech found a P1391 code, which indicates a loss of the cam and/or crank signals. The cam sensor was subsequently replaced (it draws its signal from the end of the intake camshaft), but to no avail. A trip to the local dealer produced a $1,000 estimate that included replacing the engine controller. While the owner rejected the dealership estimate, she actually bought a used engine controller from an online supplier and had the original garage install it. But the PT wouldn't even start with that engine controller plugged in.
Data Collection, Analysis
We noticed while I was holding the throttle in the service bay that the bucking and jerking would happen sometimes even in a no-load situation. The PCM had stored a P1391, along with a P0340 and a P0320, all related to cam and crank signals. We saw the new cam sensor the other shop had installed, but the crank sensor hadn't been touched. And we found that it was apparently leaking oil internally — the sensor's integral connector shell had a lot of oil in it. We replaced the sensor and cleaned its mating connector with electrical contact cleaner.We connected a timing light (to check for spark dropout) and an electronic fuel pressure gauge and drove the Cruiser again. For a while it seemed to run better, but eventually it bucked and jerked again; this time it stalled just as we were re-entering the service bay. The tachometer needle was gyrating wildly during these events, which sometimes happened in no-load situations. Oddly enough, even during no-starts, the fuel pressure was good and strong at 56 psi, the injectors were clicking and the timing light we had connected for our test drive showed that a good even spark was pulsing through the ignition cable to which the light was connected. After a very brief period, the PT eventually would start and sometimes ran just fine for a while afterward.
An Identifix search spoke of cam gear locator pin problems causing an index concern between the gear(s) and the shaft(s) and of chafing wires. Logic was taking hold now. The Cruiser ran well before the timing belt job, and I found myself wondering if there was indeed a problem related to the indexing of the gear on the intake camshaft. Because all of the elements were in place but possibly out of time at times, a scope was my next tool of choice.Connecting my OTC Solarity scope to the cam and crank signal wires, I got a peculiar looking pattern with odd, sloping horizontal lines like I hadn't seen before. But the pattern showed solid, consistent on-off signals on the cam and crank. Because, however, I had no frame of reference to determine what was normal, I contacted the Identifix hotline and spoke with one of the technicians who faxed me diagrams of what the normal cam and crank signals should look like.
Second-guessing the strange patterns I got with the Solarity, I used my Interro PDA on the second scope test and found nice crisp flat-topped patterns showing us (using the faxed diagram) that the cam and crank signals were indeed slightly out of normal sync. It appeared that the intake camshaft (the only one with a sensor) was running slightly behind where it should. To investigate this as closely as necessary, I had one of my tech students, Lawrence, remove the upper intake, the top part of the timing cover and the valve cover. We brought the shafts around so that the marks were lined up.Just for grins, I applied some gentle turning force back and forth to determine whether the gear and shaft were moving in relation to each other, as they might if the locator pins were missing or their channels were wallowed out. I could find nothing of interest during this odyssey, except that a couple of spark plug wire boots were wrapped with electrical tape where they entered their wells. Before the intake was reinstalled, the Cruiser got a new set of plug wires on general principle — the spark plugs were nearly new.
We were back to square one, and it was time to re-think everything. As I re-considered the intermittent nature of the concern (the Cruiser ran perfectly about 80 percent of the time), coupled with the absence of any movement between the intake camshaft and its gear, I began to move in a different direction with my diagnosis.
Back to the Basics — and Pay Dirt
I decided to approach the concern as if I was seeing it for the first time, with no knowledge of its repair history. We re-checked wire harness routing and condition, carefully following the harness to see if it was making its way across a sharp edge anywhere, but came up with nothing that way. Then, with the engine running and hot, I disturbed the wires at the cam sensor with my fingers and listened for bucking and skipping. I noticed the Cruiser had stalled once while we were connecting the scope to the cam sensor, but the PDA sometimes causes that kind of thing, because it seems to have a lower impedance than some vehicles will tolerate.
For example, using the PDA to check the crank sensor on an 1987 to 1990 Jeep will sometimes drag the crank signal low enough so that the engine won't start with the scope connected. Furthermore, we once checked the three accelerator pedal sensor pots on a 2005 Ford Five Hundred and got a good pattern on the screen while sweeping the pedal up and down, but the PCM wouldn't operate the throttle body motor until we disconnected the PDA.
With those two situations and other ones like them in mind, I chalked that stalling situation up to the PDA scope's mild impedance problem. I had experienced a similar anomaly when connecting to the crank sensor on the PT Cruiser. But after that, the engine started and I had retrieved the crisp green-on-black pattern that led me to check for cam gear/shaft movement.
Back to my wiggle test, jiggling the cam sensor wires didn't produce any results, so we raised the vehicle with the engine running (no scopes connected). With the engine good and hot, we found that gently pinching and twisting the wires at the Crank Sensor (CKP) connector caused skipping, bucking, and an engine stall. We lowered the Cruiser and restarted it to repeat that test, verifying in the end that the crank sensor connector was indeed the source of the concern.
Conclusion
Replacing the CKP pigtail, we tested the connector and the on-road performance of Cruiser with no repeat of the bucking, skipping and stalling, and there were no check engine lights nor trouble codes. Apparently the engine oil that had wicked its way through the sensor's innards had made its way up into the weatherproof connector to compromise the integrity of one or more terminal crimps enough to cause the crank signal to drop out. It's odd that we never saw that happen on the oscilloscope, even when the bucking and skipping were rampantly happening in the service bay.
As for the slightly out of sync cam and crank signals, I chalk that up to product variability in the field, I guess. It's not worth chasing if the Cruiser runs right.
I'm a great believer in analyzing scope patterns, voltages and PID data, but in this case, an old-fashioned wiggle test yielded the answer. Maybe I should have done that first instead of second-guessing the timing belt job, which had nothing to do with this concern. Lesson learned.
Richard McCuistian is an ASE-certified Master Auto Technician and was a professional mechanic for more than 25 years. Richard is now an auto mechanics instructor at LBW Community College/MacArthur Campus in Opp, Ala. E-mail him at [email protected].
