This problem PT was originally assigned to a fellow tech in the shop I call home. The problem was an intermittent dimming of all the electricals...visible in the lighting, and you could hear it in the blower motor. The problem was only noticeable as the electrical loads were increased, running everything that you could run. There was no pattern to the dimming either. It would come and go as it wished.
The tech checked the charging system, and noticed that system voltage would bounce from normal to battery level, and amperage would fluctuate from normal to obvious discharge. Thinking it was a problem with voltage regulation, the tech decided to condemn the alternator and ordered a replacement. Only one problem with that assumption, though...the regulation is not internal to the alternator on this system. It is contained in the ECM. The new alternator didn't fix anything, so he asked me for my help.
I don't pretend to be any sort of "super" tech, but I do welcome the opportunity to learn and figured it would be worth giving it a shot. I saw the dimming effect, and he showed me the readings he got with the charging system tester and his DMM. I wanted to see if there was any coorelation between these events that might help point us to the problem, so I decided to use the scope. Here is the first capture, covering about 25 seconds of system operation:
Blue is charging voltage measured directly at the battery, gold is current measured with a high amp clamp at the battery positive cable, and red is voltage output to the alternator field coil measured directly at the ECM. That was the first lesson. My service information said that the field was ground controlled, but the pattern I got was that of a high side control. A little digging, and we found out that the PT uses three different variations of charging system control.
Earlier models did use a ground side control to the field coil in the alternator, with power supplied from a wire spliced into the ASD relay output. In this model, the control was switched to the power side of the field, with ground being run to a body ground in the harness. Later, the ground side was run all the way through the ECM. That made me wonder if there had been issues with the system that wasn't listed in the TSBs. Time to zoom in and get some more clues:
When I first looked at this capture, I wondered about the condition of the new alternator and the condition of the ECM driver to the field. There were signal dropouts all over. With these first few thoughts, I also knew that the circuits had better be clean in order for the ECM to do its job. Time to do some voltage drop tests. With the engine running, and the loads on, I checked the voltage drops on the alternator field coil ground, and the alternator case ground. Power to the field looked OK, but I did measure a 0.50 volt drop between the B+ alternator terminal and the positive battery post. Power and grounds to the ECM all seemed fine. Now I was really leaning to a problem with the ECM...
Let's take a closer look at just voltage:
This is a minute and a half of information on one screen. You can clearly see the drop outs in voltage between what was measured at the alternator B+ terminal (red), the battery (blue) and at the ECM field pin (green). I could actually manipulate the screen to bring the three patterns together and they appeared to match perfectly. I am really leaning toward the ECM now. After all, if the field is changing the alternator output is going to change. I thought I had confirmed the integrity of the power and grounds to the ECM, as well as the input signal from the battery temperature sensor. If the inputs were right, and the power/grounds were OK, but the outputs looked weird, then it should be the ECM, right? Let's look closer...
I just wasn't comfortable ordering a new ECM quite yet...something just didn't feel right. Then it struck me...voltage drop testing is valid using the method I've shared here (and used on this PT) IF the system voltage is steady. That wasn't the case in this charging system...DUH! In fact, that was the issue...varying voltage. By looking closely at the capture, I found several cases where the voltage at the ECM field pin was almost 1.0 volts less than what was at the battery. Some readings were below battery open circuit voltage! There had to be a problem in the path between the positive battery post and the ECM.
Breaking out the DMM, I started checking the positive side of the circuit the old fashioned way...by placing one meter lead at the battery positive post and the other at the ECM pin. Sure enough, voltage drop was bouncing around with peaks just over 1.0 volts. Backtracking the circuit to the battery cable terminal at the junction box narrowed the problem down to the cable itself. There is no real difference in the readings performing voltage drop tests this way, it's just more confusing to many technicians who don't understand that the meter is measuring the potential difference between the leads. In method "a", you have to do the math and in method "b", the meter does the math for you. Had I done this the first time, I would have seen the variations in the DMM and wouldn't have even needed the scope. A replacement cable corrected the drop, and restored the system to normal.
I almost replaced the ECM for no reason. My mistake? Focusing on the trees instead of the forest, and getting tunnel vision. Good thing I caught myself!