87 Grand National driving me crazy

1987 Buick Grand National (3.8L Turbo) with an intermittent misfire. It is my dad’s car, and it has mostly sat for the last 20 years, but has had the problem intermittently since he regularly drove it 20 years ago. The problem varies in severity. Sometimes it is bucking/surging at mid throttle, and occasionally it runs so poorly that it can barely keep up with traffic. When it is really running terribly, it is spewing black smoke out the back, exhaust backfire, just barely running, but never stalling. Problem tends to be primarily when the engine is hot. No codes other than one that I caused while troubleshooting.

 I finally caught it when it was running very poorly, and I found what was happening with my scope. Primary ignition current waveform was unremarkable, but current on the fuel injector circuit has an occasional problem. It has no current when it should be firing one injector, and then after five other injector ramps, it fires with double the usual current (but with a smooth ramp). How often this happens varies at random, and in between malfunctions, the injector current looks fine. Suspecting an injector problem, I backprobed the #2 injector control wire while measuring the current waveform to figure out which injector was acting up, and found that the spacing in between the #2 injector firing is uneven when it malfunctions. The #2 cylinder fires one cylinder too late, then #2 fires four cylinders later to get back on track. So the ecm is falling out of sync, and firing injectors at the wrong time, which would explain why it is running so poorly. (Yellow line - #2 injector control wire, red line: injector circuit current)

At this point, I suspected a problem with the cam sensor or cam sensor wiring. On this vehicle, the ignition control module supplies power to the cam sensor, the cam sensor signal goes through the ICM (magnavox type 1 bypass ignition) and to the ECM. See below for the factory wiring diagram (sorry for the hard to read scan).

I measured the cam and crank signals on my scope, and didn’t see anything unusual about the shape. The cam and crank signal were both about 0-5V in amplitude. I measured the cam signal along with the injector current, and found that the missed injector and the double amplitude injector corresponded to the cam signal, but the cam signal doesn't drop out or anything. (Red line: cam signal voltage, yellow line: injector current) 

I double checked the cam sensor signal at the ECM and got the same thing. I checked the factory service manual to see what level the cam signal should be. The factory service manual said that the ICM supplies voltage to the cam sensor on the signal wire, and with the cam sensor unplugged, you should see 6-9V on the signal pin. I see 5.01V on the signal pin. The FSM also says that with the cam sensor and the ECM unplugged, key on engine off, the cam sensor signal wire at the ECM should read >6V. I measured something around 0.1V. I verified wire integrity from the ignition control module and ECM, verified power and ground to the ICM, and made the call on a bad ICM. I installed the new ICM, and checked the voltage on the signal wire of the unplugged cam sensor, and it read 5.06V (vs 5.01 before, and 6-9V that the factory service manual says). I measured the cam signal wire key on engine off, cam sensor unplugged at the ECM at around 0.1V (about the same as before, vs >6V claimed by the factory service manual). With everything connected, it fired up and ran fine for the 15 seconds I left it idling before walking away in frustration. I didn't spend hours trying to reproduce the intermittent problem, but the measurement that lead me to call the ICM hasn't changed, so I'm sure it isn't fixed.

 At this point, I’m totally out of ideas other than loading up the parts cannon, or maybe parking it for another 20 years. I have no clue whether the cam sensor measurement is normal and the FSM is wrong, if there’s something else I missed, or what. I’m pretty bummed after spending 2 days troubleshooting an elusive problem only to make the wrong call. Any advice would be appreciated, especially if anyone has experience with these older systems.
 

Have any scan tool capabilities of reading any codes from an 87? When it’ has black smoke from the tail pipe I’d be tempted to look at the block integrator.

Yeah, the BLM is low when it is running terribly (90-100ish IIRC), but it's also a little low when it is running well (I can't remember the numbers off the top of my head, though). The engine is not stock, with higher flow injectors, a bigger turbo, and a non-stock eprom, so I would expect that it would tend to be taking away injector time even if running properly. I'm pretty sure that the black smoke and backfiring is because it's firing the injectors at the wrong time, though, rather than a normal AFR problem.

Just wanted to bump this up with a more specific question: does anyone know what the crank/camshaft position sensor reference voltage should be on a gm type 1 ignition ? FSM says it needs to be 6-9V, original ICM puts out 5.01V, and new duralast ICM puts out 5.06V. Not sure of I'm on the right track trying to troubleshoot low reference voltage or chasing a red herring.