CrazyHawaiian

I have been troubleshooting a stalling/rough idle problem over the past few weeks and while troubleshooting I noticed something very strange. Car is an '89 w/ L98 so MAF TPI and a 165 ECM. I've got the car to the point where it will idle rough with all sensors connected (no more stalling out) but when I put the ECM into field service mode to check open / closed loop status, strange things happen.

1. Start the car up and it is idling rough between 1,200-1,500 RPM. Insert the paperclip into the A and B terminals of the ALDL to enter field service mode. Right away the idle smooths out and remains stable at 2,000 RPM. Idle is suddenly much better, just kind of high. The SES light flashes about once a second indicating it is running in closed loop.

2. Remove the paperclip to exit field service mode. Right away the idle goes back to being rough between 1,200-1,500 RPM. Not expecting this, insert the paperclip back into A and B and observe the SES light. Right as I insert the clip the SES light flashes fast for a split second but then things change. Just as before the idle smooths out and remains stable at 2,000 RPM and the SES light then blinks slow about once a second indicating closed loop.

3. Wondering if I was seeing things, I did this again, and again it idles rough with no clip inserted, put the clip in, SES flashes fast for a split second, then something somewhere changes and idle smooths out at 2,000 RPM and SES flashes change to slow indicating closed loop.

Now this makes no sense to me because it appears to idle better in closed loop than it does in what I'm assuming is open loop. I'm assuming the fast flashes (maybe 2 or 3 flashes in a 1/2 second) right as I insert the clip indicate it is idling rough in open loop when no clip is inserted. Sound like a safe assumption?

If it is true, there are two things that are confusing me. Number one the loop status appears to changing based on putting the ECM in field service mode (it will reliably smooth out the idle when clip is inserted). Number two it appears to idle smoother in closed loop than it does in open loop. I've done a search and came up with one very close (almost exact) problem (located here) in which Vader says the field service mode should be a reading only, it should not be forcing the ECM into either loop mode. I know Vader knows his stuff, but this is so strange I have to ask again. Has anyone out there ever seen anything similar to what I am experiencing?

I made sure to never start/crank the car with the A and B terminals grounded, but I'm seriously wondering if the ECM is bad anyway ...