Heya, been a while. Truck is a 1991 D21 5speed Base with the KA24E.

The problem is one I've been fighting the last week, and I only have a few more days with it before I have to head back to the other side of the country. This adventure started with an extremely rough idle and running very rich. Before all this started, it had a recent Bosch O2, cap and rotor, plug wires, plugs, and the injectors all replaced. Was running like a top, passed California SMOG as clean as can be, all the numbers were well below the stated averages on the SMOG sheet. Then this last few months it developed a horrible idle and extremely rich mixture. The MAF was replaced with a parts store unit, and no change. Still idles horrible, stumbles when driving, down on power, and will stall after idling for about a minute or two. Throttle body gasket was replaced along with IAC valve based on it throwing codes for that a year or two prior, no change.

Then I came back home last week and started working on it. New Bosch O2, new NGK plugs, the cap looked like it was the Mitsubishi style so I went ahead and put the proper Hitachi style on, found the plug wires were pushed up into the boots on the distributor side so pulled those out and made sure they were in correctly. New timing set, everything set per the 1994 FSM because that's what I have. New fuel pump and filter, tank was clean as can be inside and filter wasn't very bad at all. Still idled bad and had a stumble/miss at cruise, had a very slight stumble on acceleration and feels okay on power so at least some issues were resolved. It seems to idle very good while cold, but once it's off fast idle after about 20 seconds of running it gets very erratic and fluctuates between about 1200 to all the way down in the 200-300 range, choking and sputtering the whole time, and after about a minute will stall out. If I give it gas, it comes back up to around 1200 for about a second then continues it's erratic idle.

After putting it back together and finding the main issue not fixed, I pulled the MAF and sprayed it with MAF cleaner, checked connection for corrosion, and reinstalled with no change. Every vacuum line I can see shows no leaks and appears to be routed properly in accordance with the hood sticker and the FSM, and spraying carb cleaner around possible leak areas produces no changes. EGR valve was replaced as it was looking pretty bad inside. All ground connections I can see in the engine bay were cleaned and got dielectric grease applied. No codes were ever thrown during all of this besides coolant temp sensor when purging air from the system after the timing set install. Both sensors were replaced with spares anyway just because I had them already. Did the posted fix on the two crimps in the engine harness, cut and soldered both. Injectors were not replaced by me, but everything looks how it should and the harness is in the right spots. I rechecked the timing marks on crank and cam sprockets, and oil pump and shaft, and the distributor is in the correct placement. All plug wires in correct spots. Almost every connector in the engine bay was removed, treated for corrosion as required, and dielectric grease applied. Added another ground from fender to the block on the same bolt the battery is grounded to.

Now here's where things got me stumped. I ran through the diagnostics flowchart for the MAF, the harness looks good on all three pins and they all test out fine per the FSM flowchart. I decided to add on a second ground because I saw that was a thing on the NPORA forums, and I didn't think it could possibly hurt anything. Well, the ECM seems to not like that, because it put the truck into limp mode and threw the MAF code almost immediately. I checked to make sure I didn't ground out the power wire, best I can tell the solid black is ground, white is signal, and black/white stripe is power in, so that's all good. But when it threw the code and went into limp mode, it idles like a dream, no hunting and almost no misfire at all, just very rich like it's supposed to be. Has a slight stumble on throttle with no load but that's kinda expected when in limp mode. So I decided to do some testing, removed the ground I added, reset the codes, and it's back to running like s***. Touch the second ground to the battery, and it goes right into limp mode with a MAF code and idles fine again. So I'm just confused as to what's going on now, it running fine in limp tells me that mechanically everything is correct, it's just something with the electrical feedback portion of the system that is wrong. I have another parts store MAF on the way as there are no KA trucks with one near Temecula, so that's going in tomorrow morning. I just wanted to get this out there in case someone has an idea on this. I am also going to open the engine harness again and redo the two crimp connections I already redid, just in case I messed that up.

Sorry for the text wall, but too much information is better than none I guess!

Have you measured ground voltage while the engine is running? (Multimeter negative probe to battery negative post and positive probe to engine ground, body ground, chassis ground, etc.) A very low mV reading would be normal.

I had another late night thought. You checked for vacuum leaks and found none but if there was a leak at the exhaust, like upstream of the O2 sensor or possibly an EGR tube from the exhaust pipe , that too would affect the O2 sensor with air intake that it not metered by the MAF. Since the ECM ignores the O2 sensor signals until it heats up, at about the same time your fast idle turns off, it is possible you could have a leaking exhaust flange. At idle, this type of leak draws more vacuum between exhaust pulses that at higher RPMs. The pulse width between exhaust strokes is more compressed at higher RPMs so, as you described, it is less of an issue when the engine is revved higher. If you have a scan tool that can check live data from the O2 sensor, you can test this by observing the O2 sensor response if carb cleaner is sucked up when you spray it at potential areas where it could be leaking, like at the exhaust manifold to head or EGR tubes. OK. I can sleep now. Not that this has kept me up but I do hope you can figure this out and put it to rest.