I6 4.0 1991 bucks/surges on med/hard acceleration

[McQue]

Don't know the answers to your questions, but just thinking out loud here...Here is a theory for you to ponder:
Assume good MAP and MAP wire circuit and ECM.
My 01 FSM states that the "Map sensor input is the number one contributor to fuel injector pulsewidth". So I also assume if you disco the MAP and things work ok, the ECM must be determining fuel injection pulsewidth as if it is in open loop mode, or at least determining injector pulse width without the benefit of input from the MAP, or it thinks the MAP is saying "all or nothing" with respect to manifold vacumn...I would guess it is saying to the ECM, I sense no manifold vacumn.
Having said all that, when the Map is connected it is providing good input to the ECM for injector pulsewidth change, but your motor doesn't like the results for some reason.

Makes me want to really understand how your entire fuel delivery system is working.

 

[blistovmhz]

Me too. But it doesn't look like anyone has any numbers everything.
Gathering information from multiple sources.
A lot of it is wrong, but I think enough of it is right, that I'm in the right direction.
I'll check motor mounts as well, before rewiring.

 

[Ecomike]

Damaged flex plate might explain the forward - reverse, or something in the engine to tranny attachment, connection is loose, loose hardware.

The ECU uses a default table (Open Loop) for fuel injection and ignition timing if the O2 sensor or MAP signal is lost.

Have you run a a full vacuum gauge diagnostic test on the engine???? I would do that next, with the MAP disconnected, then again with it connected. If the Vacuum signal is screwy with the MAp sensor disconected, you have an engine problem, hardware. If the vacuum readings are all good with the MAP disconnected I would look for a bad MAP to ECU wire.

A bad wire or connector from the MAP sensor to the ECU that is degrading the MAP sensor signal to the ECU when something moves just the right way, like durring aceleration when the engine moves some. Rotation, angular momentum, is in the opposite direction in reverse gear during acceleration which would add to bad wire theory probablity as it may need to be stressed in a certain rotational direction to cause the signal degradation.

 

[blistovmhz]

I think I got it!!! WOOPWOOP!
I was testing my backwards driving and confirmed the problem only happens when going forward. Got my roomate to brake stand for me while I watched the engine, sure enough, it lifts counter clockwise in forward, and clockwise in reverse.
I started twisting the wire harness right where all the intake sensors tie into the main harness on the firewall, and the engine stalled. With the engine off, I twisted the harness wires again, and heard clicking from a few of the injectors, and the fuel pump breaker.
We found both the ground to all the sensors, was a bit loose, and eventually, I got the bright idea to disconnect the MAP and test some more.
With MAP unplugged, no problems. Traced the violet/something wire into the harness, and it was slightly broken where it ties into the main wire in the harness, and leading back to something else. Tore it apart, spliced back together, drove and brake standed (stood?) all up in here, no problems.
I got a little more power, and the exhaust doesn't smell sour anymore.

I still need to solder everything back together, but it so far, looks like this was the only problem. :woohoo:
I'll do some more testing once its all soldered back together.

This is sorta funny cause I'm always dissing the local mechanics, who claimed to have hooked the ecm up to a scanner, and saw no problems reported. I'll bet if they'd ACTUALLY hooked it up to a scanner, they'd have noticed that every sensor in the engine, was turning off three times a second.
Fuckarz.

Thanks for all the help guys. I hope I'm not getting ahead of myself and getting my hopes up.


Thus far, I've spent a bit over $1200 on parts and tools to track this problem.
If I'd had it on a scanner, (and watched the mechanics actually scan the ECM) I'll bet this would've cost $60 for a scan, and about $0.04 for the bit of wire I had to replace.

*EDIT*
I think the reason I didn't see this problem below -40C was because the motor mounts are stiff as hell at that temp, and the engine doesn't flex near as much. Also explains the problem recurring when I hit bumps.

 

[McQue]

That is good news! I know you have been working hard on this one!
Good luck with it, and keep you fingers crossed, but it sounds like you fixed it.

 

[dfete91]

Is your speedo working? My speed sensor went out and did the same thing. It adjust fuel trim or something. If it is good run some B12 through it and make it drive.

 

[blistovmhz]

Haha, I forgot about my posts till your reply.
I did solve the problem.
I finally figured that the cause of the problem, directly, was the engine torsion. (ie, engine torquing to one side.
When in forward gear, while doing a brake stand, the engine torques counter clockwise (if you're facing the front of the vehicle) and clockwise in reverse.
The problem was only apparent when in forward gear, so i started twisting all the wires in the harness, counter clockwise, and sure enough, the engine stalled when I twisted right near the MAP sensor.
I pulled apart the harness, and found a single strand of one of the 5V sensor leads, was poking into a the MAP signal wire.

So, in summation, what 4 Jeep shops couldn't even diagnose for $5000, I fixed for about 3 cents of wire, and 5 cents of solder.

Tada!

Thanks for all the help guys.
Jeepy Jeep runs like a Jeep now.