Intermittent lean issue, causes stalling / unstable idle *Update: Bad ECU?* Options

[rjames]

Forgive me for starting a new thread that is really just a continuation of a thread I started last year for this same issue. I've tried a couple of new things since then with no progress and figured I'd start a new thread to consolidate everything I've tried so far on one post. (I also put a link at the end of that thread to this one).

Symptoms:
Car will intermittently go into a very lean state causing the power to fluctuate, RPMs to fluctuate at idle and often stall if not opening the throttle a fair amount. Usually happens when engine is warm, but will go away mysteriously on it's own- usually after over enriching the air/fuel mixture and running in that way for several minutes.

I’ve determined it’s going lean because when this happens unplugging the TS1 / Air temp sensor and the surging / staling will cease. (Unplugging the TS1 according to pbanders site enriches the fuel mixture)
Or I can dial the ECU or MPS for a richer mixture which will also solve the issue to certain degrees, but when the issue mysteriously goes away the fuel mixture is too rich.

Symptoms are similar to when an MPS diaphragm fails. In fact, the car was running fine up until about 3 years ago when the MPS failed. I bought a rebuilt / recalibrated one from Jeff B. The car wouldn’t start with the rebuilt MPS because it was too lean (or rich, I can’t remember). I adjusted it, it started, I adjusted the AFR using a wide band sensor and ever since it’s had this intermittent issue.

Things I’ve done to try and solve the issue:
-Ensured correct matching ECU, MPS and CHT sensor.
-Have tried a correct # spare ECU
-New CHT
-Verified CHT is in spec both when engine is cold and warm. (I checked the CHT when the lean condition happened and it measured 78 ohms).
-Installed 123 distributor (issue also happens on stock distributor). I’m running retard on distributor, but have tried without.
-Installed new coil with the 123 distributor. Issue happens with old (stock) coil, too.
-Replaced all vacuum lines and verified they are hooked up correctly.
-Cleaned and flow tested injectors
-New injector seals
-Installed fuel pressure gauge in engine compartment- ~28psi. Reading is constant even when surging/stalling happens.
-New intake runner seals
-New intake runner gaskets
-New throttle body gasket
-Swapped throttle body with NOS model.
-Verified no vacuum leaks with smoke test.
-Verified AAR is opening/closing correctly.
-Verified the decel valve is opening/closing correctly.
-New injection harness
-New ignition harness
-Swapped relay board and verified continuity on all connections.
-Tried a spare rebuilt MPS that hold vacuum
-Manifold vacuum is constant at 15-17 HG
-Smoke test didn't reveal any leaks.
-Verified decal value is functioning correctly
-Monitoring the AFR using wide band sensor.
-Calibrated throttle position sensor. Tracks look good. (Unplugging the TPS has no effect when issue is happening)
-Verified grounds from the FI harness, under the relay board and ground strap on tranny is good.
-New odyssey battery (verified voltage is good)
-Verified timing and valves are adjusted to spec.

Misc info:
-Setting the idle above 13:1 at idle or thereabouts puts the idle speed at 1100rpm, and it will hunt unless you richen up the mixture, either with the ECU knob.
-Throttle body idle screw is all the way in to keep the idle at ~ 1000rpm. Verified that air gets past the idle screw from above the butterfly. Apparently not an uncommon issue according to others. Others report having a hard time getting the idle below 1000 with the 123 ignition.
-Compression is low on cylinders 1 & 2: 110, On cyl 3 & 4: 131
-Open AAR valve doesn't increase engine RPM when engine is cold. Verified AAR is opening. However, plugging the hose that runs from the AAR to the airbox will stall the engine until it's warmed up. (AAR closes as it should then)

[Superhawk996]

Pretty comprehensive list. (IMG:style_emoticons/default/biggrin.gif)

I know you said you smoke tested it - How sure are you? Have you been able to smoke test it when the issue is ocurring?
Didn't Zach just find a loose throttle shaft causing similar intermittent issues?

When you say it goes very lean, what is the AFR when that is occuring? Does that give any insight into how much excess air is entering somwhere?

Also, since you say disconnecting the ambient temp sensor TS1 helps, are you monitoring the resistance of that TS1 temp sensor when it is occuring? Could that sensor itself be going intermittent?

[rjames]

I know you said you smoke tested it - How sure are you? Have you been able to smoke test it when the issue is ocurring?
Didn't Zach just find a loose throttle shaft causing similar intermittent issues?

Thanks for the response. I'm not sure of anything anymore. (IMG:style_emoticons/default/biggrin.gif)
I was thinking about smoke testing it again. That's probably worth doing again, although wouldn't a vacuum leak with DJet result in higher rpms at idle and not a surging/stalling condition?

When you say it goes very lean, what is the AFR when that is occuring? Does that give any insight into how much excess air is entering somewhere?

When it goes lean, it fluctuates from lean to rich pretty quickly- the idle hunts from ~1100 to almost 0 (lean), and then will stall after several oscillations if I don't richen the mixture, similar to what happens if you turned the ECU knob all the ways counter clockwise. AFR gauge tries to keep up but goes up and down with the idle surge and goes past 16:1 when the RPM drop, and after several cycles will stall. I can keep it running if I keep the throttle open, but power is way down and I can still feel it oscillating.
Then there are times where I'll be driving along fine at low or high speeds with no hint that anything is wrong-AFR even looks good, but if I put it into neutral the car will stall. This is also intermittent, but when it does happen it's hard to start without opening the throttle all of the way, and then the idle oscillation behavior happens.

Also, since you say disconnecting the ambient temp sensor TS1 helps, are you monitoring the resistance of that TS1 temp sensor when it is occuring? Could that sensor itself be going intermittent?

I don't think so, because when I unplug it and then adjust the AFR accordingly, the lean issue/hunting idle/low power/stalling still happens.

I was really hoping it was just a simple fuel delivery issue, but the gauge reads a constant 28 psi at all times.
The characteristics really seem like a bad MPS (I have 2 that hold vacuum that I've already tried swapping) or a bad CHT. My CHT reading was 70 ohms last time this happened. Is that too low? Pbanders site says it should be less than 100 ohms with a hot engine but doesn't cite specific numbers. I've tried swapping the CHT out too though. :/

[Superhawk996]

I know you said you smoke tested it - How sure are you? Have you been able to smoke test it when the issue is ocurring?
Didn't Zach just find a loose throttle shaft causing similar intermittent issues?

Thanks for the response. I'm not sure of anything anymore. (IMG:style_emoticons/default/biggrin.gif)
I was thinking about smoke testing it again. That's probably worth doing again, although wouldn't a vacuum leak with DJet result in higher rpms at idle and not a surging/stalling condition?

Yeah, agree that minor leak woud be higher rpms (i.e. not able to pull higher manifold vacuum at idle). I was wondering if it could be some sort of much larger leak that was getting in and not being measured by MPS and leaning things out.

Thinking of large leak paths. It almost sounds like the decel valve might be oscillating. When you say decel valve is OK - is that measured against vacuum levels and at what vacuum it opens, or was that simply that it opens and closes with vacuum? Set point needs to be above idle condition and only opening the valve when engine is over-running with high vacuum that you won't see at idle. On over-run, modern cars will do a complete (or nearly complete) fuel shutoff for fuel economy. Not uncommon to see AFR of 17:1 or even leaner during that decel fuel shut-off conditon as fuel is feathered in & out at the DFSO transitions. Wondering if your 16:1 condition and hunting is indicating a major air bypass that is oscillating due to improper set point?

If it isn't going lean due to air bypass then the only other option is that fuel is being reduced to cause the lean condition like 16:1. Since you're saying fuel pressure doesn't vary, it seems that it is more likely to be going lean due to the airflow path. Maybe a quick check would be to rig an LED to the injectors so that you can verify that when it is oscillating that you don't have injector pulses cuting out. Ideally, I'd use an oscilloscope but betting you don't have one and/or that you won't have it when the intermittent condition occurs on the road!