sudden rich condition, code 11, fixed by EFI reset

[Victor Charlie]

I have narrowed down a problem and not sure what would be most productive to do next. I had multiple problems originally, so I'm starting this thread now that some easy ones were weeded out.

Problem: Car starts great, runs a variable amount of time - seconds to 1/2 hr. then suddenly, instantly, switches and goes rich by the smell of the exhaust, barely idles, sputters. On the road this means it stalls when the clutch is held in, won't restart easily. This is rectified temporarily by pulling the EFI fuse for a minute and resetting it. I always starts normally if the car sits overnight untouched, but usually not if it sits just 1 hr.

I think it is missing spark frequently, and that's why it is rich and sputtering, but I don't want to prejudge it. When I pulled the first plug I had spark, but wasn't clear if I had it regularly, and then I found and think I fixed that. Had trouble with aftermarket spark wires and coil pack, not picking up spark signal with timing light. That is much better since yesterday after spreading the connectors and reseating, but not perfect. i had to wiggle one to get the timing light to track this morning, then it behaved, but, strangely, the engine at that moment was running well and didn't change when I got the timing light signal back.

voltage tests at the ECU per TSRM were all normal, once TPS and idle were set, except the ac signal was odd. Setting of the AC switch on/off doesn't affect the other problems or throw a code 51, so I'm putting that down on the priority list.

codes:

code 11- on and off, I think likely related to the problem, though it doesn't perfectly match when the engine isn't running. There seemed to be some diming in the check engine light this morning for a moment after it started, as it was flashing the code 11 at me. I don't see obvious loose connectors anywhere: ECU. fuel pump relay, CPS, TPS, coil pack, ignitor. Wiggling these doesn't affect the poor running.

code 51 just when I set the idle screw to allow it to run, backed it out and code cleared, don't think it is the problem.

I haven't got a wideband yet, Tach not working yet b/c it is an NA-t swap. I feel fuel return, seems to have pressure.

Questions, unless someone just has it all figured out for me:

1) So how does the code 11 work? It isn't set by pulling the EFI fuse, obviously, since that resets the system. Can it be set intentionally by intermittent loss of voltage in the EFI circuit, like putting the fuse back too quickly?

2) Is there any part failure in the ECU, ignition circuit, or elsewhere, that can kick in randomly and throw a code 11, without stopping the engine, and then clear overnight?

Is the coil pack likely enough to be a problem that I should replace it? could it have anything to do with code 11 ( don't know how, but thought I'd ask).

If this isn't enough to point to an obvious answer or priority list for trouble shooting, I will be working on it and reposting as I get more info. Thanks.

 

[Victor Charlie]

Sorry for the long post. Looking for help prioritizing the troubleshooting process, don't expect anyone to figure it out for me.

 

[Victor Charlie]

Funny how things seem clearer on a sunny morning. After browsing around, I realized that this is kind of a dumb question until I check all the grounds. Sorry, I'm doing it and will post when I have results.

 

[Victor Charlie]

Okay I did the ignitor upgrade, cleaned the grounds from the battery to body, head to body. coilpack to coilpack ground wire to head, checked plug wire order and put rubber inserts in the hollow of the plug wire at the coil side to make even better contact with the coil packs. connectors all look tight. Haven't pulled out the ECU itself yet.

so far: no difference in behavior, still starts okay, slows a lot and seems to misfire a lot after maybe a minute. A lot of white smoke smelling like unburned gas at first when it has been sputtering before it was turned off. It threw a code 51 again, so I am checking the TPS now, but I think that started more recently as I was playing with the throttle set screw. No code 11 so far.

 

[Victor Charlie]

I'm just thinking out loud, I guess. I also found an extra plug, matches the EGR Temp sensor, which TEWD shows as CA only. Since the engine is JDM with EGR parts taken from an apparently non-california '87 GTE, I'm going to assume this doesn't matter.

Checked TEWD. I have fixed all the grounds in the TCCS circuit except the fuse box mounting screw on the left fender. Will clean this up tonight.

The RB #4 has a ground, but it hasn't been disturbed in the swap. It is an NA body, so I have to check that it is compatible with the 7m-GTE.

 

[shaeff]

Hey Victor,

I'm fairly certain that your code 11 will mask any other codes until you fix that. It's a power interruption to the ECU. The ECU is grounded on the bottom right of the intake manifold near the firewall. There are two eyelet connectors that should be free of corrosion, and securely fastened to the manifold. If those are loose, you'll get all sorts of little gremlins. Check them.

Next, see if you can get the plug wires properly connected to the plugs and coilpacks. If they're not seated properly, that'll cause you some issues as well. NGK or OEM wires fit well. MSD's tend to be kindof a pain to get to fit, so I'd stay away from them (speaking from experience)

You likely didn't see a code 51 due to the fact that code 11 will mask it, as that's much more of a problem. Your main priority should be to find and fix the cause of the code 11.

You shouldn't be messing with the idle screw, there's a specific procedure to follow (that I cannot recall right now) to adjust idle, and you're not doing it all, so leave that be for now.

There shouldn't be pressure in the fuel return. Are you running an upgraded fuel pump? Adjustable fuel pressure regulator? If so, have you drilled the j-tube out? (see here: http://www.supramania.com/forums/showthread.php?t=34548&highlight=j-tube )

You should grab the tach circuit board from another member and slap it in your cluster. It's bolt-in, and will work great. It's always nice to have a tach!

It's not likely that the coilpack is your problem, but you could find someone local and swap quickly to find out. It only takes a few minutes.

Hope that helps!

Edit: check out this thread, specifically, post #9: http://www.supramania.com/forums/showthread.php?t=64492&highlight=ECU+grounds

 

[Crazybobmundo]

I also found an extra plug, matches the EGR Temp sensor, which TEWD shows as CA only.

Do you have a picture of this plug? I have a random plug near the EGR that doesnt go to anything.
My engine was "JDM" before it was rebuilt. It had the factory EGR block off plates. So it kind of doesnt make sense if I have a CA harness and a "JDM" engine.


I read your post on my thread. You have similar problems but some are different. The different was that I wasnt getting spark at all.

I did get code 51, code 11 and code 14. I cleared them all, I have no clue how but what I did was disassemble the fuse block and swaped it with a 91 block (I had the codes still for a few days then magically they were gone and got spark). Changed the EFI relay and both codes 11 and 14 went away after I put everything together and grounded the grounds.

I also swaped the MAIN Relay underneath the dash on the drivers kick panel.

Now my car starts, revs itself, dies. I try it with the pedal depressed and as if my throttle wasnt hooked up, nothing happens and it is rigged. Im pretty much running in circles and I bet you can relate too.

If I ever figure out my problem Ill let you know and it'll hopefully help you.

 

[figgie]

Victor

the issue might be that the CPS signal (yes I know, code 11 is not that ) might be shorting the signal wires at the CPS connector itself. My car used to exhibit the same issue, it ran fine and then it sputtered, hiccuped etc. Felt like one of the cylinders was miss firing. Well it was not one. It was two due to the waste fire setup. The 7m TCCS does not have misfire detection of that level so it will continue to inject fuel when there is an ignition miss fire.

Quadruple check the CPS connector for any bare wires especially the CPS side (though check the harness side also). You might be grounding G1 or G2 to G- which will cause a miss fire on 1&6.

 

[Victor Charlie]

Victor

the issue might be that the CPS signal (yes I know, code 11 is not that ) might be shorting the signal wires at the CPS connector itself.

This is a great idea! I didn't know that the CPS could specifically affect one coil pack. I may have to break out the scope, but it is a serious, high end oscilloscope for electronics bench work, not car repair, so I'm a bit nervous to take an expensive, pristine piece of equipment into the garage. I also haven't used it for years, so there would be a learning curve.

I certainly will check that the wires are not shorted to ground or each other. I did do some harness work on those wires when I rehabilitated my harness, so it is entirely possible, and it is close to where I'm messing with the plug wires, so I might be mistaken about what is affecting my timing signal, which is very fussy.

I did work on the MSD wires to get a firm connection by taking a small silicone hose, cutting a short piece, cutting a slit lengthwise, and then spreading the connectors and filling the space with my piece of hose to keep them from closing all the way when reinsert the wires. I believe this has been effective in making solid contact.

 

[Victor Charlie]

Thanks for the flood of great help everyone. To make this more readable this is my summary:

Code 11 is no longer showing after some fussing, and may have been unrelated.

Still could use help:

1) finding the great procedure for callibrating the TPS.

I will now check/fix in this order:

1) ground for fuse block
2) ground under intake manifold for ECU, add redundant ground wire
3) check integrity of wires from CPS. Gap and resistance already tested
4) adjust and test TPS to be sure code 51 is cleared again
5) recheck timing after code 51 is cleared.
6) swap coil packs.
7) check main relay operation with DVM.
8) swap AFM
9) check EFI relay operation.

Give me a day or two to try these things. Thanks. :icon_bigg

 

[figgie]

Victor

good plan of attack.

The O-scope, forgo that for now. Do the visual check first and foremost before slapping the o-scope to that circuit.

Once the visual is done

http://www.autoshop101.com/forms/h23.pdf

page 11/16

unfortunatley it does not reveal voltages but the system is a low level signal. If anything start at 12v and then step down to 8v and if needed, 5v. time on the other hand, the 24 tooth wheel will do 2 turns for each revolution of the crank. Just do some math to find the time window needed to measure the pulse.

 

[Victor Charlie]

I completed:

1) new ground for fuse block
Observation - it was very rusty and poor contact
2) ground under intake manifold for ECU, add redundant ground wire
Observation - bolt was never tightened, my bad. the white/black
ground wire wasn't even attached.
4) adjust (per TSRM)and test TPS to be sure code 51 is cleared again
Observation - no codes.
5) adjusted idle set screw while doing step 4

Now it runs well again, full power. Idle is stable except idles fast for a minute after acceleration, then suddenly slows and surges a few times before reestablishing stable idle. This appears to be in part due to dashpot needing adjustment.

So I think my problem is solved, and I think the most important issues were two bad grounds, poorly seated MSD wires, plus TPS needing adjustment.

Thanks a million to everyone who helped me figure this out.

 

[Victor Charlie]

More driving, still the original issue with idle, though it is now drivable.

Unstable idle:
> when cold: Now it starts and idles well, but smells very rich.
> After a little driving, not warm yet: it starts holding a higher RPM after revving and laying off the throttle, then after about a minute, it starts with about 1 1/2 sec cylces of normal idle speed, which cuts off suddenly but briefly and rpm goes down, then raises up more gradually. In these cycles it is now firing consistently, no stumble or obvious misfire, seems to be running on all 6. still smells rich.
> When fully warm it idles nicely, runs well, just smells slightly rich.

Only known remaining defects: very restrictive catback from stock NA, soon to change, and timing still set 15 deg. BTDC.

The unstable Idle has been there from the start of this thread, but now is is running well enough that it doesn't stall, so the car is usable. Starts fine too, hot or cold, so far. Anyone correct me if I'm wrong, but the sudden cut off seems too fast to me be due to fuel pressure changes at idle, but cold be changes in electrical signal to the injectors. The steady cycle rate makes me think it isn't weak or inconsistent spark.

Question for the forum: pursue fuel or spark or something else next? I have a few obvious things to do still.

New priority list:

1) reset timing to stock and see if the timing light still flashes regularly when it is surging/fading.
2) get A/F guage, aftermarket boost guage, and Tach installed and functional.
3) install wideband so I don't get cancer or holes in my brain sniffing exhaust fumes.
4) adjust AFM calibration screw to optimize to stoich ratio.

 

[figgie]

the first plan of attack. Set timimg per tsrm. That might fix alot of the surging at idle.

 

[Victor Charlie]

the first plan of attack. Set timimg per tsrm. That might fix alot of the surging at idle.

Interesting. i wouldn't have guessed that. I will do so.

For posterity: forget all the cold/warm issues with idling. It turns out that the idle instability occurs within a certain RPM range. With no tach, im just guessing it's around 2000 rpm, very narrow band, then steadies out above or below that. The issues with all the other tuning, warming up the engine, and temperature are because all those things affect the idle speed and either put in in or take it out of the rpm range where it surges.

 

[Victor Charlie]

I finally got some follow up information:

I fixed the timing. To my surprise, it was set 5deg too late. I reset it and the spark advance was working. The car drives about the same, I think.

I haven't solved all the problems, but I think I have the spark working well, and I can rule out spark as the cause

remaining:

1) engine speed wavering - roughly 2000 rpm the engine surges a bit 1 1/2 seconds and then cuts out for 1/2 second still, in a regular rhythm. as soon as the throttle is pressed slightly more or less, it runs steady again.

I now know it is getting good spark when this happens because the timing gun still flashes without missing a beat. I tried pulling all the vacuum tubes off the throttle, boost sensor, EGR, and it didn't change the behavior at all. I assume this has something to do with how the ECU is controlling the injectors. it is not a problem driving at all, don't notice any problems driving, so it's just a puzzle.

2) restarting after a stall - mostly starts and idles perfectly now, but my daughter was trying to learn to drive stick in it and stalled it. It wouldn't start again. I cut the fuel by pulling the EFI fuse, crank a minute to "air out" the cylinders, put the fuse back in, and it starts right up. It rarely has this problem anymore, but this trick always works when it happens. I wonder if anyone else has used this.

3) running rich - the obvious stuff isn't getting it. it is consistently rich, revving or idling. I need to get my wideband kit, and also try to rig a leak test.


I'm wondering if I screwed up my AFM when I took it apart for cleaning. i painted the housing, then reinstalled the sensor and electronics. Some of the seal for the electronics was missing, so I used grey RTV goop to reseal it. should I have to recalibrate the AFM with the adjustment screw after this?

Plan now:

1) Swap the AFM.
2) rig some kind of leak down test for boost leaks.
3) get sensors working (tach, AF ration)
4) adjust A/f ratio with the calibration screw on the AFM.

Aside from those, I don't have any other ideas.

 

[Victor Charlie]

I have the zeitronix wideband setup working. What I found was interesting, so I guess I'll revive this thread for anyone interested.

I have two known remaining problems:
1) runs rich/bad fuel economy 17mpg
2) strange idle behavior

Wierd engine data (simulated narrowband signal switching at afr of 15.2)

a) in neutral with throttle plate held steady the engine suddenly slows at 2700rpm, and then gradually builds up again after it falls below 1800 (first figure). This looks similar to deceleration fuel cut but happens at steady throttle of 2% open. I'm thinking a shorting IDL contact would trigger decel fuel cut until rpm fell below a certain point.

b) IN and out of closed loop at incorrect times: in normal highway driving afr drifts while Vf doesn't seem to react in the right direction. When AFR drifts to near the switch point of 15.2, normal closed loop behavior for short bursts, No consistent correlation with rpm or TPS output. (second fig). sometimes same thing at idle with no controls touched(fig 3). Doesn't happen until the car warms up.

I am looking into possibly a bad THW water temperature signal leading to abnormaly low THW temp indication, high enrichment, poor warm starts, forcing open loop. I have a code 21 from disconnecting the stock O2 sensor to let the simulated signal feed the ECU, so that's probably messing things up a bit, but problem b, and also the poor fuel economy both occurred with no codes and with stock narrowband.

I will check the temperature senders and patch the AFM output into the data logger. will also check IDL signal, reconnect stock O2 sensor, clear codes. Not now, its 15 deg F out there.

 

[jetjock]

You sure like to make things complicated. Why the WB? Not needed...

 

[Victor Charlie]

You sure like to make things complicated. Why the WB? Not needed...

Just because:
1) I'm a curious person and I want to learn all about how the TCCS operates, so I love the data logger.
2) I was influenced by posts on this site saying not to even bother tuning without a wideband, and rather than waste a lot of time, I 'splurged' for $280.
3) I didn't trust that my narrowband was working, and I don't know a cheaper way to check it than to compare it to a wideband.
4) it impresses my wife and improves my sex life

EE's should know better than to ask "why?"

But seriously, I agree with you. The wideband doesn't add much. Whenever the engine is under load, which is where all the fun happens, it is in open loop anyway. I didn't understand this until I saw it graphically on the logging software. When it is in closed loop, the narrowband is adequate. I'm much more concerned about why my engine runs rich in open loop.

 

[figgie]

But seriously, I agree with you. The wideband doesn't add much. Whenever the engine is under load, which is where all the fun happens, it is in open loop anyway. I didn't understand this until I saw it graphically on the logging software. When it is in closed loop, the narrowband is adequate. I'm much more concerned about why my engine runs rich in open loop.

well

disconenct the IAC and see what it does. If the flucutations stop... there you go.make sure you disconnect it when the motor is warm though and prior to stopping the motor.

otherwise you might ahve a leak on the intake runners.