Mitsubishi 3000GT & Dodge Stealth Forum banner

Engine "cuts out" at idle and low load

6.6K views 40 replies 13 participants last post by  Hammer  
#1 · (Edited)
I did search on rough idle, and stalling etc. but have not found problem similar to mine.

This happened couple months ago, but went away. Now the gremlin is back.

When I first start the car, it idles fine at first, but as the idle slows it begins to "cut out". Just like the key was turned off/on quickly. If I give it a tad of throttle the "cut" goes away. The car accelerates normally and runs fine as long as rpm is up and/or there is some load.

If idling with the clutch pushed in, it will run at approx 1000 rpm, then 'cut' and drop to 500 or so, and continue to do this every several seconds. It completely stalled out twice at traffic lights.

When this happened a couple months ago, the 'cut' would stop when it fully warmed up. Today it continued even after warm.

'91 Turbo, basically stock. ECU was replaced before I bought it.

Any ideas?

Update 1/30/2006:
The problem seemed to disappear after replacing the ECU caps, but then came back after several weeks. . . . When it came back however, it was much more consistent in how it acted. . . . I think originally I had a issue with both the ECU and ISC, then weeks later just an ISC issue. (see post #26)
 
#3 ·
The plugs are relatively new, the wires have not been replaced since I have owned it. I also wondered about those. Maybe I should try to start it at night, and see if I get a light show. :)
 
#4 ·
3Sfever said:
I did search on rough idle, and stalling etc. but have not found problem similar to mine.

This happened couple months ago, but went away. Now the gremlin is back.

When I first start the car, it idles fine at first, but as the idle slows it begins to "cut out". Just like the key was turned off/on quickly. If I give it a tad of throttle the "cut" goes away. The car accelerates normally and runs fine as long as rpm is up and/or there is some load.

If idling with the clutch pushed in, it will run at approx 1000 rpm, then 'cut' and drop to 500 or so, and continue to do this every several seconds. It completely stalled out twice at traffic lights.

When this happened a couple months ago, the 'cut' would stop when it fully warmed up. Today it continued even after warm.

'91 Turbo, basically stock. ECU was replaced before I bought it.

Any ideas?
Two things to try

1. Do a pressure test and and fix any vacuum leaks. Typically the engine runs lean and the engine speeds up burns the fuel then drops tries to die then speeds up again and repeats the cycle.

2. You ecu was "replaced" perhaps with another old ECU with 16 year old caps that are now bad or waiting to go? Remove the ECU and visually check it.
 
#6 ·
I was thinking fuel filter and i think that really hurts it more at a set rpm and not when its cold. Thats why I thought wires, pressure test is always a must but dont see why it would get diff when its cold or warm.


Coop
 
#7 ·
I said filter because he said it would run fine as long as the rpm's are above idle. This would mean there would be more voltage going to the pump making it provide more pressure to over come a restriction such as a clogged filter. At idle the voltage would drop, thus the pressure would drop enough to where it could not provide enough fuel to keep the engine running. That was my thought anyway!
 
#8 ·
This is not a "floating" idle, it is a sudden drop, like the ignition switch is cut off. Does not feel like restricted fuel or vacuum leak. You never know though.

I ordered a set of MSD wires. I have been wanting to do that for some time anyway, to maybe fix a "spark blowout" problem I have if I turn the boost up past 12 psi. Might reduce the plug gap some while I have it apart also. (currently factory gap on NGK Iridiums)

Also ordered a fuel-pump hotwire kit and fuel filter. (Something else I have been wanting to do for a while).

Oh what the heck, the other "been wanting to do" is a pressure test, I might just have to do that also after I get everything hooked back up.

Thanks for the advice so far.
 
#9 ·
3Sfever said:
This is not a "floating" idle, it is a sudden drop, like the ignition switch is cut off. Does not feel like restricted fuel or vacuum leak. You never know though.

I ordered a set of MSD wires. I have been wanting to do that for some time anyway, to maybe fix a "spark blowout" problem I have if I turn the boost up past 12 psi. Might reduce the plug gap some while I have it apart also. (currently factory gap on NGK Iridiums)

Also ordered a fuel-pump hotwire kit and fuel filter. (Something else I have been wanting to do for a while).

Oh what the heck, the other "been wanting to do" is a pressure test, I might just have to do that also after I get everything hooked back up.

Thanks for the advice so far.
Do yourself a favor and get some NGK BCP7ES non-resistor plugs about 3 buck each and gap them to .028. Don't throw the iredums away but save them untill you get your spark blow out resolved. MSD 8.5mm are great wires. When you run your hot wire through the fire wall shild the wire with vacuum hose to prevent the fire wall from cutting it in the future.
 
#10 ·
idles that drop and die but yet run okay at 1000 rpms or more when the Throttle is held steady can also, on an old car, be the deteriorated FIAV inside the throttle body. As the car idles above 1000 rpms or more, with throttle depressed, she probably is out of the Idle system so the Fast Idle Air valve(FIAV) is not an influencing factor.
Swap in a known good TB and see if the problem is fixed, turn the BISS screw out half way at best when you try this.
 
#11 ·
Isn’t it fun trying to diagnose intermittent problems.

- two days ago, the car ‘cut’ dozens of times,
- yesterday it was bad when cold, but only cut a few times after warming up.
- today on the 20 minute drive to work, it never cut once, (and I was really watching for it) Started up, idled fine, ran flawless the whole time.

Nothing has changed, same tank of gas, weather is similar, etc.etc.

This reminds me of a couple months ago when the problem came and left.

I will continue with my plans to change the spark plug stuff, fuel filter, etc. but now I may never know if that fixed the problem, or it just went into hibernation again.

Thanks for all the advice. I will update the thread, if I notice anything of interest.
 
#12 ·
check the IAC, steps to clean it are on stealth316. That helped me with my idle issue.
 
#13 ·
The "cut" still seems to be an intermittent problem, some days it does it, other days not . . . baffling.

I put the new MSD wires on. Did not fix the cut issue, but sure did improve the idle. This car has always had a little "miss" while idling, not really a rough idle, but a little "putt" now and then. I just thought that was kinda normal. With these wires, it is more like every cylinder fires every time. :)
Should have replaced the wires a long time ago. (too lazy to take off the intake)

As I mentioned, even after the wire swap, I still had a "cut" incident. Car ran fine in test drive, fine on the way to work, but coming home yesterday it decided to act up at a stop light one time. Had to give a little throttle to keep running.

Since this is so intermittent, I thought I would pull the ECU out and inspect the caps. There is no clear evidence of leaking, but I did do a "smell test". I read a couple places that leaking caps will smell like rotten seafood or fish. There was a very faint odor similar to that. I even asked my wife her opinion (honey, smell my ECU and see if it smells like rotten seafood :D ). She also thought there was a slight odor. This ECU obviously had the caps replaced at one time before.

Does anyone know --> could the smell be from the previous capacitors, or if there is any smell, then the caps need replaced?
 
#14 ·
Do the caps anyway, only way to rule them out. I spent ages trying to fix a running fault on my mk1 and it was down to the caps. After that may be an idea to set the idle as per the service manual and check the idle switch (a lot easier with a datalogger). I found a tip off here worked very well, get the engine hot and with no accessories or electrical load set the idle bypass screw so you have the IAC on 6%/steps, worked the best out of all the methods I tried.
 
#15 ·
Still troubleshooting this. I have some caps on order for the ECU.

Attempts so far:
- New MSD SC wires
- New NGK copper plugs at .030
- Pressure tested, and all leaks fixed,
(except a couple really small ones @15 that I can’t find.)
- New fuel filter (filled filter with seafoam, to give injectors a cleaning, maybe)
- Cleaned throttle body as much as possible on the car.
- Changed the oil :D

The engine seems smoother, I think the wires helped that, but I still get the sudden cut intermittently.

This morning I was going through the parking lot at work, clutch in, coasting to parking spot. It cut, I thought I saw out of the corner of my eye -> the Scanmaster go to 0, during the cut. (Normally at this point it is cycling though values in closed-loop). Then, it did it again, same thing (I think) -> O2 volts to zero. Both times I was watching cars and people, so could not really be sure. After stopping, I watched it for a couple minutes idling, but it would not cut. :( Wouldn't cut while watching, so can't say for sure.

If that is true, (O2 volts going to zero during the cut), what does this mean? I would think that means the fuel is shut off temporarily. If it was spark miss, then seems like it would go rich, not lean.

I hope the ECU caps fix this.
 
#16 ·
3Sfever said:
I hope the ECU caps fix this.
Soldered in the new caps, and took for a short test drive today. No "cuts" so far, (but it was a short drive). . . Still have my fingers crossed.
 
#17 ·
Drove for approx 45 minutes this morning with no problems. That is the longest it has gone recently without a 'cut' incident. Still have fingers crossed.
 
#18 ·
I can't say for sure if I can close the book on this issue, (because it has gone away and returned before) . . . but I do feel fairly confident however that replacing the ECU capacitors fixed it.

It has been 3 days now since replacing caps . . . and experienced absolutely none of these intermittent "cuts" while idling etc. I drove a lot yesterday.

Another problem I have also been having for months is at WOT. I assumed it was spark blowout all this time. . . On the way to work this morning I tried a 20mph - 100mph WOT run. It was the smoothest I can remember in months. No surges, stumbling, cuts etc. I imagine part of the cure to be all the other new stuff (plugs, wires, filter, boost test) etc., . . . but . . . it was acting up 4 days ago, even with all the new stuff.

Now I can watch the logs and start cranking the boost up :D . I have been running 8-10psi for months now, waiting to "get around to" changing the plugs etc.
 
#20 ·
It has not "cut" at idle a single time since I replaced the caps. I still have problems when I try to boost past 10psi. (runs great up to that point though). It is not consistent in what it feels like. Sometimes just a "shudder", other times it "cuts" a little harder. Then oddly other times it runs great.

Mine is not a consistent RPM thing. (I skimmed thru your post earlier. That is an odd one). Mine is more "load" dependent, than RPM dependent.

I just replaced the throttle body shaft seals yesterday, (the last of my boost leaks). Took it for a test drive . . . first time I went WOH it bucked like crazy when the boost picked up. Then later, after almost home, I tried again, and it ran great up to 12 psi, where my MBC is currently set.

My next step is fuel pump hot-wire. If that does not work, I am going back to looking at ignition.

Concerning the ECU caps. I am FAR from an expert, but if I was you, I would pull the ECU. It is easy to do. Take a look for leakage. (and smell it too) :D . From what I understand the caps "clean" the power to the ECU, and if they are not working correctly, all kinds of strange things can happen.
 
#21 ·
Did you ever think to check your coil packs? My VR4 was doing the same thing under load for the longest time after changing the plugs and wires twice in a span of 3 months. Bucking at WOT, but also idling too low and stalling, etc. I was baffled seeing as how nothing seemed to be amiss with my ECU and wound up taking it to Ray...just to find out one of my coil packs was shot.


From what i hear, it's not too common but still worth it to check. Good luck :)
 
#22 · (Edited)
TrAb: 1) Just to be clear, my car has had problems when boosting past 10psi since I have owned it. The "cut" at idle was new, and seems to be totally fixed since replacing the caps. I think this seems different from what you are experiencing . . . except . . . bad caps do strange things!

2) off-topic . . . nice looking car!
 
#23 ·
Kbizzle: yes, that has been haunting me. I will be checking coil packs (and PTU) soon. Do you happen to know off-hand, is that something I can check myself (with VoltOhm meter etc.) or do I have to take somewhere? I will be researching it soon.

Thanks for the suggestion!
 
#24 ·
You can check the coil packs with an Ohmmeter....or you could try swapping them out with some known working coil packs if anyone close to you can help you out with that.

Some subtle signs that might tip you off to the coil packs for example: in my case, every time i changed the plugs, 2 plugs were consistently different in appearance compared to the other 4. I want to say numbers 3 and 6, i dont remember which ones anymore. Both plugs managed to look like this each time i changed them.
Image


It's a bad picture but up by the electrode on both plugs was completely covered in black residue also. With a bad coil pack you'll usually find your car operating without one or two cylinders seeing as how one coil fires a pair of spark plugs opposite each other in the firing order. The shudder and sputtering you feel on WOT could very well be your car operating on say 4 cylinders and the 4 working cylinders trying to overcompensate for the 2 dead cylinders. And other times, the car may feel like nothing at all is out of the ordinary. Every time the car felt normal, i tricked myself into thinking it was fixed, never thinking to check this.



Found this in a 'how to', i've never tested a coil pack myself but it seems pretty self explanatory if you know your way around an Ohmmeter:

"First, isolate the coil pack by disconnecting all the leads. Set the ohmmeter in the low range, and recalibrate if necessary. Connect the ohmmeter leads across the ignition coil primary terminals, and compare the primary resistance reading to specifications (typically less than 2 ohms). Then connect the ohmmeter leads across the coil secondary terminals and compare the secondary resistance reading to specifications (typically 6,000-30,000 ohms). If readings are outside the specified range, the coil is defective and needs to be replaced."

I hope this helps.
 
#25 ·
Kbizzle said:
Both plugs managed to look like this each time i changed them.
Image

It's a bad picture but up by the electrode on both plugs was completely covered in black residue also.
I opened the picture first and thought --> "I can't see anything, except they look black" :D . . . Then read your description, and it says "black".
.
Mine are not black, all 6 looked the same and are light "tan" in color.
.
Kbizzle said:
I hope this helps.
yep! thanks for the "test" procedure.
.
 
#26 ·
Well . . . the problem is back :( This difference is that now it is very consistent.

After some research, I am quite sure it is the ISC. . . . CurtVR-4 posted a link to a DSM site, on another thread, with an excellent writeup on the ISC. . . My car does EXACTLY what is described in this link. (surges between 1200/1500 rpm while coasting to a stop, with clutch in, but immediately quits surging when the car comes to complete stop). I can even see on logger when it cuts the fuel momentarily (INJ=0), rpm drops, then rises, cut again etc.

Time for me to look into fixing the ISC. (Some of you already mentioned that might be my problem, but I did not really know what that meant).

Check this out, if you want to learn a little about ISC.
http://www.dsm.org/how-tos/tttt/tttt3.html
Below is an exerpt from it:

Q: So what is an ISC and what is it good for?
A: The ECU attempts to control the idle speed by opening or closing the ISC. The valve on the ISC allows extra air to bypass the throttle plate. Since a stepper motor does not provide any feedback on its actual position, the computer simply counts steps and assumes that the valve lands in the correct place.

During normal operations, the ECU uses the ISC for some very interesting things besides simply adjusting the idle speed. The ISC is completely shut when you kill the engine to make sure it stops quickly. The ISC is opened completely at startup to provide plenty of air. It is also opened during normal driving to prevent the car from bucking when you take your foot off the throttle to shift. If you run the air conditioner or turn the steering wheel, the ECU compensates immediately by opening the ISC a little to keep the revs up. This also happens during cruising to mask the effect of the increased load on the speed of the car.

Q: So what, exactly, is idle surge?
A: Idle surge is the condition that occurs when something in the ECU idle speed feedback loop fails to operate as expected. Since the ECU is not quite as smart as you are and has a very microscopic view of its world, its behavior may seem very strange to you, but seem perfectly normal to the computer. The result is an idle speed that oscillates from ~1200-1500 RPM about 2 times a second and no error codes are ever displayed.

My idle surge was a result of the following situation:

1) The idle switch is depressed (foot off the gas)
2) Too much air is getting through the throttle body (ISC stuck open)
3) O2 sensor registers lean mix (too much air)
4) ECU adds fuel to achieve correct mix
5) RPMs increase due to increased fuel/air supply
6) At 1500 RPM the ECU goes into decel mode and cuts fuel (*)
7) With only air coming into the engine, the RPMs drop rapidly
8) At 1200 RPM the ECU turns the fuel back on
9) Goto 2.

*) In our cars, the ECU cuts the fuel supply completely when the car is decelerating. It determines that you are trying to decelerate because the idle switch is depressed and the RPMs are above 1200. This is done to prevent an over lean condition and to assist in engine braking.