Diagnosing rough running LT1

Hi all - the poor running characteristics that I've always had in my car persist after the engine rebuild. In a nutshell here's what I observe


Start the engine when it's cold. Runs fine at first.
Once the engine goes closed loop, it starts running rough.
Let it idle and get a bit warm, when you then blip the throttle it'll break up, stumble, and occasionally backfire before it revs up.
Sometimes it sounds like it'll stall when returning to idle

The backfire thing is a big clue since as you guys pointed out a backfire can only happen when either a cylinder fires at the wrong time or an intake valve is open when it shouldn't be. Since the valvetrain is new I doubt it's a valve problem, so it's most likely an ignition problem.

Another clue- the exhaust is really, really stinky. Smells really rich to me. Cats are new also.

I've replaced every part of the ignition system more than once by now. I even swapped out the PCM tonight just to see if maybe that had something to do with it, and it didn't make a difference. I can make the car backfire almost at will while idling once it's warm.

I have looked at the various data points on the motor and nothing looks wrong to me, but at a sample size of one I don't have anything to compare it against.

So, given that I've touched just about everything and the problem persists, I think I need to start over and make no assumptions about what's probably good and what's probably to blame. What would you guys suggest I do to start eliminating variables and isolate what's really going wrong here?

Which sensors have you replaced? Did you replace the opti, injectors, fuel pressure regulator?

Have you tracked what the idle air motor is doing during the problem stages?  If the balance between IAM is off between the throttle plate position and the amount the IAM is open, you can get weirdness when you crack the throttle.  The IAM should then open more as the volume to the engine increases with throttle opening so that when the throttle is snapped shut it still gets adequate air as slowing down.  I haven't looked at if any of the rates are settable on this programming, but it was on my old TPI aftermarket programs.  Changing cams or throttle body size can often need some tweaking on those kinds of settings.  On the TPI they called it the throttle follower.

On edit: Just looked at the DTS program and it shows 3 throttle follower settings. Two are in the constants tab and appear to be the time between IAC opening steps. One for P and N and another for in D. There is a shorter time in neutral which makes sense because engine speed is going go up faster than in drive with load on the engine. The other is a throttle follower offset in idle tables section. I don't know which way it goes but probably is taking steps out with vehicle increases for a while and then puts them back in faster, but it could be the other way to. They may be playing with engine braking or something with that. On my TPI setup I had to have the follower release quickly because it was a manual transmission or the rpm wouldn't drop fast enough when shifting. You may have noticed a lot of computer controlled cars that when you blip the throttle in neutral they go back to idle quite slowly to a soft landing. That is from the throttle follower.

Fred: The only sensor I haven't replaced in recent memory is the MAP sensor. The only part of the ignition system I haven't replaced is the engine wiring harness. I did not replace the injectors or the FPR, but I did do an injector balance test which seemed to be normal. I think re-checking fuel pressure and injector balance again would be good.

Booster: I hadn't thought of that, I'll see if I can check it with datacat. The entire throttle body presently on the car came from a crate pulloff and it's "new", so if something was wrong with the old one I'd expect some change in behavior which I don't.

I'll do some testing and report back- thanks!

If you can get the scanner to run a log of the IAC vs rpm, tps, and such you may be able to identify what is happening at the stumble points.

Don't get fooled by exhaust smelling rich, as I have in in the past. It can also smell of fuel from a lean miss sometimes. Stumbles and backfires can both be caused by leaning conditions, especially when feathering or letting of the throttle after a higher load period. When I was running a carb on the turbo 340 it would backfire occasionally when I accelerated into boost, which opened the enriching secondaries, and then when I would feather and the secondaries closed it would go lean and backfire.

I had a problem with my LT1 buick that it would stumble and hesitate under hard accel I replaced the MAP and that problem went away.

INTAKE backfire:  

unplug the maf ,.. see how it runs,.
Plug the vacuum hose going to the egr, see if runs better
pull vac booster hose plug it too,. see if it runs better
inmo you have a vacuum leak,. when the 02's see too much oxygen they dump fuel in to cover it up,.
Check the pigtail going to the ect in the front of the water pump,.if its suspect crop it back and solder in a new one,.
weak spark will show when the motor goes lean,. if you have a shitty coil (buy a decent msd),.
or poor icm  

assuming you have new spark plug and coil wires and new properly gapped spark plugs,.?  

EXHAUST backfire,.crossfiring wires,. exhaust valve too tight

put a gauge on the fpr,. and a vacuum gauge on the engine,.

Thanks all, these are helpful tips but Nick, thanks- I did what you suggested just now since it was an easy test:

Yes the backfire is through the intake.

Plug brake booster line: no change
Cap vac fitting to EGR solenoid: no change
Pull connector on MAF: well, engine died. But I restarted it, and it ran better. No backfire. Plug MAF back in, engine runs worse, backfire returned.

Swapped MAF with another junkyard MAF: no change

Ignition system is all new, ACdelco plugs with correct gap. New MSD wires. New MSD cap and rotor, new MSD coil. New ACdelco ICM. I also added a MSD 6AL, which makes no difference in the backfire whether I bypass it or not. (runs better with it enabled though)

It backfired under the same conditions before I touched any of this stuff.

ECT sensor is good. Scan data shows good temps there.

Trying to think what pulling the MAF would do, other than I suppose put the computer back into open loop mode. That'd make sense since I can't get it to backfire when it's cold before the O2s get warm enough to enter closed loop.

Also I did use a piece of hose and used it to listen around the intake for leaking sounds, and sprayed some carb cleaner around various spots and listened for changes in idle. Didn't notice anything.

Maybe the two MAFs I have are both bad? Seems unlikely, but certainly possible. What next?

lamune wrote:

Thanks all, these are helpful tips but Nick, thanks- I did what you suggested just now since it was an easy test:

Yes the backfire is through the intake.

Plug brake booster line: no change
Cap vac fitting to EGR solenoid: no change
Pull connector on MAF: well, engine died. But I restarted it, and it ran better. No backfire. Plug MAF back in, engine runs worse, backfire returned.

Swapped MAF with another junkyard MAF: no change

Ignition system is all new, ACdelco plugs with correct gap. New MSD wires. New MSD cap and rotor, new MSD coil. New ACdelco ICM. I also added a MSD 6AL, which makes no difference in the backfire whether I bypass it or not. (runs better with it enabled though)

It backfired under the same conditions before I touched any of this stuff.

ECT sensor is good. Scan data shows good temps there.

Trying to think what pulling the MAF would do, other than I suppose put the computer back into open loop mode. That'd make sense since I can't get it to backfire when it's cold before the O2s get warm enough to enter closed loop.

Also I did use a piece of hose and used it to listen around the intake for leaking sounds, and sprayed some carb cleaner around various spots and listened for changes in idle. Didn't notice anything.

Maybe the two MAFs I have are both bad? Seems unlikely, but certainly possible. What next?

You probably should change out the MAP sensor if you haven't yet although it runs on that in open loop also.

The clue, I think, might be in when you go back to open loop even with a hot engine it gets better. Open loop is across the board richer in mixture than closed loop to account for cold engines during warmup. That would add to the evidence you are looking at a lean condition, not rich and be back to air intake issues or fueling. It is probably not leaks as that would at least change with the new manifold and other parts, so back to anything that hasn't changed. Throttle body, IAC, TPS, MAP etc. Another possible is if the EGR valve itself has failed and is stock open or leaking. I would try blocking it off or doing a full delete on it. IIRC there are some specs on how far the throttle plates should be open at idle and with the correct amount of steps on the IAC that I have seen online when I was chasing my idle issues on the new engine last year. Even an incorrect or stuck PCV valve could contribute to issues. If your fuel trims look OK, you know for sure it is only in transitions when things change fast like airflow and fuel, plus timing but timing would probably also show up in open loop.

Get some MAF sensor cleaner. Seen a lot of driveability issues caused by wonky MAFs. Clean the junkyard MAF too. Forgive me if I missed it… but how old are the oxygen sensors? Maybe unplug em, and drive it for a day.

from my understanding the MAF not only measures airflow but also air temp so if it's not reporting the right air temps the computer might be miss adjusting things based on false data which could cause the problems

Ok- I did clean both MAFs with the appropriate MAF cleaner. No change. O2s were replaced when I put the headers in, so are new. ACDelco parts.

New Genuine GM made in China MAP sensor arrived. Swapped it in. It actually backfires more readily now than it did before.  

Bank 1 short term fuel trim 128, long term 129
Bank 2 short term fuel trim 129, long term 141

If I recall 128 is the "zero point" with a range from 0 to 255?

Maybe a video would help? The backfire isn't super-obvious in the video

I could try actuating the EGR and see if that does anything. I replaced that valve after I cleaned the intake in the parts washer. In any case though the behavior was there before I touched that as well.

that actually sounds exactly as my car did before I replaced the mass air flow, and I did try cleaning it first but with no result, but when I put the new MAF in now the car fires, nearly stalls then idles and runs fine

I have no experience with ‘datacat’ but I’m sure it provides the same info as a scan tool. Sometimes you have to step away from “what’s going on with this motor?”, and pivot to “what does the ECM think is going on? And what is it doing to correct this perceived problem?”.

One of the problems I have isn't in getting PCM data. I have multiple ways to look at, store, replay, and visualize as much PCM data as I want. The problem is I have no idea how to interpret any of that data.

I would be certain that someone with far more experience on this motor than I have would be able to look at the data stream and know exactly what's going wrong. That guy isn't me. If anyone knows that guy, please send me his contact info.

That's why I'm stepping back and watching this thread closely. The basics of air, fuel, spark still apply and it seems logical to re-start there.

That being said I do agree with the MAF theory after doing what Nick said and removing it from the equation. I have a reman delco MAF on the way in. I'll give that a try and see what difference that makes.

In the meantime, I'll see what else comes up as things to test to help narrow it down.

check the maf wiring,..see if the plug is corroded or dirty,make sure all the pins are straight and full length,..check where the wires go to the ecm and make sure you haven't got a wire compromised,.. so you cleaned the intake and the egr pindle and they are both clean and the pindle can seat properly?
Have you had a fuel pressure gauge on it,.? fuel pressure is also not an indication of fuel volume,.
Is the fuel rail pinched at all,. i have seen the motor lifting chain put a squeeze in it,. and same where fuel lines clip on,.
go under the car and verify the fuel lines are not kinked or compromised,.

If it runs fine in speed density,. then you have a maf problem
put a vacuum gauge on it,.

Plug the egr vacuum hose AT the intake,. the solenoid may be bad,. always plug your vacuum parasites at the source (intake)


I helped a guy one time with remote diagnosis,. then i went there and found 5-7 plug wires crossed,..
make sure you have good clean connections at your coil,.he also said it had a rough idle,. but it cleaered up as the rpm increased,..

remember its not magic,. plenty of fuel with good spark at the right time and it'll run,.

Thanks all. It was the MAF.

Despite the seller saying it's a Delco reman, I'm pretty sure that's not the case. Whatever its pedigree though, no more stumbling and backfiring, and it'll start right up when hot now also.

Here's an "after" video to compare with the "before" video.



Nick, although I checked distributor to plug wiring many times, I'm certainly aware of what you're saying. I did double-check that and although I found nothing out of order I did find that the #5 wire decided to get up close and personal with the header tube and got a little melty. It's not leaking spark (yet) but that's got to be replaced now. Good thing to find now rather than later!

all's well that ends well,.. search out the ltcc conversion add 8 coils and short wires ,.add the ford injectors guys swear by them they improve driveability quite a bit,.see if joel is still around sherlock9c1

Yes, definitely. Thanks Nick and everyone who chimed in. I am super relieved to have finally solved this particular problem.

I like the LTCC but I think I'd be inclined to go with the Torqhead system and ditch the distributor entirely. It's a lot more money but it seems like a more complete solution. I'm curious if anyone's done it in this group.

I got the note a while ago from Sherlock and Booster is going to be doing the Ford injector swap sometime soon. I'd like to see what he comes up with, but I've seen positive reports about that mod out there.

I think next I have to figure out where I'm getting an exhaust leak. If you come to a stop or are going really slow, I've got the HVAC bringing exhaust gas into the car from somewhere.

I've looked at both, the torqhead gives better options for coils, and also you can get a cleaner looking install as they have a intake mounted coil bracket with cover, and for those of us with the 94 obd1 systems it allows us to upgrade to obd2 as it includes a new plug and play pcm, but you do lose some things with the switch, primarily emissions but I think there were some other computer controlled things aswell that would be lost.

Is the Torquehead a full standalone ignition system with it's own control module. If so, you would lose some spark table covered things, I think. Idle speed control from timing, knock retard, basically anything that is in the OEM program in the spark tables and constants. The factory system looks at a lot of sensors to determine the final timing.

torqhead includes a modified LS series PCM that is plug and play with the LT1 harness, they said they have a basic tune to let the engine run but do recommend getting a tune.

it's been almost a year since the last time I really looked. but I do know it is much more in depth then the LTCC system.

Yes- the torqhead system is here if you want to take a look: [You must be registered and logged in to see this link.]

I'm really curious folks though- what was the MAF doing to misreport it's data such that it'd cause a backfire?

The only way to get pressure into the intake would be to either fire the cylinder when an intake valve is open, or open an intake valve at the wrong time. The second problem is mechanical and not in play, but with a computer controlling the engine you'd think it'd be smart enough to know not to make a spark when the intake valve is still open.

It's one thing to fix a problem but it's better to understand why it happened in the first place, I suppose.

I think what was going on with mine was the MAF was misreporting the air temp, so it might have been running rich enough that when under acceleration it messed up the timing combining it with a rich fuel mixture to compensate for colder air temps that what there actually was. Never was able to properly conclude what was going on with mine as I have no access to a reader, though I was able to burrow my cousins old snap-on scanner with obd1 adaptor and could get live feed on it later one for another issue I had.

as like with old carb engines the choke plate reduces air allowing for a more fuel rich mixture which is required to help get a cold engine going and keep it running in the cold.

The term ‘backfire’ gets attached to every lean or pre/post ignition event…don’t get distracted. If you look at live MAF data (with a ‘good’ MAF), 5.7 liter engine at idle will show roughly 5.7 grams/sec of airflow, + or - a half gram. A dirty/older/failing MAF will usually under-report actual airflow, into the engine.

This article is generic, but does a better job explaining……

[You must be registered and logged in to see this link.]

Other quick and easy data to look at is air temp and coolant. On a cold motor, when it’s 60 degrees outside….expect those two sensors to report roughly 60 *F. Also, key on/engine off, look at the MAP. Should read BARO pressure for wherever you live. If any of those sensors are off….their inputting that bad info to the ECM. Newer vehicle ECMs do rationality checks, and disregard bad info, i.e. ‘ if the air temp is 80*F, then the coolant can NOT be -40*F’. But I can’t speak to how mid-1990 ECMs react to such inputs. But unplugging the MAF entirely, is a good way to diag.

Thanks for the link, that was a good read and things are starting to make sense.

I have a log from cold start to idling to operating temp with the old MAF, and I just did the same thing with the new MAF to compare. In both cases the IAT, baro, ECT all read ambient and correct values before startup.

Upon reaching operating temperature at idle (thermostat open) the original MAF was reading 8.3g. The new one at the same point reads 8.0g. That's pretty far from the 5.7g rule-of-thumb value expected and they're not different enough that I'd expect someone would notice.

Of course that's at idle and doesn't reflect how the MAF responds to changes in airflow, etc. but I thought I'd relay my results.

It's pretty interesting nonetheless to try and dig deeper.

Also I found in Volume 2 of the FSM is a chart of (almost) all the data points and their expected values. My bank 2 long term fuel trim is out of the recommended spec. Looking at the graphs of the two O2 sensors side by side the bank 2 one looks a bit weird. I've got another O2 and I'll swap it to see if that makes a difference or not.