Engine misfire troublshooting

[imjeff]

Originally Posted by Kudzupatch

But in this case, it been quite a while since it was tuned up. It has run fine for several months.

I guess I misunderstood your original post where you said you gave it a complete tuneup "a few weeks back". I was under the impression that you made changes, then began having problems during a trip.

[ChevyTech]

Originally Posted by imjeff

I guess I misunderstood your original post where you said you gave it a complete tuneup "a few weeks back". I was under the impression that you made changes, then began having problems during a trip.

Me too!

[Kudzupatch]

Sorry about the confussion.
Tuneup was several months ago so this is a new issue.

UDATE:
Replaced ERG valve earlier this week with no change.

Today I replaced cap and rotor and as stated in previous post, checked the wires, plugs and everything I could see in the ignitions system. Now t is running much better but not quite perfect.

It backfired though the TBI one time within 2-3 minutes of starting the test drive and didn't seem much improved. I took a second drive and have driven it several miles through town and on the highway. It has not backfired again and no more rough idle either. Could it be that the computer had to adjust it's self after the repairs?

Still one issue left. It 'stumbles' or cuts out from a standing start with normal light throttle start. Not always but most time. It only does it for a couple of seconds and then it runs fine till the next stop and start. Similar to a carburetor when the accelerator pump quits. I did notice if you are moving even 4 or 5 mph it doesn't do it. Seems to be only from a dead stop.

Much better than it was though!

[Kudzupatch]

Spoke too soon. Drove it some more and stumbling or as my wife calls it, 'Stuttering' kept getting worse and it about where it was.

Now I am getting two codes 15 and 33

Engine coolant temperature (ECT) sensor -voltage high
Manifold absolute pressure (MAP) sensor -voltage high

As soon as it cools enough to remove the cap I will double check the Coolant sensor.

Decided to try it with the O2 sensor disconnected and it ran perfect. Stopped at a buddies near me and we were discussing it and looking and I saw I had left the ECT disconnected.

Hook it back up and it runs bad. Hook up the O2 also no change. Take the ECT loose and it took the computer a few seconds to adjust, then it runs fine. Have I found the problem??

When I checked the resistance it seemed to be very close to right.

Also, been thinking about the distributor play. How much is a problem? Hard to tell if there is any movement. Was thinking about putting a dial indicator on it in the AM to see exactly how much it moves. Of course it maybe (hopefully) a mute point.

[Kudzupatch]

Checked MAP sensor and seems to be working fine.

Checked the ECT again with with 2 thermometers)
125 degrees I am showing 404 ohms.

Per the chart 122 degrees should be 973 ohms.

404 ohms should be between 176 deg. and 158 deg. F.
Chart says 332 ohms @ 176 deg. F and 467 ohms @ 158 deg. F

[ChevyTech]

If you ran the engine with the coolant sensor unplugged, that will set the code 15.
If it sets a code 15 without the sensor being unplugged, then there is a problem that needs to be fixed.

Testing the coolant temp with a thermometer in the radiator is not a valid test. The coolant in the engine can be much warmer then the coolant in the radiator.

If you can't feel any sideways movement in the distributor, don't worry about measuring it.
If it wiggles much it should be replaced.

You better test the vacuum supply to the MAP sensor.

[Kudzupatch]

Originally Posted by ChevyTech

.......Testing the coolant temp with a thermometer in the radiator is not a valid test. The coolant in the engine can be much warmer then the coolant in the radiator.

Yea, I should have thought of that. I could see how there could be a 30 degree difference.

So, help me understand why the engine runs so good with the sensor unplugged? Am I wrong in thinking that it means that one of the sensors is sending a faulty signal to the computer? With the sensor unplugged it runs great. Plug it back up and all the trouble starts again.

I am really starting to relate to all the frustration others are feeling troubleshooting these engines.

[ChevyTech]

The system has detected a problem.

The system no longer trusts the sensor data and system runs at a predetermined calibration. This calibration is what will keep the system running best across the most normal conditions, without sensor data.

Yes, you are correct that the ECM is most likely getting bad data from at least one sensor.

When the sensor tests to be good and still seems to be causing a problems consider it may be a bad connection to the sensor or a bad ground or grounds. Test the ground side of the sensor too.

Here is how a oxygen sensor can or misfire can affect the system:
These systems work great when every thing is right but they have a big flaw. Once the engine misfires for any reason it throws the system in the rich direction. This system watches the O2 sensor, but the O2 only will see how much oxygen is in the exhaust stream NOT burnt gas. When every thing is right the O2 knows just how much of the oxygen should have been burned by combustion. If the system gets a little to rich all oxygen is consumed. If it gets to lean there is not enough fuel to burn all the oxygen and there is more oxygen in the exhaust. Once it misfires, there is way more oxygen in the exhaust, because none got burned in a cylinder. The system thinks it is to lean so it gives the engine a richer mixture. The richer mixture often makes it misfire more (especially if it is cause by weak ignition, bad plugs…) so it dumps in more fuel. Now it is so rich the plugs and O2 sensor get soot on them. It misses more. The system goes richer and richer.

If you don't seem to be getting a handle on what is going on, you should follow where the codes bring you. Clear the codes, run the truck, and test for codes again to see if they get reset.

The TBI system is the easiest fuel injection system to work on with the exception they did not put a fuel pressure test port on it.

[Kudzupatch]

Originally Posted by ChevyTech

Yes, you are correct that the ECM is most likely getting bad data from at least one sensor.

Great. That makes sense.

When the sensor tests to be good and still seems to be causing a problems consider it may be a bad connection to the sensor or a bad ground or grounds. Test the ground side of the sensor too.

I will go check that now. I know grounds can create lot of issues and I always forget to check that.

Here is how a oxygen sensor can or misfire can affect the system: .....

That makes a lot of sense! Very good to know!

If you don't seem to be getting a handle on what is going on, you should follow where the codes bring you. Clear the codes, run the truck, and test for codes again to see if they get reset.

Will do! Going to go ahead and put in the temp. gage sensor and I will hook up everything and see what happens. Tempted to put in a new ECT while it is drained but I am really starting to doubt that was the issue now.

Once again, thank you very much! Always wanted to know more about how these work but.... I would have prefered to learn it from a book, not first hand.

[Kudzupatch]

Saw something that gave me an idea. I took a piece of copper welding wire and it used it to probe the connections, just slide it down beside the wires till it hit the terminal. So, I have rechecked my sensors, checked for ground back to the negative post on the battery and they are good.

Borrowed a timing light since mine is MIA for some reason. Timing is set at about 4 degrees. Can't get to the nut to adjust it. I am sure Jim has a wrench I can borrow.

Reset the computer before checking the timing. With it connection back on it stays around 4 degrees at idle. With throttle it advances and is stable, not jumping around.

I climbed in the engine compartment, removed the cap and rotor and there is a small amount of play in the shaft. Very small, but from what you are saying any play is cause to replace it?

I don't mind spending the money but I just don't want to throw money at it and replace good parts. So I held off on the ECT sensor for now.

Taking it for a drive shortly and will check codes. But I am really wondering if it is the ignition/distributor now.

[speedygonzales]

Why don't you look for play in the timing chain rather than just the distributor?

[Kudzupatch]

If I understand correctly, a bad bushing in the distributor can cause the electrical components to touch and lead to ignition issues.

I am pretty sure there is at least some slack in the timing chain. I don't see it causing this issue.

[Kudzupatch]

Quick update. The rotor cap was obviously a big part of my issues. As I stated out of desperation I checked the resistance on each of the terminals on the rotor and the coil tower had over 4 million ohms resistance (EXTREMELY high). That has taken care of 95% of the problems. No more Check Engine light either.

There is the stumbling on starting up and the more I drive it, I think your right ChevyTech and it will be a cracked plug. It feels like a skip when it does it. I have pulled them but can not find a flaw. So I have a new set to put in and then I have a long trip ahead of me. So that will tell if it is fixed.

[ChevyTech]

Thanks for the up-dates.
I am keeping an eye on this thread.

[Kudzupatch]

No love for the home team. Put new plugs in this morning before leaving and it stumbled pulling it around front.

Drove it 350+ miles with a load and trailer, just like I did 3 weeks ago. MUCH improved though. Ran perfect out on the road. Just have that stumble when I start up and when we came off the interstate into some stop and go traffic I noticed it was idling a little rough too.

FWIW it was pouring rain most all the way. Didn't seem to make it any worse though.

When you step on the gas is stumbles most times. Sometimes really bad and other just barely. Stepping down on the gas will make it go. So it is only happening at very first of the throttle opening.

Time for a compression check. I found my vacuum gage so will put that on it when I get home next week.

Could it be an O2 sensor? Buddy of mine is really leaning toward that but I am not sure if that could cause it?

[speedygonzales]

A very common off idle stumble is caused by an EGR problem. My friends was leaking around the gasket.

You can try disconnecting the O2 and let it go to open loop to see if it gets better.

[Kudzupatch]

LATEST UPDATE

As stated previously I replaced the Dist. Cap and that made a big difference. I have driven it 600-700 miles on a trip. I got a solid 2+ mpg more on this trip. Same route and similar loads. The cap fixed all the rough running issues. but I still have that stumble from a standing start and computer gives me no codes.

I was pretty sure ChevyTech's suggestion of a cracked plug was it, but new plugs made no difference at all.

Just as a recap all the sensor check good. Found a leaking EGR valve and replaced it. Found a cracked vacuum fitting (vacuum leak) and replaced it.

Oh yea, before I left I found the timing is at 4 degrees. I couldn't find my wrench, so I haven't set it back to 0 yet. Probably do that tomorrow.

Only thing I haven't tried and don't know how to check is the O2 sensor. I guess I could disconnect it and see if that makes any change. I don't want to do what most folks do and just replace perfectly good sensors but this might be an exception to that. Otherwise I am out of ideas.

[Kudzupatch]

Since I am stumped and I seem to have everyone else stumped, I am working on getting WinALDL cable built. Hoping this will shed some light on my issues.

BTW I have discovered mine has an odd ball cpu, 16168625 and the terminals do not match up to the standard cable configurations. That leaves me wondering if WinALDL will even work with it?

[ChevyTech]

Here is what I have in my notes about your ECM number:

Service number 16168625 was used by GM only in 1993 in the light duty trucks.

1993 C/K Truck, Van, 4.3L, 5.0L & 5.7L engines, and S/T trucks.

If your 1993 has an automatic transmission, then it has a 8192 BAUD data stream computer and the computer is technically a PCM. WinALDL will not work.

WinALDL is capable of reading a 160 baud data stream.

WinALDL will not work with a 8192 BAUD data stream. Tunerpro can be used.

[Kudzupatch]

Thanks, I will shift gears and move toward Tuner Pro.

It's a 93 Suburban, 350, automatic BTW

[Kudzupatch]

Today I noticed a click or knocking sound. Similar to a lifter but not the same sound. Got a buddy of mine that knows cars to listen and it stumped him too.

We have decided it is coming from the top rear of the engine. Don't here it under the car but it's clear on the top.
It's not the injectors.
95% certain it is not a lifter. It is mechanical and tied to engine speed. Distributor seems like where it is coming from but I have never heard of one audible knocking?

I really want a mechanics stethoscope. Going to buy one next time I order anything.

[ChevyTech]

The distributors can get very worn on these trucks and the shaft can wiggle sideways causing running problem like I stated before. Check to see if the shaft wiggles.

Anything knocking or ticking load enough can cause the knock sensor to read knock counts and then the timing will retard. Check into what is making noise.

[Kudzupatch]

Sliced open a finger so my work on the Suburban has been limited.

So far:
Timing set to 0 from 4 degrees

Vacuum check 18" steady at idle, 20" when I rev it up.

Distributor has only the slightest detectable play in the shaft. I can just barely feel the slightest movement.

Clicking sound I THINK is coming from the fuel pressure gage I installed. It failed and even with a stethoscope I can' be certain but it is loudest there. And it failed at the same time the sound started.

Truck runs good except for the stumble, so the Dist cap was the problem causing the misfire.

Still have the stumble at start up. If it were a carburetor I would lay odds on a bad accelerator pump. It fells just like that. The one thing I want to double check is the actual voltage at the TPS. Just make sure that there is not bad position just off idle. Past that I am stumped.

[Kudzupatch]

Double checked the EGR this morning by pluging the line and no change.

Checked the actual voltage at the TPS thinking that it might be a problem just off the idle postion. I see .8 volts at idle and any movement causes voltage change. So that (appears to me) to eliminate my last ideas.

Only clue is that is worse since I back the timing back to 0 from 4.
Time to buy a cable and software?

I am stumped. Bueler? Bueler?

[ChevyTech]

Originally Posted by Kudzupatch

Only clue is that is worse since I back the timing back to 0 from 4.

This is a sign that the engine has a mechanical knock and the system is pulling timing.

The starting point is with the timing advanced more so the problem is not as noticable with the base timing set a 4 degrees BTDC.

The Knock sensor contains a piezoelectric crystal which causes a voltage to be produced. Loose components on or in the engine can cause the knock sensor to create a voltage and the system may interpret this as pinging. Lifters, cam, rod, crank and piston noises all have the potential for causing false knock signals to be produced. Even lose torque converter bolts can cause a false knock signal to be detected.