1991 5.7 TBI runs rough when cold

[Sonny Baker]

I must be missing something here, but I am about out of ideas.

I have a 1991 c1500 5.7 tbi with 115,000 miles on it. The truck misses for about 30 seconds after start up when the outside temperature is below 50 degrees. I have done all I can think of to this point and need some others opinions. Here is the thing that is so weird to me. At start up when it is missing, there is no smoke to speak of from the exhaust. As soon as it warms up slightly (and quits missing), it smokes just as you would expect. It makes me think it is something to do with the EGR system, but I have thrown a ton of parts at it to no avail. I have no CEL and it is showing no codes.
The truck does have a cold air intake on it, if that matters.

Here is what I have done so far:
Set timing (advance wire unplugged)
Ran two cans of Seafoam through the intake (1 through brake booster and 1 through PCV line)
Checked all vaccuum lines the best I can without smoking it.

New parts I have installed:
Plugs and wires
Cap and rotor
Coolant temp sensor
EGR valve and control module
PCV valve
IAC valve
Throttle Position sensor
Fuel filter
Battery
Rebuilt the TBI

The engine runs perfect once warm and will restart and run perfect, until it cools off.
I am the second owner and purchased this truck in August this year. This problem did not exist until the weather cooled off.

Can anyone give me any advice as to what could possibly be the problem?

Thanks,
Sonny

 

[Rock Hard Concrete]

Do you have any exhaust modifications? A free breathing exhaust will make a cold tbi run funny till it warms up because of the non heated o2 sensor.

My truck has a cat-back straight pipe and it will stumble a bit, but not miss when cold.

 

[sgmbald1]

I am not sure what you call rough when cold but I have been fighting somewhat of the same issue for 6 months now and the colder it has gotten, the worse it has gotten. That is until yesterday... I will explain. At startup my truck would idle for about 30 seconds and then you could smell the flood of gas to the motor. It would miss, it would stumble, idle would go up and down until the engine temp got to 150 and the ECU started taking in sensor data and then it would run perfect... I have all new sensors and a heated O2 sensor as I did a V6 to V8 swap with mid length headers and dual exhaust. Well yesterday a neighbor commented to me that one of my brake lights was out... I have new LED taillights and checked them out and nothing wrong there so I turned my attention to the brake light switch. And dummy me thought that that switch only controlled the brake lights and cruise control... but it also interfaces with the TCC of the transmission and this is what has been causing the issue...... Once I swapped it out, I started the truck and it purred. No stumbling, no jerking, no flood of gas, just purred. I took it for a quick once around the block without warming it up and it purred. So this morning was the big test. Went out fired it up, gave it 30 seconds, and started driving...... IT RAN GREAT....... Hope this helps...

 

[Ken K]

A few possibilities; #1; With 26 years and over 100k, intake valves, stacked with carbon, can soak up a great deal of fuel during cold enrichment process. When warmed up, the fuel trim can handle it as any carbon on the intake valves have soaked up all they can, no longer an issue.
#2; It is triggered by the ECM seeing engine coolant & air temp within 10 degrees of each other at start-up. Without a scan tool to verify these cold reading, we can assume they are ok without codes, but seeing the actual the sensors while cold, is very re-assuring. It is like looking at bad "BARO" and then a good "MAP" after running. It will set the wrong altitude fuel / timing schedule with no codes.
Remedy; Using a pressure pot (Kent-Moore / OTC, etc.) & 10% GM upper engine lube or ACDelco X66P with gasoline, will help clean injectors and remove or loosen valve deposits on the back of the intake valves. It is a pain to get the right fittings in front of the TBI. Many have used chemicals from sea-foam (I do not like), Berryman's B-12 Chemtool down the Throttle body to attack the intake runners & valve deposits, but not so much for injectors. Be very careful and contact the manufactures as the electric fuel pump in the tank will take a severe hit and fail quickly, if your juice is wrong for the tank. In the 80's, Berryman's had a "CAT" kit with 3 cans, 1 in the oil, 1 in the tank and 1 to sipped into the intake / carb, of a warm engine, while working the throttle to keep running. Once the can was near empty, dump it all in and kill the engine to allow soaking. Wait a couple of hours, start and drive the cob-webs out of it. It amazed me on several occasions as the old bead-type-cat had Boo-Coo exhaust back pressure and fixed it. Just be careful WHAT you put in the tank.
Top Tier fuels where just standards to establish the amount of Techron ratio to gas was agreed and promoted by 6 or more manufactures.
It works over hundreds of fill ups but cost more, it quickly became part of a GM TSB that allowed a tech to clean the injectors once at no charge, then instruct the customer to use Top-Tier fuel from now on...other wise, customer pay.
There are many "Mechanic in a Can" products, but over the years, you have "Marvel Mystery Oil", "ATF", "Water" and others. The problem as i said, happens in the tank. Fuel pumps have to use lead-free brushes, wire coatings dissolves, etc. Trucks emissions & CAFE fall into a different category with cars first. The "Bucket-Type" fuel pumps assembly's called "Module Reservoir Assembly" or "MRA's" kept fuel for cooling, lubrication and most importantly, fuel without air in low tanks that cause misfire. Instead of a top-referencing pump or pump on a stick gave way to bottom referencing with some using a high pressure bleed to the bottom of the bucket, used the venturi-effect to draw fuel into the bucket. These pumps have two filter socks, one out side and one inside on the pump. The bucket will hold dirt, dust, rust, etc. and demand inspection or suffer a repeat failure if the tank is not cleaned with hot water.
Misfire detection rules where relaxed by CARB in 1997 to stop check engine lights from blinding us all of the way home.

 

[skyd]

I had a similar issue with my 93. It ended up being the ignition control module going bad.

 

[DerekTheGreat]

Like skyd said, try unplugging that EST wire to see if it starts up just fine with that unplugged. Speaking of that, what did you set the base timing to? Should be 0 degrees from the factory. If you have a scan tool, hook it up and see what the sensors are sending to the ECM.

When you say it's missing, is it actually missing or just idling real low?

 

[Sonny Baker]

Update.......You guys are on to something.
Here is what I have done and what I have found.

I have:
Removed wires cap and rotor. I polished the distributor stem and reinstalled cap and rotor. Reinstalled wires per the Service Manual and ensured 5 & 7 wires are routed apart from each other. Problem still exists.
I verified that timing is stil set @ 0 degrees with the EST wire disconnected.

I let the engine cool down and hooked up the timing light.
On cold start the timing is advanced so much that you cannot see the timing mark on the balancer and it is firing radically (based on the light) at #1 wire. As the engine starts to warm (30 to 45 seconds) the timing comes back to what I would say is normal at cold running condition and the engine smoothes out and idles perfect.

I let the engine cool again and disconnected the EST wire.
This fixed the problem. It was idling a little slow, but no stumbling or backfiring through the TBI.
I shut it off, reconnected the EST wire and restarted. Problem came back for about 30 seconds then ran smooth.

So the problem seems to be that something is advancing the timing way out of range on cold start for 30 to 45 seconds, then it comes back to normal and runs fine. The colder it is, the worse it is and the longer it lasts.

Any ideas?

Thanks,
Sonny

 

[Sonny Baker]

I thought of something.

Since the engine does not miss/ stumble at cold start with the ESC wire disconnected, are there any other sensors that the ECM "ignores" with this wire disconnected and returns the setting back to a predetermined value?

Thanks,
Sonny

 

[camspeed01]

I know this is an old post, but has anyone else experienced this issue and found a fix? I just got my 92 k1500, and frankly I have never worked on something like this before. I was able to read the codes and got code 43, referring to a knock issue, which sort of relates to the last 2 posts on this thread.

Sent from my IN2025 using Tapatalk

 

[DerekTheGreat]

Seems to be normal to me. Like Sonny mentioned, every time I look at the timing using a light with the ESC wire plugged in, the timing is usually really advanced at idle for some reason, sometimes erratic. Neither of our trucks miss at idle regardless of temp, its just that sometimes they idle smooth as polished glass and other times they idle rough regardless of ambient or engine temps. When they're idling smoothly, the timing isn't as advanced. Why the ECM commands this is beyond my comprehension.