Intermittent very hard start and rough idle when cold - '90 350 tbi

I've been chasing this issue for a couple of days now and I'm not sure I'm on the right track, maybe I'm over thinking it, but any input you guys have would be much appreciated :happy:

Vehicle: 1990 K2500 350tbi th400 8800lb gvw (8 lug truck)
Stock tune except for timing bumped up in the +1000rpm range and smoothed fueling tables. All stock long block.

(I'm ruling out the tune because it's been starting idling and driving great for months now)

What it does: Past couple days have been cooler here in NC, in the 30s. On a cold start first thing in the morning it will crank fine but will not fire. It will start only if I hold down the gas pedal and feather it a bit. Once started it will barely idle, it will hunt and sputter between 400-600 unless I hold down the gas. It will not fast idle. If I stop holding the gas it will chug down to 400 rpm until it stalls. I can hold the gas down and keep it at 1200ish rpm to warm it up, as it warms the idle will smooth a bit until finally it will idle and run on its own normally. It drives normally after that, and I can shut it down and restart while hot without issue. While I was diagnosing I also noticed it would hunt at idle a bit, bouncing around 1-200 rpm occasionally. After messing with it the problem has inexplicably stopped and the truck is running and starting great again.

Initially I was thinking a bad IAC, but it may just be sticky because I've been unable to replicate the issue and after testing it and reseting it it seems to be functioning normally. I attached a quick snapshot of tuner pro right after I started it this morning while I was trying to get it to act up. As far as I can tell all sensors are reading normally. I diagnosed for about 2 hours last night and all the data read and looked the same

I did notice the TPS voltage did bounce between .51v and .53v and idle causing the Tps% to bounce between .2 and .7. Not sure if this is normal or could point to a failing TPS.

I checked for any vacuum leaks and even blocked off several ports but the problem went away before I could verify anything. I shot some carb clean around the base gasket and vacuum lines, no effect. I put everything back to normal and it's still functioning correctly this way.

I still lean toward sticking iac, I ohmed the pins on it and read in spec at 50ohms through the coils and infinity across. I checked the connector with the A and B ALDL pins jumpered and got voltage into all of them so it seems to be working correctly as well. I'm debating cleaning it but if it's being problematic I'd rather replace it. Maybe I'll pull it and look at it, but from the looks of the end of the pintle I could see from the top of the throttle body it didnt appear dirty. I rebuilt the TBI a couple years ago and cleaned it all then as well.

My other thought was EGR sticking open, as I did push open the valve while it was hunting to see if that affected it, but it did not seem to as it idled worse when I opened it. It's possible it was a a little sticky and I freed it up but it did move smoothly and easily.

Fueling wise it has a new ac delco EP381 pump and injector spray pattern looked pretty good. Both pulsing cones. Pressure I didn't check but after I put the pump inot it was rock solid at 13psi.

I checked timing, it was good at 0* with the ESC plug disconnected.

I'm hoping it acts up again long enough that I can pinpoint it. Right now I'm leaning toward iac, maybe egr like I said. Let me know what you guys think. :wtg:

Bad CTS?

Quote:

---

Originally Posted by Roadknee

Bad CTS?

---

Forgot to mention, I checked that too, reads correctly hot and cold. Reads fine in the data logs too.

IAC seems incorrect. IAC only adds air. You should be able to reduce throttle opening slowly and bring vehicle below 1200 rpm where it should run correctly without sputtering.

If there are no codes then you've likely got a problem with an un-monitored system, or you have a problem with a sensor reading incorrectly.

Does the exhaust smell rich when vehicle isn't running correctly? Is there a large amount of water dripping out of the exhaust? Is the tailpipe black? Those symptoms generally indicate a rich engine which would often require higher rpm to overcome. As it's temperature based I'd be looking at CTS and the temp signal. Yes, it tested ok once. But this is an intermittent problem, yes? You could have a broken wire at the sender, or the pins could be corroded or misshapen. Or the temp scale could be skewed in the sender.

If there is no sign of the engine running rich but it stalls whenever throttle drops below a certain reading then you could have an issue with the TPS. That would show up as a distinct change in the voltage, and would likely need to be tested when the problem occurs in order to catch.

It's also possible you have a misfire or two or three. Moisture can condense in a distributor overnight and cause problems. A timing light connected to each plug wire could be used to determine whether one or more cylinders are misfiring but getting a visual of the cap during the time when the engine runs badly might be the only way to confirm it's moisture related. Any signs of carbon tracking or burn through could warrant replacement "just because."

I've had aftermarket O2 sensors cause problems by sending a voltage too soon after startup. Temporarily disconnecting the sensor or disabling closed loop would help eliminate the sensor signal as a potential problem.

Your EFI system does not provide a %TPS in the datastream. That value is simply calculated from the voltage. This can cause confusion when the calculation doesn't show 0% to 100% TPS as the throttle is moved. I don't think the TPS voltage varying by 0.02V with the throttle closed is an issue.

It runs smoothly or roughly when holding the gas to have a little higher rpm? If it's smooth then maybe the IAC is the issue but if it's still rough then it's most likely something else. If you open the throttle a bit that's the same as the IAC being open more so both should have the same effect.

Quote:

---

Originally Posted by 1project2many

IAC seems incorrect. IAC only adds air. You should be able to reduce throttle opening slowly and bring vehicle below 1200 rpm where it should run correctly without sputtering.

If there are no codes then you've likely got a problem with an un-monitored system, or you have a problem with a sensor reading incorrectly.

Does the exhaust smell rich when vehicle isn't running correctly? Is there a large amount of water dripping out of the exhaust? Is the tailpipe black? Those symptoms generally indicate a rich engine which would often require higher rpm to overcome. As it's temperature based I'd be looking at CTS and the temp signal. Yes, it tested ok once. But this is an intermittent problem, yes? You could have a broken wire at the sender, or the pins could be corroded or misshapen. Or the temp scale could be skewed in the sender.

If there is no sign of the engine running rich but it stalls whenever throttle drops below a certain reading then you could have an issue with the TPS. That would show up as a distinct change in the voltage, and would likely need to be tested when the problem occurs in order to catch.

It's also possible you have a misfire or two or three. Moisture can condense in a distributor overnight and cause problems. A timing light connected to each plug wire could be used to determine whether one or more cylinders are misfiring but getting a visual of the cap during the time when the engine runs badly might be the only way to confirm it's moisture related. Any signs of carbon tracking or burn through could warrant replacement "just because."

I've had aftermarket O2 sensors cause problems by sending a voltage too soon after startup. Temporarily disconnecting the sensor or disabling closed loop would help eliminate the sensor signal as a potential problem.

---

Interesting points. Reason I think iac is if it was stuck closed, it wouldn't get the air it needed to idle until I cracked open the throttle.

It does smell rich, perhaps I need to watch the injectors when it acts up, can they hang open when clogged?

I should double check my cap and rotor too, that's a good point. Even still, it runs smooth at higher rpm even when acting up, so I would surmise the ignition system is working correctly.

Seems my only hope is it was either a cold weather fluke or if it does it again it lasts longer enough to diag. Thanks

Quote:

---

Originally Posted by lionelhutz

Your EFI system does not provide a %TPS in the datastream. That value is simply calculated from the voltage. This can cause confusion when the calculation doesn't show 0% to 100% TPS as the throttle is moved. I don't think the TPS voltage varying by 0.02V with the throttle closed is an issue.

It runs smoothly or roughly when holding the gas to have a little higher rpm? If it's smooth then maybe the IAC is the issue but if it's still rough then it's most likely something else. If you open the throttle a bit that's the same as the IAC being open more so both should have the same effect.

---

Noted about the tps, thanks.

It runs smooth when you bring the RPM up, at least 90% so.

Quote:

---

Interesting points. Reason I think iac is if it was stuck closed, it wouldn't get the air it needed to idle until I cracked open the throttle.

---

It might help to have a more complete picture. The IAC is fully opened after engine shutdown then moved in to the park position. After startup the valve is slowly closed so the idle speed will decrease. If the IAC is working when warm but sticking cold then the idle cold should be high. If the IAC is not working warm or cold you'd generally see problems with stalling when warm, especially when shifting to drive or coming to a stop, and you might have no starts or hard starts. I've seen few IAC troubles with TBI but the six liter engines in our vans do have problems. Typically the engine idles slowly warm and won't start or starts and stalls when cold.

Quote:

---

It does smell rich, perhaps I need to watch the injectors when it acts up, can they hang open when clogged?

---

I've never seen that but I can't say they won't do it. What I've seen is a poor pattern. You can watch injector pattern with a timing light. Connect the light as you would to check timing and aim it at the injectors.

So, it'd run fine at say 800rpm as long as you use a little throttle? It is starting to sound like the IAC was simply sticking and didn't open on the first cold start.

Yes. Typically the IAC works intermittently during the day causing random idle speeds when the vehicle stops, most often below specification. A driver who is not cognizant would use the vehicle despite this. At the end of the day they would park the vehicle. The next morning the vehicle would start and stall or would not start. Manually opening throttle slightly would allow it to start and run. It would run well at any throttle opening but closing the throttle would cause a stall. If throttle was held and the vehicle allowed to warm up it could usually be driven although idle speed would not be consistent throughout the day and again, would tend to be below specification.

For years it was rare to change a GM IAC. Now it is much more common for us.