What's new
CK5
Register an account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members.

That damn TBI idle variance - FIXED

C

Cricket

3/4 ton status
Joined
Nov 29, 2004
Posts
8,217
Reaction score
0
Location
Colorado
For the last 3 weeks I picked up a stall at idle, when warm only, in park or neutral only. It runs fine in gear.

It will idle low, than high, than lower, than high, than almost stall, than recover, then stall. I replaced the IAC and while it quit stalling it still wants to idle funny. I tried plugging my MAP vac line and the damn thing ran smooth.

What the heck does that mean? Normally the plugged MAP should cause it to stall shouldn't it?
 
Disconnecting MAP should force the ECM into preprogrammed values. If it works right when the MAP is disconnected, something vacuum related in the engine (or MAP) is a likely problem. Don't forget the wiring portion of it, if the correct voltage isn't being relayed based on the pressure seen by the MAP, there would be issues.

Warm only tells me that it's either some sort of switch that's solely temp based acting up, which I can't think of any on a TBI motor, or that again, something not being looked at as an input to the ECM in open loop is the issue.

Could also be a fueling issue, too much, too little, based on engine temp. O2, screwed up injector wires, etc.

Done a good visual on the engine? Plugs, wires, cap, rotor all look good? All vacuum lines?
 
Plugs, cap, rotor all new. Wiring is a new painless harness. Vac lines to both EGR and Map are new. The 02 is located just past the collector on the Hedman Elites although it almost sounds like it's cycling between open and closed loop - don't know if it's possible for it to drop in and out that quickly.

I haven't checked the voltage at the Map plug. No pinched or frayed injector wires, injectors have been cleaned by AFF recently. I'm running a 31 gal TBI tank in the bed all stock with steady pressure of 13.5 psi.

What I need to do is break down and buy a laptop so I can run Winaldl and see what's going on. :doah:
 
It'll cycle open and closed loop very quickly, typically based on engine RPM. Watched my scanner with it hooked up, unheated o2 drops out of closed loop but immediately goes back with the slightest throttle. Doesn't hurt anything.
 
In that case it might be worth moving the 02 sensor into the header, just to see if there's a change. I have it on my list of things to get done. If it makes a difference I'll post up. The Map issue bugs me, I'll keep working with that and see if I can pinpoint the problem. I tested 2 Maps and both did the same thing so that glitch is being generated somewhere else.
 
I should go search but I'm lazy. Are you running a Painless harness, and if so, did it have you ground the P/N switch? Outside chance this is an issue, but curious.
 
The painless harness wiring to the park indicator switch consists of one Orange/Black wire and one Black/White wire.

1. If you do not use a VSS then you connect the two wires together - BUT this will stop the computer from controlling the EGR.

2. If you have a VSS (which I do) then you can do one of two things.

If you have a GM column you can use the combination switch part# 15679680 and wire it as shown in figure 6.4. If you wire it as shown in figure 6.4B then you simply cap off the black/white wire.

If you are going to use the recommended switch for the computer's benefit ONLY when the car is in PARK, you may or may not want to use the black/white wire. The other end of the black/white wire is already connected to ground through the harness.


I have the stock GM column switch and it is only using the Orange/Blk wire with the Black/Wht wire capped off and taped. It is not connected to the Orange/Blk wire. It seems awfully coincidental though that I have a funky idle and a code 32 which has been there since day 1. :thinking:
 
Last edited:
Where I'm going is that the ECM "learns" IAC from the VSS as I understand it.

Code 32 is EGR, and according to the GM chart (http://www.thirdgen.org/techboard/attachment.php?attachmentid=69562) you can't have that code unless the MAP doesn't see the drop in manifold pressure as the valve is commanded to open. If you aren't running EGR, that is why you have that code.

This is why I don't like Painless. No offense on that intended, but this gets frustrating trying to figure out what they were trying to work around. :( Are those words yours or theirs? If so, then they are so dumb as to still tout that you don't need VSS, and EGR is the only downside of not running it? Eh, don't need a response to that I guess, steers us off topic.

But I digress, and step off my soapbox. If your ECM knows when the vehicle is in park/neutral or drive via the switch, then that's not your problem. But since this is only in park/neutral, it certainly sounds tied to the switch.

Still would be looking at the MAP voltage, and see if it is within spec while idling in gear, since the surging in neutral will skew the voltage. If you have no EGR, then you might as well ignore the 32, until you burn a chip, it's going to be there.

Lastly I might try the minimum air adjustment, (instructions have been posted here and at thirdgen) see if that does anything for you. It doesn't cost anything, nor can it hurt.
 
That quote is from the Painless manual sitting on my desk, abbreviated a little by me.

This swap is complete and includes all sensors and VSS. No shortcuts.
 
OK so we've got an EGR, but still getting an EGR code.

So the ECM in your setup KNOWS when the vehicle is in park/neutral, or drive?
 
Well Dorian, that is the question. :D

It runs fine in gear at idle, it runs fine in town and on the highway except for throwing the code 32 at cruise. It doesn't always throw the 32 and sometimes the code will come on and then go off 30 minutes down the road. The EGR diaphram moves and I can stall the truck out by manually compressing it. The EGR solenoid holds vacuum.

The idle in park and neutral oscillates up and down when cold but evens out after a few minutes. Then after driving for 20 minutes, if I stop and put it in park the idle oscillates up and down then tries to stall.

When the MAP is plugged the idle evens out.

Every last sensor on this thing is new. The vac lines are new. The TBI is rebuilt. The only component that has not been replaced is the distributor.
 
If you can trace the wiring, figure out if the ECM knows it's in park/neutral or drive. Should be pretty easy, assuming you are using the 1227747 ECM, the pinouts are over at diy-efi.org (gmecm, I think on Twiki now) and probably elsewhere. Find out which wire is the PN on the ECM, find out if the way your switch is connected is giving that info to the ECM, and take it from there.

I suspect the thing thinks you are in drive all the time. If it assumed park all the time, the EGR would never function.
 
Here's my $.02. I have the same painless harness, When I did mine, the column wasn't the correct one to accept the p/n switch that it called for. So I left the wires unhooked, both of them. I needed the A.I.R. to work, and with them grounded to each other it wouldn't. I had it at the Fuel Injection guys here in town, which is what they do, and they said it wasn't a problem. The only thing that it does is tells the computer when you go from park to drive, so it can compensate the idle in that split second, for the load of being put in gear. I have never gotten any codes for it, and haven't had any problems related to it. I would suggest you unhook your wire at the p/n switch and see if your problem goes away.
 
Computer needs to know park vs. drive for a lot more things than that initial idle compensation.

I'm not arguing, just sick of the so called "experts" (including painless) continuing to say that it's ok to run without a P/N switch, or saying it's only an emissions deal. Don't take that the wrong way, I'm not singling anyone here out. jagsthatrun.com felt so strongly about this that they published their VSS section of their book online.

*With that said* I know plenty of people run their vehicle without VSS and/or a P/N switch hooked up, and no problems. However, there ARE problems associated with it, and without the setup being per GM specs, you can't discount problems that you normally would. There's no reason someone should have guess if there issue is caused by incorrect or bad wiring, especially at something as basic as the P/N switch.
 
Finally nailed the problem down to my heated 02 sensor. I simply unplugged the keyed ignition wire and the damn thing runs like a clock now. I unplugged and replugged that wire twice to confirm the pattern.

Refresher: Once in closed loop and warm the idle would drop down to 550 or so, run smooth for 5 seconds and then drop suddenly and jump back up or "Blip". After 30 seconds to 1 minute it would stall.

Do you think it's possible once I switched from the longer tube Dynomax headers to the much shorter Hedman Elites that the heated 02 was causing a lean condition? It was obviously affecting the ECM. I was under the impression that the heated 02 shut off after the exhaust temp reached 600 degrees.
 
Last edited:
Heated 02 sensor should have no effect on your issue. There are many vehicles that come factory with short exhaust manifolds that run a heated 02 sensor and have no problems. The heated 02 sensor stays heated until the power is switched off to the 02 sensor has nothing to do with temperature.
 
Last edited:
4X4HIGH said:
Heated 02 sensor should have no effect on your issue. There are many vehicles that come factory with short exhaust manifolds that run a heated 02 sensor and have no problems. The heated 02 sensor stays heated until the power is switched off to the 02 sensor has notrhing to do with temperature.
Click to expand...

Well something is changing in relation to the keyed ignition side. Unplug the power to the 02 heat coil and it holds idle, plug the power back in and it wanders then stalls.
 
Spare ignition slot in the fuse box. The 02 heats up with the key on and stops heating with the key off, I've pulled it out of the exhaust twice to check function.

I've had the heated 02 for a good year and had no problems with park/neutral idle until several weeks ago. As far as I know it could be a problem with air/fuel ratio or 02 sensor position in the exhaust pipe.

The only two tests I've done that effected the idle variance where to plug the MAP line which smoothed it out but it still stalled or to disconnect the heated 02 power which smoothed it out with no stall.
 
maybe the o2 is shorted internally, so when the heater power wire is plugged in, it is feeding through to the sensor, and giving the computer some crap readings.
 
You must log in or register to reply here.

Latest Posts

Top Bottom