1990 s10 cold idle problems


noble527
02-03-2011, 04:17 PM
i have a 1990 s10 4.3 tbi, 4x4 5 sp. On cold start up it idles rough, until warm, even spits and sputtes driving down the road, once it is warmed up she runs great. Did the egr cleaning, egr solenoid replaced some old vac lines, fuel filter, new gas... still idles rough when cold...Went to put in a new temp sensor for ecm, and it through a code, (code 15) So i put the old one back in and no code no engine light. Tried another new sensor and it did the same thing...could i have a bad ecm? any ideas would be great thank you!

old_master
02-03-2011, 08:02 PM
Temp sensor for the computer is located near the thermostat housing. It has two wires: Yellow and a black. Is this the sensor you replaced?

noble527
02-05-2011, 09:11 AM
Yes it is

old_master
02-05-2011, 10:14 AM
Seems odd that the new sensor set a code, and the old one doesn't. Temp sensors have a thermistor in them. A thermistor is a resistor that changes resistance when temperature changes. If you have a digital volt ohm meter you can check resistance of the sensor to see if it's accurate. This chart is for ECT and ACT sensors. When checking resistance, measure it when cold, warm, and hot to make sure it's accurate at all temps.

http://i561.photobucket.com/albums/ss58/cwhook/ECTACTresistancetable.jpg

noble527
02-05-2011, 02:13 PM
I will check, do you think a ECM could be a issue, or do u think that could be why it idleing rough?

old_master
02-05-2011, 02:56 PM
Probably not, these ECM's are just about bulletproof and rarely need replacement.

noble527
02-05-2011, 03:19 PM
So then if i have good readings what do you suggest? I'm gonna test it with old one in there..

old_master
02-05-2011, 03:24 PM
Could be a poor connection at the ECT sensor, or even a ground. There is a ground connection near the thermostat housing. It's possible it was disturbed when you replaced the ECT sensor. The data has to make it to the ECM for it to be able to process it correctly. Certain scan tools will read what the ECM sees and that's a more accurate way of testing. You can read it using a DVOM at the ECM connector. At the ECM: The yellow wire goes to connector "C" pin #10. The black wire goes to connector "A" pin 11.

noble527
02-05-2011, 04:53 PM
What do you think about the rough running when cold issue?

old_master
02-05-2011, 06:08 PM
When the engine is warm, the ECM primarily uses the oxygen sensors for determining fuel mixture. When the engine is cold however, the ECM ignores the oxygen sensors and relies heavily on the ECT sensor, (it also uses TPS, MAP and MAF sensors). The ECM processes the data from them and uses a predetermined fuel mixture. If the ECT circuit has a poor connection somewhere, the ECM interprets the high resistance as a colder than actual temperature and it will richen the fuel mixture. That might be what's causing the rough running.

noble527
02-06-2011, 11:06 AM
ok, something else to riddle you on, this morning went out to start and it was idleing up down just about to die, like it had been, i decided to unplug map sensor, she idled up and smoothly...or atleast better! Whats up with that, its a new map sensor with a new vac line ran to it?? Any ideas?

old_master
02-06-2011, 01:49 PM
The MAP sensor sends manifold vacuum data to the ECM. The ECM primarily uses that data for determining engine load. If the engine is surging, manifold vacuum is also surging, the ECM tries to correct it and ends up making it worse. When you disconnect the MAP sensor, granted the ECM isn't receiving data from it, but at least it's not trying to constantly correct for the eroneous data. If there's a problem with the MAP sensor, it will set a DTC in memory.

noble527
02-06-2011, 02:24 PM
Dtc in memory? what's that mean? Lol so again what are your thoughts? Generally I have read where people have unplugged sensors while running and if the symptom changes an gets better then that sensor is the issue... Right now I'm pretty much clueless. Thanks for your input i really appriciate it!

old_master
02-06-2011, 04:07 PM
Just like code 15 in your first post, DTC = Diagnostic Trouble Code. Disconnecting your MAP sensor covered up the surging idle, but the MAP sensor might not be the root cause of the problem. Disconnecting sensors, without proper diagnosis along with it, will not tell you if a sensor is faulty, that job belongs to the ECM. When troubleshooting DTC's, always start with the first one. After diagnosis and repair, make sure the check engine light stays off confirming that the repair is complete. If the check engine light comes on again, retrieve the code(s), diagnose and repair the first DTC and confirm the repair. Continue like that until the light no longer comes on. The reason DTC's are diagnosed in the order they are retrieved is because the first DTC can cause others to show up. Much like a burnt out bulb in a string of Christmas tree lights: if two of them burn out, it doesn't mean all of them are bad, one bad one can cause the rest to go out. You need to fix them one at a time, in order, until all are repaired.

It's possible that the ECT signal is not being received acurately at the ECM. The ECM has no choice but to "believe" all sensors are reporting acurately. It does "watch for" readings that are out of range. If it detects an out of range condition, it sets a DTC. For example: The actual temperature is 23 degrees, and the ECT resistance is 12,300 ohms, (perfect). If a poor connection in the ECT circuit has 40,000 ohms resistance through it, the ECM will interpret the ECT resistance to be 52,300 ohms which calculates to -20 degrees. The ECM will richen the fuel mixture. After the engine is warmed up, the air fuel mixture will still be rich and a rich DTC will be set, but nothing regarding the ECT because it's still within range. Now lets say that poor connection has 100,000 ohms resistance through it, plus the 12,300 ohms, now the ECT is out of range and a DTC is set. This is the importance of testing at the ECM.

Also, have you checked ignition timing with the EST, (Electronic Spark Timing) wire disconnected? There is a connector behind the carpet just above the passengers feet. Single tan wire with a black stripe, must be disconnected when checkng/adjusting timing. Ignition timing is controlled by the ECM but base timing needs to be adjusted when the ECM is not controlling it.

noble527
02-06-2011, 04:47 PM
ok i will do some testing, however i checked for engine codes, had none (ofcoarse until new sensor was installed) Installed old cleared codes none re appeared. I will check resistance of temp sensor. Spark timing? Where is that adjusted distributer? Also what needs to be done to adjust this? Thank you again, will report back with readings.

old_master
02-06-2011, 04:50 PM
You'll need a timing light to check/adjust it. Adjustment is done by turning the distributor.

noble527
02-07-2011, 01:06 PM
At 32 degrees it was 5.04 ohms

noble527
02-07-2011, 01:09 PM
At 46 degrees the new one was around 5.240 k ohms old one at 32 degrees was 5.02 k ohms

old_master
02-07-2011, 03:56 PM
With a little room for error, that's pretty close. Was that reading taken at the ECT sensor or at the ECM? Might want to check it at operating temperature also.

noble527
02-07-2011, 09:15 PM
tonight I set the timing, it was a little off. Put the new temp sensor in for s and g's and no engine light came on... It kicked in to high idle which it never did, and ran great. However it kept idleing high and my temp gauge was being wacky. So i pulled it back into garage and pushed in on the connector (on temp sensor) and it idled down...let pressure off of it and it idled back up, still no engine light. So i pulled connector off and bent in the tabs inside of it to make it fit tighter over the pin, and it seemed to help it not idle high. But when i took it for a test drive again, it felt like it wasnt running as smooth as it just did...so im letting it set over night and we will see if it idles high in the morning...just thought id give u a reply. I also tested resistance in the wire from sensor to ecm, harly any resistance ( just checking conitnuity) also checked black wire to ground, and no resistance...so wires are good...

old_master
02-07-2011, 10:49 PM
Ok good job! Couple of things.... The temp gauge on the dash uses a different coolant sensor, single wire sensor and screws into the cylinder head between 2 spark plugs. Could be on either side of the engine but usually on driver side. Temp gauge going wacky could be an air pocket in the cooling system and if that's the case, it could also affect the ECT sensor. It sounds like you have a bad connector at the ECT sensor if touching it makes a difference. I think Advance auto parts can get them, take the old sensor in to them and ask. It will be the connector with about 6" pig tail wires on it. Solder them on, don't use crimp connectors. Let me know what happens.

noble527
02-08-2011, 06:36 AM
This morning when I started truck it didn't not idle up and idled rough again, I am going to get new end and pigtail and rewire it.. I knew they where differen sensors thats what confused me lol. I will put that on and let you know may be acouple days... Thank you

noble527
02-08-2011, 06:44 PM
Put new connector in, still doesnt work right still rough idle. Don't understand how i could get it to work right one time... Can i run a new signal wire to ECM by itself? And also a separate ground?

old_master
02-08-2011, 08:48 PM
Sounds like a poor connection somewhere. Yes, you can run new wires. At the ECM: The yellow wire goes to connector "C" pin #10. The black wire goes to connector "A" pin 11. Do not connect the black wire to chassis ground, run it right to the ECM.

noble527
02-08-2011, 10:08 PM
ok im going to run new wires to ecm, why is it you can ground to chassis though? Just curiosity not questioning just wondering. When i tie into the ground on the ecm, is it something wherei can cut the wire and connect to ecm, or do i need to tie old wire from harness in with it? I hope that fixes it, just frusterates me that it halfway worked and ran 10 times better.

old_master
02-09-2011, 04:30 PM
The black wire is "signal return" for the ECT sensor, MAP sensor and TP sensor. They all splice into it in the harness and then goes to the ECM. If you cut the black wire at the ECM and run a new wire for the ECT, the MAP and TP sensors will not supply data to the ECM. If you're going to run new wires for the ECT, Solder them close to the ECM terminals. Do not cut the black wire, just solder your new wire on to it. Then cut the original yellow wire near the ECM to eliminate the possibility of a short to ground in it. (The yellow wire runs from the ECT to the ECM, nothing else is connected to it.)

noble527
02-09-2011, 08:23 PM
ran new wires tonight soldered them both, still same problem, im getting very frusterated now, what do you think i should do now? Still no codes though with new sensor..

old_master
02-09-2011, 08:33 PM
Disconnect the ECT sensor and the throttle position sensor. Check resistance on the black wire from one connector to the other. Should be less than 5 ohms. Then check it from the throttle position sensor connector back to the ECM at your solder connection, should also be less than 5 ohms.

noble527
02-09-2011, 09:33 PM
.01 ohms of resistance so good there

noble527
02-09-2011, 10:00 PM
Unhooked map sensor again and ofcoarse idled perfect I tried looking at egr valve it's brand new but how do u know if it's working?

old_master
02-09-2011, 10:02 PM
On a cold start up, pull the air cleaner cover and see if both injectors are spraying an even cone shaped pattern.

Does the air cleaner have a spot for a tube to run down to the exhaust manifold? It would be on the underside of the snorkel, toward the inlet end of the air cleaner. Tube is a little over an inch in diameter.

old_master
02-09-2011, 10:06 PM
.... I tried looking at egr valve it's brand new but how do u know if it's working?

The EGR valve shouldn't open until the engine is at operating temperature. You can disconnect it and see if it makes a difference, but I doubt it will.

noble527
02-10-2011, 07:37 AM
sprays nce and even, no tube that goes to exhaust manifold...

noble527
02-10-2011, 07:39 AM
i will check spray pattern again, when i leave from work, right now its -11 here, so sitting 8 hrs i will see what it looks like...

noble527
02-10-2011, 12:11 PM
Well went out started it up, look like one side was sprayin ok the other was spraying and look like dribbling a bit also

old_master
02-10-2011, 03:18 PM
Sounds like you've got a dirty injector. Shouldn't be any dribbles or drips, just a fine cone shaped mist from both injectors. There's probably something stuck in the director plate or spray tip. If fuel pressure is low, it can cause the injectors to dribble or drip but, most likely both injectors would do it, not just one. Fuel pressure should be 9psi to 13psi, might be worthwhile checking it before spending $$$ on an injector. You can try a bottle or two of injector cleaner in the fuel tank; if that doesn't improve the spray pattern, you'll need a new injector. Replacing the injector is not difficult at all.


http://i561.photobucket.com/albums/ss58/cwhook/Rochesterfuelinjector.jpg

noble527
02-10-2011, 03:35 PM
I've done the bottles of injector cleaner... Explain to me how to test fuel pressure and what I need and also checking the injector... Would a faulty injector though cause it to run rough when cold and better warm? Is there any tests I can do on those? Thanks for the input!

old_master
02-10-2011, 03:58 PM
There is a schrader valve on the fuel rail next to the injectors. You'll need a pressure tester capable of testing GM TBI systems, possibly Advance or Autozone rents them, you'll have to check.

To easily and efficiently burn fuel, it must be atomized and vaporized. Atomization takes place as the fuel exits the injector under pressure. Vaporization is done in the intake manifold from the heat of the engine. When the engine is cold, vaporization is almost nonexistant, making complete atomization absolutely necessary. If fuel drips out, it's not atomized enough and requires more fuel for the engine to run properly. The ECM is programmed to provide a certain air fuel mixture on a cold start up, depending on data from the ECT. The ECM has no means of monitoring the mixture until closed loop is attained. When in closed loop, the ECM receives data from the oxygen sensors and adjusts the fuel mixture. By that time, the intake manifold has warmed up and is vaporizing the air fuel mixture. The result of a dripping injector is a lean misfire, (cold) usually on every other cylinder in the firing order.


In post #31 I asked about a tube to the exhaust manifold: The tube directs warm air to the throttle body to help with vaporization and to help prevent ice up. Some older vehicles have it, some don't, couldn't remember for sure on yours. BTW, that's one reason cold air intake systems cause more problems than they're worth.

noble527
02-10-2011, 04:35 PM
OK, i will check fuel pressure tonight and see what i come up with, that makes sense on the fuel delivery, is it a big deal to change out an injector, any gaskets i need to get if thats the case? Any way of cleaning them besides the fuel injector cleaner, or is it just easier to buy brand new. My truck does NOT have the tube hooked to air cleaner...thanks again, will report back with results.

old_master
02-10-2011, 05:08 PM
The throttle body must be disassembled to replace the injector, (not a big deal) and you will need a couple of gaskets. The auto parts store will be able to tell you what gaskets you need and probably have them in stock along with the injector.

noble527
02-10-2011, 05:41 PM
Looked around couldn't find valve? Any pics?

Add your comment to this topic!