Bad idle when its cold.. fast idle when its hot.

[Ecomike]

No,

If you hear vacuum sound when you apply the brakes it means you have a leak in the brake booster which is bad!

Have you replaced the master cylinder and bleed the brakes?

Good job you guy's on the advise! and tips, i learned a lot in my quest to trace down a high hot idle and a low cold idle problem. I also have an 89 and the brake booster does not seem to be working, the pedel is soft, goes down a long way before the brakes start to work. just replaced rear shoes, no big improvement.When you push down on the pedel shouldent you hear some vacume sound? How do you test a booster?

 

[mattbred]

Ok resistance numbers: (ground is the disconnected battery cable)

Backprobing with TPS connected:
A to Ground - 341.4 ohms
B to Ground - 0.4 ohms
C to Ground 2.65k ohms

TPS Harness not connected to TPS:
A to Ground - 0.369k ohms
B to Ground - 0.4 ohms
C to Ground - 50.5k ohms
A to C - 50.8k ohms
B to C - 50.4k ohms
A to B - 0.37k ohms

TPS itself, disconnected from harness:
B to C - 2.898k ohms
A to B - 4.35k ohms
A to C - 5.83k ohms

 

[mattbred]

I disconnected the IAC while the idle was slowing down and it stopped, and the idle was perfectly stable at 900 RPM's. I also noticed something else the other day,

When it's cold, my idle will rev up to 2000 RPM's or so, and drop down very slowly, IE itll take about 30 seconds to a minute to settle on 500 RPM's.

When it's hot, it'll rev between 1000-1500 on startup, and then quickly go down to 750 RPM's.

 

[88xjlover]

I disconnected the IAC while the idle was slowing down and it stopped, and the idle was perfectly stable at 900 RPM's. I also noticed something else the other day,

When it's cold, my idle will rev up to 2000 RPM's or so, and drop down very slowly, IE itll take about 30 seconds to a minute to settle on 500 RPM's.

When it's hot, it'll rev between 1000-1500 on startup, and then quickly go down to 750 RPM's.

Had the same problem on my 88 auto 4.0.
Changed TPS and went away. Also, check for vacuum leak at the vacuum canister down by the bumper. Mine was completely disconnected and I had to buy one.

 

[mattbred]

Any advice on them resistance values?

I think perhaps my wiring might be screwed up?

 

[Ecomike]

Loose unmounted, Brand New TPS on the ECU side itself, disconnected from harness is:

B to C - 2.360k ohms at Idle
A to B - 3.740k ohms at idle
A to C - 6090k ohms at Idle

Loose unmounted, Brand New TPS on the ECU side itself, disconnected from harness is:

B to C - 6.130k ohms at WOT
A to B - 3.740k ohms at WOT
A to C - 2.420k ohms at WOT

My working installed used TPS (disconnected from harness) is:


Pin A to B: 4360 to 4360 ohms, constant from idle to WOT
Pin A to C: 6580 (idle) to 2200 ohms at (WOT)
Pin B to C: 2330 (idle) to 6550 ohms at (WOT)

Ok resistance numbers: (ground is the disconnected battery cable)

TPS itself, disconnected from harness:
B to C - 2.898k ohms
A to B - 4.35k ohms
A to C - 5.83k ohms

Backprobing with TPS connected:
A to Ground - 341.4 ohms (This is good!)
B to Ground - 0.4 ohms (This is good!)
C to Ground 2.65k ohms ( I don't have test data on this one)

TPS Harness not connected to TPS:
A to Ground - 0.369k ohms ( I don't have test data on this one)
B to Ground - 0.4 ohms (If this is the harness, then this is real good)
C to Ground - 50.5k ohms ( I don't have test data on this one)
A to C - 50.8k ohms ( I don't have test data on this one)
B to C - 50.4k ohms ( I don't have test data on this one)
A to B - 0.37k ohms ( I don't have test data on this one)

My working installed used TPS is (with harness disconected):


Pin A to B: 4360 to 4360 ohms, constant
Pin A to C: 6580 (idle) to 2200 ohms
Pin B to C: 2330 (idle) to 6550 ohms

You said above you got:

TPS itself, disconnected from harness:
A to B - 4.35k ohms This is OK
A to C - 5.83k ohms which seems about 20% low
B to C - 2.898k ohms which seems about 30% high

Please refresh my memory, was/is this a new TPS or not?

Your TPS numbers above are about halfway between my bad used TPS and the good used TPS I am now using. They are the only 2 MOUNTED TPS sensors that I have installed data on. Recall that the mounted TPS has a more limited travel range (and more limted resistance range) that an unmounted TPS.

But based on your data I'd say the limited range of the installed TPS on yours is definately part of your problem.

 

[mattbred]

This is NOT a new TPS, just another I had laying around.

Keep in mind that our TPS levers are most likely in different positions, which causes different resistance amounts. I believe the only way to test this would be to take an un-mounted TPS and test it with the lever fully sprung (below idle) and with the lever fully extended (beyond WOT).

I'll try and get some figures tomorrow. I still don't think this problem is the TPS..

 

[Ecomike]

This is NOT a new TPS, just another I had laying around.

Keep in mind that our TPS levers are most likely in different positions, which causes different resistance amounts. I believe the only way to test this would be to take an un-mounted TPS and test it with the lever fully sprung (below idle) and with the lever fully extended (beyond WOT).

I'll try and get some figures tomorrow. I still don't think this problem is the TPS..

Then I would say its time to try another TPS, used or new.

That is correct, that is why I posted data for mounted and unmounted TPS sensors. My bad one read similar to your current one, but mine was worse than yours. By worse I mean its resistance range, versus a good used sensor was about 1/2 the range of the good used sensor, installed. I am using the good used sensor now with no further idle or power problems. So comparing your installed one to my working used installed one, and my bad used installed one, it says yours is marginal at best, and probably just far enough gone (but still sort of working) for it to be screwing with the ECU and its ability to idle properly.

But I agree, you should pull you TPS and test it unmounted and compare that to the new sensor data I just posted.

I was convinced for 2 years that my problem was NOT the TPS, and I replaced everything under the hood except the ECU, TCU, Engine and AW4 (Auto Tranny), and I still had to come up with test data to convince myself it was the TPS before I replaced mine. It made a huge differance, much more so than I previously believed possible.

Also, dont forget that if your TPS is the least bit noisy, that a slight variation of 0.02 volts from first idle to second idle, meaning repeatabilty of the idle stop voltage reading, can change the idle speed by as much as 140 rpm. I ran some tests (I posted in the long winded ReniX files thread) where I deliberatelly changed the idle TPS voltage and measured the change in the idle speed and it was something huge like 70 rpm per 0.01 volts as I recall.

While you are at it, don't you have a second used TPS? If so test it too! Seems like I recall you fixed one or 2 problems since you swaped used TPS sensors. If my memory is correct, you might want to retry the old TPS now that the other things are fixed! It may have been OK!?

 

[mattbred]

Alright well I'll test my other TPS and compare it to your data. If it's close I'll throw it in. See, I've had the exact same idle problems with both TPS's installed. So either they're both bad, or something else is bad.

Also, what EXACTLY is the formula to set the TPS idle voltage to? People have said anywhere from 0.82v-0.83v to 0.79v etc.

My input voltage from the ECU is 4.93 volts, and my TPS is set to 0.79 volts at idle.

 

[Ecomike]

I have read where people looked at setting the idle voltage and WOT voltage as percentages of the input voltage (your input being 4.93 V). It was something like .17 (17%) for idle and .83 (83%) for WOT, but there is only one position adjustment on TPS, so setting one, sets the other. I have also heard some using .16 (16%) and I think .18 (18%) of the input voltage for the idle. I have used 0.80 to 0.83 V, and settled on 0.82 V for mine. At 0.82 V, I am getting a hot idle rpm of about 750 rpm in park, and 575 rpm in drive which is helping my hot idle oil pressure with out overloading the transmission in drive (while not moving, stopped). At 0.80 I am getting lower idles, like 600 / 500 rpm and lower oil pressure in drive at idle than I want.

I would go with 0.82 V for idle, then test the WOT to see if is reaching 4.93*.83 = 4.09 V at WOT, power on but engine off! Mine was barely getting to 3 volts at WOT.

For that matter, why not test the WOT voltage before pulling the current TPS!?

Alright well I'll test my other TPS and compare it to your data. If it's close I'll throw it in. See, I've had the exact same idle problems with both TPS's installed. So either they're both bad, or something else is bad.

Also, what EXACTLY is the formula to set the TPS idle voltage to? People have said anywhere from 0.82v-0.83v to 0.79v etc.

My input voltage from the ECU is 4.93 volts, and my TPS is set to 0.79 volts at idle.

 

[mattbred]

Well my high idle returned.. After-driving idle is 1200 RPM's and after shutting if off and restarting it is 900-1000 RPM's.

However, when it was idling at 1200, I unplugged the IAC, then restarted the engine, and it idled at 900-1000, so the IAC is NOT the contributing factor to the high-high idle. I still think it's either the TPS or the wiring to the TPS, or even the ECU.

I noticed something odd though. When cold, I had my TPS set for 0.79 volts, ok, yet after hours of driving, we come back and I check it and it's more around 0.76 - 0.77 volts, with engine OFF ignition ON. hm.

I could order a TPS from that innernet site and see if it fixes the problem. If it doesn't, then I have NO idea what the hell.

-----------------------------------

Edit:

Loose unmounted, Old TPS, ohmmeter prods going right into TPS connectors

B to C - 2.175k ohms at Idle
A to B - 4.55k ohms at idle
A to C - 6.71k ohms at Idle

Loose unmounted, Old TPS, ohmmeter prods going right into TPS connectors

B to C - 6.8k ohms at WOT
A to B - 4.55k ohms at WOT
A to C - 2.09k ohms at WOT

 

[mattbred]

And for the one that was just in the jeep:

Loose unmounted, Second old TPS, ohmmeter prods going right into TPS connectors

B to C - 2.08k ohms at Idle
A to B - 4.5k ohms at idle
A to C - 6.53k ohms at Idle

Loose unmounted, Second old TPS, ohmmeter prods going right into TPS connectors

B to C - 6.53k ohms at WOT
A to B - 4.5k ohms at WOT
A to C - 2.077k ohms at WOT

---------------------------

Can you test your harness for me as well? I'd like to get your resistance values on that. thanks!
Also, how do these values compare to your bad one?

 

[Ecomike]

My TPS volatge was wandering like yours before I fixed the last of my ground wire problems. A noisy, wearing out, at the end of its life TPS can also do that. It has to be the TPS, The wire or connectors, or the ground for that idle voltage to wander, unless the supply volatge to the TPS is wandering, but that would also likely be caused by loose dirty connector pins.

That data on the old TPS sounds VERY good. Reinstall it and see if it does not solve the last of your problems, or at least change the problem, which might help the diagnosis work.

On second thought, I just looked at the other data you posted on the other TPS, they are nearly identical, and look very good.

I would focus on the wiring harness, harness connector pins (the female pins can be tightened and both M/F pins can be cleaned) and the ground. Have you checked the ground, that is between pin B on the harness and the battery negative post cable, while wiggling the wires of the harness all around the engine and the ground wires on the block and firewall to see if the ground resitance is changing any. For instance there could be a nearly broken wire under the insulation somewhere. This technique would help find it, if it is there.

At some point the C-101 bulk head connector pins also may need to be checked, but for now I would stick with the above first. You can wiggle the C-101 connector while watching the TPS idle voltage! It is a 4" long black plastic connector on the firewall just above the Brake vacuum booster. It is reported to be a source of electrical continuity problems as well (I have not had the pleasure of that problem yet with mine, only read about it).

In conclusion, looks like enough data to prove the TPS is generaly OK, except for the possiblity of a noisy, drifting idle volatge signal, for that reason alone I would still try the spare again, and even at some point buy and try a new one, as the noise test is difficult to do, requires a special, old style analog volt/ohm meter, and the noise test is not really conclusive, that I know of.

Then work on the TPS connector pins. You might try some houshold ammonia and water. Ammonia is great at removing copper oxidation off connectors and is easy to wash off with water.

One thing is for sure, the drifting idle volatge is the source of the problem!!!! So you know what to focus on. Forget the ECU, it is OK and is not the problem.

Well my high idle returned.. After-driving idle is 1200 RPM's and after shutting if off and restarting it is 900-1000 RPM's.

However, when it was idling at 1200, I unplugged the IAC, then restarted the engine, and it idled at 900-1000, so the IAC is NOT the contributing factor to the high-high idle. I still think it's either the TPS or the wiring to the TPS, or even the ECU.

I noticed something odd though. When cold, I had my TPS set for 0.79 volts, ok, yet after hours of driving, we come back and I check it and it's more around 0.76 - 0.77 volts, with engine OFF ignition ON. hm.

I could order a TPS from that innernet site and see if it fixes the problem. If it doesn't, then I have NO idea what the hell.

-----------------------------------

Edit:

Loose unmounted, Old TPS, ohmmeter prods going right into TPS connectors

B to C - 2.175k ohms at Idle
A to B - 4.55k ohms at idle
A to C - 6.71k ohms at Idle

Loose unmounted, Old TPS, ohmmeter prods going right into TPS connectors

B to C - 6.8k ohms at WOT
A to B - 4.55k ohms at WOT
A to C - 2.09k ohms at WOT

 

[mattbred]

Alright I'll check out the wiring tomorrow. I KNOW for a fact that all my ground points are 100% grounded, completley. I did a complete throughough check and shining a few days ago, even remade some wires and made DAMN sure that they're good. I'm getting 0 ohms from TPS B to the battery, as well. That was without wiggling it. HOWEVER I believe I am still getting odd issues on this:

Positive prod on voltmeter goes to pin C of TPS connector
Negative prod on voltmeter goes to pin B of TPS connector.
This gives anywhere from 0.76 to 0.79 volts, alright, yet
Negative prod on voltmeter goes straight to negative battery terminal
Will always give a different reading than if it's in pin B. It's strange.

Can you do the same test? Simply do this:

Put positive probe in pin C
Put negative probe in pin B
write down voltage.
Put positive probe in pin C
Put negative probe on negative battery connector
write down voltage.


I want to see if they're the same.

------------------
Edit:

What is the BEST way to clean electrical connectors with shelf products? IE first cleaning it, getting rid of all the bad stuff on it, and then protecting it? I want to do this to my pins. I have a couple auto stores near (Napa, gilberts auto, LordCO, Canadian Tire) and would like to do this to it.

 

[Ecomike]

I have about 0.78 ohms of resitance between pin B, 3 pin ECU side of the TPS wiring harness connector and my battery negative post, so yes the voltage reading is a little bit different when you use negative battery clamp for the ground.

If you turn the Jeep power off and use the lowest ohm scale on your meter and test resistance from Pin B of the wiring harness (3 pin, ECU side) to the battery negative cable clamp you should read something just slightly less than 1.00 ohm if your meter goes that low? As long as you have less than 1 ohm on that wire, Pin B to the negative battery clamp, even when wiggling everything under the hood, then that pretty much eliminates that ground wire as a problem source.

Sounds like your getting the hang of using that meter for testing and analysis!:yelclap:

The best way is to use a professional electronics product called De-Oxit. It is an electronics maintenance item. The company that makes it also makes a dielectric grease or lube to put on after you have de oxidized it. 5-90 posted some stuff on it here before, try searching De-Oxit.

The ammonia suggestion (handy household stuff) comes from my experience in metal finishing and chemcial engineering. Might even try Tarn-X, sold in the hardware and grocery section for cleaning jewelery and silver. It is great at removing oxides off of silver and copper.

Brake cleaner works great as the first step to get rid of any engine oil that has wondered into the connections.

Alright I'll check out the wiring tomorrow. I KNOW for a fact that all my ground points are 100% grounded, completley. I did a complete throughough check and shining a few days ago, even remade some wires and made DAMN sure that they're good. I'm getting 0 ohms from TPS B to the battery, as well. That was without wiggling it. HOWEVER I believe I am still getting odd issues on this:

Positive prod on voltmeter goes to pin C of TPS connector
Negative prod on voltmeter goes to pin B of TPS connector.
This gives anywhere from 0.76 to 0.79 volts, alright, yet
Negative prod on voltmeter goes straight to negative battery terminal
Will always give a different reading than if it's in pin B. It's strange.

Can you do the same test? Simply do this:

Put positive probe in pin C
Put negative probe in pin B
write down voltage.
Put positive probe in pin C
Put negative probe on negative battery connector
write down voltage.


I want to see if they're the same.

------------------
Edit:

What is the BEST way to clean electrical connectors with shelf products? IE first cleaning it, getting rid of all the bad stuff on it, and then protecting it? I want to do this to my pins. I have a couple auto stores near (Napa, gilberts auto, LordCO, Canadian Tire) and would like to do this to it.

 

[mattbred]

Alright. I have about 0.4 ohms or less of resistance between pin B and the negative battery post. So it can be safely ruled out as not a ground problem. And my imput voltage always stays the same. So I guess this makes it a TPS problem.

I'll try to find some of that De-Oxit stuff, thanks!

I have about 0.78 ohms of resitance between pin B, 3 pin ECU side of the TPS wiring harness connector and my battery negative post, so yes the voltage reading is a little bit different when you use negative battery clamp for the ground.

If you turn the Jeep power off and use the lowest ohm scale on your meter and test resistance from Pin B of the wiring harness (3 pin, ECU side) to the battery negative cable clamp you should read something just slightly less than 1.00 ohm if your meter goes that low? As long as you have less than 1 ohm on that wire, Pin B to the negative battery clamp, even when wiggling everything under the hood, then that pretty much eliminates that ground wire as a problem source.

Sounds like your getting the hang of using that meter for testing and analysis!:yelclap:

The best way is to use a professional electronics product called De-Oxit. It is an electronics maintenance item. The company that makes it also makes a dielectric grease or lube to put on after you have de oxidized it. 5-90 posted some stuff on it here before, try searching De-Oxit.

The ammonia suggestion (handy household stuff) comes from my experience in metal finishing and chemcial engineering. Might even try Tarn-X, sold in the hardware and grocery section for cleaning jewelery and silver. It is great at removing oxides off of silver and copper.

Brake cleaner works great as the first step to get rid of any engine oil that has wondered into the connections.

 

[Ecomike]

The De-Oxit is a little expensive and might be hard to find at the neighborhood store. While you are looking for it, if you have any ammonia (or go buy a pint bottle for $1) try about 1/3 ammonia to 2/3 water and dip the connectors in that for 60 seconds. Then rinse them with DI (bottled distilled water) and dry them with brake cleaner aerosol and see if that helps any.

I agree it does sound like the TPS.

Have you tried very small tip tweezers or needle nose pliers to tighten (decrease the diameter) of the female pins yet? If they can get a better grip when assembled (from tightening the female pin) they sometimes clean themselves a bit from the friction of assembly.

 

[mattbred]

I'll try tightening the connectors and checking out the big pin connectors sometime this weekend.

Also, can you test something for me?

Put the prods into pin B&C of the TPS harness with engine off, ignition on, and mark down your voltage. Then, bring the throttle wide open, and let go, and let it smack down back to idle. Write down the TPS voltage. Then keep doing this, letting it smack into idle slowly, then fast, then slowly, then fast etc at different speeds and see if your voltage always remains the same in the end. I know mine can vary greatly.

Hell, just attaching the prods to the TPS itself and checking resistance, it will be much different depending on the speed and such that I let the lever go. It doesn't always give back the same results.

 

[Ecomike]

I already have, mine is very repeatable at 0.82 volts each time.

One last thing, make sure the TPS is not moving, that the TPS mounting bolts are not loose before you order the new TPS.

I'll try tightening the connectors and checking out the big pin connectors sometime this weekend.

Also, can you test something for me?

Put the prods into pin B&C of the TPS harness with engine off, ignition on, and mark down your voltage. Then, bring the throttle wide open, and let go, and let it smack down back to idle. Write down the TPS voltage. Then keep doing this, letting it smack into idle slowly, then fast, then slowly, then fast etc at different speeds and see if your voltage always remains the same in the end. I know mine can vary greatly.

Hell, just attaching the prods to the TPS itself and checking resistance, it will be much different depending on the speed and such that I let the lever go. It doesn't always give back the same results.

 

[mattbred]

Alright well that's forsure me problem.

I'm positive the bolts were tight, however I will be buying some new ones as the heads on mine are stripped. What size are they? The napa said they didnt have any, and said its a special size 4 thread 75.

Also already ordered the new TSP! $50ish which is really good compared to $227. Thanks for that link! Where did you buy yours from?