1992 LS400 - Replaced starter - now runs great cold but surges and bucks after reaching temperature

[threepwood]

So, the starter has been replaced (by me) and it starts great, even appears to have fixed what I thought was a battery issue but I now know it was the starter motor.  Unfortunately, even though I am 100% certain I put every thing back (there were 45! connections that I labeled and am certain I reconnected) it has a new problem.  Cold, it runs great.  Accelerates perfectly, maybe better than before.  But when it reaches temperature it stumbles and misses badly whenever you try to accelerate.  It did not do this before the starter replacement.  At first, I assumed that since I cleaned the EGR pipes (which were solidly plugged with carbon deposits) while i had it apart that the EGR valve was now dumping to much gas back in the intake.  So I plugged the vacuum hose to the valve, no change.  Then I put a solid blank plate between it and the intake, just in case it is leaking, no change.  Then I decided to try and trick the computer into thinking it was always cold.  I pulled the connector off the temperature sensor under the right hand ignition coil and put a 2k resistor accross it.  This also did nothing and was quite surprising, as the temp gauge on the dashboard still functions normally, showing the car heating up normally.  So the temp must be coming from some other sensor???  I just don't know of any other system that is bypassed cold and kicks in hot. New problem.  With foot firmly on the brake and car hot and running, would NOT come out of park so I could leave the store I was at.  Had to use the override button to unlock the shifter.  I replaced the ECU with a spare one that I had replaced the capacitors in long ago and this problem went away.  I think this car is just begging to go to the junk yard. Next step, maybe pull the thermostat so it actually is forced to run cold all the time ???  Any thoughts would be appreciated.

[RX400h]

Could it be a faulty O2 sensor? If it causes the engine to run rich, that would be a problem after the engine is warmed up.

 

[threepwood]

Thanks.  I guess I'll just pull the thermostat and see what it does.  I can't read any codes because, for the second time, the clock display has gone blank.  

[RX400h]

I was thinking more of the o2 sensor(s). A malfunctioning thermostat that is stuck closed would cause overheating, whereas a stuck open thermostat would probably not cause drivability issues after the engine is up to operating speed.
 

[threepwood]

Sorry, I wasn't clear.    I'm going to remove the thermostat to force the car to run cold and prove it is engine temperature dependent.  If it still wacks out running cold then it is time and not temperature dependent.  I don't think it is, but this would prove it.   The only problem I have with the O2 sensors is I don't understand why changing out the starter would cause the O2 sensor to fail and it did not do this before the starter replacement.  But the O2 sensor is a good thought and I am keeping it on my list of possibilities. Thanks again for your comment.

[RX400h]

You're welcome, Guy.

[threepwood]

Ok, so let me tell you what I have done.  I fixed the check engine light, it is working perfectly.  Disconnect the temperature sensor to the computer and the light pops on,  replace it, light clears.  Therefore I believe the O2 sensors are not the problem because there is no error code.  So I tore the engine down to the starter again and all i found wrong was a water leak from the new gaskets I used on the intake manifold, possibly into one of the cylinders but definitely on top of the engine in the bay where the starter motor lays.  I did not use a sealant the first time as the gaskets I disposed of did not have a sealant on them.  I, once again, put everything carefully back together but this time using a sealant on ALL gaskets.  And, because I had a new spare, I replaced the throttle position sensor with a new one.  For a week, the car ran perfectly, hot or cold with one exception.  The idle speed in neutral was 2400 rpm.   In drive it would drop to 1000.  Today, for no apparent reason, at 40mph hot 2 miles from the house the engine just quit.  I tried to restart from neutral while rolling but it would crank but not start.  I was able to pull onto grass, put it in park, turn ignition switch completely off and back on and restarted the car!  I then drove it home, left it in the driveway idling to go get the mail and by the time i got back it had stalled again.  but from park, recycling the ignition switch, it started right up again.  I am at a complete loss to explain this.  Any thoughts would be greatly appreciated.

[RX400h]

High idle speed is often indicative of a vacuum leak. If that leak is too big, the engine could stall. I would check for cracked tubing or even an intake snorkel that is cracked. Rubber gets very brittle as it loses its plasticizer. Mechanics sometimes use a smoke generator to locate leaks.

[threepwood]

Ok, now it is getting even more strange.  I did find the hose between the ISV (idle speed) and inlet snorkel cracked, so I replaced it.   Idle came down to a slightly more reasonable 1400 rpm in neutral. still 1000 in gear.  But 30 minutes out car still died and this time would not restart no matter what I did.  Temperature was normal warm, where it always runs.  Left me stranded a mile from home so I just walked home, waited an hour, then went back to try again.  Would not start no matter what I did, but that new starter motor really is getting a work out.  Now here's the strange part.  On a lark, I pulled the temperature sensor wire to the computer.  Car started up immediately, check engine light on, and I was able to drive it home and it ran fine !!!  I believe once you pull the sensor the computer runs on a backup program.  So somehow, it appears the temperature sensor is shutting down the car even though it is not overheated ??????????????

[threepwood]

PS - I already changed out the ECU with my spare and it didn't make any difference.

[paulo57509]

SWAG:  Did you bleed the cooling system after refilling it?

[threepwood]

Yes I did.  It's takes a 17mm socket and the pipe plug is just above the thermostat.  Unfortunately, I've made that mistake before :).  And when I did, back many years ago, it did overheat.   So good thought, but not the current issue.  I'm running the car right now with connection disconnected and plan to get a resistance reading on the sensor after it warms up.  I'll post it later. thanks.

[threepwood]

Ok, so I measured the resistance of the temperature sensor with the engine hot.  It measured between 260 and 310 ohms.  I think this is normal.  It did drop to 150 at one point but I think that was my loose wires and not real cause the engine kept running.   So I decided to check the check engine code just to make sure it was the sensor disconnected code.  It was, 22.  But to my surprise it had a second code, 15.  15 is some kind of igniter fault which could, according to my book, shut down the fuel injection system., and that is possibly what is happening, as the engine just shuts off.  I guess I'll be checking for loose wires this afternoon at the igniters and coils.

[threepwood]

This is my final post on this issue so that hopefully, no one else will have this problem. After much reading of the maintenance manual I went to measure the air flow temperature sensor resistance today (part of the Mass Air flow connector) and found that apparently when I reconnected it after installing the starter one of the 2 pins bent over, not making contact or maybe making very poor contact.  Why this did not throw a code 24 is beyond me, but it never did.  After straightening the pin and reinserting it several times to make certain it had good contact the car has been running perfectly, hot or cold.  My take on this - electronic fuel injection is to bleeping complicated, give me back my holley or carter quadrajet carb :) :) :)

[GillCoop]

On machines of this type it is worth putting a high-quality car cooling system.