ankurdave

Hi everyone,

I cleaned my throttle body on my 1995 LS400 yesterday to try and fix a low idle when warm (450 RPM in D, 500 RPM in P) that was causing cabin vibrations. Now my idle is very high (1200 RPM in D, 2500-3000 RPM in P). From searching, a high idle after cleaning the throttle body seems to be a common problem, but most people's idle is not nearly this high so I thought I would see if anyone has suggestions.

Here's what I did:
1. I removed the throttle body, sprayed the inside directly with CRC Throttle Body and Air-Intake Cleaner, scrubbed it with a brush, and wiped it dry.
2. I also foolishly sprayed a fair amount of the cleaner into the air intake chamber using the same procedure, probably leaving pools of the cleaner in the inaccessible parts of the chamber.
3. I didn't do anything to the mass air flow sensor or the idle air control valve.
4. I left the battery disconnected for several hours to reset the ECU.
5. When I started the car back up, it idled at 5000 RPM!
6. After leaving the throttle body propped open for a couple hours to allow the excess cleaner to evaporate, the idle came down to its current state (2500-3000 RPM).
7. After leaving it overnight, it still idles at the same speed.

Here are some guesses:
- The ECU might just need more time to relearn the idle. The problem with waiting to see is that I'm worried the engine may overheat while driving. When I drove it for 10-15 minutes at low speeds, the engine coolant temperature from OBD2 reached 220° F instead of the normal 190° F. Surprisingly, the temperature sender displayed in the instrument cluster continued to show a normal reading.
- A common suggestion is to check for vacuum leaks. I checked all vacuum hoses several times and I don't think there are any.
- The idle air control valve could be stuck open. I don't think this is the case, since it was fine before (the ECU was able to step down the idle speed properly as the engine warmed up) and I didn't do anything to it.
- Cleaning the throttle body opened up a slight gap around the butterfly plate only detectable by holding it to the light. This gap was blocked by carbon deposits before. I think the gap is there by design.
- The throttle position sensor could have been damaged by the cleaner, or I could have switched the throttle and subthrottle position sensor connectors (my car has TRAC). I don't think either of these is the problem, because the OBD2 throttle position value is 0% when idling, as I would think it should be. (Edit: the factory service manual says it should actually be 7-11%, so maybe an adjustment is needed?)

M4rk

My immediate thought would be a faulty TPS reading supplied to the ECU. The FSM lists steps for diagnosing it based on resistance on the terminals. I would start by checking the connectors as you pointed out, and then checking the sensor itself.

This might also be of help: http://www.lexls.com/tutorials/EFI/tps.html

dicer

Hose missing and manifold leak, is my guess. Or bad tb gasket or not even or cracked that sort of thing.

ankurdave

Thanks for the replies!

dicer: I hope it's just a vacuum leak! I don't see an obvious one (picture of the engine below). If there is a leak, there are two places I would guess it to be:
- The throttle body gasket, which I just replaced a couple weeks ago but did spray with the cleaner before reinstalling yesterday because I got it dirty.
- The PCV hose that goes from the RH valve cover to the underside of the throttle body, since I was hasty putting that area back together. (Edit: This wouldn't make sense since it's upstream of the throttle plate.)
I can take the throttle body back off to check. I have a hard time with this since the TRAC throttle body is so long that the bottom two bolts are at risk of falling into the engine valley.

M4rk: The TPS might be a problem. My Bluetooth OBD2 reader shows that its reading moves smoothly when I press the accelerator pedal, but for the first 1/8 of its travel the reading stays at 0%, and it tops out at 52%. The app may be applying some range correction though. I tried running the engine with the TPS at both extremes of its adjustment position. The idle remained high in both, and the engine exhibited hunting/surging/looping behavior where it would periodically drop from 2000 RPM to about 1000. I also tried running the engine with the TPS disconnected; the idle still remained high. I don't have a multimeter to check the sensor, but I can get one. For the test, are you referring to Engine - SFI System - Service Specifications - Service Data, page EG-210? Image attached.

Overall view
TRAC throttle body and throttle position sensors. Note: I moved the hose clamps on the big IAC hose so I could adjust the throttle position sensor, but the problem occurs with them in place too.

ankurdave

While removing the throttle body, I found the problem! The throttle position sensor was being propped open by the hose clamp for the throttle body coolant hose. This was giving the ECU a partially-open throttle signal at idle, causing it to richen the fuel mix and (I think) open up the idle air control valve.

After adjusting the hose clamp so it doesn't interfere with the throttle position sensor, the car runs great, idling at close to 650 RPM in P when warm. In hindsight, the fact that the OBD2-reported throttle position stayed at 0% for the beginning of the throttle's travel was a clue.

Thanks again!

dicer

And that qualifys as a manifold leak, a huge one. As you see in the picture there is no way to prop the tps by itself as it is completely enclosed, so the throttle was open then.

ankurdave

I don't think that's true. The piece that was being propped up was a moving rotor-like part behind the black TPS housing. You can only see one side of it in the picture, it's the metallic part to the left of the black TPS housing and underneath the black valve. There's a small yellow matchmark on it.

I tried moving it by hand and it seems to move independently of the throttle plate in the open direction. Of course, in the closed direction, it internally rests against the throttle plate.

ankurdave

Never mind, I just went and checked and you're right. Propping the metal piece open does hold the throttle plate open. It just doesn't affect the throttle cable linkage on the other side of the throttle body, which is why the accelerator pedal felt normal (springing back to the top position rather than sitting partway down, as happens when cruise control is engaged).