legacycentral bbs

I've had an on-going issue that is driving me freaking nuts.

Car starts and idles mostly fine but when you drive it the idle wont stay down where it should, it gets a lot worse after hard acceleration and/or getting into boost. Here is a vid of the issue:

http://vimeo.com/86382477

I've done a smoke test (not a pressurized smoke test) but a passive one to check for leaks and I found one very small leak, fixed it and it didnt effect anything.

In the past I just blocked off the IAC and adjusted the idle screw. Which seems to work ok once its warmed up but I've been driving my Lincoln the past few years so now having to deal with that crap kinda annoys me. (Btw the IAC is completely hooked up and functional again)

I've cleaned the MAF, made sure there arent any leaks and I'm getting really fed up with this. I'm not sure if I had this issue when I was still running an ej22t, I think I may have been though.

I've got the top of the engine torn apart. The IAC is all plumbed in properly. I'm replacing the PCV valve right now and I'm going to re-do all the PVC lines as I cracked two of the hoses while removing them. I was thinking maybe the valve was getting stuck after being pressurized from boost. But alas, the old valve seems just fine and isnt stuck.

I have no CELs and like I said, when its driving the car pulls hard and drives fine. Just idles high. (And yes I've readjusted the idle screw back down to where it should be and I've calibrated the IACV back to where it should be.)

Could it be that something is wrong with the TPS and its sending the wrong signal to the ECU so the ECU is telling the IACV to open and allow more air in raising the idle?

While not the exact same symptoms, a bad (or slow responding) engine coolant temp sensor can lead to symptoms similar to a bad TPS or AFM.

I've considered that... But since the problem really shows itself after acceleration... Wouldnt a bad Temp Sensor be more erratic?

Have you checked the TPS's resistance and adjusted idle switch? The idle switch adjustment is the most important.

http://main.experiencetherave.com/subar ... sting1.jpg
http://main.experiencetherave.com/subar ... sting2.jpg
http://main.experiencetherave.com/subar ... sting3.jpg

Are there any codes in the ECU?

Does your throttle body have this bypass screw on the top?

http://main.experiencetherave.com/subaru/images/tb

You may also want to check that the bypass or blow off valve is operating correctly.

Mine does not have that screw. I'm going to buy a multimeter tonight so
I can test it.


No check engine lights or codes.

I think I've checked the bpv. I can't blow any air through any of the inlets or outlets on it including the vacuum line port.

I have the same problem man, its irritating I've concluded its IACV or some kind of voltage deficiency.

Well I've done everything conceivable to the IACV so I doubt it has anything to do with that at this point.

Is the butterfly plate on the TB moving freely and closing fully at 0% throttle? I would check that the throttle cable is moving freely without binding.

Both of them are moving fine

Ummm... I'm sorry for being special... But how the hell do I use this thing?

Disconnect the electrical connector from the TPS. You may wish to remove the TPS, but not really necessary. If you remove it you WILL need to adjust the idle switch.

Using the above pic, identify all the pins. Double-check.

Set your meter to OHMS, 20K. Place both the meter leads together and make sure the meter reads zero, or very close.

Put the meter leads on pin #2 and pin #3. The reading should be about 12000 ohms. It might read as 12.xxx depending on how your meter works. If it is too far off, then the TPS is bad.

Next, put the meter leads on pin #2 and pin #4. With the throttle closed it should read between 10000 (10K, 10.xxx) and 12000 (12K, 12.xxx). With the throttle open is should drop to between 3000 (3K, 3.xxx) and 5000 (5K, 5.xxx). Also, the change in numbers should be smooth and rise gradually and in proportion to the throttle movement. Your looking for spots that jump up or down suddenly, usually returning to the 'normal' curve just as suddenly. Out of range readings or any bad spots indicate a TPS that is worn out.

Do a final check with the meter between pin #3 and pin #4. This is the same as the previous test, but the results should be reversed....Closed reading 3-5K and open reading 10-12K. Treat it as a double check to the above test to verify what you saw.

Leave the red and black wires where they are. That must be an older multimeter, because it doesn't have autorange. No problem.

Ohms.
Ohms is checking for resistance, as in, how much signal is lost between the black and red wire when you test something. The different settings under Ohms is your anticipated range. Eg. If you anticipate a sensor has a resistance of 50 ohms, then set it to 200 (this means it displays 0-200 ohms). If you are looking for something that has a resistance value of, say, 1k Ohms, you would set it to 2000 (0-2000 ohms, but it shows ohms in the 3 and 4 digit range most precisely). And so forth.

When the service manual says something should have 1 megaOhm resistance, that just means there is no connection between the two pins, and hence it is "open."

Also under ohms, is a most useful setting, the black one. That tests for continuity, as in, is there a direct connection between two points. Great for testing for broken spark plug wires, etc etc.


VDC
The other settings you will commonly use on the car is VDC, or DC voltage. VAC is household AC current, and won't work here. With VDC you select the voltage range your car will likely have. 20 in this case. Or at least I hope your car doesn't put off more than 20V.


I would recommend getting a multimeter that has Autoranging, so you don't have to select the anticipated resistance or voltage range: it does it for you. This one will work, but is more challenging when you are learning how to use it lol.

I really don't know what you do or don't know about electronics, so don't take this as me talking down to you, Spence. Just trying to be helpful. I am a 1st year electrician apprentice and don't claim to know everything, yet. That comes in my 3rd year I hear, lol.

edit: Phyrram ninja'd me, lol. He is a mad electronics guru!

First off, I bow before your electrical epic-ness and thank thee for thy expertise greatly.

Secondly here are my readings.

1st - Testing pins 2 and 3.

Results - 10.06, doesn't change with throttle opening or closing

2nd - Testing pins 2 and 4

Results - 9.37 when closed, goes smoothly to 2.25 at wide open


Last test - testing 3 and 4

Results - 1.02 when closed goes smoothly to 8.37 at wide open




My assumption is that my TPS is bad and/or calibrated improperly.

PhyrraM wrote:

Disconnect the electrical connector from the TPS. You may wish to remove the TPS, but not really necessary. If you remove it you WILL need to adjust the idle switch.

Using the above pic, identify all the pins. Double-check.

Set your meter to OHMS, 20K. Place both the meter leads together and make sure the meter reads zero, or very close.

Put the meter leads on pin #2 and pin #3. The reading should be about 12000 ohms. It might read as 12.xxx depending on how your meter works. If it is too far off, then the TPS is bad.

Next, put the meter leads on pin #2 and pin #4. With the throttle closed it should read between 10000 (10K, 10.xxx) and 12000 (12K, 12.xxx). With the throttle open is should drop to between 3000 (3K, 3.xxx) and 5000 (5K, 5.xxx). Also, the change in numbers should be smooth and rise gradually and in proportion to the throttle movement. Your looking for spots that jump up or down suddenly, usually returning to the 'normal' curve just as suddenly. Out of range readings or any bad spots indicate a TPS that is worn out.

Do a final check with the meter between pin #3 and pin #4. This is the same as the previous test, but the results should be reversed....Closed reading 3-5K and open reading 10-12K. Treat it as a double check to the above test to verify what you saw.

And thanks Mike for the walk through on how to use it... As far as my electronics knowledge goes its very limited. I can handle mechanical stuff (because lets face it, anyone can follow directions and turn a wrench) but electrical escapes me.

Spencer, Those readings are close enough that I wouldn't automatically dismiss the TPS. The ECU is actually reading voltages and depending on the ECUs 5 volt reference those small differences in resistance could still provide the ECU with proper voltages.

The next step is to A) check and adjust the Idle switch and B) check the TPS voltages. I prefer to do this with a scantool of some sort (Select Monitor, VRG3s B10, Evoscan) to read the parameters directly from the ECU, but can also be done fairly easily with the meter.

I'll try to do a quick write up after work today, but time is tight and I might not get through it before the weekend. The process is in the service manuals that are posted for download in the "stuff you need to know" forum (http://bbs.legacycentral.org/viewtopic.php?f=12&t=14485)

Quick update. I reconnected the TPS and turned the ignition to "On"

The white signal wire is showing 1.0v at Wide Open Throttle and 4.63 and closed. It gently goes between 4.63 and 1 without any major jumps and doesnt appear to be "stuck" anywhere.

I'll check the idle switch next

I checked the continuity between Pins 1 and 2... Which I believe is for the idle switch. Not continuity when the throttle is either open or closed. So does that mean the idle switch isnt working?

If you checked continuity with the connector disconnected, then yes, that means the idle switch is either bad or not adjusted correctly.

I would loosen the screws and, at idle, see if it will rotate and close the switch. If you can get the switch to close (have continuity), and with the TPS in the new position, verify that the *voltages* match what is listed in the service manual.

But if I adjust the TPS wont that throw the voltages out of adjustment?

Yes. That's why you need to double check them after the switch is adjusted. If you can't find a sweet spot where everything jives, then the TPS is likely bad.

Sometimes you can nudge the TPS to get the idle switch working and keep the voltages in spec..sometime you can't. That is all assuming the idle switch works at all.

This is the proper procedure for adjusting the idle switch. Follow it, and then recheck your resistances.

I'll give you guys an update tomorrow. Crazy work schedule the last two days so I haven't been able to adjust the idle switch…

Thank you for the spelled out instruction Josh!

It was definitely the TPS!

Got it adjusted and no more hanging idle!


But I cracked one of the pcv lines and now I've got a leak. Oh how I love this shit lol

Glad to hear that adjusting the TPS brought the idle back into check.

Fixed the broken PCV lines…

The car has never idled this well!!!