Higher temp = Higher idle -- running out of things to check

[betsy325e]

Greetings, smart people on the internet with answers

Just like the title says, the rpms on my 325e start at a perfectly smooth 700ish rpm idle at start up to a 1300-ish idle at operating temperature in park and neutral. Pretty constant 750-800 in drive or reverse (which is significantly higher than it used to be). Runs pretty well (I hadn't driven it through the winter, so i'm not 100% confident in comparing it to as I remember).

Trying to sort this out I've done everything:

*Have checked for vacuum leaks: Nada. All new hoses, all new gaskets on intake manifold, throttle body, cold start valve, valve cover, o-rings on tube of death, brake booster o-ring, charcoal canister, etc. Everything. I've checked and rechecked and rechecked because I was really hoping this would be the issue. Nope.

New coolant temp. sensor -- checked prior to and after install: working.
New coolant temp. switch -- checked prior to and after intall: working.

Cleaned ICV // Swapped out ICV with known operating one. Neither solved my issue. Both ICV's have the adjustment screw all the way in -- that produces my 750 rpm. Prior to ripping the car apart looking for my vacuum leak that originally lead me to replace everything that could leak, I never had to set that screw all the way in. Unplug at idle produces rev up, then cut out, rev up, cut out as has been described in other posts. *How can I be sure the ICV is getting the correct signals from ICM?

Checked MAF using 9-volt battery test: operating fine (was super glad I didn't have to play with anything in there)

Swapped out ICM with like 9 other green ones ones I've picked up from junkyards: no luck. Tested every connection at the harness end of ICM as per Bentley/what I've seen on the internet. Everything is fine; hot, cold, everything good.

Did every check I could think of pin in/pin out at ECU. Everything seems fine.

Did the Bentley check on the O2...got continuity as Bentley says I should (I'm not really sure what that proves, though).

Checked and adjusted throttle, checked with feeler gauges for close. All good. Checked TPS -- continuity at close, none at open, continuity at WOT at the other pins.

And on and on. If there's a potential solution for this problem on the interwebz and someone put it on a forum, I've read it and tried it.

I know a lot of you will say just to let it go...and if it were a manual I would. However, I have the dreaded automatic and have already put a used one in after my inspection mechanic revved the crap out of it in park after driving it and burned the clutches. I'm worried my high idle (it shoots RIGHT up once you put it in park or neutral) is going to cause tranny failure.

I've checked more stuff than I've mentioned, but I'm unable to recall right now. I've been thorough about this. Not trying to waste anyone's time on their millionth high idle thread.

Things I've just happened to notice: pulling oil dipstick makes car run like crap. Pulling oil filler cap at temperature makes car idle perfectly. I don't even comprehend that...hopefully someone will. Pretty strong gasoline odor from cap (again, this is subjective...I don't really know if it's running rich or not).

Also, after getting to op temp and shutting off engine (for coolant bleeding purposes) I noticed that when car starts, temp. gauge is nearly 3/4 and quickly came down as fan cooled the engine. At 3/4 it had a nice 750 rpm idle for a split second and then shot up as the temp. gauge got to 1/2.

I'm not sure what to make of either of these.

Wtf is wrong with this car? It's driving me CRAZY that I can't figure it out.

Ideas? Things to check? Anyone else have this problem?

Thanks.

[Klug]

hummmm.. I don't know how it might happen- but have you checked to make sure that the A/C is off and remains off when it is heated up? Typically there is an idle bump when the A/C is engaged.

[betsy325e]

A/C doesn't work -- wasn't getting 12 volts at the ICM on or off, but I didn't think to check it with the car hot. Sort of doubt that's going to be it since it doesn't work, but, like I said, I'm completely out of ideas. Will check and report back.

[ahlem]

Take out the Coolant temp sensor and clean it with a scotch brite pad.
The brass colored area can get some crud on it over time and while the unit will function properly, it might be insulated by the oxidation.

[betsy325e]

Brand new sensor. Tested on stove. Works.
Old one worked too. Tested once I swapped it out and it worked too. Not gonna be the physical sensor.

[betsy325e]

I posted this in a different thread, but does anyone know:

Do you know if any resistances change coming out of the ECU to into ICM?

For example, if I measure a temperature sensor at the sensor I get X ohms. If I check it at the ECU I still have X ohms. Checking it at the ICM I've got significantly less than X ohms.

Does that sound right, or do I have a wiring issue?

[betsy325e]

Update:

Tested wiring as follows.

Coolant temp switch ground wire to G103: Solid ground. No problemo.
Checked all grounds for ECU and ICM: No problemo.
Checked continuity from ECU pin 13 (coolant temp sensor) to ICM pin 11 (coolant temp sensor): no problemo.

Checked continuity from positive end of connector to both ECU pin 13 and ICM pin 11: Problemo? I'm picking up about 400 Ohms, while I think the wire's resistance should be none. Is that right? It should have zero resistance in the wiring, and the 400-500 Ohms is going to throw off whatever the ECU/ICM does with that resistance measured at the coolant temp switch, right?

I assume that end of the harness goes to a connector/junction of some sort before heading down to the ECU. 1987 325e (manu: 9/86). Does anyone know where that connector is and what pin the temp. sensor is? I'd like to try to track down a fault in the wiring.

[betsy325e]

Update:

Cut open the connector going to the coolant temp. sensor. The positive brown/red wire has a resistor in the connector -- that's where my extra 500 ohms between the sensor itself and the connections at the ECU and ICM are coming from. It doesn't appear in any of the wiring diagrams, which is super handy... I'm going to rule out the coolant temp. sensor.

Anyone got any other ideas?

[betsy325e]

Update:

1. Checked gas cap seal just for the hell of it: fine.

2. Checked TPS continuity through positive closed throttle wire at connector to #2 ECU and #12 ICM: no problem.

3*. Checked TPS voltage between ground and closed throttle bin: only 3.5/4.0 DCV. Bentley says minimum of 5 DCV.

* Does anyone know if that could be my issue, and if so, what could be causing it?

4. Checked charcoal canister vacuum line to throttle body. I suspect maybe it is collapsing (serious 90 degree bend in it). Don't know if that could cause my issue but I'm going to buy a solid elbow connector and replace.

5. Checked cold start gasket: fine. '

6. Checked throttle body gasket: fine.

7. Checked tube of death seals: seemed fine but didn't rip the manifold off (I just replaced o-ring seals on tube of death and intake manifold gaskets).

8*. Pulled plugs, thought maybe I'd find a leaking injector. Didn't smell gas, but plugs 4,5,6 had a decent amount of oil on them. 1,2,3 were ok.

*Could this have anything to do with what I suspect is a crazy amount of vacuum in the crankcase? (Remove oil cap while idling @ 1200-1400 RPM at operating temperature brings it down to a nice 750 RPM). Cleaned plugs and re-installed finger tight: going to check for gas tomorrow morning.

9*. Brake booster vacuum lines seem good, but worried about check valve.

*How do I test this?

10: Checked vacuum lines from manifold to FPR: snug and no leaks.

*11. I also noticed a decent amount of thin oil coating on the floor of my intake manifold plenum. I cleaned it out very thoroughly prior to installation and ran the car a total of maybe 30 minutes total since then.

*How normal is this?


Anyone got any ideas about numbers 3, 8 and 9?

Losing my mind trying to figure this out -- even really bad ideas are welcome at this point haha.

Thanks.

[downforce22]

If you have a spare ecu, maybe try to swap it and see if that does anything different. Maybe one of the connections in the ecu is starting to develop a hairline crack and mess with a temp signal or something?

[betsy325e]

Really wish I had a spare ECU lyin' around, but no such luck. That said, all the resistance readings are normal between the prongs on the sensor and what I get back at the ECU and ICM (+500 ohms from resistor at wiring connector for coolant temp. sensor). I have also swapped out a total of 3 green ICMs to zero effect.

[betsy325e]

Update:

1. Replaced fuse #9 even though it checks out fine.

2. Checked all relays for corrosion. Gave them a little super fine sandpaper treatment. All seemed good.

3. Rechecked bitch tube/tube of death: pulled up washer and checked o-ring on lower: fine. Checked out upper o-ring: fine.

*4. Ran out of ideas, checked fuel injector wiring which I assumed I could not have fucked up because the engine idles so smoothly. Yep, was a hasty idiot and connected connector #3 to injector #4 and likewise #4 to #3. Was super pumped that this was finally going to maybe be my issue. Put everything back together in non-idiot fashion, turned on the car, and let her idle.

*CHANGED ABSOLUTELY NOTHING. Idle speed came up as temperature came up. Still idles just as smoothly.

#3 and #4 are on separate injector controls from the ECU from what I can see. How did this not change anything? Is my engine in some weird default/get you home safe mode? WTF IS THAT ABOUT?

I'm completely out of ideas at this point. Gettin' angervated.

Anyone?

Thanks.

[Klug]

Man, this sounds frustrating...

Have you checked your CPS? Verified correct ohms? Also correct distance from CPS to crank wheel and it isn't dirty? Looks like your early model should be 960 +/- 96 ohms.

I just ran into a different random misfire issue and my CPS checked out good but was too far away from the crank wheel. However, the bentley states that early models are not adjustable.

Just throwing out ideas of stuff maybe you might have over looked

[Klug]

3*. Checked TPS voltage between ground and closed throttle bin: only 3.5/4.0 DCV. Bentley says minimum of 5 DCV.

* Does anyone know if that could be my issue, and if so, what could be causing it?

Try bench testing the AFM with a 9 volt battery, there is a DIY out there

8*. Pulled plugs, thought maybe I'd find a leaking injector. Didn't smell gas, but plugs 4,5,6 had a decent amount of oil on them. 1,2,3 were ok.

*Could this have anything to do with what I suspect is a crazy amount of vacuum in the crankcase? (Remove oil cap while idling @ 1200-1400 RPM at operating temperature brings it down to a nice 750 RPM). Cleaned plugs and re-installed finger tight: going to check for gas tomorrow morning.

A bit of oil is usually normal...

9*. Brake booster vacuum lines seem good, but worried about check valve.

*How do I test this?

Try doing a home-made smoke test to find vac leaks. Or get the engine running and get a can of carb cleaner, spray all vac unions and see if RPM's bump

[betsy325e]

I did the 9-volt testing on the AFM. Everything was fine.

I was actually inquiring about the voltage running to the TPS. Bentley says minimum of 5 DCV. I've got like 3.5 - 4.0 DCV. Not sure why you brought up the AFM?

A smoke test isn't going to tell me anything about the inner workings of the check valve.

Did remove the CPS, did all the ohms testing. Connector was a bit knackered, but no issues a few zip ties couldn't handle until my new one arrives next week.