1989 325is (M20, 5-speed) – high idle ~1,750–2,000 rpm with CEL

[wilburb]

Car ran great a month or two ago when I last drove it. Wednesday this week it was hard to start, then idled rough and high with the CEL on, in the ballpark of 2,000 rpm. I replaced the rotor and distributor cap today, as I knew they were pretty worn from this spring's timing belt replacement. While it feels smoother running, idle is still around 2,000 rpm.

What I've checked so far:

- Intake boot: inspected, no cracks. All vacuum hoses traced by hand, nothing obvious except the following.
- Brake booster elbow at the throttle body is loose (press-fit, can move it with a finger). Sprayed carb cleaner on it with the engine running: no rpm change. Same with the ICV line into the throttle body: it's loose and I can push it with a finger or two, but spraying carb cleaner there didn't change the idle either. I assume these have been loose for a while and are not the cause of this newfound high idle.
- ICV: removed, it was gummy. Cleaned it and bench-tested with a battery; it snaps fully open and fully closed. Reinstalled, no change in idle.
- DME coolant temp sensor: ~1.63k ohms on a mostly cool engine, 4.89V at the plug with key on. Unplugging the sensor makes no noticeable change.
- Throttle: moves freely and closes fully. Idle switch clicks; both idle and WOT contacts show continuity on a meter.
- Charging: steady 13.7–13.9V running. Engine-to-body ground voltage drop is ~30–37 mV.
- Fault codes: I've read you can turn the key on and press the pedal 5x to get a CEL flash code readout, but doing so results in no flashing. No access to a scanner.

Questions:
1. I'm not a mechanic, and my experience beyond the basics is what I've learned working on this E30 over the last couple of decades. Are there obvious ideas here that I'm missing? Again, the last time I drove it a couple of months ago it was fine. It sat in the garage until I tried to start it on Wednesday, and replacing the cap and rotor today didn't fix the idle.
2. Any reason the pedal code readout wouldn't work on a 1989 US 325is?
3. Anything specific I should also check to get a better idea of what is causing this?

Thanks for any ideas.

[88Sedan]

The hose for the purge solenoid hides very well and I believe no clamp is used there. If the hose is old it can let a lot of air through. It is hard to see under the intake setup but it's a possibility.
Keep checking all vacuum lines, and maybe while running try and squirt something around injector holes to see if theres an air leak. I swear the valve covers on these motors start leaking if you look at them funny.

[Mick Delane]

On #2: that pedal method isn’t how a US 325is with Motronic 1.1 reports. It’s key only. Turn the key to ON, engine off, and watch the Check Engine light. After about 3 seconds it flashes any stored codes, one burst per code, with a 3-second pause between them. Key off and on again to repeat.
1 flash: airflow meter
2: oxygen sensor, or mixture too lean or too rich
3: coolant temperature sensor, or engine running too hot or too cold
4: throttle switch
Only emissions-related faults light the CEL, so no codes doesn’t mean no faults. Once it’s fixed, they clear after five or more starts.
On the idle: 2,000 rpm is a telling number. It’s roughly what this engine does with the idle valve unplugged. So before chasing anything else, find out whether the valve is actually being controlled:
1. Engine warm, accessories off. The valve should hum, and you should feel it vibrate.
2. Pull its plug at idle. If the rpm doesn’t move, the ECU isn’t driving it.
3. Ignition off, plug off: about 40 ohms across the two outer pins, about 20 from the centre to each outer.
4. Ignition on, at the plug: battery voltage on the centre terminal, about 10 V from the centre to each outer. No voltage there points past the valve, to the wiring or the control unit.
5. With the valve off its hoses and still plugged in, switch the ignition on. The piston should move to about half open and stay there.
Your battery bench test proves the piston can move. It doesn’t prove the car is commanding it.

[dice2012]

I thought 173 ECUs with Motronic 1.3 came around late 88/89? Check the ECU to see which one it is. If it's not a 153, then you have the stomp test. Keep trying, you'll get it.

Also, no offense, but bot?^

[Mick Delane]

dice2012 Respectfully, you can make your point without being weird and calling someone a "bot" because you disagree. There's a real person behind this account, and I don't post anything I haven't checked against the page it came from. If something I post is wrong, call it out on the facts and I'll correct it in the thread.

On the ECU, fair point, and it's the fastest way to settle it. The Bentley puts the 325i and 325is on Motronic 1.1, with codes flashed on the Check Engine light at key-on and nothing about pedal presses. If some later cars got a 173 box, that's a running change the Bentley doesn't show, and the label on the car beats any book.

@wilburb: read the Bosch number on your control unit's label and post it here. That tells us which system you've got before anyone argues about the readout. Either way, don't let the codes hold up the idle valve checks. At 2,000 rpm that's still the first place I'd look.

[Mick Delane]

Correction to my #3. I said a 325is reads codes key-only, as Motronic 1.1. For your year that's most likely wrong. BMW's own parts catalogue lists Motronic 1.3 control units for the US 325i from mid-1988, so on an '89 dice2012's pedal method is the one to try. The Bosch number on the label still settles it for your car.

[dice2012]

Mick Delane
this you?