Another Idle Surge Thread

[Floxer]

Not sure about e36, but all other vehicles I've worked on it was like this: If the ECU pulled all the timing it could and it still knocks, then the CEL lights up

[bmwstephen]

Fully closed.


If knock sensor's pulling timing, the CEL would indicate so?

It would normally post a check engine light. a stomp test would reveal a code 1225 or 1226. That's what happened to me but I was on original knock sensors at 130k. they were literally cracked all around and I didn't see it until I went to pull the intake manifold off when doing cooling system

[Rugbyfan]

When I first did my e30 swap, I had, for a lack of a better term, “ghosts”. Super random intermittent idle issues. My engine harness had seen better days. These harnesses get crusty over the years. Doubly so if a PO hacked into it. You may want to grab a cheap harness from a junkyard and spend an hour replacing it. It couldn’t hurt.

[Province]

So I bought an evap smoke machine to do a more "proper" test only to find no leaks / marginal leaks.
I revisited my TPS, I slotted my TPS even more so that at closed, I'm hitting 1.13k. I started the car up, no idle surge at all. I drove the car while its open loop and at around 2k at low throttle input, the car does a HARD cut. Don't know whether or not ignition/fuel. I didn't drive it any more because of that hard cut. I will probably do it again and see how it acts during closed loop when its daytime. Resistance sweeped the sensor, no dead spots.


The only thing I can make out of this would be my maf has gone bad but no codes what so ever.

Edit: Moved tps to 1.2k but car was already warm and in closed loop. 1.1k had 1216 code and the 1222 code dropped (likely due to the stuck open purge valve being bypassed). At this point, it has to be the tune thats bad. EEPROMs are hard to come by and ecus are expensive.... sheesh

[Rugbyfan]

I may have missed it but:
-grounds?
-leakdown/compression?
-how’s overall voltage? Alternator going intermittently out causing a drop and wonky electronics?

What’s a red label 413 go for these days? They used to be super cheap. I’m sure someone would loan you one.

Engine harness? Again should be cheap in a junkyard.

[Province]

Grounds are good, no electrical gremlins.
Leakdown good, holds about 10.2 +- 0.1 /cyl with strong 19inhg vacuum. No bleedoff on either.

Im on silver 413. I think I found my remedy:
Im gonna set tps to a resistance position where its just ever so slightly out of range of the tune so it idles good open loop that way it wont affect throttle response.
Edit: Fail, still lose response due to dme having to switch fuel/timing maps intermittently.

Does dme trigger some sort of idle circuit on tps feedback? If so, then the tune might have an adjustable parameter for this.
I'm also considering FPR being stuck closed. Don't have a gauge to T into the setup so I'll unplug the vacuum to it and observe, maybe push vacuum from external source to verify things.

[Floxer]

Does dme trigger some sort of idle circuit on tps feedback?

it does. use RomRaider and see what it does.

[Province]

Pulled vacuum from FPR, with the air leak car runs smooth (confirms rich idle), I plugged the manifold outlet to fpr; car ran rough still. Applied vacuum to fpr, no change until 16inhg where the car misfired less. Potential stuck fpr? It's cheap so I'll give it a shot.

[Floxer]

16hg sounds like something that could be reached during idle, maybe more, and the FPR worked, as your car ran better. Why does that point toward stuck FPR?.

Pulled vacuum from FPR, with the air leak car runs smooth (confirms rich idle)

it confirms that you've got unmetered air entering the engine (removed hose), and also more fuel than planned, because you removed the hose from the FPR. How can you rule out that these two couldn't cancel out by accident.

[Province]

See

I plugged the manifold outlet to fpr; car ran rough still.

Rough being the same surging idle.

When I applied vacuum to fpr, the engine started to warm up (temp gauge being at edge of blue) and thats usually when my surging stops. I only realized my doubts now so Ill revisit when the engine's dead cold.

I never mentioned this but on greentops, on dead cold engine; the surging is hard but 400 - 900. When I switched to pinktops, the surging became extreme to where it'll kill the engine, or go between extremes ie: 200 to 1200. Idling open loop, fuel should be constant as per fuel pressure. This is why Im suspecting fpr.

[Floxer]

datalogging would be useful. you'd see ignition angles, injection times, icv, tps, all kinds of data at the moment of the surging. romraider logger and a 20 bucks cable is all you need.

[Province]

Final update;
I realized I have to press down pedal 5 times during code reading to move on to the next code. Obtained 1286 code so I probed continuity on the circuit, no breaks either way. Bought a 325i 413 dme, replaced the chip with my turner, car started up cold without a single misfire or surge. When I drove, the car does not pull timing @ 2k-2.5k rpm on open loop and closed loop. Consider it all done.
I looked at my old dme and saw no water spots and looked through general areas for cold solder/cracked solder, mainly the connector bridge and several caps and yielded nothing.

Now I ask any ideas of preventative measures? I put down foam padding on plate the DME sits on but beyond that, can't think of anything else to do.

[catalyst.]

Well damn, thank you for the update, I will probably try this after I look at everything again, especially the ICV. My issue is that the car will not idle at all... starts and seems to run perfectly fine, but it just dies, doesn't even attempt to idle. My other thought is it is a bad 540 maf, but maybe it is a dme issue like yours was. I think i replaced the ICV though, i cannot even remember, need to check it. It has been so long since i messed with the car I have to look at a list to even remember which parts are on it.