High Idle After Warm Start

[jmp88]

On cold starts I seem to have the parameters dialed in for a nice, smooth idle, but when I turn off the engine and restart it hot, it idles at least twice as high, and just stays there.

The part I can't understand is that it doesn't settle back down to a normal idle after any amount of time running. If I started it warm, it wants to idle at like 3,000rpm until I turn it off.

Any ideas? Thanks in advance!

Cheers,
Jonathan

[billr]

Those are some big numbers in the spark advance table. Does MS see "0" at crank TDC, or has that been faked, for some reason? Post a run log, also.

[jmp88]

I know, right? I didn't think those sorts of values should be possible but I checked the timing, and MS really does see 0 at TDC (okay, the marker on the flywheel is worn down, so it's sort of hard to tell, but I'm pretty confident that those numbers are accurate).

I'll get a run log this evening. Unfortunately I don't have internet at home right now so I have to come in to work to post here. I'll get it posted tomorrow at the latest.

[jmp88]

Here are two run logs, one on a cold start, and one on a warm start. I like the way it runs on a cold start, but I don't like that the idle stays above 2k after a warm start. Ideas?

Suggestions on my tune, in general, would be helpful too. It seems to run pretty smoothly but, as billr pointed out, the spark advance in particular is quite high. How is that possible? Am I unknowingly compensating for that in another paramter?

Thanks again everyone,

-Jonathan

[Matt Cramer]

That looks like WAY too much spark advance - which can indeed make the idle stick high. Have you confirmed you're really getting 70 degrees of advance with a timing light?

And if so, what engine in the name of Otto actually needs 70 degrees of spark advance? I'm guessing this is some sort of highly unusual motor or running some sort of unconventional fuel, so please post the details of your engine build.

[jmp88]

There's nothing out of the ordinary about the engine. It's a BMW M10 engine with stock pistons and cam, running regular gasoline.

I checked the advance with a timing light a few months ago and had some trouble finding the marker on the flywheel, but I finally did. And if I remember correctly those numbers are accurate.

I started with more conservative values and edged them up cautiously to their current values. After I did that is when the high idle thing started, so it would make sense that the advance caused it, but I still don't understand why the idle averages ~1000rpm higher after the warm start, than after the cold start.

For instance, on the cold start log, once the engine is warm the idle floats around ~1150. On the warm start log, the idle is consistently at ~2300. Is the spark advance used differently by MS-Extra in either of these cases?

[Matt Cramer]

If it lands in a cell with too much advance, the excess advance can keep it trapped in the cell until something loads up the engine enough to pull it out of the cell.

[jmp88]

Holy crap, I think that explains it!

The only time I've gotten it out of the high idle after a warm start was once I tried engaging the engine with the brake on until it almost stalled out, and for some reason when I put the clutch back in, it magically stayed at a lower RPM.

What does a cell with "too much advance" mean? and why would it get stuck there? Is it specifically just one of the cells it hits while cranking on the warm start that does it, or do I need to change my whole table?

Thanks again for the info!

[jmp88]

Actually, now that I think about it, if that were the case then shouldn't the spark advance stay constant through the second log? Or is it only being affected by the RPM and stuck in one MAP row or something?

[Matt Cramer]

If the RPM comes up high enough to get into the excessively high spark advance section of the map, it gets stuck there. My guess is this is only happening on hot starts and the RPM does not spike high enough on cold starts. Fix your spark table.

[jmp88]

If that's the case I just don't understand why the spark advance isn't constant in the second log. If it were stuck, shouldnt the "SparkAdv" variable stay fixed in the log? It seems to be changing as I would expect WRT engine speed and MAP.

Also, I'm hesitant to change the whole spark map because it's actually running quite well, besides this issue. Is there another variable that is maybe accidentally compensating for these values that I should change simultaneously when I adjust the spark map? Like, lower spark advance=lower VE or something?

[grom_e30]

i would be be setting the engine to fixed timing that you have a reference mark for eg 0 deg if you only have a tdc mark, and confirming that you have the tooth1 angle correct by using a strobe timing light, as the table looks way to advanced maybe you are compensating with big numbers as the tooth1 angle is off, m10 wants 25-30 btdc at about 3000 rpm according to a manual i have for an e30.

[jmp88]

I'll check the timing again this evening if I get a chance.

[Matt Cramer]

I did not mean it was stuck at one exact value - just that the engine was staying above the RPM where the huge amount of advance came in, and unable to come down in RPM.

[jmp88]

Ah, I see.

But while I'm driving it I take it through the same range of loads and speeds regardless of how it wants to idle. Why wouldn't that break it out of the high advance areas in the warm case, or conversely, put it there after driving around on a cold start?

[jmp88]

Okay. I need to verify this through some other method, but I think my tooth 1 angle is wrong. (see discussion here: http://www.bmw2002faq.com/topic/157124- ... -be-wrong/ )

That shifts my whole spark table down by 26.9 degrees, but doesn't really answer this question. It just makes the numbers look more reasonable.

[grom_e30]

is the hot start issue repeatable? if so from the cold start log the advance the engine is ticking over at its about 56-57 deg, where as the warm start is 71-72 deg, start the engine have it running fast and use fixed timing at say 56 and if the speed drops then you have pin pointed the issue to be timing related.

[billr]

"I know, right? I didn't think those sorts of values should be possible but I checked the timing, and MS really does see 0 at TDC"

Remember this from the beginning of this thread? How did you "check it" and now think you may be 27 deg off? There is no substitute for using a strobe light at the beginning of your "first start" procedure.

[jmp88]

"I know, right? I didn't think those sorts of values should be possible but I checked the timing, and MS really does see 0 at TDC"

Remember this from the beginning of this thread? How did you "check it" and now think you may be 27 deg off? There is no substitute for using a strobe light at the beginning of your "first start" procedure.

Thanks for the snark, bill. The flywheel is offset by 45 degrees, and i had timed it to the only mark within any reasonable range of where i expected to find the real mark. You can read about it in the link i posted earlier: http://www.bmw2002faq.com/topic/157124- ... -be-wrong/

[jmp88]

is the hot start issue repeatable? if so from the cold start log the advance the engine is ticking over at its about 56-57 deg, where as the warm start is 71-72 deg, start the engine have it running fast and use fixed timing at say 56 and if the speed drops then you have pin pointed the issue to be timing related.

Good call. I'll test this tomorrow and see what happens.

Cheers,
Jonathan