Removing rev flare on startup

[DocWalt]

Hi everyone,

I've not spent too much time on this so I may be missing something obvious (either in the tune or how an engine behaves) but I'm trying to kill a rev flare that happens on the first start of the day. I'm attaching my tune and two logs. The first log shows the rev flare and the second log shows what happened immediately after shutting the engine off and restarting it after the first log. It's not very cold at the moment, so the rev flare isn't quite as exaggerated as it is sometimes. I'm trying to kill the rev flare because I've had some bearing issues and I'm trying to avoid any unnecessary revs before it at least has full oil pressure.

The vehicle is a '92 VW GTI 16v, 2.1L stroker with TT 288* cams. MS2 Extra w 3.57 board with a jbperf PWM mod driving the Bosch 2 wire PWM valve and the basic mods to drive the coil directly using the factory VW hall sensor. Runs VERY well other than this issue (though the VE table isn't quite perfect yet).

Thanks everyone in advance

[billr]

Try setting the "Idle Cranking Duty/Steps" lower. Or, maybe higher, if your IAC valve works "backwards"...

[DocWalt]

Try setting the "Idle Cranking Duty/Steps" lower. Or, maybe higher, if your IAC valve works "backwards"...

Already done that. If I go lower the car struggles to start after the first start. Not sure why it starts so rapidly on the first start and revs up so easily.

[billr]

Maybe it is too rich when cranking, a combination of the priming and cranking pulses... or maybe because of that weird MAT correction curve.

[Matt Cramer]

Also try less spark advance at the point where the "rev flare" happens.

[DocWalt]

Maybe it is too rich when cranking, a combination of the priming and cranking pulses... or maybe because of that weird MAT correction curve.

That's a good point. I may reduce the cranking pulses a bit with the idle valve closed a bit more to control the spike.

The MAT correction curve is due to some bizarre heatsoak issues (MAT sensor is after the idle valve exit from the intake tube... so at idle there's almost no airflow over the sensor) I have where it would eventually lean out to the point the engine wouldn't run and EGO correction couldn't correct for it. I haven't driven the car in traffic this summer so I don't know how much of an issue it actually is, but I assume it still has the same issue. It's off topic, but how would you recommend fixing it (obviously moving the MAT sensor would work, and I'll probably do that)? For the record, the left half of the MAT correction curve is 100% "stock" from MS Extra base map, I only touched the hot MAT stuff.

Also try less spark advance at the point where the "rev flare" happens.

I can try that too.

[DocWalt]

I'll bump this because this has cropped up again. When my car exits cranking the PWM idle immediately jumps up to a really high value and causes the revs to rise drastically. It immediately drops a LOT and when that happens the engine tends to stall.

This is leading to various issues on cold starts, I can't open the valve more or the engine revs like crazy and risks bearing damage on dry starts. With the valve not being open wider it doesn't turn over well unless I hold the throttle cracked open, and when it does start it just stalls. After a few starts it finally stays running long enough to build some heat and I can back off the throttle.

I'm using MS2 v3.57 board w/ MS2/Extra 3.4 (I know the firmware is old, but this behavior has existed for a while and I've seen nothing in the changelogs that makes me think anything has changed to affect/improve this, but I could easily have missed something.

What am I miss here?

edit: funny, this seems to show the issue perfectly seen here: http://www.msextra.com/forums/viewtopic ... 34&t=67895

There's no PWM voltage compensation during cranking so when the engine exits cranking the PWM valve is suddenly opened a LOT and bad things happen.

[billr]

Did you remove a MSQ posted much earlier in this thread? Now I see that the VE table is also quite low, and was wondering why I didn't notice that before. So, I went back to compare, but there is no previous MSQ now. Matt C. had suggested reducing spark advance down in the idle rpm range, but it is still pretty high. Did you ever try lowering that advance?

Personally, I wouldn't worry about it flaring to 2800 at first start. Nothing in that engine is really dry of oil, even if it has been sitting for weeks or months. Is your oil filter oriented so you can pre-fill it and get oil pressure almost instantly after a filter change?

PS: People get in a hot-sweat about priming so that the engine fires immediately. Maybe for the bragging rights, "my engine fires up quicker than yours"? I have no priming, so it cranks for a couple of revolutions before firing; by then oil pressure is already up enough to feed the bearings.

[DocWalt]

Did you remove a MSQ posted much earlier in this thread? Now I see that the VE table is also quite low, and was wondering why I didn't notice that before. So, I went back to compare, but there is no previous MSQ now. Matt C. had suggested reducing spark advance down in the idle rpm range, but it is still pretty high. Did you ever try lowering that advance?

Personally, I wouldn't worry about it flaring to 2800 at first start. Nothing in that engine is really dry of oil, even if it has been sitting for weeks or months. Is your oil filter oriented so you can pre-fill it and get oil pressure almost instantly after a filter change?

PS: People get in a hot-sweat about priming so that the engine fires immediately. Maybe for the bragging rights, "my engine fires up quicker than yours"? I have no priming, so it cranks for a couple of revolutions before firing; by then oil pressure is already up enough to feed the bearings.

I was pretty sure I had a tune and log uploaded before, I don't remember deleting it but I may have.

If I let the car sit for more than a day or two it takes 5 to 10 seconds while running to prime and turn off the low oil pressure light... I don't want to be at 2000+ RPM while that's happening. There's nothing wrong with the oil filter or oil system, plenty of flow and pressure, oil filter has an anti-drain back valve, etc.

This tune is a LOT richer at idle than my old engine (stock) and the idle advance seems normal to me. The cams in this engine are gigantic and ideally I'd be even lower VE at idle... FWIW Matt suggested less advance where the flare is happening, not idle. That's not really an option because it revs up high enough to be in "normal" driving conditions. Dropping the idle advance much lower and the engine doesn't like to catch itself while returning to idle and also turning the wheel causing a load on the power steering (there's no idle up switch for the P/S). The massive cams make this a big hassle, this was pretty easy to tune around with the stock engine.

FWIW I don't care if it fires immediately, I don't want it to rev its nuts off beating up my rod bearings because the PWM idle valve isn't working properly. I'd suggest looking at the log file to see the PWM idle valve behavior at startup versus the battery voltage. IMO this isn't behaving properly. If I need to modify the firmware for my usage case, I will. I'd also like to use MAT for cranking duty of the idle valve, kill two birds with one stone...

[billr]

I don't think I'm going to be able to help here. I'll wish you good luck and bow-out.

[prof315]

Welcome to big cams with a plenum manifold.

[DocWalt]

I don't think I'm going to be able to help here. I'll wish you good luck and bow-out.

Thanks for the help, Bill! I apologize if I came off rudely, but I've been chasing this PWM issue for years and it's getting old.

Just today I fired the car up in 22* F weather and it surged up and then immediately fell flat on its face. Fired it up a second time and it started beautifully and stable. If I can get the firmware to behave like I think is correct then I'll make another thread if I'm still having startup issues.

Welcome to big cams with a plenum manifold.

I don't think the poor PWM behavior here has anything to do with big cams. I'm going to have to correct the firmware for my setup and be done with it. It's one line of code, I believe. Changing the cranking duty to look at MAT instead of CLT looks to be a little more complicated, I'll worry about that later this winter.

but.... I should definitely do ITBs to get better behavior but at this point I'm frustrated with the reliability of the car and don't want to spend more money. I've eaten enough rod bearings thanks to a subpar oiling system due to various shops dropping the ball after claiming to have the fix... Finally fixed that, I think.

[Blown88GT]

...I was pretty sure I had a tune and log uploaded before, I don't remember deleting it but I may have....

Uploads are automatically deleted after 1 year.

[turbo conversion]

Can you post a data log of a failed cold start, the one posted is a hot start.

From my experience when a cold start flairs then dies the ASE is not set high enough.

Every engine is different but to me your ASE seems to be kind of low.

Something else to consider is when the engine fires it goes in the 14.7 afr range not 12.5 afr (as the idle cells are set) so it will lean out a fare amount.

David

[DocWalt]

I made the change to the firmware to use voltage compensation during cranking, and it seems to work fairly well on warm/hot starts. I'll log and see if it fails tomorrow morning on a cold start (should be about 20* F). I let the car sit and got down to 60* F and when I started it up there was a pretty big jump in RPM (up to a bout 2500 RPM) like before, but it started without touching the throttle at all. Once it starts consistently I'll attack the ign. advance as suggested above to try and control the big flare up. As I mentioned above, it ends up in areas of the map that I use while driving so I'm a little hesitant to drop the advance there too much

I've attached the latest MSQ & warm-ish startup log.

Yes, I know I need to tune the MAT/CLT blend based on airflow, the MAT sensor heatsoaks badly and I bodged a "fix" by messing with the MAT table to richen it up when it's hot out... it leads to slightly inconsistent behavior when it gets cold, doh. I also know it's pig rich at idle with likely too much idle advance, but leaning it out or pulling timing makes it stall too easily when I turn the wheel at a stop when I have my wide/sticky track tires on the car.

Thanks for the help folks!

[DocWalt]

Much better now. I replaced the battery with almost no improvement in starting in cold temps even with an extra 110 CCA, so I guess I need a new starter It chugs along at about 150 RPM but only gets a few turns then just stops. Whatever, when it's above freezing it works great.

I reduced the ignition advance to help it stop revving so high, but the firmware changes work great. When I get a starter that works properly in colder temps I'll work on colder starts and make further tweaks.

I believe the next improvement that I need to make is using MAT for opening the idle valve on startup instead of CLT. I believe this is needed because the idle valve controls airflow that varies in density based on air temp, not coolant temp. When the engine is hot but the air is cold it opens more than necessary. This looks to be a bit more involved of a change, unfortunately.

[turbo conversion]

If it were me I would be more worried about not seeing oil pressure for 5 to 10 seconds on a cold start.

There has to be an issue with the oil draining for it to take that long to build oil pressure on a cold start.

I had this type of issue with K&N oil filters so I tried Bosch and fixed the problem.

I sent 6 filters to K&N and they told me 5 out of the 6 filters the drain back valve was not seating correctly.

A simple test is to blow through the center hole on a filter, if you don't have total resistance then the valve is not seating correctly.

David

[DocWalt]

If it were me I would be more worried about not seeing oil pressure for 5 to 10 seconds on a cold start.

There has to be an issue with the oil draining for it to take that long to build oil pressure on a cold start.

I had this type of issue with K&N oil filters so I tried Bosch and fixed the problem.

I sent 6 filters to K&N and they told me 5 out of the 6 filters the drain back valve was not seating correctly.

A simple test is to blow through the center hole on a filter, if you don't have total resistance then the valve is not seating correctly.

David

It's when the car sits for a few weeks at a time that I'm most concerned, in normal use it's half a second or so for the light to flick off. I've been using the flood-clear to crank and build pressure, but there's no guarantee that the oil is everywhere yet.

I've been using Purolator filters, I'll check them out next time I change oil. I've never had an issue before but you never know for sure. Thanks for the suggestion.

[krisr]

OEM's use timing to control start-up flare and use a factor table as you dont want full retard when cold. This is something that I would definitely +1, because my 400 has the same problem, big cam, needs alot of air to get started, but being a high stall automatic, it's not as simple as just pulling timing where the flare occurs because that area is actually used for driving.

1-startupflaretiming.jpg (285.35 KiB) Viewed 1301 times

2-startupflarefactor.jpg (225.79 KiB) Viewed 1301 times

[DocWalt]

That's interesting...

I adjusted the parameters for my "Idle Advance" to enable themselves shortly after startup and when the parameters are satisfied early enough (happens nearly 90% of the time) it controls the ignition advance and there's no rev flare.

More things to play with Thanks!