[Killerfish2012]

On Friday, I was At the Autobahn Country Club, with Hooked on driving. It was the first track day at this facility since last fall, and the track surface was cold to the touch. This meant no cornering traction for everyone, except the AWD monsters. The rest of us struggled to break 3 mins. I was in between a mustang boss (3:03), and a FBO E90 M3 (3:06), with a time of (3:05). Normally times for those cars are 2:49-2:53 in June-August.

About Limp Modes: This was my first time using Cobbs OTS 100 octane race map. It had truly impressive power gains over Stage 2+ aggressive, which I ran last fall. I had only one limp mode all day due to a 30FF. I think I've finally figured out how to keep 30FF, and 30FE from ruining your track day. It seems our ECU uses a PID algorithm to figure out boost. I read up on PID, and since it stands for Past, Present, and Future, I figured that the ECU, must be using the data stored from all your recent street runs, in combination with the stuff it's logging as you race on the track, to throw boost leak, and over boost codes, over the most minor issues. I decided before each session, to reset ECU learning, using Cobb. This has yielded very good results. Of course you should check for boost leaks before your track day, and you should upgrade your vacuum Lines, etc. But barring any Major issue, this is a sure fire way of keeping the overly sensitive ECu from ruining your track day! Anyways here's the video. Please ignore the crappy tractionless cornering.

[TurboBimmer]

Have you considered getting a pro-tune to optimize the maps for track use and get rid of the limp modes? Might be worth a try.

Nice driving btw! Track looks like a lot of fun.

[Killerfish2012]

Originally Posted by Autobahn335i

Have you considered getting a pro-tune to optimize the maps for track use and get rid of the limp modes? Might be worth a try.

Nice driving btw! Track looks like a lot of fun.

Thanks! I'm not interested in a pro-tune at the moment. Guys with the pro-tunes are still having under/over boost codes. It's less to do with maps, and more to do with ECU sensitivity during tracking.

[TurboBimmer]

Originally Posted by Killerfish2012

Thanks! I'm not interested in a pro-tune at the moment. Guys with the pro-tunes are still having under/over boost codes. It's less to do with maps, and more to do with ECU sensitivity during tracking.

I'll give the ECU clearing method a try on my next trackday. Can't hurt I guess

[Killerfish2012]

Originally Posted by Autobahn335i

I'll give the ECU clearing method a try on my next trackday. Can't hurt I guess

Do it before each session. Once we start seeing 30FE/30FF, you keep seeing it, even when you switch maps or flash back to stock. So it's definitley not map related, and most likely the PID algorithm.

[everlast]

PID is not really past, present and future like you're thinking. It stands for proportional integral derivative, and is terribly complicated. I find closed loop tuning some of the most difficult and challenging aspects of tuning, and if not fully understood, PID can lead to disastrous results lol. It could be very likely there are targets for failure in the ECU that have a variety of parameters to "hit" that trigger the codes, and by resetting the ECU you're resetting the counters involved, but that's not really anything to do with the closed loop boost control.

Are you able to use a coding cable to read any stats out from your car that might be useful? If you can identify the parameters that are triggering, you can perhaps use some electronics to trick the sensors into seeing more sensible data or something. I dunno, I'm just guessing while drinking a coffee.

http://en.wikipedia.org/wiki/PID_controller

[demontwinz]

kfish didn't u buy an e36 m3?

wait maybe it was jbass ... the fish is mix'n me up

[Killerfish2012]

Originally Posted by everlast

PID is not really past, present and future like you're thinking. It stands for proportional integral derivative, and is terribly complicated. I find closed loop tuning some of the most difficult and challenging aspects of tuning, and if not fully understood, PID can lead to disastrous results lol. It could be very likely there are targets for failure in the ECU that have a variety of parameters to "hit" that trigger the codes, and by resetting the ECU you're resetting the counters involved, but that's not really anything to do with the closed loop boost control.

Are you able to use a coding cable to read any stats out from your car that might be useful? If you can identify the parameters that are triggering, you can perhaps use some electronics to trick the sensors into seeing more sensible data or something. I dunno, I'm just guessing while drinking a coffee.

http://en.wikipedia.org/wiki/PID_controller

Past, Present, and future, as in error preditiction mechanism. If the error is present in the past, and slightly more in the present, it's future occurance is higly likely. This is how the ecu helps you deal with boost leaks and over boost issues, as they get worse over time (or at least I think is what is going on). Clearing learning is showing great results, but you're right in noting that it isn't an ideal fix. Renntech is the only company I know of that dealt more directly with this issue via Klennfeld corrected values at the track, to deal with track boost PID.

[Killerfish2012]

Originally Posted by demontwinz

kfish didn't u buy an e36 m3?

No, holding out for american muscle: 11 Stang GT, Boss, or C6 Z06, but thats a definte maybe I'm having too much fun right now chasing the above mentioned cars with my wavetrac LSD!

[DallasBoosted]

Also make sure that your map isn't requesting boost that you car can't deliver; remember that it will get hot and the turbos won't be able to pump as much hot air as they can when the IC is working cold. Which means you have to taper the boost at higher rpm more.

I don't know what the threshold is but when I had my car trying to hold more boost to redline it would get 30FF at the track @ high rpm now and then; since I changed the map to taper the boost down following what the turbos can actually do, it doesn't throw that code anymore. I still get about the same amount of boost to redline, but the requested boost isn't 2-3psi higher than the actual, which is what causes the 30FF.

Hope this helps, my .02, etc

[Killerfish2012]

Originally Posted by DallasBoosted

Also make sure that your map isn't requesting boost that you car can't deliver; remember that it will get hot and the turbos won't be able to pump as much hot air as they can when the IC is working cold. Which means you have to taper the boost at higher rpm more.

I don't know what the threshold is but when I had my car trying to hold more boost to redline it would get 30FF at the track @ high rpm now and then; since I changed the map to taper the boost down following what the turbos can actually do, it doesn't throw that code anymore. I still get about the same amount of boost to redline, but the requested boost isn't 2-3psi higher than the actual, which is what causes the 30FF.

Hope this helps, my .02, etc

The stuff I'm talking about here isn't map or even hardware related. On more than one cobb 335I we've seen the exact same 30FF/30FE thrown even after we've unistalled the tune entirely, and ran stock. It's entirely related to boost PID, managed through Klennfeld Values (Engine management) which cobb, being a partial tune, isn't really compensating for.

[DallasBoosted]

Originally Posted by Killerfish2012

The stuff I'm talking about here isn't map or even hardware related. On more than one cobb 335I we've seen the exact same 30FF/30FE thrown even after we've unistalled the tune entirely, and ran stock. It's entirely related to boost PID, managed through Klennfeld Values (Engine management) which cobb, being a partial tune, isn't really compensating for.

No I get what you're saying; I think that will help too. Resetting the learning isn't a bad practice, especially on our stock turbo cars where the turbos are just running their little hearts out trying to make 18psi.

I'm just also pointing out that setting up your map to try to request 18psi past about 5000rpm will also cause 30FF because the turbos just can't do it. I got rid of some nagging 30FF codes (which only really popped up on the track, sustained periods of "not making requested boost") by taping my targets down considerably, to match the boost that I was making anyway.