EBC is misbehaving.

[LowR3V'in]

i posted this on the ms forum and i know matt is a supper busy guy

hopefully some of you cool dudes can get some thoughts in as well.


Basically at about 5,7k boost falls off target hard.
at first it seemed to be bad PID tuning.


upgraded to latest code and it allows to use a slider and has a set PID

it follows. Pretty bad ass. But it still does it.

Follows target until about 5,7k and shits the bed.


here's the thread with msq and datalogs.

http://www.msextra.com/forums/viewtopic.php?f=134&t=71254&p=541694#p541694

screen shot of the problem


graph by jet jones, on Flickr

[varg]

I wish I could help you but I really can't, I'm just interested to see if you fix this. Despite a fair amount of MS experience over a decade and a half of messing with them them I have never been able to get closed loop boost control working correctly on my E30. It has a different failure mode though, holds 100% duty until the boost threshold like it should, then drops to 0% before going into closed loop initial value every time. The result is extremely slow boost response and it caused me to just give up on it.

[LowR3V'in]

Thanks.
EBC always worked well for me
i had it on the m20 since about 5 years. same exact settings too.
Was doing 29-30psi on meth. followed target perfectly, even on the m30.

Now it's doing this all of a sudden.
going to do a boost leak test while i change the clutch tomorrow.
a boost leak somewhere in the EBC system seems like it could be it
BUT it dipping like that and unable to boost back up i don't think that likely.


also, ya. That's how mine works too. it'll hit close to 100% when i floor it and then 0% like it waits to naturally
spool up on spring when it gets close to target then start to control it again. resaulting in a slower spool vs MBC.
was that what you where saying?

do you have latest code?

(Closed-loop) Boost pressure must be within this many kPa of the target boost before closed-loop control will activate.

Outside of this range the valve is held wide open (keeping the wastegate shut) for fastest spooling.

There is a boost control lower limit (kpa) setting that looks to be what can fix exactly that.
it wasn't in the code that i was using.

[varg]

also, ya. That's how mine works too. it'll hit close to 100% when i floor it and then 0% like it waits to naturally
spool up on spring when it gets close to target then start to control it again. resaulting in a slower spool vs MBC.
was that what you where saying?

do you have latest code?

There is a boost control lower limit (kpa) setting that looks to be what can fix exactly that.
it wasn't in the code that i was using.

Yeah, that's exactly what it does. Holds 100% duty, spools, gets to the Boost Control Lower Limit Delta setting you're talking about, which I have set to 28kPa (~4psi), then duty drops to zero for a very short period of time (but enough to open the wastegate and kill spool) then boost duty rises again to the initial value. Firmware 3.4.2

I even posted about it on msextra forums and was told to update the firmware (useless exercise) and someone even suggested that it might be a solenoid issue, which is completely impossible since this is obviously an output issue and has nothing to do with the passive device that is receiving the signal.

[SLEEPYDUB]

Using closed loop control, that is exactly how my graph looks as well. I target 20psi, it overshoots to about 22, drops to 18 and then stabilizes at 20. If you move the slider up more, it should oscillate more rapidly, and sometimes you can even hear it oscillate. Once it does it rapidly like that, then move the slider back down till that stops.

I've tried this, and it still doesn't seem to work well. Like varg, ive relied on open loop operation since I can get a solid curve out of it that way. But I would love to be able to use closed loop control

[LowR3V'in]

i haven't really experimented with it since my car is down.

The thing is it worked perfectly on the m20 with 30psi
and on the m30 with lower psi up to where i put in a bigger cam and upped the boost.
Another strange thing it only does it at that rpm
so i'm thinking maybe the backpressure is too much for the wastegate spring.
The exhaust is a crush bent 3'' so it goes to a 2.5'' on each bend.

The MSextra forum guys think it could be the BOV opening up but i leak tested to 30psi tial Q held shut.
strange strange strange

[pazi88]

i haven't really experimented with it since my car is down.

The thing is it worked perfectly on the m20 with 30psi
and on the m30 with lower psi up to where i put in a bigger cam and upped the boost.
Another strange thing it only does it at that rpm
so i'm thinking maybe the backpressure is too much for the wastegate spring.
The exhaust is a crush bent 3'' so it goes to a 2.5'' on each bend.

The MSextra forum guys think it could be the BOV opening up but i leak tested to 30psi tial Q held shut.
strange strange strange

I too didn't got the closed loop boost control to work in ms2. In 3.2.5 the closed loop control was joke and solenoid had 100% duty all the time. In 3.4.2 fw I got it working as op has, so first at 0%duty and then duty has that huge sawtooth look. Not very accurate boost control at all. Then upgraded to ms3 and all closed loop boost control issues went away and it follows target perfectly with nice looking duty graphs. No changes in hw. I too think it's problem in ms2 code, but ms guys seem bit too proud to admit that there might be problems in the code. And for me it looks like we will not get updates to ms2 code any time soon (same in ms3...) so solution for op would be to upgrade to ms3 processor.