Increasing line pressure by increasing reported torque

[Sbrach]

Some of you probably saw on FB in the DCT group but I have been measuring DCT clutch and system pressure via INPA in an attempt to get some verified data regarding the differences between E and F series tq holding capacity.

My car is a 2011 E82 w/ PS2 running around 22psi. EGS is flashed with the 135is file via WinKFP.

Popular opinion is that the F series cars hit 18bar on the clutches and that E series was limited to something less. As you can see in the screenshots, I hit 16.76bar at around 800nm.

Increasing load to torque in an attempt to report the max 1024nm at the clutch resulted in a NMAX torque limiter (rev limit) that didn’t actually reduce load req, timing, or anything until 6200rpm where it dropped load req hard. Very weird.

I’m back on stock load to torque now but if anyone with an F series or an e9x wants to try to grab the same data points I think the comparison would be interesting.

 

[houtan]

Could that nmax torque limiter at 6200 rpm be the reason why I hit limp mode around the same rpm (6000ish) when I have boost above 19psi, even though my log looks perfect?

 

[Jake@MHD]

Steve, what's the approx reported torque where the 2048 lim popped up?

 

[Sbrach]

I’ll post the L2T table I was running but 2048 popped up as soon as full throttle was applied. I think it was hitting a driver torque request limit. Not sure why it didn’t do anything to reduce tq until the limiter dropped off at 6200 but I have a lot of other things changed. Would be interesting to test just increasing the L2T on an otherwise stock car.

 

[Sbrach]

https://datazap.me/u/sbrach/rev110e?log=0&data=4-13-18-33-34-35&solo=18-33-34-35

 

[RSL]

Nice info. How well does log output line up with the actual INPA values? I'd go out and run mine, but laptop is dead and adapter is in storage

It may be rev limiter limiter at 6200rpm, but you can hit 2048 at barely any throttle/RPM (E N54) and not even in prep for/during a shift, so classifying 2048 strictly as a rev limiter limiter may be a bit of misnomer. It's obviously triggered during other scenarios as well.

I assume you have torque cap and limit offsets maxed. Not sure on N55, but N54 stock cap is 1000nm and will toss a limiter 1 (which is also DTC) when you reach it. 998.6nm is no limiter, 1000nm is limiter 1. Everything else goes as planned (boost, load, etc.) during the limiter. I was just messing around a year or 2 ago to see if I could reach the max without issue. I didn't try raising the cap or shifting closer to redline (let off at 5600 here), just called it done and flashed back to normal lol

 

[Sbrach]

I think 2048 is strictly rev limit but various things can cause a reduced rev limit. For example before I found some wheel speed plausibility tables the logged speed value would drop to zero past 171mph. This caused a revlimit tq limiter because at 0mph the rev limit is reduced.

 

[Jake@MHD]

but you can hit 2048 at barely any throttle/RPM (E N54) and not even in prep for/during a shift, so classifying 2048 strictly as a rev limiter limiter may be a bit of misnomer. It's obviously triggered during other scenarios as well.

What? lol. Tq Lim 2048 is definitely max rpm.

 

[RSL]

I'll try to dig up logs, but they're years old. I could easily hit 2048 at part throttle and less than 3000rpm. Torque monitor adjustments averted it. In my case, seems like a monitor caused a limit reduction like Sbrach said, but it was actually a torque problem, despite a rev limit limiter being called. I could hit a 2048 pulling out of the driveway if I tried lol

When I think rev limit limiter, I think shifts and high RPM, that's just not always the case.

 

[Sbrach]

Lots of things can induce a lowered rev limit. Vehicle speed, oil temp, gear, EGS or DSC issues, etc.

 

[aus335iguy]

Could this be used for 2 step then ;P

 

[Sbrach]

Jake would be the guy to talk to about NLS, LC, Anti-lag, etc that needs custom logic.

I’m just working on getting stock calibration tables defined and understood so we can fix as much as possible via calibration rather than custom logic. The good news is based on the 1000whp S55 with MEVD17 I think we can get a lot done with just calibration compared to N54.

 

[RSL]

Lots of things can induce a lowered rev limit. Vehicle speed, oil temp, gear, EGS or DSC issues, etc.

Thanks. The usual things I was aware of, but honestly hadn't given much thought to the limit being pulled down past that and mine revolved solely around torque changes, which made it difficult to think its response was actually a rev limit. Makes sense though and seems blanket solution for a lot of situations, where ever they actually stem from. I could hit it at will cold/warm, 5-70mph, any gear, etc. in varying states of testing, but found my notes for it (2015 lol) and they indicate a 2D59 set, so definitely a torque issue, and dropping the rev must be the go-to response.

Back to the actual line pressure testing...

 

[Jake@MHD]

@RSL yea it sounds like whatever monitor you were exceeding caused a limp style rpm limit to get set.

Could this be used for 2 step then ;P

No lol. Even regular DCT launch control as is doesn't keep the throttle fully open, which I've had to change.

 

[RSL]

Not sure anyone is still following this, but in case anyone runs across it later, pressures that I logged. v1 is stock INA0S, about 500nm peak torque actual, v3 is about 540nm peak torque actual and v5 is about 775nm peak torque actual with a lot of other torque related DME params changed.

No one will ever see 1024nm engine torque in the TCU without some other changes and you'd immediately exceed 1024nm at the clutch if you ever did get it. Clutch torque target always exceeds TCU engine torque (so far). If it's possible to walk the line to actually get 1024nm at the clutches though, clutch pressures would be on track to easily exceed 20 bar. It's up ~6.3 bar in +275nm and would be another +250nm to get to 1024nm. I've seen 25 bar main line pressure spikes (not at WOT), so it at least reads high enough for that much pressure, but 25 bar may be max. It was flat lined there IIRC.

I'm testing a lot of individual settings now to see what/how the TCU responds, if at all. That will take a long while, but the wanted to confirm the N54 DCT pressure swing with torque actual for now.

 

[Sbrach]

Not sure anyone is still following this, but in case anyone runs across it later, pressures that I logged. v1 is stock INA0S, about 500nm peak torque actual, v3 is about 540nm peak torque actual and v5 is about 775nm peak torque actual with a lot of other torque related DME params changed.

View attachment 25251


No one will ever see 1024nm engine torque in the TCU without some other changes and you'd immediately exceed 1024nm at the clutch if you ever did get it. Clutch torque target always exceeds TCU engine torque (so far). If it's possible to walk the line to actually get 1024nm at the clutches though, clutch pressures would be on track to easily exceed 20 bar. It's up ~6.3 bar in +275nm and would be another +250nm to get to 1024nm. I've seen 25 bar main line pressure spikes (not at WOT), so it at least reads high enough for that much pressure, but 25 bar may be max. It was flat lined there IIRC.

I'm testing a lot of individual settings now to see what/how the TCU responds, if at all. That will take a long while, but the wanted to confirm the N54 DCT pressure swing with torque actual for now.

I logged 17.x bar clutch pressure reporting 900+ nm clutch tq. I believe there is a pressure limiting valve that limits it to 18bar. System pressure limit is 30bar I believe.

This was on N55 and was in response to popular opinion that M3/M4 cars run 18bar clutch pressure but E series runs much less. With increased reported tq on my E82 that did not appear to be the case.

 

[RSL]

I logged 17.x bar clutch pressure reporting 900+ nm clutch tq. I believe there is a pressure limiting valve that limits it to 18bar. System pressure limit is 30bar I believe.

This was on N55 and was in response to popular opinion that M3/M4 cars run 18bar clutch pressure but E series runs much less. With increased reported tq on my E82 that did not appear to be the case.

That is entirely likely/possible on a limit, have not gotten that high yet while logging. I wouldn't be terribly surprised if M DCTs are less limited. I assume the F series 18 bar were also tuned and they don't mean its stock pressure. Would be great if some of the E series M DCT conversion peeps logged theirs just out of curiosity.

Either way, E series is obviously in the 18 bar ballpark after some tweaks.

 

[JohnDaviz]

This is a great thread that somehow died..

The changes done to the torque tables. Was there any kind of testing for differences?
So far this is the only thread i found with information that raising reported torque to increase clutch pressure is possible and proof by measuring things.

There are probably other threads but my guess is that here with @Sbrach and @RSL we have to guys who know what to do.

 

[Stokes]

Would be great if some of the E series M DCT conversion peeps logged theirs just out of curiosity.

What do we use to log the pressures? I'm still M-DCT and E-series.

 

[RSL]

What do we use to log the pressures? I'm still M-DCT and E-series.

I use Testo (need a laptop), but it's a bit of a pain to set up (if you want English). It's basically a fork of Ediabas that records the values, so if you're familiar with Tool32, it's a plus. Need to add your DKG .prg to the ECU list in the folder and I basically just do the default read or detail read on it, I don't remember which. You could probably do a custom job and setup limited specific params only, but I just select what I want out of the full list on the datalog screen.

I think DeepOBD will log it if you don't have a laptop, but I have never used it.

I think we've come to the conclusion non-M DCT is going to be limited to around 18 bar at the clutches. Main line will hit 25-30 on pressure spikes though. The request was more out of curiosity at the time, but if you get the info, would still be curious to know 1) what your torque actual or tranny input torque is (can log it) and 2) what clutch pressures you see at it.

Feel free to shoot me a message if you run into any snags getting it going.