Turbo boost and MSS54 torque and load managers

[rmz001]

Hi all, I'm in the middle of tuning my e46 m3 turbo's DME to run under boost.

I seem to keep tripping the emergency program for some reason. At times I'll get a full-throttle run-in, but once I slow down, the emergency drive program will kick in by design. (It's a very neat safety system)

Here is what I've tried so far.

1. Increased the values in KF_MD_MAX_MD_IND_OPT (MAX OPTIMAL INDICATED TORQUE), the RF axis now goes from 0.07 to 3
2. Set K_MD_SK_MAX to 900nm
3. Set K_MD_SK_LLR_MAX to 900nm
4. Set K_HFM_DIAG_ML_MAX to 2500 kg/hr (load manager)
5. Both lambda controllers deactivated, (all 4 o2 sensors plugged in)

The car will suddenly, have a 4k rev limit. (otherwise feels fine)

Do I need to also adjust the KL_RF_N_AQ_REL for >1 RF value since I wonder if the Alpha-N (RF) lookup is used as a check against the HFM/MAF computed load/RF?

Anyone got any tips?

Thanks!

[Tomba]

Which error code do you get?
Are you running alpha-n or MAF?

[rmz001]

Which error code do you get?
Are you running alpha-n or MAF?

MAF - pmas hpx

The only dtc that was thrown was the low fuel cat protection code.

I have cat protection disabled via setting the fuel trim factor to 1.0

I’ve checked the fuel levels in the saddle fuel tank and it all seems fine. With the passengers side mostly full.

So I disabled the dtc for low fuel.


After some thought. I’m thinking it might be my lambda sensors. I’ve got both lambda controllers deactivated as I have a wide band and have been tuning with that. All 4 o2 sensors are plugged in but not in the exhaust system and dtc codes disabled.

Just to avoid the dual bank fuel trim issue with a turbo manifold. My theory is that having no lambda signal at all (0 mV) may cause issues as the sensors signal might still be used by one the safety programs. So I’m going to install a pair of simulators for the pre-cat sensors as a test.

Other than that the car runs great on e85 and boost for a bit before I get capped at 4k.


Sent from my iPhone using Tapatalk

[S54B32]

MAF - pmas hpx

The only dtc that was thrown was the low fuel cat protection code.

I have cat protection disabled via setting the fuel trim factor to 1.0

I’ve checked the fuel levels in the saddle fuel tank and it all seems fine. With the passengers side mostly full.

So I disabled the dtc for low fuel.


After some thought. I’m thinking it might be my lambda sensors. I’ve got both lambda controllers deactivated as I have a wide band and have been tuning with that. All 4 o2 sensors are plugged in but not in the exhaust system and dtc codes disabled.

Just to avoid the dual bank fuel trim issue with a turbo manifold. My theory is that having no lambda signal at all (0 mV) may cause issues as the sensors signal might still be used by one the safety programs. So I’m going to install a pair of simulators for the pre-cat sensors as a test.

Other than that the car runs great on e85 and boost for a bit before I get capped at 4k.


Sent from my iPhone using Tapatalk

I HIGHLY recommend not disabling DTC where you not know where they come from. („Low fuel..“). If remember correct, they are triggered by knock sensors if there is pre detonation or no detonation. Ecu will think there is to low fuel/fuel pressure, so for ecu the tank is empty. Had this dtc also with way to less fuel pressure. Maybe others will chime in and can be more precise on this particular dtc.

I had driven LOTS of miles with no problems without a single o2 sensor connected, only with wideband gauge and logger on both banks. But nothing connected to ecu. Disable fuel trims/force lambda controller not ready and delete dtc. Working 100%.

[rmz001]

Thanks for the color. I’ll remove the o2s from the car and restore the low fuel dtc and report back.


Sent from my iPhone using Tapatalk

[Tomba]

I doubt your lambda sensors have anything to do with the problem you are currently interfering. Your ECU measures load from MAF and adjusts fueling to it based on values in the tables. I would recommend to change the load axis since you are running above 100% load.
Do you have logs and ori and modded file I can have a look at? Kinda guessing without this information.

[Tomba]

Just see the PMAS MAF is different from stock. Did you change the linearization kenline in the ECU?
x axis would be voltage, y values would be air mass in kg/h.

[rmz001]

Attached are the MAF transfer, Load Base, and Torque map I'm using.

As you can see I've updated the Load axis to go up to 3, and the MAF values I've been fitting/tuning with my wide band as the pmas transfer function needed to be updated for the load table to match what I see on the wide band.

I've got EGT disabled, and the lambda controllers set for forced open loop (disabled the controllers)

I'll PM my file directly.

Thanks for the help!

[Martyn]

Have you logged load / RF to see what you're hitting?

[rmz001]

Hi Martyn, I have the boost running on wastegate spring pressure so far. It's only around 2-3 psi. With that boost, when I can get a full-throttle run, I hit around 1.3 RF.

Typically, once I start driving, I can get one full-throttle pull, but once I come to a stop, the emergency program kicks in. Based on my read of the MSS54 function document pdf, this is by design as the DME does not want to disrupt the driving condition, so it waits till you've come to a stop, then the limits kick in.

I found the table(s) that relates the emergency run program to torque and engine speed limits. I made the torque 900nm and engine speed 7600 vs. 4000. I think I'm in emergency program 2, as the limit is 4k. That said, even though I've hacked the emergency run program limits as a test, I still get capped at 4k...


Just so I am clear, my goal is not to trip/hack the emergency drive cycles, but I am experimenting/troubleshooting.

[rmz001]

Fixed ! Here is what I did and discovered.

Step 1 - Took a step back and applied some advice from this thread.

1. Restored many all of my torque manager and emergency control programs back to stock.
2. Restored all DTC's

Hardware changes.
1. Since I have both lambda controllers disabled via the control flag parameter. I removed/disconnected them.
2. This tripped the o2 sensor heaters, updated the DTC's accordingly
3. Replaced a faulty EGT sensor, and also turned off the EGT control.

I noticed when reviewing each parameter in the tune VS stock.. I found that I had set the lambda, FMIN/FMAX to 0 instead of 1. I had also not changed both the LA and LAA parameters.

After that.. no DTCs, many full throttle pulls with low boost ... Wideband logs looking good on e85, timing looking stable, etc.

Next up, tuning the off-idle tip in as it leans out..

Thanks all.. more to follow as I move up the boost spectrum.