Upgrading Firmware from pre3.3alpha5 to 3.4.2

[robd]

Hi All, I'd like to update my firmware from pre3.3alpha5 MOD gslender 2.7 to the latest 3.4.2 release. My car is a '96 BMW E39 528 turbo running MS2 extra on 6cyl wasted spark using stock 60-2 crank wheel. I've had the old firmware on the ecu for a long time(used on a previous car) and the car runs really well on it, but I want to update and make use of the improvements including decent CL Boost Control with the initial table values etc.

I've tried updating to the 3.4.2 firmware a couple of times, but the car just doesn't seem to run right afterwards. Initially my AFR's are much leaner and the car misfires constantly under boost. I've spent a while dialling the AFR's back in but it's the misfiring under boost that puts me off continuing to tune with it as I figure I must be missing something, probably obvious!

My question was, other than the changes to Baro Correction and Mat Correction, is there anything else that would change my AFR's so much and anything else I need to be aware of that I should be changing for it to work? I've gone through checking all the settings are correct like Falling Edge, Going High etc and I also chose to use the old baro calc as well.

It's possible I just need to stick with it and completely dial the tune back in, but it just seems strange it runs so badly and as soon as I switch back to the old firmware it's fine. It's back running the old firmware now.

Any suggestions would be great. I was hoping to be able to update the firmware, check settings etc make a few adjustments and then focus on the CL boost tuning.

I've attached a copy of my working tune on the old firmware and the new tune I was trying with the new 3.4.2 firmware.

Thanks very much
Rob

[rickb794]

Have you re-calibrated your sensors since installing the new firmware?

Did you start a new project and reconfigure the new default msq for the firmware change? If you try this you can import the tables from your previous tune.

I'm not familiar with that old version, but I suppose it is possible that fueling calcs have changed significantly.

[Matt Cramer]

Does the fueling behave normally with 3.3.x release firmware, or is it running like it is on 3.4.2 on that code as well? The GSlender fork is not as well known as the standard version.

[robd]

Have you re-calibrated your sensors since installing the new firmware?

Did you start a new project and reconfigure the new default msq for the firmware change? If you try this you can import the tables from your previous tune.

I'm not familiar with that old version, but I suppose it is possible that fueling calcs have changed significantly.

Yeah I re-calibrated all the sensors. I used my existing project and just updated the firmware and then loaded in my old tune and went through all the settings. The only things I had to change were Baro Correction and the MAT Correction values were different. Everything else appeared fine as most settings in the GSLender 3.3 code were similar to the 3.4.2 code.

Does the fueling behave normally with 3.3.x release firmware, or is it running like it is on 3.4.2 on that code as well? The GSlender fork is not as well known as the standard version.

Sorry do you mean to try a standard 3.3 release firmware instead of the GSLender code? Fuelling is fine on the pre3.3 alpha5 gslender code, car drives really well. I like how it drives on this code but the main issue is its not using the better CL boost control among other things.

Cheers
Rob

[Matt Cramer]

Correct, I'm asking you to try a standard 3.3.x release. There's a lot in the GSlender fork that is hard to find documentation for, so I'd recommend narrowing down if this is a 3.3.x vs 3.4.x difference, a GSlender vs standard code difference - or maybe a bit of alpha code weirdness.

[robd]

Ah ok, thanks very much, I'll try a standard 3.3.x release then and see what it's like and update.

Cheers
Rob

[robd]

Just noticed a comment in the 3.3.3 release notes about a previous bug with the tooth#1 angle...

"18. Some "trigger wheel" users may need to change their tooth#1 angle if it was close to 360deg or 720deg. (A previous bug was fixed.)"

My tooth#1 angle is 334 so not sure if that will effect me, but I'll check it once I load the 3.3.3 firmware and see if it's off. Maybe that could explain why the car was feeling generally under powered on the 3.4.2 release if the timing was off!

[robd]

I sorted the issue out. I noticed that when setting the trigger angle on the old pre3.3alpha firmware it didn't actually adjust the timing when in fixed timing, but once I switched back to 'use table' it did have an effect! I'm now on the 3.4.2 firmware and I just had to enter the trigger angle again and confirm the timing was correct, as it had been out by 10*!

Thanks for the help.