Some of you will have read about my engine failure in this thread.
https://f80.bimmerpost.com/forums/sh....php?t=1726570
Back then I didn’t want to go too much into detail as to what happened exactly until I had more information. Well the engine tear down is finally complete and I’d like to share more about the damage that happened and hopefully gain some insight into what actually caused the damage and how to prevent it again in future.
I hope the discussion in this thread can be informational and perhaps more members with failed engines can share more about their own experiences as well
Engine mods:
2015 F80 M3
FBO
BM3 Paul Johnson tune
JB4 WMI (dual GPH12 Prometh nozzles in charge pipes)
100% meth
Carillo conrods
CP Pistons
Pure Stage 2+ turbos
Gintani hub
I purchased the car used, engine had already been built by previous owner. I started a new tune process with Paul Johnson because previous owner had an E30 tune on it with port injection and I wanted to run only pump fuel + meth.
The tune process was fairly smooth and Paul was very responsive throughout the entire process, starting from 23psi boost building up to the final 28-29psi revisions after each log (at least 8-10 revisions to get there). During this time I had also experimented with different WMI nozzle locations, as I had slight issues with uneven fuel trims between banks as pointed out by Paul. The final charge pipe locations seemed to even out the fuel trims the best compared to J-pipe and post intercooler locations.
The engine failure happened during one of my highway logging sessions. I had already done about 4 logs in 3-4th gear up to redline and shifting into 5th. On the final run, the engine started stuttering badly and seems to be misfiring, there was no fault code and I managed to drive the car home. These are my BM3 logs of the actual night of engine damage:
https://www.bootmod3.net/log?id=5ec6...729b509821b4bf
https://www.bootmod3.net/log?id=5ec6...729b511c4e6fda
https://www.bootmod3.net/log?id=5ec6...0b434c28f5b419
https://www.bootmod3.net/log?id=5ec6...0b434cf6b2b340
* one thing to note here is that I suspect that BM3 logs number the cylinders backwards. I’ve seen past logs that all show my #2 pulling timing consistently, and ended up that my #5 cylinder had the most damage. Paul has mentioned he does remember noticing sometime like this before as well on another car. I reached out to PTF about this and they say their parameters are exactly as BMW designed
I’ve to mention that I’m not very technically knowledgeable in engine tuning and what I’ve learnt up to then was purely from Paul’s feedback on tuning parameters. What is important to note what I had requested a track map from Paul specifically something that would be able to sustain extended highway runs at high RPMs. Up to and after the engine failure, Paul has maintained that the map he wrote for me was not an aggressive map and in line with what I had asked for, even after looking over all the past logs.
At this point there have been various assumptions and theories online about other engines being damaged under Paul’s tuning but I will not comment about any of that and will only discuss my own personal experience. Many have said that Paul de-sensitizes the factory knock sensors to run higher ignition advance. I specifically asked Paul about this and he replied that he had not changed any settings on my knock sensors at all.
This was his reply in his own words:
Quote:
“No we don’t desensitize knock on pump gas tunes. Looking at your plugs it seems that all of your issues happened in only one cylinder. I suspect a fuel injector or meth distribution problem during that particular pull causing a lean condition in that one cylinder resulting in the spark plug heating up and causing pre-ignition (becomes hot like a glow plug). It could have been an injector clog or failure, or methanol cylinder distribution at the time of that particular pull, possibly due to meth momentarily stopping flow or reduced flow resulting in no atomization, possibly tank low resulting in air being pulled into lines under g-forces, etc. The O2 sensors see 3 cylinders at a time 1-3 and 4-6, so if only one cylinder has a fueling issue and the other two on that bank are fine the dme will not know about it or be able to compensate for it. An overall fueling situation the dme can compensate for but if only one particular cylinder goes very lean from the conditions I mention it will not show up as the O2 is reading mixture from 3 cylinders combined. This is how one particular cylinder could under certain conditions get much hotter causing the glow plug pre-ignition I described.
I really try to stay away from the forums and the drama they cause. Everyone has different opinions and theories.
I hope my thoughts on this helps. I’m sure you have see throughout the tuning process that I do my very best follow safe practices to keep the motor healthy. Starting low with the base map and ensuring the logs indicate that everything is running properly and clean before making any increases in a revision. Then only making small increases at one time.”
|
In order to facilitate a quick return back to the road, I purchased a BRAND NEW S55 engine from UK (BIG shout-out to AP Motorsports for the great help with fast shipping). Once the car was back up and running, we started the tear down process.
Firstly the spark plugs were pulled and #5 had completely melted off.
Dismantling the head showed the following damage:
#4 exhaust port severely misshapen/ melted, some damage to spark plug threads
#5 spark plug seat cracked/ melted, valve seat damage, exhaust port damage
#1-3, 6 looked ok.
Piston slight marked on the top. Block still looked ok. Pistons all looked really black which makes sense, seeing that the engine was previously consuming 1l of oil per 1,000kms. Previous engine builder did not do a great job with the forged pistons install.
Several workshops I spoke to all suggested damage was caused by running too lean in that cylinder, and causing extreme temperatures to be able to melt the cylinder ports. But the actual cause of lean is hard to determine. I have tried to get my fuel injectors bench flow tested to no avail. I’m still looking into it. It could be failed/ clogged fuel injectors, uneven distribution of meth or simply tuning error.
Another theory would be that the engine was already not healthy when I bought it, and my tuning process/ logging pushed it over the edge. This was my thought after looking at how bad condition the cylinder head was in. But of course I can’t prove this either.
So now the engine is junk and I’m parting out for spare parts/ accessories.
I’m currently in the process of Ecutek tuning and setting up a new direct port injected WMI system with Cortex controller.