Quote:
Originally Posted by Boss330
Don't remember having had arguments with you... 
I'm not saying that I know what caused his strange dyno reading, but not everything adds up and some claims just seem to contradict engineering basics on engine management, ignition timing influence on torque, knock only happening at high RPMs and the DME not being able to reduce boost as a function of knock or fuelling.
But, I'm keeping a open mind and look forward to more in detail explanations on what was meant. |
Bringing you guys up to speed on proper analysis here is a rather exhausting unpaid job for me. I think the best thing would be for you to strike up a conversation with BMS on their forum if you want more in depth discussion on the subject. I will answer a few final questions for you and bow out of the conversation.
First the DME is perfectly capable of reducing load as a function of knock. It's simply not part of BMW's logic path. It's not programmed to work that way. When knock is detected the DME reduces timing by 3 degrees in the cylinder it was detected in. Boost remains the same. When repeated knock is detected in the same cylinder the DME again reduces timing 3 degrees. If knock continues the DME decides the cylinder is experiencing preignition rather than detonation and a zero boost limp code is triggered. As knock is not detected timing is slowly increased back up to its learned maximum. The learned maximum is based on a long term octane adaption trim.
If you want to test this for yourself monitor boost/timing while on 93 octane, then pump in a few gallons of 87 octane, note how timing shows knock indications, and how boost is unchanged.
When I say torque is largely independent of timing what I mean is that as it goes through this process of reducing timing 3 or 6 degrees those variances have a subtle impact on peak torque figures. But a dramatic impact on peak HP figures. The "torque loss" in the dyno is not consistent with knock in my professional opinion. If you'd like some additional evidence here please take it up with BMS directly on their forum.
Here is a screen shot of the load to knock table from an MSD81 DME. The more negative the index the higher the "knock". As knock increases at lower RPM the DME actually raises boost slightly. Even at it's most extreme boost reduction would be around 6% (this factor is multiplied by a load target that runs from vacuum to full boost). I happen to know the DME limps out with "super knock" codes at around -70 and never actually gets to the point that the logic dictates lowering load significantly. This is how it's been programmed from the factory. The s63tu uses the same basic table values. Again, if you chose to doubt this, please take it up with BMS directly.
Finally, neither of you answered my question as to why the 100% stock M5 running 94 octane, DCT, in 5th gear, "lost" 50 wtq in the midrange in the dyno chart I posted above. There was zero knock during this run. It "lost" torque because the DME decided to lower its load target in the midrange to protect the DCT trans. I believe the same thing happened during the dyno we are discussing.
Mike