The firing pressure transducers I deal with at work are supposed to only be reliable for around 100 hours, so I have my doubts that a production type transducer that is expected to last thousands of hours is realistic, and a problem waiting to happen. Like you, it seems like every time we set up firing pressure it is finicky. I've spent far too much time swapping transducers and troubleshooting.
They were a real pain to work with. I learned quite a bit and there was some great information in there, but yeah, difficult.
I haven't kept up with what BMW is doing, but I thought they were gleaning some information from feedback in the ignition system rather than directly measuring firing pressure?
I don't recall any details, I just recall that they were using something to dynamically change spark advance based on real-time conditions, rather than lookup maps. I had thought they had some sort of sensor that fit between the spark plug and the cylinder head to determine that, but this is a very vague memory that I read somewhere some years back, so don't quote me on it.
I've done valvetrain testing using accelerometers, I hadn't considered using it to try and measure detonation. I have wondered if strain gages and high speed acquisition couldn't be used to get an estimate on when you reach peak firing pressure relative to crank position, in lieu of having a dyno/test cell to run in. It might be interesting to try.
Like I said, in the old days (when all of our aircraft engines were certified) that's how it was done. The same concept is used in most "modern" (1990s onwards) cars that have knock sensors, but I suspect those were instrumented better. The accelerometers did the job, although the setup did have a lot of human intervention and thus room for error. When I was using probes we were able to set the computer to record detonation events and get much more repeatable results.
I have another question for you Ted, since you're probably the only one here who has actually run an instrumented aircraft engine in a test cell. Can you actually reach MBT on a typical non-turbo aircraft engine or will it start knocking before you get there?
The simple answer is that the factory ignition settings were actually not too far off from MBT at rated power (full throttle, max RPM, and the best power mixture setting). I tended to find on the higher power engines (which was most of what I ran) that advancing the timing further had no difference on torque (at least, a difference that was within the margin of error of the dyno). On the low power engines you couldn't make the things detonate on 100LL.
Now for the more complex answer. One thing to note with aircraft engines is that we have a wide range of CHTs to deal with, and detonation is heavily influenced by CHTs. So is power output. At high CHTs, your power output goes down (one of the man reasons why good baffling on aircraft engines is so important). Plus we have a wide range of mixture settings to deal with.
So, the answer to your question is more complex since it depends on those factors. When doing typical power tests, the cooling air was typically more at a typical cruise power type setting, which kept CHTs in the mid 300s (fairly typical for the aircraft that the engines were put in). This is different front detonation testing, where CHTs had to be close to redline (I forget the exact number, I think the hottest head was required to be within 10F of redline), oil temp within some temperature of redline, induction air 100F, etc. For detonation tests on the higher power engines, you typically were limited by detonation before you got to the best power setting. But at max power you're typically running at full rich mixture, not at best power mixture.
I think you're probably asking about when doing detonation tests rather than the standard power tests with more cooling air, in which case the answer is that you can't run a best power mixture and MBT, you'll get into detonation for higher power naturally aspirated engines, but you can for lower powered engines.
The air cooled aspect really does make things interesting from an aircraft side. With a water-cooled engine, you more or less have a constant engine temperature, or at least constant for a set of conditions. With air-cooled, you can have huge swings.
