Probably a good idea to just cryogenically freeze them until needed /ubbthreads/images/graemlins/devil.gif
I've been following this thread for a bit and even sourced a possible alternative replacement. I created a plug n play harness for the Miata sensor. The sensor is made by mitsu and looks exactly the same as ours down to the goo and all. Plugging it in to see if it throws a CEL or not is easy enough, but...
Biggest issue, as pointed out in the tuners thread, is determining if it has the right frequency filters. No one has been able to get any manufacturers to confirm if the the filtered frequencies are the same, and there aren't any very scientific ways to confirm it. As of now I plan to run a known good DSM knock sensor and make a few logs, then swap and try to replicate and compare results. If anyone wanted to test the Miata sensor, I would be happy to work with them, but can make no guarantees about it working, just speculation based on someone from tuner's experience. As an alternative, I may try a standalone knock detection in conjunction with both sensors to see if they pickup the same knock detected by the standalone. Its a little risky to run these while making a pull with potentially bad knock detection, but there aren't a lot of alternatives for confirming.
Ultimately, it's not a very scientific approach. Knowing from Mitsubishi what the filter frequencies for these sensors is would be the best, but short of that, I assume someone would need an oscilloscope or similar equipment to test the filters. No one is stepping up with a scientific approach, just a lot of speculation (Miata and CSM sensors look the same, made by Mitsu) and I think we're all really hoping something works.