Well, the normal megasquirts do have a filter circuit on their power supplies (after the 5v regulator) but it certainly couldn't hurt to try installing a beefier filter on the 12v line such as the one you linked to. If you are in fact getting processor resets on the primary CPU, it's conceivable that the aux CPU could also be flaking out. No easier place to start than at the power supply.

Please tell me if this idea is feasible:

I could temporarily install my stock Mazda ECU in the car - and keep the MS in there, as well.

What I have in mind is providing all power and ground inputs necessary to make the stock ECU operational, and just use it as a voltage regulator, for testing purposes.
That way, the MS will be (temporarily) relieved of its voltage regulating duties, and I will be able to focus on other aspects of its operations.

Would I be able run the stock ECU that way, with just power, ground and alternator input/output and nothing else?

I would suspect that you will also need the crank and cam sensors to be shared, so that the stock ECU will know that the engine is running and the alt needs to charge. Unfortunately, you can't share the crank/cam sensors.

The above is only a theory, but you can test it if you want.

Also, check your email for good news.

Thanks for the good news! YAY!

Wait.. why would a regulator circuit need to know if and at what rpm the engine is running?

I mean external regulators have no such provision. Voltage in, Voltage out, right? What am I missing here?

The stock ECU will probably not try to charge at idle and most importantly, during cranking. This helps with smooth starting (less load on the engine).

Not sure if the Good News™ email obviated the need for the stock ECU, but from what I understand, the NB ECU also requires an IAT signal in order to run the alternator. You could potentially fudge this with a 1.8k (or thereabouts) resistor between the stock IAT pin and ground, which will give the ECU a reading of somewhere around 90 degrees F, which is a normal intake air temp.

You can't just parallel them, as the stock ECU's pullup will throw off your calibration.

You don't really need it. The ECU will run and charge at a constant ~13.7-14.1V.

This is contrary to what I've heard from folks running stock ECUs, but I'll trust your experience here as my hands-on knowledge of NBs is limited.

Well, it was just an idea..

(Joe the mail thing is an entirely separate deal, not even remotely related to what we are talking about here, sorry for the confusion.)

I had a chance to observe the voltage/miss issues a bit further...

First off, I was able to get to about 14 psi with no voltage issues. I did it ever so incrementally on the freeway, bringing boost up very, very slowly. I also kept an eye on the voltage, kept dicking around at 13.7 to 13.9 Volts.

I then brought the car back to under 100 kpa, and got on it... This time I just got an ignition miss, and no voltage spike.

Then, I drove the car some more today, and got to 8-10 psi, with no problems.

I will throw in a set of iridium plugs I have, and test it again tomorrow.

Funny thing is, the car used to charge at 14.2 Volts with the MS. Now all I see is 13.8... Every once in a while it goes up to 14.1, and back to 13.6.

Drives me nuts.

Problem solved (I think)...

Well, I was partly right, and mostly stupid, with a dash of lazy.

I replaced the current BKR7E-11 plugs with the stock 1.1 mm gap with a set of iridium plugs I had laying around, and all symptoms are poof, gone.

So, my guess is the wide gapped plugs were just flaming out under the new found higher boost, and the coils were creating havoc back in the ECU.

The thing pulls like a (insert glorifying adjective here) all the way to redline, under full throttle now. No misses, just me and the Mrs. now... ;)

Wait.. Here's a couple logs, too.

Yay!