It's been way too long for updates on this. The delay was due to having no really good news in 2015 and for the first half of 2016. The car was a crap show of problems that culminated in a complete failure as I was pulling out of the garage to load the trailer for MRLS2015 last September. That thread is commemorated here:
http://www.mazda-speed.com/forum2/index.php/topic,31728.0.htmlThe MSM was a champ and shocked a few caged track beasts over the course of the weekend. I pretty much shelved '90MSM for the first part of the winter but at some point, I started digging back in and figured out the issue. Here is the abridged version starting about 14 months ago:
The first sign of trouble was the miss at the end of the dyno and subsequent failure at the track mentioned in the 6/29/15 post below. I don't remember all of the details but a new MAP sensor got things running better. A wire on the IACV broke at some point and during this period, the car wasn't idling particularly well either. My next trip out, I dropped a valve which ended the day. I don't have a lot of excuses or explanation - I must have left one of the keepers slightly out of whack. I pulled the head and swapped in the spare without much excitement. The machine shop fixed the original which was back in action sooner than I had hoped as you will soon read...
The next trip out (I think - last summer of was blur of bad memories I'd rather forget) I developed a strange misfire that would cause the motor to vibrate horribly and coolant to spray on the windshield. A full discussion of this is probably beyond the scope of this thread but in short, I believe ECU was going haywire in random circumstances and causing the engine to fire at strange points which forced coolant out a suspect head gasket. Compression was fine and I tried some things to hopefully fix the misfire. That day was cut short and I ventured out on my next attempt where the misfire continued randomly and continued the coolant on the windshield. When that wasn't happening, the car was running fantastic but I resisted the temptation to finish the day. I later determined that was a loose crank trigger wheel. Despite blue loctite, it had come loose but at least we found the problem, right?
I got back to the garage and pulled the head finding that there was indeed a coolant path out past the HG. I cleaned up the block, set a new head gasket and placed the original head that had been repaired after the dropped valve. This yielded balanced compression numbers and the car stared up but didn't run well. I was still chasing an electrical gremlin even though the trigger wheel was back in place. I turned 9 laps at The Ridge Motorsports Park on 9/26/15 - just days before leaving for Laguna. I don't remember what happened to cut the day short, but I thought I had fixed it because I don't remember thrashing in the garage leading up to the depature. I got all the way to the morning of the load - started the car, let it idle for a few minutes and then, shifting into reverse... it died. That was the last electricity to flow correctly through that ECU harness.
At some point last fall, I found another Adaptroic 420c thinking that was the problem. I went this route because a) used 420cs are pretty cheap and b) it would be the easiest plug and play fix if it worked. I plugged it in and the car started right up... for a minute and then died just like it had before Laguna. Hot Damn!!! The problem must be in the harness. That adapter harness had been around since I started messing with ECUs in 2009 so it made sense that it was failing. I took a look at the PCB abomination and realized I would never find a clean one and Adaptronic didn't make them any more. Fortunately, the ECU I had bought came with an NB1 adapter harness and while it didn't fit my engine harness, it had the female ends that go into the ECU. I bought a generic NB2 harness and started soldering.
Using the original harness for inspiration in addition to the Adaptronic schematic and the MSM wiring diagram, I painstakingly wired up a new harness one wire at a time. When I found two wires I thought I should join, I checked them for continuity on the original harness. Near the end of the process, I found the lead to the Cam Angle Sensor on the old board that didn't have continuity... then it did... then it didn't... The wires were correct, but the trace on the board was broken. Eureka I thought!!!!! The whole problem is solved!!!! That explained a lot of the symptoms but I was very close to being done with the new adapter harness so I finished it up. This CPS trace was another false breakthrough... or was it?
I plugged the new harness in and the car started right up. It didn't idle very well but you will recall the IACV was suspect so I didn't worry... until the car shut off spontaneously mid idle. Nuts. I actually said a 4 letter word a little stronger than 'nuts' but forum policy suggests I don't transcribe it here. I tried the other 420c and the car would not fire at all. At this point, I considered shipping the car to FM or Trackspeed or KO Racing or anywhere else with a credit card and a note taped to the windshield but I didn't. I already had a brand new Adaptronic Select ECU I got from Andy Down Under back when I was first building '90MSM that I never installed. I had planned to put it in the street car at some point since it was truly Plug-n-Play but hadn't got around to it. What should I do?
Clearly, the car has a gremlin somewhere so I wasn't going to do a damn thing with this engine harness. I bought a new complete MSM harness and started converting the wiring. Doing the new harness from scratch gave me some time to get acquainted with the only complicated part of the process - The X-04 connector. The MSM engine harness (and likely all NB harnesses) have 3 plugs at the ECU that don't really get modified and the X-04 plug that interfaces it with the rest of the car's wiring harness. It has 42 wires in it and about a dozen of them need to be wired into whatever your are doing to get the engine to run. Here is a thread I put over at MT to discuss my frustrating progress:
http://www.miataturbo.net/general-miata-chat-9/wire-fn-wire-hoping-turn-angry-lap-september-89208/* Use caution posting on MiataTurbo. They have the temperament of a pack of feral cats and don't take kindly to new people if they perceive a question to be beneath the level of the group...
I got the new harness in and 4 power leads, 3 ground points and 6 'other' items (neutral switch, fuel pump, etc.) wired up. I got on the Adaptronic forums and got some help converting my 420c map to the Select and plugged it in. Backfires... lots of cranking and backfires. We started the long and painful process of figuring out what was still wrong with the car. Sensors passed a sanity check, trigger lights were blinking but I couldn't even check base timing because the car wouldn't fire enough to run for even a minute. Mark Gool from Adaptronic support traded emails with me every day for a week or two and was an absolute rockstar. A screwdriver and 'Let's check everything' finally got '90MSM off the operating table and back to the track.
Checking the triggers, we saw a good crank trigger but the cam trigger was still acting strange. I pulled the CAS out of the car and waved a screwdriver in front of it and everything blinked fine. With it back in the cam cover, nothing consistent. It was then the sinking feeling hit me. I took the cam cover off and found that the nubs on the intake cam were on the exhaust cam -
I had put the cam gears on backwards. Can you believe it? I sure didn't...
Back up to the previous summer when I popped the other head back on - That was when I made the mistake but
the car ran after that point. It turned laps on the track after that point. What the hell? It appears the 420c has a failsafe mode Mark didn't even know about that ignores the cam sensor if it isn't there. Going all the way back to the first hiccup on the dyno in June 2015, my issue was some derivative of a cam angle sensor problem - either a dry socket confusing the ECU, the missing signal plus the dry socket killing the car and finally the new ECU refusing to fire without a clean CAS signal. Like I said earlier... Nuts... If the car had failed to start that day after putting the new HG on, I would have found my error in under an hour. The fact that it started and ran pretty well completely through me off the trail of thinking about a mechanical problem. Nuts...
Mark and I had a few more adventures setting up a 36-2 trigger wheel since the 4 tooth one cobbled onto the ATI damper had come loose all those months earlier and I wanted something better anyway but after that, I went out to the track and had my first successful day in '90MSM since October 2014. There were no lap records but also no electrical issues. I did have a few comical things happen I'll skip for today but in short, '90MSM is back to full speed ready to make mischief at MRLS in October. Who's coming out to play?
