To be honest, I've got the Z3 engine with the crank sensor on the block, so my experience is different. Just going on experience/stories with other engines. Is there any chance that there's an O-ring or other crap in the bottom of the sensor hole that could be preventing it from seating fully?

We may be chasing our tails if that voltage on the graph isn't right. I'm thinking maybe some kind of signal filtering/interpretation issue. But these sensors should really be pretty binary in operation - either you get good signal or you don't. It doesn't make sense to me that these issues would get more frequent if it was really a misconfiguration. So yeah, if the sensor is within the proper range of the toothed wheel and not loose, AND you're able to get a strong signal when the engine is running, why would it go all cattywampus?

Are the ECU and PDM grounded together? I don't expect your proposed ground wire to do anything, but with the engine running, measuring the voltage between the ECU ground and PDM ground might be interesting.

 

So, I think I fixed it... and I'm kicking myself for not thinking of this sooner.

I went back and went through my wiring notes, and remembered that the IACV was originally powered off the same circuit as the Cam and Crank sensors (yes, that was dumb). The IACV (aftermarket) melted down almost immediately on the dyno and we disconnected it (the tuner tuned around it, and the car idles fine). After a quick check of the melted connector to make sure there wasn't continuity between the pins, I zip tied the connector to the harness to keep it from flopping around and we continued with the dyno tuning. Then I completely forgot about it.

I was thinking about things that were physical in nature and could degrade over time, because that's how this problem has felt from the beginning... not like electrical interference, but a loose/intermittent connection. At 3AM this morning, I woke up and the IACV connector popped into my head. This morning I cut it out of the harness... and no more misfire. I took it out for a drive and ran it to redline a few times, and it ran perfectly. When I got back I ran a trigger scope at 5k rpm to compare to the one I posted earlier (which I'll repost for easy comparison).

Before, showing the 'mystery spike' in red:


After:



So, here's the suspect connector:



No matter what I tried, I couldn't get continuity between the contacts, but something had to have been going on in there... because it appears to have solved the problem.

I looked at the log for my 8 minute test drive and had 4 trigger errors (none of which I noticed while driving)... the last log from an attempt to drive it with the problem had nearly 200 trigger errors in about 1 minute.

So... TLDR - I think it's fixed, but I'm not smart enough to understand exactly what was going on...

 

That’s quite the find! My only speculation is that something has to be touching somewhere in that connector with vibration at higher RPM. Can you try cutting it open carefully and inspecting it?

 

Could you take a look at the log and see where these errors happened? If they were right after cranking then nothing to worry about, but otherwise something is still amiss.

 

That's what I figure as well, but I couldn't reproduce it. I'll try to cut it apart and see what I can see...

 

They were all high rpm, nearly at the limiter and on different gears, so pretty widely separated.

 

Holy cow. Any chance that IAC connector was able to ground out on something? I'm not sure how else to explain that, unless it was intermittently shorting internally from the meltage.

Anyways, good find, hope this is it!

 

It was zip tied to something pretty solid, and nothing in the vicinity to make contact, so it had to be some kind of internal short.

 

Damn, nice find!

Having the information previously present, I don't think any of us saw that one coming lol.

 

Yeah, me either. I just wish I had thought of it before wasting two track weekends...

 

While super uncommon to run across, I have seen a previously melted connector that can make intermittent connection like this... My running theory is that black plastic uses carbon/graphite for colorant/additives and that the initial short/melting event can create a carbon trace short path between power and ground with the right voltage even if the metal isn't physically touching (knowing that graphite is a good conductor of electricity).

Really nice job of catching that and good to hear cutting the connector off seems to have fixed things. Excited to see what you think of the swap once these teething issues are sorted. It took me a track day or two to get my growing pains addressed on the swap and it's been green flag racing ever since.

 

Thanks. I'm now thinking that the low voltage a multimeter uses wasn't enough to 'bridge the gap' but 12-14V while the car was running was enough to short across the connector and cause the voltage spikes in the CPS signal.

I am really liking the car with the K engine. The power curve is almost exactly like an n/a BP, so the car doesn't feel much different to drive, it just gets through the gears a lot quicker. It'll sneak up on you if you're not careful, because the braking points are not where they used to be. The vibration is on a whole 'nuther level compared to a BP, even one with Blackbird competition engine mounts and delrin diff mounts. If there's any downside to this swap, that's it. That and the tight clearances...

I'm signed up for an AutoX this weekend. Not really my thing, but it will be useful as a shakedown test to see if the CPS problem is fully resolved. If it is, then I get to fix all the neatly loomed wiring I messed up trying to find the problem...

 

My Capri is on my lift, so the NA went up on stands for its pre-event inspection... everything looks good. AutoX Sunday... fingers crossed that the CPS issue is truly solved.

 

Good luck, man! Fingers crossed it's smooth sailing from here.

 

Good news! No crank trigger related issues at AutoX!



AutoX isn't really my thing... 8 hours in the sun for 6 minutes of track time is a pretty poor ROI, but it was a test/tune opportunity, and it was close to home. I set my cold tire pressures a little higher than usual, and didn't touch anything on the car all day. The surface (local police/fire training pad) is relatively new concrete, and slick. It was also very dusty, and they gave the first group (my run group) a 'clean-off' run before timed runs. Even then, there was no grip. All of the fastest runs were set by the second group after the track was cleaned off and had a little rubber on it. At the end of the day, I was fastest in my run group, 1.5 seconds off FTD (CAM-S C5 Vette) and 9th overall (out of 27). I was entered as SSM, and the only other SSM car was a Rotrex supercharged NB2 on 275 Hoosiers... I beat him by over 3 seconds. Overall, I was pretty happy with my performance, not being an AutoX regular, and happy to show the AutoXers that us road racers can wheel a car too...

I haven't had a chance to look at data yet, but I'm interested to see lateral G numbers, because they were way lower than I normally experience on track. I'm looking forward to actually quantifying the lack of grip.

 

So glad to read this, huge W! The cars looking great, nice work showing those cone chasers what's up! My biggest complaint about autox (other than the ROI like you said) is that my eyes don't see lines out of the cones. Just a big jumbled mess of orange, which definitely gives me a disadvantage. I've done decently in some auto-x as well but I imagine it would take a lot of experience to get "eyes" for it.

 

That’s fantastic news @Roda ! So glad for you. And also nice job on the win!

I’m interested to see some footage and data of your weekend when you get around to that.

 

Dug into some data this morning... early runs had lateral Gs in the .9-1.0 range, last run was 1.1-1.15. Typically see 1.2+ g on track, so definitely not a lot of grip.

Unfortunately, I didn't get any good video. Might have been a setting issue on the Smartycam.

 

Well, since I don't already have enough distractions, I bought another Miata...





In all seriousness, we had '13 Club PRHT that was my wife's DD from '2014-2018, and we've been wanting to get back into an NC2/3 PRHT for our 'street' Miata. As much blood, sweat and $$$$ as I have in our NB, the NC PRHT is just a better street driver. I've been looking for a while, but choking on the current NC prices. I spotted this 33k mile, 2011 Grand Touring in Tucson, and yesterday we drove up and made the deal. Still paid a little more than I would have liked, but I could probably make $4k if I flipped it on BAT right now. Not going to do that, this one's a keeper.

Now I just need to find a buyer for my NB....

 

ONE OF US! ONE OF US!

Congrats dude. Can't comment on the NB as I've never owned one, but tit-for-tat, my stock-ish NC is much more comfortable on the street than my stock NA was.