1986 Mazda RX7 NA / 13b / staged inj / stock ign

[muythaibxr]

Details:
1986 Mazda RX7
NA 13b with stock ignition
12+1 NipponDenso crank trigger (stock)
Running the stock leading and trailing ignition systems

Started out today with the engine starting right up on ms3 after swapping out the ms2. There were a few issues though:

Closed loop idle speed control's target was not being looked up properly, and there was a table lookup error on idle advance. After a while figuring those out and fixing them, I got it back out in the car, and TPSdot was going crazy (wasn't doing this before). Turns out some code James committed to the rtc section broke the time counter I was using in the rtc section to calculate TPSdot, which made the car run like crap. James fixed that, and now the car runs just as well as it ever did on ms2...

Actually, driving it around a bit in the neighborhood, it seems like it's running even better than ms2. It's just a bit more responsive, and runs just a little smoother. This may be the placebo effect, as I have no quantifiable data that shows it runs better, but it does feel better.

So chalk up another car running on MS3!

I'll be driving it a longer distance tomorrow to work on the corolla a bit, so may update when I get back tomorrow.

Ken

[muythaibxr]

So I drove it today about 80 miles total, and all I can say is that there's not much to say. It ran mostly as well as the ms2 although a few spots need some tuning. I logged on the SD card as well. It didn't log the first time I plugged it in, but it worked fine after that. I'm using "insert" mode, so that might be a factor. I might've not gotten it plugged in correctly the first time.

Ken

[dyoungre]

So ... any updates?
I've built an MSII, and had planned to put it on a '91 RX7 ITS car, but observed some resolution concerns with PNP on a Miata. The AFR variation as the engine goes through the RPM band was varying more than I'd like to see, and I presumed it was due to the table look up limitations / bin point resolution.

I'm expecting that you have much cleaner AFR control than before, right?

Are you using any NON-stock hardware for input / output on your RX7? Any concerns with a stock crank trigger?

[jsmcortina]

The AFR variation as the engine goes through the RPM band was varying more than I'd like to see, and I presumed it was due to the table look up limitations / bin point resolution.

Even though MS3 is faster, more precise, more featureful etc. we haven't seen any evidence of the tables being a problem on MS2/Extra.

James

[muythaibxr]

So ... any updates?
I've built an MSII, and had planned to put it on a '91 RX7 ITS car, but observed some resolution concerns with PNP on a Miata. The AFR variation as the engine goes through the RPM band was varying more than I'd like to see, and I presumed it was due to the table look up limitations / bin point resolution.

Yeah, if you're seeing a lot of variation, I'd look at the hardware or mechanical side of things, or check the distribution of your bins, opening time voltage correction, etc. With my rx7 on ms2/extra, the AFR pretty much doesn't vary at all, especially with the new staging algorithm. It only really varies during accel enrichment. Things like getting opening time voltage correct, MAP jitter, etc.. can all contribute to an unstable AFR. You must tune these. The PNP is based on ms1, which calculates rpm in 100 RPM increments, and controls PW in .1ms units. MS1 is nowhere near as accurate as ms2 or ms3 (which calculates rpm in 1 rpm increments, controls pw in 1 usec or less, and calculates PW in .01 usec units), so I would expect going to ms2 to be a pretty big improvement, but even ms1 tuned properly can give very stable results. I believe the PNP can also be upgraded to MS2 without too much pain.

I'm expecting that you have much cleaner AFR control than before, right?

Not particularly. I didn't have any issues there on ms2/extra though. I didn't see much difference here from ms2/extra until I added the new MAP sampling code to ms3. That is now incorporated (although in simpler form) in ms2/extra so even that probably isn't a difference anymore. The only real difference I noticed was that the ms3 seems to run the engine a bit smoother. That could be the placebo effect though.

Are you using any NON-stock hardware for input / output on your RX7? Any concerns with a stock crank trigger?

I'm currently running it the same way I did in in ms2/extra. I'm using one of my Zeal Engineering daughtercards for VR conditioning and to run the BAC valve and the fuel pressure regulator solenoid, and pullups on the LEDs to drive the ignition. I'm using the ms3 in upgrade mode. Once we (the developers) get ms3x boards, I intend to switch over to that.

Ken

[muythaibxr]

Just another quick update. I've been keeping the code up to date on the rx7, and it still runs great. If anything it's running smoother than ever since I got rid of a few of the mechanical problems that were causing issues here and there.

Ken