Zaphod

I would be in too (as you know)

AbeFM

iTrader: (3)

Ok, time for an update. After trying it on several people's cars with bad wiring, unknown sensors, etc, and waiting several weeks for people who said they would let me test on their cars, I spent half a morning getting all the wires sorted on my car, and kinda started to do a test.

Results (good):
TPS, CLT, and similar sensors all worked.
Ignition inputs seem to work
Car will try to start, sort of catch, cough, and die again.

Results (bad):
I think the FAN, VICS, and IDLE are all messed up. Not sure. The fans are on, and I didn't figure out why yet. The VICS actually seems to be working, but it's hard to tell since I need 5000 RPM to make it switch (I could change it in software, but that's not what's limiting me)
The idle valve isn't humming, which tells me there's a real problem, and yes, one which would keep it from starting.

Also, thanks to the MSe people, I'm having real troubles finding matching versions of everything. Sometimes things are by date, sometimes by version number, sometimes they report the wrong info. Right now I seem to have version 2.1.0o in the computer and 2.1.0p in the software - I don't care what version I use, I just with it were consistent, at the same time I don't want to change the one on my regular DD MS, so I can still get to the beach. :-)

Anyway, I'm going to take this board to the lab and debug it - I can power it off my voltage "ref" lines, so I don't really need a jimstim. that part is cool. Will keep everyone posted. Thanks all!
-Abe.

Marc D

iTrader: (7)

cool, good with the testing. I guess you werent able to test out the alternator circuit cause you couldnt get the car to idle at all

AbeFM

iTrader: (3)

Ok, interesting, my pulse widths seem to be at 33 ms. 33 ms on start up, with 100* coolant? What the hell? That's 550 cc injectors (req fuel 6.0).

Something is weird. Any ideas?

edit: That's on the Stim, my board/car isn't invovled.

AbeFM

iTrader: (3)

I can't test it anyway, without more digging into my wiring than I want to do right now.

Marc D

iTrader: (7)

hmmm... thats strange its at 33ms, that would mean its flooding the engine with fuel. I would suggest the MAP sensor might have been disconnected, but since it was done on a jimstim, im a little stumped myself.

For the idle, would it be possible to do the mod on the main board, then just pass the line directly to the adapterboard then to the harness plug? maybe that might clear up the idle problem?

AbeFM

iTrader: (3)

Yeah - I'm not yet sure where the idle issue is. I'll just have to dig into it with an oscope.

JustinHoMi

Post the schematic of your idle circuit, if you like!

Marc D

iTrader: (7)

oh abe, as far as the correct version to use, you should download the most recent "release" version. It was released on 6/27. I think it willl give a good start for most people

AbeFM

iTrader: (3)

Just got back to it, I think I'm going to dig a bit. Looks like the idle is working on the Stim (so the MS is ok there), but not on my board. So I think it's two unrelated problems.

hustler

iTrader: (6)

miataturbo > *

AbeFM

iTrader: (3)

I just want to say how amazingly awesome the little row of signal interrupts is. :-) I've been able to put the jimstim in in parallel, making debugging both the code and the board much more easy.

I'm not as much a fan of my jumping the idle valve transistor, since, if you plug it right into the stim, the CPU has to sink all the current from the LED, which is fails to do, dropping the LED to only ~2.3V. Won't be doing much more of that!
-Abe.

P.S. For those keeping track, the idle valve seems to be working (if not the MS's "idle valve test" function - not sure why yet), so now I need to check I didn't just make a no-brainer mistake and put the idle out on the wrong OEM harness pin.

Marc D

iTrader: (7)

If you did put it on a wrong pin, that would really suck. Probably need to make more boards again?

Zaphod

There is your answer about the test mode:

"That feature might be broken.

Ken"

AbeFM

iTrader: (3)

Everything's on the right pin, and the signals pulls through fine - I pulled it up with a resistor and it acts as it should. Perhaps it IS working, it just seems more quiet than before.

What's got me more concerned is this factor of ten in the fuel delivery, and the fact that one of the two ignition LED's never quite goes off. This is more trouble shooting the MS than the board!
-Abe.

AbeFM

iTrader: (3)

Ok, new transistors are on the way, I should have everything I need to kit whole boards here within a week. Hopefully by then I get it running on my car, find a volunteer to verify the alternator and AC controls work, and could get stuff out to all of you soon! Very happy about this, and not just because I need the space. :-)

lsc224

iTrader: (13)

Sweet....I'll be looking forward to it.

redrider706

iTrader: (4)

Great news Abe. I am looking forward to getting my hands on the board.
-Raj

Marc D

iTrader: (7)

What was the earlier problem with the pulse widths related to? Wrong type of transistor?

Thanks a bunch for doing this, I've been aching to get MSII set up, trying to read up on Frank's way, but his modifications look very tedious to set up and getting the correct parts might be a pain too.

Speaking of alternator circuit, Frank said that the alternator circuit on his board worked IIRC, so it should work just fine

AbeFM

iTrader: (3)

At the risk of jinxing myself... The pulse widths were from taking my MSQ and putting in to an MS built differently.

My knock sensor in line was *shared* with the barometeric correction. So, any bouncing on the line moved the fuel all over the place. In software I set baro to initial read, and put the knock on it's own pin (and set the internal pullup - nice of them to include it in software) and all is well.

Now everything looks right on the stim. Time to wedge it into my car once more, and try to track down the idle thing, then I'm all good.
-Abe.