WestfieldMX5

So, here's how the install went:
I tested the MS on the JimStim and couldn't get a tach signal. I had the two-dot version and this apparently has the wrong code for miata mode. Since I didn't want to test the MS in the car without seeing it work on the JimStim, I asked Matt to send me the three-dot version. I believe this is supposed to be working fine. Only, on my JimStim, I still couldn't get a tach signal. So I was assuming that my MS had a hardware problem. I bought an oscilloscope to check the board and found no problems at all. All circuits proved to be 100% OK. So at that point I concluded that there had to be a problem with the JimStim (or its software) and decided to plug the MS in the car anyway and see what happened. Well, nothing happened. No fuel, no spark, no nothing. Uh. It had to be a wiring / software issue but at that point I was fed up with MS. MS went on sale, got sold, but the buyer didn't pay. MS went into a box waiting for me to charge my internal batteries.
At that point, I knew the board was ok, so it had to be wiring / software.
When my batteries were charged, I gave it one last attempt. Since I didn't get the FP running, I jumped it in the diagnostics connector. I than tried to start but didn't catch. It was pretty obvious that my timing was 180° off. I had switched both ign outputs. After switching them, it would start, but my software settings were way off. Wrong settings for injectors, VE table that was way off (50% to low numbers). Also, stupidely, I had forgotten to connect the map sensor. So when I started it for the first time, it would idle at 100kPa. Funny enough, I didn't immediately notice this because the VE tabel was so lean, that my 100kPa numbers got it idling. Once I revved it, it fell flat on its face because much to lean. When I connected the map sensor, it went to lean that it would hardly start, let alone idle. Of course, by that point I had figured out that I needed to use the 100kPa numbers in the 20ish kPa row.
Oh yeah, somewhere in the process, I also had forgotten to connect the IAC valve, which didn't help either .
I installed the WB yesterday and hope to tune the VE table in a couple of days.

Zaphod

I just had a quick look - you didn't get the fuelpump running because of the darn immobilizer unit. (The cars without the immobilizer unit have the FP pin at 3N (LG) the cars with the factory immobilizer have it at 3P (R/W).

The R/W at 3P goes to the FP and there is another R/W at 1J which goes to the immobilizer unit and that in a way switches the main relais.

I don't understand the circuit though. Maybe someone could help me out there (but I don't wan't to post that information on the internet - I don't want anybodys car stolen because of that) it would help me with my standalone project.

Greets

Marc D

iTrader: (7)

how is the alternator circuit working?

WestfieldMX5

Never knew it was an immobiliser thing. I thought it was a euro / us thing. In a sense it is, because euro cars have the immobiliser and us cars don't.
Do you have a euro wiring diagram you could send me? I used the US one, hence my issues with some of the wiring. I have a 2000 spanish wiring diagram, but it has a 4 connector ecu.

The ecu needs to get a coded signal from the black box under the instrument cluster. If the ecu gets that coded signal, the car can be started, otherwise it won't. I do think the ecu does more than cutting the fuel pump though. If that was all it did, it would be extremely easy to override in the diagnostics connector. They probably cut the ignition as well. It would be easy to test this by unplugging the black box, jumping the fuel pump and see what happens (running the stock ecu).
Or you could wrap some tin foil round the ignition key, disconnect and reconnect battery and try to start the car.


It's working perfectly. Jason did a great job designing that circuit. I did get slightly different results on the testbench than on the car. On my power supply, it would charge to 14.3V. In the car, it'll charge to 14.6V. I'm using an el cheapo DVM, so 14.6V is probably off by a couple of tenths. I'm gonna measure what the stock ecu is charging at on that DVM so I can finetune Jason's circuit to the same output voltage. But like I said, it's working great. Battery voltage is 12.something and a soon as I start the car, it goes up to 14.6V.

AbeFM

iTrader: (3)

Huh, looks like I'll need to add that in on the next spin of my board. At least it'd be easy as anything to jump on the board itself, two adjacent pins and just drop some solder over them. I'd probably run a trace which people could cut out if they wanted.

WestfieldMX5

Don't think it would be a problem to bring FP to both pins. On my car (with immobiliser), the L/G wire in 3N is there, but it terminates in a blue connector under the dash, so no problem to shortcircuit 3P and 3N.
OTOH, I don't know if R/W in 3P is there on a us car (read without immobiliser) and where it goes.

Off topic, I thought that my DVM was way off when I measured my alternator voltage because it says my stock ecu charges to 15.1V! I tried a different DVM (thought the first one was off) and much to my surprise, it read a tad over 15.0V.
Seems awefully high. This was with the fan running and at idle, so it might even be a bit higher without fan @ higher rpms. Sjees.
Anyway, I have Jason's circuit charging at 14.6V which is probable a bit too high still. Anyone else ever measure what their stock ecu charges at?

Edit: Factory manual says voltage should be 13-15V at idle.

AbeFM

iTrader: (3)

I wouldn't freak out at 15v. It's high, but reasonable. All the better, really. My Big Stereo would probably prefer it.

Reverant

iTrader: (10)

Stock ECU charges at 14.1-14.7, depending on external temperature (uses the MAT sensor). My RX-7 FD alt likes to charges at 14.4 to 14.9. Some head units freak out and stop working after 15.5. Also, you can expect light bulbs to start blowing up faster.

Jim

WestfieldMX5

Hah, I read about this in the manual as well, but always ignored it (lots of other stuff to get right first).
Well, I finally got round to tuning my opening time as well. I was using 1.0ms with my stock injectors which turned out to be way off, because changing from 2 to 4 squirts resulted in an AFR change of over 1.0!
I settled on 0.88ms wich gives me identical idle AFR's on both 2 and 4 squirts. Idle is much more stable as well now. Where I first needed 13.3 to get a stable idle, it now idles rock solid at 14.3. Good stuff Abe.
I hope it still works as good in the morning
I guess I'll have to redo the entire VE table now? Bah.

On a sidenote, the difference in AFR readings between MT and my Techedge 1.5 WB seem to differ from one day to the other. Huh. WB is grounded right at the MS connector. Power is coming from the fuel pump relay. Stumped. Maybe the WB is going south? I've had it for about 8 years, but it's never been used the last 6-7 years or so. Do the sensors age?? I free air calibrated the WB and it wasn't too far off.

Reverant

iTrader: (10)

Wow, very old WB there. Is it the one in the long gray box?

Jim

AbeFM

iTrader: (3)

Huh. That's in the manual? Anyway, glad it worked. I'm inspired to take another look at my own car.
Not much - remember, in a 10 ms squirt (midrange/upper range), 0.22 ms is a small difference. You might might see it, but it's not a huge effect there as it is at idle.

AbeFM

iTrader: (3)

I had a bunch of issues here, too. Turned out to be the scale on MT's readout, not the wideband. You can tell by reading voltage directly.

There's some well known error in the code, which, while well known, they don't fix. I put a "generic, linear wideband" in, and set that 0-5, 10-20 and it works pretty well. Actually, I got it to drive a narrowband gauge very well, from 17-12:1 by giving it a known signal and setting it right.

AbeFM

iTrader: (3)

So you're still using 5.5*? I've checked this on two different cars, and 4 was always it. What RPM did you test at?

Zaphod

I am also using 5 or 5.5°.

WestfieldMX5

yup, 5.5°. I tested at idle and while revving up to about 2500-3000rpm.
Mind you, the stock ecu ran ±12°. Maybe euro cars run higher timing? This could account for the 1.5° difference we're seeing.

AbeFM

iTrader: (3)

That's my feeling, that you guys have a different trigger wheel. It didn't seem worth the expense to me.

We do get worse gas than you, I believe, which would justify the timing change. Still, I'd assumed it would be in software.

I'm curious what pin your Alternators are on in the OEM harness....