rider384

iTrader: (3)

Mods are in my sig. I'm running a brain-built MS-I.

The problem occurs in 1st and 2nd gear, or when I mash the throttle in high RPMs. It's very odd. Consistant, but I can't think of what would cause it.

If I floor it in 1st or second gear, it will accelerate just fine until higher RPMs, at which point it will brutally stop accelerating and the tach will drop to zero. It happens so fast that it doesn't have time to fully go to zero. It just drops rapidly and shoots right back up.

I can accelerate smoothely at WOT from 3rd to 5th with no problems, but if I start in high RPMs in 3rd it will also do it. I can, however, accelerate smoothly to the top of 3rd if I start from low RPMs.

I'm almost 100% sure it's a problem with my coilpack, because of the tach as well as the loss of spark. Would replacing the coilpack solve this issue? Has anyone else ever had it?

rider384

iTrader: (3)

Here is a YT video showing the problem:

rider384

iTrader: (3)

Bump, I really need an answer to this.

chicksdigmiatas

iTrader: (11)

If i have spark cut revlimit on, thats what it does. It looks like your coils are cutting out, or signal to them is being cut somewhere. I would say coil, but I don't want to be wrong and have you waste your moola. But those are the only things i could think of that would cause that. Well, I take that back, my car did that when it was wet out, and it ended up being the ecu ground. I took it apart and cleaned it and that solved the issue. I don't know how you would test the signal to the coil while the vehicle was moving though.

rider384

iTrader: (3)

I'm quick to jump to the conclusion of bad coils because if it was being cut anywhere I doubt it would happen at the same time (The misfiring and the tach dropping). If it was just loosing spark I doubt the tach would drop and if it was just loosing the tach wire it wouldn't not fire.

I've checked the ECU ground, that was the first thing I did. Thanks for the advice though. Coils aren't too expensive, I think I'll just throw down the 40 bucks for new ones.

Joe Perez

iTrader: (8)

I think you've got noise or dropouts on one of your CAS signals. Check your logs. See a big spike or dip in logged RPM when it happens?

rider384

iTrader: (3)

Bump for update: I tried to get a log of it happening but the one day I bring my laptop it doesn't do it. However, the problem has now expanded to non-WOT in 1st and 2nd and the WOT part has expanded to 3rd and 4th, though still not fifth.

Note that I also turned down the boost 'till I figured this out, and it still does it. Exhibits the same behavior.

Anyways, I'll take another stab at getting a log of it. But before I do, I'm beginning to reach the conclusion that something is dying because the problem has expanded to 3rd and 4th. At the next payment (September 1st) I'm going to replace either the coils or the CAS depending on what the log says.

Let me just make sure I have this right:
-If the RPMs drop on the log (And also on the tach as it's been happening) it's most likely the CAS OR a bad connection/interference somewhere.

-If the log shows no drop in RPMs and the only signal of drop in RPMs is the tach dropping, it's the coils/a ground? This is the one I'm not sure about, it's just an educated guess since both the tach signal and spark come from the coils.

Joe Perez

iTrader: (8)

Huh?

On a '96 Miata, the signal to drive the tachometer on the dash shouldn't be coming from the ignition coils. That was one of the changes that happened in mid 1995- they added the crankwheel, dropped the IGf signal, and took a feed from the ECU to drive the tachometer. I suppose my data here could be in error, but I didn't think that any of the US-spec 1996 model year cars still had the old four-wire ignition coils.

Go look at the connectors on your ignition coils. If they have four wires each, then I'm wrong. But I'm betting that they only have three wires each.

rider384

iTrader: (3)

Dammit, you're probably right Joe, I thought they all came from the ignition coils. But now that you mention it, that makes sense since I had problems getting MS to work the tach at all.

So, now that that theory is down the drain, if the log shows no dip in RPMs but the problem has happened, what could that entitle? Could it really be anything but the CAS or a ground? I've already gone through the car and the only ground I found to be bad was one for the EGR system, which is no longer in use on my car.

Joe Perez

iTrader: (8)

Did that actually happen, or are you trying to twist my head with random hypotheticals? If you opened up the air filter and found that it was filled with cat *********, what would that indicate? Well, it would indicate that it's a damned stupid question, since the ********* would have all been drawn through the compressor, pulverized, and would probably be plugging up the hot side endtank of the intercooler. ********* just don't stay in an air filter for no good reason.

Who the heck knows? Could be damn near anything, up to and including a bad keyswitch. Get some documentation first, and then we'll start philosophizing about what it means.

rider384

iTrader: (3)

Update: log shows no dip in RPMs

Braineack

iTrader: (62)

then it's an issue with coils and/or plugs. start with the $5 spark plugs swap.

basically the MS is saying spark and it seems like both coilpacks are saying **** NO. It doesn't show a drop in the MS tach because technically the cam is still spinning that fast and not sending garbage, however, you see it drop on the tach because it's being driven by the coils which are missing and not sending the signal.

Joe Perez

iTrader: (8)

I thought we decided that this was a '96?

This just isn't adding up. If the tach signal is coming from the coils, then he's got some bastardized wiring setup. I still want him to go out and physically look at the electrical connectors going to the ignition coils, and tell me whether each one has three or four wires.

OTOH, if the tach signal is coming from the MS as we'd expect it to on an OBD-II car, then tach dropout + misfire = faulty CAS trigger. That this would not show up in the log as a spike/dip in RPM is hard to conceive.

Braineack

iTrader: (62)

oh yeah, i keep forgetting the '96 thing.

rider384

iTrader: (3)

The thrilling conclusion is.... 3 wires per coils!

I also just checked, the log definitely shows no drop in RPMs.

God damn, this is fucked up right here. FWIW, it's also getting worse, it now does it in 5th, so it's infected my last good gear.

Joe Perez

iTrader: (8)

So, first off, post up the log. And the MSQ. If this turns out to be overboost protection, I'm gonna reach through the tubes and kick your ***.

Next, let's assume it's not overboost protection, or some other stupid software function that's set wrong. No log dropouts means the CAS trigger is fine. And no log resets means that power is fine. And yet a dropout of the tach means that the ignition system itself isn't at fault.

So, we've eliminated everything before the MS and everything after the MS.

rider384

iTrader: (3)

Can I trim logs? I took a log but I accidently left it running from the startup to the end of my journey, I meant to stop it right after the problem but I forgot.

EDIT: Ok, at the exact moment it's happening, there's a big bar in the log file that says "MARK 001 - RESET - *date*". So it appears the MS resetting is what's causing this, but the question is, what's causing the reset?

rider384

iTrader: (3)

Attached is the log file - sorry about the horrendous length. Just jump to the max throttle position because I only floored it once on the drive.

Joe Perez

iTrader: (8)

Ah HA! Now we get the critical piece of information.

Ok, random resets can be a bit of a bitch to troubleshoot, as there's roughly 2.7e14 possible things that could be causing it, but here are some of the more common ones:

1: Faulty power supply, by which I mean the supply of power from the car to the MS. This could be anything, from a loose wire to a faulty ignition switch to a faulty main relay to a faulty (...). If you want to experiment with this, you could run a wire direct from the battery to the MS's +12 input.

2: Bad grounds. I know you said you'd already done some work here, I'm just mentioning it. You can never be too well grounded.

3: Noise, the electrical kind. A few years ago, I rather lazily connected both a relay coil and an input to the MS in parallel to a switch contact. Part of my WI failsafe circuit. I suspect it was back-EMF from the relay that was doing it, but every single damn time that circuit operated, I got a processor reset. Scared the living dogshit out of me the first time it happened.

4: Some random **** loose inside the MS itself. Doesn't seem real likely, but anything can happen.


DIY has gotten into the habit of installing big-*** capacitors across the MS's internal +12 supply, in what, on a 3.0 board, would be equivalent to between the banded end of D10 and GND. Supposedly helped then out with resets related to supply irregularity. Can't remember what value they said they use, something like 4,700-10,000uF. (Once you get into that neck of the woods, size-wise, precision isn't exactly crucial- we're killing ants with A-bombs here.)

rider384

iTrader: (3)

Thanks Joe - I just did some wiring work around the car (To be honest I don't remember some of the stuff I did, I was a little out of it) but when I drove it and floored it in second, nothing happened (Out of the ordinary, that is. It still hauled ***). So something I did fixed it (At least temporarily).