gospeed81

iTrader: (51)

I hate starting a new thread, but I searched as much as I can with boss coming in and out, and I feel like I'm probably missing something obvious.


Without any consistency, my car will buck and die at 5Krpm when I'm running at high loads (full throttle or near ful throttle).

Symptoms:

-tach drops (faster than actual rev drop)
-spark LEDs(outer two) turn off momentarily
-AFRs stay constant

I got it to happen once while my fans were on, and the warm up light(center LED) stayed lit, so it wasn't a full reset.

I shouldn't be anywhere near my overboost protection (4psi overhead) and this is of course way under both rev limiters.

I will try and datalog it tomorrow, but was in a hurry to work this morning. I'm starting to think it's not MS related, but want to pick your collective brains before I go about replacing the CAS or anything.

Thanks in advance,

-Damon

Jeff_Ciesielski

iTrader: (10)

Check your plugs first, I had nearly the exact same thing happen to me. It didn't always register as a full reset when I was logging, but my plugs were over-gapped and it would buck and the tach would drop out when I went WOT.

gospeed81

iTrader: (51)

I had replaced plugs and wires a while back, and then replaced plugs again last week after losing spark in cylinder 1 only.

The time that this started creeping up would coincide. I'm getting good spark on all 4 now, but occasionally get this shutdown.

It's very violent though, although it will feel like it's just pushing through it sometimes.

I've got spark gap pretty low (0.025in) and exactly where I had it before.

Jeff_Ciesielski

iTrader: (10)

Do you have tunerstudio installed on your laptop? If so, there is a diagnostic panel that can log trigger events. Very helpful for this sort of thing, at least you could rule out the cas.

gospeed81

iTrader: (51)

No, may try and get it just for that though.

You've got me thinking plugs again though, just because it's too much of a coincidence. Maybe I've got them gapped too low???

It is odd how it does it at some times, and then other times runs to 7K like a bat out of hell.

Joe Perez

iTrader: (8)

This is a problem with the CAS input signal. Sounds like the same one, incidentally, that drove me bat-**** crazy until I built the 36-1 wheel setup.

I believe it is caused by radiated noise from something like the injector lines being received by the CAS lines and interpreted as false trigger inputs at the MS. They all run in the same bundle from the engine compartment to the ECU.

One thing I never did try was to run the CAS signal lines in their own shielded cable, separate from the harness bundle that contains the noisy stuff. If you have access to some shielded two-conductor audio cable and want to do a little wiring, try clipping the white and yellow / blue wires where they exit the CAS, run them through the inner conductors of a piece of shielded audio cable, and connect them straight to the MS. Connect the MS end of the shield to ground, leave the other end floating.

Jeff_Ciesielski

iTrader: (10)

I doubt they are gapped too low. I have mine gapped at .025 as well and havn't had any problems. Its an easy check though. Maybe pull them to inspect and try gapping them to .030? Also, check the connections on your cas, coilpack and plug wires. My friends car was experiencing some bogging the other day due to a loose wire.

gospeed81

iTrader: (51)

I will check connections, and if Joe is right (god I hope not) then Tunerstudio should tell me.

My CAS doesn't have a cover on it(thank you Meximechanic), which is why my first assumption was a dying CAS. I only suspect MS issues since it does consistently happen at the same rpm.

Joe Perez

iTrader: (8)

The fact that the tach drops out means that it is not the plug gaps, or anything else in the secondary ignition system. Period. This is a trigger problem.


Yup, sounds familiar. If I was gentle on the throttle I could take it to redline. If I was WOT, it'd misfire pretty reliably between 4000-4500.

One other option you might consider would be to build a new input circuit for the MS similar to Abe / Jason_C's NB input circuit. The way the second trigger is normally done with the MS is kind of hokey what with the single capacitor trying to act as a filter, whereas theirs have some proper hysteresis that gives you good noise immunity. I never did try this when I was having the problem, I decided to solve it with the lathe rather than the soldering iron.

Jeff_Ciesielski

iTrader: (10)

I'm not trying to start a pissing contest, this is strictly personal experience. (Seriously, I've got CRAZY respect for the joe perez goodness) But when my plugs were over gapped, my tach would drop to zero and the car would buck when under load.. Every time. I swapped them out, and the problem went away.

Joe Perez

iTrader: (8)

On a 1.6?

Huh. That kinda busts one of my pre-conceived notions about the architechture of the igniter...

Ok, I'll accept that.

Jeff_Ciesielski

iTrader: (10)

Honestly, it confused the **** out of me. I read all of your posts about the cas problem, and started beating my head against a wall because I didn't want to have to switch to a toothed wheel setup or **** around with another input circuit. In a way it kind of made sense after I figured out it was the plugs. If the tach is triggered off of the igniter on our cars, if one of the coils didn't fire ( I believe its the 1/4 coil ) neither would the tach. Or maybe I'm talking out my *** and need a good schooling . Either way ->

[/threadjack]

Joe Perez

iTrader: (8)

Well, the reason I say this is that the igniter really has no way of knowing whether or not a spark actually happened. It is connected only to one end of the primary windings of the coils. It switches them to ground, they charge, it opens the connection, and the field in the coil collapses into the secondary, inducing a high voltage and (hopefully) causing a spark. On the '94-'95 I could accept that it's theoretically plausible (though unlikely) that they might be monitoring secondary current, but with the 1.6 I just can't see how the igniter would know whether or not the plugs actually fired.

Still would not explain the outer LED on the MS turning off when it happens.

gospeed81

iTrader: (51)

This crap's still eating my lunch. Car runs hard and fast sometimes, but will cut out like it swalled a stone right at 5Krpm sometimes.

I've noticed it does it more when cold...which kinda leads me back to signal noise from the injectors since warmup enrichment would make the pulsewidth longer, but I think by the time I'm boosting up to redline coolant is already up to temp.

I drove the car for a few days with the #1 spark plug not firing before I swapped all 4 (again) and had this problem crop up. I'm trying to come up with a theory that explains something being damaged by running on 3 cylinders and killing my CAS signal. Doesn't make any sense to me though in a wasted spark setup since #4 would have always been firing anyhow.

I really need to log it, just have been seriously busy with work and life. I don't really want to switch to a trigger wheel, but will do so if necessary, and have been reading up on this as well as your input circuit mods Joe.

You're right about the outer LEDs dying, just keeps telling me CAS signal is being interrupted somewhere. I may try the shielded wires first.

Savington

iTrader: (31)

I had this exact problem right after I turboed my car. Serious misfires at high load/high RPM, accompanied by the tach dropping out as well. Got progressively worse. A fresh set of plug wires solved it.

(get some COPs.)

Joe Perez

iTrader: (8)

Either that, or build Abe's (or Jason's) NB-style input circuit. Those will have much better noise rejection than the typical NA-MS1 mod.

gospeed81

iTrader: (51)

That's what puzzles me, I did put new plug wires on it. I've put 4 new wires and 12 spark plugs on it since going turbo 3 months ago.

I do really want to do COPs though, I'd just hate to do it and still have CAS signal issues.

JOE: I will look for those mods, I've only read your thread so far, and didn't see the final result. I'd rather not crack open the MS though. I did get Scott to build it for a reason...

gospeed81

iTrader: (51)

Whatever it is...I think it's .msq related. I swapped to an older (newer) map I had stopped using when my AFRs started getting jumpy for some odd. I also swapped to Brain's .vex for spark table.

Could not get the problem to crop up this morning. Took some logs, will try and upload one when I get home.

gospeed81

iTrader: (51)

Ok, I was wrong....it's still happening. I should have known that Joe knows better.

It is happening less, but this could be due to any number of causes. Either way I don't ever want to see it again, so I'm going to attempt a trigger wheel (36-1) setup.

Joe, I read your thread, and it's really a work of art, so I didn't want to crap it up. I'll keep my newb questions over here:

1. Has anyone else replicated your setup?

2. Where did you get your circuit design (black box)? If you built it yourself do you have a schematic for it?

3. Any idea on total cost? I can do my own hub adapter machining as well, and may even be able to make my own trigger wheel (waterjet). Electronics I worry about.

4. Can I do this without opening MS casing? I hate to ruin Scott's work, but have no problem soldering one wire to another. I just don't want to build a circuit inside the controller. Judging by your setup I can build an external circuit and wire it into the DB37.

Any (guido)ance much appreciated you gold-chain wearin', chest hair flauntin' Floridian.

Seriously, I admire your work, and just don't want to bite off more than I can chew.

Joe Perez

iTrader: (8)

Wow. Have you at least tried running a new piece of shielded cable from the CAS to the MS, or using an op-amp based input circuit such as Abe's? Either one will likely fix the problem, and frankly, if I'd have been thinking in that direction from the start I probably wouldn't have done the 36-1 wheel. Not that I'm at all unhappy with it, but it was a lot of work to get right.


Not to my knowledge. There are at least two other Miatas I've seen that are using custom crankwheels, however neither of them were based on my setup. One used the EDIS wheel cut from a stock Ford pulley and either welded or pressed onto a Miata pulley, and the other used an aftermarket wheel similar to mine, but was not a derivative of my setup.


I simply copied an existing design. I prototyped both the stock VR circuit from page 3 of the MS 3.0 schematic as well as the LM1815N-based circuit described in the MSExtra "second VR wheel" documentation. I wound up using the design from the 3.0 schematic, as I had trouble with missed teeth on the LM1815 design.

The only reason it was constructed in an external box is that I hadn't populated the VR portion of the MS board during initial assembly, and had subsequently glued a large capacitor to the board in the space the VR circuit would normally occupy.



Not much, actually. The wheels are $25 from DIY, I bought the hunk of aluminum from Online Metals for about $20, the sensor was around $30 from NAPA (though you can get them much cheaper off eBay) and the rest of the materials were scrap that I had lying around.



With a slight modification, yes. If you add an open-collector output stage to the end of the circuit, then you can connect it to pin 24 (TachIn) of the DB-37 and simply change the software settings.


Hahaha. Chest hair, check. Hawaiian shirt, check. Gold chains, negative.