Help me-Did I hurt my Squirt? aka Mumpkin problems

As recently discussed in this thread: The Mumpkin We had a very frustrating attempt at testing this weekend that was axed by the car not firing cylinders 2 and 3. Swapping in a good igniter, a good set of coils, and even swapping to a homebrew COP setup using a Bosch 4 channel igniter and CBR coils produced a no spark on cylinders 2 and 3 condition. The car is a turbo 91 Miata running a homebuilt MS2 V3.0. It is using a stock igniter and coil setup which fires two-two terminal coils in wasted spark mode. 1 and 4 are paired and 2 and 3 are paired.
The car ran fine a few days before the event. The possible events that may have caused problems since it last ran well are:
1) the stock igniter was removed to build the COP setup and two seperate times the +12V wire at the igniter plug briefly shorted to ground.
2) The key was left in the ON position for an extended period of time without the car running. Due to the design of early Megasquirts this can roast ignition parts.
3) The Challenge Gods are testing our will type randomness.

We came to the conclusion that the problem lies either in the Cam Angle Sensor, the wiring either before or after the Megasquirt, or the Megasquirt itself.

I have the Megasquirt and SVReX has the car so I need help diagnosing the squirt.

Here she is:

Heres a melty looking part:

Any suggestions for how to proceed for someone who is but shouldnt be a MS newbie?

Do you know if the key was in the ON position when the igniter plug was shorted to ground?

I don't think it is the Cam angle sensor. I am pretty sure it sends only 1 signal to the MS- It determines when Cyl #! is at TDC.

If this is correct, it would not only effect the signal for cylinders 2 and 3. It would have screwed up the signal for all 4 cylinders.

In reply to SVreX:

I believe the 12V+ is switched, so yes.

Wow this seems familiar...

I assume you're using the LED outputs to fire the spark with both the coil packs & CoPs? If so, sadly, that suggests that something in the MS is fried. You can test the LED outputs directly to confirm.

In reply to GameboyRMH:

How would we test those LED outputs?

Just wire up some LEDs to them as if you're going to use them for indicators, and see if they flash when the spark should be firing. This is assuming you're getting a good input from the CAS.

Edit: Actually for a MS2 V3 board, the wiring is different for LED indicators and LED ignition...basically you put the LEDs on the pins that would normally go to the CoP trigger lines, with the correct polarity. They should flicker one way or another when they get a signal to fire.

In reply to GameboyRMH:

Define "good" input from the CAS. How can we test this without an oscilloscope?

Does the CAS send 2 separate signals to fire the cylinders, or 1? (IE: FIRE: 2&3, FIRE: 1&4, or is it more like 1 signal the MS interprets, like #1 is at TDC NOW, NOW, NOW)?

I know... I sound like an idiot. That's OK. Treat me like a 4 year old.

Everything I see and read about Megasquirt makes me want to never get involved with one.

In reply to GameboyRMH:

Are you saying there is a different setup for trigger lines for the outputs for a CoP vs stock?

In reply to Lof8:

This poor megasquirt was installed in '08 and has been handled with neglect ever since. It, and the whole car, have been amazingly durable considering.

I'm not expert with MegaSquirts either but a cam angle sensor usually has a wheel with some number of teeth, with one missing which lines up with TDC.

I don't know what the stock wiring setup to run the Miata's engine was, but this is how an MS2 v3 should be wired to run CoPs (for individual firing):

This is the setup for wasted-spark CoPs:

From here:

http://www.msextra.com/doc/general/sparkout-v30.html

That's an MS1 V3.0, so notes on the MS2 may not be valid and can cause some confusion. The MS2/Extra has some diagnostic tools that would be handy here, too.

But I'd first check the unit on a JimStim and see if you can get a good RPM signal and see if the outputs pulse there. The JimStim will go a long ways towards making this easy.

If you don't have one, make a data log of an attempted start and see if you get RPM.

The part you posted is part of the injector drive circuit. We have a comprehensive troubleshooting guide for that circuit here:

http://www.diyautotune.com/tech_articles/megasquirt_injector_driver_troubleshooting.htm

Oooh oooh, I have this!

The squirt has always just done it's job and other people have wanted to do the tuning so I am embarassingly clueless about it.

It looks like I have a few flux corrosion bridges. Looks like it needs a bath.

MrJoshua wrote: It looks like I have a few flux corrosion bridges. Looks like it needs a bath.

Denatured alcohol is default safe flux remover.
Other solvents will remove flux, but also dissolve plastics, silk screen ink, and such used on the electronic components.
An old toothbrush helps scrub away debris/corrosion.
(High proof grain alcohol, Everclear etc. but even Vodka in a pinch)

Lof8 wrote: Everything I see and read about Megasquirt makes me want to never get involved with one.

Because nobody ever writes about how they installed their system in '08 and it's been fine ever since. That stuff is boring.

It's like the OMG E36 CHASSIS CRACKING which happens on maybe .5% of the cars out there.

So I've been playing with the stim. If I set the stim to ~1k rpm and test the ignition with an LED like gameboy described I get different behavior for each of the two ignition circuits. The circuit for plugs 1&4 blinks along merrily and steadily while the the circuit for 2&3 is almost fully on with a very slight flicker or pulse. Super exciting ignition test video! Now i need to look for diagnostic procedures for components in the ignition circuits.

Might be worth reloading the firmware on the CPU and verifying the settings in the configuration as well.

Gave the board an spa day featuring an alcohol bath with a warm water rinse to be followed by a day of rest and relaxation to be sure its completely dry.

Before:

After:

I will retest later today to see if that did anything.

Depending on igition mode you may get odd behavior on a "regular" stim like posted above. Change to basic trigger to chack for clean rpm and set tomsingle coil for troubleshooting. Use the output lorts settings in advanced to check the leds, just set them to turn on above x rpm and watch the leds on the unit. If checked ok set back to your original config and test on the car.

Also cbr coils really need a cdi.

With that Stim version and an MS1, you'll want to use "Stim for wheel" mode for RPM testing.

Do alcohol after water, unless you're using distilled. Even then, you can't ignore the drying effects of alcohol.

Paul_VR6 wrote: Depending on igition mode you may get odd behavior on a "regular" stim like posted above. Change to basic trigger to chack for clean rpm and set tomsingle coil for troubleshooting. Use the output lorts settings in advanced to check the leds, just set them to turn on above x rpm and watch the leds on the unit. If checked ok set back to your original config and test on the car.

I will try that, thanks.

Paul_VR6 wrote: Also cbr coils really need a cdi.

I've just researched them enough to know some people swear they do and some swear they don't. Some of the Honda guys claim to be pushing serious boost with big plug gaps running them non CDI.

Our current problem occurred with the stock setup installed prior to ever installing the COP setup. We will be trying to solve this problem with stock ignition components before chasing the COP install.