JCH13 - Turbo Miata Rehab Thread

[JCH13]

No chance at capacitive/inductive cross coupling anywhere? You have a scope?

I'd doubt it, being such a widely used conversion (even with my ECU) that sort of thing should be (I hope) a known issue.

No scope, which makes me a sad panda.

 

[exdeath]

You have the original monolithic igniter I assume, how is that split and distributed between coil groups? What does the primary input to the dead coil look like?

Yeah scope = yawn its almost cheating heh.

 

[JCH13]

My setup is using two transistors in the MS1 to drive the two coil pairs, the stock igniter module is 100% gone. The four coils share one 12V,one GND, and one Tach signal wire, only the signal wire changes. I have no way of scoping that to make sure it's okay, short of 'liberating' the Tektronics unit from work, which I'd rather not do.

 

[exdeath]

Ah ok. So one coil does not appear to be getting primary signal from the ECU. Do you have a test light or something? Can at least see strobing, esp if you limit to cranking speed. Try at both the coil side and at the transistor output on the ECU, compare to the working channel for reference.

Cranking rpm / 60 = pulses per second since its every 360 degrees on each channel.

Nothing stupid like a bent or loose trigger wheel? Heh.

/pulling off the wall causes out of my arse.

 

[JCH13]

Hrm. I think I'll try putting a spare plug in the two suspect coils, grounding them to the head, and just looking for a spark...

I think very first thing to check is the plugs, to see if they're just drenched in gasoline...

 

[exdeath]

Didn't do that first ? :awe:

 

[JCH13]

Didn't do that first ? :awe:

I did the most recent de-bugging (and chopping out the connector and re-splicing the wire) on my lunch hour and ran out of time...

 

[JCH13]

I get the initialization spark (when the key is first turned and the ECU comes on) but I get no spark while running. I pulled a coil out and stuck a plug in it and looked at it while the car ran, zilch. When the key is cycled the same plug/coil sparks.

The hardware/coils are good, but somehow the ecu isn't getting the message when the engine is actually running? I don't even...

My brain is going to explode...

 

[JCH13]

Replaced CAS with no change... other Miata owners reported losing spark on 1&4, or 2&3 being related to a bad CAS.

Sigh. Calling it quits for the second night in a row with no idea why it's not working.

 

[Zargon]

 

[exdeath]

Sounds like a successful conversion to German Miata!

Telling you man, scope and check vital I/Os from both inside the ECU and the terminal harness ends. Its THE only way to be sure.

How long have you run this ECU? Did you change settings in the ECU to run the dual channel wasted spark setup in 4 cyl mode? Is the working coil signal twice per engine cycle or 4? Granted it doesn't explain the pevious intermittent behavior. Check the transistor base-emitter/collector against collector -emitter flow, wonder if the intermittent shorting you described did something...then again power up spark...

Could try swapping ignition channel pins on the ECU harness side temporarily to isolate harness.

 

[ino uno soweno]

Put your thinking suit on,

 

[jlee]

Put your thinking suit on,

:awe:

 

[JCH13]

Sounds like a successful conversion to German Miata!

Telling you man, scope and check vital I/Os from both inside the ECU and the terminal harness ends. Its THE only way to be sure.

How long have you run this ECU? Did you change settings in the ECU to run the dual channel wasted spark setup in 4 cyl mode? Is the working coil signal twice per engine cycle or 4? Granted it doesn't explain the pevious intermittent behavior. Check the transistor base-emitter/collector against collector -emitter flow, wonder if the intermittent shorting you described did something...then again power up spark...

Could try swapping ignition channel pins on the ECU harness side temporarily to isolate harness.

I basically did check the spark control output voltage, they're wired through two output LEDs on the ECU and one wasn't lighting up very brightly (I had to pull out the ECU to see them). I replaced that control transistor and upped the two spark pull-up resistors from 1kohm to 100ohm to improve the controllers' current supply capacity and *magic happens* it works!

I've never seen this sort of 'semi-failure' of a transistor before. I've always seen them work, or not work. Never before have I seen one 'sort of work'. Learn something new every day I guess.

Most of the car is buttoned back up, I need to swap back to my old CAS, it is a 1.8L version and provides a cleaner, more reliable, cam angle signal than the 1.6L version I swapped it out for.

TL;DR - Replaced a spark control transistor and two resistors inside my ECU and it runs great now!

 

[manimal]

I basically did check the spark control output voltage, they're wired through two output LEDs on the ECU and one wasn't lighting up very brightly (I had to pull out the ECU to see them). I replaced that control transistor and upped the two spark pull-up resistors from 1kohm to 100ohm to improve the controllers' current supply capacity and *magic happens* it works!

I've never seen this sort of 'semi-failure' of a transistor before. I've always seen them work, or not work. Never before have I seen one 'sort of work'. Learn something new every day I guess.

Most of the car is buttoned back up, I need to swap back to my old CAS, it is a 1.8L version and provides a cleaner, more reliable, cam angle signal than the 1.6L version I swapped it out for.

TL;DR - Replaced a spark control transistor and two resistors inside my ECU and it runs great now!

Awesome man! Glad you figured it out. Problems like this one really humble me and I always end up calling in friends. Glad you got it!

 

[JCH13]

Awesome man! Glad you figured it out. Problems like this one really humble me and I always end up calling in friends. Glad you got it!

Thanks! This one really baked my noodle. Definitely humbling, but I feel better that all of the MS wizards on MSEFI.com or miataturbo.net appeared stumped as well.

Also, I have pictures of the install to post up later tonight. I fear I will no longer be able to use my strut bar. The upside of that is that I get to make my own, absurd, strut bar now. Though I don't think any of the titanium tube scrap at work is even remotely the right size

 

[exdeath]

Transistors, despite common belief, are VERY linear (read: analog) devices hehe. All kinds of wierd half on half off whatever shenanigans can occur. Especially with power amplifiers, eg ignition control.

Depending on the failure mode it can be possible to work like a capacitor, conducting briefly on power on then stopping once it biases = power on spark?

Did you go back and reinstall the connectors and tug on the harness again to replicate ealier issue? I'd be concerned if a problem in the harness killed the transistor or if it was just a wacky coincidence. Driver transistor output needs a load on it or it could fry if something bypasses that load and allows unlimited emitter-collector current.

 

[JCH13]

Transistors, despite common belief, are VERY linear (read: analog) devices hehe. All kinds of wierd half on half off whatever shenanigans can occur. Especially with power amplifiers, eg ignition control.

Depending on the failure mode it can be possible to work like a capacitor, conducting briefly on power on then stopping once it biases = power on spark?

Did you go back and reinstall the connectors and tug on the harness again to replicate ealier issue? I'd be concerned if a problem in the harness killed the transistor or if it was just a wacky coincidence. Driver transistor output needs a load on it or it could fry if something bypasses that load and allows unlimited emitter-collector current.

It definitely did act like a cap, very weird. But, I guess, thinking back on my circuit/electronics classes from college I should have known something like this was possible just not very common.

In theory the pull-up resistors and a few other safe-guards prevent things from overloading inside the ECU. I believe the transistor might have been weakened from soldering too close to it (legs were VERY short) and the 1k pull up resistors didn't provide enough voltage (resulting in a current spike, or even just an inability to proprly energize the transistor) given the new and different loading of the COPs vs the stock igniter.

Hooray electrical engineering!

 

[exdeath]

Hooray electrical engineering!

Otherwise known as teh magikz!

Pull up/down resistors are there to keep a gate pulled to a known state and prevent floating when the load isn't present or goes high Z, they aren't going to protect the driver from burning up when the active high current output has a direct path to V/GND with no resistance. If you short your drain the transistor is gonna be like lulz F UR resistor! /sizzle every time it turns on.

In general, of course, I really have no clue how your ECU is built. I'm assuming those are beefy power MOSFET transistor (look like voltage regulators with metal tabs) for driving the primary coil that can handle the inductive kick back when it fires (eg same type of transistor as an ignitor). In which case I've been using BJT terminology incorrectly anyway.

 

[JCH13]

They're just cute little guys that switch more powerful transistors in the COP's internal igniter. A transistor to flip a transistor.

I've always heard these resistors as being there to control current sent to the COP/Igniter transistors.

Here is the install guide: http://www.diyautotune.com/tech_articles/how_to_megasquirt_your_mazda_miata.htm

Here is some in-depth discussion about a problem similar to mine: http://www.miataturbo.net/general-miata-chat-9/more-cop-problems-help-27324/page2/

I have done ECU igniter transistors in a MegaSquirt before (GSXR600 engine) but that SUCKED. I would never want to do it again. Self-ignited (if that's the right term) COPs have nice internal bypass features (usually) to keep things from exploding. I smoked a few coils and more than my fair share of transistors trying to get that system to work. A microsquirt has those sorts of safeguards built into it, it's fine for driving wasted spark systems.

Toyota COPs for me from now on plz! Much less of a PITA to deal with overall.

 

[exdeath]

Ah ok internal igniters or what they call "ignition amplifiers" in COP speak.

I was thinking the GM style wasted spark with standard "dumb" coils for some reason. I guess that's not even COP really.

 

[JCH13]

So, a few pictures. Here is the aluminum plate I made to bolt the coils down to. 1.5inx0.25in. I drilled clearance holes for the coils, through-holes for the three valve cover bolts I grabbed, and tapped holes to bolt the coils down to. I swapped out the valve cover bolts for studs, just because.

I tried a sample of both a brushed finish and engine turning. The engine turning turned out (hah!) pretty well very quickly, but I figured it would look weird with only a swirl or two showing at any given point. The sample bit is on the left, and the un-finished coil bracket is on the right.

The plate, as installed, with a partly complete wiring harness and the old coils/wires still there.

The ECU bits where I was replacing resistors and a transistor.

Two pictures of the final install. I can't put my strut bar in anymore (boo) but I think I can make a new one, or modify this one, to work (yay).

Before:

After:

 

[exdeath]

Your car sounds funny, what's that whistling sound?

 

[JCH13]

Your car sounds funny, what's that whistling sound?

Don't worry about it bro.

:awe:

 

[JCH13]

I wired in a little magic sauce in between the 4 coils. It's PFM how much a difference this little guy makes in how my car runs. It's totally different, starts better, idles better, develops smoother power. It's a beautiful thing.

I am now VERY happy with how the COP conversion turned out. Highly recommend to any Miata owners.