Rpm signal loss

[AkRev]

Good luck tomorrow, I hope it works and solves your problems. Jeff is nuts, he is so tired and ready for the car to be done. I should clarify he is my brother in-law's brother. But ya, I am going to get a lecture on my car if I do not powder coat everything and clean it to death.

 

I will keep an eye out tomorrow and see if there is anything more I can help with.

Good luck

[seattlejester]

Changed settings, no dice.

 

Tach output wire is still showing simultaneous spark via test mode.

 

Even during cranking, none of the led's are flashing.

 

Any diagnostics anyone wants me to run while I'm still here?

[AkRev]

Must admit, that bummed me out when I read that last night. 

Had some time to reread and look at this a bit harder.

There is a bit of conflicting information out there about the Logic level settings between "Going high" and "Going Low." I am going to read more on this and try and find a definitive answer. 

One thing I want to check with you, what pins are you using to connect the logic drivers to the coils? Your work order specifies Pin 36 for spark A (DB37), Pin 10 for spark B (DB15), and Pin 11 for Spark C (DB15). Just want to make sure you have the Spark A on the 37 pin and the other two on the 15 pin.

[AkRev]

So,

Read some more.... and noticed something else

Please go to your Ignition Settings - Ignition Options- Wheel decoder.

Take a look at your setting on Number of Coils.

Please change your setting from Wasted Spark to Wasted COP.

 

Excuse me why I go beat myself.

[AkRev]

Also, one last thing. 

This part is is from http://www.diyautotune.com/catalog/megasquirtii-ems-system-smd-pcb357-assembled-ecu-p-171.html

"

• 5 volt ignition output (JS10 jumpered directly to pin 36) for EDIS, Bosch 124, or GM HEI ignition control"

I know that you you have a work order that shows how they have spark A go to pin 36.

But in case there is anyone else reading this I want to provide more information so this is not misleading to anyone.

 

In looking at this picture.

 

In the upper left you see the jumper from pin 36 to;

 The bottom right wire connection on that picture (PAD1). Also if you look at this picture, you will notice that JS10 is jumped to S12C and is used for other options that SeattleJester purchased from DIY. 

The stock ms2 3.57 is jumpered so that the settings will run JS10 and not D14. SeattleJester's is modified for D14. 

 

BTW SeattleJester, your boost control is jumpered on J11, I noticed that you do not have that turned on yet in tunerstudio, when you do please select J11.

 

Good luck

Edited April 13, 2014 by AkRev

[seattlejester]

Yea, I went in with high hopes, but yea....

 

Yes, output pins have been verified, one from DB37 pin 36 I believe (brown wire labeled spark A), the other from DB15 pin 10 and 11.

 

DIY actually told me to set it to wasted spark when I had it on wasted COP. I was told it does not make a difference? I can try giving that a change and see if anything happens.

 

Boost control is not connected so it is not outputted. Plan is indeed to set it to J11 when the time comes.

 

I'm giving it one last effort, and then going to either ask to send it back to DIY for a checkup to see if the board is outputting as it should, or if I somehow hurt only the ignition portion of the board, or take the board to my local MS installer, and see if they can shed some light. If the board checks out, then I guess I have to replace the coils and take the entire harness apart.

 

Thanks for the assistance guys, I really wished it was something we could have figured out, but alas...

[Matt Cramer]

Let me know if you need to set up a return. If we made a mistake and the spark outputs are shorted together, it will be fixed under warranty - if not, will need to charge a diagnostic fee for testing.

 

How exactly are the coils wired up externally?

 

 

I'm giving it one last effort, and then going to either ask to send it back to DIY for a checkup to see if the board is outputting as it should, or if I somehow hurt only the ignition portion of the board...

Edited April 14, 2014 by Matt Cramer

[seattlejester]

Thanks Matt,

 

Will do. I really don't think you guys sent the board out with problems. I'd be more then happy to pay a diagnostic fee if this solves this conundrum.

 

The coils are wired via...

32097-upgrading-coil-plugs-all-years-cop-writeup-cop_pins-jpg

1 ground
2 trigger
3 tach
4 +12V

 

found on here

 

http://www.miataturbo.net/useful-saved-posts-8/upgrading-coil-plugs-all-years-cop-writeup-12704/

 

Part of my last checks will be heading up and inspecting the output pins on the coil packs to the pad outputs to make sure each one is outputting to it's own coils (paired outputs). Once that is confirmed, I think wiring can be 100% crossed off the table, and then it will just be the settings/board gremlin somewhere.

[Matt Cramer]

Ok, if that doesn't pan out, email me and I can get you the return paperwork.

[seattlejester]

So a bit of an update.

 

With some remote testing and testing the outputs with the coils unplugged, it seems the ECU is happy. Matt has been very patient and helpful with that process. We still haven't crossed an undetectable fault off the board, but it seems supremely unlikely that just the ignition portion of the ECU is unhappy when plugged into all three coil channels. This points us back towards wiring and COPs.

 

Test mode with all 6 coils unplugged are outputting as it should. Plugging in one input at a time behaves as it should (i.e. plugging in only coil 1 and firing channel A only fires coil 1 and 6, removing the plug for 1-6 and plugging in the others does nothing). This points to the problem being isolated to the coil on plugs firing simultaneously when more than 1 channel is plugged in to the coil on plugs. Continuity tests further points towards the wiring harness being without fault. The wires I am using also have an additional oil protection coating so cross talk through 4 layers seems unlikely. Just in case, the wiring harness will be separated as a last resort to see if that is a source of the problem.

 

From further reading wasted COP setting is used when you have 6 coils wired into outputs A, B, C, D, E, F, and want to pair the outputs. Setting wasted COP will have the ECU fire two of the coil channels. With the way things are wired it actually does not make a difference whether I choose wasted spark or wasted COP as with wasted spark the 3 channels fired are A, B, C, while with wasted COP it would be A-X, B-X, C-X. I think for simplicity I will revert back to wasted spark.

 

If the LED's are to be believed, then the outputs when unplugged from the COPS are going as they should, test firing A has the PAD1 led blink in going high etc etc. 

 

I have ordered another set of COPs, annoyingly they only really come in sets of 4. I bought two sets last time so I have a total of 6 unused, but just in case, I will use just 3 of the cops from the new set on cylinder 1-2-3 to see if the COPs were the problem.

 

So the questions are...

The difference between going high and going low. If the LED's are to be believed, going high means a pulse of power is sent to the COP as the led blinks when firing spark. Using the same logic, going low the LED's stay on and I imagine if you had sensitive enough equipment you could detect a very subtle blink when the circuit is triggered. The way the board is wired and the COPS I am using suggests I should run going low (logic output, built in igniter coils). On the flip side, some people are saying to run going high on the miata forums. 

 

One thing that is a bit of a concern is a slight fluctuation in voltage while the ECU is connected. My voltmeter in the car shows no fluctuations, but in tuner studio I am seeing maybe a rise and drop of 1/2 a volt or so with just the ignition on, not running anything. What could cause the pulsing voltage? Have other people noticed the slight bouncing of the voltage in tuner studio? I am running the VR delete and an internally regulated alternator, so perhaps until I get above running voltage I won't see anything?

 

Akrev: This one is for you, would you happen to know if Jeff followed the diagram below? You mentioned not needing one for the ground side, it seems like the capacitor is supposed to be bridged between the power and the ground?

 

So current theories in order of likeliness:

COPS are bad (I did run them in the going high configuration for a while)

Strange Voltage (perhaps the fluctuations are causing some kind of problem?)

Wiring is bad (cross talk of some sort)

ECU is bad (highly unlikely it seems)

 

I think this may be my last week before I'm done playing with it. I have a bit of money saved up, the shop that installs megasquirt has gotten back to me saying they are willing to help, and they are only 20 miles away and I should have my AAA renewed later today.

Edited April 24, 2014 by seattlejester

[AkRev]

Still going through stuff to find the pictures from Jeff, I sent you the one that I had that is not on his build. 

One thing I remembered looking back at things, was when you put in the higher amp fuse for the coils. This was when you had the settings on "going high." 

I do not know if your coil got damaged in the process, but I have a hard time understanding why you had triggers triggering when they should not have been. To be honest, wiring seem the most likely. I would unplug the connectors and first test for continuity between sets on pin #2. 

 

One other possibility is that one of the coils went bad heated up and damaged wiring in the coil. Could also explain the blown fuse. 

Any way you could pull the coils and "bench" test them with megasquirt?

Edited April 25, 2014 by AkRev

[seattlejester]

Yes that has been weighing on my mind, that and the incident when the battery just quit. Granted the engine did run which is the most confusing of this whole scenario, and the toyota COPs are supposed to be smart and have built in protection from being overloaded. Granted I'm sure they can only tolerate a certain amount of stupid.

 

The fuse is on the higher side of what I have heard, but I hear in perfect working order they should see 7-8 amps functioning correctly. If I needed to go all the way up to 30, it definitely wouldn't be outlandish to think that the coils are unhappy, hence why I ordered the new set. I'm hoping to run 3 of the new coils  only and see if they output correctly and then use the last three spare coils to pair the outputs. I think if all works at that point, I will have to assume one or more of the original 6 are bad and I can go about testing them.

 

Coils can come out easy enough. Not sure how much they are going to like firing without an exit nearby (spark plug). I'm just not sure on the best way to ground test the coil on plug setup. If I can find a way to bench test the coil on plugs I would definitely like to try it, most of the tests I'm finding though involves swapping coils or having a magnetic pickup tool.

 

Trust me I wished it was just my wiring, my life would be much easier at that point. For the wiring to exhibit the way it does, the spark outputs would have to be continuous to each output. As in continuity from A-B-C when testing any of the output pins. And they show no continuity between inputs or output pins. The only time it looks a little problematic is when actual COPs are connected.

[AkRev]

Understood.

This is more what I meant by bench testing https://www.youtube.com/watch?v=Z12HPGaw25k (personally I like to pull the coil and plug and lay them across grounded metal so they can spark. (And its fun to take time lapse pictures ) Not testing them with out the plugs in there.

I am trying to think about a way to test the coils. I can think of two ways, the first is if you have a temperature reader and can read the temps of the coils ( the bad one should be hotter in running conditions.) The other one is to hook up ground and power to the coil, and see if you can get any grounding signal from the trigger pin.

 

When do your new coils come in? 

[seattlejester]

That seems easy enough to do.

 

Grounding signal from the trigger pin. Would that be to check if there is continuity from the trigger pin to the ground? I'm pretty sure I would need an oscilloscope to capture the actual firing event. Sorry still not all to comfortable with electronic aspects.

 

Coils look like they are coming in next Monday...although I bought them from someone less then 100 miles away on Tuesday.

[NewZed]

Following your thread just for the complication of the problem...  Browsing around for other views on COP I came across another (one of billions) Megasquirt page that seems informative.  They make an interesting statement about dwell control and "over-dwelling" (my term).  The coils discharge to save themselves.  

 

Considering your blown fuses and the fact that you have too many sparks, it might be worth consideration.  The statement and the link are below.  A simple test might be to turn the dwell setting way down to no-spark, then bump it up.  You'll either get controllable correct spark or be right back to where you are.  I would do this with the new coils anyway, if it turns out you've burned the ones you're using.

 

"Note that some coils have a built-in overdwell protection feature. If given too much dwell the coil will automatically spark. This can give a dangerous advanced spark. Be sure to strobe your timing at high revs to ensure this is not happening."

 

 

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

[seattlejester]

That's a good point, I was running a pretty unreasonable dwell timing for the spark based off of internet threads. I think it would indeed be best to start from the bottom and work my way up. I am finding some definitive safe dwell times for people using toyota cops and MS2. I really should have read more on the miata forums then the toyota forums it seems.

Edited April 25, 2014 by seattlejester

[seattlejester]

So yea...I'll write up everything when I am in front of a computer instead of my phone...but yea pretty much a level 1 mistake was the cause of my problems.

 

Current issue 3000rpm idle. Most likely cause vacuum leaks probably have to take the manifold off. Reroute lines etc.

[AkRev]

Congrats!

[madkaw]

I must of had a little luck because so far the MS has been flawless!!

It doesn't really matter what level of project you are doing- the mistakes seem to be the easy ones. Keep plugging at it.

[seattlejester]

So, it all goes back to my method of checking which coils were firing. The cheap timing light. I don't know how or why, but that seems to be the source of a long goose chase.

 

I replaced all the coils without problem, tried starting the car to no luck. Took out the spark plugs, gave them a good clean, and plugged them back in to no dice.

 

Changed the setting from wasted COP back to wasted spark, and I was getting at least signs of life, but still the stalling etc. Checking with the timing light confirmed the same behavior as previous...except, this time the light started blinking and then just stayed on. Running test mode, the light would just stay on. I realized the 10$ light was junk and that I had yet to actually check the spark coinciding with each channel. I had bought some inline coil testers at harbor freight (easiest way to check a coil on plug by the way, very visually friendly, only problem is the pin doesn't quite fit into the COP, I imagine it is american size so it needs to be turned down a bit). Even with a poor fit, it would be the best route to ground and so at worst it would easily ark for a quick test. I had bought three to check the actual channels. So I checked A, coil 1 firing, checked B, and coil 3 was firing, and I checked coil C and coil 2 was firing. Looked up to quickly confirm that firing order should be ABC, reordered the two channels. And she started right up.

 

The error was the bloody equivalent of the wrong plug order! Freaking level 1 mistake there. I really wanted to shoot myself...

 

I think the error came from when I soldered in the pins for the spark output on the DB15. I do remember there was some confusion with pad B and pad C being out of order that I mentioned somewhere. I had ran both wires of the same color and labeled the ends. At some point the ends were removed and the wrong plug was plugged into channel B and the remainder must have been plugged into the open port. Most of the wires that I added were grey so I am guessing the confusion is coming from there. I had only checked continuity on the spark A output on the harness as I had assumed if they were firing at the same time, the continuity would be present in the harness. I guess my biggest mistake was trusting the timing light as reading true. It must have been picking up a stray signal and actuating the flash, I'm willing to bet my coils from my old set are probably perfectly functional.

 

Things to note was that, the car would not spark in wasted COP mode. I'm not sure exactly why. Maybe the board recognizes the lack of channels DEF and cannot pair the ABC inputs and doesn't want to fire or something of that nature. I've lowered the amps to 20 and the coils are still firing very happily. 

 

Literally two or three cranks and she fired right up. Really caught me off guard with the 3000rpm idle though, I thought the throttle was open so I quickly grabbed the accelerator with the front of my shoe and pulled up only to find it was still running 3xxxrpm with the tps at -0.1% (tps was calibrated, so closed tight). I'm not to uncomfortable with saying the intake manifold is at fault here as I did notice a puff coming from the cold start injector block off plate when the car had backfired in the past. Engine is getting air from somewhere to idle that high, and quite a lot of it. Once I check the surfaces on tighten the hose clamps on the boost reference lines, I'll test with a bottle of brake cleaner to look for vacuum leaks and then adjust the idle plate to turn it down.

 

Thank you very much for the assistance guys. From the beginning I had a feeling it was going to be something stupid on my end. Thanks for being patient with me and going over several settings and entertaining my ups and downs. Car is registered and insured, so hopefully I'll be able to wrap up this thread with the final settings for my given setup and a success story at the end.