Megasquirt and Variants Running Thread

[jpcturbo]

All power grounds go to the head and then to the body behind the battery with the exception of some relays located where the PM was.

So the head grounds to the body and then somewhere the battery is connected to the body? If that's correct, how about skipping the body or adding a dedicated head to battery line?
Also how much load does the load pro apply? I read one place that it was only 1/2 amp - is that enough? How I test grounds for O2/Wideband gages is that the reading should not change when I turn the headlights on and off.

[cordes]

So the head grounds to the body and then somewhere the battery is connected to the body? If that's correct, how about skipping the body or adding a dedicated head to battery line?
Also how much load does the load pro apply? I read one place that it was only 1/2 amp - is that enough? How I test grounds for O2/Wideband gages is that the reading should not change when I turn the headlights on and off.

The negative terminal from the battery grounds at the same spot on the head which everything else grounds to via a 2ga wire. A 2ga mil-spec wire goes from there to the body. This works swimmingly well in Joy's Daytona, and from what I've read, it's the "ideal" way to do things.

The Load Pro is a 25 ohm resistor, so it doesn't pull a huge load. In practice, I've seen dramatic drops in voltage when I use it for circuits with too much resistance. It's pretty crazy really. I would have been chasing a headlight problem in our Neon before I came across that tool. It was pretty neat.

[c2xejk]

I did some tests today. When the key is flipped forward, the sensors go high, but in the time the fuel pump primes, there's some interesting stuff going on with the cam and crank waves. Why does it get pulled down repeatedly like that? I was able to duplicate it many times by just flipping on the key.

The only change I made today was to take all of the grounds and run them to the head. So now literally everything with the exception of the wideband gets grounded to the head. The wideband is grounded in the kick panel area where the stock ground is located.

The frustration level is not going down. Good times.

So how is the power feed during this test? (these are hall-effect sensors, right? Pull-up or Pull-down?) Also, at what point do you start cranking? If you have the channels it would be nice to see the 12v supply, signal and crank signal

- - - Updated - - -

I did some tests today. When the key is flipped forward, the sensors go high, but in the time the fuel pump primes, there's some interesting stuff going on with the cam and crank waves. Why does it get pulled down repeatedly like that? I was able to duplicate it many times by just flipping on the key.

The only change I made today was to take all of the grounds and run them to the head. So now literally everything with the exception of the wideband gets grounded to the head. The wideband is grounded in the kick panel area where the stock ground is located.

The frustration level is not going down. Good times.

So how is the power feed during this test? (these are hall-effect sensors, right? Pull-up or Pull-down?) Also, at what point do you start cranking? If you have the channels it would be nice to see the 12v supply, signal and crank signal

[cordes]

OK, here is the key flipped forward. It didn't seem to do much at all while cranking, so there weren't exciting pictures to show. Ch1 was 12v feed, Ch2 was 5v reference signal to sensors, Ch3 was crank signal, and Ch4 was either the sensor ground or power ground, depending on which time I was testing, as I wanted to see both at different times.

Of note was the fact that this repeated grounding of the crank sensor seems to happen about the time the fuel pump is done priming. Sometimes it'll throw off two such patterns. Other times, just one.

ETA: The site doesn't like the format for the photos unless I put them into paint, save them again, etc. etc. so here it all is as google drive links.


https://drive.google.com/file/d/1sEp...usp=drive_link
https://drive.google.com/file/d/1bEI...usp=drive_link
https://drive.google.com/file/d/1H-b...usp=drive_link
https://drive.google.com/file/d/1ED4...usp=drive_link
https://drive.google.com/file/d/1wlG...usp=drive_link
https://drive.google.com/file/d/1g2B...usp=drive_link
https://drive.google.com/file/d/1V2s...usp=drive_link
https://drive.google.com/file/d/1-l2...usp=drive_link

[wallace]

Have you tried posting this issue on MS extra forum? It sounds like a flyback issue if it happens when the fuel pump shuts off.

[cordes]

Have you tried posting this issue on MS extra forum? It sounds like a flyback issue if it happens when the fuel pump shuts off.

I did, and there weren't a ton of replies, though I haven't kept it very updated.

I'm glad someone else brought up flyback voltage. I was assured on my initial call with them that this should not be an issue, but I guess I could check by pulling the pump fuse next time I conduct the test. I'll also disconnect the WBO2 next time. It hasn't put out a good signal since shortly after this whole ordeal started. I know whatever is causing this must be a low probability issue at this point.

["Top Fuel" Bender]

I'm using this one from Littelfuse

https://www.digikey.com/en/products/...pGI2NBBEAugL5A

ETA: My alternator either died, or the alt control in the MS stopped working. I'll probably take the alternator to get tested tonight. I'll be sending off the ECU to DIYautotune, as I'm fairly certain after further testing that the unit is bad.

That sucks you're having issues with it

2 things Jaren had issues were using a battery charger while trying to start his
he said it killed his crank sensor
I believe he's using the same ZF brand Cherry Hall sensors as me GS100701

He also had alternator charging issues
EFI Source walked him though this and he ended up with a 1-wire Mopar Alt. (not cheap though)
I'm running an external regulator with no issues on mine ,though it is charging high, gonna get an adjustable regulator

I forget , our both cars running different models of MegaSquirt ?
Would be nice you could swap them around .
If you want to eliminate the cam signal you could always run it on wasted spark too

[cordes]

That sucks you're having issues with it

2 things Jaren had issues were using a battery charger while trying to start his
he said it killed his crank sensor
I believe he's using the same ZF brand Cherry Hall sensors as me GS100701

He also had alternator charging issues
EFI Source walked him though this and he ended up with a 1-wire Mopar Alt. (not cheap though)
I'm running an external regulator with no issues on mine ,though it is charging high, gonna get an adjustable regulator

I forget , our both cars running different models of MegaSquirt ?
Would be nice you could swap them around .
If you want to eliminate the cam signal you could always run it on wasted spark too

Thanks for the feedback, Bruce.

I have Microsquirt on Joy's Daytona, and an MS3 Pro Mini on the Shelby Lancer. It seems like the alternator is once again charging with little change in anything I've done. I guess it could have been a poor connection at the alt itself, but I think that's low probability.

I do have a second MS3 Pro Mini on the way to put on another car. It will also settle the question of this actual unit being a bit flaky vs. something in the wiring causing the problem.

I had previously read that one should not attempt to start the car with a battery charger on it, so I've avoided that entirely during this process. Recently, I had fired up the car without remembering to plug in the cam sensor. It exhibited the same issues as before with the only exception being that it did show a cam sensor fault. I think that rules out the Littelfuse cam sensor being the primary culprit at this time.

I think the new unit should be here any day now, so it won't be long until that's tested.

[wallace]

Did you change out the little fuse sensors for the diy ones?

["Top Fuel" Bender]

I have the coils mocked up with some wires on them. I modified the distributor today also. In the future I may tap the dist for the 12x1.0 thread rather than use the two nuts alone.

Hey , just noticed which LS Coils you have there
You may wan to use ANY other of the LS coils besides THOSE
The LS2 D585 "Truck" Coil easily identified with the heat sinks
have been known to cause issues

they have a built in protection that will randomly discharge the coil pack to protect itself if the dwell is too high
or it feels it's gonna overcharge
which becomes a BIG PROBLEM if it discharges at the wrong time causing pre-ignition or knock

also the 1st gen LS1 coils have the connector and plug wire on the same side
I was gonna do similar to yours , they tuck in better for packaging but not as readily available as the LS2 coils
The connectors are also different

[cordes]

Did you change out the little fuse sensors for the diy ones?

I changed out the crank sensor. It made no difference. It also made no difference with just the crank sensor plugged in. There is definitely something else afoot here.

Hey , just noticed which LS Coils you have there
You may wan to use ANY other of the LS coils besides THOSE
The LS2 D585 "Truck" Coil easily identified with the heat sinks
have been known to cause issues

they have a built in protection that will randomly discharge the coil pack to protect itself if the dwell is too high
or it feels it's gonna overcharge
which becomes a BIG PROBLEM if it discharges at the wrong time causing pre-ignition or knock

also the 1st gen LS1 coils have the connector and plug wire on the same side
I was gonna do similar to yours , they tuck in better for packaging but not as readily available as the LS2 coils
The connectors are also different

After some reading, and a little advice from guys who use them, I wasn't afraid to use the coils, as I don't think I'll ever get to dwell levels necessary for them to auto fire. The idea of the plug wire and connector being on the same side is interesting though. Perhaps a little more dielectric grease wouldn't be a bad idea.

[wallace]

I've used the 582 coils on multiple cars with MS without issue. You do need to set the dwell when you set the ECU up.

[cordes]

I've used the 582 coils on multiple cars with MS without issue. You do need to set the dwell when you set the ECU up.

That's good to hear. I leave the dwell on the default setting, which should be more than enough, and from what I've read it's also under half of what it would need to be for them to auto fire.

[cordes]

The new ECU is exhibiting the same symptoms as the other one. Not a huge deal since I wanted a spare anyway. However, that leaves me more perplexed regarding this issue.

If I pull the fuel pump fuse from the bus, it won't show the temporary ~300RPM when I flip the key forward. It doesn't seem to help the crank trigger signal either though.

[wallace]

Are you using the MS 5v reference for the crank and cam sensors?

[cordes]

Are you using the MS 5v reference for the crank and cam sensors?

I was, but the way it was explained in the MS3 Pro Mini manual, it seemed like it should get switched 12V. When I made that change, it ran better, but still had a miss. I'm thinking of going back to that.

[wallace]

The ground for the sensors is connected to sensor return wire and not to an actual ground? I can't remember I think I'm using 5v source for mine.

[cordes]

The ground for the sensors is connected to sensor return wire and not to an actual ground? I can't remember I think I'm using 5v source for mine.

Correct. The crank and cam sensor grounds are tied into the sensor signal ground wire.

ETA:

OK, so I figured out why it would only intermittently start, and I'm blown away that I was able to get this thing to run about everywhere before I had any problems. I had the 12v switched feed for the crank sensor teed into the wire which powered the injectors. That was powered off of the fuel pump relay. Of course, that's not energized the entire time the key is forward, so I must have been getting the car to start by cranking immediately after flipping the key forward in order to have the crank signal keep the FP relay on.

Today I hooked up my 5vref wire again at the crank, it has started with every attempt. I can only assume my quick starting style got me through after the wiring change. I am hoping that Wallace's tip of replacing the crank sensor with the "Cherry" type sensor which DIYautotune sells solves my problem of the occasional sync loss.

Now, I just need to figure out how to test the WBO2 sensor which is currently reading full rich. It's a Spartan unit from 14point7.com and works over the can bus. This will be another new adventure which I'm not necessarily looking forward to.

A quick How To along with a couple videos which guided me to this solution are to follow in my next post.

[cordes]

How the crank sensor works with MS3 Pro Mini:

The computer generates 5v and sends it down the signal wire to the hall sensor at the crank. The sensor then grounds this signal when the teeth of the crank wheel (usually a 36-1 configuration) go past the sensor.

In order for the transistor in the hall sensor to be able to function thus pulling the signal down as the teeth pass, there must be a good 5 to 12v source and a good ground path on the sensor signal ground wire.

To test the sensor wiring with a cheap oscilloscope or volt meter:

With the ECU unplugged, Test the sensor 5-12v power wire by running 12v to the place it initially receives power and pulling a load on the wire at the connector at the crank by either hooking up a spare headlight bulb, or one can use my preferred tool, which is a Load Pro. This will verify that the wire has a good path to the sensor connector with no shorts or too much resistance.

With the ECU unplugged, Test the sensor ground wire by running 12v to the connector at the crank and pulling a load on the wire at the ECU connector by either hooking up a spare headlight bulb, or one can use my preferred tool, which is a Load Pro. This will verify that the wire has a good path back to the ECU with no shorts or too much resistance.

Power up the ECU and ensure that there is ~5v at the signal wire at the connector to the hall sensor (if there isn't voltage there, the ECU is probably bad). Ground that signal wire and measure the voltage on the wire at the connector to the ECU. I used the sensor ground wire as the ground for my scope to replicate the path the sensor would use from the ECU connector and out through the power grounds. The voltage should reliably drop to zero and then rise back up to ~5v each time you ground the sensor signal wire. If the voltage reliably goes from 0 to 5v when repeatedly grounding the signal wire, the signal wire should be good.

To test the sensor after wiring tests good:

After one has verified that the wiring is good, then the same connection at the ECU for the signal wire and sensor ground can be used to measure the signal from the sensor while cranking. It is ideal to have a scope in order to view the square wave generated by the sensor as it goes from 5v to 0v. However, a multimeter will average this signal outputting something greater than 0 and less than 5v. If a signal is seen, then the sensor is good.

HOWEVER: It would be wise to double check that the sensor is seeing a good 12v source WHILE CRANKING (this one bit me badly) and a good sensor signal ground as previously mentioned.


Here are two good videos by my new best friend, Scannerdanner: