So I've been assembling my MS3/MS3X for my 90 miata using Frank's guide. Got pretty far, was able to update the firmware, connect to tuner studio, and flash a map. I then tried to adjust the VR circuits and could NOT get the upper leg of R54 to go lower than 4.9v, and couldn't get the lower leg of R17 above 2.3v... no matter how much I turned the potentiometers (I probably turned them too much). All of this was with the daugterboard plugged in, btw.

Then all of a sudden, after adjusting the pots some more, R54 and R17 read super low, like 200 mV. Checked the 5v out on the voltage regulator/U5 and it was also really low. U5 got super hot, too. Like too hot to touch. I pulled off the daugtherboard, and voila... U5 shot back up to 5v, and the overheating stopped.

What can you guys make of this info? I feel like I must have fried the daughterboard, and I must have ruined the pots as well. Is it bad to adjust them while the megasquirt is powered on?

Thanks a lot for your guys' help, it means a lot.

 

I'm going with you shorted it somehow from assembly, and it took a while for the short to heat up and break something.

typically when you short the daughterboard it's instantaneous, but the cpu itself typically heats up, not 5 volt regulator.

Makes me think you put a diode/cap in wrong or something.

 

<p></p><p>Post a high-res picture of the board</p>

 

<p>Did you ever remove those diodes you had?</p><p>I had the same issue of shorting the power supply assembling mine.</p><p>Doubt the daughterboard is fried, other stuff might be though.</p><p>Also post pictures of everything.</p>

 

 

These pictures okay?

Braineack: I double-checked all the diodes on the main board and didn't see anything off. Tried checking the capacitors but it's really hard to discern polarity.

aidandj: I did remove the diodes. Connected S12 on the main board to D1 on the MS3X, just like Braineack said.

As you can see the ECU connecter is already hooked up. So the MS3X is powered along with the MS3. Not sure if that matters but thought it worth mentioning.

 

you installed r56 backwards...


there's writing on the three caps with polarity that points to the the 12v lead. (there should be at least, I couldnt tell from your pics)

 

Didn't know the pot was directional. Thanks. I'll flip it around right now, but I assume the damage has already been done. Any suggestions on how to narrow down the damaged components?

I couldn't see the writing on the caps at first, but I see it now. They are on correctly.

 

im sure it's fine, just introducing 5v fault to ground.

 

I flipped the pot around and the problem is still there. Voltage regulator now outputting 0.4 volts.

 

<p>When unpowered is there continuity between ground and the output of the 5v regulator?</p>

 

No, there isn't any continuity.

 

looking again, i don't see how that could fry something. i think they may be unrelated.

it would basically just keep the voltage at r54 at 5v like you were seeing.


double check everything you did in your wiring "harness"

 

Double-checked my harness and I don't see anything wrong. I followed this diagram:


The only difference is, I have a 1k resistor bridging pin 22 with pins 8 & 9 for ground (for the stock TPS). Pin 22 is not connected to the ECU harness at all.

Also I remembered that, when I powered up the megasquirt to adjust the VR circuits, I accidentally hooked up the ground source to one of the sensor ground pins (main pin 1 or 2) instead of the power ground pins (main pins 8-11). Didn't think anything of it since they're both ground. Could this have damaged it?

 

No. I think it's something dumb with a wire going to the wrong place causing a fault and once you find it all will be okay.

 

Remove the 1k resistor and unplug the stock tps

 

Pin 22 of the main board. It says in Frank's guide to connect the tps input to ground (for those planning not to use a variable tps), so that's what I did. I actually started a thread a few weeks back for clarification regarding this matter. I plan to run a vtps eventually but I want to make sure it runs well completely stock first.

So I can just unplug the tps instead of running the 1k resistor, huh? And the car will run fine with the tps unplugged? So much simpler.

 

Leave the resistor. Keeps the tps value locked in place, otherwise that input floats

 

Turns out the daughterboard WAS fried. Replaced it and it seems fine now. I guess the daughterboard doesn't always heat up when it goes bad.

I still have the problem with the VR circuit on the MS3X board though. Lower leg of R17 maxes out at 2.33 volts. Anyone know how to troubleshoot this?

 

Looking at this you probably figured it out, but if you haven't I noticed the same thing.

Don't worry about it. I was able to get a cam signal with three turns clockwise from full counter clockwise. I believe this put me at about 1.5 volts.