Reset/Lock when serial is connected

[hybrid]

I have moved my Microsquirt over to my Mustang project, and I'm getting resets when the serial is connected. This was happening on the bike too.

When it happens, the notebook loses connection to the ECU, and the engine stalls, but the fuel pump keeps running until I turn the power off and on again.
It's also happened on a notebook with a serial port, and a USB to Serial converter.
Anyone seen this issue before?

My ignition trigger is a hall sensor connected to opto-in using 12v signal. On the bike I was also using opto-in, triggering from a tach signal from the original CDI.
I have seen some suggestions of putting a power filter on the 12v input to the ECU - has this helped anyone?
I'm assuming this should go as close to the ECU as possible to filter the noise before the ECU?

[Bruce Bowling]

Question - are you using the internal ignition coil drivers? This has caused some installations grief due to the 400V flyback pulse, and the serial connection often gets the brunt of this and causes resets, etc. If you are using the internal drivers then you may have to make sure the ignition drive wires are well separated from the other wires, which is a bit hard to do on the single AMPSEAL.

- Bruce

[hybrid]

Thanks for the reply Bruce.

I'm using one of the internal ignition drivers, but it's only driving a MSD 6A, not a coil, so I'm assuming the flyback would not be an issue?
When it was doing it on the bike, I was only using the Micro for fuel - ignition was controlled by an aftermarket programmable CDI.

Also, it seems I can have the serial cable physically connected and the car running, but I cannot have Megatune or TunerStudio running while the car is running. So I guess it's once the serial port is opened on the notebook that the issue pops up.
Is it worth trying to separate out the serial wires as close to the ampseal as possible?
I'll only be able to go as far back as the tape on the DIY Autotune harness though.

There would only be a few inches of the serial cable in the split tubing I'd say.

[Bruce Bowling]

Hard to know exactly how the noise is coming back, but if there are resets then somehow the uS power is getting disrupted somehow. Yes, separating the serial lines from the rest of the bundle may help. Is the laptop running on batteries, or is it plugged into the AC mains?

If you can get your hands on a scope, if you could look at the Vref signal while this is occurring that would be of interest to see if there are dips, etc. Also, just look at the +12V power rail itself, I have seen where CDI boxes spew all sorts of noise on the vehicle +12V.

- Bruce

[hybrid]

Laptop has been running on battery lately, but I've had my old one on AC power (both saw this issue).

I've now put a stereo noise filter on the 12V line as well, so I'll see how that goes.
I think krisr is going to bring his scope over, so we can scope vref and 12v. I have a scope that I left at my parents place when I moved out, but it's an old big "buddy" thing. Kris has a nice hand held one.

[hybrid]

For the record, this seems to be fixed by tying the bootloader wire to TPS VREF as suggested by James on the msextra forums.