Megatune causing resets SOLVED

[JustinL]

Hello, I've got my car running with the microsquirt now, but I'm getting resets when I plug in the db9 cable and run megatune. I've tried 2 laptops, one with a USB-DB9 converter and yesterday I tried an older laptop with a serial port and the same problem persists. The car runs very smoothly on µS even with all the serial plugs connected to the laptop. It's only when I open megatune that it starts with the resets and the stalling. This makes me think that it's the flow of information in the wire that is the problem. Would there be a solution in the comms section of megatune to alleviate this? Do I need shielded cable?

Thanks!

Justin

[efahl]

Justin,

If I'm reading this correctly, you do not have the problem when the cable is plugged in and MT is not running? That's very strange, usually these problems are due to the cable acting like an antenna and pumping noise back into the ground plane or something like that, so they occur irrespective of MT running or not.

Hmm. Would you be willing to cut up your cable? I'm thinking it would be interesting to cut out all the wires at the uS end of the cable except for the three that are actually used (2, 3 and 5)... http://www.not2fast.com/megasquirt/doc/cable.html I'm guessing wildly here, but maybe some of the other wires are telegraphing signals into the connector???

Eric

[efahl]

Oops, as usual I forgot to answer part of your question...

What version of MT are you running? The 2.25p1 version was recently updated to 2.25p3 by the addition of some better initialization in the comms section and that might help.

Eric

[JustinL]

Thanks for the ideas Eric. I have a spare cable, so I can cut one of them up. I'm also using P1 so I'll update to P3 and see if that helps. I tried changing the comms data rates and it still resets with any rate and delay. I'll try software first, and then slicing and dicing the cable.

Reporting back soon,
Justin

[JustinL]

I updated to P3, and found that my cable has only 3 wires in it, so there was no dissecting that I could do. Still no resolution to the resetting problem. I removed all but one of my ground wires one at a time thinking this might remove any grounding loops, but this didn't work either. It still runs fine as long as I don't try to look at any data. I'm not sure what my next step should be.

Justin

[efahl]

Justin,

You're way out of my area of expertise... Maybe try shielding the wire by wrapping it with aluminum foil? I remember Andy Whittle doing that on some MAF lines and it was a good diagnostic.

Eric

[JustinL]

Tonight I tried wrapping the wire in foil, new spark plug wires, new resistor plugs. Still resetting. Is there any way for me to narrow this down further? Is there another program that interfaces with microsquirt or just fires up the comm port to see it I can generate resets a different way?

Thanks for all the input,

Justin

[Matt Cramer]

There's a program called PortCheck.exe bundled with MegaTune that sends signals through the COM port to see if it responds. That may help test things out.

[efahl]

Or maybe just good old hyperterm?

All Programs -> Accessories -> Communications -> HyperTerminal

You can tell if you are connected properly because "S" and "Q" typed into hyperterm should cause MS-II code to echo back their signature and version info, respectively. (That's basically all that portCheck does inside a loop that opens every port at 9600 and 115k baud...)

Eric

[SQLGUY]

you don't by any chance have the car on a battery charger and the laptop running off of A/C power, do you? If so, get rid of one of them. You can get ground loops through the laptop when it's plugged into the wall.