Bug: 1.4 Firmware and IO Extender

[krisr]

Hey guys,

Just noticed and can reproduce a condition after flashing my test rig to the new 1.4 firmware which is hooked up to an IOX box over CAN.

For testing I setup the IOX HOUT and LOUT outputs to become active if clt > 20 on the stim. With the old 1.3.4 firmware, the LED's will operate as they should and stay on if the conditions are met. With the 1.4 firmware the LED seems to just "flash" and not stay on after the conditions are met. With either MS3 firmware, I can configure the IOX itself general purpose outputs to trigger if the MS3 reads clt > 20 so it looks like the IOX is getting the data over CAN comms normally. I flashed back to 1.3.4 and the IOX controlling the stim LED behaviour was as expected and working as before.

Attached are MSQ's, not sure if a datalog would help though?

Kris

[racingmini_mtl]

I can reproduce your issue. But if you disable the CAN input port polling on the MS3, then the outputs go back to working correctly. There seems to be a conflict between the input port polling and the output port writing. Please try this to confirm my findings.

I don't have the source code so I can check what the issue might be. But it must be in how the CAN packets are sent with those two options.

Jean

[krisr]

Ok I can confirm your behaviour of turning off "Enable Input Port", the outputs then work again and ADC's work too. It's something that's not urgent for me right now as I was going to setup my shifter inhibitor switch (for in/out of gear spark table switching) and brake light switch to the IOX inputs but the car is apart at the moment so that wont be happening for a couple of months.

[jsmcortina]

I don't have the source code so I can check what the issue might be.

I'm planning on releasing that imminently.

James

[racingmini_mtl]

I've had a look at the MS3 source code and found the issue. I've sent James the solution.

Jean

[krisr]

Nice Thanks Jean/James!

[jsmcortina]

I applied Jean's suggested fix and krisr confirms that it works. This will be in the next beta code version.

James

[Speedy_G]

could somebody post the fix here, so that we can apply it onto the source code.

Or maybe easier for all, to reupload the source code with the bug fix?

[jsmcortina]

There will be another beta within a month or so. Historically it has been shown that minor bugs are always detected only after a major release, so I'll wait a period of time to collect them up before putting out 1.4.1.

James