compiler error

[Michael Burger]

dsPIC33EV256GM106

My FlowCode 8 updated on 1 June 2020 now if I compile some of the analog channels don't work but they simulate fine.
If I load the pre-update compiled HEX file of the same program it runs perfectly and all the analogs works?
I have reported this before and the FlowCode team fixed it before, I think something went wrong with the last update?


Kind Regards

Michael Burger

[Michael Burger]

Quick Fix, if you have a backup
I replaced the (01.06.2020) CAL folder "C:/Program Data > MatrixTSL > Flowcodev8 > CAL" with a backup copy of the folder (09.04.2020) and it has fixed the compiler issue with the dsPic xx EV xx's analougue channels not working.


Kind Regards
Michael Burger

[medelec35]

Hi Michael,
You can also revert back to a known working version by following these steps

[Benj]

Hello,

Are all ADCs problematic or only specific channels using the latest CAL ADC file? Can you tell us which ADC channels aren't working? I'm looking at the code now and so far I've not spotted any problems but there have been some small changes to try and improve efficiency.

[Michael Burger]

Hello Ben,

With the CAL file from 2 June 2020 AN16 simulates but does not work in the real world once compiled, I will check them all with the latest CAL over the weekend and let you know.

Why is the speed for this device limited to 48MHz, it is a 70 MIPS machine and according to Microchip, it is capable of 70MHz?

Kind Regards

Michael Burger

[medelec35]

Hi Michael.
You should be able to manually enter a value as well.
What I would is select the 48000000,
then just highlight the 48 and enter new value

[Michael Burger]

Ben,

This is the test result with the analogs

dsPIC33EV256GM106 48MHz

CAL 9/4/20 CAL 1/6/20

AN19 Good Good
AN18 Good Good
AN17 Good Bad
AN25 Good Good
AN26 Good Bad
AN27 Good Bad
AN56 Good Good
AN48 Good Good
AN3 Good Good
AN30 Good Good
AN16 Good Bad

Kind Regards

Michael