Code only runs fully with USB plugged in - SOLVED

[jmarcelino]

@smssms
What's powering it when not connected to USB?

[smssms]

@jmarcelino
XP JCJ1012S05 1.5A 5V output DC/DC converter into Vin/Gnd Plus 10uF capacitor for smoothing as it also has to run a stepper motor. The max. draw is only 800mA, so well within the supply's limits. The stepper motor is running in the first 1/3 of main.py too with no problems. But without USB, it just seems to hit a point where it freezes with no error messages (that I can capture).

http://uk.farnell.com/xp-power/jcj1012s05/converter-dc-dc-10w-5v/dp/1637871

[jmarcelino]

@smssms
Interesting setup thanks.

So the DC/DC converter is mounted on a separate board which then powers the WiPy on the expansion board or do you have a custom board for the WiPy?

Can you connect some cheap FTDI-style USB - Serial TTL (@3.3v) adapter to the TX + RX + GND pins on the WiPy in order to get the serial output without powering the board too?

[smssms]

@jmarcelino
Yes, the DC/DC is on a breadboard and jumpered across to the WiPy Expansion Board.

Don't have the parts here now to try comms only, but will look into it - thanks.

When it is running without USB (before it fails), I can get the code's serial output over Telnet, but there is never an error message when it stops...

[jmarcelino]

@smssms
Have you tried running it without the stepper motor working (i.e. motor disconnected from power)

[smssms]

@jmarcelino
Yes, this is the annoying thing:
If I take the stepper motor out of the equation (commenting it out) the rest of the code seems to run ok.
But, it can run the stepper, it does so in the first 1/3 of main.py consistently with no problem.

[livius]

@smssms
i suppose that @jmarcelino mean do not remove it from code
only disconnect it from power source

[smssms]

@livius @jmarcelino
Ah, I see.
Tried that and yes, the code seems to run ok with the stepper unpowered.

[livius]

@smssms
It would be nice to see how it is connected
Because if you power the motor and microcontroller from a single source then you must fight with disturbance
Some ceramic condenser, diode, RC/LC, etc.

minimal protection will be diode:

some hint also:

[smssms]

SOLVED - I powered the stepper motor separately and all is good. I'm guessing there were transient demands in excess of the 800mA I could measure that were causing problems.