I'm using Numato 16 GPIO board to switch relays on 8 Channel Relay Module. Until no 220V load is connected to the relay all works fine and I can switch on/off relay using Serial port command to GPIO as many times as I want. But once I connect actual device (in my case pump) through the relay - relay switches on (or off) the device and GPIO board stops communicating through COM port. After reconnecting GPIO USB cable it works again one time. Is my micro bieng stung by arcs radiating? Do I need some sort of snubber?

Here is image of the current wiring: Wiring

Previously I had GPIO Board feeding from USB, but was experiencing exactly same problem. Thought maybe it's not getting enough power from USB so rewired to use external 5V supply as it is now.

I wonder maybe somebody had similar problem or has an idea what can be done to solve it?

  • \$\begingroup\$ Not sure of this is related to your problem but the relay module is spec'ed to require 15-20 ma but the GPIO module can only source 8ma. \$\endgroup\$
    – DoxyLover
    Jan 2, 2016 at 23:05
  • \$\begingroup\$ @DoxyLover: Can we have a look at your wiring layout? (Upload a decent photo or two.) There's a 'schematic insert' button on the editor toolbar if you want. Edit your question - don't post the additional information as comments. Welcome to StackExchange. \$\endgroup\$
    – Transistor
    Jan 2, 2016 at 23:59
  • \$\begingroup\$ @transistor not my question. You meant Algis. \$\endgroup\$
    – DoxyLover
    Jan 3, 2016 at 5:15
  • \$\begingroup\$ @DoxyLover: I have unit tests written for hundreds of switch on/off cycles using all 8 ports at the same and works fine. This behaviour starts appearing only when relay load is connected. \$\endgroup\$
    – Algis
    Jan 3, 2016 at 22:08
  • \$\begingroup\$ @transistor: updated question and added photo of the current wiring \$\endgroup\$
    – Algis
    Jan 3, 2016 at 22:11

1 Answer 1


It seems that it's an EMC problem rising from the surge current of your load (a motor pump). If you supply the relay board with the same 5v as the Numato board, it's possible to have some return spike on the 5v that hangs the Numato CPU when the pump start. Try to put some ceramic capacitors (one 10uF and one 100nF) next to the 5V connector of the relay board.

  • \$\begingroup\$ I had Numato board feeding from USB and relay board from 5V suplly - was experiencing same problem... \$\endgroup\$
    – Algis
    Jan 4, 2016 at 20:24
  • \$\begingroup\$ Ok, then try (with this configuration) to separate the mains wires from the logic-level ones and test and re-test (ex: with same load on differents outputs). \$\endgroup\$
    – papyDoctor
    Jan 5, 2016 at 20:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.