I've finally come up with what looks to be a manageable design for this circuit where an Arduino is controlling 9 solenoid valves through solid state relays. I've attached both the schematic and photo for the setup for what it would look like for two solenoid valves. The setup will be exactly the same for the additional 7 valves.

My question is whether or not I'll have any issues with this grounding setup. Is there any problem putting the ground from the SSRs to the 24V power supply (positive input is from digital pins on Arduino).

For a little more info: in the photo those two wire molex plugs will have the lines to the solenoid valves hence the blue "S". The black "+" is the positive input from the 24V power supply, the black "-" is the ground for the solenoid valves to the power supply, and the red "-" is the ground for the SSRs also to the same power supply.

enter image description here

enter image description here

  • \$\begingroup\$ Can you revise your schematic to show where the MCU's 5V/3V3 supplies and GNDs are connected to. \$\endgroup\$
    – TonyM
    Commented Apr 13, 2017 at 16:58
  • \$\begingroup\$ YOur fly-back diodes as backwards \$\endgroup\$
    – Trevor_G
    Commented Apr 13, 2017 at 18:16
  • 1
    \$\begingroup\$ @Trevor darn I just realized that, in my real setup they're correctly oriented. \$\endgroup\$
    – clifgray
    Commented Apr 13, 2017 at 18:22

1 Answer 1


The resistors in series with the LEDs in the solid state relays must be connected to the ground of the processor.

Think about it. The current to drive a LED goes out a processor digital pin, thru the LED, thru the resistor, and has to get back to the processor ground somehow.

The SSRs give you isolation, so there is no need for the 24 V power supply to be connected to the processor power or ground in any way. While a single connection, like tying the grounds together, is possible, there is no need for that. To reduce noise from the solenoids getting back to the low voltage circuitry, I would keep the two power regimes isolated.

  • \$\begingroup\$ Thanks for the answer! So I understand the last paragraph, I suppose that limiting noise is a good enough reason not to ground my low voltage circuit on the 24V power supply, but I don't understand the first two. Why would the current driving the LED in the SSR need to get back to the processor's ground as opposed to a different ground other than limiting noise? \$\endgroup\$
    – clifgray
    Commented Apr 13, 2017 at 18:08
  • \$\begingroup\$ @clifgray 3 reasons. 1. Isolation, 2. you are generating a ground loop when you do not need to, and 2. If there is a voltage in the ground line on the solenoid side due to large amps the SSRs may behave erratically due a different ground level on the micro side. \$\endgroup\$
    – Trevor_G
    Commented Apr 13, 2017 at 18:21
  • \$\begingroup\$ Okay that all makes sense, the reason I had initially planned to do it this way is that one of my lab mates had suggested that the Arduino Gnd might have some funky behavior with all 9 lines into it. Is there any reason in your mind not to just use the Arduino Gnd? \$\endgroup\$
    – clifgray
    Commented Apr 13, 2017 at 18:42
  • \$\begingroup\$ @clif: The current going out the power supply, into the processor, out a pin, thru a LED, and thru the resistor has to get back to the negative input of the power supply somehow. \$\endgroup\$ Commented Apr 13, 2017 at 18:42

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.