Sizing zeners to control flyback voltage

I plan to control a 12VDC motor that is rated at 250 mA under no load, using an Arduino Uno. I have a Grove 2-coil latching relay rated at 3 amps that will provide the forward/reverse selection, and then a STSP relay between the Grove and the motor to control on/off. This is new territory for me, and so I am hoping for some feedback on a few points:

  1. I understand that using two zener diodes nose-to-nose (or back-to-back) in parallel with the STSP relay provides effective flyback protection while minimally degrading the relay contacts. I have had a great deal of difficulty finding a clear explanation for how to spec the diodes. I did find a post here that gave this rule-of-thumb: voltage rating roughly equal to that of the relay coil, and a current rating that is about 10% of the coil's draw. Does this sound right?

  2. Do I need to similarly isolate the motor? I am assuming the Grove unit can handle it directly and so nothing is needed for it. Is that right? If I do need to isolate the motor, would the rule-of-thumb above apply here also?

  3. I'm assuming I can use the Grove board without any additional buffering. Or should it too be isolated?

As you can see, I have only a marginal grasp on the principles at play here. Any advice would be greatly appreciated.

You can just use a plain diode to provide flyback protection . I would only use zeners on the motor if it was used to go forwards and backwards( 1-2volts above supply volts).
Having said that , if driving it forward/backwards with a H bridge , you can usually put diodes across the transistors/contacts .
0.1micro farad across the motor terminals will help too ( or two - one each from the terminal to the motor case)

Yes, the motor will be used forward/backward -- that's what the Grove 2-coil latching relay is for. I am assuming (but am not at all sure) that the Grove board has shielding for its relay. It's not clear to me why I wouldn't want to shield the other relay, or what the spec would be for the zeners.

With a 2-coil latching relay you can just use a simple diode across each winding. Only a one-coil
latching relay needs anything more complicated. The zener voltage would need to be above the
max supply voltage, not equal to it (otherwise it will conduct when the winding is activated), and
the peak current rating is the only one to worry about.

And its SPST, not STSP… Far more search matches if you get that right!

Use a bridge rectifier (or four separate diodes).
The AC terminals to the motor, and the +/- terminals to the supply.

I would think a factory made relay board would have the diodes already populated, can you post a link or Seeed part number?

MarkT -- thanks for the tips and clarification.

Outsider -- here's the relay board:

MarkT -- thanks for the tips and clarification.

Outsider -- here's the relay board:

Well its got loads of diodes and stuff, and its open source hardware by the logo so you can
find out exactly what it does.

Pity the relay itself is a tiny signal relay, 125V at 1A ??? That's definitely not mains-rated.

I've used latching relays that do 10A, 240Vac, yes they are bigger, but still pcb mount.

True, it's not heavy duty, but I'm running a motor that draws 250 mA without load. With the motor connected to the board, and (I think) the board well-protected, is there any need for diodes in parallel with the motor?

It will reduce arcing of relay contacts, prolonging their life - signal relays have small contacts so its
probably a good idea.

Thanks for the advice. Would this be an appropriate choice: