Here is the schematic for the motorView attachment 234137
And here is the schematic for the arduino power
View attachment 234138
The high side diode is to prevent the motors from stealing charge from the arduino, and the low side diode is to prevent the motors from moving the ground plane. I've tried without the low side diode, and there's still an issue with the restarts.

I'm running out of theories for what could explain this issue. It's very temperamental, sometimes it's quite stable and sometimes it immediately resets. I'm thinking that there may be some interference issue, possibly from the relays. But I'm not sure how to test that theory, and how I would fix the issue.

I have problems like this with arduino mega use ethernet shield and 8 channel 5v optocoupler relay, sometimes arduino is restarted with mysterious reason I don't know..., if I try in my lab all fine, when I move it to terrace near water pump, everything doesn't work properly, but now without change everything was works for over 7x24 hours, with same ethernet wire, same power supply, same code, only I place it in top of brick and near AC 220v wire also near water pump, maybe something interfere my arduino, and I was scare to change my arduino code, risking my arduino not work properly again....

Moderator edit: New thread created from this.

 

I have problems like this with arduino mega use ethernet shield and 8 channel 5v optocoupler relay, sometimes arduino is restarted with mysterious reason I don't know..., if I try in my lab all fine, when I move it to terrace near water pump, everything doesn't work properly, but now without change everything was works for over 7x24 hours, with same ethernet wire, same power supply, same code, only I place it in top of brick and near AC 220v wire also near water pump, maybe something interfere my arduino, and I was scare to change my arduino code, risking my arduino not work properly again....

In my case arduino isn't restarted but ethernet shield and also LCD doesn't work properly... but anything excepts that is totally fine, really mysterious problem...

 

For the Arduino power supply, try using a 100 nF ceramic capacitor in parallel with the C1 and C2 electrolytic capacitors. The explanation is that different types of capacitors behave differently at high frequencies. The electrolytic capacitors, having metallic electrodes, are said to exhibit inductive effects at those frequencies.

 

Once I used an Intel microcontroller to control a film recorder. The film was in a Polaroid film pack and the film was ejected from the pack, going through rollers to spread the developer on the way out. Every Nth photo the controller would go nuts and keep the motor running, spewing the Polaroid film all over the floor.

I came to find that noise from the motor brushes was working its way back up the cable, through the drive circuit, and into the microcontroller and occasionally making it lock up. If your motor has brushes, use every trick you know to keep the noise from the motor from getting back to the controller. Once that's done you can try removing some of the fixes. You are not looking at simple control pulses, if there is arcing going on in the motor you are looking a broadband noise -RF if you like.

Or the problem might be something simpler and easier to fix. Let's hope so.

 

Once I used an Intel microcontroller to control a film recorder. The film was in a Polaroid film pack and the film was ejected from the pack, going through rollers to spread the developer on the way out. Every Nth photo the controller would go nuts and keep the motor running, spewing the Polaroid film all over the floor.

I came to find that noise from the motor brushes was working its way back up the cable, through the drive circuit, and into the microcontroller and occasionally making it lock up. If your motor has brushes, use every trick you know to keep the noise from the motor from getting back to the controller. Once that's done you can try removing some of the fixes. You are not looking at simple control pulses, if there is arcing going on in the motor you are looking a broadband noise -RF if you like.

Or the problem might be something simpler and easier to fix. Let's hope so.

and that pump motor was destroy two of my fotek SSR relay, firstly I think SSR is a problem, I used new relay then they still broken after use to turn on and off pump about five times, but I think there is not enough noise, because water pump is only about 0.3kW, 220v
Do you think it's because water pump is ungrounded, so make damage to my SSR ?
*Solid State Relay/SSR

 

and that pump motor was destroy two of my fotek SSR relay, firstly I think SSR is a problem, I used new relay then they still broken after use to turn on and off pump about five times, but I think there is not enough noise, because water pump is only about 0.3kW, 220v
Do you think it's because water pump is ungrounded, so make damage to my SSR ?
*Solid State Relay/SSR

Water (fluid) pumps are the worst for long duration locked rotor start currents due to incompressible fluids. IMO you really need some sort of soft-start SSR with make/break AC snubbing.
https://www.omron-ap.com/service_support/FAQ/FAQ02035/index.asp
https://www.mouser.com/datasheet/2/357/Altistart_01_ATS01N103FT-1786423.pdf
This is why I still like old school contactors.

 

Water (fluid) pumps are the worst for long duration locked rotor start currents due to incompressible fluids. IMO you really need some sort of soft-start SSR with make/break AC snubbing.
https://www.omron-ap.com/service_support/FAQ/FAQ02035/index.asp
https://www.mouser.com/datasheet/2/357/Altistart_01_ATS01N103FT-1786423.pdf
This is why I still like old school contactors.

Nice idea, but I was used 5.5kW contactor, and it's contactor coil is controlled by arduino 5v relay, it's seems overkill for 0.3kW pump with 5.5kW contactor, but it's works very nice, and I don't worry for transient/surge current come to my arduino when pump stopped