arduino + relay + motor unit noise

Thread Starter

LAOADAM

Joined Nov 21, 2018
166
Use arduino + relay + motor unit got electronic interference and the unit lost control, search out that need separate power supply for Arduino and load. my question is if they still use common ground? can the common ground separate the noise?

Thanks
 

danadak

Joined Mar 10, 2018
3,701
"Normally" you spilt grounds and join them at board edge.

Clamps and snubbers on inductive loads like relay coils.

MCU pins, especially inputs, terminated in low z where possible to minimize
C coupling of noise.

Not all Cs perform equally as bypasses for same value -



Use you scope, hopefully DSO, to trigger ~ 1 diode drop outside
Vdd and Vss rails, to see if you are getting any triggers on any
pin. As a troubleshooting method. If you do get a trigger that pin
is being driven by a transient violating its allowable range of V.

Trigger test V is > Vdd + 0.7, < Vss - 0.7.


Regards, Dana.
 
Last edited:

djsfantasi

Joined Apr 11, 2010
5,838
Supply decoupling is still required to prevent the motor and relay from putting spikes on the power supply. Post a schematic.
Exactly. You’ll need a snubber or a simple reverse diode back EMF protection in your motor.

Also a capacitor across the motor terminals may help.

Since you’re using a relay, it needs to be protected in a similar manner as the motor. Another snubber or diode.

If the motor doesn’t need to reverse, you could use a MOSFET instead of a relay, which will prevent noise.

If it does need to reverse, I’d suggest an H-bridge instead of relays. You want to eliminate as many noise generating components as possible.

Your ground connections need to be in a star configuration. Otherwise, increased noise propagation will occur.

A large capacitor across your power supply should be added.

Since you’ve decided not to post a schematic, which of these suggestions cannot be evaluated. Post a schematic and you’ll get better assistance.
 

Thread Starter

LAOADAM

Joined Nov 21, 2018
166
"Normally" you spilt grounds and join them at board edge.

Clamps and snubbers on inductive loads like relay coils.

MCU pins, especially inputs, terminated in low z where possible to minimize
C coupling of noise.

Not all Cs perform equally as bypasses for same value -



Use you scope, hopefully DSO, to trigger ~ 1 diode drop outside
Vdd and Vss rails, to see if you are getting any triggers on any
pin. As a troubleshooting method. If you do get a trigger that pin
is being driven by a transient violating its allowable range of V.

Trigger test V is > Vdd + 0.7, < Vss - 0.7.


Regards, Dana.
Thank you Dana.
I'll try it.
 

Thread Starter

LAOADAM

Joined Nov 21, 2018
166
Exactly. You’ll need a snubber or a simple reverse diode back EMF protection in your motor.

Also a capacitor across the motor terminals may help.

Since you’re using a relay, it needs to be protected in a similar manner as the motor. Another snubber or diode.

If the motor doesn’t need to reverse, you could use a MOSFET instead of a relay, which will prevent noise.

If it does need to reverse, I’d suggest an H-bridge instead of relays. You want to eliminate as many noise generating components as possible.

Your ground connections need to be in a star configuration. Otherwise, increased noise propagation will occur.

A large capacitor across your power supply should be added.

Since you’ve decided not to post a schematic, which of these suggestions cannot be evaluated. Post a schematic and you’ll get better assistance.
Thank you.
The motor does need reverse, I used two relays to control the direction.
I'll add snubber or diode.
 
Top