EMI reducing capacitors in DC motor

[neazoi]

Hi I have a DC motor spinning really close to an AM radio and I fear of it creating inrerference.

I have seen some toys use a capacitor connected across the motor contacts. Is it for interference reasons?
I have also seen each motor contact to be bypassed with one capacitor to the ground.

Which one of the two is better?
I am thinnking of using a 100nF, will that be good for suppressing interference from 150KHz to 30MHz, or do I need more?

 

[KlausST]

Hi,

DC motors have commutators inside. They create sparks. Those sparks create HF.
So a capacitor connected next to the motor terminals keeps this "HF loop" small, thus radiation is small.
It reduces differential mode HF current flow through the wires.

You talk about "toys" and "capacitor to the ground". There are several grounds. System ground, Earth ground, shielding ground.....

Toys often are not connected to Earth Ground.
And for HF radiation true EARTH ground does not play a big role, because of lengthy and high impedance wiring.

So maybe shielding ground.... Connected to the metal case of the motor, or a metal enclosure, or cable shielding..
Here the "capacitors to ground" may help to reduce "common mode" HF.

BTW: be sure the DC motor is not controlled by PWM. With PWM the capacitor may be counter productive.

Klaus

 

[neazoi]

Hi,

DC motors have commutators inside. They create sparks. Those sparks create HF.
So a capacitor connected next to the motor terminals keeps this "HF loop" small, thus radiation is small.
It reduces differential mode HF current flow through the wires.

You talk about "toys" and "capacitor to the ground". There are several grounds. System ground, Earth ground, shielding ground.....

Toys often are not connected to Earth Ground.
And for HF radiation true EARTH ground does not play a big role, because of lengthy and high impedance wiring.

So maybe shielding ground.... Connected to the metal case of the motor, or a metal enclosure, or cable shielding..
Here the "capacitors to ground" may help to reduce "common mode" HF.

BTW: be sure the DC motor is not controlled by PWM. With PWM the capacitor may be counter productive.

Klaus

No the motor is pure DC no PWM. So you think there is no need to go for 2 capacitors each shunted from the motor terminals to "ground"? A simple capacitor across the motor terminals will do?

Well the minus terminal of the motor will be connected to the minus (ground) of the circuit after all, so the capacitor across the motor terminals is in fact a single capacitor that shunts + to ground.

I am trying to find out if a single cap across the motor terminals is enough, or if two series caps across the terminals, with their center point to ground is better. But I think since the motor minus is to be connected to the circuit ground, there is no need to use the second option at all, what do you think?

I remember I have seen in the past a capacitor across a HV switch, to reduce sparking. Can a capacitor across the motor terminals help to reduce internal sparking as well and hence increase motor life?

 

[KlausST]

Hi,

I remember I have seen in the past a capacitor across a HV switch, to reduce sparking. Can a capacitor across the motor terminals help to reduce internal sparking as well?

in the case of the motor the capacitor can not avoid sparkling, but it can suppress the HF problems caused by sparkling.

It can not avoid, because the spark is generated when the commutator opens, while the current was going through the coil. Thus a high voltage is generated by the coil. If you want to suppress it you have to "capacitively close the loop of the just opening commutator", but you only have access to one commutator switch side.

Klaus

 

[neazoi]

Hi,

in the case of the motor the capacitor can not avoid sparkling, but it can suppress the HF problems caused by sparkling.

It can not avoid, because the spark is generated when the commutator opens, while the current was going through the coil. Thus a high voltage is generated by the coil. If you want to suppress it you have to "capacitively close the loop of the just opening commutator", but you only have access to one commutator switch side.

Klaus

All right, a single capacitor (100nF for example) across the motor terminals.

I remember a diode is placed across the relay coil to protect the driver transistors. Can a similar arrangement be applied in the motor case (of course there is no driving transistor in my case)

 

[KlausST]

Hi,

the diode across a relay coil is not because of continous switching noise.
It is because of the back-EMV.

Back EMV voltage on a coil is reveresed to the previously applied operating voltage.

Back EMV voltage of a DC motor is considered in same direction (not reversed) as applied voltage.
This is because of the back EMV of a motor is defined as the voltage caused by the motor working as a generator when rotating.
But still in the moment of opening a motor´s switch there will be a short negative pulse caused by stray inductance of the motor.
So if you control a motor by a semiconductor you should use a protection diode.


Klaus

 

[betwixt]

Practically, the commutation process will probably absorb most voltage spikes by itself as it will be "make before break" by design. If the motor case is 'ground' and also negative, I would suggest a 100nF ceramic capacitor wired as close as possible to the positive connection and case. Feed the positive through either a small choke or a ferrite bead with a few turns through it and add a second ceramic capacitor before that. Effectively making a pi filter with low DC resistance.

Brian.

 

[neazoi]

Practically, the commutation process will probably absorb most voltage spikes by itself as it will be "make before break" by design. If the motor case is 'ground' and also negative, I would suggest a 100nF ceramic capacitor wired as close as possible to the positive connection and case. Feed the positive through either a small choke or a ferrite bead with a few turns through it and add a second ceramic capacitor before that. Effectively making a pi filter with low DC resistance.

Brian.

I have just put a capacitor across the motor contacts now (100nF) to test it. It does produce interference up to 30-40cm away on my sensitive receiver.
The pi filter you suggest is for removing RF from the cables.
I have not tried grounding the body of the motor to see the interference level when doing so.

 

[FvM]

Capacitors between each motor terminal and the metal "can" and a third capacitor between the terminals are often seen.

 

[betwixt]

The pi filter you suggest is for removing RF from the cables.

Yes, the source of interference is entirely inside the shield formed by the motor body so interference is being radiated along the connecting wires. Adding a capacitor directly across the motor connections is helping to reduce high frequency components being radiated and if you increase the RF filtering it will be reduced even more.

Brian.