Need Help w/ specs for RC Snubber Circuit

[ptc0007]

Hi - I have built a circuit to reverse the direction of a DC motor. The motor is 12 VDC and draws a max of 42 mA and the windings have a resistance of 17.2 Ω. My relay is a DPDT dual 5 VDC latching relay rated at 1 amp DC. The motor rotates a wheel which eventually actuates a NO limit switch. The closure of this switch allows current to flow to one of the coils in the relay which switches the contacts in the relay such that the voltage to the motor changes polatity and the motor moves the wheel in the opposite direction until it actuates another NO limit switch which allows current to flow to the other coil. This shifts the contacts back to the original position reversing the voltage polarity to the motor which causes it to reverse and move the wheel back in the other direction. I am aware that putting an RC circuit across the contacts will help prevent them from arcing due to the abrupt halt of the current flow through them. I am trying to get an idea for the values of the capacitor and resistor that would best serve this function and still be very cost effective. So far I thought that a non-polarized mica capacitor of .01 µf value rated a 150 V and a 1/2 watt 47 Ω carbon resistor in series across each contact would do the job. BTW, I know this is not an ideal situation. Turning off the current first and the switcing the contacts would be a better way to do this. However until I can get around to re-designing the circuit, I am looking for a stop gap measure to protect what I now have running. I would welcome any comments pro or con.

 

[Mr RB]

I answered this in your OTHER post.

The problem wont be fixed by a snubber. You are instantly reversing the voltage to the motor while it is at full power, causing massive current surge.

You could try putting a resistor in series with the motor provided that doesn't slow its normal running speed down too much, use the LARGEST ohms value you can.

 

[dknguyen]

A snubber would help by giving the flyback current a path to flow during the period when the relay is switching over when the contacts are open which would reduce arcing.

But no RC values can be given because you need to know the inductance of the motor and a few other things. It's much easier to go trial and error. However, I will say this. I don't think you need a resistor. THe resistor is there to limit the power dissipation, but your event only occurs at a very low frequency, not kHz like PWM so the energy dissipation is much less. So you could probably use a much larger capacitor for better damping (which also increases power dissipation) but your low frequency offsets this. A resistor might still be needed, but those are easy enough to adjust.

Or you could use a bidirectional TVS diode that is rated a little bit higher than 12V in parallel with the motor. It will breakdown and allow the flyback current to flow between motor terminals through the diode rather than by arcing across the contacts. Make sure it's the proper power rating to survive. A bit hard because of the way TVS diodes are rated, but your application is very low frequency so that should help.

 

[mneary]

The starting current of your motor is about 12/17.2 or 0.7A, and your instant-reverse current is about twice that. If you want to reduce the reversing surge to 500 mA, place 31 ohms in series with the motor (33, 1/2W will do just fine). The resistor will receive about 8W when the motor reverses, but you said this doesn't happen very often. The primary determinant of the time this pulse length is the mass of the load on the motor.

The 33 ohm resistor will reduce the running voltage to the motor by about 33*.042 = 1.3V (a little over 10%). Your speed will decrease accordingly.

 

[ptc0007]

Thanks for the feedback. Mr. RB has convinced me to work on a new circuit that turns the motor off before switching contacts. As far as my immediate problem goes, I don't see what the large resistor will do except drop the effective voltage that the motor sees which will make the problem less severe by making the circuit less efficient (the voltage drop across the resistor and the resulting heat dissipated). I'm probably missing something here. The suggestion from Knguyen has more appeal to me although I have never heard of a bidirectional TVS diode. I think I understand the concept and I will do some research on this part. It certainly would be easier to implement; just one part across the motor terminals. Thank you both for taking the time to reply. This was my second post on the same subject since my first post seemed to have run out of replies.