Eliminating inductive kick

[davidd31415]

I am using relays to turn 24 small DC motors (250mA each) on and off thousands of times per day in a testing application. The relays seem to fail after several months of use and I am suspecting inductive kick as the cause of the problem since there is no snubbing circuitry wired into the testing device.

What should I consider when designing a circuit to suppress the inductive kick.

I have seen solutions involving diodes in parallel with the relay as well as tank circuits. Is there a disadvantage to using only a diode (this is the way I was planning on going)? Should I choose a diode that can handle the same amount of current that the motor requires? Are there any other specifications I should look at?

Any help will be appreciated. Thanks,

David.

 

[Skogsgurra]

If you switch a lot of DC motors on/off like that, I think that starting current is getting your relays. You have 24 motors at .25 A each. That is 6 A rated current. Starting the way you do it means havung starting currents well in the 60 A range (armature resistance is very low and no counter-emf at stand still). Are your relay contacts rated for that current? No snubber or diode can help if you are overloading the contacts.

Gunnar Englund

http://www.gke.org

 

[sreid]

Go with the diode. 1N4004 diodes (1 A, 400 V) are cheap. However, the fact that your relays are failing after hundreded of thousands of operation may not be surprising.

 

[itsmoked]

Can you see the relay contacts? Clear cases? Then look at the arc size-shape-intensity and then add the 1N4004 and see if you see a difference. If you do then the life will likely improve greatly. If you don't, then skogs is probably correct in that you are over-current-ing them. This may mean you need to hunt up a different solution like some form of current limiting or different relays.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

OK. I see that you are using relays (several), not one relay. So the overload might not be too bad. But starting current is still quite high in comparison to rated current and you have to select contact rating that can handle that current.

BTW. A motor that is running does not produce much of an inductive kick when you break the circuit. It is not at all like breaking current to a coil. I do not think that a diode changes much.

How many motors do you have per contact? What relay type are you using?

Gunnar Englund

http://www.gke.org

 

[biggeorgetechnical]

Regarding the failure mode: If it is the relay contacts = Might be worth checking the relay manufacturers spec sheet. I have seen cases where manufacturer specs relay contacts at 50,000 cycles at rated load - in your case that might be attained in months. In the case I was exposed to, it took about 4 years to raeach 50,000 cycles. You might want a specification for 1,000,000 cycles - that would give you about 1-1/2 years per relay. Solid state relay with zero crossing current would eradicate that kick - relay only opens at zero-crossing current point.

 

[nbucska]

I suggest adding a series R to reduce the starting current
and using solid state switches.


Plesae read FAQ240-1032
My WEB: <

http://geocities.com/nbucska/>

 

[IRstuff]

Perhaps you need beefier relays:

http://www.keithley.com/data?asset=394

Note the 10⁸ closures rated life, but they degrade to 10⁵ closures at max signal, which is 1 A in this example.

TTFN

 

[BobM3]

Do you have a diode across the coil? That will slow the contacts when releasing and can cause burning of the contacts.

 

[2dye4]

Trash can the relays.

Replace with MOSFETS. Suitable parts can handle
the current for about $3.00

Also do put a diode across the motor so that stored
energy circulates in the diode and not produce a
kick.