Relay contact arc open vs close

[Jester]

I'm looking at a circuit where a relay contact makes and breaks up to 25A. (the current can be either ac or dc, and can be up to 250V, however when switching 25A, the voltage is much lower,  10Vmax)  Relay-->:  https://b2b-api.panasonic.eu/file_stream/pids/fileversion/293
This relay does not have a dc rating, however a 5A FET based SSR is connected in parallel with the contacts and driven by a XOR gate that turns the SSR on briefly whenever the relay opens or closes. The SSR has a surge rating of 25A, however only for 10ms.

Obviously breaking DC can be really harsh on relay contacts, and assuming the SSR timing is correct should allow the relay to survive switching DC. That being said I don't see the need for the SSR for contact closes, am I missing something?

[drvtech]

Relay contacts bounce when they make so you can get arcs on both closing and opening. Shorting the contacts with a semiconductor while they are settling will prevent those arcs.

[Berni]

Yep getting a relay to switch >200V at 25A will definitely not be handled very well by any general-purpose relay.

Large high voltage relays use this very trick to combat acring. They have one set of big bulky high power contacts that carry the actual current once switched on, while having a second parallel set of contacts that is designed to move very quickly and far in order to break any arcs that form. So when the relay is opening first the large contacts opens, transfering load to the small contacts that then quickly spring open shortly after.

When the voltages are low enough the 2nd arc resilient contact can be replaced with a semiconductor. But if you are to have AC across the contacts this usually means you will need two transistors in series in order to interrupt each of the directions. Isolated gate drive is also required to turn them on and off while floating, but a simple gate drive transformer should work well for this.

But do also give some thought into what happens if this system fails. With the sort of voltage and current at DC the failiure of the arc supression system could quickly ignite an arc on the relay upon next turn off and keep the arc going as the relay is open, eventually burning through the relay housing, the fire helping sustain the arc even more, melting the relay itself and drawing the arc onto something near the relay continuing the dangerous angry hissing DC spark fire until the entire device is burning.

[bdunham7]

I'm looking at a circuit where a relay contact makes and breaks up to 25A. (the current can be either ac or dc, and can be up to 250V, however when switching 25A, the voltage is much lower,  10Vmax)  Relay-->:  https://b2b-api.panasonic.eu/file_stream/pids/fileversion/293
This relay does not have a dc rating, however a 5A FET based SSR is connected in parallel with the contacts and driven by a XOR gate that turns the SSR on briefly whenever the relay opens or closes. The SSR has a surge rating of 25A, however only for 10ms.

Obviously breaking DC can be really harsh on relay contacts, and assuming the SSR timing is correct should allow the relay to survive switching DC. That being said I don't see the need for the SSR for contact closes, am I missing something?

The relay does appear to have a DC rating of 22A @ 30 volts.  I would hope and expect that the relay would nominally meet the requirements for switching whatever power is in the circuit and that the SSR snubber is there mainly to increase the lifespan of the relay.  It would seem unacceptable to have a design where the SSR makes up for a relay that is wholly inadequate, for the reasons that Berni mentioned. 

When the relay closes, it 'wipes' the contacts a bit, a sideways movement as it finishes closing that renews the surface a bit, but wearing it at the same time.  Having significant current going through the contacts during wiping causes a bit of arcing and greatly increases the amount of that wear.  The snubber will keep that current fairly low (but not zero) during that wiping period.

[CaptDon]

If switching DC and if the load is inductive the spike voltage could cause your SSR to fail shorted. MOV's or Transorbs may be required to protect your SSR also.

[Jester]

I'm looking at a circuit where a relay contact makes and breaks up to 25A. (the current can be either ac or dc, and can be up to 250V, however when switching 25A, the voltage is much lower,  10Vmax)  Relay-->:  https://b2b-api.panasonic.eu/file_stream/pids/fileversion/293
This relay does not have a dc rating, however a 5A FET based SSR is connected in parallel with the contacts and driven by a XOR gate that turns the SSR on briefly whenever the relay opens or closes. The SSR has a surge rating of 25A, however only for 10ms.

Obviously breaking DC can be really harsh on relay contacts, and assuming the SSR timing is correct should allow the relay to survive switching DC. That being said I don't see the need for the SSR for contact closes, am I missing something?

The relay does appear to have a DC rating of 22A @ 30 volts.  I would hope and expect that the relay would nominally meet the requirements for switching whatever power is in the circuit and that the SSR snubber is there mainly to increase the lifespan of the relay.  It would seem unacceptable to have a design where the SSR makes up for a relay that is wholly inadequate, for the reasons that Berni mentioned. 

When the relay closes, it 'wipes' the contacts a bit, a sideways movement as it finishes closing that renews the surface a bit, but wearing it at the same time.  Having significant current going through the contacts during wiping causes a bit of arcing and greatly increases the amount of that wear.  The snubber will keep that current fairly low (but not zero) during that wiping period.

Thanks, I missed that DC rating hiding in the safety section, I only looked in the contact rating in the 2. Specifications section. The actual relay is actually the high capacity type so 33A @ 30Vdc.  So it looks to have some margin even without the SSR circuit.

[Terry Bites]

AC and DC behave very differently in mechanical disconnect. AC current drops to zero at double the line frequency allow the arc to self extinguish. DC is steady state and just loves to make an arc lamp from your contacts. Use a Solid State Switch for anything over a few amps and avoid the head aches.  Mosfets  in a well designed circuit are far more reliable than a relay.

DC relays cost a bomb! Check out the G9EA-1 from Omron a snip at $150 (US). It uses smart magnetics to reshape the arc. Kerchunk!

Great test of a 12kA DC Breaker https://www.scibreak.com/?p=95121