Relays - AC vs DC specs

[zappedagain]

It seems I learned this a long, long time ago but I'm drawing a blank. A quick Google search didn't help so...

Why are a relay's DC specs so much lower than the AC specs? I'm looking at a relay that can switch 7A at 250VAC (1750W) on a resistive load, but can only handle 3A at 30VDC (90W). Why the difference?

I am selecting a relay to disconnect servo motor phases in the event the E-Stop is activated. The driver is rated for 6.4A peak, 750 VA peak, and a 300mH load. Thus I don't believe the above rated relay will work, because my servo motor could well be stationary and looking like a DC load.

John D

 

[zappedagain]

I think I found it in a datasheet from Omron:

"When a relay breaks an inductive load, a fairly high counterelectromotive force (counter emf) is generated in the relay’s contact circuit. The higher the counter emf, the greater the damage to the contacts. This may result in a significant decrease in the switching capacity of DC-switching relays. This is because, unlike the AC-switching relay, the DC-switching relay does not have a zero-cross point."

So the AC relay has the smarts to switch at the low current point, whereas a DC relay doesn't have that option.

John D

 

[dpc]

As mentioned, on an ac circuit, the current goes to zero twice a cycle and the arc created by opening the relay contact is much easier to extinguish and to keep from restriking. With a dc circuit, the contact has to do all the work.

You can increase the dc breaking capability of relay contacts by putting multiple contacts (from the same relay) in series. This increases the effective arc length.

David Castor

http://www.cvoes.com

 

[ScottyUK]

Bigger DC contactors use magnetic blowouts, either in the form of permanent magnets or in the form of coils which use the power circuit current itself to generate a magnetic field. In both cases the magnetic field forces the arc up the contacts as they open, onto an arc chute where it is segmented, weakened and eventually extinguished.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[cranky108]

There is an option of using a MOV, or capacitor across the contacts. If it were a true DC circuit, you could also use a diode (in reverse connection) across the coil.
I guess you can also use the capacitor or MOV across the coil, rather than the contacts.

There are devices that use this concept, and some electronics to "extent" the contact interrupting capability.

Also, if you don't expect 10,000 operations, you can rerate the contacts for fewer operations (Don't ask, I haven't done this myself).

 

[zappedagain]

Unfortunately, my servo drive could have either polarity when I need to turn it off. So far the DC relays that I have found have polarity defined; probably for improved specs ("Wiring with incorrect polarity may result in inability to turn OFF the Relay or loss of functionality.").

E-Stop won't be used that often, so de-rating is possible. Not somewhere I really want to tread with a safety circuit though. It wouldn't be good to have my safety circuit weld itself inactive.

John D

 

[ScottyUK]

Can you interrupt further upstream where the current flow is unidirectional? I assume your servo driver is a full bridge or similar - maybe put the relay in the supply to the bridge rather than the output?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[zappedagain]

No, it is an AC powered driver, so I'd have to dig into the guts of the driver. That will be my last step, as then I'm voiding the warranty on my driver before I ship it to the customer.

The driver manufacturer is recommending disconnection of the phases and disabling of the driver; this seems redundant, so maybe there are other ways to add redundancy without having to disconnect the phases. That gets back to meeting their other safety recommendation - "Never disconnect or connect the product while the power source is energized.". Nice.

John D

 

[zappedagain]

What about a silicon solution (solid state relay)? I have to switch about 200Vpp and 10A signal. Should I start looking into TRIACs, or are there other devices you recommend?

Thanks,

John D

 

[ScottyUK]

Solid-state devices generally aren't qualified as safety devices due to their imperfect shutoff (leakage current) and habit of failing short-circuit under fault conditions. Having said that, the variable speed drive manufacturers have put a lot of effort into getting silicon-based E-stop circuits qualified for use in safety systems. I don't think it is a trivial task by any means and is likely to prove expensive unless you're making a lot of these units.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[zappedagain]

Thanks Scotty. These were all looking like expensive solutions.

I read the requirements again and realized that it calls for removing power from the output connector to the motor; that's a bit difficult unless I'm inside their box. I called their FAE and his response was "don't disconnect the phases, you'll blow the driver on your first E-stop!". He doesn't know why the recommendations got written the way they did, it seems to be some incoherent combination of the European and North American standards.

So I'm back to a simpler, common sense approach. I'll disable the driver so the motors come to a controlled stop. If I have to go one step further I'll shut off power to the driver after a delay.

Thank you all for your responses.

 

[fangas]

E-stopping a servo by shutting down the supply is (was) used to be pretty common with FANUC.

Here is FANUC's E-stop setup. In the description, the term "Magnetic" is used on Page 3/5 around line 50. My recollection is that "Magnetic" is FANUC's term for Contactor. It's been awhile seen I looked at a FANUC servo chassis, but again my recollection, is that it does have a 3-pole contactor that allows the drive to be powered up.

http://tinyurl.com/y8mb785

And a safety summary from FANUC. Note item 6

http://tinyurl.com/y8nwulb

Ed

 

[Compositepro]

Most servo motors I've worked with that have e-stops will switch-in a dynamic braking resistor on e-stop.

 

[sreid]

You can try Gigavac.

http://www.gigavac.com/products/contactors/datasheets/gx11/index.htm

 

[BobM3]

Combine your relay with a solid state device (in series) and add some delay to the coil relay circuitry so that the solid state device is the one that turns off the current.

 

[Thedroid]

I installed a few AB relays awhile back on a FAA strobe light. It wasn't an ice cube relay, and it had reversible contacts, which could be upgraded for higher capacity. I'm pretty sure they could handle your requirements. I don't remember the bulletin but i'll check.

 

[Thedroid]

Have a look at the Allen Bradley bulletin 700. They are rated up to 20A at 600VDC, and are very flexible in there configuration.