Help Me Choose A Relay

[Les]

In another thread I am discussing making some modifications to a VL500 80V. As part of this project, I will need a relay with a 120v coil voltage and the ability to switch a 1,000w lamp. I know this is getting in to relay vs contactor territory, but I'm positive that there has to be a small package out there that will work for me.

I was looking at this relay, but the concerning thing to me is that on the datasheet under "Ratings", it lists a max 840w resistive load. However, current is listed at a maximum of 30A. This is confusing to me.

Can someone who knows more about relays help me out? Also, the one I have chosen is an SPST. This seems like the logical choice for switching on and off one circuit, right?

 

[porkchop]

The 840W rating is for DC. 8310 VA rating is for AC. Looks to me like the relay you've selected would work. Nice and compact and no base to rattle loose out of.

FWIW this is the relay Clay Paky used to use in Stage Scans for pretty much this exact purpose. You probably don't need this much, but just for compairison sake:
http://www.jsfrance.com/en/relais/4...-pour-stage-scan-zoom-neuf-2100000025770.html

 

[Les]

This clears it right up. Thanks! I bought one to test with and will order the other 4 later. The dimensions are at the upper limit of fitting in the space, but I think I can make it work. Now I just need to figure out that snubber circuit thing. Apparently it's one of these. So many choices!

 

[JD]

Are you looking for a snubber on the coil or the contacts? A resistive load would not require a snubber but an induction load like a ballast can be harder on relays without one.
Depending on how you are driving the coil, a different snubber can be helpful. If your relay coil is driven by an AC source, a small cap can help but is probably not needed. If it is driven by a DC source, throwing a reverse polarity diode across the coil (Like a 1N4007) helps protect whatever is driving the coil. Note, the diode should be installed across the coil in the opposite direction of normal current flow so that is does not conduct. When the coil powers down, it generates a reverse polarity flyback voltage that will be snubbed by the diode. This flyback spike is what can pop the transistor that may be driving the relay.

 

[Les]

Are you looking for a snubber on the coil or the contacts? A resistive load would not require a snubber but an induction load like a ballast can be harder on relays without one.
Depending on how you are driving the coil, a different snubber can be helpful. If your relay coil is driven by an AC source, a small cap can help but is probably not needed. If it is driven by a DC source, throwing a reverse polarity diode across the coil (Like a 1N4007) helps protect whatever is driving the coil. Note, the diode should be installed across the coil in the opposite direction of normal current flow so that is does not conduct. When the coil powers down, it generates a reverse polarity flyback voltage that will be snubbed by the diode. This flyback spike is what can pop the transistor that may be driving the relay.

Thanks, @JD ! I watched a YouTube video on snubber networks last night and while I didn't understand a lot of it, I recognize what you said from that video - especially the portion about flyback voltages. The more I learned, the less I was sure about exactly where I needed to put the snubber — or whether or not I needed multiples.

This relay has a 120v coil which will be getting its power directly from the mains input of the fixture (there is a terminal block where it splits between that and the low voltage supply powering the MCB, motors, etc). The input side of the relay will be powered from an external dimmer and the output will be to a 1,000w halogen lamp (Philips EGT). So, since I'm not dealing with motor or ballast loads, it sounds like a snubber won't be necessary? They're pretty cheap so I'm not opposed to adding them for peace of mind, but if it's totally unnecessary, that's one less part to order!

Also, in most cases the relay will not be switched under load since the load will be a lamp on a dimmer. Of course dimmers do have some leakage current so there will probably be some voltage present, but ideally, the fixture would already be powered up before bringing the dimmer channels up. It would also (ideally) only be switched once per use of the fixture/once a day, so it wouldn't be like some automation operations where it has to perform several actuations in a short amount of time.

 

[JD]

Doesn't sound like you need a snubber in either location. Mains/AC 120 volt for the coil eliminates the need there, Halogen lamp load doesn't need one either, so if I was doing this circuit, I would be comfortable without using one.

 

[Les]

Excellent. In that case, I won't let it stress me out too much.

Here's a question on ferrite cores. In the original design, the leads exiting the dimmer and heading for the lamp included two ferrite cores situated between the dimmer PCB and the inner arm of the fixture. Should I try replicating this? I know it has to do with keeping the cable from acting like an antenna for high frequency noise, but I have assumed that it was more due to the presence of the dimmer. Since the dimmer will be inactive, I figure I can just do straight leads and not worry about the ferrites.

 

[FMEng]

No need for ferrites if there is no dimmer producing RF noise.

 

[JD]

You indicated there was an external dimmer being used. I'm sure that dimmer contains it's own filter so they should not be needed. In the original design, those ferrites may have been the only filtering of the dimmer circuit and needed to suppress RF noise. If you still have them and stick them in there no harm will be done. I doubt you will have any problems if you leave them out.

 

[Les]

Thanks guys, this has been very helpful. I had a feeling about the internal dimmer. After the modification, the lamps will be powered by a dimmer rack, so no worries there. I would leave the ferrites in, but I guess some of my fixtures are older versions because only 3/5 have had them. I checked on VL's site and sure enough there was a Rev 2 that added ferrite cores. I'll just keep it simple.