Picking a relay to control transformer

Greetings,

It has come to this - my first post. For help.

Setup. I have landscape 12V lights driven by a transformer. I used to control it with an X10 XPFM Fixture Module. But the X10 is not very reliable, and so lame now that I got into Arduino. My plan is to use my Uno to control a relay to switch on/off the line circuit to the transformer. I thought it would be fairly straightforward, but as I began researching all kinds of variables and unknowns created an impass…

What I’ve found.

  • The transformer is rated at 120 VAC, 3A on the input, and 12VAC 250W on the output. The secondary, low voltage winding is on top of the primary, line voltage one.
  • It had a light sensing circuitry, which I gutted out. It was dipped in some black hardened plastic (moisture protection?), but it had a light sensor (photo resistor?), and a relay, which (if enabled) I am quite sure (but not 100%) switched on and off the transformer. I removed some of the black plastic, and the relay is rated for 10A 250V. Looks like a generic one you’ll find on many PCBs.
  • The X10 XPFM fixture is rated for 15A 120VAC. I opened it up. It has a somewhat crude latching relay with fairly hefty contacts.

Concerns.
Publications I read about relay applications have warnings about inductive load. And the inrush current these could create that would be 10 -15 times the rated nominal input current. My transformer is 3A, so I am looking for a 45A relay? And consequently the need for snubber circuit. At this point I am way beyond my circuitry knowledge.

As of now.

  • I like the Panasonic ADW1205HLW, mostly because it’s double latching. But the numbers fit as well. What I can’t get my head around is how can those tiny terminals not burn out with the rated 16A, let alone 100A inrush current?
  • I’ve looked at SSR. Tend to be on the spendy side, and are not latching. I don’t have any solid argument, but don’t like the idea of running some current through a pin all night.

Questions:

  • Can I safely use the ADW1205HLW? I am OK driving it through transistors. But if I need to add a snubber, I don’t know what values to use for the RC.
  • If the above is no-go, what would you recommend?
  • How would I connect the 12 awg transformer wire to the puny relay terminals? Just solder them extra thick?

I would appreciate any help or comments.

References:
ADW1205HLW - https://www.mouser.com/ProductDetail/Panasonic-Industrial-Devices/ADW1205HLW?qs=sGAEpiMZZMtSzCF3XBhmW2w01ZqnWsnJNQawQzXt%2FT4%3D

How much current do your landscape lights require? Are they incandescent bulbs or LEDs? Are you switching the transformer on the primary (230V) side or secondary (12V) side?

  • Lights current: I am estimating 18A max.
  • Lights type: It's a mix of incandescent and LEDs.
  • I'll be switching the primary (120 VAC) side. That's what I was doing with the X10, and what the internal light sensing circuit was switching as well.

The maximum switching current of the relay is 16A, they do not specify, afaik, the switching of the inductive current inrush rating. They have designed the unit to perform at it's rating. The pins will be soldered to a more sunstantial pcb track, so the heating effect of 16 amps through that pin, will be conducted away.
I would use a ssr, and control latching/whatever via arduino. I've a 12V 85A toroidal transformer used as a resistance soldering power source. I switch that maybe 5 times a second (using a Pic uP- not recommended switching on and off at that speed, but it does what I wanted and the ssr is still OK. Its a pd2435, rated at 25A, but has screw terminals. I bought them years ago, much less than the price now, so I expect there is a cheaper, more modern version around.

@raymw, thank you for your answer.

I read some more about transformer loads. As I understood it, the load from the primary coil (the one I am switching) mimics the load on the secondary. In my case the lights are a resistive load, so to the relay the load will look like resistive one as well. The worst case is switching off the primary coil when there’s no load on the secondary. That would be pure inductive load on the relay with all the bad consequences, and methods to remedy them. So no testing on the work bench with lights disconnected! The outcome for me was that I stopped worrying about a snubber circuit, and didn’t add one.

I used a T9AS1D22-5 relay, which has nice solid terminals. It has been working just fine for about three weeks now. Hopefully it will last.

:slight_smile:

The transformer has a winding it is an inductive load, no doubt about it. I assume you have experience and are comfortable working with mains voltage. If not STOP! A 10 amp relay would be just fine and much easier to work with. Since you are using a relay a snubber is not needed. The relay you picked should work without any problems. Why are you using 12 AWG wire, at 3 amps 20 AGW would be ok. Give it a try it should work just fine. If this goes outside use appropriate packaging and grounding. The simplest way of driving a relay is use an Avalanche rated N-Channel logic level FET with enough current capability to operate the relay coil. Hook up is simple: Connect one side of the relay to the + supply connect the other side of the relay to the drain of the mosfet, connect the source of the mosfet to ground and the gate either directly to the micro pin or put a resistor in series with the gate preferably less then 50 ohms. When they tell you that you have to use a flyback diode or whatever ignore it, the fet is designed to operate without one. The flyback when the fet is in the circuit will be give or take a little about -2 Volts. You can find the UIS (Unclamped Inductive Spike) capability in the data sheet, and if it more then the solenoid load it is ok. Check this data sheet: https://www.vishay.com/docs/91017/91017.pdf Remember when an inductor is switched off the polarity reverses and the voltage will rise until quenched.
Good Luck & Have Fun!
Gil

A transformer with no secondary load will act as a large inductor, but with a heavy load on the secondary will reflect the load impedance (scaled by the square of the turns ratio) and may not look very inductive at all.