Relays with universal coil?

[Dago]

I'd basically need a (normally connected) relay which would work (the coil that is) with both 120VAC and 230VAC. Surprisingly I do not seem to be able to find such relays. They just need to switch like maybe half an amp or something max.

The closest thing I managed to find was this solid-state relay from Crydom ( http://www.crydom.com/en/Products/Catalog/c_l.pdf )  which supports 90-250VAC control signal but the problem with this is the "massive" minimum load current of 150 mA RMS which at 230 VAC equates to like 35W and the equipment in it's minimum configuration will draw maybe half of that.

I could rewire it to use a separate PSU (with universal input) and a relay with a low voltage DC coil but there are tens of these units so it is not something I'd like to do.

Does anyone know of any manufacturer who has relays that have this kind of universal coil?

[David_AVD]

How about a small SMPS (90V-265V input) feeding the coil of a 12V DC relay?

[Dago]

How about a small SMPS (90V-265V input) feeding the coil of a 12V DC relay?

That is my backup plan (mentioned it in my OP) but it would require more rewiring (and not sure about space) so preferably not.

[Marco]

Rectify the AC signal voltage into a small capacitor, use a make shift linear voltage regulator (resistor, zener, BJT) to drive a DC coil. Small signal non latching relays seem to go down to around 5 ma, so that would consume about a Watt when you're turning off the relay (mostly heating up the BJT).

[DanielS]

Since the current through a plain old coil connected to AC is dictated by the coil's impedance, a "universal coil" would have to be designed to activate the relay at the lowest end of its input voltage spec and be over-engineered enough to survive the higher-end of its spec. Such a universal relay would draw 5X as much power at 240V as it does at 100V.

While it should be possible to build universal AC-powered relays, there aren't many applications where one would want relays to cope with a 5:1 coil drive power range and that probably explains why such relays are rare - probably in the realm of custom-order stuff.

[Dago]

Since the current through a plain old coil connected to AC is dictated by the coil's impedance, a "universal coil" would have to be designed to activate the relay at the lowest end of its input voltage spec and be over-engineered enough to survive the higher-end of its spec. Such a universal relay would draw 5X as much power at 240V as it does at 100V.

While it should be possible to build universal AC-powered relays, there aren't many applications where one would want relays to cope with a 5:1 coil drive power range and that probably explains why such relays are rare - probably in the realm of custom-order stuff.

I think such a relay would have some kind of active circuitry inside, not just a coil. I would assume this is not a wildly special requirement though so I'm kinda surprised I'm unable to find any such relays.

[T3sl4co1l]

Going off guaranteed specs (not typical behavior), you're going to have a hard time I think.  You could make a current limiter out of a transistor (pair end-to-end, or single inside a FWB), so it works for AC, use the 120V coil, and have it limit the current just beyond what's normal.  The excess is dissipated as heat.

It would be nice to have a nonlinear component (inductor or capacitor) that can do this naturally; the coil itself might work (most likely has a laminated iron core), but it's hard to say as the inductance probably isn't well controlled (and who knows if the manufacturer decides to modify the design later).  The operational idea would be something like a ferroresonant transformer (supply the coil with a capacitor, such that the resonant frequency is near line frequency; as the core saturates, resonant frequency rises and amplitude -- and current draw -- levels off).  Pull-in might even be hampered with such an approach, because coil inductance varies with operating state (it's very low initially), so the tuning would be all off.  There may be no single approach that works real well here.

To me, this doesn't sound like a common "electrical control" problem, the kind of thing where you'd use relays -- if you could tell more about the application, maybe something better could be suggested.

Tim

[Dago]

To me, this doesn't sound like a common "electrical control" problem, the kind of thing where you'd use relays -- if you could tell more about the application, maybe something better could be suggested.

Well we have some travel cases which have some space for rack mounted equipment (everything works with universal power) that are used in both EU and US. They mainly contain fiber optic to SDI (broadcast video) converters. I think the boxes also have a plug to which you can plug whatever you need on the field (up to a few amps I guess). This relay is the main power switch for the whole thing that you can use to turn it off.

Not built by us and there are potentially tens of them to fix so the more plug and play solution there would be the better... Currently someone is checking if the power input was specified to the original manufacturer (their responsibility to fix them in that case).

[DanielS]

I think such a relay would have some kind of active circuitry inside, not just a coil. I would assume this is not a wildly special requirement though so I'm kinda surprised I'm unable to find any such relays.

How often do you run into equipment that requires universal-input AC-driven relays? Line-driven machinery (AC motors, heaters, lamps, etc.) is generally not designed for universal input - at least not without some modifications - so relays are just one more part that requires swapping between 120V and 240V models while electronics-driven machinery usually ends up driving relays with the 5-24VDC it already had to run point-of-load regulators and analog circuitry.

Another possibility might be a PTC device: the coil eats the initial higher current on 240V but the PTC warms up and picks up most of it after a few seconds so the coil does not overheat. The PTC may need time to cool down before the relay can be activated again after being turned off though.

You could also simply share the same low-voltage relay supply across multiple racks instead of giving each relay or rack its own supply. If all the racks are meant to switch on/off together, you can then simply turn the low-voltage supply on/off to control the racks' relays.

[kolbep]

What about using 110v relays, and then for the drive circuitry that gives 230v,, use a resistor potential divider to dividethe voltage in half.

[DanielS]

What about using 110v relays, and then for the drive circuitry that gives 230v,, use a resistor potential divider to dividethe voltage in half.

That boils down to the same apart from the ~10W extra wasted in the resistors.

[Alex Eisenhut]

What about dual coil relays? Is there such a thing? Wired in series on 220, parallel on 110?

[wraper]

I have seen contactors working from 100-250V AC, but no miniature relays. http://uk.farnell.com/abb/af09-30-10-13/contactor-3p-100v-250v-4kw/dp/1846141

[NiHaoMike]

Some 240V relays will pull in with 120V. Some testing will be needed to determine if there's enough margin to deal with low mains voltage.

Another idea is to just use the SSR you listed, but with a motor run cap on the output to act as a "lossless" dummy load. Use the zero cross version of the SSR and it might be a good idea to add some resistance (e.g. a NTC?) in series with the cap to avoid inrush.

[SeanB]

You can use a mains rated capacitor, a bridge rectifier and a 24 or 48VDC relay. Will use the capacitor to make a somewhat crude constant current supply and the relay will pull in on this. Capacitor has to be rated for the highest mains voltage and type approved. Value is a compromise so that it will pull in at lowest mains voltage at 50Hz (relay spec determines that current) and not exceed the max voltage at max mains at 60HZ.

This will work, but is best with a very sensitive high resistance relay. Typical in street light photocontrols with a 0.22uF capacitor as the cap and a 48V relay with a 5k coil resistance. Often they derive a 60VDC rail across the bridge rectifier that is regulated down to 15V to drive the electronics with a simple zener diode and transistor regulator. then you can have a 5V regulator to power a MCU if needs be at low power, as typically the total power draw of the electronics is 10mA maximum.

[David Hess]

It would be nice to have a nonlinear component (inductor or capacitor) that can do this naturally;

Another possibility might be a PTC device: the coil eats the initial higher current on 240V but the PTC warms up and picks up most of it after a few seconds so the coil does not overheat. The PTC may need time to cool down before the relay can be activated again after being turned off though.

The right incandescent lamp in series with the right coil might have enough negative resistance to make this work perfectly.  It would be easy enough to test.

I am sure it would be easy enough to make a simple threshold circuit using a high voltage zener or a couple of neon bulbs to switch extra reactance in series with the coil when driven with 240 volts AC.

[T3sl4co1l]

Duh!  Back in the day, there was such a thing as a barreter, usually an iron wire in hydrogen atmosphere.  Iron of course has a lovely tempco.  The DC characteristic was surprisingly constant current.

Tim

[Seekonk]

I went back to see the folks at a place I used to work for.   They started producing universal 24-265V AC DC modules.  They used a nifty little trick, a LED driver chip.    That fed a zener shunt so they had constant voltage and load.  It wasted a little power but it was universal.

[Magicmushroom666]

What about using 2 relays, one 110 and one 220 wired together?

[Dago]

What about using 2 relays, one 110 and one 220 wired together?

Wouldn't work, then with 220 the 110V would use five times more power than it was designed for and at 110 the 220V relay will star oscillating because it kinda manages to switch but doesn't hold so it goess BRRRRRRR.

We decided to swap the relays for 110V ones for some of the boxes for now and look into sending the rest of the boxes to the OEM for modifications for universal power.