-
How can I convert a 42VAC relay to 24 or 12VDC
Posted by
dbendeke
on 26 Jan, 2022 06:49
-
I have a welder with a gasrelay that has failed. It uses a 42VAC relay which is very hard to find. I do have a selection of DC relays I would like to use.
I have 12VDC and 24VDC relays but are having trouble change the control voltage from the original 42VAC to ex. 12VDC. I have tried to measure current supplied to these relays and it is fairly high - appr. 1A . Stepdown modules usually have a limit of 30V and can't handle the conversion and there must be other solutions than using transformers etc. as space is limited.
I have a basic understanding of electronics but a solution eludes me. So I am hoping someone here can point me in the right direction. It would be nice to have the welder back in action
-
#1 Reply
Posted by
Psi
on 26 Jan, 2022 07:24
-
if it's 42V AC you could half-wave rectify using one diode and one capacitor (it will have to be a large cap, like 4000uF, or multiple in parallel)
That will give you ~18-21V average under load, that should be fine for a 24V DC relay.
Or, if you put even more filtering capacitors on it you can drive a DCDC module to step it down to 12V.
You might want to check it actually is 42V AC with 50% duty cycle at 50/60hz, make sure its not some oddball AC waveform or frequency
Have a look on YouTube or Wikipedia about half-wave rectification if you're unsure. There are lots of videos on it.
-
#2 Reply
Posted by
Ian.M
on 26 Jan, 2022 08:03
-
Nominal 42V RMS could go over 46V during high line conditions and due to the transformer's regulation factor, even higher if lightly loaded. You therefore need a solution that can cope with 70V or even more after rectification. One approach would be to PWM the relay coil with a low side MOSFET, adjusting the duty cycle to average 24V across the coil. The PWM controller could be as simple as a NE555 running from a dropper resistor + shunt Zener supply, set to give a fixed duty cycle, or it could be much cleverer, measuring the rectified but unsmoothed 42V and adjusting the duty cycle continuously to maintain 24V average across the coil, removing the need for a large reservoir capacitor.
However the easiest way out of the hole you are in would be to see if you can get a solid state relay beefy enough to do the job, controlled by a dropper resistor from the 42V RMS signal or if not, use a small SSR to control coil power to a relay with a mains voltage AC coil.
I'm not too fond of Psi's ideas - half wave rectifying 42V RMS gives you a waveform that peaks at close to 60V, + a lousy power factor. In fact the RMS secondary current will be over three times the DC current, so you'd better be sure the 42V secondary is good for it. The 60V peak is more problematic. to get the average down to 24V +15% to keep a 24V relay happy, its essentially going to have to drop to zero between half cycles, so unless it has a slugged coil, that relay will be buzzing like an angry wasp, hammering itself to death. Also 4000uF is far too much - with 1A load current that's only going to give about 5V pk-pk of ripple, when fed from a half-wave rectified 50Hz supply. Note that the ripple current in the capacitor will be very high - comparable to the load current.
>70V rated buck converters are available, but certainly wont be cheap as a pre-built module. Sourcing the OEM relay may well be cheaper . . .
Edit: corrected a couple of numbers for 50Hz supply
-
#3 Reply
Posted by
Psi
on 26 Jan, 2022 08:06
-
I thought of that, but the relay it's always on when the 42VAC relay power is present.
So the rectifier output is always loaded down.
And a relay is more concerned about average current/voltage. So some ripple and overvoltage is ok.
-
#4 Reply
Posted by
Ian.M
on 26 Jan, 2022 08:17
-
Yes, up to 20% overvoltage *may* be acceptable, and as its only fed 42V AC when the relay is on, we can *probably* ignore the transformer's (unknown) regulation factor. However even if the relay can tolerate it, an average of 30V at the relay coil is problematic due to the I RMS and relay chatter issues I mentioned.
-
#5 Reply
Posted by
Psi
on 26 Jan, 2022 08:21
-
-
#6 Reply
Posted by
Zero999
on 26 Jan, 2022 08:52
-
Use an ordinary 24VAC (not DC) relay and a series resistor.
R = RCOIL*(42-24)/24
So if the relay has a coil resistance of 500Ohm
R = 500*(42-24)/24 = 500*18/24 = 500*0.75 = 375R so go with a 390R resistor, the nearest standard value.
The resistor needs to have a high enough power rating. P = V2/R.
P = (42-24)2/390 = 182/390 = 0.83W so use a 1W, or higher power, resistor.
-
#7 Reply
Posted by
Ian.M
on 26 Jan, 2022 08:55
-
The second DC-DC converter you linked looks reasonable + a bridge rectifier and enough high ripple current reservoir capacitance to keep it running between half-cycles. The only thing I don't like is its derating with temperature, as its 50 deg C limit for full power could be problematic inside a welder. However the O.P's in Denmark, not the southern USA, so its probably manageable if it can be located near an air intake.
-
#8 Reply
Posted by
Ian.M
on 26 Jan, 2022 09:04
-
I agree the dropper resistor idea is also viable, provided the 42V supply has enough spare current capability, as I would expect the the coil current of the original 42V relay to be around 60% lower than that of a comparable 24V one.
-
#9 Reply
Posted by
Kleinstein
on 26 Jan, 2022 09:07
-
One could use a transformer to reduce the voltage, though it could be a bit hard to find a suitbale one of the shelf. With some reduction in efficiency and larger size one could use somerhing like a 110 to 60 V (2x30V) to get about 24 V AC. With some filtering and series resistance one can adjust the votlage to the DC relay both a little up an down. Alternatively one could use only the secondary side, like a 2x24 V winding to half the 42 V without isolation. 21 V AC and a little filter cap behind the rectifier could be OK for a 24 V DC relay.
When looking for a SMPS type converter, one may have to look at converters more made mains use. So more like at parts made to work from 110 V AC and than adjust the circuit towards lower voltage.
-
#10 Reply
Posted by
Zero999
on 26 Jan, 2022 09:33
-
I agree the dropper resistor idea is also viable, provided the 42V supply has enough spare current capability, as I would expect the the coil current of the original 42V relay to be around 60% lower than that of a comparable 24V one.
Another possibility is a 24VDC relay, a bridge rectifier and dropper capacitor. The dropper capacitor doesn't have to continuously pass the relay's full current rating, only a couple of times the holding current. There will be a nice big pulse, when the power is first applied, to turn the relay on, but from there after, the current can fall to a sufficient level to keep the relay on, through the zero crossing point of the AC cycle.
Unfortunately this will require some experimentation, because the inductance, when the relay is open vs closed is often not specified.
Don't try a capacitive dropper with an AC relay, because it can result in resonance with the inductance and a much higher current through the relay.
-
#11 Reply
Posted by
SmallCog
on 26 Jan, 2022 21:16
-
48VAC relays are reasonably common
If it was my welder I'd buy one and see if it works at your 42VAC
I'd be surprised if it doesn't
-
#12 Reply
Posted by
langwadt
on 26 Jan, 2022 22:57
-
I have a welder with a gasrelay that has failed. It uses a 42VAC relay which is very hard to find. I do have a selection of DC relays I would like to use.
I have 12VDC and 24VDC relays but are having trouble change the control voltage from the original 42VAC to ex. 12VDC. I have tried to measure current supplied to these relays and it is fairly high - appr. 1A . Stepdown modules usually have a limit of 30V and can't handle the conversion and there must be other solutions than using transformers etc. as space is limited.
I have a basic understanding of electronics but a solution eludes me. So I am hoping someone here can point me in the right direction. It would be nice to have the welder back in action
what kind of relay needs 10s of watt? what does the relay switch? can you use a solid state relay instead?
-
#13 Reply
Posted by
BMK
on 26 Jan, 2022 23:58
-
Just get 2 (identical) 24V relays and put the coils in series.
-
#14 Reply
Posted by
BMK
on 27 Jan, 2022 01:31
-
Hang on... By gasrelay do you mean:
-Solenoid valve -uses electricity to open gas flow? Or..
-Relay that makes and breaks an electrical circuit?
-
#15 Reply
Posted by
SmallCog
on 27 Jan, 2022 01:55
-
Hang on... By gasrelay do you mean:
-Solenoid valve -uses electricity to open gas flow? Or..
-Relay that makes and breaks an electrical circuit?
I read it as a relay that controls the gas
If it's a solenoid, choose an available one and slave it with a common 48VAC relay
-
#16 Reply
Posted by
m3vuv
on 27 Jan, 2022 05:59
-
can you not run the 48v ac thru a load of diodes to feed an lm7824 voltage reg and use that?,or does the relays you have need ac?
-
#17 Reply
Posted by
Ian.M
on 27 Jan, 2022 06:29
-
can you not run the 48v ac thru a load of diodes to feed an lm7824 voltage reg and use that?,or does the relays you have need ac?
That combines the disadvantages of simple rectification and the dropper resistor to increase the RMS current required from the 42V supply past that required by either alone. If the 42V supply powers other systems so the 'relay' is a minor part of its normal load, it could work but would be less reliable and less efficient than a simple dropper resistor. However if BMK is correct, and the solenoid is the main or sole load on the 42V winding, you risk a burnt-out secondary.
-
#18 Reply
Posted by
m3vuv
on 27 Jan, 2022 06:36
-
i wonder if its actualy anything like 42v when its actualy welding,i guess its maybe 50% of that when its actualy welding,maybe good to try welding,or at least get an arc going and measure the voltage.maybe getaway with a zenner clamp?
-
#19 Reply
Posted by
Ian.M
on 27 Jan, 2022 08:02
-
Just as an idle exercise I cooked up this 555 based PWM sim in LTspice. It reduces the on time in proportion to the unsmoothed full-wave rectified supply voltage to reduce the effective voltage variation at the solenoid or relay coil. The RC network on the 555 Reset pin is quite critical as to avoid blowing the MOSFET, the 555 mustn't start the PWM till there is enough Vcc available to drive the MOSFET hard on. If there isn't enough coil inductance to carry it through the zero crossings, it may need just enough reservoir capacitance added to the bridge output to keep it above 12V at the ripple trough. N.B. the ripple current will be high - make sure the reservoir capacitor is adequately rated.
-
#20 Reply
Posted by
MarkF
on 27 Jan, 2022 08:43
-
Try another search for 48VAC relays. Paying attention to the pull-in/operate voltage. Most relays will pull-in far below their rated coil voltage.
Example:
-
#21 Reply
Posted by
CaptDon
on 27 Jan, 2022 18:14
-
The coils of your substitute relays shouldn't draw anywhere near 1 amp of coil current!!! Even motor contactor relays don't draw that much, the coils would burn up!!! Also, B.T.W. Many 24vdc relays will actually work fine on 40-50vac. The inductive reactance of the coil limits the current when run on A.C. I have had good luck running the little P&B 'Icecubes' in ways they were never intended. When you are losing $10k per hour in production losses you'll really go that extra mile of "Lets see if this will work"!!!
-
#22 Reply
Posted by
Zero999
on 27 Jan, 2022 21:45
-
Many 24vdc relays will actually work fine on 40-50vac. The inductive reactance of the coil limits the current when run on A.C.
A DC relay won't reliably work on AC, because it doesn't have a shading coil, which continues to provide magnetic flux, through the zero crossing point. Perhaps you might have more luck with a large relay/contactor, which has more inertia, but I wouldn't count on it.
https://en.wikipedia.org/wiki/Shading_coilOddly enough, the picture of the relay in the Wikipedia article is 42VAC.
-
-
A 120V-24V transformer should provide about 8 VAC, or very nearly 12VDC rectified and filtered if driven by 42VAC.
-
#24 Reply
Posted by
BMK
on 28 Jan, 2022 14:01
-
OP never came back. I think this thread is on a tangent about relays after misunderstood (or poorly stated) question. Maybe he was scared away by the enthusiasm!
99% sure It's not a relay, but a solenoid valve he's talking about. Gas valves with 42V AC coils are common thing in MIG welders. They are usually direct acting valves and are somewhere in the 5VA to 20VA range..
-
#25 Reply
Posted by
elekorsi
on 28 Jan, 2022 20:43
-
I am sure that op has a solenoid (valve coil) and as CaptDon said, 24VDC solenoid will work on 48VAC. In fact many solenoid coils are rated for 48VAC and 24VDC. I am 98% sure that it will also work on 42V
-
#26 Reply
Posted by
Zero999
on 28 Jan, 2022 21:06
-
I am sure that op has a solenoid (valve coil) and as CaptDon said, 24VDC solenoid will work on 48VAC. In fact many solenoid coils are rated for 48VAC and 24VDC. I am 98% sure that it will also work on 42V
No, there's no guarantee a DC solenoid will work on AC. If you power a solenoid designed for DC, from AC, it will vibrate. It's true an AC solenoid will work on DC, but at a lower voltage. The difference is a solenoid designed to work on AC will have a shading coil, where as one designed for DC alone won't.