Is it possible to operate a 12 VDC relay from a 24 VAC supply?

(snip)

For the same coil power, the 24 volt relay will need about

1/4 of the copper to operate over the distance. AC relays are not quite as efficient as DC relays are and have an inductive current component added to that which powers the coil, so the actual improvement may be closer to 3 to 1 or even 2 to 1.

But 100 feet is not so far for a couple watt 12 volt coil. How fine a wire do you want to use to drive the coil? Measure the coil resistance and use a wire table to figure out how many ohms you can afford in series with that (maybe

1/20th of the coil resistance for 200 feet of wire.

For instance if your have a relay that has a 2 watt coil (many are lower than this) that would be about a 288 ohm coil. So you would loose only about 5% of the drive voltage if the wire loop added another 288/20=14.4 ohms. You could use higher resistance wire if you had a bit more than 12 volts to drive the loop.

Taking a look at a wire table,

the size that has less then 14.4 ohms for 200 feet would be AWG 28, which is really fine. Something like cheap 22 AWG speaker wire would have a resistance of only about 3.2 ohms for the 200 feet, and that would drop only

12*3.2/(288+3.2)=0.13 volts or about 1% of the 12 volts.

You really do not have much of a problem to solve.

--
Regards,

John Popelish

On Fri, 15 Feb 2008 20:19:36 -0800 (PST), HC wrote:

:Hey, all, I'm not sure if this can be done but here's what I am trying :to do and how I've tried to go about doing it. : :I want to be able to control 120VAC devices some distance away from a :controller, say, up to 100 feet or so. I would like to run the 120VAC :to the device through a switch (relay) at the device with no other :switches or control devices in line from the breaker. Then I would :like to have the controller turn that relay on and off to control the :device. The idea is that I could run the thicker, high-voltage lines :directly to the device and then use smaller wire to operate a relay at :the device to turn it on and off instead of running the high-voltage :wire to each switch I would like to use. : :Since I have a boat-load of low-cost 12VDC relays that can switch up :to 250VAC and 15 amps and since 24VAC relays seem, in my searching, to :be a lot more expensive and harder to come by (they seem to be, in my :searching, related to HVAC and other "industrial" uses; they're not :like the overly-abundant 12VDC relays we have for our cars and such) I :would like to use a 12VDC relay at the device. However, I'm afraid :that if I attempt to use 12VDC to control these relays over a distance :like I mention of up to 100 feet that the line-loss will be :significant (on 12VDC). I was thinking that using 24VAC would be much :better (it's higher voltage and it's AC, so line-loss should be quite :a bit less than 12VDC). : :I tried this: I took the 24VAC and rectified it with a single diode :(half-wave) then buffered it to "ground" with a 2,200 uF cap. I ran :that buffered output through a LM78L12 (with input and output caps as :detailed in the datasheet I was reading) and that output to the 12VDC :relay I wanted to operate. When I apply 24VAC to the circuit the :relay turns on like a mousetrap: SNAP! But, when I remove the 24VAC :the relay turns off like a marshmallow; slow and makes some light :clicking noises. So, the input cap (2,200 uF) is still powering the :relay coil and is letting it down slowly; at least, that's my hack-boy :assessment: I'm not an expert at any of this stuff. I tried whatever :I could including "pull-down" resistors (if I'm using that term :correctly); I put a 10k resistor from the relay input to ground. I'm :afraid the slow turn-off is going to cause arcing and fry the relay :contacts. : :I don't have a lot of caps that are rated at 35 volts or higher (which :could handle the 24VAC, rectified) so experimentation was limited in :various cap sizes (like, could a 1,000 uF input cap allow the relay to :turn off quickly?); somewhere I got the idea that if you put more :voltage across an electrolytic cap than it can handle it can "explode" :or "pop" or "blow-up" or whatever so I'm reluctant to use 16V :electrolytics on what should be 24V or higher. : :Anyway, I hope I've done a good job of explaining what I'm trying to :do and what I've tried to do to achieve it. Is there an effective way :to run a 12VDC relay from a supply circuit of 24VAC? : :Thanks in advance. : :--HC

The first thing to sort out is whether your "boat load of 12V relays" is suitable for controlling 120Vac devices. Current rating sounds reasonable but the critical detail is "isolation voltage rating" between the 12Vdc side(coil) and the 120Vac side (contacts). Any relay which is to be used for this purpose must meet certain standards and automotive types are not suitable. Many "industrial" types will be suitable.

Your approch to producing the dc coil operating voltage is ok and your reasoning for the long release time is also correct. The release time problem is easy to overcome. When you want the relay to release, you don't switch off the 24V supply at the ac wall switch, you simply disconnect the relay coil from the dc output voltage of your power supply.

John Popelish gave you the answer - you won't have a problem driving those 12 volt relays with 100 feet of wire.

But as a learning/experimental/fun thing, you could try using a zener & transistor circuit with the 24 volt approach you mentioned.

Ed

table,

Hey, John, thank you for your reply. That wire link is cool and I've bookmarked that.

What I've been intending to use as the "messenger" wire for the relay is some 24 gauge wire I have already (from where and when I don't recall; it's followed me for about 10 years now).

If I understand you correctly, the resistance of the wire for this distance and for this load (the relay coil) is not significant enough to worry about, within some parameters. I went and measured my relay's coil resistance and found it to be 308 ohms. I then applied

11.89 volts (according to my DVM) from my proto-board and measured 37.7 mA (using the same DVM) across the relay coil with that voltage. I guess all I needed to measure was the 308 ohms, but since I was there it seemed useful to measure the rest. I ran it through the equations I have for power (Watts) and Ohm's Law, just to see if my measurements were right and got numbers that seem good.

1/20th of the 308 would be 15.4 ohms, so, you're saying I could add up to 15.4 ohms of round-trip "messenger" wire to the relay, on a 12VDC supply, and it would be okay? So, looking at the chart you linked to,

24 gauge has 38.958 feet per ohm so I could run, round-trip, roughly 600 feet?

I really am an amateur at this stuff and I've seen your posts all over the place helping people so I ask the following with no ill-will or attitude towards you; you said that a 2W coil would be about a 288 ohm coil. I tried to get that (because I want to be able to calculate this information for other relays and to better understand this stuff) assuming a 12 volt DC supply and I did a calculation for current with P = I E so I = 2W / 12V and got I = 0.1667A. I put that into E = I R solving for R = 12V / 0.1667A, R = 72 ohms. What have I done wrong?

Thank you very much again for your reply. It helped a lot. I will set up a test rig to verify that the numbers I work out on paper are accurate but I'm confident this will work. This is going to simplify my project quite a bit (not having to use a 24 VAC supply to try to run 12 VDC relays).

--HC

Sounds about right. 300 feet out and back, with about a 5% voltage loss. Most DC relays will pull in with less than

90% of their rated coil voltage.

You caught me making a math error. I was using P=V^2/R, or

2W=12V*12V/R, but I hit a wrong key, somewhere. Its getting late, here. Sorry.

By the way, this means that your 308 ohm coil will consume only about 12*12/308= 0.47 watts. That makes the wiring resistance a lot less of a problem.

You should also test your relay with a slowly rising coil voltage and make sure it pulls in at well below 12 volts.

--
Regards,

John Popelish

reasoning

Hey, Ross, thanks for your reply. You make a very good point about the isolation rating/capabilities of the automotive-style relays. I do, also, have some 12VDC relays that are rated for 250VAC and 10A and a good supplier of those, so I will still have some 12VDC relays I can use if the auto style won't work. I'll check to see what the isolation is on those auto relays; I bet you're right that it won't be enough.

The problem that I see with disconnecting the relay coil from its power is that switch would need to be a relay, as well; all of this is supposed to be remote. On the bench I did what you suggested (breaking the connection from the coil to the PS) and it works great; the problem is how to switch that off remotely without using a relay or similar device. What John Popelish has suggested is going to work for me, I think; the loss across the wire won't be as significant as I originally feared.

--HC

Hey, Ed, I'm game. I did some thinking along the lines you suggested and the best I can come up with is this: rectify and buffer (diode and capacitor) the 24VAC which gives me a little over 30DCV. Run that through my voltage regulator (LM78L12) to get 12VDC. That can power my 12VDC relay coil. I need a way to break that connection crisply at the relay coil so I could use a Zener diode from the (roughly) 33VDC supply (rectified and buffered) to the base of a transistor. I would select a Zener that could pass, say, over 30 volts (1N4751 might work according to the table I found here:

and, with proper resistors to bias the transistor, it would turn on when power was supplied, switching on the relay coil. When power was cut the voltage/current from the capacitor would start to drop relatively slowly. There would be enough for the LM78L12 to continue to provide 12VDC to the relay coil but would quickly fall below the Zener voltage and turn the transistor off sharply in turn shutting off the relay coil sharply.

Is that right? I had not thought about something like that. I'm not sure I got that right but if it is right that is pretty cool.

Thanks for the suggestion. I don't have any Zeners in my collection of stuff and my supply store is about 100 miles away so testing is going to be impossible until I make the trip. I dunno, I might get a wild hair and go later today.

--HC

Assuming the relay contacts can handle the load, do this:

|\\|

---------------| |----------------+---------------+ |/| | | | |

24 ac | | in relay ----- coil / \\ | --- | | | | | |

----------------------------------+---------------+

where the goofy looking things are any old power diodes, 1N4001 or whatever.

The relay coil will see pretty close to the equivalent of 12 volts.

John

Hey, John, I've been thinking about what you've said here and I think there is a concept I had not realized before: the voltage drop across the relay coil will determine its ability to work; no enough voltage and it won't activate or won't activate fully. The wire before and after the relay coil (what comes from and returns to the power supply) will act as series resistors with the load (the coil). So, assuming that 90% of the rated coil capacity is sufficient to activate the coil fully (and I can do some testing to be sure) then the coil needs to "see" 10.8VDC minimum.

So, if I'm right: 10.8VDC across a 308 Ohm coil (as measured on my relay) gives me a current of 0.035A. Knowing that I can drop 0.6 volts across each leg of wire going to this remote coil for a total of

1.2 volts dropped I can calculate the resistance allowable in the wire: 0.6V / 0.035A = roughly 17 Ohms. So, I could have a resistance in series with the relay coil of a total of 34 Ohms. It should not then matter if this resistance is actually one or more "discrete" (if I'm using that term correctly) resistors or just the resistance of the wire that the circuit is comprised of. I could then use that information to calculate the maximum wire length for ANY relay as long as I know it's coil resistance, minimum activating voltage, and the wire resistance.

Is that correct?

I will test my relay to find out what the minimum activating voltage is.

Thank you again.

--HC

John, thank you for the reply. I put that into Notepad (to see it in a fixed-width font) and that's pretty simple. But won't the voltage across the relay still be AC? I'll try this a little later today and see what happens. John P has given me information that shows that I could run these relays remotely with 12VDC without a problem (I had originally feared drastic line-loss) but I'm still going to try to find a way to use the 24VAC for no other reason than that I hate to try something and fail. So, I'll probably wind up using the 12VDC route, but I'm going to spend some more time trying to use 24VAC "just because". Ed suggested using a Zener and a transistor and I've come up with an idea that might work so I'm learning some stuff in the process.

--HC

--
No, it\'ll be 1/2 wave rectified AC, and John\'s saying that by using
only the positive portion of the sine wave the relay will dissipate
the same power it would if it were hooked up to 12VDC.  That\'s a
trick often used to good advantage with devices like heaters, but I
don\'t think it\'s a good idea for your application because the period
of the signal will be about 17ms, which may make the relay chatter.

A better way, IMO, would be to full-wave rectify the AC to get the
period of the DC pulses down to 8ms and then use a 300 ohm 1 watt
resistor in series with the coil in order to get the DC down to 12V.

Either that or (since you\'ve got a lot of them lying around) use
another relay coil in series with the first one to drop the voltage.
That way you\'d also have twice the current carrying capacity of the
contacts if you wired them in parallel.

24AWG wire has a resistance of 2.57 ohms per 100\', so your 200\'
(100\' out and 100\' back) will have a resistance of about 5 ohms,
which will have an insignificant effect on the operation of the
relay(s).

You understand it correctly. There's a little more to it, however. You would want to drop the relay when the supply is voltage has dropped considerably lower than 30 volts.

If you want to experiment with it, you don't need a zener in the neighborhood of 30 V - use a 6V zener, and add a two resistor voltage divider. Here's a diagram:

• ---+---+---7812---+------+------+ | | | | | | c | [C1] | [C2] [Relay] [D2] 1N914 R1[1K] | | | | | a | +----+-----+ +------+ | | 6v / +------------[Zd]---| NPN | | \ R2[1K] | | | | | Gnd -+--------+------------+

C1 & C2 are the standard caps used with a 7812, typically .33 uF for C1 and .1 uF for C2. The 1N914 (can be a 1N4148,

1N4001 etc) diode is installed "backwards" (that is, with the striped end toward +) across the relay coil to prevent the inductive spike that occurs when the relay is suddenly de-energized from doing damage.

R1 and R2 form a voltage divider, so the zener "sees"

1/2 the supply voltage. That also limits the current into the base and reduces the dissipation in the zener. As long as the supply voltage has not dropped below about 12 volts, the relay stays energized, but when the supply drops below 12 volts the relay drops out. A general purpose NPN transistor, like a 2N2222, will work well with your relay.

By the way, the R1/R2 voltage divider is not mandatory if you change the zener diode. You could eliminate R2 and use a zener rated around 9 to 12V. It is also possible to design the circuit with the zener on the output side of the 7812. There's a lot of room for experimentation. :-)

Ed

It probably won't. The second diode keeps current circulating in the coil inductance between half-cycles, and most DC relays have a pretty high pullin:dropout ratio.

John

Yes. now you are designing. You could also use this analysis to determine what DC supply voltage you need to have 12 volts left at the coil after the wiring drops its voltage.

Then derate it by making sure you give it at least 5% more than that to make it operate even when it is hot (which raises its coil resistance).

--
Regards,

John Popelish

That circuit uses the inductance of the relay as an averaging mechanism for the half wave rectified voltage. For half a cycle, the current increases as the half sine wave pushes current through the coil. Then, for a half cycle, the inductance tries to reverse voltage as the current falls, but the right diode turns on and shorts it out, so that the current falls more slowly than if the coil was left open circuit. The open question is whether the current ripple is smooth enough that the relay will not buzz and burn the contacts. And that depends on the L/R time constant of the coil inductance (L) and the coil resistance (R). If that time constant is much larger than the line cycle's period, then the current will be fairly steady.

--
Regards,

John Popelish

On Sat, 16 Feb 2008 09:16:08 -0800 (PST), HC wrote:

:On Feb 15, 10:41 pm, Ross Herbert wrote: :> On Fri, 15 Feb 2008 20:19:36 -0800 (PST), HC wrote: :>

:> :Hey, all, I'm not sure if this can be done but here's what I am trying :> :to do and how I've tried to go about doing it. :> : :> :I want to be able to control 120VAC devices some distance away from a :> :controller, say, up to 100 feet or so. I would like to run the 120VAC :> :to the device through a switch (relay) at the device with no other :> :switches or control devices in line from the breaker. Then I would :> :like to have the controller turn that relay on and off to control the :> :device. The idea is that I could run the thicker, high-voltage lines :> :directly to the device and then use smaller wire to operate a relay at :> :the device to turn it on and off instead of running the high-voltage :> :wire to each switch I would like to use. :> : :> :Since I have a boat-load of low-cost 12VDC relays that can switch up :> :to 250VAC and 15 amps and since 24VAC relays seem, in my searching, to :> :be a lot more expensive and harder to come by (they seem to be, in my :> :searching, related to HVAC and other "industrial" uses; they're not :> :like the overly-abundant 12VDC relays we have for our cars and such) I :> :would like to use a 12VDC relay at the device. However, I'm afraid :> :that if I attempt to use 12VDC to control these relays over a distance :> :like I mention of up to 100 feet that the line-loss will be :> :significant (on 12VDC). I was thinking that using 24VAC would be much :> :better (it's higher voltage and it's AC, so line-loss should be quite :> :a bit less than 12VDC). :> : :> :I tried this: I took the 24VAC and rectified it with a single diode :> :(half-wave) then buffered it to "ground" with a 2,200 uF cap. I ran :> :that buffered output through a LM78L12 (with input and output caps as :> :detailed in the datasheet I was reading) and that output to the 12VDC :> :relay I wanted to operate. When I apply 24VAC to the circuit the :> :relay turns on like a mousetrap: SNAP! But, when I remove the 24VAC :> :the relay turns off like a marshmallow; slow and makes some light :> :clicking noises. So, the input cap (2,200 uF) is still powering the :> :relay coil and is letting it down slowly; at least, that's my hack-boy :> :assessment: I'm not an expert at any of this stuff. I tried whatever :> :I could including "pull-down" resistors (if I'm using that term :> :correctly); I put a 10k resistor from the relay input to ground. I'm :> :afraid the slow turn-off is going to cause arcing and fry the relay :> :contacts. :> : :> :I don't have a lot of caps that are rated at 35 volts or higher (which :> :could handle the 24VAC, rectified) so experimentation was limited in :> :various cap sizes (like, could a 1,000 uF input cap allow the relay to :> :turn off quickly?); somewhere I got the idea that if you put more :> :voltage across an electrolytic cap than it can handle it can "explode" :> :or "pop" or "blow-up" or whatever so I'm reluctant to use 16V :> :electrolytics on what should be 24V or higher. :> : :> :Anyway, I hope I've done a good job of explaining what I'm trying to :> :do and what I've tried to do to achieve it. Is there an effective way :> :to run a 12VDC relay from a supply circuit of 24VAC? :> : :> :Thanks in advance. :> : :> :--HC :>

:> The first thing to sort out is whether your "boat load of 12V relays" is :> suitable for controlling 120Vac devices. Current rating sounds reasonable but :> the critical detail is "isolation voltage rating" between the 12Vdc side(coil) :> and the 120Vac side (contacts). Any relay which is to be used for this purpose :> must meet certain standards and automotive types are not suitable. Many :> "industrial" types will be suitable. :>

:> Your approch to producing the dc coil operating voltage is ok and your reasoning :> for the long release time is also correct. The release time problem is easy to :> overcome. When you want the relay to release, you don't switch off the 24V :> supply at the ac wall switch, you simply disconnect the relay coil from the dc :> output voltage of your power supply. : :Hey, Ross, thanks for your reply. You make a very good point about :the isolation rating/capabilities of the automotive-style relays. I :do, also, have some 12VDC relays that are rated for 250VAC and 10A and :a good supplier of those, so I will still have some 12VDC relays I can :use if the auto style won't work. I'll check to see what the :isolation is on those auto relays; I bet you're right that it won't be :enough. : :The problem that I see with disconnecting the relay coil from its :power is that switch would need to be a relay, as well; all of this is :supposed to be remote. On the bench I did what you suggested :(breaking the connection from the coil to the PS) and it works great; :the problem is how to switch that off remotely without using a relay :or similar device. What John Popelish has suggested is going to work :for me, I think; the loss across the wire won't be as significant as I :originally feared. : :--HC

Are you referring to the operating method suggested by John Larkin with the relay powered from 24Vac?

If so, that is a resonable and simple method of using a higher voltage to activate the 12V coil, but as John Popelish hinted there is a sort of balancing act to be considered in order to avoid relay buzz.

I don't see a problem with using another relay to switch the DC supply. In fact that would be an ideal use for one of your cheap 12V automotive relays. You could use the suggested circuit from John Larkin to power that relay. Even small gauge wire would not produce too much potential drop in the operating loop of the power switching relay.

--
"It probably won\'t"?

I\'ve found the opposite to be true, empirically.

Here:

                       COIL R  COIL L       
  MFG      PART NO      OHMS   HENRY    REF
--------|------------|-------|--------|-----
AROMAT   JT1a-DC12V    150      0.2      A
 
AROMAT   HC3-P-DC12V   160     0.27      B

ORIGINAL SRUT..12VDC   430     0.41      C

AROMAT   JW2EN-EDC12V  275     0.52      D

ECI      8501-3023-12   50    0.148      E


For your circuit:


ACIN>--[D1>]--+-----+
              |K    |
            [D2]  [COIL]
              |     |
ACIN>---------+-----+
                  
I got:

       PULLIN   DROPOUT
 REF    VRMS      VRMS
-----+--------+---------
  A     24.3     24.3

  B      >30      N/A

  C      >30      N/A

  D      >30      N/A

  E     14.8      9.3


For my circuit:

       +-----+
ACIN>--|~   +|--[Rs]--+
       |     |        |
       |     |     [COIL]
       |     |        |
ACIN>--|~   -|--------+ 
       +-----+

where Rs is a fixed resistor equal to the DC resistance of the relay
coil, I got:


       PULLIN   DROPOUT
 REF    VRMS      VRMS
-----+--------+---------
  A     15.1      9.8

  B     15.5      9.3

  C     14.6      8.8

  D     14.5     10.9

  E     18.5      7.5 

I just picked the first five 12VDC relays I could find here and
tested them using a VARIAC to vary the input voltage, an AC
voltmeter to measure it and an ohmmeter to determine contact closure
and opening.

So, it looks like my circuit got all five relays to work, while
yours only got one to work, the ECI unit, which is a 12PDT T-BAR
relay.  Not something you come across every day.

I have added a third column to your relay data that is L/R, the time constant of the coil. I think this needs to be larger than 1/F (1/60Hz=0.0167s) for the two diode circuit to work well (little vibration).

The last one has the longest L/R time constant, but it still is too short for me to trust to not vibrate the contacts with the two diode circuit, unless the supply frequency was

400 Hz (0.0025s/cycle). I am a little surprised it worked as well as it did on 6 Hz. It must have a rather massive armature whose mechanical time constant helped smooth the bumps.

--
Regards,

John Popelish

For certain values of probably. A fast relay is going to need a capacitor somewhere, since no power is available at zero crossing, no matter how many diodes you use.

Your circuit adds the Rs resistance to the coil circuit, decreasing L/R by about a factor of 2:1, which negates the improvement of going full-wave. Better would be to put the resistor *before* the bridge, so the diodes clamp the coil voltage close to zero.

John