230V Variac Wiring Question

[CarsonReidDavis]

I recently acquired a rather beefy variac, and I was hoping for some clarification on the correct ways to wire it before I mess anything up.

It is a Powerstat Type 1226, 2.5KVA, 230/115V.

My other variacs have wiring diagrams on the body somewhere. This one does not. Here is a link to a product catalog that has related information, but about the X-1226, which appears to have different ratings. http://lcweb2.loc.gov/master/mbrs/recording_preservation/manuals/Powerstat%20Catalog%20P258G.pdf

Thanks for the help!

[UPDATE] I have uploaded an album of testing photographs and a similar wiring diagram at this link https://imgur.com/a/1zh3Fn0
[UPDATE 2] I have uploaded a nice size comparison to show how huge this variac is at this link https://imgur.com/a/B4tsuq6
[UPDATE 3] I have uploaded detailed calculations and testing of the unit at this link https://imgur.com/a/DYEUdZj

[CarsonReidDavis]

So I have done a few tests but I am still not satisfied I completely understand what is going on. Pictures have been added to the album here https://imgur.com/a/1zh3Fn0

I am using the two unlabeled posts as ground for the input and output and the post labeled output as the output.

• When connecting the input to the post labeled 115V, the variac produces around 1:2.32.
• When connecting the input to the post labeled 230V, the variac produces around 1:1.16.

So it seems the unit was intended to be able to accept 115V and output up to 230V. It was also intended to accept 230V and output up to 230V.

I think I might be able to use either of the two bottom posts as ground for both input and output. Similar units have the words common ground written here.

As a side note, when I fed the unit much more than 115V on the 115V post, say 135V, it began to make an odd humming noise.

Finally, would I be correct in assuming that it can handle around 22A at 115V, based on the 2.5KVA rating? This appears to be incorrect.

[rbm]

Although the power rating for the 1226 you have differs from the X-1226 or O-1226 in the catalog to which you linked, the wiring diagram in diagram E on page 77 seems to match your configuration.

What are you trying to accomplish?  115V fixed input with up to 115V variable output from the wiper?

[Shock]

I have a different model but has exactly the same ratios. On mine 240V input and 0-280V output which is 1.16 times. This is better than 1:1 as you can make up for transformer or line losses if you need to.

You may want to use it to step up 2.32 times but it's more flexible and permanent I think to have a dedicated step up isolation transformer.

[Ian.M]

Variacs have two limiting currents - the max. winding current and the max. wiper current.   In the absence of manufaacturer's model specific rating data, all you can do is assume the two limits are the same.    You can calculate the max winding current   from the VA rating divided by the highest nominal input voltage, so for this  Powerstat Type 1226, 2.5KVA, 230/115V, variac, that's 2500/230 = 10.87A.    If you wire it using the 230V input tap (either for 230V in, 0 to 267V out, or for half-voltage operation 115V in, 0 to 133V out), the max. output current is the max. winding current.    However if you wire it for 115V in, boosted to 0 to 267V out, above 230V (double the input voltage) you have to derate it linearly by 1 / (boost_ratio - 1) to keep the current in the lower half of the winding below the input tap under the max. winding current.   Therefore at the max. boost  of x 2.32, the max output current is 10.87/1.32 = 8.23A.

Therefore you need to fuse the output at either 10A if using the 230V tap or 8A if using the 115V tap.  In practice a resettable thermal circuit breaker makes more sense than a fuse, as it will be reasonably resistant to short term overload.   If using it on a 230V circuit capable of more than 10A, the input should also be protected with a 10A slow-blow fuse or circuit breaker.  When using  the 115V in tap, it should have a 20A input fuse or circuit breaker.   That's the same or more than the panel breaker rating for a normal U.S. domestic NEMA 5-15R, or NEMA 5-20R 115V outlet so can be omitted at the variac.

You've also got some metal-bashing or 3D printing to do to make up a terminal cover with cord grips unless you plan to panel mount it in a larger enclosure.  Its chassis *must* be grounded.
 

[CarsonReidDavis]

Although the power rating for the 1226 you have differs from the X-1226 or O-1226 in the catalog to which you linked, the wiring diagram in diagram E on page 77 seems to match your configuration.

What are you trying to accomplish?  115V fixed input with up to 115V variable output from the wiper?

I added the wiring diagram you referenced to the original album. However, I think it isn't exactly the same as my unit. I believe posts 2 and 4 need to be swapped.

• The 230 volt input which produces 1:1.16 should be tapped near the end of the coil.
• The 115 volt input which produces 1:2.32 should be tapped before the middle of the coil.
• The common ground and output appear to be the same.

I'm not sure how I plan on wiring it in the final configuration. Since all my other variacs are only 0-140v, I will probably input the wall 120 volts into the 115 v input in order to output 0-280 v. Unless there is some kind of current handling benefit to wiring the 120 into the 230 for only 0-140 volt output but at higher current.

[CarsonReidDavis]

@Ian.M

I have a Powerstat 116-B which has the following listed specs:

• 120V in
• 0-140 out
• 1.4KVA
• 10A

So it would seem that in this instance the manufacturer is calculating current rating as (KVA rating)/(max output voltage across whole coil)=(current rating)

So I'm not sure I fully understand why I can't wire the theoretical wall 115V into the 230V input, which will cause the variac to output 0-133V. Since 133V would be my maximum voltage in this configuration over the whole coil, wouldn't my rough calculation be (2.5KVA)/(133V)=18.8A?

In reality I get about 122V out of my wall, so my actual max voltage in this configuration would be 141.5V giving me a current rating of 17.7A.
If I wired it in the 1:2.32 configuration, I would expect a max voltage of 283V and max current rating of roughly 8.8A. As explained below, even though power is the only listed rating on the variac, it seems that current is the limiting factor.

[Ian.M]

@Ian.M

I have a Powerstat 116-B which has the following listed specs:

• 120V in
• 0-140 out
• 1.4KVA
• 10A

So it would seem that in this instance the manufacturer is calculating current rating as (KVA rating)/(max output voltage across whole coil)=(current rating)

So far, so good.

So I'm not sure I fully understand why I can't wire the theoretical wall 115V into the 230V input, which will cause the variac to output 0-133V. Since 133V would be my maximum voltage in this configuration over the whole coil, wouldn't my rough calculation be (2.5KVA)/(133V)=18.8A?

In reality I get about 122V out of my wall, so my actual max voltage in this configuration would be 141.5V giving me a current rating of 17.7A.

No, because the current rating of the wire its wound from doesn't magically go up when you decrease the input voltage to maintain the same VA.   

Its still only rated for 10A out, so in that configuration, with 122V in, at 133V out, you only can draw 1.33KVA while staying under the 10A limit.

If I wired it in the 1:2.32 configuration, I would expect a max voltage of 283V and max current rating of roughly 8.8A.

Lets check the maths.

122V in x 2.32 = 283V  output voltage
8.8A x 2.32 = 20.4A input current
20.4A - 8.8A = 11.6A  lower part of winding current.

That's over 10A so unless you have it in WRITING that its rated for 10A out at full voltage when wired for 115V input, *DON'T* *DO* *IT*!   Of course,  its possible they over-designed or underrated it, but that's not the way to bet.

Up to 2x the input voltage its good for the full 10A out.   Above 2, in the absence of any other data, to keep the current in any part of the winding under 10A, derate by dividing by (ratio -1).   So at max output of 283V, stepping up from 122V, unless specified otherwise, its good for fractionally under 7.6A

[Myrv]

When using  the 115V in tap, it should have a 20A input fuse or circuit breaker.   That's the same or more than the panel breaker rating for a normal U.S. domestic NEMA 5-15R, or NEMA 5-20R 115V outlet so can be omitted at the variac.

Is this correct?  I thought you'd have to fuse the input to 10A regardless of input voltage as you need to limit the current in the winding (not the power).

[Ian.M]

The input fuse or breaker is to minimise the damage if there's a shorted turn.  The output fuse or breaker protects the winding against overload.

At 230V, 10A out,  the input current at 115V must be 20A, as power is conserved.   There must be 10A flowing in the upper half of the winding.  Therefore (by KCL) there must be the difference between the input and output currents ( 20A-10A = 10A ) flowing in the lower half of the winding.

Of course the wire between the half-voltage input tap terminal and the winding must carry the full 20A, but its in free air, and is either the two half-winding ends loosely twisted together, each good for 10A, or the winding ends are soldered together at the toroid and a heavier gauge insulated wire is brought out to the input terminal.

[ArthurDent]

If you look at various Powerstat model numbers you will see numbers like 1126 or 1226 depending on whether it is a 120 volt or a 230 volt model. A 1126 model could be used on a maximum of around 120 volt input and cannot be used on 230 volt but the 230 volt models could be used with 120 volts input safely.

Ones like yours can be used with 230 volts input on the 230 volt input terminal to output 0-280 volts at about 10 amps. It could also be used with 120 volts input on the 230 volt terminal to output 0-140 volts at 10 amps because a transformer is all about ratios. Your model could also be used with 120 volts input on the 120 volt terminal to output 0-280 volts, not at 10 amps but only 5 amps.

The wire used in the windings on this transformer can handle 10 amps safely so that is a prime consideration in how you wire and fuse it. No wire in the winding or the wiper can safely handle more than 10 amps continually. A mistake a lot of people would make is to only fuse the input for 10 amps thinking that that totally protects the Powerstat but if you were to have the output set to only 24 volts and the output shorted, in order to blow a 10 amp input fuse with 240 volts input, you would need about 100 amps flowing through the wiper and the output terminal because of the 10:1 ratio of input:output. This is why a number of variable transformers are burned out on the lower end of the windings. If there is a shorted output and you have the knob set to the lower end, the current can become great enough to burn out the windings at the lower end. No terminal on this Powerstat should see more than 10 amps current and to safely do this BOTH the input and output leads should be properly fused at no more than 10 amps. I realize this isn’t the way most people wire Powerstats or Variacs but if you really want to protect them this is something you should consider. 

[Ian.M]

If you fuse the input to the 115V tap at 10A, that limits it to 1200VA at 120V in.  That's going to limit the output when set to 240V to 5A. 
To know if that's correct, you need the manufacturer's derating curve for the low input tap for the specific model of Variac.   

However, by KCL and transformer ratios, when drawing 5A at 240V neither the part of the winding above the 115V tap nor the part of the winding below it, nor the wiper is carrying over 5A.   

[ArthurDent]

"If you fuse the input to the 115V tap at 10A, that limits it to 1200VA at 120V in.  That's going to limit the output when set to 240V to 5A." That is correct. If you have 120 volts in and fuse the input at 10 amps then the maximum power out is 1200 VA. If you have 240 volts in then a 10 amp input fuse will give you a maximum of 2400 VA. The wire size of the windings will be what will limit the possible power output.

If you are wiring a house with 12 gauge wire, each circuit will be required to have a breaker to limit the current in the wire to 20 amps max for either 120 or 240 volt circuits although you obviously can get much more power out of the circuit if you use 240 instead of 120 volts.  The wire size is the determining factor. 

…”when drawing 5A at 240V neither the part of the winding above the 115V tap nor the part of the winding below it, nor the wiper is carrying over 5A.”  That is incorrect. If you are outputting 240 volts at 5 amps that is 1200 VA.  Assuming 100% efficiency in the transformer, you will obviously need 1200 VA input to get 1200VA out. If you have half the voltage on the input then you will need twice the current. That is why high tension lines use such high voltage, they can use smaller wire to carry very large amounts of power.
That is why one of the common failure modes of variacs is to have the lower end of the winding burn out from very high current even though the input is properly fused.   

Read ‘Performance and Limitations’ at this link ( http://sound.whsites.net/articles/variac.htm#s1 ) where it says: “One of the most common complaints about variacs is catastrophic failure. Such complaints can be found on newsgroups and forum sites when Variac problems are discussed. Almost invariably, the failure can be traced to the user failing to understand that the rated current cannot be exceeded safely, regardless of output voltage. If a fuse or circuit breaker is used (highly recommended), it should be in series with the output, not the input. A second fuse in the input circuit is also a good idea, especially if the Variac is used to boost the mains voltage (over voltage mode). Slow-blow fuses of approximately the variac's maximum current rating should be used.”

[Ian.M]

…”when drawing 5A at 240V neither the part of the winding above the 115V tap nor the part of the winding below it, nor the wiper is carrying over 5A.”  That is incorrect. If you are outputting 240 volts at 5 amps that is 1200 VA.  Assuming 100% efficiency in the transformer, you will obviously need 1200 VA input to get 1200VA out. If you have half the voltage on the input then you will need twice the current.

How do you reconcile that with KCL and conservation of energy?

5A out with the wiper above the input tap, means there *MUST* be 5A flowing through the winding between the tap and the wiper, as KCL says current cant magically appear or disappear at a point.   Transformer action forces the input current to be the output current multiplied by the ratio of output to input voltages, and by KCL the difference between the input and output currents must flow in the pat of the winding between the tap and neutral.   As long as the ratio is between 1 and 2, the current flowing through the winding between the tap and neutral, must therefore be 5A or less, in the opposite direction to the current through the other part of the winding.

I agree the input current will be over 5A - but except at ratios extremely close to 1:1 no part of the winding carries the full input current.

[ArthurDent]

Please read the link I posted and explain how it differs from what I have said. As far as conservation of energy is concerned, if you are getting 240 volts out of a transformer at 5 amps then you have to have 120 volts at 10 amps going into a transformer, half the voltage, double the current, as I explained. Keep in mind that this is a transformer and ignoring inefficiencies the power in equals the power out. If you have a 2:1 ratio or a 20:1 ratio the same laws apply. 

If you have a transformer that puts out 1 volt at 120 amps the current into the 120 volt primary will not be 120 amps from the 120volt line but only 1 amp. The linked article is correct and explains variacs well.

[Ian.M]

My problem is with:

…”when drawing 5A at 240V neither the part of the winding above the 115V tap nor the part of the winding below it, nor the wiper is carrying over 5A.”  That is incorrect.

where you quoted part of my reply #11 and disagreed with it.

I assume we can agree the wiper must carry the output current.

So, to further simplify the problem, lets consider an ideal plain 1:2 ratio step-up autotransformer with a load of 5A, and no higher.  5A out at 240V is 1200VA, so the input current must be 10A at 120V, as for an ideal transformer the input VA is equal to the output VA.   Do we agree so far?

Now please tell me what section of its winding wire is carrying over 5A and explain why . . . .
 

[ArthurDent]

Please read the link:

"The regulation is worst at 50%, because there is the maximum amount of wire resistance in circuit with the current. Current in the lower half of the winding is double that across the remainder. Leakage inductance, which is generally quite high, also affects the regulation to a small degree. The biggest limitation though, is the maximum output current. For long-term reliability, the current rating stated on the nameplate should not be exceeded, regardless of output voltage.
Unlike a traditional transformer, the VA rating is of minimal use to you. Just because the output voltage is set for one tenth of the input voltage, this does not mean that the current can be 10 times that stated in the specifications. A 2A Variac can safely supply 2A at any voltage at or below the mains voltage (100% setting). If used in over voltage mode where the output voltage is greater than the mains (e.g. 120% setting), then the VA rating must not be exceeded. A Variac used at the 120% setting can output a maximum of about 0.83 of the rated current. Assume a 1kVA, 230V Variac ...
Current = VA / Voltage
Current = 1,000 / 230 = 4.35A   -   Maximum continuous current for any voltage at or below 230V
Current = 1,000 / 276 = 3.62A   -   Maximum continuous current at 120% of input voltage"

[Ian.M]

Please read the link:

"The regulation is worst at 50%, because there is the maximum amount of wire resistance in circuit with the current. Current in the lower half of the winding is double that across the remainder. Leakage inductance, which is generally quite high, also affects the regulation to a small degree. The biggest limitation though, is the maximum output current. For long-term reliability, the current rating stated on the nameplate should not be exceeded, regardless of output voltage.

Leaving aside the sloppiness of describing a current as being 'across' a winding rather than 'through', assuming VA is conserved, the statement in red violates KCL and is therefore *WRONG*.

Perhaps a sim will help you, (below), or take a real autotransformer and put ammeters in series with the input, output and neutral terminals.

[ArthurDent]

"Leaving aside the sloppiness of describing" ...etc.)

Sorry no one can explain the obvious in terms you feel are appropriate but the bottom line is if the OP wants to wire up his Powerstat so it works correctly and safely I'd advise he follow what the expert references agree on. Feel free to contact the author of the link I referenced and tell him about his mistakes and how it is really done. Have a good day.

[CarsonReidDavis]

So I believe my fundamental misunderstanding was thinking that the variac would be rated for the listed power at all possible operating voltages. Thereby giving a different current rating for different voltages.

It seems that in reality I should use the maximum voltage anticipated by the listed rating, 266.8V (=115*2.32), in conjunction with the power rating of 2.5KVA to obtain my max current rating of 9.37 A (=2500/266.8 ). I assume this current rating is the limiting factor of the wire or some other conductor in the variac, and would not increase even if the voltage were reduced.

Without touching on any of the finer points of the above discussion, should I be safe if I fuse the input AND output at 10A? Or should I go lower?

I was planning on wiring my wall voltage of 120ish volts into the 115v input in order to output 0-280ish volts.

Due to conservation of power, I assume that in this configuration, I would only be able to actually reach 10A output while outputting less than 120 volts, other wise the input fuse would blow. I would need to rewire it with 240 input on the 230 terminal to get 10A 240 output.

[CarsonReidDavis]

Any advice on the following?

• What brand/type of fuses should I buy?
• Should I look for anything specific when I buy my wall plug?
• How about the receptacle?
• Enclosure safety/build tips?
• Integrated current meter/volt meter suggestions?

I'm a fan of the current aesthetics of the variac, so I don't really want to tack something onto the front. I think that I would rather build a separate enclosure and maybe mount both to a frame.

[Shock]

I was planning on wiring my wall voltage of 120ish volts into the 115v input in order to output 0-280ish volts.

That sounds ok. Don't take this the wrong way, I think you need to reread how the derating works until you understand it well enough to explain it.

Here is an ohms law problem as a sanity check. Say you have a load of 10A on the output at 280V so 2800W how much current would you expect to see on the 120V input?

[CarsonReidDavis]

I was planning on wiring my wall voltage of 120ish volts into the 115v input in order to output 0-280ish volts.

That sounds ok. Don't take this the wrong way, I think you need to reread how the derating works until you understand it well enough to explain it.

Here is an ohms law problem as a sanity check. Say you have a load of 10A on the output at 280V so 2800W how much current would you expect to see on the 120V input?

Yep, I'm not sure why I can't just use conservation of power.
PowerIn=PowerOut
120V*CurrentIn=2800W=10A×280V
CurrentIn=23.3A

Which is why you have to fuse the output AND the input to the max current of the wire inside the variac, or whatever the bottleneck actually is.

I dont really understand why the above isn't valid. I guess in real life there will be power losses to heat or something else I don't know how to calculate. If I was curious I would meter voltage and current on the input and output and model those losses as the efficiency of the variac. [I actually did this later in the thread. My variac appears to have about 85% efficiency]

[Shock]

I dont really understand why the above isn't valid. I guess in real life there will be power losses to heat or something else I don't know how to calculate. If I was curious I would meter voltage and current on the input and output and model those losses as the efficiency of the variac.

The sanity check should tell you with a 280V 10A load on the output you're drawing far too much current from the 120V input. No sure about you but I wouldn't just wack on a fuse and hope for the best in that scenario.

Did you check the manufacturer derating curve to determine the maximum current draw on the output?

[CarsonReidDavis]

I dont really understand why the above isn't valid. I guess in real life there will be power losses to heat or something else I don't know how to calculate. If I was curious I would meter voltage and current on the input and output and model those losses as the efficiency of the variac.

The sanity check should tell you with a 280V 10A load on the output you're drawing far too much current from the 120V input. No sure about you but I wouldn't just wack on a fuse and hope for the best in that scenario.

Did you check the manufacturer derating curve to determine the maximum current draw on the output?

I fully see the problem, but to be honest I thought the solution was to fuse the input as well as the output. That's an extra step that wouldn't be necessary with a 1:1 or 1:1.16 variac. Is there a more comprehensive solution than fusing the input and output to 10A?

Is there some reason to fuse the output to to a lower value like 5A? That would cripple my output for the 90% of the time when I want 120 in <120 out. [edit: Fusing the output to 5ishA would effectively double fuse the input to around 10A when inputing 120 and outputing 280. As calculated later, the real number would need to be 4.3ish without accounting for losses.]

So far, I haven't been able to find any manufacturer specs on this model. Do you have access to the datasheets?