Hi all,

Infrequent poster- more mechanically than electrically inclined. Hoping to learn something from you guys.
I got the idea to build a variac and happened to have a rotor, about 4" by 4", from an electric motor. I carefully wound it with 12ga wire, figuring it'd be adequate to carry 10-15 amps. It would trip a breaker so I put a water heater resistor element in series to deal with inrush but it still usually trips the breaker when I cut out the resistor.
1. The motor rotor consists of a stack of laminated plates of, I imagine, silicon steel, but I've learned that variac cores are made of a single strip wound like in a spiral. Could this be part of my problem, and if not, will it adversely effect operation or efficiency if I get it to work?
2. Is 12ga wire too heavy for the capacity I'm shooting for? If so, what would be the appropriate guage- 16 maybe?
3. I wound about 55 turns around the core in a single level on the inside, figuring this would give me about a 2v jump from one turn to the next, which is good enough. But I've noticed all the commercial variacs have many more turns stacked two or three high on the inside. Will it not still work with fewer turns like this?

And before someone suggests that I just buy one, I've already considered it. I just get more satisfaction out of building than buying. Weird, huh?
Thank you very much for any advice.

Rick A

 

Not enough turns = not enough inductance = too much magnetising current = blows circuit breaker.

 

Hi all,

Infrequent poster- more mechanically than electrically inclined. Hoping to learn something from you guys.
I got the idea to build a variac and happened to have a rotor, about 4" by 4", from an electric motor. I carefully wound it with 12ga wire, figuring it'd be adequate to carry 10-15 amps. It would trip a breaker so I put a water heater resistor element in series to deal with inrush but it still usually trips the breaker when I cut out the resistor.
1. The motor rotor consists of a stack of laminated plates of, I imagine, silicon steel, but I've learned that variac cores are made of a single strip wound like in a spiral. Could this be part of my problem, and if not, will it adversely effect operation or efficiency if I get it to work?
2. Is 12ga wire too heavy for the capacity I'm shooting for? If so, what would be the appropriate guage- 16 maybe?
3. I wound about 55 turns around the core in a single level on the inside, figuring this would give me about a 2v jump from one turn to the next, which is good enough. But I've noticed all the commercial variacs have many more turns stacked two or three high on the inside. Will it not still work with fewer turns like this?

And before someone suggests that I just buy one, I've already considered it. I just get more satisfaction out of building than buying. Weird, huh?
Thank you very much for any advice.

Rick A

Hi,

The previous post summed it up pretty well

If you dont have enough turns the no load current will be so high you cant use it for anything because it can draw way too much current. So that's an experiment point. You have to do some testing probably with a real variac so you can vary the input voltage, or else you've got to be willing to use some resistance in series with the winding in order to test, like a 100 watt light bulb or something. If the bulb lights to full brightness the core is drawing too much current, but of course a current reading would be better.

You could estimate the permeability and then go from there. The permeability of the steel may be as high as 1000 but we'd also have to know if there are any gaps which would reduce that significantly. Knowing the original voltage of the motor might help too.

Variac's are very hard to make BTW, but you could make a multi tap transformer where you could choose what tap you want to use to raise and lower the output voltage.

 

Use a resistor or light bulb in series with the winding and measure the voltage across the winding.
That will give you a rough idea of how much voltage the winding can tolerate before saturation and how many more turns you need to add to operate at the full line voltage.

 

Buy one. They are not a device that anyone can just whip up out of random electric motor parts.

However if you insist on experimenting your stator core is the better part to use not the rotor. Fill the slots with epoxy and wind around it.
It still wont be ideal but it will be better.

 

You realize that a Variac, typically is wound on a large cylinder, At the "back" of the cylinder the wire is flattened and the varnish insulation is removed while attempting to keep the wires from touching. Full voltage is applied to the winding.

But, Then a small thin spring-loaded brush runs along this exposed winding creating the tapped transformer.
The wiper is always fused, since the wire determines the current. At 1V, 10A or 120 V, 10A. In one case 10 W and in the other 1200W, you exceed the 10A and the design is for 10A, the Variac van blow.

The center shaft is usually thoroughly coated with a thick insulator. You have the carbon brush for the winding and another slip ring type of contact to be able to bring that winding out.

 

What I don't get with a variac is that the brush must cover more than one wire else the voltage would be zero in between wires. So there must be turn(s) shorted by the brush and why does that not cause trouble?

 

The brush is fairly thin. A shorted turn or two has to be part of the design.

 

What I don't get with a variac is that the brush must cover more than one wire else the voltage would be zero in between wires. So there must be turn(s) shorted by the brush and why does that not cause trouble?

That brush contact is a very, very special material as can contact two or more windings causing short circuit in windings, highly undesirable on a variac. That material has the property of conducting in one axis much better than transversally. Cannot be replaced with a carbon brush. It is a very weird property material that should be taken in account.
I will try to search the information and edit later here if found.

The brush is fairly thin. A shorted turn or two has to be part of the design.

Nope. Does not work that way.

Edited: Grain-oriented anisotropic graphite.
---> http://sound.whsites.net/articles/variac.htm

(Rescued from my old post at http://forum.nutsvolts.com/viewtopic.php?f=45&t=13819&hilit=variac )

 

yuk! I just was thinking of this sort of brush: http://www.ebay.com/itm/191490459308

 

Thank you all for the input. I found it pretty hard to wind the high number of coils some of you mentioned. They have to be stacked two or three high in the inside of the core, which was, for me at least, hard to keep neat and tight. Probably easier with smaller gauge wire, which is why I asked about that.

The idea of a multi-tap xmfr occurred to me- would need a LOT of taps in order to get much of a selection of voltages. Most of the references I found address a few taps a few volts apart and near the nominal output voltage, not across the whole 0-120v range.

I'm curious why the stator would work better than the rotor. It would certainly make a substantial variac and I may still have it lying around- if I don't get talked out of the whole project! I don't disagree that whatever I could make would undoubtedly be much inferior to the real thing. May give up and keep my eye out for a used one after all.

Rick A

 

Edited: Grain-oriented anisotropic graphite.

Thank you for that. A very cunning plan indeed

 

I'm curious why the stator would work better than the rotor. It would certainly make a substantial variac and I may still have it lying around- if I don't get talked out of the whole project! I don't disagree that whatever I could make would undoubtedly be much inferior to the real thing. May give up and keep my eye out for a used one after all.

Because given the input voltage and the amperage range you want to work with plus the wire size you are using it will need to be that big to work with a common 60 Hz line frequency.

120 volts times 15 amps is 1800 VA which would require a similar core of cross sectional size as a motor of similar wattage power such as 2.5 - 3 HP unit. Also it will likely need to have ~2 - 3 turns per volt as well which is only going to fit on so small of diameter core anyway.

 

Use a resistor or light bulb in series with the winding and measure the voltage across the winding.
That will give you a rough idea of how much voltage the winding can tolerate before saturation and how many more turns you need to add to operate at the full line voltage.

Hi,

Now where did i hear that before...oh yes, in the post just before yours

 

Thank you, tcmtech. I think I could learn more about electricity hanging around these forums than I did in HS and college physics!

Rick A