Turning autotransformer to isolation transformer

[Simranjit]

Hi, is it possible to turn this auto transformer into isolation transformer by adding my own winding on top of it. I bought this 5kW thinking that it might have two windings. I wanted to use it for isolating tesla coils. I don't know what to do with it now.

[klugesmith]

Might work. You need to remove the central bolt and see how much core area is open in the middle.
If there's enough room for an extra 5 kVA winding, I suspect the designers chose a core bigger than necessary.
It's tedious to make toroidal windings with many turns, nice and tight, without a toroid winding machine.

/>
A test winding of a few turns, with transformer operating at nominal 240 and 120 V, can tell you the volts per turn.

[Simranjit]

Might work. You need to remove the central bolt and see how much core area is open in the middle.
If there's enough room for an extra 5 kVA winding, I suspect the designers chose a core bigger than necessary.
It's tedious to make toroidal windings with many turns, nice and tight, without a toroid winding machine.

/>
A test winding of a few turns, with transformer operating at nominal 240 and 120 V, can tell you the volts per turn.

I don't have spanners big enough to open the bolt at the moment. But I carefully poked through the rubber seal on top with a screw driver There is almost 65mm diameter of free space in center. It seems doable but I guess it will be project for another week. I have to mentally prepare myself 😁😁😁 for winding it manually. I will just keep it if I want to run something on 220v. Anyways thank you very much for your help. Another fun thing I noticed if 220v switch is selected on back and 110v is applied to input. I get 60v and 110v on outputs. I guess this will also come in handy eventually. Although selecting 110v switch on 220v input would be bad idea I guess.

[klugesmith]

Although selecting 110v switch on 220v input would be bad idea I guess.

Core would saturate, primary current might exceed 100 A, and the transformer should immediately blow a fuse.

[rikkitikkitavi]

Considering 5000VA @ 220V = 22Amps => aprox 3 mm diameter winding ( @ 3A/sq mm)

A typical large transformer like this is around 1,5 Volt/turn => 220V = 220 / 1,5 = 145 turns

So these 145 turns needs to fit in a 65 mm hole.

145*3 = 435 mm , but the hole circumference is only pi*65 = 200 mm. This is not including the insulation tape.

Thererfore they will overlay each other at the centerhole , reducing dissipation capacity.

I think it will be difficult to get the full 5 kVA from this transformer.

To be honest , judging from the pictures it looks quite small for a 5 kVA transformer. Perhaps is it thermally designed for around 2kVA continous load.

I have a 5kVA variac and it is 35 cm in diameter, 10 cm tall and with 3 mm wire. Weighs about 15kg, maybe even 20kg.

[klugesmith]

Yay for Rikki's words of practicality and experience.

I just noticed that in my linked video of toroid-winding machine, we see a non-twisted multifilar winding going on.  Instead of thick magnet wire, it looks like about 4 strands of thinner magnet wire wound as one, and lying side by side against the core.   Might that be just for convenience in termination, with the strands electrically in parallel, as opposed to independent coils with extremely low leakage inductance (close to 100% flux sharing)? I see nothing to identify the individual strands, whose free ends are hand-twisted for a short distance before operator starts loading them onto the big bobbin.

[edit] It might be tricky to get the strands all wound with about the same tension, unless there's a bit of stretching on the way from bobbin to toroid.  Anybody know the amount of residual tension in typical windings?  It generally reduces conductance, as we see in strain gauges.  Winding copper wire on iron to make electromagnets has been practiced for 200 years!

[rikkitikkitavi]

Leakage for a 50Hz transformer on a toroid is not so crucial in my experience. The core shapes the field, but it is considered good practice to fill the entire core circumference of course.

It could be that the transformer in the video has 4 individual secondaries, but for manufacturing purpose they are twisted during winding in the beginning and end? The wire does not seem to be thick enough to motivate anything but a single-wire-per-winding type. I guestimate the core to be around 500VA and around 50-60 secondary turns, ie 20-25Volts.
Or it is just that the winding machine can only handle up to a certain diameter ?

Anyhow, multifilar windings is easier to handle than single strands when it comes to high current secondaries of course. Easier to find aswell, but the utlisiation of the core window might suffer.

I have no idea about tension, but most transformers have quite a tense secondary but they undwind or release when starting to unwind in the end.
The effect on conductance, or winding resistance is probably neglible as compared to the thermal coefficient?

The windings must not stretch anywhere nere to breaking/floating as it will damage insulation. I think the tension is just sufficient for the windiings not to start humming along with the magnetic field.