Opening a Winsson 62–1225 Power Transformer

What is this beast? The label reads “Winsson Enterprises Co Ltd. Manufacturer: Zhongding CO, Part No 62–1225, Date Code 1003LF-B, China, S/N: 02222”. This transformer was salvaged from an UPS, as previously seen in Building an H-Bridge from a salvaged Uninterruptible Power Supply. This article takes a closer look at this massive transformer.

Searching for datasheets

Winsson Enterprises specializes in several industries including magnetics, their website shows this torrid winding machine:

and a transformer resembling this one, but part 62–1374 instead of 62–1225:

Searching for “winsson 62–1225” finds several pages, mostly vendors selling:

• SPW Industrial: Winsson ZHONGAING, Power Transformer 62–1225, $150
• eBay csilverblue2008, same title and picture, for $50
• This page you are reading (“Building an H-Bridge from a salvaged Uninterruptible Power Supply”), so I expanded this section

Unable to find a datasheet. It may not be feasible to locate one, since Winsson offers custom transformer manufacturing. This 62–1225 transformer was likely specifically made for the Tripp Lite BC350. Indeed, it is clear from SPW Industrial and csilverblue2008's photos this transformer was salvaged from an UPS, notice the extra battery wire leads?

Testing coil resistance

Anyways, without a datasheet, we are on our own to find this transformer’s specifications. To do so, first watched Electronics: Transformers 1965 United States Air Force Training Film. From the video it is clear this is a power transformer, with a metal core. A closer look at the transformer I salvaged:

Testing with an ohmmeter, measured the following resistances:

• White to black: 5 Ω to 5.1 Ω
• Red (lower/small) to fat red (top/battery): 0.2 Ω to 0.3 Ω

Red to brown and red to yellow also measure 0.2 Ω. As does brown to yellow, fat red to brown, fat red to yellow, they are all 0.2 Ω. The air force training film says you can measure coil resistances to calculate the turn ratio, but comments in Hackaday Retrotechtacular: Step Up and Get Your Transformer Training note that this is only valid of the same wire gauge is used for both coils. Assuming it is, this transformer has a ratio of 5/0.2 = 25:1, or accounting for inaccuracies in measurements: from 5.1/0.2 = 25.2:1 to 5/0.3 = 16.6':1. Is this accurate? As we’ll see below, no.

The fat red wire connects to the 12V battery positive terminal (the battery negative terminal connects to the circuit board, not directly to the transformer, of course, transformers can only transform AC). This suggests this power transformer is used to step-up the voltage to mains voltage. If so, white/black is the secondary coil (high output voltage), others are primary.

But why the (alleged) high ~25:1 ratio? Hint from Ken Shirriff’s blog:

You might wonder why the DC voltage inside the power supply is so much higher than the line voltage. The DC voltage is approximately sqrt(2) times the AC voltage, since the diode charges the capacitor to the peak of the AC signal. Thus, the input of 100 to 240 volts AC is converted to a DC voltage of 145 to 345 volts internally.

There may be additional losses or voltage regulation to causing the requirement for a higher step-up transformer ratio. At 25:1, only 120/25=4.8V input is needed to output mains voltage. Could this transformer be designed to transform the five volts common with low-powered logic circuits up to AC mains voltage?

To test our hypothesis, we need to disassemble.

E-block disassembly

How about taking it apart, to confirm the coil composition or reuse the copper coil wire for other purposes? Found this Instructable: Disassemble an E-block transformer, recommending a pair of needle-nose pliers and gumption.

Other approaches are to grind (useful for microwave oven transformers) or smash (from Jack the Scrapper). I opted for using a pair of pliers, a screwdriver, knife, and brute force.

First, it is easy enough to peel off the black tape on the front of the transf to reveal the copper winding:

There is more beyond the subsequent layer of yellow tape:

Now for the hard part: removing the silicon steel. This comes in laminated layers, stuck together by some kind of (RoHS-compliant) resin or glue. The basic strategy is to stick something underneath the layers to separate them, then pull apart outside of the transformer.

Easier said than done, but the first several layers are the most difficult. An X-acto knife was useful to get underneath the layers, then worked it at with a screwdriver. Since I was unable to pull out the first several layers, I just bent them back and forth and tore them off, making progress:

A strong pair of pliers can be used to bend the metal and help separate it, even starting multiple layers at once:

However the E-blocks are interlocked with each other, so they still have to be separated:

though sometimes you can get away with leaving some of the I-blocks on. A substantial amount of force is needed to start out, mangling the metal:

but eventually, the pieces can be separated more cleanly, revealing the namesake “E” shape of this E-block transformer:

Halfway through, it is all downhill from here:

When complete, you are awarded with a pile of metal from the laminated steel core:

Reusing the transformer steel

The laminated strips are made from electrical steel, which has this structure:

This metal is built specifically to increase magnetic permeability, so the magnetic field flows through it into the other coil.

But what can we do with it? One possibility is reusing for arts and crafts:

The Coils

Back to the transformer, now that it has been disassembled the construction is clear:

Both red wires (thin and thick) connect to the same contact! The 0.2 Ω resistance measured was not the coil resistance, but the wire resistance itself. Red, brown, and yellow connect to thick wires on the secondary coil.

Black and white (measured resistance between them was 5 Ω) connect to another coil with smaller wires, another reason why the resistance ratio does not correspond to the turns ratio:

Unwinding Outer Coils

There are 4 magnet wires soldered onto the red lead, 2 on the brown lead, and 2 on yellow. Notes while unwinding:

Yellow #2 coil wire, unwound 15 turns then got stuck, then 25 total to red #3
Red #1, unwound 1/4 turn then stuck (reversed), then 23 to
Red #2, unwound 1 turn, threaded between brown #1/2 and red #3/4, cut then able to unwind 2 turns, and then 24 until stuck, then 27 to brown #2
Yellow #1, unwound 13 turns stuck, then 26 to

I may have miscounted some of the turns, it appears there are ~25.

Sometimes the windings reverse, and other wires have to be unwound first. This is visible as a kink:

Down to the last windings, the other coil starts to show through.

Finished removing the high-voltage coils, there are a total of 4:

The red leads connect to all four of these coils, and the brown and yellow connect to two each.

What good is this wire? Anything magnet wire is good for, because that’s what it is. The polymer film/enamel insulation can be stripped by a knife or possibly a hot soldering iron, and the wire could be used whatever you use wire for, or for other purposes such as magnets, inductors, transformers, or recycling.

Here’s some other thinner aluminum magnet wire I had for comparison:

Inner Coil

Finally we arrive at the last innermost coil:

It is made of finer gauge wire and heavily insulated:

removing the outer insulation reveals:

then there is another paper insulation layer. Carefully removing it reveals the nice and shiny copper wire:

This could be unwrapped to calculate the turns ratio, but I kinda like this coil as-is, don’t want to ruin it. Perhaps could be useful as an electromagnet?

Schematic symbol

Could be drawn something like this (excuse the crude image):

Conclusions

Although I wasn’t able to find a good use for this transformer yet, I now know what it is made of and how each of the wires connect. A final image showing an unopened transformer beside its broken-down components: