Usage UPS transformer "backwards" (step-down)

[MarkL]

The two rows of MOSFETs are implementing a half bridge.  They connect one side of the winding to either battery+ or ground.  The other set does the same thing to implement a full bridge to drive the transformer winding.  The winding sees battery+, 0V, or battery-.

If you have a dual channel scope you can look at each of side of the MOSFET chain and the timing relationship to eachother.  And you can use the subtract function on the scope to see what the transformer is seeing across the winding.

Many (all?) cheap UPSes have a "modified sine wave" output.  Below is a screen shot of one.  Those are +/-150V peaks, 0V in the middle marked by the "+".

[Cliff Matthews]

OK, I got it.. At first, for each x-former lead I saw two rows of 5 FET's (page 3) each run by dissimilar drive circuits, and well... the amp guy in my brain went linear (don't know why, but I did see 2 invertors driving one and 1 on the other, but the old elevator didn't go up to the top..) 

[npelov]

It's good that Cliff Matthews did the tests, but every transformer is going to be different. And mine is from an unknown brand. The video that mcinque posted made me think about measuring the inductance and calculating the impedance at 50Hz to see the current at no load. Yes, he probably thought I would connect high voltage at the low voltage side (which is secondary in a regular transformer). No, I wanted to connect line voltage at the HV side.

And yes, it's probably a low frequency transformer. I've seen high frequency transformers from big power supplies, that's not one of them. It's quite heavy.
Many mentioned that usually the transformer is used backwards to charge the  battery. That's probably true. I didn't see second power supply. But I don't have the PCB... it's long gone.
Well the transformer is berried in the basement (but not lost - it's described in a catalog) until I have a purpose for it. Usually you would want linear power supply if you want low noise. But these days it'll be hard to get anything clean from the power line. So to get low noise out of a transformer + linear power supply you need to filter the power line. To me it seams the same as putting few LC filters on a switching power supply. I've tried to filter SMPS and one LC filter is enough to bring it down to 1-3 mV (rms). Putting one more makes it so low that I can't reliably measure it even with short ground spring on the probe.

But I can't think of a use for it. Why would I want to have higher currents from a line transformer. It's inefficient for today's standards and this one is probably even worse. But I couldn't just throw it away. Maybe I can get the copper wire from the low voltage side and use it for some power inductors.