Transformer regulation

[dhaen]

I was prompted to thought by this thread:
transformer in parallel - help
but did not want to pollute that thread.
What if one connects 2 slightly different secondaries in parallel?
AFAIK current will flow from the higher voltage one to the lower one, causing heat to be generated within the transformer.
The off-load voltage will be somewhere inbetween the 2 secondary voltages.
What happens when we apply a gradually increasing loading?
Current is taken initially only from the 'high" secondary, until it's voltage falls to the same as the "low" secondary. By that time, there will be no current flowing between secondarys, since no PD exists.
As the load is increased, current starts to be drawn from both secondaries.

Net result: Improved regulation (at the expense of heat).

Am I onto something?
Or mistaken?
Or, am I re-inventing the wheel?

[wintermute]

Interesting idea.... I have a second transformer that I have been wondering If I can use to beef up my PS (I had Thought about using one transformer for the +ve rail and the other for the -ve rail wiring each transformers two windings in parallel for extra current), didn't really want to use one for each channel as that would require extra PS caps...... I worried about wiring them in parallel as you have suggested, because I didn't really know what effects it would have.

One thing, though. What would happen if the two transformers were paralleled but out of phase? Would they cancel out giving zero volts? and if so would that be like having a short? I mean where would the energy go

hmmmmm I should have followed the link.... Centauri answered my question about phase in the original thread, I guess the interesting thing there is, how do you make sure they are wired in phase without a smoke test

Tony.

[Richard C]

To establish correct phasing of the secondaries: connect one end of each secondary together and measure the voltage between the free ends.

If they are in phase there will be very little voltage between the two.

[Sch3mat1c]

Looks like a simple school problem to me

First, consider each winding as a perfect voltage source with series resistances Ri1 and Ri2. (AC or DC, it doesn't matter.) The ends of these are connected together and this node is loaded through RL. With a mismatch in the voltage sources, you get a current (V2-V1)/(Ri1+Ri2). This flows from the higher to the lower voltage. Now decrease RL until it draws an appreciable current: it will cause more flow from the higher source, and reduce the current flowing *into* the lower. With a heavier load, there will be zero current into the lower voltage; any more and both will now be supplying current to the load.

It's easy to see why paralleling transformers is a questionable practice!

Tim

[dhaen]

Yes, most of us know what happens if secondaries are parallelled out of phase, or of very different voltages: Smoke!

My original question is whether parallelled secondaries of slightly differing voltages can actually aid regulation.

[officeboy]

Why not just connect after rectification?
The diodes should keep the power where it’s supposed to be (out of the weaker transformer).