Voltage ratings on power transformers

[Robert Kesh]

If I'm happy to put 240VAC into primaries listed as 110/115VAC, with the commensurate increase out the secondaries, I get a huge increase in my choice of power transformers. I'm wondering if this is safe and reliable.

Voltage ratings on transformers don't seem to be maximums, just an expression of their ratio. We have the VA as a useful maximum rating, and I obviously won't exceed this, or even exceed half of this. So I'm wondering how I can find out if I'm going over voltage.

The only rating I can find on datasheets that might be relevant is dielectric strength: which is often in the thousands of volts. Is this the rating I should be considering?

[DUG]

Originally Posted by Robert Kesh

If I'm happy to put 240VAC into primaries listed as 110/115AC, with the commensurate increase out the secondaries, I get a huge increase in my choice of power transformers. I'm wondering if this is safe and reliable.

Voltage ratings on transformers don't seem to be maximums, just an expression of their ratio. We have the VA as a useful maximum rating, and I obviously won't exceed this, or even exceed half of this. So I'm wondering how I can find out if I'm going over voltage.

The only rating I can find on datasheets that might be relevant is dielectric strength: which is often in the thousands of volts. Is this the rating I should be considering?

1. no, not safe nor reliable. It has to do with core flux density. When the core saturates, there will be no more flux per increase in voltage...0 inductance...just the wire...guess what happens...lots of current.

2. look for smoke or blown fuse or it may open cct due to too much current

3. no, that is between primary and secondary

[Robert Kesh]

Thanks, but I'm thinking for 2 that measuring power losses from primary to secondary might be less extreme.

[DF96]

Originally Posted by Robrt Kesh

The only rating I can find on datasheets that might be relevant is dielectric strength

You are not looking hard enough. Most mains transformer datasheets specify the primary voltage. You can exceed this by maybe 5-10%, but you should assume that some of this headroom will be used by mains voltage variations so your design should not exploit it very much. Doubling it will definitely not work - see post 2.

Not safe. Not reliable. If you are lucky you will blow fuses. If you are unlucky you could start a fire.

[Robert Kesh]

Originally Posted by DF96

You are not looking hard enough.

Of course I'm looking hard enough, see first post, I'm just not understanding, having believed the numbers to be more an expression of ratios than ratings.

[DF96]

Sorry, I was trying to inject some humour.

[macboy]

Flux density in the core will be proportional to the integral of the voltage waveform. If you double the voltage, you will get double the flux density. If you decrease the frequency, you increase the flux density. For both of these reasons, using a 60 Hz, 110/120 V transformer with 220/240 V 50 Hz is a terrible idea. You will saturate the core and things will not go well for you to say the least.

Many transformers have dual primaries, which are paralleled for 120 V operation and put in series for 240 V operation. As long as the transformer is also rated (or derated) for 50 Hz, then you are good to go.

[GoatGuy]

Macboy, ... the integral of the voltage is almost correct! The actual answer is that the magnetic flux of the core depends exactly on current flow (and not voltage!) ... and that for an idealized inductor (having no resistance, no upper limit of amperage), that I = L · &integ; V(t) dt ... the integral of voltage over time times the inductance.

Note: I say "ideal inductor". Real-world inductors have both resistance (significant) and capacitance (usually small, but important with increased frequency), as well as flux-saturation (max current) limits. Above Imax, additional current cannot increase flux, which to current flow looks like a huge drop in inductance. Current flow can rapidly rise, causing winding heating, causing transformer failure.

SO the moral of the story is... do not use 110 volt transformers on 220 volt mains. Period. The minority which are so generously designed to be running at half, or a third their maximum flux (which could take the 220 volt overvoltage) are so rare, as to not be worth looking for them. Competent magnetics designers design transformers to work safely up to 25% overvoltage, not 100%.

THE OPPOSITE THOUGH ... running a 220 primary at 110 volts... is perfectly safe, and an entirely reasonable thing for an experimenter to do, to obtain half-output when rummaging through the parts box for a quick experimental setup. If "it works", then an appropriately sized replacement transformer can be obtained later.

GoatGuy

[Robert Kesh]

Originally Posted by GoatGuy

Macboy, ... the integral of the voltage is almost correct! The actual answer is that the magnetic flux of the core depends exactly on current flow (and not voltage!) ... and that for an idealized inductor (having no resistance, no upper limit of amperage), that I = L · &integ; V(t) dt ... the integral of voltage over time times the inductance.

Current was what I thought, which is why I wondered if running well under current would make it all roses. But I guess not.

[Elvee]

Originally Posted by Robert Kesh

Current was what I thought, which is why I wondered if running well under current would make it all roses. But I guess not.

A way to picture it is to imagine the primary is in parallel with a varistor having an AC rating similar to the mains voltage.
Whether you draw current or not changes nothing: the varistor will conduct when its voltage is exceeded.
The non-linear behavior of the varistor is similar to that of the core (actually there is an integral difference between the two, but let's not obscure the picture)