Inability to "reverse" (backfeed/step-up vs step-down) small transformer <?>

Hello - I'm a bit baffled by a no-load test on swapping the input/output on a small power transformer: 'Been doing electronics for years, but never had occasion to feed the secondary, versus primary, to get (modest plate current) HV B+ from surplus former transistor amp xformers...

In slowing ramping up the input voltage to this ~2Ampere 24vac xformer (e.g. 5:1 ratio), I was not able to get even 20vac into it - e.g. 100vac out - before it began overheating; again...no-load?!?

I regularly see 120->12->120 small ampt PS's ...but, are there some (unusual?) windings on standard power transformers that would preclude pushing them backwards - e.g. heating up on their own eddy currents? Thanks in advance for any practical experience and findings...!
 
The convention for a transformer windings is just this: a convention. In real world, any winding may be the primary (only one at a time) while other(s) may be the secondary(ies).

But there may be some differences. For example, a 220V to 24V AC transformer, when feeding the 24V winding may give 200VAC in the other winding, because sometimes the "primary" has a bit less turns and the "secondary" a bit more in order to compensate losses in winding resistances.
 
  • Like
Reactions: Haze Head
The main problem is the regulation: for example, a 230V-->24V 10VA transformer is in fact a 230V -->30V, no load. With a 10VA load, it will deliver its nominal 24V.
It also works in the opposite direction: with 24V at the secondary, it will deliver 190V on the primary, under no-load conditions. When loaded, it will deliver 170V or even less
 
To clarify, I had every reasonable expectation to slowly increase the voltage to the (as-designed) secondary, with the primary delivering 5x the applied ac voltage from the variac. I was shocked to overheating at less than half of line voltage - absent any load. I would still appreciate a technologically thorough explanation of why (at SUBSTANTIALLY less than any hipot test voltage) this occurred... THX
 
Or the transformer is really bad designed/built. The idea of using a transformer to increase heater voltages for low current stages or for bias has been in use from decades ago. Here is one taken from a ARRL Handbook 1954 in Spanish.
 

Attachments

  • 1671196402927256093529552763412.jpg
    1671196402927256093529552763412.jpg
    362.3 KB · Views: 55
Last edited:
Good thoughts - but with the source being a simple variac following an iso-trans, I'd be at a loss where rectification of any sort could occur. I'll try a few more small xformers just to satisfy my curiosity (and for fun, monitor on the scope <'good call>). 'Might have a chance to do some scholarly research this weekend. Hmmmm