Understanding step down transformers

Nov 24, 2010 10 Replies

I could use some help in understanding why some transformers heat up even when nothing is plugged into them.



Ex. Heavy duty shredder - heats up big time when shredder isn't attached (15V and 20 watts) - so I unplug it after each use



Have a great day, Andy



formatting link
formatting link


My guess; A cheap transformer can be made using less copper wire, that causes an increase in the magnetizing current, producing extra heat, even though it has no load.

"Mint"

** The answer is the copper and iron losses due to magnetising current.

There is always some current flow in the primary winding of a transformer, even with no load. This is called the "magnetising current" - naturally there is also an amount power lost as heat due to this current aka " I squared R " loss.

The designer of a transformer can allow the magnetising current level to be any value he or she choses - including one that makes the unit get quite hot.

The advantage is that the design will be smaller and cheaper if this is done.

Usually, you only find it done when the unit will be under full load at all times - eg the transformer inside a microwave oven.

.... Phil

Bad quality, or when there are some shorted windings.

Because the magnetization that occurs in a transformer, and which changes sign with each AC half-cycle, is MAXIMUM when there is no current drawn from the secondary winding. Even in the absence of resistive power loss (current in the windings) that changing magnetic field causes heat generation in the iron core.

There's also, of course, at least a little bit of current in the windings; the magnetization does draw power in and of itself, though the major part of the energy of magnetization just comes back in the next phase of the AC cycle.

- so I unplug it after each use

Interesting.

I will just put the transformer on a power strip.

Andy

(...)

I wonder if this circuit or something similar would limit power to your shredder until it is turned on...?

Otherwise it dissipates a small amount of 'pilot power' into your transformer to sense when the your switch is turned on.

/------+-----------------\ | | | ---- o\ /-+-- | Relay nnnn o | \-. | | | | --- )|| | | | | --- )|| | Current uuuu | | /-' | Sense ---- | \-+-- | Transformer nnnn | | Resonant | | | | | Tank | | | | | | Line ------/ \-+------/ | /------+ Load | | Transformer .-. | on plug strip Neon ( X ) \-. ,- '-' )|( | )|( Neutral -----------------------------+--------' '-

Current sensor powers plug strip only after load is turned on

AACircuit

formatting link

--Winston

I don't see that. How do you get to your statement from examining a hysteresis loop? The current in the next cycle has to flip the magnetic domains right around again. That would entail a net loss of energy, not a gain and also cause heating of the core (adding to that caused by induced eddy currents)

It's easy; the area enclolsed by that loop is exactly the quantity of energy lost to heat in each AC magnetization cycle. The closure of the loop means that the cycle also periodically removes the memory of the past, which requires that the magnetization energy be removed from its storage as a net magnetic field in the core. Then replaced, in a different orientation... and so on.

The remagnetization is caused by current in the wiring, which generates heat by Ohmic resistance. Even though the magnetic energy (total energy of magnetization) isn't really lost, there are resistive losses in the wires that deliver that energy and remove it.

There are also eddy currents in the core which contribute to the heating of the transformer.

True; the effect is also to add a second H-induction to the external circuit, so that the B-H curve tracer (really just a voltage/current measurement of the transformer core primary) will show a bit of open area due to the external-H being phase shifted from the sum of external-H and eddycurrent-H. The only way to be sure the loop's eye is a hysteresis effect is to look at it in the low-frequency excitation limit. For accurate measurements, I've seen this done at circa 1 Hz.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required