Transient power rating of a small (24VA) 50Hz pcb mounted potted transformer?

[max_torque]

Take a typical pcb mounted, sealed transformer something that looks like this:





What is a reasonable assumption for it's short term power capability?


Let say it has a nominal data sheet rating of 24v, 1amp output ie 24VA

1) I assume that is it's continuous rating in the hottest ambient it is rated for?
2) How much current could i expect to pull for say 30 seconds, 10 seconds or 5 seconds?


I have an application that mostly, requires about 500mA, but occasionally (think around once every 5 min or less) requires up to 2 amps for a max of 10 seconds. Is it foolish to assume the transformer could support twice it's rating for that short period with the large gaps in between?

(at the moment, i'm not worried about the voltage sag under those high current conditions, but obviously that is also a consideration.  Critical stuff is powered from a 5v supply via a switch mode dc/dc, so the output can drop a LONG way before any issues.

The duration of the high current is really just a bit too long to buffer with a large amount of capacitance, but some capacitance will obviously help level the load a bit.

Anyone have any experience in establishing these sorts of ratings?   

[Le_Bassiste]

best way is to call manufacturer directly and discuss with him the "mission profile" of your application. that is the safest way to assure that your transformer is up to your requirements.

short overload cycles are basically buffered by the mass of the transformer, so you could guestimate something like specific heat capacity of the transformer (iron and copper, the copper doing the extra heat which is then dissipated into the iron core) and the extra energy that you apply to the windings (the extra I^2 *Rwire * overload time) to figure out the actual temperature rise of the transformer. but, as i said, that's just estimated guessing. won't save your ass if you get in trouble.

[Kleinstein]

For a transformer as small as in the picture (1VA), they are usually essentially short circuit poof and the current would be limited mainly by the ohmic resistance of the windings (not just the secondary, but also the primary - so the effective resistance is more like twice the secondary, mayb a little less for the small ones).

With larger transformers they are usually limited by the internal temperature. The time constant for the transformers is usually quite large - more like minutes.  For a short times a higher than nominal load and thus loss is possible, especially if the transformer is not very hot already.
So if normally 500 mA are used and occasionally 1 or 2 A, the transformer should be rated for well more than 500 mA (e.g. 700 mA).
At twice the rated current the power loss is about 4 times as high. Still this can be OK for 10s of seconds, and more for a large one.

For the current rating one should also consider that the simple rectifier and filter  circuit causes an AC current that is considerably (e.g. a factor of 1.5 for small transformer - up to 2 for a large one)  higher than the DC current.

[Benta]

24 VA PCB transformers are available as short-circuit rated.

[T3sl4co1l]

Seconds at a time is no problem.  There's lots of metal inside there.

I would be more concerned that you won't be able to draw nearly as much power as you want!  These transformers usually have quite high impedances, sometimes being short circuit tolerant as suggested above ("impedance protected").

For example, the "24V 1A" transformer might be 36VAC open circuit, 24VAC at nominal 1A load, and therefore 3A short circuit.  Your 2A load would draw it down to 12V, delivering the same power as at 24V 1A!

This may be an overly pessimistic example -- I don't remember what's typical.  I don't think you're likely to see a fault current of >=10 times rated current though.

You may consider:
- A conventional ("shell") type construction, which has lower impedance (good regulation).  If it's not potted, mitigate that some other way (IP66/7 enclosure?).
- SMPS module (assuming DC output is needed).  Mitigate EMC considerations as needed.
- Compensation capacitors, to improve the transformer's regulation.  This assumes leakage inductance is a dominant limiter of output power.  I would guess this could spare a couple more watts, but consider this an act of desperation.

Tim

[schmitt trigger]

So that is what "impedance protected" means.

Thanks for clarifying the term. 
I had seen it, mostly as applied to small electrical motors, but wasn't sure of its meaning.

[T3sl4co1l]

Indeed, the same applies to motors, which are merely rotating transformers.   They tend to have poorer coupling though (e.g., shaded pole type, where the core wraps around the rotor -- a leakage path), so the impedance is mostly inductive in that case.  Same mechanism applies though!

I think for transformers, it's mostly the huge primary resistance (100s to kohms?), so they still get hot under fault conditions.  You've probably seen a somewhat melty wall wart before.  Well, UL doesn't much care if it melts, or too hot to touch, just that it doesn't burst into flames...  Well, they do care if it's too hot to touch, but that's another matter I suppose...

Tim

[max_torque]

Ok interesting thanks all.  I looks like the best option is to test a few transformers and see what the output looks like vs load, and to perhaps make a stab at  measuring there thermal stability (self heating) under various loads!  I won't be able to recreate the actual installed environment, but i can at least look ratiometrically at the thermal performance etc

[soldar]

I would be more concerned that you won't be able to draw nearly as much power as you want!  These transformers usually have quite high impedances, sometimes being short circuit tolerant as suggested above ("impedance protected").

I agree. I would put a zener diode just after the rectifier bridge and the transformer resistance was high enough that a half watt diode could handle the peaks on its own. Then someone tried to improve my design by putting a much bigger transformer and it was blowing zeners like crazy.

I have quite a few of these transformers in my junk box ready to be used in future projects. Unfortunately they flow into the junk much faster than I get to use them. It seems traditional PSU's with transformer, bridge and filter cap are going the way of the dodo bird. It is easier to provide a wart SMPS.

Another thing about these small transformers is that people really do not understand what X amps means. It is not X amps DC after rectified and filtered. It is X amps AC and if you have a rectifier and cap with very poor power factor that's your problem.

Indeed, the same applies to motors, which are merely rotating transformers.    

Indeed, they transform electrical energy into mechanical energy.

[soldar]

I just got to use one of my tiny blue transformers. I wanted to monitor a 230 Vac doorbell and close a contact when it was activated. I could have used a direct 230 Vac relay but I just prefer an extra degree of separation between 230 V mains and delicate electronics. On the pins of a relay they are just too close for comfort. So I put the transformer in parallel with the doorbell and the secondary is rectified and activates a 12 Vdc relay.

I guess I could substitute a 12V SMPS wart if I had to start from zero but I already have the tiny transformer, diodes, etc.

The ones I have are generally good up to about 100 mA and are good for simple battery chargers and other simple things like that.