Power amp for measuring coils

[xaccto]

I want to build a power amp that is driven from a frequency generator
1-300kHz. sine and square wave.
The output impedance of the freq. generator is too high to do anything
useful with it for these intentions.

The purpose is to drive various inductors and transformers I experiment with
in order to measure their parameters, determine resonant frequencies, etc,etc
I'll be observing the waveform with an oscilloscope.

So sometimes the amp would have to drive into virtually a short circuit.
Haven't really gotten anywhere with ideas from just googling various terms
so thought to ask here, if anyone done similar or got some advice ?
I think
the amp should be class A
Should it use mosfet or bjt ?
How should the transistors be protected ?
bandwidth to 0.5MHz ?

some of the switching transformers I experimentally built are intended for
12-20V up to 10amp,
so what sort of power would I need to drive them in a measuring setup
surely I shouldn't have to use 100watts ? or do I ? to obtain the parameters
at the specification the windings will be running at ?

Thanks

 

[chuckey]

I would look for a ferrite transformer, that can handle the power (or a substantial part of it) and the bandwidth. Thirty years ago you could get ferrite cores that were flat 50KHz to 30MHz. They worked just like transformers should, 1 turn for the 50 Ohm circuit, 4 turns for the 800 collector circuit. So you need similar but with a bit more permeability, wind up some (6?) with 4 turns, carefully labelling start and finish of the windings, then thread them all on to a piece of 16SWG wire which is your high current output. Connect all the primaries in series. So now you have a 8:1 (64:1 impedance) transformer, try it again with your signal generator. From these results, you could determine where you should go.
Unless you are after the absolute power or voltage ratings of a device, your results from the signal generator should be scalable, i.e. 1mA into your coil gives 100mV across it, 1A in should give 100V.
Type of transistor? mosfet should be easier, use a current limited PSU to protect PA and catch diodes to protect against against any back-emf.
Frank

 

[FvM]

Most parameters except losses or core satuaration can be determined in small signal operation, e.g. with a standard LCR meter. Are you sure, that you'll need a power amp for your measurements?

On the other hand, it's difficult to specify a power level without knowing the exact measurement conditions and transformer data. The intended frequency range may be mostly achieved by state-of-the-art power audio amplifiers (linear class AB, not switching type).

 

[xaccto]

Thanks chuckey. That shoulda been an obvious solution....though I would be
looking for results in very low voltage levels, and noise. I might try it. Perhaps wind some toroid for that.

Fvm, yeah I had a moment thought of perhaps a audio amp design, so since you
say so i'll look look for one..don't need a complex high fidelity one though

I would use it on all kinds of inductors so hard to be too specific so it needs to
cope with it.