Quality testing output transformers.

[linguist]

What is the best proceedure for testing Audio ouput transformers, every second thing you seem to read mentions different ways.

When testing Audio output transformers is it best to sweep through the frequency range or test at frequency increments working up to the maximum frequency.

Also, some say to test with a sinewave & others say test with a squarewave.

Some seem to test with very low amplitude input signals compared to the actual working enviroment.

It all becomes some what confusing when you read so many different techniques.

I would have thought that at a minimum you would have to test with at least the same amplitude & frequencies as the output transformer will be subject to & transient test by sweeping though the frequency range with a similar signal to what the transformer will see, not static testing.
Maybe I am completely wrong also?


Could someone help me with an explanation of the correct way to look into this & what to look for regarding what you do NOT want to see when testing.

Cheers

 

[Boncuk]

I'd test the entire AF-equipment by connecting the suitable speakers to the transformer output and "read" the sound using an electret microphone with a wide band amplifier and a connected scope to both the sound source and the mic amplifier.

Boncuk

 

[linguist]

Hi Bonuk, thanks for the reply.

Yes, I get what your saying, but the trouble is the volume levels are very high with some amplifiers so a dummy load is required.
I thought of using a Microphone to record the sound output & then read it with a Spectrum anyliser but from zero to full volume is required--very loud.

How do the manufacturers test there audio output transformers, they must bench test somehow?

Cheers

 

[Nigel Goodwin]

I would imagine they are tested 'in-circuit', using a dummy load on the output of the amplifier - basically test the amplifier just as you would any other.

I wouldn't imagine tests on it's own would prove very much?, it's too influenced by how it's connected.

 

[linguist]

Thanks Nigel,

Yes, testing in circuit makes a lot of sense, thank you!

Cheers

 

[Gary B]

Hi Linguist,
You start out asking about testing just the transformer but then switch to the amp. Do I take it correctly that you are attempting to test/repair audio amplifiers and was just trying to eliminate the transformer as a source of the problem? If so, I have always been told to do all tests and alignments at normal power levels and do a straight forward signal trace. It started with Air Force Tech School and nobody has ever contradicted it.

If on the other hand, your intent is to characterize a transformer for a new design, I would suggest that either sweeping it or injecting pink noise and doing a spectrum analysis of the output would produce the same results. Just about as through would be the spot check method if your chosen frequencies are close enough together to produce a differentiable curve over the entire operating range. If there are no stray capacitances that produce a resonance point in the graph of the transfer function, three points would be plenty. The choice of method would be determined by what test equipment you have available.

 

[linguist]

Hi Gary B,

Yes, I switched to the amp basically because it was mentioned to test the output transformers in the circuit itself which is a reasonable suggestion.

I would prefer to bench test them, I am attempting to wind a few of my own & need a good reliable way of testing them against each other.
Basically looking for the best characteristics of a good quality sound output & saturation points etc etc.

I am open to suggestions, the in circuit idea has some good points about it being able to drive the transformers easily to full load.

I have a sweep generator, sine & squarewave output 10Hz to 20kHz, an Oscope, Spectrum analyiser.
Driving the transformers to full power would require me to amplify the signal so I guess the in circuit idea is maybe the best?

I am not sure what the spot test is?

Cheers

 

[Gary B]

I think a good bet would be to do low power design testing to characterize the device. Once it appears to be what you want, then run one or more full power tests. The only factor you should have to change at that point is the amount of metal in the core if the high power test does not follow the low power characterization. Of course, you could always just over-design the core. Am I correct in assuming that we are talking audio frequencies here and not something like an antenna coupler for a television transmitter?

I meant tests at selected spots along the operating bandwidth.

 

[grim]

you can do lots and lots of bench tests, and at the end of the day pick an 'inferior' design because it sounds better. the human ear being some what organic, you cannot rely on what looks good on a scope.

any performance testing needs to run the tranny at the power levels it's designed for, unless you are into the b#llsh#t specs some people quote, where they have a 100W amp and specify various parameters like frequency response and distortion at 100mW

Also what sort of amp is it, guitar or hi fi? as that makes a big difference.

It's also very possible to over specify a tranny. I design them for the biggest valve amp company in the UK and we regularly see people saying they replaced the trannies for something costing £100's and how much better they sound. firstly we know from back to back testing, they don't (kings new clothes) and secondly we have seen how a better top end can cause instability in the feedback. i guess it's like chucking a V8 into a fiesta. it will go faster, and then the tyres will explode.......

when you wind it, keep the windings tight and flat. no crossovers in the winding, and keep any solder joints to the outer wrap. i dread to think how much it would cost to make them yourself, once you buy a winding machine, varnish plant and of course materials.....

more sections you have the higher the top end. more lams you have the lower the bottom end. 25 years experience in one sentence............

 

[audioguru]

I think 50 years ago I saw an output transformer on an audio amplifier. I haven't seen one since. Good riddance.

 

[grim]

Solid state has it's place in the business. it's a cheap way for people to play back awesome guitar rifts recorded by people using proper amps

 

[linguist]

Thanks for the replies,

Gary B,

Yes Audio frequencies only.


grim,

The point you made about people having a 100W amp & specifying things at 100mW is why I have asked the questions on correct test proceedures, I also have found this to be rediculous.

I see what your saying about the human ear & how much time can be spent on output transformer design only to find out that the quick easy one sounds the best.

This is why I would like to check things out for myself & find suitable designs for my application.

The application is Guitar Valve amps, I have noticed that some transformers I have used sound like tin cans & others a lot better in the same circuit.
I would like to study this a bit to find what I am looking for.

I would like to get your thought's on output transformers for valve guitar amps & how they differ from Audio transformers.

Any help would be greatly appreciated.

Cheers

 

[chemelec]

Tube Amp, Audio Transformers Probably Differ Mostly In Input Impedance Values.

Plate Circuits in Tubes are HIGHER Impedance than Collector Circuits in Transistors.

For a Good Amp, Proper Impedance Matching is Important.
As well as the Ability of the transformer to work at Full Load without Saturation.

So Testing these transformers "IN the Amplifier Circuit", Also will reflect on Proper Matching at both low and high power outputs. (And can also show any Saturation problems.)

Its been about 25 years or so since I did any tube designs, but hope this helps.

 

[linguist]

Thanks chemelec,

Yes, every bit of information is of great help.

Is there a good site somewhere or information that I can read up on matching Impedances correctly for tube amp design.
As you say the plate Impedance is high.

Cheers

 

[chemelec]

Here is a Page from an OLD Hammond Catalogue, with Typical Audio Transformers and there Spec's.

Sorry, Most of my Other Tube Amplifier Info is LONG GONE!

 

[linguist]

Thanks chemelec,

The attachment is of use.

cheers