Thread: Line level iso transformer

I managed to snag a couple of pretty decent looking 600:600 ohm output transformers from a piece of equipment and I would like to make something along the lines of a Radial Twin-ISO. I've go the box, connectors, etc. all figured out, I'm just trying to figure out the wiring. I wanted to keep it simple.

The schematic on the Jensen site for their iso boxes shows some resistors and caps in parallel with the transformer, but I am not entirely sure what these are for. Some sort of impedance correction?

Is it as simple as connecting the input and output leads of the transformer to a set of XLR's and calling it a day, or should there be some other components?

FWIW, these are not the Jensen's. Just some generic transformers from a piece of audio equipment.

Transformers are simple devices in concept, but when you get specific about very good performance, how they are driven and how they are terminated matters to the marginal performance you will realize. Deane Jensen (RIP) was an early pioneer in using computers to help him measure and design transformers and he was good about characterizing his transformers wrt how they needed to be driven and terminated for optimal transfer function.

The Jensen data is not likely to apply to your XYZ transformers.

If you have time to burn, and a relatively stable application, you can probably make some measurements and perhaps dial in a little better performance, for general purpose 1:1 ground lift band aid for emergency use, perhaps don't worry too much about it.

It might be worthwhile bench testing to determine their LF saturation headroom, to advise you about when and where not to use them. If you want to attempt some generic dial in for the high end response, most line inputs should be 10k Ohm or greater, and most modern professional outputs << 600 ohms, but I'm not sure this is a good use of your time. (Perhaps push a 500Hz or 1kHz square wave through one and look for overshoot.)

JR

[edit- if you pulled these from a working unit, look at how they were driven or terminated in that application as a starting point. /edit]

Originally Posted by John Roberts

Transformers are simple devices in concept, but when you get specific about very good performance, how they are driven and how they are terminated matters to the marginal performance you will realize. Deane Jensen (RIP) was an early pioneer in using computers to help him measure and design transformers and he was good about characterizing his transformers wrt how they needed to be driven and terminated for optimal transfer function.

The Jensen data is not likely to apply to your XYZ transformers.

If you have time to burn, and a relatively stable application, you can probably make some measurements and perhaps dial in a little better performance, for general purpose 1:1 ground lift band aid for emergency use, perhaps don't worry too much about it.

It might be worthwhile bench testing to determine their LF saturation headroom, to advise you about when and where not to use them. If you want to attempt some generic dial in for the high end response, most line inputs should be 10k Ohm or greater, and most modern professional outputs << 600 ohms, but I'm not sure this is a good use of your time. (Perhaps push a 500Hz or 1kHz square wave through one and look for overshoot.)

JR

[edit- if you pulled these from a working unit, look at how they were driven or terminated in that application as a starting point. /edit]

These came from an FSR audio silent audio switcher. The printing on the unit actually advertised it as having 600 ohm transformer iso outputs. Inside the unit, there were 2 identical transformers. They were hooked up "backwards" with respect to each other. By that I mean, they used the primaries and secondaries interchangeably for each channel, so from that I deduced that the primary and secondary must have the same impedance if it didn't matter which set of windings they used for input and output.

The output on both transformers went directly to the barrier strip to the +/- outputs. The audio switching circuit that fed the transformers was pretty simple, I seem to recall it had an op amp, a handful of passive components, and maybe a few small general purpose switching transistors. That's about all I can remember though.

I'm thinking I can probably just wire this right in with the XLR's without any associated circuitry and run some tests to check performance. Chances are this will be more of a band-aid in unique situations like interfacing with questionable hotel ballroom and church PA's, so I am not too worried about performance.