Question About Audio Transformer Uses - RG?

[Paul Marossy]

I was looking thru the Mouser catalog at audio transformers and noticed that on the 42TM*** series, there are four typical uses listed - coupling, input, driver and interstage. Can anyone decipher these terms for me?

[Mark Hammer]

I was looking thru the Mouser catalog at audio transformers and noticed that on the 42TM*** series, there are four typical uses listed - coupling, input, driver and interstage. Can anyone decipher these terms for me?

Coupling = permitting signal flow without being directly connected; much like "isolation"
Input = usually higher output than input impedance; for matching low-impedance device like 150ohm mics to later stages.
Driver = Usually higher input than output impedance
Interstage = Usually all those other variations of impedance that might exist which involve matching two different impedance ranges within a circuit.

I *think* I'm right about that, but others will certainly chime in and tell you and I where I'm wrong.

[Paul Marossy]

Thanks! That gives me a little better idea about what that terminology means.

[Paul Marossy]

OK, now here's a stoopid question:

If you were building a DI box and used a 10K:10K, it wouldn't work, right? The ones I have seen have use a 25K:600ohm transformer. I assume that the 10K:10K transformer is way too high of an output impedance to drive the usual 600 ohm XLR load?

[R.G.]

A bit about transformers.

Transformers do not have an impedance as such. They have an impedance ratio. Considerations about the transformer's internal real impedances come from the target impedances, but those are purely external considerations.

In the case of the 10K:10K you mention, it's really a 1:1 ratio transformer which functions best at about 10K loading on the secondary, at which point it reflects 10K load to the primary. If you load the secondary with 8K, the primary sees 8K, and if you load the secondary with 47K, the primary sees 47K.

So if you hook the secondary to 600 ohms, the primary sees 600 ohms. Or would if the transformer itself was perfect.

And it's here that the "Oh, yeah, it's not perfect." stuff comes in. Inputs to mixing boards which are intended to be 600 ohm impedances are usually pretty close to that. So used as a DI, the 10K:10K would reflect 600 ohms through the transformer-only part.

The imperfections of the 10K:10K as opposed to a 600:600 are that for the same power level (which equals the same core size):
- the primary has more turns of finer wire, as does the secondary
- that runs the primary and secondary wire resistances up
- that also increases the leakage inductance, a lot.
- that also increases the primary inductance a lot.

Compared to the 600:600 in a 600 ohm loading circuit, the 10K:10K will have more resistive losses of signal voltage, increased bass response and decreased treble response, assuming the other things about the transformers are equally well worked out.

So yes, a 10K:10K will work as a transformer input on a DI. It will have a lot of insertion losses, coming from its primary loading down whatever drives it on the primary, and also from its wire resistance eating a lot of signal voltage from the higher currents demanded by the secondary 600 ohm load. It will also have poor treble response. The loading on a signal right out of a guitar is likely to be disastrously much.

You can pep this up a bit by putting a buffer in front of it. The buffer would presumably not load down whatever was driving the transformer, and would be able to drive the 600 ohms plus wire resistances that the primary would look like to the buffer. Now you get good low frequency response, better (ish) high frequency response, and only the wire resistances versus the board's 600 ohm loading to reduce the signal. If your buffer has some gain, you might even make up for the resistor losses.

But it's a hard, not very recommended way to do a DI.

[Paul Marossy]

Huh, thanks for the insight on that RG. I'm trying to build a stereo breakout box with a built-in direct box for use with a stereo output Parker Fly guitar (piezos on bridge and magnetic, and can be used simultaneously with two different amps), and I was going off of a incorrect memory that I needed a 10K:10K transformer. So, when I was testing it last night, it wasn't working like I thought it would - the outputs were acting really quirky, and the only thing I could see that I did wrong was to use the wrong audio transformer.

The other thing about the Parker Fly is that it has a "smart jack", meaning that it know if you plug a mono cord into it, it will sum the piezo and magnetic signals or if you plug a stereo cord into it, then it sends seperate outputs to two different amps via a stereo cord. The quirky part of my testing last night is that it was acting more like a summed signal than sending two signals, really something in between when I tried the XLR out option. I'm not sure if that can be attributed to the audio transformer or not, but that's the only variable as the splitter box works fine if I run the mono output into a DI box that I have. And, yes, I have checked out all the grounding and made sure that what needs to be isolated is.

[zachary vex]

Coupling = permitting signal flow without being directly connected; much like "isolation"
Input = usually higher output than input impedance; for matching low-impedance device like 150ohm mics to later stages.
Driver = Usually higher input than output impedance
Interstage = Usually all those other variations of impedance that might exist which involve matching two different impedance ranges within a circuit.

I *think* I'm right about that, but others will certainly chime in and tell you and I where I'm wrong.

The particular series mentioned seems to have high input impedances on the primaries of the "Input" transformers.  You're right, though, in the case of mic transformers, it is the reverse.  In the case of DI transformers you'd need high on the primary side and low on the output, but I don't know if you refer to one of those as an "Input" transformer when it's more of an impedance matching transformer.

[Paul Marossy]

In the case of DI transformers you'd need high on the primary side and low on the output, but I don't know if you refer to one of those as an "Input" transformer when it's more of an impedance matching transformer.

Agreed. I think it's just how Mouser categorizes of them.

[Mark Hammer]

Yeah, it is always going to depend on the context where you use them, as I understand it.  For instance, the typical use for a 1200R:8R transformer is going to be between a transistor of op-amp and a speaker or perhaps something filling the role of a speaker, like a reverb pan.  The semiconductor can't comfortably drive the speaker/transducer directly, so primary is high and secondary is low in that particular application.  But what if it's one of those wacky intercoms where the speaker can serve double duty as the microphone?  There, you need a winding that is quite low, so as to accommodate the speaker, and a secondary that is higher so as to accommodate the input stage.

I think this illustrates the principle that transformers come in assorted primary-to-secondary ratios anticipated by "classic" scenarios, and we and manufacturers classify them according to the scenarios where those ratios are common.  They do not HAVE to be classified that way, however.  The application is what the application is, and we simply select a transformer that suits the application.

[chunks717]

I was just going to post a question about xformer coupling in to/ out of a digital recorder chip. I thought it may help with some of the static artifact issues, and it did.......(or is that just high end loss...... )

Somewhere along the way I bumped into a piece of blanket advice involving 'transformer into digital' safety which may have been as simple as a limiter diode or so across the secondary.............

Any thoughts or ringing bells?