Switch wiring

[groselicain]

Brace yourselves, folks—this may be stupid on my part.

This isn't a stompbox, but I'd much rather pose my question here than on other forums I frequent. The majority of the helpful posters are, well... just that!

I build simple transformer units for myself and friends to use when mixing records. Here's a very simple diagram of what we're talking about:


This is all fine and good, but we've gotten tired of plugging and unplugging these things. Initially, I thought I would just add a switch to bypass the transformer, but then someone suggested that it would be nice to add a switch that would allow us to select different primaries and secondaries—up 'til now, we've simply used 600:600. At this point, I realized I was way in over my head. I'm as novice as a novice can be with electricity, and even worse with switches.

What switch would I need to use, and how would I wire it, if I wanted to have a bypassed signal, and two other options—e.g. 600:600, 150:600—in this sort of setup?

Thank you so, so much for your help. Thank you even more for the inspiration you guys have given over the years.

[PRR]

Welcome.

> What switch would I need

Very complicated.

I don't want to be that un-helpful person you avoid, but.... what are you doing?? Why even have "600" in a modern studio? (10K:10K makes more sense.) (Unless you are deliberately stressing output stages.)

Are you breaking ground loops?

[groselicain]

PRR,

     You're not the same PRR from group-diy, are you? Getting back to the questions you asked, I understand that 10k is the standard line input impedance. 600:600 was what Farmelo was writing about in TapeOp all those years ago¹, so i think most of the studio cats like me, having no EE background, just followed suit. Breaking ground loops happens, I suppose, but that's not the intended purpose.

1. http://www.farmelorecording.com/in-the-press/using-transformers-to-transform-audio/

[groselicain]

Here's a friendly bump.

[vigilante397]

Any particular reason a run-of-the-mill 4PDT wouldn't work? Pin 1 of your XLR connection is going to be grounded, put pins 2 and 3 on the middle-left lugs of your 4PDT switch, then the primaries of the first transformer on the top-left and primaries of the second transformer on the bottom left. Secondaries of the first transformer on the top-right lugs, secondaries of the second transformer on the bottom-right, then pins 2 and 3 of your XLR output will go on the middle-right.

A picture is worth a thousand words, but I'm in class right now and can't really draw up a picture.

[thermionix]

600:600 vs 10K:10K?  I thought transformers were all about ratios, and both of those are 1:1.

I guess I have a lot more to learn about transformers.

[wavley]

Personally, I would just get a patch bay for bypassing, that way any of the primary and secondary is easier switching that may or may not give you anything good, if it were me I would just patch it and maybe have switchable attenuators, no reason to overthink this.

He probably talked about the use of 600 ohm transformers in that article because that's a common value for stuff salvaged from old gear or being made to clone old gear, just my guess.  I remember that article, it wasn't that long ago was it?  Maybe it's just me getting old.

[groselicain]

Any particular reason a run-of-the-mill 4PDT wouldn't work?

That was my thought as well, but then I got confused when the option of multiple ratios was brought up.

A picture is worth a thousand words, but I'm in class right now and can't really draw up a picture.

If that time ever comes, I would be forever in your debt!

600:600 vs 10K:10K?  I thought transformers were all about ratios, and both of those are 1:1.

I guess I have a lot more to learn about transformers.

Same here, on both counts. I defer to PRR, though, because—at least if he's the same PRR from group-diy—I assume he's got his reasons.

Personally, I would just get a patch bay for bypassing...

We all have patchbays at this point. The idea is to pass audio from say, a stereo mix, through the transformers. The switch is far faster for A-Bing, which is much more helpful when we're doing this stuff. It's the same reason any compressor–after the sixties, that is—worth its salt these days has a bypass for checking.

[vigilante397]

A picture is worth a thousand words, but I'm in class right now and can't really draw up a picture.

If that time ever comes, I would be forever in your debt!

Deal

[groselicain]

Vigilante, you are a sweetheart of the highest order! Thank you so, so much!

[PRR]

For short lines and cheap gain, the optimum is low-Z sources and hi-Z loads.

Modern stuff runs 100r out and >10K in.

When gain was eXpensive (one tube a week's pay), "matching" reduced the amount of power gain needed.

For very long lines (miles), line resistance and reactance define a "line impedance" which will give optimum power delivery. This is non-issue in bedroom studios, and even in large studios. However a LOT of music audio is derived from telephony and specifically the transcontinental networks for radio broadcast. Open-line ran 300-1K ohms. Twist-pair runs closer to 100r. Also there was a need to monitor levels throughout large networks, but lo-Z means low voltage and poor rectifier performance. An older standard was based on 500r at the meter. In 1939 the VU meter standardized on 600r. 600:600 was the nominal interface. At least one major network used 150r internally, transforming where needed.

FWIW, now that very-long analog lines do NOT exist, the modern broadcast standard is about 60r source and very-hi load. Higher than the 10K-22K commonly found in modern gear. Basically to save 19 cents per input.

The sound-changing properties of transformers were appreciated long-long ago. Radios that were clear on headphones got dull when put through a power amp with three transformers. Tap-dance that was clear in the recordist's monitor were all splashy after mixing through a console with a literal ton of transformers. Nevertheless, old gear was FULL of transformers. Not all that stuff sounds wonderful.

_I_ think most audio improved as transformers were marginalized to the few places where they still offer advantage.

Farmelo offers a different perspective.

I do not think he was hearing "impedance". He was hearing the imperfect reproduction of pretty-good transformers. Blagona knew a tranny with softness that complemented the irritation in the record.

Every transformer is different. Like an artist with only one pigment, it may not be what the work needs.

I sure would do a LOT of listening with clip-leads before worrying about the switch.

[groselicain]

There is an old Inge quotation that I'm sure you're familiar with: There are two kinds of fools. One says, "This is old, therefore it is good"; the other says, "This is new, therefore it is better."

While you may feel that we're all the better for removing as many transformers from the signal path as possible, there are some of us who will gladly take those unwanted transformers off your hands. I for one am still tickled pink by the character imparted by a pair of Triad A-67Js I wired up three years ago across my master buss.

Have a good rest of your night, PRR. Thanks again to everyone for their help.

[vigilante397]

_I_ think most audio improved as transformers were marginalized to the few places where they still offer advantage.

Based on this I don't think Paul is saying transformers make things sound worse in general, it's just that the way electronics operated years ago necessitated transformers in different stages of the signal path that are no longer necessary today. There is still benefit to be had, especially from quality transformers, but we don't need a million of them scattered around the signal path.

But as per the original topic and Paul's suggestion with the clipped leads, if you have some transformers that you know you like the sound of and want to A/B them, the switch diagram I posted should take care of it.