DCAudioDIY.com

What you describe is classical pro audio practice. Maximum power transfer into a matched impedance load.
200 ohms is the European standard. 600 ohms is thee US standard, based on the impedance of parallel conductors on a telephone pole, "infinitely long," an actual transmission line. (Transmission lines in RF are considered to be many wavelengths long.)

Remember, early sound engineering came out of telephone engineering. These guys were network engineers. Even the design of horns was based on impedance matching from the microphone or phono needle to the mouth of the horn. POWER TRANSFER.

See this image: http://images-mediawiki-sites.thefullwi ... 554454.png from Harrison's patent and this discussion of mechanical modeling in sound reproduction: http://www.thefullwiki.org/Mechanical_filter --halfway down the page.

Harry Olson's later work on mechanical analogies in speakers was an extension of this approach.

Another aspect of this sort of constant impedance system engineering was the quest for low system noise figure. The idea was to have excess gain at each amplification stage, 500 ohm output impedance terminated in a resistive pad which introduced loss and presented a stable termination in both directions (T or H attenuator). Everything was locked in and referenced to 600 ohms (or whatever standard).

There is also a sort of buffering aspect to this...everything sees a non-reactive termination at input and output, so instead of making an output stage that doesn't care what impedance it sees or an input stage high impedance enough to not care about the source impedance, you tell it what to see. Any hum and noise picked up by wiring was attenuated by 10dB or whatever the pad loss happened to be.

There is plenty of good info on this sort of thinking in Tremaine's "Audio Cyclopedia," to name one old text.

Basically, the impedance scheme we know in hifi, low driving impedance into a much higher input impedance is a consumer approach, allowing for most things to be plugged into most other things without thinking about it much....unless you try to plug a plate loaded 12AX7 into a 10kohm input Z amp!

6 ohms is ridiculously low output Z but "more is always more" in hifi, right?

The higher system impedance we see in hifi do make for greater sensitivity to wire characteristics, but when this was all emerging, they didn't think there were any sneaky wire characteristics. RLC, and that's it. Bandwidth wasn't even that wide, so HF rolloff wasn't a big problem.

Beyond that, I curse these ancestors for standardizing on the damn RCA plug for unbalanced connections!

And there is also the question of Miller effects with a high input impedance and high gain triodes, gone with low impedance inputs.

The drawback of the old school pro approach is that power output stages are necessary (but that's good for sonics) and the entire system has to be designed around a constant impedance topology, so you can't mix and match with consumer gear.

Also it was expensive since input and output transformers were required.

Walt, in your case, if you have enough gain to throw away, you can design attenuation pads to give you any constant impedance termination you want, in either direction. Check T attenuator design. Audio Cyclopedia is again a great resource for this.

J-ROB wrote:Check T attenuator design. Audio Cyclopedia is again a great resource for this.

Exactly. Wire becomes pretty much a non-issue done this way. I've got some 600ohm t-pads I used as volume controls at the output of a line stage. And you can pry my Audio Cyclopedia from my cold, dead fingers

Roscoe

There are lots of nice pre-built pads around on ebay and junkboxes, made by Daven, Cinema Engineering, Altec, Langevin, etc.

I had some old Daven -10dB pads that were made of card wound non inductive resistors. Really nice stuff.

They don't even use this approach for pro very much any more. A lot of it employs the undefined low Z into undefined higher Z approach now, although the impedances are shifted down from the old tube days.

Basically viewing everything as an op amp. High input Z, low output Z.

Hey, if anybody wants to play with non inductive wirewound precision resistors, my bud Wiggins down in Fredricksburg has a big stash and sells on ebay ID gotham_analog Shoot him a note. He might have what you need. I have a stash too and would gladly hand some out.

You might also want to look at some of the new thick film power resistors out there. The TO220 style case makes mounting and sinking easy. I'm using some from Arcol in the crossover for the eggs.

You might find the Mobius resistor interesting -

https://en.wikipedia.org/wiki/M%C3%B6bius_resistor

You make them from flexible resistor cards.