"Modified phantom-powered microphone circuit reduces distortion," Dimitri Danyuk

[mediatechnology]

Note my edit where I missed typing "2" for 20 dB. Yeah I can't think of a single reason to go deeper than 20 dB. But maybe there is.

On SSL's 4K/6K pad it looks like they used an optional 4th resistor (R3?) in the switch network to maintain the termination impedance for pad depths in which the constant termination impedance couldn't be maintained. I haven't done the math above 20 dB but it look like at values >20dB this might have been a problem. Calculating all the values I did from 10 dB to 20 dB I didn't have much trouble keeping the target load impedance. I think in the 20dB/2K calcs the delta R was like 30 Ohms.

R3 here: http://www.ka-electronics.com/images/SSL/ssl_82E149.pdf

[JR.]

That is interesting.... I don't think I've seen pads done that way but yes it appears you can mange both input and output terminations that way.

Thinking about this a little more, there is one possible application where you might want to maintain termination impedance coming from the pad, and that is with a transformer input where the transformer may be tuned for the mic source impedance.

JR

[mediatechnology]

Since I am arguing that a lower output termination is generally harmless and could even deliver lower input noise...

That's one of the things I strived for in the DN-140 pads: To keep the impedance seen by the preamp low when the pad was engaged. With an open-circuit input and high gain (the showroom floor noise test) the padded noise performance is actually better than without the pad in. With terminated inputs the effect isn't as big but it still helps.

I don't think SSL ever stufffed R3 and R4 in consoles they delivered to recording studios but they may have had broadcast customers that used them. The 82E01 (transformer input) isn't built this way.