Originally Posted by justanoldmuso
gordon...heres a challenge.
one of the problems with diy mic pres useing op amps is need for + -
supplies for the op amps as you know.
see if you can come up with a easy to build mic pre useing single supply op amp with low noise and high gain with xlr in for mic...
plus variable input impedence. ive never seen a schem yet...its all dual rail supplies.
All techie-stuff again, feel free to ignore.

There are various ways to build a single-supply pre-amp, but they still all need at least a DC block(*) (e.g. capacitor) on the output and almost always also similar on the input. Most single-supply Op-Amps go nowhere near the 48V of phantom power. I doubt any of the few are low-noise, but it's possible. The choices for best low-noise, low distortion, OpAmps is surprisingly modest (or was the last time I looked deeply at this). Interestingly, JFETs have a functional advantage here, because, like valves, they can have their input biassed at 0V, which can remove the need for a separate input DC blocker.

I suspect a discrete solution with one of those low-noise JFETs as the input and gain device, with a constant current load for linearity and a bipolar output buffer would actually perform pretty well in comparison with pretty much any OpAmp solution and would not want a split supply. Condenser mics use an arrangement like that, though usually have a higher voltage signal out from the diaphragm to the FET than a dynamic has, but that diaphragm signal is at far too high an impedance to send down the cable unbuffered.

Addendum...
Interestingly, having written that, I had a quick look for more information on the FETHead and found a "what's inside", that shows four FETs (probably Toshiba 2SK209-GR or 2SK880-GR)(***), three resistors and a couple of other devices that look like either diodes or capacitors, though initially I can't see why they would fit either of those ... static protection, maybe?(**) With so few resistors, that suggests to me very limited negative feedback, so not the lowest of distortions, but maybe that's an acceptable trade-off. It looks like it's probably a very simple pair of feedback pairs in a bridge arrangement. I'm intrigued enough that I may try simulating what I think it looks like. If it works as well as people say it's an excellent example of form-follows-function minimalism. Respect.

(*) Apparently not ... the FETHead doesn't seem to have any, though as I guess the pre-amp must have DC blocks against the phantom power, they can get away without them in their module. That also supports my "bridge" thought.

(**) Edit: They'll be diodes or VDRs, primarily to protect the FET's gates from the 48V power at switch-on/plug-in.

(***) Hmm ... despite the package marking suggesting those parts, they're actually MOSFETs, not JFETs! Curious; Toshiba list them under MOSFETs then describe them as JFETs.

Last edited by Gordon Scott; 12/29/23 03:58 AM.
Jazz relative beginner, starting at a much older age than was helpful.
AVL:MXE Linux; Windows 11
BIAB2025 Audiophile, a bunch of other software.
Kawai MP6, Ui24R, Focusrite Saffire Pro40 and Scarletts
.