Hamptone gain block / mic preamp article

[Scodiddly]

Sort of like the Cloudlifter?

This started for me because I didn't have anything that made dynamics useful for quiet sources. Also because I had some transformers laying around.

[trodden]

This concept is intriguing to me. These boxes you all are making and this expensive version..

https://soyuzmicrophones.com/launcher-d ... PZEALw_wcB

[winky dinglehoffer]

The Soyuz box says it has 2 transformers in each signal path, so maybe they have a 1:x input, followed by a switchable gain circuit, followed by an x:1 output transformer. (EDIT: possibly one of the transformers is 1:1, and that's all that's used in "line" mode or whatever they call it.) It looks clever. But $$$.

[Scodiddly]

Iron is expensive! Though I did want to try building one with a door bell transformer at some point.

[The Scum]

This has sort of come full circle here.

At the TapeOp conference in New Orleans, there was a DIY breakout session. Most of the group was new to DIY, and was led through the construction of one of these gain stages by Scott H himself. There were a bunch of little baggies of parts that were distributed and soldered together.

Then there was a "big kids table" for those of us coming in with some experience, led by Dave Hill. He talked about a bunch of different subjects, but among them was his assessment of the JFET gain stage the other group was building - reaching the conclusion that with another JFET and maybe a resistor or 3, it becomes a little discrete opamp. And that opamps are much easier to deploy than fixed gain stages, because inverting feedback makes them much more controllable and predictable.

(it also dovetails into the ol' Nelson Pass DIY opamp paper, and a bunch of the basic audio opamp designs floating around)

So I brought all that home and played with discrete opamps, and basic circuits using them, filters, EQs, mixers, preamps, whatever. The Opamp Labs catalog and Nat Semi Audio/Radio Handbook.

And I've got a doorbell transformer preamp among them. It works. The transformer isn't shielded, so it's sensitive to hum - the preamp lives in an ammo can, with an external 24V DC power supply...uses the 24 for phantom power with reduced resistors (2k7 maybe?). It also led me to the conclusion that you have to go to really extreme lengths to make a mic preamp that's easily discernible as different from all the others...like purposefully messed up frequency response or added distortion stages...

[Scodiddly]

I'd definitely be curious to see a schematic for a discrete opamp based on a couple JFETs.

Sure, it's kind of hard to improve on what's already out there. This turned out to be a handy and easy to build way to bring a dynamic mic up to a usable level with a prosumer interface, so I'm happy with it. I'm not expecting to make another run of boards, though.

[The Scum]

I'd definitely be curious to see a schematic for a discrete opamp based on a couple JFETs.

Here ya go:


Or better yet, download the PDF and follow through as he iterates on the design: https://www.passdiy.com/download.php?do ... yopamp.pdf

Like the Hamptone design, he works into active biasing territory (current sources) to reduce distortion & add supply voltage independence. I've had decent luck mixing and matching JFET & BJT blocks. He also sticks with modest output stages - single ended class A, no arc welders, no 600-ohm drive.

You might also look over the Cloudlifter patent...it's amlost an opamp.
https://patents.google.com/patent/US9668045B1/en

[Scodiddly]

Thanks!

That long-tailed-pair-ish stuff is something I'm still trying to wrap my head around. I understand them, sort of.

[The Scum]

Get 3 Ammeters and play with one of the Pass diff pairs (or try Spice...2 JFETs or BJTs and a few resistors are chump change, but 3 meters with decent mA tracking aren't). You'll build some intuition by watching them together as you change the input voltages.

[Scodiddly]

Get 3 Ammeters and play with one of the Pass diff pairs (or try Spice...2 JFETs or BJTs and a few resistors are chump change, but 3 meters with decent mA tracking aren't). You'll build some intuition by watching them together as you change the input voltages.

It's one of those things where Kirchoff says it has to work, but I haven't figured out how to visualize it properly yet.

[The Scum]

...visualize...

If you squint at the NPN pair long enough, it looks like a differential gearbox.

[Scodiddly]

If you squint at the NPN pair long enough, it looks like a differential gearbox.

I think the issue is more that I still tend to see a transistor as an input, an output, and a power supply pin.

OK, so here's probably the last one, the single-channel version powered by a wall wart.

Dynapusher Cardo.jpg (146.77 KiB) Viewed 3100 times

[winky dinglehoffer]

That looks awesome! My last one is in the works--finished painting the case today, just have to get it all wired up.