Hamptone JFET - mic preamp with parts on hand

[Scodiddly]

I've been sitting on a couple hundred dollars worth of Cinemag iron - pairs each input and output transformers, don't even remember exactly why I bought them.  So now that most of my projects are based around using up stuff I already own, why not build a pair of mic preamps? 

I've got a couple-three options.
1) Opamp.  Got TL072 and 4556 types on hand, plus everything I need to build a +-15vdc supply.  That's the easiest route.
2) FETs.  Years ago I ordered most of the parts for a FET-based mic preamp in TapeOp (Scott Hampton design), might still have enough pieces to hack together the gist of the design.
3) Tube.  I've got a couple 12AX7's and such, could likely scrape up coupling caps.  Power supply - I'd have to cobble something scary together from my box of discarded wall warts.  Something like putting 6v and 12v transformers back to back to produce 240 vac so I can get something near a plate voltage.

Hmmm.... 

 

[Bo Deadly]

Post the specifics of the mic transformer. That will dictate what must follow for optimal noise / response.

 

[gyraf]

Agreed. That is what will decide for you..

 

[Scodiddly]

Bear in mind I'm not necessarily looking for "optimal", got that in plenty.  Looking for "interesting", really.

Input transformer is CMMI-7C  https://cinemag.biz/mic_input/PDF/CMMI-7C.pdf
Found another pair from a group buy, CM-75101A  https://cinemag.biz/mic_input/PDF/CM-75101A.pdf
Output is CMOQ-2 https://cinemag.biz/output/PDF/CMOQ-2.pdf

 

[user 37518]

The 4556 does not have a input current noise spec but assuming a range between 2-4pA, I would say that it would work with what you've got, the TL072 will work with the 1:10 transformer but the TL072 is a lot noisier (18nV/rtHz VS 10nV/rtHz of the 4556). In my opinion the 072 is only good for one thing (if any) and that is DC Servos. I'm aware that many have used it successfully in the past but nowadays practically any FET opamp is better than the 072. None of the opamps you have are my first choice.

You don't mention which FETs do you have but you can also try using the FETs in the following pre configuration by Sam Groner https://www.nanovolt.ch/resources/microphone_preamplifiers/pdf/E_r1.pdf

 

[NOON]

I've built some of the Hamptone JFET preamps and found them to be very interesting, that would be my recommendation.

 

[Scodiddly]

So now regarding discrete FETs, I see that I've got some J202, some 2SK170, some PF5102.  And a whole boatload of 2SK879 for an ongoing product.

Bear in mind that I'm not totally adverse to buying those THAT chips.  What I'm really lacking in mic preamps are ones that are good with ribbon mics. 

 

[Scodiddly]

Of course I should point out that I could use these transformers with several different preamps, that's all just a matter of packaging and connectors.  Right now I'm thinking more along the lines of trying to build 2-3 simple designs just to see where it goes.

And yes, I might even try a 12AX7 design with whatever plate voltage I can scrape together.  ;D

 

[user 37518]

So now regarding discrete FETs, I see that I've got some J202, some 2SK170, some PF5102.  And a whole boatload of 2SK879 for an ongoing product.

Bear in mind that I'm not totally adverse to buying those THAT chips.  What I'm really lacking in mic preamps are ones that are good with ribbon mics.

With the 2SK170 FETs you can make the pre I mentioned in my previous post. Ideally you should use NE5532s or similar but the 4556 might do the trick.

 

[L´Andratté]

One of my favourite preamps I built is the Studio Sound one (s. document section or attachment for my reduced version), with NE5534 on input for which 1:7 xfrm is just right, or if you have a couple of bc547 flying around (which everyone should have), you could build this wonderful sounding circuit:
https://groupdiy.com/index.php?topic=64069.msg811214#msg811214

What is needed for a ribbon preamp?
- high gain?
- low mic output z so higher z stepup for optimum source Z of preamp?
I only have Bumblebee R6 Ribbon it just works fine with all kinds of pres-

 

[user 37518]

What is needed for a ribbon preamp?
- high gain?
- low mic output z so higher z stepup for optimum source Z of preamp?
I only have Bumblebee R6 Ribbon it just works fine with all kinds of pres-

Besides high gain I would add very low noise. Adding 50-60dB of gain to a ribbon mic can really raise the noise floor to an audible level, thats a downside of using ribbons on low SPL instruments such as acoustic guitars, you can't escape the noise. Even with a noiseless pre amp with 60 dB of gain you get around -70dBu of noise level across 20KHz, that is very audible, so you either select a higher sensitivity mic or live with it, but I wouldn't want my pre to add extra noise.

 

[Scodiddly]

You don't mention which FETs do you have but you can also try using the FETs in the following pre configuration by Sam Groner https://www.nanovolt.ch/resources/microphone_preamplifiers/pdf/E_r1.pdf

That's an interesting circuit.  I found my file on the Scott Hampton JFET preamp and there's a bit of parts difficulty with the output stage, so I was starting to think of a discrete input stage and op-amp output stage.  It makes a lot of sense to have the FET on a +- supply to reduce/eliminate the need for a coupling cap, too.

 

[Matador]

I have a handful of samples of the NJM2122 op-amp.  Although it's limited to +-10V rails, it has an impressive noise figure of 1.5 nV/sqrt Hz.

 

[john12ax7]

With those parts I would go 1:10 transformer with discrete 2SK170 gain stage

 

[Scodiddly]

Well hey, I went and really looked in my inventory spreadsheet and I do in fact have all the actives for the Hampton JFP!  Which I bought when the article came out... in 2004 or so.  :

Thinking back I had started to breadboard the power supply and got distracted by something else.  Looking it at now, I really just need to start with the simplest possible PSU (maybe a 24vdc wall wart from my junk box) and see where it goes.

 

[user 37518]

Well hey, I went and really looked in my inventory spreadsheet and I do in fact have all the actives for the Hampton JFP!  Which I bought when the article came out... in 2004 or so.  :

Thinking back I had started to breadboard the power supply and got distracted by something else.  Looking it at now, I really just need to start with the simplest possible PSU (maybe a 24vdc wall wart from my junk box) and see where it goes.

Just saw the pre amp you are refering to https://i0.wp.com/wiki.diyrecordingequipment.com/wp-content/uploads/2011/02/hamptone_fet_scan.jpg, the basic amplifier cell is a JFET common source amplifier followed by an emmiter follower with current sink, I bet the distortion specs are not stellar since there is not a lot of open loop gain or global NF, but hey, if thats the kind of thing that you like go ahead, it just seems too crude for my taste. Noise specs aren't as good as a 2SK170, with a 1:10 transformer you get something like a 20K impedance at the secondary, and according to the 2N5457 datasheet the Noise figure is something like 7dB. With a 2SK170 you get virtually close to 0dB Noise Figure with the same source resistance.

I still think that Sam Groner's amp is far superior than the Hamptone... the 3mA drain bias of the 2SK170 seems to be the sweetspot for low EIN and low dissipation.

 

[Scodiddly]

I really don't have anything that's known for some distortion.  Though of course the parts are so cheap (aside from the iron) that I could build a couple of each and see what I like.  Have a switch to choose, even.

 

[user 37518]

I really don't have anything that's known for some distortion.  Though of course the parts are so cheap (aside from the iron) that I could build a couple of each and see what I like.  Have a switch to choose, even.

If you make the Hamptone amp you should try using the 2SK170 instead of the 2N5457, you might have to play with the bias point, check for minimum noise.

 

[Matador]

I still think that Sam Groner's amp is far superior than the Hamptone... the 3mA drain bias of the 2SK170 seems to be the sweetspot for low EIN and low dissipation.

What is the advantage of wrapping feedback around that initial JEFT as opposed to simply using a FET input op-amp directly from the transformer secondary (maybe something like an OPA827, which is specified at 4nV/sqrt(Hz)?

Like this one?

http://www.nanovolt.ch/resources/microphone_preamplifiers/pdf/D_r1.pdf

 

[user 37518]

In this case it makes very little difference, the 2SK170 has a noise voltage of 0.85 nV/rtHz @ 3mA Drain current, and the opamp you mention has 4nV/rtHz, HOWEVER, in this case the source resistance noise dominates and the contribution of the opamp or the FET makes little difference. 

I just think that Sam Groner, just like the OP, had some FETs and 5532s at hand and wanted to put them to use.