• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

DIY Fet microphone preamp

Dilettante

Member
Joined
Jun 5, 2022
Messages
87
Likes
36
There are a few designs posted over the internet (4 FETs in line). Is there a preferred one given the components availability in 2022?
Or "just buy the Fethead from Amazon" is the way to go?
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
3,458
Likes
9,151
Location
Suffolk UK
Firstly, The Fethead doesn't pass on phantom power, so not usable with condenser microphones, which to me would be a major limitation.
Secondly, they have fixed gain, and as I've found to my cost on occasions, microphone amplifiers very much need variable gain, as the dynamic range of real life is huge, and with fixed gain one either runs into noise, or at the other end, distortion. For controlled speech use with dynamic microphones they may be OK, but I would much prefer something more controllable.

As to DIY, unless you particularly want to do it for the satisfaction and instruction, I would buy one of the many mic amps around, for less money than it would cost to make one. They can be direct to USB, such as the Focusrite Scarlett range, or cheaper, any of the Behringer interfaces. Alternatively, there are again any number of small analogue mixers, I use a Behringer Xneyx mixer which has some pretty good mic amps, with decent enoughEQ.

S.
 

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
so not usable with condenser microphones, which to me would be a major limitation.
But. you don’t need a "Fethead" for a condenser microphone since they have build in active amplification.

Secondly, they have fixed gain, and as I've found to my cost on occasions, microphone amplifiers very much need variable gain, as the dynamic range of real life is huge, and with fixed gain one either runs into noise, or at the other end, distortion. For controlled speech use with dynamic microphones they may be OK, but I would much prefer something more controllable.

As to DIY, unless you particularly want to do it for the satisfaction and instruction, I would buy one of the many mic amps around, for less money than it would cost to make one. They can be direct to USB, such as the Focusrite Scarlett range, or cheaper, any of the Behringer interfaces. Alternatively, there are again any number of small analogue mixers, I use a Behringer Xneyx mixer which has some pretty good mic amps, with decent enoughEQ.
I think you have got the wrong idea about how a "Fethead" is used?
They go between an Particularly low output level Dynamic microphone and an Interface/mixers.

Normally an interface/mixers has quite a wide gain range + 30dB to +60dB and some have a negative gain range/pad of 15-or20dB

A Fathead offers around 18-27dB of extra gain (depending on model and interface)
So if you know your gain staging you would only used the inline preamp if the gain of your interface alone is not sufficient.


I know how to use Google
Oh ok... so what question can’t be answers by google?

They have 1500 in stock...

I don’t know how to use my crystal orb to read your mind.
You did not specify your application or you design goal.
 
Last edited:
OP
Dilettante

Dilettante

Member
Joined
Jun 5, 2022
Messages
87
Likes
36

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
LCSC /JLCPCB is probably the cheapest and easiest and they have the Jfet in stock.

very mush functionally the same.

XLR connector i would go with Neutrik / Amphenol.
Don't know if they are in stock at LCSC, but you can get the ones for cable mounting locally and just solder cables to the PCB.
This also keeps mechanical stress away from the PCB.

For production i would not do this but for DIY it’s ok.

Edit:
I recommend reading this:
LSK489 Application Note - LSK489appnote.pdf
JFETs: The New Frontier, Part 1 | audioXpress
JFETs: The New Frontier, Part 2 | audioXpress

The basic circuitry is common source amplifier:
185px-N-channel_JFET_common_source_degeneration.svg.png

the unusual part is "Rd" is the Phantom power impedance of your interface.

to reduce input capacitance its in a "cascode"
Figure14B_JFETs%20The%20New%20FrontierPart%202.jpg


the lower input Jfet has little AC voltage swing over DS so its operating point is fixed.
the upper JFET is handling the whole voltage drop over DS.


I think it’s kind of hard to see / confusing in the typical shared schematic:
Alctron.MA.1.jpg



I made this version:
1654440170711.png

I thinks this makes it way easier to see whats actually going on.

btw: @pma is the local expert on this kind of thing maybe he has a opinion on this.
 
Last edited:

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
3,458
Likes
9,151
Location
Suffolk UK
But. you don’t need a "Fethead" for a condenser microphone since they have build in active amplification.
I don’t know how to use my crystal orb to read your mind.
You did not specify your application or you design goal.
My Condensers have a FET buffer for the capsule, but still output at microphone levels, not line level, so do need amplification, and do need phantom power.

Furthermore, I would never use a fixed gain booster in the way you suggest as it's just too easy to overload the booster during wide dynamic swings in music, or indeed speech recording.
S.
 

DVDdoug

Major Contributor
Joined
May 27, 2021
Messages
3,016
Likes
3,966
What are you going to use it for? Usually preamps are built-into a mixer or audio interface so it's rare to need a stand-alone preamp.

And I assume you know that "computer mics" are not interchangeable with stage/studio mics? They need different preamp designs. And if you're going to use studio condenser mics, the preamp has to supply 48V phantom power. Dynamic mics don't need power. Electret condenser computer mics get 5V from the soundcard/soundchip.

Is there a preferred one given the components availability in 2022?
I'd use a low-noise op-amp. Op-amps are super-easy and it's easy to get good performance. It might be possible to get lower noise with discrete FETs or transistors, but as an amateur hobbyist I'm sure I could do better with an op-amp. And, op-amps are often used in critical instrumentation amplifiers where low noise is important so they should be good enough.

For something phantom powered like the FetHead or Cloudlifter I'm not sure, because 48V is too much for most op-amps. That's something I'd have to look into (if I wanted to build something like that).

My Condensers have a FET buffer for the capsule, but still output at microphone levels, not line level, so do need amplification, and do need phantom power.
Right. All studio condenser mics that I know of have a built-in head-amp. Usually it's 48V phantom powered, but tube mics require their own separate power supply.

Condensers typically put-out 20dB more than a dynamic (or ribbon) mic, but it's still line level and in any case it depends on the loudness of the sound into the mic.

Most regular audio interfaces are designed for condenser mics (and they provide phantom power) and they sometimes don't have enough gain for a dynamic mic, depending on the loudness of the sound. That's where you might need a FetHead or Cloudlifter. (Of course, you can get plenty of signal if you stick an SM57 in front of a loud guitar amp or on a snare or kick drum.)

A pro stand-alone preamp usually has plenty of "extra gain" for a dynamic or ribbon mic, again depending on what you're recording.
 

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
My Condensers have a FET buffer for the capsule
Sure, all Condensers have?
but still output at microphone levels, not line level,
Waht is a microphone level?
A SM7B is -59 dBV/PA
A NT1 is -32 dBV/PA
So about 27dB difference coincidentally the fixed value boosters have about 27dB* gain...
*Depending on Phantom voltage ,impedance and input impedance

Furthermore, I would never use a fixed gain booster in the way you suggest as it's just too easy to overload the booster during wide dynamic swings in music, or indeed speech recording.
If you use it or not is up to you.
But this how they are used and what they are deigned for.
If you have signal that peaks over -30dBV or so you would not need/ use a "booster" to begin with.


A pro stand-alone preamp usually has plenty of "extra gain" for a dynamic or ribbon mic, again depending on what you're recording.

Even most cheap interfaces have enough gain this days.
But Dilettante is giving us very little information about what he actually wants to do...
 
OP
Dilettante

Dilettante

Member
Joined
Jun 5, 2022
Messages
87
Likes
36
@Lambda , @sergeauckland , @DVDdoug ,

I should have clarified why this is needed in the first place. I have rather insensitive dynamic mic which is used for voice applications exclusively. As practise has shown, getting reliable -6 dB is challenging without stressing the DBX / interface preamps. Software processing is unfeasible due to numerous reasons.

All-in-all, ~30dB of additional gain should improve the situation greatly. As the microphone is dynamic, utilizing available phantom power is preferable, but not a prerequisite. I can supply a PSU if necessary.

Hope this makes things clearer, but let me know if I haven't covered something of importance.
 

MrMin

New Member
Joined
Jul 25, 2023
Messages
1
Likes
0
Im interested to see if we can make the some improvements to the circuit. Maybe parallel up more 2sk209s.. to improve noise. I need this for fixed gain approx 27db for sm7b speech. Need ultra low noise.. i got a fethead already but its built rather badly.. xlr wobbles like crazy
 

fineMen

Major Contributor
Joined
Oct 31, 2021
Messages
1,504
Likes
680
I should have clarified why this is needed in the first place. I have rather insensitive dynamic mic ...
Symmetric signal without ground reference, low impedance. Won't a modern bipolar OpAmp do it? I've seen devices with noise figures well below 2nV/sqrt(Hz) for less than the price of a can of beer. To phantom power it is as easy as it gets, full or pseudo symmetric output, adjustable gain, small, predictable, reliable for decades to come.
 
Top Bottom