Yesterday I listened to a Youtube of 2019 BAF where Linear Audio was presenting and Nelson commented that he wanted to redux The Beast and would appreciate lower cost JFETs that did not require lowest noise figure.
Has there been any discussion about possible configurations of The Beast using parts that can be acquired from Digikey or Mouser?
I have been thinking of possible parts to use and running parts through the curve tracer. More parts on order to also run through the curve tracer.
A possible proof of concept would be a 200mA bias version for a headphone amp.
The JFETs will like have to be SOT23 package. I have an initial favorite for both N and P. I will post measurements when I have all the measurements finished.
My thinking is to cascode the JFETs using the hockey pucks from the F4 Beast. This will allow using higher Idss JFETs and operating the JFETs at ~3V Vds and also operating the JFETs un-degenerated. Most of the heat winds up in the pucks.
The voltage rails can be normal First Watt rails if desired.
Just getting started with possibilities.
Has there been any discussion about possible configurations of The Beast using parts that can be acquired from Digikey or Mouser?
I have been thinking of possible parts to use and running parts through the curve tracer. More parts on order to also run through the curve tracer.
A possible proof of concept would be a 200mA bias version for a headphone amp.
The JFETs will like have to be SOT23 package. I have an initial favorite for both N and P. I will post measurements when I have all the measurements finished.
My thinking is to cascode the JFETs using the hockey pucks from the F4 Beast. This will allow using higher Idss JFETs and operating the JFETs at ~3V Vds and also operating the JFETs un-degenerated. Most of the heat winds up in the pucks.
The voltage rails can be normal First Watt rails if desired.
Just getting started with possibilities.
Really like the idea of a beast headphone amp!
The size would be manageable compared to any other headphone amp.
Early estimate is 13 JFETs to obtain ~200mA bias. Stereo push-pull followers would be 52 JFETs total. If JFETs are $0.22 each, Total JFET cost is $11.44.
Nice, that’s more reasonable than I thought. And I like the SMD choice for this. Look forward to watching your progress!
Idea
Hello codyt. Please see post #33 in the thread Different headphone amps. woofertester's suggestion to use multiples is interesting.
Best
Anton
Really like the idea of a beast headphone amp!
Hello codyt. Please see post #33 in the thread Different headphone amps. woofertester's suggestion to use multiples is interesting.
Best
Anton
Very nice, and perfect timing actually. I’ve got parts ready to breadboard a B1 rev2, and I see in your schematic that it would take very little to give yours a try with nearly the same parts. I’ll give it a listen, thanks!
Actually there is such a piece, now in the hands of another DIYer.
Also, Mr. Fullrange and I are in the process of a 1,000 fet version construction -
a little more work to go.

Also, Mr. Fullrange and I are in the process of a 1,000 fet version construction -
a little more work to go.

Hello codyt. Please see post #33 in the thread Different headphone amps. woofertester's suggestion to use multiples is interesting.
Best
Anton
Hello Anton
Yes what I am doing is similar. One difference is that the inexpensive JFETs I am looking at have much higher Idss than 2SK170BL. At 15V Vds, the power dissipation would be too close to maximum. So, the JFETs need to be cascoded.
I like your idea of trying just one pair. Easy to do on a breadboard. Thanks
cheers
Hello codyt. Please see post #33 in the thread Different headphone amps. woofertester's suggestion to use multiples is interesting.
Best
Anton
Actually there is such a piece, now in the hands of another DIYer.
Also, Mr. Fullrange and I are in the process of a 1,000 fet version construction -
a little more work to go.
![]()
Nice to hear. I look forward to seeing some schematics and pics of the official versions. In the meantime, play, play, play.
I have pics, if not schematics........ but I'm not allowed .......
practically, I didn't even wrote this

practically, I didn't even wrote this

where is ZenMod when we need him? ZM can you create a PCB with current trannies?
**** I would definitely have a go at a balancec version of this
**** I would definitely have a go at a balancec version of this
I built this...nothing special, needs more jfets
What were your voltage rails?
Did you use this for headphones or speakers?
Already done that 10 years ago with 10 pairs of K170V/J74V for a preamp output stage.
+/-12V rails; common heat sink.
Power amp is more interesting.
For example 500x 2SK209GR in parallel, as one power device; 1.5~2A bias.
2 needed for each channel, either as circlotron, or TCS push-pull.
You can also use multiple LU1014 triode cell in the ZV9 in parallel to build a follower.
Just need a gain stage upfront, and some NFB if you need to have low Zout.
Patrick
+/-12V rails; common heat sink.
Power amp is more interesting.
For example 500x 2SK209GR in parallel, as one power device; 1.5~2A bias.
2 needed for each channel, either as circlotron, or TCS push-pull.
You can also use multiple LU1014 triode cell in the ZV9 in parallel to build a follower.
Just need a gain stage upfront, and some NFB if you need to have low Zout.
Patrick
Actually there is such a piece, now in the hands of another DIYer.
Also, Mr. Fullrange and I are in the process of a 1,000 fet version construction -
a little more work to go.
![]()
You and Mr. Fullrange are gifted with patience, focus and awareness [plus great eyesight] to assemble and solder this many SMDs. I envision a robot doing this work to produce commercial Beasts.
1000x SMD FETs like 2SK209 cost more than 150USD.
That is before matching & soldering.
So commercially not viable.
Patrick
That is before matching & soldering.
So commercially not viable.
Patrick
What were your voltage rails?
Did you use this for headphones or speakers?
+/-3 volts just as its in the attached schematics
headphones only, it can not drive speakers
We've come a long way
The attached picture speaks to the progressing genius of the human race for advancing this technology and downsizing its components/systems.
The green PCB is double sided; for an [RF] SONY stereo headphones which connect to a TV for private listening. The orange-dot parts are LEDs from a spent 800 Lumens Philips household lamp.
I keep these amazing creations because I feel bad trashing them.
Best
Anton
The attached picture speaks to the progressing genius of the human race for advancing this technology and downsizing its components/systems.
The green PCB is double sided; for an [RF] SONY stereo headphones which connect to a TV for private listening. The orange-dot parts are LEDs from a spent 800 Lumens Philips household lamp.
I keep these amazing creations because I feel bad trashing them.
Best
Anton
Attachments
I built this...nothing special, needs more jfets
Hello adason and all,
Your JFET-style HA is also useable in reverse as I show in the attached schematic of a prototype which sounded great. This approach has another objective.
The signal generator does not deliver current to the phone. All of this current comes from the JFETs.
If the output stage of the signal generator is cascoded, this HA works like that described in the patent by Mr. Pass shown in my previous post.
Are you still using it?
Best
Anton
Attachments
Last edited:
You and Mr. Fullrange are gifted with patience, focus and awareness [plus great eyesight] to assemble and solder this many SMDs. I envision a robot doing this work to produce commercial Beasts.
I treat all this as a meditation. I like spending a few hours with a
repetitive task like testing jfets or soldering up beasts.
Boredom is my friend.

- Home
- Amplifiers
- Pass Labs
- Beast with 1000 JFETs redux?