A JFET-MOSFET Single Ended Class "A"

[Avro_Arrow]

I was fooling around with Richard Murdey's JFET-MOSFET amp as presented here and here.
This is what I came up with:
 

 
Any comments or suggestions?

 

[nikongod]

Replace R3 with an SS CCS and R1 with an SS voltage source. 
 
I'd replace R6 with a CCS too. At a minimum move it to the other side of R5. 
 
If you set it up with a string of LEDs the whole thing could be simplified into one more CCS and a bunch of LEDs. Who doesn't love moar LEDs?

 

[wakibaki]

The circuit looks workable and probably an improvement on the original, but what are you trying to achieve? The original had distortion which was probably below the threshold of audibility.
 
Of course it's zero feedback and discrete which counts for something with some people.
 
You could do a simulation to see if the performance is markedly improved, but class A or not it's only a buffer, the voltage swing is dependent on the source and with the original's output impedance ~3 ohms you can easily exceed the performance with a couple of paralleled opamps. It's got those great big electrolytics coupling the output too, and most people will prefer to avoid those.
 
w

 

[Avro_Arrow]

My goal, at this time, was to improve on the simple resistor CCS with a transistor CCS.
I realize that this, as with very many amplifiers we know and love, is far from "the perfect
amplifier", but we build them and love them anyway.
As for the big, nasty output cap...better that than a third channel for virtual ground and this
design does not lend itself to dual rail power supplies.
Besides, I like the way the Elna Silmic II sounds...

 

[akgfan]

I like the way my RJM amp sounds and I wouldn't want any improvements. Why would I? If I would, I could have built another amp. But somehow, I never liked opamps. And this looks simple, like tube clone.

 

[rjm003]

Hey, someone other than me actually built that circuit! I never knew.
 
The other day I decided to reappraise the JFET-MOSFET Buffer (http://phonoclone.com/diy-head.html), give it a bit of a dusting after not looking at it for 12 years!
 
It actually stands up quite well. I mean, it's not actually doing much - unity gain class A - and it never made any pretense of being efficient, so at the end of the day there isn't too much to get wrong.
 
The original JFET I used (a 2N4116 apparently, unless that was a misprint and I meant 2N4416 - no way I can remember now) is obsolete, but any N-channel with a Vgs=0 current of 5-10 mA or so will be fine. The source resistance (10 ohms) is way too low - at the the time thought it sounded better like that rather than the more sensible 30 ohms, but frankly it's just loading the output. 30 ohms would be the correct value. I'd probably choose to run it at slightly higher voltage of 15 V rather than 12 V, but for 16 ohms headphones it's fine like it is.
 
I'm working on an update. By coincidence (I just spotted this thread now) my new version also explores adding a transistor or two. The schematic attached is "pre-alpha" - the compenent values are placeholders only. I will develop the circuit over at diyaudio (http://www.diyaudio.com/forums/headphone-systems/206744-j-mo-mk-ii-headphone-amplifier.html)