Pass JFET Power Amplifier

[Nelson Pass]

Well, I guess the cat's out of the bag.

First Watt is in production with a JFET power amplifier, with
delivery next month, using the Lovoltech vertical power JFETs.

The DIY version of it can be found on

www.passlabs.com/np

under the title ZV8_POWER_JFET.PDF

This is the final draft, but you have an opportunity to correct
my grammar and errors of fact.

Enjoy.

😎

 

[PSz.]

Wow,
Nelson are you trying to fill an entire issue of AudioExpress this month?

Ok, now I've got to go read this one!

Thank you NP!

regards,
PSz.

 

[moe29]

With all the hard work you put into your creations, i hope that
you weren't scooped without permission....

 

[Nelson Pass]

I would say my timing was impeccable.

For those of you who want to cut to the chase without
downloading a 2M pdf, here's one of the schematics:

 

[mpmarino]

😱 I'M IN OVERLOAD

Slow down! Digikey isn't that fast!


---we will enjoy---

Thank you NP.

 

[darkfenriz]

Hi Mr. Pass
Why do you analyse and put much attention to Vds-Id curves, comparisons to triodes, pentodes... if final result is cascoded JFET?

 

[paulb]

"Figure 1 shows a simplified schematic
of the Yamaha B1, [which] remains a highly regarded product."

 

[Raj1]

Hi Mr. Pass
Why do you analyse and put much attention to Vds-Id curves, comparisons to triodes, pentodes... if final result is cascoded JFET?

Maybe cause nearly every posting made gets dissected, from every conceivable type of viewpoint.

 

[tschrama]

Great news .... another Pass article .... you are spoiling us Nelson, Thx!

haven't read it yet, gonna read it at once.... 🙂 🙂 🙂

cheers,
Thijs

 

[Steve Eddy]

Well, I guess the cat's out of the bag.

First Watt is in production with a JFET power amplifier, with
delivery next month, using the Lovoltech vertical power JFETs.

Ah, that's what I thought.

I'd looked at those some time back but their input capacitance was way too high for how I'd originally thought of using them (their already high input capacitance would have been magnified by a factor of 100). And oddly enough, I was looking at them again just the night before I saw Srajan's blurb on 6 Moons.

In any case, coolness! 🙂

se

 

[Steve Eddy]

With all the hard work you put into your creations, i hope that
you weren't scooped without permission....

Well he did EMail his announcement of it to a member of the press. And I'd like to think that if he'd have asked Srajan not to say anyting publically about it yet that Srajan would have honored that request.

se

 

[jan.didden]

Hi Mr. Pass
Why do you analyse and put much attention to Vds-Id curves, comparisons to triodes, pentodes... if final result is cascoded JFET?

Good question! If I understand you correctly, that particular Vds/Id characteristic really is irrelevant in the cascode topology. So, that would mean the results would be as good with any old MOSFET, or even a darlington BJT to keep the high input impedance?

Jan Didden

 

[Nelson Pass]

Why do you analyse and put much attention to Vds-Id curves, comparisons to triodes, pentodes... if final result is cascoded JFET?

Good question. In fact there is some load line action in the amp
by virtue of the limited transconductance of Q2, and we can
easily modulate it further. I was saving that particular adjustment
for another time, not wanting to give you a headache right
away.

😎

 

[jan.didden]

Good question. In fact there is some load line action in the amp
by virtue of the limited transconductance of Q2, and we can
easily modulate it further. I was saving that particular adjustment
for another time, not wanting to give you a headache right
away.

😎

Right! Makes a lot of sense. Both 😉 . So we can look forward to a version with some amount of pos feedback? (Watch all those cats Nelson!).

BTW, Clever circuit. The art of leaving out.

Jan Didden

 

[Nelson Pass]

Mr. Harrington gave me a smaller copy of the ZV8 for those
of you with dial-up service, and to save Choky the trouble.

It;s at www.passlabs.com/np with almost the same name.

😎

 

[Nelson Pass]

So we can look forward to a version with some amount of pos feedback?

BTW, Clever circuit.

Thank you. I know you don't dispense praise lightly.

As to the positive feedback, the modulation is similar to "muting"
the cascode. Some small variation is allowed through, but it
works to slightly decrease the voltage across the JFET on the
negative output swing and increase it on the positive, running
counter to the natural tendency of the transconductance to
increase at higher current and decrease at lower. This gives
some cancellation of the two forms of nonlinearity, which is what
I was referring to when I mentioned working the triode load-line.

😎

 

[jan.didden]

Thank you. I know you don't dispense praise lightly.

As to the positive feedback, the modulation is similar to "muting"
the cascode. Some small variation is allowed through, but it
works to slightly decrease the voltage across the JFET on the
negative output swing and increase it on the positive, running
counter to the natural tendency of the transconductance to
increase at higher current and decrease at lower. This gives
some cancellation of the two forms of nonlinearity, which is what
I was referring to when I mentioned working the triode load-line.

😎

You're welcome.

Coincidentally, that was also what I was referring to when I mentioned "some amount of pos feedback".

Jan Didden

 

[Zen Mod]

Mr. Harrington gave me a smaller copy of the ZV8 for those
of you with dial-up service, and to save Choky the trouble.

It;s at www.passlabs.com/np with almost the same name.

😎

sorry ,Master-I can't see smaller file at ********/np .
in any case -420KB version of ZV8 is at

http://czsr_bac.on.neobee.net/ZV8.pdf

cheers to all!

 

[jh6you]

What's up with him . . . ?


Don't cry Lumanauw 😀


Regards
jH

 

[woody]

I just wonder how hard to find these power jfets are going to
be to find.