Vintage V-FETs in class-A amplifier with complementary push-pull output

[R-K Rønningstad]

I thought is a nice idea to set up a thread dedicated to amplifiers based on old so-called V-FETs from the 70-ties (and 80-ties maybe). Transistors made by NEC, Sony and Yamaha, and to my knowledge in amplifers from Hitachi/Lo-D, JVC, Sansui, Sony and Yamaha way back then.

We have now and then touched it in the First Watt J2 thread FirstWatt J2.

I have not concluded whether the output shall be common-source or common-drain. But maybe some will do this and some that, for all I know.

I found the Sony TAN-5550 service manual, I thought I had the TAN-8550 also, the one which is refereed to in the AES preprint I briefly flashed in another thread. But this one is good enough, just to show how a push-pull V-FET amp was built way back when..


Rolv-Karsten

[R-K Rønningstad]

And here is the actual circuitry from the 1975 service manual of the Sony TAN-5550. This is decidedly class B.
I expect us to end up in class A, probably lower wattage, simpler circuit and all that.

RK

[R-K Rønningstad]

With the block diagram we should be ready to leave class-B, if someone does not need the AES preprint, that is.

RK

[R-K Rønningstad]

Which should have been attached....

[Nelson Pass]

The original Sony designs are a little odd - right around
that time we were starting to see some bold Class A
designs appear as concept pieces in Japan, and Stax
managed to get one into the U.S. marketplace.

One later Sony amplifier appears to have used these parts
to cascode its bipolar output stage - go figure.

My intent is to see what I can make of a complementary
pair of these as followers in Class A. As a concept it's easy
enough, we'll just have to see about the performance.

[R-K Rønningstad]

I haven't dug too deep into this, but I feel these amplifiers are overly complicated really. I wonder, in the AES preprint, these V-FETs are said to not have any extreme input capacitance, and should be easy enough to drive. There are several stages in cascade, open loop gain in excess of 80 dB.
I do not need hundreds ow Watts, I need some juice for ESL-63s. F5 would probably be enough for ESL-63 in most cases. So 20 - 60 Watts. And I do not think you need all those stages, but I guess you (NP) and others are in front of me thinking about this.
BTW I believe cascoding bipolars with V-FETs seems the opposite of what I would do

[woody]

Nelson, what about a cascode output stage like in your cascode amp article
but with the power jfets as the followers.

[R-K Rønningstad]

But is there anything to gain in cascoding these, really?

[Nelson Pass]

Maybe. I could easily write a whole article on this, addressing
the voltage and current relationships on a given part and
their contribution to performance, and then expand that to
a discussion of complementary versions of those parts in
push-pull.

Cascoding might be a better approach, but I haven't gotten
that far with these parts.

I will though...

[R-K Rønningstad]

Mine have not arrived yet. Looking forward to these V-things though....

Do you have datasheets from way back when?