On Sun, 16 Jan 2005 19:36:48 +0000, Nick Gorham
wrote:

Iain M Churches wrote:
"Eiron" wrote in message
...

How strict are the negative feedback rules?

It has to be single ended with no NFB as far as I recall.

I presume a Darlington pair or triple is allowed but if I use a triplet
of NPN/PNP/NPN or even an NPN/PNP pair then someone is going to argue
that it has a load of negative feedback even though it behaves like a
high gain transistor in emitter-follower mode.

The SET uses two halves of a 6SN7 and then a 300B, so three stages
per channel.

I may join in then, my 211 only uses three stages per chan as well :-)

But doesn't that have a *lot* more power?

My paper design does indeed have three stages, a central MJL4281
voltage gain stage which dominates the 'sound', flanked by emitter
followers at input and output. Four transistors in all, very simple
design, 5-8 watts of Class A operation, no loop feedback. Frankly, I
don't care what Turner et al think, it is a KISASS design which uses
BJTs in a way suited to that technology, in the same way that the
6SN7/300B is used in a way suited to that technology.

--

Stewart Pinkerton | Music is Art - Audio is Engineering

"Stewart Pinkerton" wrote in message
news
On Sun, 16 Jan 2005 20:03:09 +0200, "Iain M Churches"
wrote:

You could follow the KISS thread in RAT so far, and see if you agree with
what Andre has written to date. RAT is a friendly NG with a large number
of
very well informed subscribers.

Indeed it is - and some extremely ill-informed ones! :-)

It is also a pleasure to follow a group (RAT) where
each and every thread does not turn into
"all good SS amps sound the same -
identical to the input" :-))

Kunniottaen

Iain

In article , Eiron
wrote:


How strict are the negative feedback rules? I presume a Darlington pair
or triple is allowed but if I use a triplet of NPN/PNP/NPN or even an
NPN/PNP pair then someone is going to argue that it has a load of
negative feedback even though it behaves like a high gain transistor in
emitter-follower mode.

This thread did make me wonder if it might be interesting to try making a
power amp with 'one device' - e.g. a SAP15. :-)

There should be some design parameters set before I raid my junk box. To
make it fair for the valvies, perhaps I should only use germanium
transistors. :-)

Don't use AL102's. 8-]

....although if you find any that have survived so long, they're probably
the reliable ones, so should be OK. :-)

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html

In article , Iain M Churches
wrote:

"Stewart Pinkerton" wrote in message
news
On Sun, 16 Jan 2005 20:03:09 +0200, "Iain M Churches"
wrote:

You could follow the KISS thread in RAT so far, and see if you agree
with what Andre has written to date. RAT is a friendly NG with a
large number of very well informed subscribers.

Indeed it is - and some extremely ill-informed ones! :-)

It is also a pleasure to follow a group (RAT) where each and every
thread does not turn into "all good SS amps sound the same - identical
to the input" :-))

This brings to a mind a question I have been meaning to ask you, but have
only now got around to... :-)

When producing/balancing/mixing/etc a recording and adjusting the results,
monitoring what you then get (hear) I am wondering what you have in mind in
the following terms:

When you then output a set of waveforms (e.g. in the form of sets of sample
values to define the waveforms to be recorded/distributed on CD's, etc) is
the implict assumption (or consious intent) that these should be the
waveforms that will be presented to the loudspeaker terminals of the
listener who has a 'good' system? Or is it that this represents the
pressure waveforms you wish the speakers to radiate to their ears? Or what,
exactly?

The question really boils down to, when you decide a given result
heard/observed by you during this process is 'right', how do you set about
providing that for the customer (buyer of the CD, etc). Or is it your
assumption that what you are doing has no defined relationship with what
the customer should hear? I am trying to clarify what the actual purpose of
the balancing/mixing/monitoring, etc, is in terms of what you aim to
deliver to the customer.

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html

Eiron wrote:

Iain M Churches wrote:
It has no global feedback, but otherwise no details are known, indeed
it's not known if it has physical existence.

It would be difficult to run a series about the design of an amplifier
that doesn't exist:-)


It would seem prudent, therefore, to postpone my making any comments on
either its design or its 'performance', or attempt to compare it with
anything else. :-)


You could follow the KISS thread in RAT so far, and see if you agree with
what Andre has written to date. RAT is a friendly NG with a large number of
very well informed subscribers.

How strict are the negative feedback rules?
I presume a Darlington pair or triple is allowed but if I use a triplet
of NPN/PNP/NPN or even an NPN/PNP pair then someone is going to argue
that it has a load of negative feedback even though it behaves like a
high gain transistor in emitter-follower mode.

There should be some design parameters set before I raid my junk box.
To make it fair for the valvies, perhaps I should only use germanium
transistors. :-)

--
Eiron.

The challenge to use TWO bjts, ie 2, one plus one.
No more than 2. Less than 2 if you like.


Since it is impossible to get Ro down to 1 ohm without some sort of NFB,
we will allow enough to do this and no more.

5 watts output.
1 volt input, at least 50k input Z.

I doubt it can be done with TWO BJTs.

But a mosfet and j-fet will allow it,
so the circuit can be as simple as the triode amp.

Patrick Turner.

Iain M Churches wrote:

"Eiron" wrote in message
...

How strict are the negative feedback rules?

It has to be single ended with no NFB as far as I recall.

I presume a Darlington pair or triple is allowed but if I use a triplet
of NPN/PNP/NPN or even an NPN/PNP pair then someone is going to argue
that it has a load of negative feedback even though it behaves like a
high gain transistor in emitter-follower mode.

The SET uses two halves of a 6SN7 and then a 300B, so three stages
per channel.

Andre's design called for a triode driver tube with a gain of about 40,
so that only one triode drive tube is needed between the CD player,
gain control and 300B.



There should be some design parameters set before I raid my junk box.
To make it fair for the valvies, perhaps I should only use germanium
transistors. :-)

Most people seem to think that the SS design is a lot more difficult
than it may at first appear. It will be interesting to see what you come
up with.

Its easy with a mosfet and j-fet driver, with a low impedance
16 Ohm : 6 ohm OPT.

Its very difficult with BJTs.

Patrick Turner.



Iain

Nick Gorham wrote:

Iain M Churches wrote:
"Eiron" wrote in message
...


How strict are the negative feedback rules?


It has to be single ended with no NFB as far as I recall.


I presume a Darlington pair or triple is allowed but if I use a triplet
of NPN/PNP/NPN or even an NPN/PNP pair then someone is going to argue
that it has a load of negative feedback even though it behaves like a
high gain transistor in emitter-follower mode.


The SET uses two halves of a 6SN7 and then a 300B, so three stages
per channel.


I may join in then, my 211 only uses three stages per chan as well :-)

I have a 13E1 design which makes 25 watts, but has two drive stages and 15 dB
global NFB.
This design like your 211 is too complex to be considered simple.

Simple is where there is no NFB, so one gain stage can be omitted, and
sensitivity
is such that there is no preamp needed, and so just a triode driver with 300B

and a horn speaker is all one needs. Power needed is about 0.5 watts max most
days
although the ceiling is 5 watts.
THD is low because PO is so low.

Patrick Turner.



--
Nick

"Life has surface noise" - John Peel 1939-2004

Stewart Pinkerton wrote:

On Sun, 16 Jan 2005 19:36:48 +0000, Nick Gorham
wrote:

Iain M Churches wrote:
"Eiron" wrote in message
...

How strict are the negative feedback rules?

It has to be single ended with no NFB as far as I recall.

I presume a Darlington pair or triple is allowed but if I use a triplet
of NPN/PNP/NPN or even an NPN/PNP pair then someone is going to argue
that it has a load of negative feedback even though it behaves like a
high gain transistor in emitter-follower mode.

The SET uses two halves of a 6SN7 and then a 300B, so three stages
per channel.

I may join in then, my 211 only uses three stages per chan as well :-)

But doesn't that have a *lot* more power?

My paper design does indeed have three stages, a central MJL4281
voltage gain stage which dominates the 'sound', flanked by emitter
followers at input and output. Four transistors in all, very simple
design, 5-8 watts of Class A operation, no loop feedback. Frankly, I
don't care what Turner et al think, it is a KISASS design which uses
BJTs in a way suited to that technology, in the same way that the
6SN7/300B is used in a way suited to that technology.

We know you don't care, but it doesn't worry us.
Your'e using FOUR bjts, not TWO.
Initially, you said you'd achieve simplicity with no loop FB.
But I betcha you are using a lotta NFB to get a result.
I think you have to, because those naughty bjts are unlistenable without it.

Have you considered two fets?

Feel free to use a choke feed to the mosfet drain, and cap couple to the load.

Patrick Turner.






--

Stewart Pinkerton | Music is Art - Audio is Engineering

It is also a pleasure to follow a group (RAT) where each and every thread does
not turn into "all good SS amps sound the same -
identical to the input" :-))

I think there's only one ng in the known universe that operates on those
'principles'.

=== Andy Evans ===
Visit our Website:- http://www.artsandmedia.com
Audio, music and health pages and interesting links.

"Andy Evans" wrote in message


It is also a pleasure to follow a group (RAT) where each and every
thread does not turn into "all good SS amps sound the same -identical to
the input" :-))

I think there's only one ng in the known universe that operates on
those 'principles'.

Not hardly.

It's a shame that so many people in RAT think that imperfect is perfect and
near-perfection is to be intentially avoided.