High-End Tube preamp with ECC88

[Johan Potgieter]

Originally Posted by DF96

Sorry for OT, but Shoog started it!

No - I did! Sorry; bad habit of me. Back etc.

[Shoog]

Originally Posted by DF96

Yes. I blame postmodern culture and poor science teaching in schools. People enter adult life thinking that truth is just a matter of opinion. Then they discover that NFB and noNFB sound different (as they should), and prefer noNFB because that is what 'free-thinkers' (i.e. non-thinkers, in some cases!) seem to do. Then they get upset when people like me suggest that perhaps they like the sound of low-order distortion and restricted bandwidth. Of course, any amp where you can just 'dial-in' your preferred level of feedback could be poorly designed anyway and almost certainly has the wrong amount of feedback.

Sorry for OT, but Shoog started it!

The point been made is that Allen Wright manages to produces an extremely low distortion wide bandwidth amp without resorting to Global Negative feedback. It sounds excellent and in my experience better than a gNFB design.

Take it how you like - but thats reality - global negative feedback is not essential and has a generally negative outcome for many people who have made comparisons. Do not assume that I, or anyone else who comes to the same conclusion, is preferring the sound of increased distortion.

Tough if some disagree, but thats your right. You can blame it on a modern culture if you like, but then you would be wrong.

Shoog

[kevinkr]

I have not felt compelled to use gNFB in most of my designs for over a decade, I haven't found it necessary and I am able to achieve quite good linearity, bandwidth and noise performance based on stated design goals.

I do use gNFB in my power supply designs, and in addition in large PP amplifiers using beam tubes.

[artosalo]

It is possible to do a good tube amplifier without GNFB.
However, with GNFB it will be further improved.

By the way, what are the "extremely low distortion and wide bandwidth" of Allen Wright amplifiers ?

[Shoog]

Originally Posted by artosalo

It is possible to do a good tube amplifier without GNFB.
However, with GNFB it will be further improved.

By the way, what are the "extremely low distortion and wide bandwidth" of Allen Wright amplifiers ?

Spin over to his site and see what the measurements are - however let it be said that his designs are based upon HP oscilloscopes front ends which were expected to operate from DC to Mghz ranges with vanishingly small distortion.

Its also the primary reason why people go to all the trouble of using DHT in order to take advantage of their inate linearity.

My experience is that with each decibel of gNFB you can hear the life and presence drain out of the music. I choose valves because I can do without gNFB with valves.

You may be able to achieve 0.01% distortion measurements by slapping on the gNFB - but frankly it's not worth the cost.

Shoog

[revintage]

Note, FVP5A has about 9dB of NFB in the linestage due to the unbypassed cathode-resistor. As I understand it the CF is heavily feedbacked to get lower Zout. The only reason this underbiased E88CC has a chance to give us low distortion is due to the local feedback. Seems like a compromise just to get lower gain, as the E88CC has far to much gain for a lineamp.

Talking about lack of GNFB does not have so much relevance here, as there is only one gain stage. Allthough GNFB over more stages makes it all worse.

[Shoog]

As I have said - local feedback done well works well.
Ultimately it doesn't matter to me - but don't disparage a reason choice of approach by suggesting its an illogical fashion statement - thats a personal insult.

Shoog

[AJT]

just add a buffer ahead of amps that uses GNFB, and you get rid of the "stigma" of global negative feedback......there, easy and quick fix and it works......

recently i made a 6C33 set it is a non global negative feedback amp.....before that i made a pp 6V6 amps, this one uses gnfb....guess what, they both sound musical to my ears.....

making a sweeping statement that GNFB ruins the music is meaningless as well as useless to everyone else except the one making the claim....

[Shoog]

Its not really meaningless to say gNFB kills the sound with every decibel added. Low gNFB will sound better than high gNFB. Many amps will sound terrible without gNFB, which is not necessarily a complement of gNFB.

Far better to design an amp which is innately linear and use's judicious amounts of local feedback mechanisms to improve on that. Only as a last resort would I resort to gNFB, and at that point I would be considering chucking the design out and starting again. As was pointed out, many of the large pentode amps will not work without gNFB - and that is why I don't build with them generally.

My point overall is that a high gNFB amp approaches the sound of a transistor amp, which to my ears is not a good thing. Overall it shows that you are listening to gNFB rather than a valve amp.

Ultimately the proof of the pudding is how many really outstanding valve preamplifier's employ gNFB ??

Its not off topic to be discussing this fundamental design decision when deciding on an approach to building a High-End preamp.

Shoog

[artosalo]

Originally Posted by Shoog

.... Low gNFB will sound better than high gNFB....

Shoog

I can not find this phenomenom, but I do not say that somebody else can not hear the difference.
What you think is the reason to this ?

I understood that you think local NFB does not spoil the sound.
What you think makes the difference ?