Going on from simons 45 thread

[steve s]

Paul's had me thinking, a dangerous thing I know, but if an output valve driving an identical output valve would cancel their own 2nd harmonics, why don't we see this implemented more. Does it really work in practice.
The coupling stage would need to be pretty phase stable at all frequencies.

I know most output valves are too lower gain and require a voltage amp stage.
But not all, so a low output two stage amp is possible, but is there any value in it

I would also guess that the 2nd order distortion may get get substantially worse near clipping as the wave form would become even more asymmetrical than the output valve can handle, but there could be a possibility of a low distortion sweetspot that is of benefit.

[Nick]

The problem with the idea is the driving valve load and voltage swing will be different so the cancellation will be varied at best. If you want to cancel 2nd harmonic using two of the same valves then a push pull output stage would be the way.

[steve s]

so it's really an urban myth..

Good point about the push pull though..

[andrew Ivimey]

Yeah but doesn't push pull throw out a different kind of distortion or..... it's the way I make'em.

[Nick]

Well, there will be some combination of the distortion and there will be points the levels are reduced, but they may vary in a less natural way than two or more dissimilar values. It may be that using two values in this way produces a particular distortion result, and it may be that that result is preferred, but silver bullet its not (IMHO).

There is some irony in that there are well known and understood way of reducing distortion, but they are in some way disliked at the same time something like this is admired, I suspect its a variation of the appeal to authority fallacy.

[Nick]

Yeah but doesn't push pull throw out a different kind of distortion or..... it's the way I make'em.

Without measuring something you have made I cant say, but push pull is a combination of two identical values that can be used to reduce 2nd harmonic distortion. That's all I was replying to.

[andrew Ivimey]

And I agree but I thought that negative harmonics came into play. 3rd fifth and so on

[Neal]

Reducing distortion created in the output stage but not if present in the input signal

[Nick]

Reducing distortion created in the output stage but not if present in the input signal

Yep, we went around that house some time back. Almost by definition, the input signal has no distortion.

[Nick]

And I agree but I thought that negative harmonics came into play. 3rd fifth and so on

Well, they are there anyway, but become more apparent because the higher levels of 2nd are reduced.

But I refer everyone again to that remark in the video by John Curl about the 7th Harmonics being the evil one.

https://www.audio-talk.co.uk/phpBB3/vie ... 07#p197607

[andrew Ivimey]

7th eh! Okay.

[Nick]

I just added the link to the post in question and my list of what tones the various harmonics relate to (or don't in the case of the 7th)

[Neal]

Input signal has no distortion, can you explain that one Nick

[Nick]

Well, its is what it is, that's what I mean by no distortion. It may be guitar through a fuzz box distorting the guitar to hell, but that's the signal, not the signal plus distortion.

[steve s]

Years ago I was offered a large and grand looking audio distortion analyser, at a probably a bargain price
I've regretted turning it down ever since.