Increasing 2nd HD

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello everyone,

is there a way to slightly increase the second harmonic distortion of a Zen or similar single ended amp? Should it be done in the driver stage or the output stage?

I simply want to experiment and find what differences I can or cannot hear. However, I want to do it in a way that leaves the amp performing at its best in all other ways.

Thanks for any help,
Grant
 
AX tech editor
Joined 2002
Paid Member
Can you tell us, do you want only 2nd harmonic (difficult) or increased even order harmonics (much easier).
I guess you also want to know exactly (more or less) how much it increases, so you can establish levels of audibility?

You can for instance insert back-to-back diodes in the forward or feedback chain to get extra even order THD. Can you measure the levels or do you need a way to specifically set the THD level?

I know, questions, questions...

Jan Didden
 
The one and only
Joined 2001
Paid Member
Try varying the bias current up and down for any of the
Zen amps, or on ZV2 and 5, adjusting the value of the AC current
gain of the current source.

No procedures will add 2nd harmonic without any other
changes, but these are easily accomplished.
 
One trick some pedals use...

Is to create a dc offset in the signal. Then, when in the early stages of overdrive when the amplifier just starts to squash the signal, the waveform will be slightly asymmetrical (eg squashed differently on the top and bottom). When the waveform is asymmetrical, even order harmonics are present. I learned this reading a very informative paper written in for the AES by Russell Hamm at the recording studio where I used to intern. They're die hard vacuum tube purists there and they really did their homework on what makes a difference in the sound produced by the different amplifier technologies:

"Tubes Vs. Transistors -Is There an Audible Difference?" a PDF of the original paper with diagrams and artwork can be downloaded at:
http://www.milbert.com/tvt.pdf

-It won't tell you how to get even order harmonics, per -se, but it will tell you just about anything you ever wanted to know about the nature and causes of amplifier distortion.
I highly recommend it to everyone.

-Erik.

:cool:
 
janneman said:
Can you tell us, do you want only 2nd harmonic (difficult) or increased even order harmonics (much easier).
I guess you also want to know exactly (more or less) how much it increases, so you can establish levels of audibility?

You can for instance insert back-to-back diodes in the forward or feedback chain to get extra even order THD. Can you measure the levels or do you need a way to specifically set the THD level?

I know, questions, questions...

Jan Didden


Thanks for the suggestions, all.

Jan, adjusting the even order harmonics would be fine. Ideally I would measure it, but lacking decent instruments, I would at least like to know when it is increasing. As for the back to back diodes, do you mean the diodes are in series with each other, and in series with the feedback chain?

Nelson, thanks, I assume you mean I should apply less bias current or current gain for more 2HD? And is the 2HD only going to increase at clipping levels, or at all levels?

Erik, the DC offset idea is neat, but I expect you are talking about increasing 2HD at clipping levels.


I am more interested in being able to adjust the 2HD at moderate output levels, say 0.1 to 1 Watt (I have efficient speakers), and without the amp clipping. And I am only referring to distortion levels of the order of 0.25% to 1% or 2%. Similar to the likely distortion levels of the speaker drive units.

More of an experiment in subtlety than in gross audibility.

Grant
 
overdrive vs. clipping

nowater said:

Erik, the DC offset idea is neat, but I expect you are talking about increasing 2HD at clipping levels.

More of an experiment in subtlety than in gross audibility.
Grant

Yeah, that can often be the case.
The paper I mentioned explains this much better than I can, but I'll give you what I understand of it:

The kind of harmonics being emphasized (eg lower order and even vs. odd and higher order) will be much more euphonic to the ear, (being musically related to the original signal -even harmonics=octaves, fifths, etc...) which can tolerate significant percentages of the stuff before the signal begins to sound distorted. It just sounds fuller, and more "choral" to use an imperfect description.

The kinds of harmonic distortion must be differentiated. There is the slight compression of a signal as it begins to near the limits of the active device. This is a rounding of the waveform or "squashing" which occurs in tubes to the greatest degree, followed by MOSFETS, then finally bipolar transitors and op amps. The interesting thing is the range you have to play with before hard clipping occurs. In tubes you can get almost 20 dB into this range, before a significant number of odd and higher order harmonics become audible enough to become objectionable. I'm not sure how far you can go with FETs, as they were not tested in this paper, but bipolar transistors halved that range to about 10 dB, while Op Amps were around 5 dB. Roughly, beyond those ranges the amp would indeed hard clip, and you would get the kind of undesireable sound we normally associate with "distortion".

Again, I'm only giving you the quick and dirty summary of what's in the paper. Russell Hamm makes it all clearer.

I don't know if any of this is useful to you, but I thought it might be possible to coax the more subtle early overdriven sound out of your amp on purpose, while keeping it below hard clipping.

-Erik
 
Re: overdrive vs. clipping

e.lectronick said:


<<snip>>

The kinds of harmonic distortion must be differentiated. There is the slight compression of a signal as it begins to near the limits of the active device. <<snip>>... before hard clipping occurs. In tubes ...20 dB, .....bipolar ...10 dB, Op Amps ...5 dB.

.... quick and dirty summary of what's in the paper.

...I thought it might be possible to coax the more subtle early overdriven sound out of your amp on purpose, while keeping it below hard clipping.

-Erik

Thanks again Erik. I did read the paper and it's very interesting. But it is trying to "tell me the answer" of how higher 2HD sounds, instead of letting me find out for myself, and its effect on *my* speakers.

Say I'm listening to music at 1W output level with a 10W-20W amp, and distortion is maybe 0.2% 2HD (single ended amp). I want to tweak it up (along with other harmonic distortions if necessary) to 5 or 10 times that amount of 2HD while still at 1W output (approx).

I'll experiment with several of the suggestions received earlier, except where they involve clipping the output stage. Maybe if I apply Erik's DC offset idea through an earlier gain stage e.g. preamp? (ugh)

Grant
 
i thought i remembered reading this in one of the Zen articles.
i tend to refer to someone who knows more about this stuff than
i do. This is from the original Zen amp article.


It is occasionally misunderstood that single-ended amplifiers
intentionally distort the signal with second harmonic in order to achieve
a falsely euphonious character. This is not true. Low distortion is still an
important goal, and it is my observation that deliberate injection of
second harmonic into a musical signal does not improve the quality of sound.


just thought i'd mention this in case you hadn't read the article.


m.
 
I heard a Cool Anecdote

I don't remember who said it, but I heard some one say, once that "The ideal amplifier is a straight wire with gain."

I kind of liked that. It sort of brings you back to center when the idea of High Fidelity gets muddled with concerns of which distortion is better, etc...

-Erik.

P.S. I think designers of musical instrument amplifiers whose intent it is to produce a euphonic distorted sound would be the ones to actually want to design that capability into their circuits. Cascading multiple gain stages, using diode clamps. etc...
 
Some more ideas

Nelson Pass said:
Try varying the bias current up and down for any of the
Zen amps, or on ZV2 and 5, adjusting the value of the AC current
gain of the current source.

No procedures will add 2nd harmonic without any other
changes, but these are easily accomplished.


Nelson, didn't you also mention in your Balanced Zen Line Stage article that the subtle mismatching of gain devices would produce 2nd order harmonics to a degree at output levels around ~1 volt?

I wonder if you went out of your way to mismatch the FETs to a greater degree if that might accomplish the goal of increasing 2nd order harmonics while retaining all the other characteristics of the amp? Then, perhaps cascade a couple of these stages in series (depending on how much 2nd harmonic you want), and crank up the gain in the power amplifier to make up for volume.

-Of course, this would only work in a balanced preamp or power amp-

None of this would seem to be in danger of clipping, as the BZLS and Son of Zen amp would just be doing their jobs as normal.

-Erik.
 
The one and only
Joined 2001
Paid Member
Re: Some more ideas

e.lectronick said:
Nelson, didn't you also mention in your Balanced Zen Line Stage article that the subtle mismatching of gain devices would produce 2nd order harmonics to a degree at output levels around ~1 volt?

I wonder if you went out of your way to mismatch the FETs to a greater degree if that might accomplish the goal of increasing 2nd order harmonics while retaining all the other characteristics of the amp? Then, perhaps cascade a couple of these stages in series (depending on how much 2nd harmonic you want), and crank up the gain in the power amplifier to make up for volume.

I imagine you can do all this, but I can't help but wonder what
the point is.
 
Re: Re: Some more ideas

nowater said:


Say I'm listening to music at 1W output level with a 10W-20W amp, and distortion is maybe 0.2% 2HD (single ended amp). I want to tweak ..up ...2HD while still at 1W output (approx).

I'll experiment with several of the suggestions received earlier, except where they involve clipping the output stage. Maybe if I apply Erik's DC offset idea through an earlier gain stage e.g. preamp?

Grant


Nelson Pass said:


I imagine you can do all this, but I can't help but wonder what
the point is.


I'm glad you asked. Lets say the speakers produce 0.1-0.5% 2HD in the midband at moderate SPL. If the amp produces similar 2HD at that output level then, allowing for random phase differences, at *some* frequencies the distortions will sum to less than the speaker alone with a "perfect" amp. This cancellation effect is apparently measurable. If the frequency band of cancellation is in the band where the listener is most sensitive to hearing distrotion products (and the band of higher distortion is outside the most sensitive band for that particular listener), then the listener may overall hear less distorted music than with that "perfect" amp.

There is a fair bit of randomness in this, which is whay I want to "experiment" rather than try to "design" it all in. If the 2HD products happen to be summing in the critical band, then reversing the speaker leads will change it to cancellation. Which is very handy. But it is all for nought if the 2HD from the amp and speaker are not of similar magnitude. So I want to have a way to adjust the amp's 2HD without going anywhere near overload of the output stage.

So the point of increasing 2HD is to get better perceived (and actual for that critical band) fidelity!!!!

Grant
 
Hi Grant

If I understood you right you want to test the audibility of different levels of 2nd order distortion.

Why don't you leave your ZEN's performance as good as it is and put a line-level circuit in front of it that generates a controllable amount of 2nd order harmonics ?
Such a circuit could probably consist of an analog squaring device and a summing circuit. Or maybe a circuit with a controllable bias point.

Regards

Charles
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.