These are five variations on a simple SE. Small differences, but which one would you go for in the real world? It is mostly about 2H wrt 3H.
You'd never hear any difference between them with the human ear at that low of a distortion level, if this is what the as bult amp actually does. At least not the differences in THD. If your goal is making under 0.1% THD, I'm curious why you are messing with a SE tube amp, as it's not going to sound like one.
I'd rather see the circuit rather than the results of some simulation that may or may not reflect the reality to the as built results. Like I said, if these were the actual results, the human ear wouldn't be able to decern the deference between 0.04 and 0.07 THD.
Forget about THD! It is not about the circuit, it´s about harmonic pattern, mainly about level between 2nd and 3rd. Why post if you are not capable of answering the simple question: ”Which one would you go for?”
Last edited:
None of them.
Simulation based on mathematical formulas, which are has finite variables.
It's approaches the reality, but shows the ideal picture.
Can point out the gross differences, defines corner points in operation, but due to the "finite" the "picture" is the hand painted (idealized by designers), not "hardcopy" of origin functions.
If you see real measurements, each of them differs from it's simulation. The contours, the primary colours are the same, but the shades are differs.
With simulation good audio products can be made, but nobody knows how will it sound.
Sample: typical 300B amplifier at moderate 4W power.
As can you see, the difference between two output tubes (and OPT) larger (in the practice the gates FFT almost identical), than you shows in your simulations.
Simulation based on mathematical formulas, which are has finite variables.
It's approaches the reality, but shows the ideal picture.
Can point out the gross differences, defines corner points in operation, but due to the "finite" the "picture" is the hand painted (idealized by designers), not "hardcopy" of origin functions.
If you see real measurements, each of them differs from it's simulation. The contours, the primary colours are the same, but the shades are differs.
With simulation good audio products can be made, but nobody knows how will it sound.
Sample: typical 300B amplifier at moderate 4W power.
As can you see, the difference between two output tubes (and OPT) larger (in the practice the gates FFT almost identical), than you shows in your simulations.
The harmonics generate from the DHT power tube can easily be found with use of the loadline calculator or through simulation like LTspice. However, the major issue would be the harmonics from the driver stage that will further mix with signal at power stage. At full output power level, most of the driver tube would have approximately 4~5% distortion signal by itself.
Hey guys, why not concentrate on my question about 2nd vs 3rd harmonic.
It is NOT about measuring, it is NOT about sims and NOT about loadlines.
It is is about HARMONIC PATTERN, which was ony exemplified
So, any opinions about a suitable relationship between the gradually falling harmonics?
It is NOT about measuring, it is NOT about sims and NOT about loadlines.
It is is about HARMONIC PATTERN, which was ony exemplified
So, any opinions about a suitable relationship between the gradually falling harmonics?
Your questions are related to the virtual world without any connection with real
When you attach the amp to the real load the shapes of the harmonics change in relation with it.
No way
Ad this aspect is not related to a specific circuit, less or more any amps have a own results.
Your is only a theoretical discussions without any chance to get a perfect answer.
The best test is the THD vs Frequency at different level so you can see how it vary , mainly for tube amp where the limits at low and at high frequency are easy to see.
Walter
Get it! It is an attempt to have a theoretical discussion about harmonic pattern in general! So far it seems hopeless.
So if I say that 3rd and 5th harmonic should be a lot higher than 2nd, you wouldn’t mind 😉?
So if I say that 3rd and 5th harmonic should be a lot higher than 2nd, you wouldn’t mind 😉?
This is impossoble to see in a SE ; in case, you haven't an amp but a square wave generator
Your assumption is wrong as question
Walter
Lars,
You didn't look closely enough my attached picture in #6.
Two identical (bought as factory selected pair, and measured its as pair -in ONE operating point-) 300B producing same H2, H3, H4, then other harmonics begins to differ.
One tube (red) H5 10 dB lesser than another, and it's H6 about 7 dB lesser. This tube's other components disappears.
The strictly monotonous of "green" tube line up to H6, then falling.
The "red" has lower THD, than "green"... but the difference about 0.06%.
The "green" has strictly monotonous FFT components (H2..H6) but H7... too .... 100dB lower than 1kHz carrier.
The "red" has strictly monotonous FFT components (H2..H4), but then equal size H5, H6, lower H7 .. and nothing.
Is "green" better, than "red"?
Is FFT components over H4 - more than -90 dB! - are very important?
IMO not.
If the first three component are "OK" -monotonous- (measuring!) presumable that the design (tubes, non linear components, layout, even harmonic cancellation etc.) also "OK", no too big mistakes.
But this is my opinion ... based on more than forty years practice ....although I could be wrong. 🙂
You didn't look closely enough my attached picture in #6.
Two identical (bought as factory selected pair, and measured its as pair -in ONE operating point-) 300B producing same H2, H3, H4, then other harmonics begins to differ.
One tube (red) H5 10 dB lesser than another, and it's H6 about 7 dB lesser. This tube's other components disappears.
The strictly monotonous of "green" tube line up to H6, then falling.
The "red" has lower THD, than "green"... but the difference about 0.06%.
The "green" has strictly monotonous FFT components (H2..H6) but H7... too .... 100dB lower than 1kHz carrier.
The "red" has strictly monotonous FFT components (H2..H4), but then equal size H5, H6, lower H7 .. and nothing.
Is "green" better, than "red"?
Is FFT components over H4 - more than -90 dB! - are very important?
IMO not.
If the first three component are "OK" -monotonous- (measuring!) presumable that the design (tubes, non linear components, layout, even harmonic cancellation etc.) also "OK", no too big mistakes.
But this is my opinion ... based on more than forty years practice ....although I could be wrong. 🙂
Waltube, so you can answer my last question that was intentionally wrong. Why not give your view on how 2nd and 3rd, 5th and on should be related?
My main issue was 2nd to 3rd and how the fall in harmonics should look, which I exemplified.
They can’t be related because in the real world they continuosly change ( not predictable) with level and the real load
Of course starting from a good project
Du har fått PM...
Nähä, det var inte möjligt.
Jag har en del funderingar över detta och tar gärna en diskussion per telefon eller mail. Skicka något till flex2@live.se
För att svara på din fråga.
Den sista.
Den stämmer mest överens med dom preferenser jag har personligen.
Nähä, det var inte möjligt.
Jag har en del funderingar över detta och tar gärna en diskussion per telefon eller mail. Skicka något till flex2@live.se
För att svara på din fråga.
Den sista.
Den stämmer mest överens med dom preferenser jag har personligen.
Last edited:
I don't think the differences between your 5 variations is enough to be audible. There's other stuff in a Distortion profile which is missing which have big influence on audibility & preference.
You need to show how this harmonic profile changes with level & frequency too.
You need to show how this harmonic profile changes with level & frequency too.
It is hopeless because you are either asking the wrong question - or asking the right question the wrong way!
Looking at the answers you got this should be clear to you.
If you had started out asking a clear question about what balance people prefer between harmonics and why - as a thesis.
And also related to other influencing parameters
Then this could have been a fruitfull discussion.
The actual clear answers you got from several people so far, is that none of your five "THD simulations" hold sufficient information to valid a clear answer on preference. - Got it? 😉
But I could ofcourse be wrong, and you really just want us to recommend one of your tuned circuits, for an upcoming build based just on the curves.
That will then be like asking people how curved they prefer their banana.
Please take this in a positive manner - it is somehow related also to the H2 generator experiments, from which there may be a lot of experiences to harvest.