|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
76.90.36.216
In Reply to: RE: That really won't make any difference. posted by Ralph on October 23, 2020 at 12:22:52
You don't believe me, you don't believe Harry Olson (shame on you) and you probably won't believe this so that's leaves me with not much to say.
Now don't try this with a triode wired 6550 or similar or a triode wired power pentode and don't try this with a 6as7g, none of them are linear enough. You will end up with a lot of distortion.
Direct heated triodes are special.
.
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 10/23/20Follow Ups:
its just that his charts don't depict what's happening- he didn't have access to test equipment with greater resolution.
Your load line appears to be a simulation. Have you measured what the amp actually does?
I've seen simulations off enough that I've learned to be distrustful... I can give you a number of anecdotes but that's all they are.
"its just that his charts don't depict what's happening- he didn't have access to test equipment with greater resolution."
He had the resolution to show the upper ordered harmonics on the pentode but not on the 2a3?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
I think the point is that these plots go down to "only" -80db and there is a lot going on below that level that could still be audible.
Ralph is right that there are still some higher order harmonics being generated by DHT SET, especially when the power goes up but they are balanced by the pattern of a relatively monotonic decay, which seems to be as important as the raw distortion harmonics themselves.
The 'First Watt' may really be the 'First 100 Milliwatts'.
Even listening to my arguably 105dB (sorry, not 106dB) Edgarhorns, offending odd order distortion becomes evident with but a bit of cranking up on the volume knob, at least with 2 watt indirectly heated triodes.
Will soon be bi-amping each driver with its own 2 watts of SET magic. Should give me an additional 3dB of 'headroom'. :-)
And with active x-overs I can get rid of 3dB of the 6dB padding-down on the JBL 2441. Might help a bit as well.
Make that the 'First 10 Milliwatts' :-)
You need to get them a good 20dB less and then they start to be 'nearly' inaudible. That's why its almost impossible to put enough feedback on a tube amplifier- usually the phase margin is exceeded. We make tube amps with arguably the best phase margin out there and its hard with our stuff too.
horns at CES some decade or so past.
A bit of headroom there.
If you keep the amp power down (and with horns in a room like that, its rarely going over 1 watt) then you don't run into problems.
The bottom line, DHTs sound cleaner that any other tube I have ever heard.
And I have heard a lot. I've been building for 40 years.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
You won't get any objection from me that this is largely a true statement (although the implementation matters).
I am interested in WHY that is the case.
If the 2nd and 3rd are in sufficient quantity, they will mask the presence of the higher orders. In addition, it appears that they also add to the sense of detail and soundstage. IMO more research is needed as to why and how of that last bit.
The 3rd is treated by the ear/brain system in the same way as the 2nd- it contributes to 'warmth' and 'body', unlike any of the other odd orders.
This masking seems to be why a tube amp (and especially one with triodes running no feedback) is able to sound smoother than solid state, even though its higher ordered content is at a higher level than solid state amps often have.
I'm really using some short hand here with no discussion of phase margins, gain bandwidth product and the like and the significance of that to the use of feedback, but that's been covered pretty well elsewhere. I'm just mentioned it to point out that the use of no feedback is part of the recipe.
2nd and 3rd order will not mask higher orders until the volume goes op (the masking effect is level dependent). At lower levels those higher orders will be exposed to our hearing.
I think the difference with tube amps (without feedback...with feedback it can sound as strident as SS amps can) is that at lower power outputs the high order harmonics are much lower than typical in an SS amp relative to the lower orders.
There seems to be more than masking at work as it seems the pattern itself seems to matter (hearing is much more than just the mechanism in your ear...there is anticipation of what to hear based on experience).
Jean Hiraga discussed this (he called it monotonic, I would call it exponential decay) and Cheever also highlighted this and goes into a lot more depth about the perception aspects and level dependence.
I was nutshelling it; we're on the same page here.
About the only way I can see around this is to have so much feedback that the amp can compensate for the distortion added by the feedback itself. But that takes a lot of feedback (over 35-40dB) which is really impractical in most tube amps. For that matter, its been impractical in most solid state amps too, which is why there is over a half century legacy of bright harsh solid state amps with few exceptions.
'This masking seems to be why a tube amp (and especially one with triodes running no feedback) is able to sound smoother'
this may sound [or be] stupid but I liken that to 'up-sampling' where what's missing is added by a synthesized continuation of the waveform
it has to be a more complex phenomena in the analog domain but that's how I visualize it, as an 'additive effect', and when listening your ear / brain does the rest of the task
regards,
but the 2a3 didn't have anything passed the 4th.That is my point.
BTW We don't need to see HD clear out to the 81st, a tiny bit of 7th and it's game over. 7th sounds really bad.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 10/26/20 10/26/20 10/26/20
And with any tube amp its almost impossible to put enough feedback in the circuit to really do what needs to be done. So the next best thing is no feedback at all IMO. But you wind up with higher output impedance.
However, Harry's gear didn't show anything above the 4th, but it was there nonetheless. Any modern test gear would show it.
-
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: