|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
108.168.121.47
In Reply to: RE: I would say nearly all amplifier designs are conceptually WRONG! posted by hahax@verizon.net on September 18, 2015 at 20:43:04
When accounted for, this is pretty minor.
I guess you aren't aware that triodes are the most linear voltage amplifying devices available?Just to clarify, the "pretty" second harmonic distortion, is actually the source material being more realistically conveyed.
What you are hearing is the lack of detrimental distortion.It's not that SETs sound "rich" per se, it's that other typologies sound cold.
The additional harmonic content from other typologies suffers more from spectral regrowth. (Harmonic distortions of harmonic distortions) It is these compounding distortions that null out and interact some of the intended material, rendering it cold and harsh sounding.
It gets even worse when the input signal is increased.
△ᴉʇɐuᴉɯnllI oᴉpn∀△
Edits: 09/18/15Follow Ups:
Sounds rich is correct. It always sounds rich because of high 2nd harmonic. What I find more important is that SE can't even do a simple sine wave correctly much less a complex music wave form. This means, since it is the nature of an SE circuit, that it can NEVER be correct. Perhaps other formats can be since this simple wave form distortion does not seem to be endemic to push/pull although, I admit, that there may be other problems that are unsolvable in push/pull.
And yes I have known for decades that a triode is the most linear amplifying device. But I would use it in an amp that is push/pull all the way(except for a 1st stage transformer used for phase splitting) without negative feedback. Not perfect but much preferable by me to SE. And I could at least get some real power from such a configuration.
"Sounds rich is correct. It always sounds rich because of high 2nd harmonic. "
-While rich sounding amps have high 2nd, it is not the high 2nd directly that makes it sound rich.
"What I find more important is that SE can't even do a simple sine wave correctly much less a complex music wave form."
-While not perfect, it is still better then anything else available.
"And yes I have known for decades that a triode is the most linear amplifying device. But I would use it in an amp that is push/pull all the way(except for a 1st stage transformer used for phase splitting) without negative feedback. Not perfect but much preferable by me to SE. And I could at least get some real power from such a configuration."
-A triode in PP has the exact same non-linearity you are concerned about as the SE. PP does not change the transfer function of the tube.
All you have created is symmetry, not linearity.
Your configuration is suitable if you need the power, can't argue that, but realize you have created a worse situation then having a single SEnded triode .
△ᴉʇɐuᴉɯnllI oᴉpn∀△
As a friend said 'color me perfect'. You choose your color and I'll choose mine. But I'll take the amp that's more dynamic(at all levels) every time and it isn't an SE especially one with just a few watts which is almost every SE. And you'll choose another color. But they are at best just different and you have no way to substantiate your color is better except for your personal use. You haven't proven at all which is the 'correct' road to follow.
And at least a PP can do a single sine wave so the negative half looks like the positive half flipped.
"And at least a PP can do a single sine wave so the negative half looks like the positive half flipped."
Actually, it distorts them both equally...and you get 3rd harmonic plus others...
of course but that's the point an unsymmetrical sine wave is an awful reproduction. Besides the asymmetry implies overt non-linearity which according to some very knowledgeable designers is fundamental to reproduction. David Berning considers linearity fundamental.
nt
try it! you know you want to!
OK, show us a device that is more linear then a triode, I won't hold my breath.If you find something, I recommend you quickly patent it.
△ᴉʇɐuᴉɯnllI oᴉpn∀△
Edits: 09/19/15
You are correct about the tube but not the circuit. And some triodes are more linear than others. The fabled 300B is just very good. The 2A3 is more linear and I'm told the 45 more linear yet. The problem is power.
However a screen drive 509 or 519 is a very linear triode and a push/pull pair is capable of well over 200 watts. And on another front the old Sony VFETs are a very linear solid state triode and they don't need an output transformer. Their curves look like a 2A3 and you can get power out of them too.
As I mentioned in my post to start this thread...the problem may not be so much the device...although even the VFets and screen driven 509s are not as linear as a good DHT, but using it in push pull. So, Nelson's SIT amp might be the answer but even that, I am told, doesn't converge with DHTs. It is also power limited.
I got pretty close with my NAT Symbiosis, which was a Class A SE(T) hybrid. Input and driver were tubes and the output was a single big industrial MOSFET mounted on the bottom of a vertically mounted monster heatsink. It delivered a true 100 watts of Class A power with no negative feedback and had the tone of a nearly perfect SET and SS like bass control and drive. HOWEVER, it was still missing something that was difficult to pin down and in the end I preferred true SET amps over it. I wish I could find some measurement data on it but there seems to be none because it was probably close to ideal but missing in a key area.
In terms of imaging it was more like great SS than holographic 3d.
According to a friend who loved and used the 2A3 the Sony VFET curves looked about as good as a 2A3. And from a friend who made commercial amps a screen drive 519 did not just show decent triode curve but very good ones.
By the way one of my big questions in audio and you do it too is class A SET. What is a non class A audio amp? SET has to be class A in a classic analog amp. There's no other real option.
The problem with VFETs and SITs are what's available power wise. There's not enough call from audio for VFETs. I think big ones were used in TV power supplies in the 70s. Audio alone can't support manufacture. But if it could we'd have high power triode performance without transformers.
On your hybrid you didn't have triode output so I'm not surprised you didn't quite like the sound. A MOSFET isn't a triode device while a VFET is a solid state triode.
nt
try it! you know you want to!
As far as I can see you have just two things in THIS world - the picture of the electronic puck as your signature, and that cheesy commercial line from the '50s under it.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: