So, any amp distorts. Some in a way some in other way, some less, some more.
The difference is made by the listener. Some listeners love a mix of some kind of distortion no matter which, no matter how much, some listeners do love another mix of distortions.
This could be why sometimes listeners who don't like much distortion agree with measurements and why sometimes listeners do not. Since, the second ones love a lot of second harmonic distortion, forget the rest.
Is that it?
The difference is made by the listener. Some listeners love a mix of some kind of distortion no matter which, no matter how much, some listeners do love another mix of distortions.
This could be why sometimes listeners who don't like much distortion agree with measurements and why sometimes listeners do not. Since, the second ones love a lot of second harmonic distortion, forget the rest.
Is that it?
purity ...
Hi Brett,
polemic question...
A harmonics compensation can be made in SE topology, too.
Often you see that the driver stage makes an opposite distortion to that one
of the output tube.
Like in PP you have to match tubes and adjust the biasing to get "excellent results",
means a "non distorted" signal at the output.
As mentioned above (#23) you don't need triodes to get that result.
Thanks SY
It is not that I like a distorted sound,
I want to get what the triode gives, as unadulterated as possible.
This is why I don't like non linear driver stages and PP to get "excellent results", .
Hello lowtherdream,
all components participate more or less in the final sound.
It is in your hand how much surrounding components participate in the final sound.
A well designed circuit can reduce the influence of surrounding components very much.
E.g. the Western Electric (repeater amplifier)
http://www.nutshellhifi.com/library/Tube_Fest_Talk.html
a closed output current loop and
no influence of the output current in the driving voltage makes this possible.
You will see that most circuits don't care about this aspect.
Kind regards,
Darius
Hi Brett,
polemic question...
A harmonics compensation can be made in SE topology, too.
Often you see that the driver stage makes an opposite distortion to that one
of the output tube.
Like in PP you have to match tubes and adjust the biasing to get "excellent results",
means a "non distorted" signal at the output.
As mentioned above (#23) you don't need triodes to get that result.
Thanks SY
It is not that I like a distorted sound,
I want to get what the triode gives, as unadulterated as possible.
This is why I don't like non linear driver stages and PP to get "excellent results", .
Hello lowtherdream,
all components participate more or less in the final sound.
It is in your hand how much surrounding components participate in the final sound.
A well designed circuit can reduce the influence of surrounding components very much.
E.g. the Western Electric (repeater amplifier)
http://www.nutshellhifi.com/library/Tube_Fest_Talk.html
a closed output current loop and
no influence of the output current in the driving voltage makes this possible.You will see that most circuits don't care about this aspect.
Kind regards,
Darius
Larry Lomax said:So, any amp distorts. Some in a way some in other way, some less, some more.
The difference is made by the listener. Some listeners love a mix of some kind of distortion no matter which, no matter how much, some listeners do love another mix of distortions.
This could be why sometimes listeners who don't like much distortion agree with measurements and why sometimes listeners do not. Since, the second ones love a lot of second harmonic distortion, forget the rest.
Is that it?
It could be true if some people were born in reality of mechanical oscillating systems that produce sounds, but some people were born in electrical oscillating systems. But we all are from mechanical world where all oscillating, reflecting, conducting sound media generate low order harmonics: the louder the sound, the higher is their level, the wider is their specter. A pristine clean air is non-linear sound conductor. It distort. But we don't perceive such distortions as distortions. If the amp generate such distortions they are not perceived as distortions, but rather as higher loudness.
Wavebourn said:
Thanks Anatoliy,
This is I think the heart of the matter.
What is it that makes some amplifiers, speakers, sources,... just
sound more beautiful?
Klaus Heyne says he thinks the successful microphone capsule and
amplifier has a certain "euphemism" for the way the ear hears
sounds.
Now there is a difference if it's Klaus' ear or mine...
Cheers,
Michael
If second order distortion is crucial to musical beauty,
and matters truly what mic and pre-amp are used...
Then stands to reason that the distortion polarity of
a given SE amplifier either adds or sutracts from the
2nd order in the original recording...
Not all mics and instruments necessarily recorded in
the same distortion polarity, reversing the amp could
actually change the harmonic content.
And then the matter of Louspeakers and air are also
essentially 2nd order distortion devices...
Where does the madness end?
and matters truly what mic and pre-amp are used...
Then stands to reason that the distortion polarity of
a given SE amplifier either adds or sutracts from the
2nd order in the original recording...
Not all mics and instruments necessarily recorded in
the same distortion polarity, reversing the amp could
actually change the harmonic content.
And then the matter of Louspeakers and air are also
essentially 2nd order distortion devices...
Where does the madness end?
kenpeter said:
Ken,
I try not to think about it too much but, during my last
amp change/mod/bring-up with a goofy bias scheme I
made an observation:
Female voice waveforms are frequently f2-rich to the
extent that the waveform almost looks like a PP with
one side gone bad. (Norah Jones, Diana Krall, etc.)
So I noticed that the output level where my amp
misbehaved was noticeably different depending on the
input polarity. I found this out by playing the same
section of the track over and over through my
instrumentation while varying different things.
Then I noticed that on some tracks the vocal waveform
seemed consistently inverted, with the "peaks" going
the other direction. Throughout a given track the polarity
of the f2 was consistent.
Is this something inherent in the female voice or am I
seeing a tube mic or what?
Cheers,
Michael
lowtherdream said:
I built a Loftin White 801 amp, which you can find with a quick search here. Based on my results with that, I would put in a plug for Darius's design, as found here:
http://rl12t15.blogspot.com/2008/03/ecc83-300b-darius-loftin-white.html
It's not complicated, though conceptually it's pretty sophisticated. And to simplify, you can eliminate all the stuff between the 8 uf supply cap and input cathode. That's a circuit to cancel filament hum from the 300B. In conversation,(I spoke to Darius earlier today - he was nice enough to call and discuss a little the concept of the "triode design factor", which doesn't seem to have a formal name anywhere.) he mentioned that with a 2.5V filament version of the 300 (or with a 2A3), compensation for AC filament noise won't be necessary.
Note that because the supply ripple is canceled at the input tube cathode, it is simple and uses small cap values, which can all be film. Two stages, direct coupled, input to output.
Sheldon
Something to read
Since you seem to be interested in SET design and full range single driver speakers, I thought that you would enjoy these articles and discussions:
http://www.diyaudio.com/forums/showthread.php?threadid=111973
http://www.audiopax.com/papers_part1.htm
http://gboers.xs4all.nl/daisy/home/g3/139/g1/loudspeakers/systemdistortion.html
I am not writing to give you any simple answers.lowtherdream said:
Since you seem to be interested in SET design and full range single driver speakers, I thought that you would enjoy these articles and discussions:
http://www.diyaudio.com/forums/showthread.php?threadid=111973
http://www.audiopax.com/papers_part1.htm
http://gboers.xs4all.nl/daisy/home/g3/139/g1/loudspeakers/systemdistortion.html
Some random thoughts on the designs you were looking at.
If you decide on one of them, build it without modifications.
My rule of thumb from a design perspective is:
Build an existing design. - Moderately Easy
Modify a design after its built and improve it = Moderately Hard
Design your own - Harder than Modifying a working design.
Modify another design - Harder than designing your own.
Some of the designs have features that are not apparent unless you have both background and experience.
All of the designs you posted have intriguing features.
Thanks for pointing them out.
On a lighter note, I think you will really enjoy your Simple SE.
Good Luck.
Doug
If you decide on one of them, build it without modifications.
My rule of thumb from a design perspective is:
Build an existing design. - Moderately Easy
Modify a design after its built and improve it = Moderately Hard
Design your own - Harder than Modifying a working design.
Modify another design - Harder than designing your own.
Some of the designs have features that are not apparent unless you have both background and experience.
All of the designs you posted have intriguing features.
Thanks for pointing them out.
On a lighter note, I think you will really enjoy your Simple SE.
Good Luck.
Doug
I'll just toss on a couple more briquettes. Hope you don't mind some opinionated rants from an old fart on the other side of the boat.
First, if you haven't built anything then it might be a really good idea to build a well documented kit. I'm saying this in regard to getting a physical experience and feel for good practice in terms of design, layout, construction and safety. Many experienced builders liked studying the old military electronics manuals and one good reason why is that they never put a bunch of stuff in a lesson that wasn't essential to learning the lesson they were trying to teach, so you don't get overloaded with indigestible details . . . . .and every section or chapter winds up with a few questions that lets you know very clearly whether you were actually getting it or not. Building a kit has the same sort of feel. You might be absolutely sure you did it right . . . .so why don't it work? The feedback is clear and with a well documented, well supported kit you can learn a lot with help for working out your own particular oversights.
There are many kits out there as you know. Personally I think the Bottlehead road is a good one.
Second, Don't let anybody put you off the Axiom. No I haven't heard one but the guys who built it aren't stupid and the design is beautiful and elegant within it's own genre.
I liked the look of it from the first moment I saw it and later built a sort of mini version with only two tiers , 6BX7 driver and 6336 output. It was a worthwhile project and I learned a great deal from doing it . The diodes do make a difference though maybe not necessarily the one YOU want. In my amp the cathode diodes worked 'miracles' doing a lot to clean up the sound of an output tube with distortion that is fairly high relative to the popular DHT's (300B, 2A3, 45, 845 et al.).
On the other hand I didn't like the sound of the diode in series with the plate. It did make the sound smoother but there was also a certain subtle ethereal airiness to the sound that I found dissatisfying. I'll stress again though that this was in an amp I built and with very different tubes, transformers and design. Of all the reports I've read by people who have actually heard the Axiom I didn't see any strong negatives.
My only reluctance for the Axiom as a starter project is (as somebody posted above) that while each tier doesn't look like a lot of voltage , the potential between the top and bottom is high (was it 730VDC?). My personal feeling is that it's something akin to choosing to do your first driving lesson in an F1 racer.
Just a technical point - The 300B's bias does not come from the resistor shown in the schematic. That's a 100 Ohm hum balance pot. With the two halves of 50 Ohms each in parallel the pot is only adding 25 Ohms to the circuit. Not enough to bias a 300B.
And one last swing, In response to your "I guess the Axiom is not what I expected it to be. Probably not worth $1500 to find out if good or not". Don't let anybody talk you out of being interested in the circuits you're genuinely interested in. You have to start somewhere , You can't get it all in the first go, and you may as well be building the one you're building because you have a heart-on for it !
Well there you go. Hope you don't mind these meanderings.
First, if you haven't built anything then it might be a really good idea to build a well documented kit. I'm saying this in regard to getting a physical experience and feel for good practice in terms of design, layout, construction and safety. Many experienced builders liked studying the old military electronics manuals and one good reason why is that they never put a bunch of stuff in a lesson that wasn't essential to learning the lesson they were trying to teach, so you don't get overloaded with indigestible details . . . . .and every section or chapter winds up with a few questions that lets you know very clearly whether you were actually getting it or not. Building a kit has the same sort of feel. You might be absolutely sure you did it right . . . .so why don't it work? The feedback is clear and with a well documented, well supported kit you can learn a lot with help for working out your own particular oversights.
There are many kits out there as you know. Personally I think the Bottlehead road is a good one.
Second, Don't let anybody put you off the Axiom. No I haven't heard one but the guys who built it aren't stupid and the design is beautiful and elegant within it's own genre.
I liked the look of it from the first moment I saw it and later built a sort of mini version with only two tiers , 6BX7 driver and 6336 output. It was a worthwhile project and I learned a great deal from doing it . The diodes do make a difference though maybe not necessarily the one YOU want. In my amp the cathode diodes worked 'miracles' doing a lot to clean up the sound of an output tube with distortion that is fairly high relative to the popular DHT's (300B, 2A3, 45, 845 et al.).
On the other hand I didn't like the sound of the diode in series with the plate. It did make the sound smoother but there was also a certain subtle ethereal airiness to the sound that I found dissatisfying. I'll stress again though that this was in an amp I built and with very different tubes, transformers and design. Of all the reports I've read by people who have actually heard the Axiom I didn't see any strong negatives.
My only reluctance for the Axiom as a starter project is (as somebody posted above) that while each tier doesn't look like a lot of voltage , the potential between the top and bottom is high (was it 730VDC?). My personal feeling is that it's something akin to choosing to do your first driving lesson in an F1 racer.
Just a technical point - The 300B's bias does not come from the resistor shown in the schematic. That's a 100 Ohm hum balance pot. With the two halves of 50 Ohms each in parallel the pot is only adding 25 Ohms to the circuit. Not enough to bias a 300B.
And one last swing, In response to your "I guess the Axiom is not what I expected it to be. Probably not worth $1500 to find out if good or not". Don't let anybody talk you out of being interested in the circuits you're genuinely interested in. You have to start somewhere , You can't get it all in the first go, and you may as well be building the one you're building because you have a heart-on for it !
Well there you go. Hope you don't mind these meanderings.
Axiom ...
Good morning Forum,
Hello Hearinspace,
thank you for the describing the influence of the diodes in the sound.
This is what I wanted to point out in my post #28.
Do you understand why the diodes are used?
The 6BE3 makes biasing for the 6J5.
The other diodes 6CG3 are delayed switches.
The biasing voltage comes from the voltage drop at the coil resistance.
Please note the DC Voltage drop at the coils, the coils acts as wire wound resistors.
This is what I wanted to say in post #24.
The previous tube sets the biasing for the following tube.
This is why the delayed (diode) switches are needed.
No "axiom", no "mystery", just timing (post #28).
Kind regards,
Darius
Good morning Forum,
Hello Hearinspace,
thank you for the describing the influence of the diodes in the sound.
This is what I wanted to point out in my post #28.
Do you understand why the diodes are used?
The 6BE3 makes biasing for the 6J5.
The other diodes 6CG3 are delayed switches.
The biasing voltage comes from the voltage drop at the coil resistance.Please note the DC Voltage drop at the coils, the coils acts as wire wound resistors.
This is what I wanted to say in post #24.
The previous tube sets the biasing for the following tube.
This is why the delayed (diode) switches are needed.
No "axiom", no "mystery", just timing (post #28).
Kind regards,
Darius
Hi Darius .
I don't know the DCR of the winding so don''t know what percentage of the bias it shares with the diode. I'll email Dennis and ask if he has time to say anything here.
Have you read the development notes on the Axiom over on the Chimera labs site? It's interesting stuff. They talk about switching out cathode resistors for diodes and getting reduced distortion. . . . .
I don't know the DCR of the winding so don''t know what percentage of the bias it shares with the diode. I'll email Dennis and ask if he has time to say anything here.
Have you read the development notes on the Axiom over on the Chimera labs site? It's interesting stuff. They talk about switching out cathode resistors for diodes and getting reduced distortion. . . . .
More from the Department Of Free Advice...
Imho, ur getting too fine and specific.
You'll quickly find that things do not usually work as advertised, especially when you build it yourself according to someone else's plans. Two big reasons: your components are probably different, and their idea of what sounds good is not yours at all.
Another big problem: let's say you copied a very good SE design, say one that uses a "300B". Now let me ask you a question - which "300B" are you going to use?
I can guarantee that a Western Electric 300B will sound quite noticeably different than say an EH 300B. For me, perhaps to the point where I don't want to listen to an EH 300B, although it is a good tube.
So, again the choices that inevitably must be made, and the compromises that end up being made have a significant effect on the results. It's not as simple or straightforward as many of these 'writers' (out there on the web) would lead you to believe.
Plus, once you yourself actually build some amp and can substitute in tubes, resistors, caps and even change the circuitry - you'll start to notice just how variable the whole thing actually is.
This is much more true for a ZFB (zero loop feedback) type amp, in this case a tubes ZFB amp, although it applies to virtually any audio device.
It just ain't so simple.
OTOH, as I said earlier, stop worrying about any of this stuff. It makes nil difference to you at this point. Simply build the thing that you can build given your resources and parts, nothing else matters. Start.
The easiest thing to do is to look about for an old tube type receiver, and scavenge the power transformer and output transformers (be careful, you might run into a valuable one - in which case you sell it on ebay and buy parts), build something. If you don't want P-P, then "parafeed" the output iron as a SE so you can use the PP iron... or buy some random iron on ebay, or even buy some new Hammond... etc... the hardest part of a tube amp is the iron. So, get iron, and that dictates the design possibilities, fit your schematic/circuit accordingly - even if that means copying a "published" design.
The only caveat I'd give is "avoid pentodes". That's unless you like the "pentode sound" - I generally don't, ymmv.
There are quite a few tube-o-philes lurking in Montreal, btw, and there is a pretty big show up there each year - you might meet others by going? Dunno.
Imho, ur getting too fine and specific.
You'll quickly find that things do not usually work as advertised, especially when you build it yourself according to someone else's plans. Two big reasons: your components are probably different, and their idea of what sounds good is not yours at all.
Another big problem: let's say you copied a very good SE design, say one that uses a "300B". Now let me ask you a question - which "300B" are you going to use?
I can guarantee that a Western Electric 300B will sound quite noticeably different than say an EH 300B. For me, perhaps to the point where I don't want to listen to an EH 300B, although it is a good tube.
So, again the choices that inevitably must be made, and the compromises that end up being made have a significant effect on the results. It's not as simple or straightforward as many of these 'writers' (out there on the web) would lead you to believe.
Plus, once you yourself actually build some amp and can substitute in tubes, resistors, caps and even change the circuitry - you'll start to notice just how variable the whole thing actually is.
This is much more true for a ZFB (zero loop feedback) type amp, in this case a tubes ZFB amp, although it applies to virtually any audio device.
It just ain't so simple.
OTOH, as I said earlier, stop worrying about any of this stuff. It makes nil difference to you at this point. Simply build the thing that you can build given your resources and parts, nothing else matters. Start.
The easiest thing to do is to look about for an old tube type receiver, and scavenge the power transformer and output transformers (be careful, you might run into a valuable one - in which case you sell it on ebay and buy parts), build something. If you don't want P-P, then "parafeed" the output iron as a SE so you can use the PP iron... or buy some random iron on ebay, or even buy some new Hammond... etc... the hardest part of a tube amp is the iron. So, get iron, and that dictates the design possibilities, fit your schematic/circuit accordingly - even if that means copying a "published" design.
The only caveat I'd give is "avoid pentodes". That's unless you like the "pentode sound" - I generally don't, ymmv.
There are quite a few tube-o-philes lurking in Montreal, btw, and there is a pretty big show up there each year - you might meet others by going? Dunno.
Hi Darius,
If you'll take another another look at the schematic you'll see that the 6CG3 heaters are not conducting the cathode current as DHT filaments would, so there's no DC current through the heater supply secondaries and so no biasing VDC drop across them. The connections shown on the schematic at F15,F10 and F7 can make us think more than we have to - they're just connecting the centre tap of the windings to their respective circuit's local ground. The bias is coming from the drop across the 6CG3s.
Powering the amps up, all heaters and filaments are switched on together first. B+ is applied separately after the heaters are warmed up.
Thanks
If you'll take another another look at the schematic you'll see that the 6CG3 heaters are not conducting the cathode current as DHT filaments would, so there's no DC current through the heater supply secondaries and so no biasing VDC drop across them. The connections shown on the schematic at F15,F10 and F7 can make us think more than we have to - they're just connecting the centre tap of the windings to their respective circuit's local ground. The bias is coming from the drop across the 6CG3s.
Powering the amps up, all heaters and filaments are switched on together first. B+ is applied separately after the heaters are warmed up.
Thanks
Hearinspace said:
The voltage drop is across the inductors (CHA 1100 and 1200) in series with the cathodes. They act as cathode resistors, bypassed to B+ for the respective stage. So bias is determined by the diode and inductor in series. The inductor AC resistance will be high - almost a current sink/source, so I would imagine the bypassing cap is a little more important than with a regular resistor?
Sheldon
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.