SY said:
That's basically what we are discussing, a shunt reg with a few precautions against potential instability - did somebody think choke here?
Could anybody post the Futterman schema? Some of you guys are pretty good at coming up with stuff like this. Is Futterman a modern circuit? If we can learn from it... right on!
Joe R.
A CCS in the tail of an LTP. not too uncommon. I use an IC, either a 10M45 or an LM334. This is mentioned in Morgan's book.
The zener or floating regulator between the screen and plate has been discussed on this forum before. One such discussion is here:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=97384&highlight=
There was another thread (which I can't find right now) where a similar subject was discussed. I pointed out that a zener or gas tube can be used to drop the screen voltage so that a sweep tube with a low screen grid voltage rating can be used in triode or UL. The drawback is that the screen grid can be driven to zero or negative as the tube attempts to conduct heavilly leading to nonlinearity.
I have a new driver board under development that will allow the AC drive and DC bias voltage on each grid to be adjusted independently. Yes there is some considerable sand in this one.
Zero feedback is common in SE amps. I use no feedback in my 300Beast amd which is push pull, but uses triodes (300B's).
The LEM. I admit to being curious, but with no further details, I can assume that you are adding a high frequency bias in a similar fashion to the bias added to linearize magnetic tape recording. Curiosity may force me to connect a signal generator into one of my amps.
The picture of the amp on your web site gives that secret away.
Antripodean said:
He has informed me that he has the amp. It is based on the very schematic - topology - that I have posted above. He has emailed me, but he is now reluctant to go ahead and post his experience with the amp and how it sounds. I gather that he likes it.
As the basic circuit and how it works IS posted, it should be OK for him to go ahead?
Joe R.
Someone asked about this one before. I explained it here:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=139765&highlight=
tubelab.com said:
...damn, it dropped the 'twisted cascoding' comment. George, it looks like you've done a good job of describing a cascode to me...
I'd be quite tempted to implement this one with DN2540 FET's...one set as CCS the other as the follower. Couldn't care less what the PS above is doing, but making it ripply for the amplifier stage seems unwise.cheers,
Douglas
tubelab.com said:
Quite true, but in tube amp circuits it is 'modern' and hence does fit this thread. And we did it before MJ wrote about it and had a discussion re the use of Fets etc used as CCS in tube circuits. But again, in terms that most tube designs, especially based on Pentodes, are really not that much different from the fifties Hafler years. You know that because I know you are an innovator. Now that is the key word here and that makes it 'modern.'
So I have noticed. We did likewise. www.enjoythemusic.com/Magazine/viewpoint/0104/aachapter52.htm
You've got it! Indeed, according to what I remember Menno telling me, both were patented in the early fourties. Tapes took off but the other got lost, maybe in the haze of war? So the application is modern. Does that qualify? Me thinks so.
I suggested idea this to Menno (not knowing about the patent, honestly, at which point he set me straight), after asking what sort of bandwidth he aims for in his txs. His answer was 100KHz and I suggested using the upper part of its bandwidth to keep the tx busy. But do your calculations re the level. Taking advantage of the step-down ratio means we can aim for quite a few volts peak-to-peak on the Anode/Primary side (which is where it's needed) but still making sure that the Tweeter does not dissipate more than 10mW or so. Higher ratios are obviously an advantage, but I have done most 20:1 or more.
So give it a try and report back.
My experiments are mentioned in this AES 2007paper: www.mennovanderveen.nl/nl/download/download_1.pdf
Joe R.
Lots Of Time? Spend It Wisely!
Wow, guys, you seem to have loads of time (unlike me). Either you are old(er) and have adult children, or your wives do all the house-work and care for the children like in the 50s (tube guys, eh?) -- or you are single and have all the time in the world to devote to yourselves.
Now I understand why they were trying to disrupt our power grids 10 years ago! They tought we would be crippled by that for years to come... "what a mistake to make"...
Anyway, I am sorry to hear that. And happy to hear that, because it make me less sorry for not living in the land of opportunities. At least I get to AC heat my DHTs withouth fear of surges and brownouts
It would have done, if we were here to show our provess with tubes and circuits. Instead, I just wanted to point out the difference: I am happy to hear that you get to do what you want, although it is difficult to make ends meet doing it. I do not do what I like to do, and if I were doing it in some of your neighbourhoods, I would be quite well off, maybe deciding that it is better for me to be well off than to do what I like. That is the irony.
BUT, what I was trying to point out, is the approach to things. Not that you are into marketing for the big gain, like some SONY or PANASONIC, but since you do it commercially, you are inclined towards finding fancy names for things, and making relatively simple things look more complicated to the eyes of those not skilled enough to do it on their own. As an example, each and every diyer is capable to solder a resistor between points A and B (i.e. anode and B+) but if instead of a resistor we have to put a CCS or similar, things get complicated for most -- both in terms of designing it, and making it. That is where we differ: I am interested in the simplest approach that works, and try to improve in such a manner that everyone can replicate it; on the other hand, you facilitate your public with i.e. CCSs or complete kits -- that way they can make your designs just as easily as they would be able to make some of mine. The Bottlehead is a similar example.
I hope this was not too long an explanation as to allow confusing ideas one more time. Although, frankly, in some threads I get the impression that everyone is there for the confusion...
Just imagine how many new designs and improvements, how many new circuitry to carry your names could be made in the time spent on those forums?
Wow, guys, you seem to have loads of time (unlike me). Either you are old(er) and have adult children, or your wives do all the house-work and care for the children like in the 50s (tube guys, eh?) -- or you are single and have all the time in the world to devote to yourselves.
Now I understand why they were trying to disrupt our power grids 10 years ago! They tought we would be crippled by that for years to come... "what a mistake to make"...
Anyway, I am sorry to hear that. And happy to hear that, because it make me less sorry for not living in the land of opportunities. At least I get to AC heat my DHTs withouth fear of surges and brownouts
It would have done, if we were here to show our provess with tubes and circuits. Instead, I just wanted to point out the difference: I am happy to hear that you get to do what you want, although it is difficult to make ends meet doing it. I do not do what I like to do, and if I were doing it in some of your neighbourhoods, I would be quite well off, maybe deciding that it is better for me to be well off than to do what I like. That is the irony.
BUT, what I was trying to point out, is the approach to things. Not that you are into marketing for the big gain, like some SONY or PANASONIC, but since you do it commercially, you are inclined towards finding fancy names for things, and making relatively simple things look more complicated to the eyes of those not skilled enough to do it on their own. As an example, each and every diyer is capable to solder a resistor between points A and B (i.e. anode and B+) but if instead of a resistor we have to put a CCS or similar, things get complicated for most -- both in terms of designing it, and making it. That is where we differ: I am interested in the simplest approach that works, and try to improve in such a manner that everyone can replicate it; on the other hand, you facilitate your public with i.e. CCSs or complete kits -- that way they can make your designs just as easily as they would be able to make some of mine. The Bottlehead is a similar example.
I hope this was not too long an explanation as to allow confusing ideas one more time. Although, frankly, in some threads I get the impression that everyone is there for the confusion...
Just imagine how many new designs and improvements, how many new circuitry to carry your names could be made in the time spent on those forums?
Interesting paper on tube OP Amps from 1947, don't see a CCS in this one, or even use of the negative supply for a long tail:
http://www.philbrickarchive.org/operational_amplifier.htm
More:
http://www.philbrickarchive.org/
Mention of tube OP Amps as far back as 1943 here:
http://www.uoguelph.ca/~antoon/gadgets/741/741.html
Pete B.
http://www.philbrickarchive.org/operational_amplifier.htm
More:
http://www.philbrickarchive.org/
Mention of tube OP Amps as far back as 1943 here:
http://www.uoguelph.ca/~antoon/gadgets/741/741.html
Pete B.
Joe Rasmussen said:
Hi Douglas
Finally, the -10V on the Suppressor Grid. Can only be done to one tube I know of. Can you guess which one? Puts the Suppressor into cut-off mode relative the Cathode and should be get out of the way. Electrons only go one way.
Joe R.
I think you should get out more. Separate supressor grids for only one tube? Try HY69, or perhaps 8001, or H-K 57, or the 5-125B...let alone the TV sweeps; lots have a separate connection.
cheers,
Douglas
Joe,
Does VTO impose an additional limit on anode voltage excursion
(swing) and therefore anode efficiency?
What does unity coupling require of the loudspeaker system?
Are there small-signal measurements that show the impact of
the improved available permeability due to HF bias?
What are the tradeoffs associated with injecting the HF bias
at different stages in the amp, i.e. at the OPT directly vs. the
input? Does it do anything helpful or harmful in the low signal
stages?
Anatoliy,
Do you think the Russell Hamm paper is hogwash? What kind of
microphone preamp do you prefer? Do you think he has a point
wrt the dynamic range of microphone signals? Sure, real world
data would make a stronger case.
I don't buy the "electromechanical" idea at face value, but it
seems to me that the dynamic range of the music being
recorded has a fair chance of being greater than what can
be accommodated in the recording. Does a vacuum tube mic
and/or recording channel provide a more musical response to
the inevitable overloading on peaks?
I can't explain why tubes may be better to drive loudspeakers
but it does seem to be the case that I have been able to achieve
better results, e.g. less ear fatigue for one, using a good match
of tube amp to loudspeakers than I could with the usual SS
amp driving low efficiency monitors. Accuracy? I hear more
detail with the tube amp setup which is what matters to me
in mixing.
My $.02 on modernity:
Modern tube amp design is by definition tube amp design using
modern tools, methods, and components (Spice, FFT, transistors...)
I don't see how silicon pollutes a tube amp circuit any more or less
than iron, nickel, cobalt, carbon, paper, oil, polypropylene, mica... if
used in such a way as to achieve the tube circuit designer's goals.
How much signal swing will you get without the energy storage
in the coupling choke or primary winding? Why not use a MOSFET
and power supply to do the same thing?
I don't buy the signal path argument though (been there recently).
Everything's in the signal path; the question is how do you get gain?
Controversial idea:
Linearity is second order to sound as sound is by definition nonlinear
Cheers,
Michael
Does VTO impose an additional limit on anode voltage excursion
(swing) and therefore anode efficiency?
What does unity coupling require of the loudspeaker system?
Are there small-signal measurements that show the impact of
the improved available permeability due to HF bias?
What are the tradeoffs associated with injecting the HF bias
at different stages in the amp, i.e. at the OPT directly vs. the
input? Does it do anything helpful or harmful in the low signal
stages?
Anatoliy,
Do you think the Russell Hamm paper is hogwash? What kind of
microphone preamp do you prefer? Do you think he has a point
wrt the dynamic range of microphone signals? Sure, real world
data would make a stronger case.
I don't buy the "electromechanical" idea at face value, but it
seems to me that the dynamic range of the music being
recorded has a fair chance of being greater than what can
be accommodated in the recording. Does a vacuum tube mic
and/or recording channel provide a more musical response to
the inevitable overloading on peaks?
I can't explain why tubes may be better to drive loudspeakers
but it does seem to be the case that I have been able to achieve
better results, e.g. less ear fatigue for one, using a good match
of tube amp to loudspeakers than I could with the usual SS
amp driving low efficiency monitors. Accuracy? I hear more
detail with the tube amp setup which is what matters to me
in mixing.
My $.02 on modernity:
Modern tube amp design is by definition tube amp design using
modern tools, methods, and components (Spice, FFT, transistors...)
I don't see how silicon pollutes a tube amp circuit any more or less
than iron, nickel, cobalt, carbon, paper, oil, polypropylene, mica... if
used in such a way as to achieve the tube circuit designer's goals.
How much signal swing will you get without the energy storage
in the coupling choke or primary winding? Why not use a MOSFET
and power supply to do the same thing?
I don't buy the signal path argument though (been there recently).
Everything's in the signal path; the question is how do you get gain?
Controversial idea:
Linearity is second order to sound as sound is by definition nonlinear
Cheers,
Michael
Many of my amps use low cost OPT's where the HF roll off is too close to the audio band, so this technique may not help them. I do however have a monster set of Plitron toroids that I got from their clearance corner. The upper 3db point is about 75 KHz, so I may experiment on them the next time I have them hooked up.
Yes, kids are grown and gone. Wife is gone for weeks at a time taking care of her mother who has cancer and lives 1200 miles away. I do however spend 50 - 60 hours a week trying to keep my job, which has limited my tube time lately. Life is not always full of extra time though. I had 6 cancer surgeries last year which totally killed all of my tube time for 2008. I have just about caught back up to where I was a year and a half ago. I should have some extra time soon and I will be doing some experiments.
Random ripple would be a bad thing, but in this case the "ripple" is an exact copy of the output signal. The PSRR of a cathode follower is 20 to 40 db so little of this "ripple" makes it to the output. If it did it would just add to the output signal. This PSRR is the reason I can get away with a less than exact copy in the DSP design. My A/D is only 10 bits.
The desired effect is a reduction in distortion since the plate to cathode voltage across the tube remains constant. This reduces the distortions caused by the variations in tube parameters with applied voltage. It will also remove phase distortions caused by voltage varying capacitances (really important in a mosfet follower). The CCS forces the tube current to remain constant reducing distortions caused by the changes in tube parameters with applied current. We now have a follower with a truly constant transconductance.
Michael Koster said:
Not controversial to me. Try this one for size:
All amplifiers of high damping factor connected to real Microphones
and/or real Loudspeakers will clip. Because the impedance of those
mechanical devices are moving targets.
When an amplifier inevitably clips, does it produce music, or noise?
All harmonics are music, IMD is usually not. Can we additively inject
pure harmonies that tug the signal away from clipping and IMD?
Indeed tube stages may sound more natural when overloaded. But clipping and audibility of fine details are a bit different things, Michael?Michael Koster said:
I don't buy inevitable overloading as the main reason for differences. But electromechanical idea of driving complex loads is valid: output resistances of amplifiers are not constant and are not linear. The same about microphones, but the article shows condenser microphones connected to different consoles. Also, the article says nothing about power amplification ratios required for both microphone and power amps, but that is very significant.
You may compare non-linearities of electronics with non-linearities of mechanical media that human beings got used to no matter evolved they from single-cell organisms, or made by a Supreme Beings. Do we hear distortions added by non-linear air that transfers the sound, by walls that reflect it? No, we hear clean sound altered by physical media, that is always natural. So, why should we hear distortions of an electronic equipment when they have very similar character?
But when they differ we hear them well... "Blanket on the sound", "loss of details", in both cases levels are low, far from clipping...
The TV sweep tubes brought out the supressor grid to a seperate pin so that they could be biased slightly positive to eliminate Barkhausen oscillation. This voltage is specified in the data sheets. For most tubes the spec is 0 volts minimum, and 30 volts maximum with a few tubes rated for 50 volts.
Now the sweep tube tube manufacturers clearly stated 0 volts minimum. Most readers know how well I observe those ratings! Again one more experiment to try.
My thoughts, I am an engineer, I will use the technology that is available to me regardless of its age to build what I believe to be the best sounding amplifier that I can make given the usual engineering compromises (budget, size, weight, thermal dissipation, etc.). I am not afraid to try something new, and indeed only a few of my ideas actually work. I have learned to use simulation to eliminate many of the "less than perfect" ideas before I build them. Most of my ideas, experiments and working designs have been discussed on this forum and many have been posted to my web site in case someone wants to experiment further.
I have built far more solid state amplifiers in my lifetime than tube amps. That was during the time that I measured quality in watts, and most of them were guitar amps. I may again build some more solid state amps, but I don't have any inclination to do that now.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.