Thanks, this is indeed one tough cookie. Will seriously look into it.
I tend to like the DSP part as it allows for "software tweaking", but than again I myself need not put "show off" electronics. I will evaluate both designs and will share my thoughts.
.
I agree. Modern designs must include new components, otherwise it probably not realistic to invent something new from existing components, especially since no new tubes are being researched or developed for audio application.
For me, a tube amplifier is an amplifier where tubes drive the speakers.
Adding solid state to the peripherals is OK by me.
Me too, as long as there is an output tube and output transformer, I am good.
I always have a question about tube SE, if we are going to drive the grid of the output tube with a sand cathode follower, can we even eliminate the prior tube stage and replace it with a step up transformer. Just one stage, pretty simple.
Tube sound
Long ago I read a research that studied the "tube sound" phenomena. (I'm trying to locate the reference research)
Their conclusions:
1. Tube interact with electromechanical elements (Guitar pickups, Microphones, Speakers)in a quite distinct way that is different from the way solid sate do. most people prefer the tube interaction.
2. Tube tend to produce even harmonics which are more pleasant to the human ear.
3. Tubes distort sound in a sonic attractive way.
Conclusions:
1. When no electromechanical devices are involved , tubes have no influence. This means that tube buffer stages, have no real "tube sound" effect.
I can verify this from my own humble double blind tests.
2. Since PP eliminates 2nd order harmonics, only SE has the real "tube sound"
3. Solid sate devices can preserve the tube sound (Tube based guitar amplifiers are connected to solid state PAs) and in theory could be built to sound like them.
I think this would add some perspective to the "pure tube"debate.
Long ago I read a research that studied the "tube sound" phenomena. (I'm trying to locate the reference research)
Their conclusions:
1. Tube interact with electromechanical elements (Guitar pickups, Microphones, Speakers)in a quite distinct way that is different from the way solid sate do. most people prefer the tube interaction.
2. Tube tend to produce even harmonics which are more pleasant to the human ear.
3. Tubes distort sound in a sonic attractive way.
Conclusions:
1. When no electromechanical devices are involved , tubes have no influence. This means that tube buffer stages, have no real "tube sound" effect.
I can verify this from my own humble double blind tests.
2. Since PP eliminates 2nd order harmonics, only SE has the real "tube sound"
3. Solid sate devices can preserve the tube sound (Tube based guitar amplifiers are connected to solid state PAs) and in theory could be built to sound like them.
I think this would add some perspective to the "pure tube"debate.
Re: Tube sound
Well, it's your thread, so I guess it's OK for you to ruin it.
From your first post it sounded like you were seeking more advanced designed that made use of SS's superiority in certain applications to improve fidelity. Now it turns out your just into that audiophile babble. So a tube PP amp with vanishingly low distortion measurements scores poorly on your "tube sound" criteria?
xenu said:
Well, it's your thread, so I guess it's OK for you to ruin it.
From your first post it sounded like you were seeking more advanced designed that made use of SS's superiority in certain applications to improve fidelity. Now it turns out your just into that audiophile babble. So a tube PP amp with vanishingly low distortion measurements scores poorly on your "tube sound" criteria?
Either one will put out an output that is independent of mains voltage variation, so I don't see why I should mention it.
All the more reason to regulate it since it will then be absolutely fixed at your preferred 10% high voltage, with no danger of ever going higher.
Not when I design it.
That is the first time anyone has insulted my mains voltage. Take it back or I'm going to die laughing...
I am not calling anyone anything. Each is entitled to their own opinion. If we didn't have an opinion, some marketing wizard would have us all listening to MP3's.
I learned most of my electronics on tubes, they were the only thing taught in an under funded public high school electronics program. Now I spend my days designing (and defining) state of the art RF communication equipment. Tubes are a welcome escape from the world of 0204 SMT components, with 1.8 volts of B+.
Those are not 45's. They are NX-483's, a tube that not many people have ever heard of. I found a rusty radio chassis lying in the dirt at an outdoor flea market for $20. It had a sticker that said that all of the tubes were replaced in 1931. I do have a few 45's. They were all obtained from Ebay before the prices went into the stupid money range. Some for $5 each!
I am probably just as happy with my Tubelab SE as you are with your RH amps. I find that driving a DHT with a low impedance source like a mosfet follower is the best economical solution for solving the overload recovery issue so common with the usual coupling capacitor design. Granted a really nice interstage transformer will work good, they are just not in my budget.
Here in South Florida the line voltage can range from 119 to 126 within a few minutes. That is on a good day. During a summer thunderstorm (every afternoon) lightning strikes can cause 1 KV surges and 75 volt brownouts regularly. The power distrubution system was 60% destroyed by a hurricane 5 years ago and is still patched together regardless of what our power company says.
Restating the request.
The two statements are in direct conflict with each other, but tend to imply that you are after a 60WPC SE tube amplifier with no sand in the output stage. Now you are asking for something even harder. To get 60 watts from a SE amp requires either a big transmitting tube an even bigger high voltage power supply and an expensive output transformer, OR a low impedance design using several tubes in parallel. Either way these projects are complicated. Such a project is not a good candidate for the inexperienced tube amp builder since the probability of success is low, the cost is high and there are serious safety concerns.
If you have never built a tube amp before, first off read, then read some more. This forum , Morgan.......
Then build something with a high probability of success. Build it and learn. You will learn a lot by building a simple design, and probably save yourself a lot of time, money and frustration down the road.
tubelab.com said:
It ROCKS? Hey, that's what really matters.
We all stand upon somebody's shoulders. Ever heard of a guy called Lee de Forrest? You are standing on his shoulders.
Same re Tektronix and HP, but they were hardly into high-end audio.
Maybe just the credit given for taking the constant current constant voltage cathode follower into the modern age, that's good enough. But when we started to develop the idea with modern (relatively speaking) tubes, nobody was using it. And if someone shows that somebody else was, no problem sharing a little glory. We did develop several versions of course but the basic idea the same - it's a trade-off between performance and complexity. The one that is kind of my favourite, not because it's the best, is the one that has split voltage, can be used with as little as -/+ 24V. The one in the Tube Gainclone DIY version used -/+35V. The beauty was the basic concept: Lots of people had cut their teeth on SS circuits, being scared witless by high tube voltages. My motive was pure DIY, here was a chance for those guys getting into tubes and not be worried about the demon HTs. I would love to know the number that did this just as I intended, this was their first tube DIY project. Since then, having gained the confidence, many have moved on to bigger things.
So it's not necessarily about making the world change, just move.
Just a correction. The DIY version used a simple CF on split rails. But the SLCF schematic sans exact values was published for those who wanted to be more venturous and add the required elements. If UG stage was sufficient, that little SLCF was often put into systems where ten grand preamps were replaced. It not only did OK, it did more than that.
I love it when this sort of thing happens.
BTW, we did not look beyond small scale signal levels and not seriously high current, so the field is and alway will be, open.
Joe R.
PS: Enjoyed your analysis on the other thread on the topic.
tubelab.com said:
I agree that I may ask for too much. I brought the reference to the SE, since some (you among them..) seem to prefer the SE sound, which go inline with what the article mentioned.
I unfortunately have no experience with tubes, and I take your advise to start with something simple.
I do thank you for the two references, they were what I was looking for in term of new designs. It may well be that a 60W SE is way beyond my abilities to DIY. Even NETA looks complicated (the HV part)
Re: Re: Tube sound
My first post did indicate what you mentioned, that I was seeking a modern design that preserve the tube sound but overcomes some of its problems with SS.
I brought the reference to the seemed superiority of SE over PP, since some seem to prefer the SE sound, which go inline with what the article mentioned.
As Tubelab mentioned, I probably ask for too much and need to compromise either on the power or on the method (SE vs. PP)
leadbelly said:
Well, it's your thread, so I guess it's OK for you to ruin it.
My first post did indicate what you mentioned, that I was seeking a modern design that preserve the tube sound but overcomes some of its problems with SS.
I brought the reference to the seemed superiority of SE over PP, since some seem to prefer the SE sound, which go inline with what the article mentioned.
As Tubelab mentioned, I probably ask for too much and need to compromise either on the power or on the method (SE vs. PP)
Re: Re: Re: Tube sound
When I first got back into tubes, my father handed me the amp he build while in high school. It was an 80-100W amp using 8417s running as pentodes. I needs a lot of work to be safe to use and it also has no power transformer (long story). I became rather obsessed with building a 100WPC tube amp and collected a large quantity of iron to do so. Somewhat more recently I have come to the conclusion that I was much better off spending a fraction of that cost on a good set of used speakers with decent sensitivity. Mine are Klipsch KLF-10s that I got off of Craigslist for $150. They are 98dB, but anything over 95dB will help a great deal. It is amazing what you can do with 2WPC with speakers like this.
If your speakers are not that sensitive and you are not sure what kind of amp topology you prefer, I would start with a basic push-pull design. If you have the money, get one of the Dynaco Kits from Bob Latino, Triode Electronics, or dynakits.com. You get to build it all yourself, but everything that you will need is included along with step-by-step instructions. I built the ST-70 clone from Bob Latino and it is a great amp...very versatile with the kinds of tubes you can try in it. 35WPC is PLENTY to push my older, less sensitive speakers to ear bleeding levels.
http://www.tubes4hifi.com/bob.htm
Just some food for thought....
Russ
When I first got back into tubes, my father handed me the amp he build while in high school. It was an 80-100W amp using 8417s running as pentodes. I needs a lot of work to be safe to use and it also has no power transformer (long story). I became rather obsessed with building a 100WPC tube amp and collected a large quantity of iron to do so. Somewhat more recently I have come to the conclusion that I was much better off spending a fraction of that cost on a good set of used speakers with decent sensitivity. Mine are Klipsch KLF-10s that I got off of Craigslist for $150. They are 98dB, but anything over 95dB will help a great deal. It is amazing what you can do with 2WPC with speakers like this.
If your speakers are not that sensitive and you are not sure what kind of amp topology you prefer, I would start with a basic push-pull design. If you have the money, get one of the Dynaco Kits from Bob Latino, Triode Electronics, or dynakits.com. You get to build it all yourself, but everything that you will need is included along with step-by-step instructions. I built the ST-70 clone from Bob Latino and it is a great amp...very versatile with the kinds of tubes you can try in it. 35WPC is PLENTY to push my older, less sensitive speakers to ear bleeding levels.
http://www.tubes4hifi.com/bob.htm
Just some food for thought....
Russ
ray_moth said:Good article, interesting to read. PSYCHOACOUSTICS - is this something that audiopaths listen to?
In the article they describe "an air" and "details", stressing several times "on low levels", but concentrate on a severe overload only.
Also, they explain significance of mic and power amps by interaction with mechanical devices, while microphones they used have no mechanical devices connected directly to consoles.
Good attempt, but no credit.
I posted a diagram and a bit surprised that it did not raise a single comment. There are a few less than common features. So I decided to highlight them and repost.
Joe R.
An externally hosted image should be here but it was not working when we last tested it.
Joe R.
Joe Rasmussen said:
It looks like the LTP's ccs is *QUITE* dependant on a clean -50V supply; any ripple there will change V_b-e and thus current passed through. Get that issue addressed and it would be a fine circuit...
The diode drops to the triode-ed g2 is used from time to time, and for a variety of reasons; what in particluar are you thinking of?
cheers,
Douglas
Bandersnatch said:
Hi Douglas
I wouldn't just use a straight zener there, but it simply demonstrates the idea. It has been called "Virtual Triode Operation" and I had an email exchange with Bill Perkins re the idea (Pearl Hi-Fi - and planet10 Dave knows Bill quite well). Bill scoured old sources and came across the idea. His way of doing it is a bit different from mine, but the idea is the same. Basically both the Anode and Screen see the same Z (the cap needs to be largish) but separated DC wise while actually seeing the same AC and Z. Watch (listen) to the dynamics jump. The Anode/Primary interface is a modulated High-Pass filter that needs to be kept as dynamically stable as possible. This does it.
SY was there in Holland (ETF2006) when Menno v. Veen gave a discussion on this exact topic. Later Menno heard it, VTO, with an amp I sent over to a client of his. Triode is the way to go but the Screen gets in the way. The usual resistor modulates the Screen (makes it a renegade Grid and audibly slows it down). Connect it directly to Anode and it sounds great, but not for long... poof. It needs to be protected as shown. VTO does the same AC wise without blowing up the Screen.
It has been tried by others but not with reported success in every instance. But...
But when combine with HF AC Bias - the two complements each other (note the input voltage stage cathode input - yes, the cathode is an input). VTO ups dynamics etc but can be a bit couth. I believe it shows up permeability problems in output tx. But the L.E.M. supplies the 'tickle' that keeps permeability levels from dropping too low by constantly exciting the transformer. I worked with Menno on this and the LEM is the result. This plus VTO is like 1 + 1 = 3 or 4. They complement each other.
Note no feedback - nothing comes off the secondary. This configuration sounds much for the worse if feedback is applied. The idea is simple: This is designed as a Unity Coupling Amplifier - ask Lynn Olson and he will nod approvingly because his amp is as well and he understands the description. Any feedback will force the amp to behave as a voltage amp - or nudge it in that direction. Results in high immunity to back-EMF and the kind of sound that Nelson Pass ascribes to in current (send) amps.
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.
PS: I have a couple of other 'modern' circuits I could show. If there is a bit of a discussion on this one I might post them. One of them is so off the wall as it relies on making the power supply have an extremely poor bandwidth - it's a bottler and this one I am quite sure you won't find in any text book or such.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.