In another thread on SE OTFs Angel (surely a deceptive tag with these sort of questions) asked if I had considered SV572-3 over 300B. Given I know nothing about valves and chose 300B due to the reputation and level of support available on the web I thought I'd start another thread.
I want to build the best amp I possibly can, I don't want to build amp after amp. (may not end up true but I'd like to start there)
What is the best set of tubes to base my amplifier on keeping in mind that I really want to use an existing design and not have to become a valve engineer.
At this stage I'm not considering cost but woundn't mind feedback on value for money.
Thanks a lot
Ralf
I want to build the best amp I possibly can, I don't want to build amp after amp. (may not end up true but I'd like to start there)
What is the best set of tubes to base my amplifier on keeping in mind that I really want to use an existing design and not have to become a valve engineer.
At this stage I'm not considering cost but woundn't mind feedback on value for money.
Thanks a lot
Ralf
I'm glad you have decided to search for more information before choosing a path.
First off, if you are considering value for money, the 300B is certainly not the way to go. It is one of the most expensive valves around, largely due to its cult-like following.
Secondly, you need to decide whether you want a romantic sound, or a neutral/accurate sound. A romantic sound is typically acheived by using valves that have relatively high levels of distortion, and not properly using them to their maximum potential.
If what you want is neutral/accurate reproduction, I would suggest you forget the 300B. While it is more linear than many valves that are used for audio, it is not nearly the most linear valve available. And its power output is limited to about 8W when operated with a sensible load.
Taking cost into account, you have certain alternatives. I'll list them here, and bear in mind that power output for push-pull is typically four times that of single ended. I'll be listing approximate values for single ended.
(A) You could go with a triode of some sort. In doing so, I would carefully consider linearity, lifetime, availability and cost. Good examples of readily available triodes for audio use, that are not overly expensive, include the SV572-3 (25 watts), SV811-3 (15 watts), 6C33C or 6C33C-B (15 watts), 6AS7 (2-3 watts) or 6C45P (1 watt). There are other valves that would be even better, such as the broadcasting triodes by Svetlana, but those require special care, and are likely unsuited as a first project.
(B) You could go with a pentode. Mostly the same concerns, but they tend to be even more common. Pentodes have more odd ordered distortion than triodes, but universally labeling them as bad is just yet another audiophile myth. I could nominate e.g. 6550 (15 watts), EL34 (10 watts), EL84 or SV83 (3 watts). Note that the EL34 designs tend to be a bit low on detail, and slightly 'fat' sounding. I'm not convinced this needs to be so, and the Lundahl application note makes a good case for the EL34. Note also that the 6550 could easily be accomodated by changing that design slightly, which I'm sure people on this forum can help you with.
If you intend to use this amplifier for high-sensitivity speakers, I would recommend going with the 6AS7 (very available, low cost, rugged and durable, offers 2-3 watts, which is enough for speakers with a 90dB sensitivity or better, if you don't require very loud listening volumes), the SV83 (very available, low cost, rugged and durable, also offers 2-3 watts) or the 6C45P (not so available, high cost, extremely rugged, but only about 1 watt, which ups the requirement to about 95dB sensitivity)...
For medium sensitivity speakers, I would go with the 6550, 6C33C or 6C33C-B. Either way, you get about 15 watts of single ended power at low distortion. The 6550 was used to great effect by Cayin in one of their amplifiers, while the 6C33C-B is routinely employed by Balanced Audio Technologies. Both these valves are readily available at reasonable prices. The 6C33C-B has a low life expectancy, however, so beware.
For regular speakers, the SV572-3 is the only 'regular' valve with good availability at reasonable prices, capable of delivering enough power, about 25W in single ended configuration, to drive speakers whose sensitivity can be as low as 85dB (if you don't need a disco in your home), or maybe even 80dB.
If going for a push-pull solution is acceptable to you, the options are a bit freer, ranging from a few watts with two 6C45P, to a whooping 120W with two SV572-3. In push-pull, you also get the (realistic) option of building output transformerless amplifiers, although I would not recommend it out of concern for their reliability.
Considering that this is your first project, I would suggest you go for either the 6AS7 or 6C33C-B, as they are quite comfortable with voltages under 200Vdc. 6550 if you want to be a bit more ambitious about the voltages involved. The SV572-3 is absolutely lethal, with a recommended voltage in the range of about 800V for high output power.
Another question you're going to bump into, is that of driver valves. Along the same lines, for the 'low voltage' designs, I would recommend the 6N1P or 6N30P. These are readily available (although the 6N30P is a bit expensive), and have a medium gain, which is okay, since you can use two voltage gain stages, or an interstage transformer. (Both have low impedance, and are therefore suited to stepping up with a transformer)..
For semi-high voltage designs, the 6SN7, 6FQ7 or 6CG7 (the latter two are technically the same valve) should do, as far as triodes are concerned. For pentodes, I seem to recall there are a great number of low-capacitance, relatively linear, high voltage pentodes available.
I am unfamiliar with suitable driver valves for high voltage designs, such as the SV572-3, and would recommend an interstage transformer to solve that problem.
Either way, single end or push pull, high voltage or low voltage, triode or pentode, I would recommend a small amount of local feedback, at least around the driver stages. Valves have a high gain variation, and if your gain is off in the slightest, stereo imaging will suffer. For a first project, you don't want to get into having to rebalance your amp all the time.
As an example, the 6N1P has a nominal gain of 33. Using local feedback (a cap and a resistor will do this for you), you could cut it down to a gain of 20-25, which is less than 6dB feedback, but is more than sufficient to compensate for variations between the tubes.
When selecting output transformers, there are a number of good choices out there, and I doubt you would go wrong with any of the ones recommended here. In case no-one else responds, I will recommend Lundahl, who have single-ended and push-pull output transformers with simple wiring and excellent performance, in addition to their more flexible offerings. These simpler transformers come with impedance ratios of 5Kohm to 8ohm, 3Kohm to 8ohm, and 5500ohm to 5ohm.
The 5Kohm to 8ohm will be suitable for a 300B, and should be suitable for the SV572-3 and a number of others. The 3K version is useful for somewhat lower impedance valves. If your speakers are 4 ohm, the 3K:8 version is (IIRC) about ideal for the 6C33C-B.
Hope this helps.
First off, if you are considering value for money, the 300B is certainly not the way to go. It is one of the most expensive valves around, largely due to its cult-like following.
Secondly, you need to decide whether you want a romantic sound, or a neutral/accurate sound. A romantic sound is typically acheived by using valves that have relatively high levels of distortion, and not properly using them to their maximum potential.
If what you want is neutral/accurate reproduction, I would suggest you forget the 300B. While it is more linear than many valves that are used for audio, it is not nearly the most linear valve available. And its power output is limited to about 8W when operated with a sensible load.
Taking cost into account, you have certain alternatives. I'll list them here, and bear in mind that power output for push-pull is typically four times that of single ended. I'll be listing approximate values for single ended.
(A) You could go with a triode of some sort. In doing so, I would carefully consider linearity, lifetime, availability and cost. Good examples of readily available triodes for audio use, that are not overly expensive, include the SV572-3 (25 watts), SV811-3 (15 watts), 6C33C or 6C33C-B (15 watts), 6AS7 (2-3 watts) or 6C45P (1 watt). There are other valves that would be even better, such as the broadcasting triodes by Svetlana, but those require special care, and are likely unsuited as a first project.
(B) You could go with a pentode. Mostly the same concerns, but they tend to be even more common. Pentodes have more odd ordered distortion than triodes, but universally labeling them as bad is just yet another audiophile myth. I could nominate e.g. 6550 (15 watts), EL34 (10 watts), EL84 or SV83 (3 watts). Note that the EL34 designs tend to be a bit low on detail, and slightly 'fat' sounding. I'm not convinced this needs to be so, and the Lundahl application note makes a good case for the EL34. Note also that the 6550 could easily be accomodated by changing that design slightly, which I'm sure people on this forum can help you with.
If you intend to use this amplifier for high-sensitivity speakers, I would recommend going with the 6AS7 (very available, low cost, rugged and durable, offers 2-3 watts, which is enough for speakers with a 90dB sensitivity or better, if you don't require very loud listening volumes), the SV83 (very available, low cost, rugged and durable, also offers 2-3 watts) or the 6C45P (not so available, high cost, extremely rugged, but only about 1 watt, which ups the requirement to about 95dB sensitivity)...
For medium sensitivity speakers, I would go with the 6550, 6C33C or 6C33C-B. Either way, you get about 15 watts of single ended power at low distortion. The 6550 was used to great effect by Cayin in one of their amplifiers, while the 6C33C-B is routinely employed by Balanced Audio Technologies. Both these valves are readily available at reasonable prices. The 6C33C-B has a low life expectancy, however, so beware.
For regular speakers, the SV572-3 is the only 'regular' valve with good availability at reasonable prices, capable of delivering enough power, about 25W in single ended configuration, to drive speakers whose sensitivity can be as low as 85dB (if you don't need a disco in your home), or maybe even 80dB.
If going for a push-pull solution is acceptable to you, the options are a bit freer, ranging from a few watts with two 6C45P, to a whooping 120W with two SV572-3. In push-pull, you also get the (realistic) option of building output transformerless amplifiers, although I would not recommend it out of concern for their reliability.
Considering that this is your first project, I would suggest you go for either the 6AS7 or 6C33C-B, as they are quite comfortable with voltages under 200Vdc. 6550 if you want to be a bit more ambitious about the voltages involved. The SV572-3 is absolutely lethal, with a recommended voltage in the range of about 800V for high output power.
Another question you're going to bump into, is that of driver valves. Along the same lines, for the 'low voltage' designs, I would recommend the 6N1P or 6N30P. These are readily available (although the 6N30P is a bit expensive), and have a medium gain, which is okay, since you can use two voltage gain stages, or an interstage transformer. (Both have low impedance, and are therefore suited to stepping up with a transformer)..
For semi-high voltage designs, the 6SN7, 6FQ7 or 6CG7 (the latter two are technically the same valve) should do, as far as triodes are concerned. For pentodes, I seem to recall there are a great number of low-capacitance, relatively linear, high voltage pentodes available.
I am unfamiliar with suitable driver valves for high voltage designs, such as the SV572-3, and would recommend an interstage transformer to solve that problem.
Either way, single end or push pull, high voltage or low voltage, triode or pentode, I would recommend a small amount of local feedback, at least around the driver stages. Valves have a high gain variation, and if your gain is off in the slightest, stereo imaging will suffer. For a first project, you don't want to get into having to rebalance your amp all the time.
As an example, the 6N1P has a nominal gain of 33. Using local feedback (a cap and a resistor will do this for you), you could cut it down to a gain of 20-25, which is less than 6dB feedback, but is more than sufficient to compensate for variations between the tubes.
When selecting output transformers, there are a number of good choices out there, and I doubt you would go wrong with any of the ones recommended here. In case no-one else responds, I will recommend Lundahl, who have single-ended and push-pull output transformers with simple wiring and excellent performance, in addition to their more flexible offerings. These simpler transformers come with impedance ratios of 5Kohm to 8ohm, 3Kohm to 8ohm, and 5500ohm to 5ohm.
The 5Kohm to 8ohm will be suitable for a 300B, and should be suitable for the SV572-3 and a number of others. The 3K version is useful for somewhat lower impedance valves. If your speakers are 4 ohm, the 3K:8 version is (IIRC) about ideal for the 6C33C-B.
Hope this helps.
The only tube-amp I have ever listened to is my own SEKT88-amp. As I understand, the 300B is magical and probably not of this world. It was given to the the people of earth by kind extra-terrestrials who heard our cry for the highest possible form of fidelity. But that's just what I've heard 😉
No, seriously... The 300B seem to be a great tube, but I suspect that are other tube that also sound good, but maybe not as good as 300B. I don't think I'll ever build another amp than my KT88-amp. 😀
No, seriously... The 300B seem to be a great tube, but I suspect that are other tube that also sound good, but maybe not as good as 300B. I don't think I'll ever build another amp than my KT88-amp. 😀
The main problem with valves is that there are many ways to skin a cat. A person such as yourself may build a great 300B amp, and be happy with it forever....
Some of us, especially on a DIY site like this, have a lot of bits sitting around, might have some random transmitter tubes that we know a guy has, plus some interesting tube from the 1920s, and a constant current source, and amplifiers become Works In Progress for many years!!! 😀 Its a hobby thing.
300Bs certainly have a great reputation amongst single ended tubes. They also have a decent amount of power available, unlike many SE amps. You can't go wrong with them really, with the simplicity of circuits and the knowledge available. They are also not that expensive, maybe $US 100 for a pair, not too bad.
Its all in the output transformers if you ask me... Don't skimp too much on them, and you can reuse them if you want to.
BTW My 6B4G SET amp sounds fantastic, and does have all that "ultra smooth midrange" that they talk about... SETs can be very sexy to listen to.
Just make a decision and build the thing already!! 😀
Some of us, especially on a DIY site like this, have a lot of bits sitting around, might have some random transmitter tubes that we know a guy has, plus some interesting tube from the 1920s, and a constant current source, and amplifiers become Works In Progress for many years!!! 😀 Its a hobby thing.
300Bs certainly have a great reputation amongst single ended tubes. They also have a decent amount of power available, unlike many SE amps. You can't go wrong with them really, with the simplicity of circuits and the knowledge available. They are also not that expensive, maybe $US 100 for a pair, not too bad.
Its all in the output transformers if you ask me... Don't skimp too much on them, and you can reuse them if you want to.
BTW My 6B4G SET amp sounds fantastic, and does have all that "ultra smooth midrange" that they talk about... SETs can be very sexy to listen to.
Just make a decision and build the thing already!! 😀
Just one last point regarding actively supported projects, you might try searching for Andrea Cuifoli (spelling?). He has a ton of different projects.
As to the SV572-3 being out of production, that is not the case, as far as I can tell. Besides, either way it will be readily available for years to come.
The 300B does have a cult following.
However, its 'magic' has nothing to do with reality. It is a colouration. A darned cozy one, mind you, but a colouration nonetheless. And not a very subtle one.
I, for one, prefer to have my valves reproduce the soul of the music. Not infuse it with one that is not its own.
You would get a lot less colouration with the 6C33C-B, with more output power. Same goes for the 6550. You could also use the SV572-3, and get more than twice the output power.
Proper use of feedback, with a good output transformer (one with low static resistance, and a low, constant group delay), can linearize just about any valve, however.
As to the SV572-3 being out of production, that is not the case, as far as I can tell. Besides, either way it will be readily available for years to come.
The 300B does have a cult following.
However, its 'magic' has nothing to do with reality. It is a colouration. A darned cozy one, mind you, but a colouration nonetheless. And not a very subtle one.
I, for one, prefer to have my valves reproduce the soul of the music. Not infuse it with one that is not its own.
You would get a lot less colouration with the 6C33C-B, with more output power. Same goes for the 6550. You could also use the SV572-3, and get more than twice the output power.
Proper use of feedback, with a good output transformer (one with low static resistance, and a low, constant group delay), can linearize just about any valve, however.
I have seen this mentioned many times but AFAIK it has no bearing in reality. The source of the supposed short life time of the 6C33 tube is from what is mentioned in the original Russian data sheet where it is mentioned life time of between 400 and 1000 hours depending on type which is indeed very low, BUT this is the guaranteed life time when fulfilling the complete specification including for instance the ability to withstand a 100 G shock and other spec points that is important when the tube is used as a series voltage regulator in a MIG29 fighter aircraft.
However when used sensibly as in an audio amplifier there is no reason to expect that a 6C33 will not have similar or probably better life time than other power tubes, this is also reported by many users including BAT which uses quite a lot of these tubes.
I have used the same set of 6C33C in my OTL now for more then 3 years without any drift in bias or any tendencies towards failure.
Regards Hans
I've done a very quick search on the SV572-3 using google and while it turns up some references there certainly aren't the nice full on stories of how someone has put it together. With 300Bs I've turned up about 6 full projects with a lot of design discussions. That is pretty important to me.
I will go out and buy some good valve book to see if I can deviate from 300B - want to fully research rather than just jump in. Any recommendations for a particularly good book(s) to start with.
I'm not worried about voltage - was an electrician in a previous life before becoming a Technical Officer in a power station - worked on a lot of control circuits. I worked with very high voltage so that's not an issue. Also worked in microwave comms for a while.
I want to thank everyone, particularly Angel for the comprehensive coverage of alternatives. Will spend the week end researching.
It would be really nice if there was just a really weel accepted design with parts that are recognised as excellent to start out with - dare to dream.
Regards
Ralf
I will go out and buy some good valve book to see if I can deviate from 300B - want to fully research rather than just jump in. Any recommendations for a particularly good book(s) to start with.
I'm not worried about voltage - was an electrician in a previous life before becoming a Technical Officer in a power station - worked on a lot of control circuits. I worked with very high voltage so that's not an issue. Also worked in microwave comms for a while.
I want to thank everyone, particularly Angel for the comprehensive coverage of alternatives. Will spend the week end researching.
It would be really nice if there was just a really weel accepted design with parts that are recognised as excellent to start out with - dare to dream.
Regards
Ralf
Some people really love the 300B, but I'm not one of them. It's not at all "magical" although it can be very beguilling on a lot of material. However, it's way too expensive for what it is, and an 845 amp for the same money (cheaper tubes but more expensive PSU) would likely better it* and prove a cheaper long term solution.
* to my ears and tastes.
* to my ears and tastes.
300B SETs
Ralf,
Those are very wise words from Angel, and I will be bearing them in mind for my next project.
However, I'd still suggest a 300B for your first project for several reasons.
Mainly, there is a mass of information out there on 300B SETs. Circuits galore, but more importantly the acquired expertise on how to drive it well; and many discussions on finer points such as ac vs dc heating, fixed vs cathode bias, RC vs LC vs IT coupling, etc.
You will have no difficulty finding OPTs that are an ideal match for 300Bs.
Given all the above, I think you have a better chance of success if you go for the 300B.
A 300B amp can have plenty of 'drive' and good sound quality; results are pretty well known and virtually guaranteed.
From people's circuits and descriptions of the sound they get, you can try to pick factors to give you the 'flavour' of sound you prefer.
Romantic or neutral; dynamic/forceful or polite; etc etc.
I appreciate the concern that the 300B valve, in itself, may be nonlinear and 'romantic', but with respect I don't fully agree. I'd suggest that the distortions may be more to do with the driver stage.
For a neutral result, I suggest a driver stage has to be very linear; has to work over +/- 150V (yes, that's much more than in theory); and has to have a low output impedance so as to drive the grid of the 300B (with its high Miller capacitance) *and* a low value grid leak resistor.
A tall order, almost needing a power amp as a driver.
With such a driver stage, a good PSU, say TJ mesh 300Bs (neutral and extended), and a good OPT at 5kohms, a 300B amp can be highly neutral, fast, clean, dynamic and *powerful* rather than warm and cuddly.
And *musical* with it; really awesome. 🙂
OK, experts like Angel can do better by intelligent use of different valves, but I don't think it's a good first project.
Ralf,
Those are very wise words from Angel, and I will be bearing them in mind for my next project.
However, I'd still suggest a 300B for your first project for several reasons.
Mainly, there is a mass of information out there on 300B SETs. Circuits galore, but more importantly the acquired expertise on how to drive it well; and many discussions on finer points such as ac vs dc heating, fixed vs cathode bias, RC vs LC vs IT coupling, etc.
You will have no difficulty finding OPTs that are an ideal match for 300Bs.
Given all the above, I think you have a better chance of success if you go for the 300B.
A 300B amp can have plenty of 'drive' and good sound quality; results are pretty well known and virtually guaranteed.
From people's circuits and descriptions of the sound they get, you can try to pick factors to give you the 'flavour' of sound you prefer.
Romantic or neutral; dynamic/forceful or polite; etc etc.
I appreciate the concern that the 300B valve, in itself, may be nonlinear and 'romantic', but with respect I don't fully agree. I'd suggest that the distortions may be more to do with the driver stage.
For a neutral result, I suggest a driver stage has to be very linear; has to work over +/- 150V (yes, that's much more than in theory); and has to have a low output impedance so as to drive the grid of the 300B (with its high Miller capacitance) *and* a low value grid leak resistor.
A tall order, almost needing a power amp as a driver.
With such a driver stage, a good PSU, say TJ mesh 300Bs (neutral and extended), and a good OPT at 5kohms, a 300B amp can be highly neutral, fast, clean, dynamic and *powerful* rather than warm and cuddly.
And *musical* with it; really awesome. 🙂
OK, experts like Angel can do better by intelligent use of different valves, but I don't think it's a good first project.
The best amp or the best SE amp? As soon as you limit yourself to SE, you've made a bunch of compromises- you're going to have relatively low power, relatively high harmonic distortion, and relatively high source impedance. OTOH, if that's the sort of coloration you're seeking, a good push-pull design will assuredly NOT give that to you.
Not that I'm a 300B lover....
The SV572-3 is a nice valve indeed. I've used it before.
However, to realise it's full potential, a lot of attention must be paid to the driver stage.
A 572-3 with mediocre driver stage will give less power than one of these steroid fed 300B's.
There is also more transformer choice with 300B's.
I was under the impression that no SV572's have been made for a number of years now.
It's a shame that 300B's are so overpriced. 2A3's are less so, but then the power output is rather less, unless you parallel or P-P them.
The SV572-3 is a nice valve indeed. I've used it before.
However, to realise it's full potential, a lot of attention must be paid to the driver stage.
A 572-3 with mediocre driver stage will give less power than one of these steroid fed 300B's.
There is also more transformer choice with 300B's.
I was under the impression that no SV572's have been made for a number of years now.
It's a shame that 300B's are so overpriced. 2A3's are less so, but then the power output is rather less, unless you parallel or P-P them.
tubetvr said:
Thank you for setting me straight on this one. I shold also have mentioned that running the filaments 3% low, limiting the inrush current, delaying the HT supply, and operating at a conservative dissipation, should at least double the life expectancy, all other things being equal. But these things might be tough for a starting project, especially considering the 6.6A filament current (if paralell operated) and half-amp HT current..
Konnichiwa,
Well, much has been said here. First, the least Expensive 300B (Reflector Russia) when purchased directly from russia are dirt cheap and sound quite decent and NOTABLY better than EL34/KT88 in SE (Triode), from experience.
With the 300B you have a Valve that is comparably easy to drive and has modest HT and Heater requirements (making for easily managable circuits that still can be made affordable with real premium components) coupled with enough power to manage in the "real" world and not only with super high sensitivity speakers.
Further, "300B equivalent" Valves are available from many manufacturers, making it together with the KT88/6550, EL34/84 and 12A..7/ECC81 to ECC83 Series probably one of the last Valves to become unavailable.
Svetlana has not made the 572/811 "dash" series for a long while and remaining stocks are drying up, plus they sound pretty pedestrian, compared to a cheap chinese 845 or 211.
Having heard several amplifiers with very similar execution, plus a variety of Amplifiers where I had my grubby mitts in the design and have SOME idea what went inside and on I find the 300B the best all around valve, the 45 and 10Y (no - they do not sound like each other but they do share a lot of features compared to the 300B) the best when power is NOT an issue and I do not much like the Bi-Anode 2A3 but Single Anode Types (only heard Sovtek/Reflector and KR) somewhere between these and 300B, but usually lacking in available Power compared to the 300B and lacking the detail and finesse of the 45 and 10Y.
From there on I feel the 211 the best of the rest. As for 845 I did not like the souind very much nor did I like the two 833A Amp's I came across all that much, though I guess the are okay using rather low sensitivity and low resolution speakers which would struggle with a 300B Amplifier anyway. This may have been at least partially down to the Amplifier implementations, not sure.
I have not heard the 6S33 in SE apart from the hideously expensive (and looking) Lamm which I felt was outperformed by a less expensive 300B Push-Pull Amp for a lot less money.
So, in absolute terms where speaker sensitivity is no issue 45 or 10Y would be my take, 300B for all roundiness and 211 where a (comparably) lot of power is required.
Sayonara
(PS, playing at the moment in my system is a transformer coupled 45 Push-Pull Amplifier on 95db/W/m Open Baffle Speakers, ears good mon, righteous dub)
DVDHack said:
Well, much has been said here. First, the least Expensive 300B (Reflector Russia) when purchased directly from russia are dirt cheap and sound quite decent and NOTABLY better than EL34/KT88 in SE (Triode), from experience.
With the 300B you have a Valve that is comparably easy to drive and has modest HT and Heater requirements (making for easily managable circuits that still can be made affordable with real premium components) coupled with enough power to manage in the "real" world and not only with super high sensitivity speakers.
Further, "300B equivalent" Valves are available from many manufacturers, making it together with the KT88/6550, EL34/84 and 12A..7/ECC81 to ECC83 Series probably one of the last Valves to become unavailable.
Svetlana has not made the 572/811 "dash" series for a long while and remaining stocks are drying up, plus they sound pretty pedestrian, compared to a cheap chinese 845 or 211.
Having heard several amplifiers with very similar execution, plus a variety of Amplifiers where I had my grubby mitts in the design and have SOME idea what went inside and on I find the 300B the best all around valve, the 45 and 10Y (no - they do not sound like each other but they do share a lot of features compared to the 300B) the best when power is NOT an issue and I do not much like the Bi-Anode 2A3 but Single Anode Types (only heard Sovtek/Reflector and KR) somewhere between these and 300B, but usually lacking in available Power compared to the 300B and lacking the detail and finesse of the 45 and 10Y.
From there on I feel the 211 the best of the rest. As for 845 I did not like the souind very much nor did I like the two 833A Amp's I came across all that much, though I guess the are okay using rather low sensitivity and low resolution speakers which would struggle with a 300B Amplifier anyway. This may have been at least partially down to the Amplifier implementations, not sure.
I have not heard the 6S33 in SE apart from the hideously expensive (and looking) Lamm which I felt was outperformed by a less expensive 300B Push-Pull Amp for a lot less money.
So, in absolute terms where speaker sensitivity is no issue 45 or 10Y would be my take, 300B for all roundiness and 211 where a (comparably) lot of power is required.
Sayonara
(PS, playing at the moment in my system is a transformer coupled 45 Push-Pull Amplifier on 95db/W/m Open Baffle Speakers, ears good mon, righteous dub)
I've done a very quick search on the SV572-3 using google and while it turns up some references there certainly aren't the nice full on stories of how someone has put it together. With 300Bs I've turned up about 6 full projects with a lot of design discussions. That is pretty important to me.
Alessandro Galalotti has a design with the SV572-10. At this site, you will find Andrea Ciuffoli's home page, with any number of valve designs. He appears to be going commercial now, however.. On the Svetlana Technical Bulletins page, you will find several designs. Lundahl references the Ciuffoli page, and has some designs at this page. The Lundahl site also has a schematics section, which has some good designs.
You need to search a bit more, there are a number of good designs out there. You might want to consider looking for designs that use the other SV572 valves (-3, -10, -30 or -160, designations of the mu), or even the SV811 valves that are supposedly more linear, but have half the power. Both are direct heated triodes, as the 300B, and both use high voltages.
I will go out and buy some good valve book to see if I can deviate from 300B - want to fully research rather than just jump in. Any recommendations for a particularly good book(s) to start with.
Unfortunately, my book list appears not to be on this computer. I fear my S.O. has been somewhat overzealous in moving files to CD-R's to free up space.
There should also be enough information out there on the web to get working with some sort of Spice simulator, and figure out a decent design of your own.
I'm not worried about voltage - was an electrician in a previous life before becoming a Technical Officer in a power station - worked on a lot of control circuits. I worked with very high voltage so that's not an issue. Also worked in microwave comms for a while.
Okay, good. I am always a bit concerned about recommending that people start fiddling around with kilovolt supplies (as is ideal for the SV572-3, for minimum distortion and plate resistance, as well as permitting immense power in single-ended class A1). I don't want anyone to get themselves killed over my advice.
I want to thank everyone, particularly Angel for the comprehensive coverage of alternatives. Will spend the week end researching.
Thank you. Research is good. Otherwise, you might as well order a kit from Audio Note, Edison (triod-ljud), or some other supplier. It is always better to know what you're building, and why. Have a look at the single-ended design on the Lundahl page, it gives you a nice walkthrough of the design, explaining key points. It also employs a rather interesting kind of feedback. While it employs the EL34 pentode, you could easily refit it to the 6550, or even a triode, such as the SV572-3 (which matches excellently with the 5K:8 output transformer from this company).
It would be really nice if there was just a really weel accepted design with parts that are recognised as excellent to start out with - dare to dream.
I doubt you will find one, but I guess members of this forum could help you design an amp. How about starting a thread, asking people to help you out? One might, for example, use the 6C33C-B as an output valve, and the 6N1P as a driver. Two or three stages should be enough, and the valves operate with the same voltage, allowing you to avoid the hassle of two HT supplies. Want me to provide a concept drawing?
Alternately, if you want to go more complex, alter the design to have it operate in single ended mode up to about 5W, then push-pull up to about 45W, and finally single-ended with significant compression for the last 5W up to 50W or so. This could be realized by compressing the signal above a certain level, to match the increased transconductance from transitioning to push-pull, and gating one valve below that level. This way, you also avoid the need for a huge air gap in your transformer. Remember, the first watt is the most important in any case.
Alessandro Galalotti has a design with the SV572-10. At this site, you will find Andrea Ciuffoli's home page, with any number of valve designs. He appears to be going commercial now, however.. On the Svetlana Technical Bulletins page, you will find several designs. Lundahl references the Ciuffoli page, and has some designs at this page. The Lundahl site also has a schematics section, which has some good designs.
You need to search a bit more, there are a number of good designs out there. You might want to consider looking for designs that use the other SV572 valves (-3, -10, -30 or -160, designations of the mu), or even the SV811 valves that are supposedly more linear, but have half the power. Both are direct heated triodes, as the 300B, and both use high voltages.
I will go out and buy some good valve book to see if I can deviate from 300B - want to fully research rather than just jump in. Any recommendations for a particularly good book(s) to start with.
Unfortunately, my book list appears not to be on this computer. I fear my S.O. has been somewhat overzealous in moving files to CD-R's to free up space.
There should also be enough information out there on the web to get working with some sort of Spice simulator, and figure out a decent design of your own.
I'm not worried about voltage - was an electrician in a previous life before becoming a Technical Officer in a power station - worked on a lot of control circuits. I worked with very high voltage so that's not an issue. Also worked in microwave comms for a while.
Okay, good. I am always a bit concerned about recommending that people start fiddling around with kilovolt supplies (as is ideal for the SV572-3, for minimum distortion and plate resistance, as well as permitting immense power in single-ended class A1). I don't want anyone to get themselves killed over my advice.
I want to thank everyone, particularly Angel for the comprehensive coverage of alternatives. Will spend the week end researching.
Thank you. Research is good. Otherwise, you might as well order a kit from Audio Note, Edison (triod-ljud), or some other supplier. It is always better to know what you're building, and why. Have a look at the single-ended design on the Lundahl page, it gives you a nice walkthrough of the design, explaining key points. It also employs a rather interesting kind of feedback. While it employs the EL34 pentode, you could easily refit it to the 6550, or even a triode, such as the SV572-3 (which matches excellently with the 5K:8 output transformer from this company).
It would be really nice if there was just a really weel accepted design with parts that are recognised as excellent to start out with - dare to dream.
I doubt you will find one, but I guess members of this forum could help you design an amp. How about starting a thread, asking people to help you out? One might, for example, use the 6C33C-B as an output valve, and the 6N1P as a driver. Two or three stages should be enough, and the valves operate with the same voltage, allowing you to avoid the hassle of two HT supplies. Want me to provide a concept drawing?
Alternately, if you want to go more complex, alter the design to have it operate in single ended mode up to about 5W, then push-pull up to about 45W, and finally single-ended with significant compression for the last 5W up to 50W or so. This could be realized by compressing the signal above a certain level, to match the increased transconductance from transitioning to push-pull, and gating one valve below that level. This way, you also avoid the need for a huge air gap in your transformer. Remember, the first watt is the most important in any case.
Re: 300B SETs
Mainly, there is a mass of information out there on 300B SETs.
This is very true, and a valid point. But there are also a number of established designs employing the KT88, 6550, EL34, EL84/SV83, 2A3/6B4 and a number of other valves. Why decide before having a look at all of them?
Circuits galore, but more importantly the acquired expertise on how to drive it well; and many discussions on finer points such as ac vs dc heating, fixed vs cathode bias, RC vs LC vs IT coupling, etc.
It bears mentioning, however, that there are also a number of myths out there. Not all the information you find is valid, or pertinent. Part of the problem is that it is so easy to get a reasonably good result with valves, leading many people to think that they've got a great design, while in fact they are only utilizing a fraction of their components' potential.
You will have no difficulty finding OPTs that are an ideal match for 300Bs.
Or any other popular valve, for that matter. EL34 comes to mind.
A 300B amp can have plenty of 'drive' and good sound quality; results are pretty well known and virtually guaranteed.
The main question being whether he wants a sound which is magic, even if you feed it white noise, or a sound which is faithful to the source material, and its intent. If the former, then, yes, the 300B is a guaranteed winner. If not, then it is almost a noncontender.
From people's circuits and descriptions of the sound they get, you can try to pick factors to give you the 'flavour' of sound you prefer.
Romantic or neutral; dynamic/forceful or polite; etc etc.
This is highly subjective, and depends on the prior experience of the person describing the sound. Unless you are familiar with their point of view, and former experiences, you have no grounds for comparison, and their description becomes nearly worthless. Some of these people have been living with cheap A/V receivers and decades old 14-bit dacs, while others go to concerts every week, and use that as a reference.
I appreciate the concern that the 300B valve, in itself, may be nonlinear and 'romantic', but with respect I don't fully agree. I'd suggest that the distortions may be more to do with the driver stage.
The driver stage frequently compounds the problem, as does a poor PSU design, and bad choice of output transformers. But the 300B in itself does not appear to be linear enough to defend its high cost.
A tall order, almost needing a power amp as a driver.
As long as you keep the almost in there. Many people actually drive it with something that would qualify as a power amp in its own right, and that is overkill. I would recommend using local feedback to solve this problem. It introduces some phase shift, yes, but will probably work out to be less offensive than introducing an unneccessary cathode follower or such, as is often done.
OK, experts like Angel can do better by intelligent use of different valves, but I don't think it's a good first project.
I'm by no means any expert. But I feel that there are equally good first projects that will leave him with more of his money unspent, and available for later projects, improvements, tweaks, or even (gasp) records.
On a side note, should you go with the 300B, consider that the 6SN7, and the 6FQ7/6CG7 (exclusively used by Stax in their high-end electrostatic drivers), can operate off the same supply voltage, and should be capable of driving the 300B with sufficient force.
Mainly, there is a mass of information out there on 300B SETs.
This is very true, and a valid point. But there are also a number of established designs employing the KT88, 6550, EL34, EL84/SV83, 2A3/6B4 and a number of other valves. Why decide before having a look at all of them?
Circuits galore, but more importantly the acquired expertise on how to drive it well; and many discussions on finer points such as ac vs dc heating, fixed vs cathode bias, RC vs LC vs IT coupling, etc.
It bears mentioning, however, that there are also a number of myths out there. Not all the information you find is valid, or pertinent. Part of the problem is that it is so easy to get a reasonably good result with valves, leading many people to think that they've got a great design, while in fact they are only utilizing a fraction of their components' potential.
You will have no difficulty finding OPTs that are an ideal match for 300Bs.
Or any other popular valve, for that matter. EL34 comes to mind.
A 300B amp can have plenty of 'drive' and good sound quality; results are pretty well known and virtually guaranteed.
The main question being whether he wants a sound which is magic, even if you feed it white noise, or a sound which is faithful to the source material, and its intent. If the former, then, yes, the 300B is a guaranteed winner. If not, then it is almost a noncontender.
From people's circuits and descriptions of the sound they get, you can try to pick factors to give you the 'flavour' of sound you prefer.
Romantic or neutral; dynamic/forceful or polite; etc etc.
This is highly subjective, and depends on the prior experience of the person describing the sound. Unless you are familiar with their point of view, and former experiences, you have no grounds for comparison, and their description becomes nearly worthless. Some of these people have been living with cheap A/V receivers and decades old 14-bit dacs, while others go to concerts every week, and use that as a reference.
I appreciate the concern that the 300B valve, in itself, may be nonlinear and 'romantic', but with respect I don't fully agree. I'd suggest that the distortions may be more to do with the driver stage.
The driver stage frequently compounds the problem, as does a poor PSU design, and bad choice of output transformers. But the 300B in itself does not appear to be linear enough to defend its high cost.
A tall order, almost needing a power amp as a driver.
As long as you keep the almost in there. Many people actually drive it with something that would qualify as a power amp in its own right, and that is overkill. I would recommend using local feedback to solve this problem. It introduces some phase shift, yes, but will probably work out to be less offensive than introducing an unneccessary cathode follower or such, as is often done.
OK, experts like Angel can do better by intelligent use of different valves, but I don't think it's a good first project.
I'm by no means any expert. But I feel that there are equally good first projects that will leave him with more of his money unspent, and available for later projects, improvements, tweaks, or even (gasp) records.
On a side note, should you go with the 300B, consider that the 6SN7, and the 6FQ7/6CG7 (exclusively used by Stax in their high-end electrostatic drivers), can operate off the same supply voltage, and should be capable of driving the 300B with sufficient force.
Re: Not that I'm a 300B lover....
However, to realise it's full potential, a lot of attention must be paid to the driver stage.
This can be said of any valve, but, yes, the SV572-3 can be demanding. I recommend operating it at higher plate voltages, to lessen the requirements on the driver stage, by not entering the grid current region.
There is also more transformer choice with 300B's.
Any 5K transformer capable of withstanding the current should do for the SV572-3. As I've repeated ad nauseam, the Lundahls can withstand the current and voltage, and are rated for the 25W that the SV572-3 will produce.
I was under the impression that no SV572's have been made for a number of years now.
Quite possible, though I've not heard it. In any case, the valve is easily available, and will remain so for quite some time, at reasonable prices.
However, to realise it's full potential, a lot of attention must be paid to the driver stage.
This can be said of any valve, but, yes, the SV572-3 can be demanding. I recommend operating it at higher plate voltages, to lessen the requirements on the driver stage, by not entering the grid current region.
There is also more transformer choice with 300B's.
Any 5K transformer capable of withstanding the current should do for the SV572-3. As I've repeated ad nauseam, the Lundahls can withstand the current and voltage, and are rated for the 25W that the SV572-3 will produce.
I was under the impression that no SV572's have been made for a number of years now.
Quite possible, though I've not heard it. In any case, the valve is easily available, and will remain so for quite some time, at reasonable prices.
Re: Re: Build with 300B or something else???
Konnichiwa,
Hello.
... sound quite decent and NOTABLY better than EL34/KT88 in SE (Triode), from experience.
This experience wouldn't happen to include a number of 'traditional' designs with these valves, would it? I agree that the EL34 tends to sound rather 'fat'. But the 6550 (which I understand is a variation of the KT88) can be made to sound excellent, even transistorish, with sparkling upper mids.
Svetlana has not made the 572/811 "dash" series for a long while and remaining stocks are drying up, plus they sound pretty pedestrian, compared to a cheap chinese 845 or 211.
Please elaborate on your use of the term 'pedestrian' in this context. It makes no sense.
And the 845 and 211 are typically considered to be about as coloured as the 300B, while offering more power.
I have not heard the 6S33 in SE apart from the hideously expensive (and looking) Lamm which I felt was outperformed by a less expensive 300B Push-Pull Amp for a lot less money.
Again, you have not been clear about what you're looking for in terms of sound. So 'outperformed' is rather vague.
Also, have you listened to e.g. the Balanced Audio Technology amplifiers that use this valve?
Konnichiwa,
Hello.
... sound quite decent and NOTABLY better than EL34/KT88 in SE (Triode), from experience.
This experience wouldn't happen to include a number of 'traditional' designs with these valves, would it? I agree that the EL34 tends to sound rather 'fat'. But the 6550 (which I understand is a variation of the KT88) can be made to sound excellent, even transistorish, with sparkling upper mids.
Svetlana has not made the 572/811 "dash" series for a long while and remaining stocks are drying up, plus they sound pretty pedestrian, compared to a cheap chinese 845 or 211.
Please elaborate on your use of the term 'pedestrian' in this context. It makes no sense.
And the 845 and 211 are typically considered to be about as coloured as the 300B, while offering more power.
I have not heard the 6S33 in SE apart from the hideously expensive (and looking) Lamm which I felt was outperformed by a less expensive 300B Push-Pull Amp for a lot less money.
Again, you have not been clear about what you're looking for in terms of sound. So 'outperformed' is rather vague.
Also, have you listened to e.g. the Balanced Audio Technology amplifiers that use this valve?
Fabulous response on this thread I must say.
I have some really good starting points for my week end research.
Just to clarify - I have decided nothing yet - so all options are open. I do not have any parts at this stage so there is no "can I use these bits somewhere".
From my reading I gathered that SE was the way to go - should I be corrected here?
I play a variety of music so lean toward a neutral sound but on the warm side. I don't have speakers for this system yet - I plan to put this one in a different room from my other system which is a surroundsound system also. So, my speaker options are open - if I build a lower power amp I'll buy efficient speakers if I build a higher power amp I have more options when I go out and buy.
I would like at least 8W though so that I am not limited to much in my speaker selection.
When I have considered a design to start with I will post it for improvement suggestions.
Thanks and Regards
Ralf
I have some really good starting points for my week end research.
Just to clarify - I have decided nothing yet - so all options are open. I do not have any parts at this stage so there is no "can I use these bits somewhere".
From my reading I gathered that SE was the way to go - should I be corrected here?
I play a variety of music so lean toward a neutral sound but on the warm side. I don't have speakers for this system yet - I plan to put this one in a different room from my other system which is a surroundsound system also. So, my speaker options are open - if I build a lower power amp I'll buy efficient speakers if I build a higher power amp I have more options when I go out and buy.
I would like at least 8W though so that I am not limited to much in my speaker selection.
When I have considered a design to start with I will post it for improvement suggestions.
Thanks and Regards
Ralf
300B amp as first amp project
Just want to tag in on the end here and add my thoughts:
I agree with Mike C that a 300B is good first amp project.
I finished my first amp in November: a 2A3/300B convertible amp. 6SN7 as input and triode-strapped 6V6GTs as drivers.
And with respect to the recommendation of single manufacturer russian tubes (although by reputation some of them are excellent). One great thing about 300Bs is the broad selection of available tubes at all different price points: from EH at <$100 a pair to EML at >$500 a pair; and a bunch of chinese manufacturers in between. And of course WE at >$700 a pair. With the Sovteks or Svetlanas, they are what they are and you are married to them for the life of the amp. And if the company starts to have problems, then availability becomes an issue.
As as far as power output is concerned; if its a problem then its a problem with your speakers and room, not with the amp. Since you have yet to select speakers, then this should not be a problem.
FWIW, I would recommend the following:
Since, you asked for book recommendations. Valve Amps by Morgan Jones, Beginners Guide to Tube Audio Design by Bruce Rozenblit, Principles of Power by Kevin O'Connor, back issues of Sound Practice, any issues of Stereo Sound Tube Kingdom, a Japanese magazine, Vacuum Tube Valley, and RDH. And do a boatload of research on the web. Answers to every question you may have about tube design is available to you using the above resources.
Long-term value for the money is basically building the *right* amp once and NOT after a few half-assed attempts. As DIYers we are all capable of building the best amp we can the first time we build it without any mistakes or problems. Its a question of doing the proper R&D prep work. If its a question of experimentation and study then many take a different approach and build and rebuild and rebuild...
Another road is to buy a commerical kit and build someone else's design. Its easier than complete DIY, maybe more appropriate and less costly for a first-time builder, but understand that its not your amp but one of many others.
If you are serious about building the best amp you possibly can and are committed to the project...and considering this is a first amp project:
First look at all available 300B schematics. Decide on a two or three stage approach. I prefer a 3-stage design myself.
Choose what amplifying tube to use as input, I chose a 6SN7.
Choose a proper driver, NOT a 6SN7, another triode like 2A3, 45, or some triode stapped pentode. I chose a 6V6GT in triode mode.
An example: my last amp was a commercial 6SN7 DC 300B komuro-style amp. I bought a Emission Labs 300BLXS from Bob Ungemach at AMR, which according to EML can be used as a "drop-in" replacement in any 300B amp. I dropped the pair into the amp and all I got was distortion up the waazho. Blamed the tubes, of course. I then finished building my amp, running the 300Bs at the same or similar op of the prior amp, but with different drivers, and all I got was sweet music. Yes, its not an apples to apples comparison, and its an extreme example, but it did teach me something about drivers.
Choose coupling methods, RC coupling, LC coupling, IT coupling or direct coupling. I chose the following: 6SN7 RC 6V6GT LC 300B.
Since this is a SE design, focus on the power supply. A cheap power supply is just that, a cheap power supply. Build separate supplies for the output stage and the input/driver stage. Use separate PTs, separate CLCLC filters, use hybrid rectification using hexfreds or schottkys AND tube rectifiers like 5AR4/GZ34 or GZ37s, or whatever. Use fixed bias on the ouput AND driver tubes, if you can. Schottky bridge to RCRCRC. Outboard the power supply. Any vibrations caused by the power supply will affect the audio circuit. If outboarding the supply, us DC to heat the filaments. Schottky bridge to RCLC.
Select parts for quality, period.
A non-audio example: Being a cyclist, I had a schwinn aluminum-framed road bike for about 10 years, which I decided to replace with a new bike last year. After a bit of research I decided on a custom built titanium/carbon fiber road bike from Dean USA. After my first long ride on my new bike, I found myself less fatigued than I had ever been in the past. Also, during the ride I was able to go faster and longer, because I was more comfortable and less stressed on the bike. This experience taught me that materials count, manufacturing quality is important and that customization is critical.
So for my amp I went with custom PTs and chokes from Electraprint, custom Amorphous C-Core OPTS and choke from Tribute, chokes from Bartolucci, Lundahl and S&B. Most of which are potted. Silver wire for signal, silver plated OFC for DC, tinned copper for AC and ground. Roederstein, Mills, Kimawe resistors. El-Con, Panasonic, black gate, elna and Jupiter caps.
DIYing from scratch can be a great experience or a horrible chore, depending on how you approach it.
One last thing, the *romantic* *coloured* reputation of the 300B has a lot more to do with ubiquitous use of 6SN7s (or other lame driver of choice) as drivers for these tubes. If there were more properly designed amps out there, then more people would agree with Thorsten's assesment that it is the best all around tube available for DIY.
And no, I do not find my 300Bs coloured or romantic in MY amp; although I do prefer my 2A3s at times (but, at 3watts I do run out of steam with my medium efficient 94dB speakers).
Just want to tag in on the end here and add my thoughts:
I agree with Mike C that a 300B is good first amp project.
I finished my first amp in November: a 2A3/300B convertible amp. 6SN7 as input and triode-strapped 6V6GTs as drivers.
And with respect to the recommendation of single manufacturer russian tubes (although by reputation some of them are excellent). One great thing about 300Bs is the broad selection of available tubes at all different price points: from EH at <$100 a pair to EML at >$500 a pair; and a bunch of chinese manufacturers in between. And of course WE at >$700 a pair. With the Sovteks or Svetlanas, they are what they are and you are married to them for the life of the amp. And if the company starts to have problems, then availability becomes an issue.
As as far as power output is concerned; if its a problem then its a problem with your speakers and room, not with the amp. Since you have yet to select speakers, then this should not be a problem.
FWIW, I would recommend the following:
Since, you asked for book recommendations. Valve Amps by Morgan Jones, Beginners Guide to Tube Audio Design by Bruce Rozenblit, Principles of Power by Kevin O'Connor, back issues of Sound Practice, any issues of Stereo Sound Tube Kingdom, a Japanese magazine, Vacuum Tube Valley, and RDH. And do a boatload of research on the web. Answers to every question you may have about tube design is available to you using the above resources.
Long-term value for the money is basically building the *right* amp once and NOT after a few half-assed attempts. As DIYers we are all capable of building the best amp we can the first time we build it without any mistakes or problems. Its a question of doing the proper R&D prep work. If its a question of experimentation and study then many take a different approach and build and rebuild and rebuild...
Another road is to buy a commerical kit and build someone else's design. Its easier than complete DIY, maybe more appropriate and less costly for a first-time builder, but understand that its not your amp but one of many others.
If you are serious about building the best amp you possibly can and are committed to the project...and considering this is a first amp project:
First look at all available 300B schematics. Decide on a two or three stage approach. I prefer a 3-stage design myself.
Choose what amplifying tube to use as input, I chose a 6SN7.
Choose a proper driver, NOT a 6SN7, another triode like 2A3, 45, or some triode stapped pentode. I chose a 6V6GT in triode mode.
An example: my last amp was a commercial 6SN7 DC 300B komuro-style amp. I bought a Emission Labs 300BLXS from Bob Ungemach at AMR, which according to EML can be used as a "drop-in" replacement in any 300B amp. I dropped the pair into the amp and all I got was distortion up the waazho. Blamed the tubes, of course. I then finished building my amp, running the 300Bs at the same or similar op of the prior amp, but with different drivers, and all I got was sweet music. Yes, its not an apples to apples comparison, and its an extreme example, but it did teach me something about drivers.
Choose coupling methods, RC coupling, LC coupling, IT coupling or direct coupling. I chose the following: 6SN7 RC 6V6GT LC 300B.
Since this is a SE design, focus on the power supply. A cheap power supply is just that, a cheap power supply. Build separate supplies for the output stage and the input/driver stage. Use separate PTs, separate CLCLC filters, use hybrid rectification using hexfreds or schottkys AND tube rectifiers like 5AR4/GZ34 or GZ37s, or whatever. Use fixed bias on the ouput AND driver tubes, if you can. Schottky bridge to RCRCRC. Outboard the power supply. Any vibrations caused by the power supply will affect the audio circuit. If outboarding the supply, us DC to heat the filaments. Schottky bridge to RCLC.
Select parts for quality, period.
A non-audio example: Being a cyclist, I had a schwinn aluminum-framed road bike for about 10 years, which I decided to replace with a new bike last year. After a bit of research I decided on a custom built titanium/carbon fiber road bike from Dean USA. After my first long ride on my new bike, I found myself less fatigued than I had ever been in the past. Also, during the ride I was able to go faster and longer, because I was more comfortable and less stressed on the bike. This experience taught me that materials count, manufacturing quality is important and that customization is critical.
So for my amp I went with custom PTs and chokes from Electraprint, custom Amorphous C-Core OPTS and choke from Tribute, chokes from Bartolucci, Lundahl and S&B. Most of which are potted. Silver wire for signal, silver plated OFC for DC, tinned copper for AC and ground. Roederstein, Mills, Kimawe resistors. El-Con, Panasonic, black gate, elna and Jupiter caps.
DIYing from scratch can be a great experience or a horrible chore, depending on how you approach it.
One last thing, the *romantic* *coloured* reputation of the 300B has a lot more to do with ubiquitous use of 6SN7s (or other lame driver of choice) as drivers for these tubes. If there were more properly designed amps out there, then more people would agree with Thorsten's assesment that it is the best all around tube available for DIY.
And no, I do not find my 300Bs coloured or romantic in MY amp; although I do prefer my 2A3s at times (but, at 3watts I do run out of steam with my medium efficient 94dB speakers).
DVDHack said:
Amp selection before you have the speakers is often a waste of time. The speaks will have a far larger effect on the sound in your room than the amp will. The amp can be chosen first if you have a lot of experience with different designs and how they work with a variety of speakers.
For instance, I like horns and he clarity and ease they bring to the presentation, and would much rather listen to my horn rig with a crappy receiver than my "best" amp driving most cone'n'dome squeezeboxes.
- Status
- Not open for further replies.