Why go with PSE 2A3 at all? A single 300B will net you slightly more power and depending on tube choice will be comparable in cost.
My experience with PSE 300B, paralleled triodes in general is that they tend to give up something in terms of low level resolution and dynamics - what audiophiles often term micro-dynamics, etc.. Hard to explain, but in many instances the single tube version just sounded better overall, in otherwise very similar amplifier designs..
My experience with PSE 300B, paralleled triodes in general is that they tend to give up something in terms of low level resolution and dynamics - what audiophiles often term micro-dynamics, etc.. Hard to explain, but in many instances the single tube version just sounded better overall, in otherwise very similar amplifier designs..
I am very happy with my SE 2A3 mono's. I only use one 7A4 triode to drive it. I am using RCA black plate 2A3.
The good thing about 2A3 is that it is very quiet even with AC 2.5V filament supply.
I am using Hammond 1627SE OPT which is 2.5K primary impedance. It's has very good bass because of its high inductance. The treble is not as good as the Plitron but it is good enough for me.
Johnny
The good thing about 2A3 is that it is very quiet even with AC 2.5V filament supply.
I am using Hammond 1627SE OPT which is 2.5K primary impedance. It's has very good bass because of its high inductance. The treble is not as good as the Plitron but it is good enough for me.
Johnny
kevinkr said:
You can get a pair of Sovtek 2a3's for about $70, while the cheapest 300B's I know of (EH) are about $120 per pair, so the quad of 2a3's actually only end up about $20 more. Moreover, the sovtek 2a3's have a pretty good reputation while the EH 300B's don't.
Plus, you get a quite lower Rp with the pair of 2a3's, which might be important.
kevinkr said:
Here, however, I think you are probably right. No matter how much better the cheap 2a3 is compared to the cheap 300b, if a PSE setup is inherently detrimental, it won't save you. I've used PSE with small signal tubes (6n1p, 5687, 6n6p, 6dj8, etc.) and not noticed an issue of sound degradation, but it may be different in a DHT. I'd tend to trust Kevin' ears and experience here.
Kevin said:
"My experience with PSE 300B, paralleled triodes in general is that they tend to give up something in terms of low level resolution and dynamics - what audiophiles often term micro-dynamics, etc.. "
I don't have much experience listening to flea power amps since running Dahlquist and Vandersteen speakers for years. So am I to assume that NOS American 2A3s, that are actually two triodes in a single plate structure, fall into this catagory? I have dozens of this style and only one old single plate version. Or does it have more to do with the extra wiring involved or even slightly miss-matched tubes?
Victor
"My experience with PSE 300B, paralleled triodes in general is that they tend to give up something in terms of low level resolution and dynamics - what audiophiles often term micro-dynamics, etc.. "
I don't have much experience listening to flea power amps since running Dahlquist and Vandersteen speakers for years. So am I to assume that NOS American 2A3s, that are actually two triodes in a single plate structure, fall into this catagory? I have dozens of this style and only one old single plate version. Or does it have more to do with the extra wiring involved or even slightly miss-matched tubes?
Victor
The 2A3 you have got, are not actually double triodes. They just share two series of filaments. If you notice the pins inside the two plates are connected to each other. It is just the new type of 2A3 compared to the old single plate ones you have got. Maybe the old ones are 45?
I have proposed to use Sovteks as a very good ballance of price and sound quality and I read in this thread that you propose them too. Besides they are single plate tubes, which maybe gives an advantage? For sure it is more pleasant to the eye... Are they really a good replacement in a relatively low price?
To speak more specificly, I am thinking of making the PSE circuit shown here http://www.audiodesignguide.com/my/pse3.html
The italian is good, using simple topology an interstage transformer.
There is even an SE version of his here
http://www.audiodesignguide.com/my/2a3_amp.html
But a single 2A3 has 3.5W of output pwr, I think it is too low for my non-horn based speakers, although the room is very small...
That is why I was thinking of the PSE solution, so I would really love to write me some info about the PSE and SE issue related to 2A3. (sonically basically)
I have proposed to use Sovteks as a very good ballance of price and sound quality and I read in this thread that you propose them too. Besides they are single plate tubes, which maybe gives an advantage? For sure it is more pleasant to the eye... Are they really a good replacement in a relatively low price?
To speak more specificly, I am thinking of making the PSE circuit shown here http://www.audiodesignguide.com/my/pse3.html
The italian is good, using simple topology an interstage transformer.
There is even an SE version of his here
http://www.audiodesignguide.com/my/2a3_amp.html
But a single 2A3 has 3.5W of output pwr, I think it is too low for my non-horn based speakers, although the room is very small...
That is why I was thinking of the PSE solution, so I would really love to write me some info about the PSE and SE issue related to 2A3. (sonically basically)
I've used single plate Sovtek 2A3 in one of my amplifiers and was quite pleased with the result working into a 3K transcendar opt.. Not quite as good as the somewhat rare dual box plate Kenrads I currently use, but very close for about 1/4th the price.
The original RCA 2A3 was a single plate design with Harp style filaments and is considered legendary. It's also unobtainium at any price approaching reasonable, and rare enough as it was only made for a short while. The 2A3 that replaced it should technically be considered a dual triode as there are two complete triodes internally connected in parallel.
Paralleling 2A3 requires a transformer with a primary Z of around 1.25K to achieve maximum power output, if like I do you buy custom made transformers for your projects this is not much of a problem - and Doug is right the lower rp is beneficial, however off the shelf transformers are a bit of a compromise.. Probably a UBT-1 at 1.6K would be a fair compromise.
Interestingly enough JJ does make a 2A3 on steroids which is basically nothing more than a 300B with a 2.5V filament.. Boost plate voltage and current and you should be able to get a lot more power out of one than is typical for the type. (Plate dissipation is rated at 40W.) Note get these burned in and fully tested.
SOT: (Slightly off topic)
I am using older JJ 300B (probably >5yrs old) and have been delighted with their performance in my 300B amplifiers. Yes, the WE300 is a little better, but it's surprisingly close for the huge difference in money. The JJ filaments use 1.5A rather than the 1.2A common in other 300B which is a problem for interchangeability with other brands if you use ccs for heating as I do.
BOT: (Back on topic)
My opinion of PSE is mostly based on the AN Keegon which I found to be a very disappointing amplifier compared to other 300 B SE amplifiers I was working on at the time. I have used up to 4 parallel sections of 5751 in phono stages before discovering high transconductance triodes and found their performance quite acceptable - I was careful to allow each section the luxury of self bias - this incidentally does not seem to work well with leds or fixed bias - only with cathode resistors (bypassed or not) as they no longer operate at the same point on their respective load lines due to variations in tube to tube transconductance. Independently biasing each section improved distortion performance and increased the overall gain significantly with no other change.
PSE with cathode bias you might want to provide each tube with a separate filament winding and cathode bias network, if fixed bias (my approach) the bias should be independently adjustable for each output tube again with separate filament windings to facilitate measuring the idle current.
One thing the 2A3 does have going for it is the voltage drive requirements are easily met in a robust single stage, not quite the case with the 300B unless an interstage transformer is used. (D3A/5842 driver.)
The original RCA 2A3 was a single plate design with Harp style filaments and is considered legendary. It's also unobtainium at any price approaching reasonable, and rare enough as it was only made for a short while. The 2A3 that replaced it should technically be considered a dual triode as there are two complete triodes internally connected in parallel.
Paralleling 2A3 requires a transformer with a primary Z of around 1.25K to achieve maximum power output, if like I do you buy custom made transformers for your projects this is not much of a problem - and Doug is right the lower rp is beneficial, however off the shelf transformers are a bit of a compromise.. Probably a UBT-1 at 1.6K would be a fair compromise.
Interestingly enough JJ does make a 2A3 on steroids which is basically nothing more than a 300B with a 2.5V filament.. Boost plate voltage and current and you should be able to get a lot more power out of one than is typical for the type. (Plate dissipation is rated at 40W.) Note get these burned in and fully tested.
SOT: (Slightly off topic)
I am using older JJ 300B (probably >5yrs old) and have been delighted with their performance in my 300B amplifiers. Yes, the WE300 is a little better, but it's surprisingly close for the huge difference in money. The JJ filaments use 1.5A rather than the 1.2A common in other 300B which is a problem for interchangeability with other brands if you use ccs for heating as I do.
BOT: (Back on topic)
My opinion of PSE is mostly based on the AN Keegon which I found to be a very disappointing amplifier compared to other 300 B SE amplifiers I was working on at the time. I have used up to 4 parallel sections of 5751 in phono stages before discovering high transconductance triodes and found their performance quite acceptable - I was careful to allow each section the luxury of self bias - this incidentally does not seem to work well with leds or fixed bias - only with cathode resistors (bypassed or not) as they no longer operate at the same point on their respective load lines due to variations in tube to tube transconductance. Independently biasing each section improved distortion performance and increased the overall gain significantly with no other change.
PSE with cathode bias you might want to provide each tube with a separate filament winding and cathode bias network, if fixed bias (my approach) the bias should be independently adjustable for each output tube again with separate filament windings to facilitate measuring the idle current.
One thing the 2A3 does have going for it is the voltage drive requirements are easily met in a robust single stage, not quite the case with the 300B unless an interstage transformer is used. (D3A/5842 driver.)
For my next amp I'm going with the JJ 2A3-40.
About $209 on ebay or Triode Electronics (call for a price, they told me they will match the price you find on ebay).
As Kevin said, this is a 300B with 2.5V heaters....
So if you want the power of a 300B without the hum from 5vac heaters this may be the way to go.
About $209 on ebay or Triode Electronics (call for a price, they told me they will match the price you find on ebay).
As Kevin said, this is a 300B with 2.5V heaters....
So if you want the power of a 300B without the hum from 5vac heaters this may be the way to go.
Well in Greece is not too easy to find this tube.. But the best brands of tubes you can find, are from overseas allways indeed..
2A3 is one of the most sweet tubes (first being 45 with a much more low power that makes it more impractical to use).
Besides there are 2A3 like sovtek monoplate that have higher dissipation and can be driven more roughly to give a greater output power (although I have not test the compromise in sound above nomimal levels).
As far as I have read, using a PSE2A3 can give 7-8W of power which is fair enough for not so extremely sensitive speakers >92db, yet still maintaining the sweetness of this tube type...
2A3 is one of the most sweet tubes (first being 45 with a much more low power that makes it more impractical to use).
Besides there are 2A3 like sovtek monoplate that have higher dissipation and can be driven more roughly to give a greater output power (although I have not test the compromise in sound above nomimal levels).
As far as I have read, using a PSE2A3 can give 7-8W of power which is fair enough for not so extremely sensitive speakers >92db, yet still maintaining the sweetness of this tube type...
Is WE 300B really worth this much (1K each, used)?
From a local CL, http://sfbay.craigslist.org/sfc/ele/430998781.html
From a local CL, http://sfbay.craigslist.org/sfc/ele/430998781.html
Re: Is WE 300B really worth this much (1K each, used)?
Yes, unfortunately, early 1950's vintage WE 300B that test as new generally go for about this sort of money. Later vintage ones command just a little less, and earlier ones somewhat more. The Japanese tend to snap these up, and I would be more than a little surprised if he sells them successfully on Craig's List given the local nature of the list.. Audiogon would be a more suitable low cost venue for those tubes, and eBay would get much more international exposure - at a substantial additional cost to the seller.
pchw said:From a local CL, http://sfbay.craigslist.org/sfc/ele/430998781.html
Yes, unfortunately, early 1950's vintage WE 300B that test as new generally go for about this sort of money. Later vintage ones command just a little less, and earlier ones somewhat more. The Japanese tend to snap these up, and I would be more than a little surprised if he sells them successfully on Craig's List given the local nature of the list.. Audiogon would be a more suitable low cost venue for those tubes, and eBay would get much more international exposure - at a substantial additional cost to the seller.
2A3 sovtek are good to use like test tubes for 1st start amplifier...
If You get lucky to find matched closly...
(My friend tested Sovtek 300b and from 5 pieces 3 was very fast vent to cathode break...)
for parallel SE it is for me no sense to use rare single plate 2A3 because of the fact that
topology is already paralell...
If You get lucky to find matched closly...
(My friend tested Sovtek 300b and from 5 pieces 3 was very fast vent to cathode break...)
for parallel SE it is for me no sense to use rare single plate 2A3 because of the fact that
topology is already paralell...
I was lucky enough to obtain a factory platinum mached quad of sovtek 2A3.
This means that they have been operated for 24 hours at the factory and then matched.
Besides 2A3 SP sovteks have a higher dissipation. They are like small 300B in look.
Well now, this does not mean that I will drive the tube harder but this means practically that the distortion levels will be less at the 7W output power than a lower dissipation tube... I think..
This means that they have been operated for 24 hours at the factory and then matched.
Besides 2A3 SP sovteks have a higher dissipation. They are like small 300B in look.
Well now, this does not mean that I will drive the tube harder but this means practically that the distortion levels will be less at the 7W output power than a lower dissipation tube... I think..
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.