This seems like the right place to get an opinion about plate loads and driver stages for a 4P1L amp. My junkbox is overflowing with parts that really should be assembled into something useful, like a 4P1L PSE. Just to mention the more interesting items:
12x 4P1L
4x Loctal sockets
2x James output transformers, 2,5/3,5k:4/8R 10W (70mA DC IIRC)
2x LL1660 10mA interstage transformers
2x LL1671 50mA interstage transformers
2x 50H 40mA plate chokes (or was it 40H 50mA?)
2x Toroidal Power transformers, 210 + 105 + 11V 100mA + 100mA + 3A (estimated current ratings)
6x Double C-core chokes ~55mH Rdc >1 ohm 1A DC (estimated)
2X Double C-core chokes ~20mH Rdc >0,4 ohm 3A DC
+ various low voltage transformers, filter chokes, caps, input/driver tubes, HV mosfets, 10+ pole mil spec circular connectors, a multitude of large power resistors (some really big 40-300R), vintage CC resistors and much more.
I´m thinking triode wired PSE with 3,5k plate load for maximum linearity and damping factor, mosfet regulated B+, class A2 via mosfet "Power Drive", choke filtered filament supplies and separate chassis for the PSU.
12x 4P1L
4x Loctal sockets
2x James output transformers, 2,5/3,5k:4/8R 10W (70mA DC IIRC)
2x LL1660 10mA interstage transformers
2x LL1671 50mA interstage transformers
2x 50H 40mA plate chokes (or was it 40H 50mA?)
2x Toroidal Power transformers, 210 + 105 + 11V 100mA + 100mA + 3A (estimated current ratings)
6x Double C-core chokes ~55mH Rdc >1 ohm 1A DC (estimated)
2X Double C-core chokes ~20mH Rdc >0,4 ohm 3A DC
+ various low voltage transformers, filter chokes, caps, input/driver tubes, HV mosfets, 10+ pole mil spec circular connectors, a multitude of large power resistors (some really big 40-300R), vintage CC resistors and much more.
I´m thinking triode wired PSE with 3,5k plate load for maximum linearity and damping factor, mosfet regulated B+, class A2 via mosfet "Power Drive", choke filtered filament supplies and separate chassis for the PSU.
You won't regret it. I've had mine on 24/7 for over 2 years. Best amps I ever built, and no desire to build anything else. Replaced 300b, 2a3 and 6B4G SETs.
My speakers allow me to use just a single 4P1L in the output stage, so 5K OPT. But for 3.5K you should go PSE as you say.
Be sure to use filament bias right through - no cathode bypasses!! And run the 4P1Ls at least at 20mA. The 10mA LL1660 won't work, it will sound very thin. Interstage or plate chokes both good - I'm now using plate chokes with FT-3 teflon caps.
2 chassis is good, or even separate filament supply chassis as well.
My speakers allow me to use just a single 4P1L in the output stage, so 5K OPT. But for 3.5K you should go PSE as you say.
Be sure to use filament bias right through - no cathode bypasses!! And run the 4P1Ls at least at 20mA. The 10mA LL1660 won't work, it will sound very thin. Interstage or plate chokes both good - I'm now using plate chokes with FT-3 teflon caps.
2 chassis is good, or even separate filament supply chassis as well.
Sounds good! My front loaded horns would probably be happy with less than the 4-5W I´m expecting from this setup, but a bit of surplus power never hurts ( a phrase seldom used when talking about SE triode amps...
) I guess the 10mA IT´s can be connected with the primaries in // to allow 18/20mA, that should be enough if I want to use another 4P1L as input/driver tube? DH input tubes and IT coupling is of course tempting, but I ca´n´t help thinking that "Power drive" a´la Tubelab might be an easier way to push the output tubes grids into A2 operation without irregulaties at the zero voltage crossing point.
) I guess the 10mA IT´s can be connected with the primaries in // to allow 18/20mA, that should be enough if I want to use another 4P1L as input/driver tube? DH input tubes and IT coupling is of course tempting, but I ca´n´t help thinking that "Power drive" a´la Tubelab might be an easier way to push the output tubes grids into A2 operation without irregulaties at the zero voltage crossing point.
Sounds good! My front loaded horns would probably be happy with less than the 4-5W I´m expecting from this setup, but a bit of surplus power never hurts ( a phrase seldom used when talking about SE triode amps...
You really should use a DHT in the first stage. You should only need 2 stages if you are putting in a 2v signal from a DAC or other. The good thing about 4P1L outputs is not only filament bias and no cathode bypasses, but also the high mu of around 10-11. Plenty of gain to choose a DHT for the first stage with a mu of 10 to 15. So another 4P1L (and you could add a little gain with a small step-up in an IT if you needed) or even a 3A5 or 6J6, though I'd prefer the 4P1L.
If you can manage with slightly less mu in the first stage like around 8-9, then you have the 26 though I'd again prefer a 4P1L. You have the 10Y/801 which is certainly going to be good but expensive and here I'd again stick with a 4P1L.
But what you do have is Ale's 01A Gen 2 circuit. There isn't much doubt that this sounds better than a 4P1L first stage, in fact it sounds stunning. But I don't know that it would drive the 4P1L output(s) sufficiently with its low current. Ale and I did kind of try it and it was not obvious. I'm sure he'd tell you more about this. There may be some fixes to use this front end, and if so it would be a bit of a leap forward for a 2 stage design. It works like a charm in a 3 stage design, e.g. driving a 4P1L into a 4P1L.
Thanks for the input. Maybe something like Michael Kosters 4-65A amp with a gyrator loaded input stage, direct coupling and stacked power supplies would work? My toroids with 210 + 105 + 11VAC windings should fit quite well.
As soon as I´ve finished a couple of other project (might take hours or years...) I´ll start building a prototype ona breadboard.
As soon as I´ve finished a couple of other project (might take hours or years...) I´ll start building a prototype ona breadboard.
01a into 4P1L PSE
Hi there,
as Andy well says, a winning combination is 01a driving the 4P1L (PSE) stage. Given the low driving requirements of the 4P1L (what a beautiful valve) you can use the 01a and achieve great performance within the limitations of the 01a as a driver. However, I'd highly recommend the addition of a follower stage (of your choice either cathode follower or a sandy version as in diagram below). Otherwise you may encounter some slew rate challenges and a loss of the treble - been there done that. I think this is a great circuit to implement:
You could implement it with filament bias though, but I prefer fixed bias in my experience.
BTW: I had a monster 814SE Amp based on Michael Koster's design. An amazing amp which is now retired due to my 1 year old daughter
Hi there,
as Andy well says, a winning combination is 01a driving the 4P1L (PSE) stage. Given the low driving requirements of the 4P1L (what a beautiful valve) you can use the 01a and achieve great performance within the limitations of the 01a as a driver. However, I'd highly recommend the addition of a follower stage (of your choice either cathode follower or a sandy version as in diagram below). Otherwise you may encounter some slew rate challenges and a loss of the treble - been there done that. I think this is a great circuit to implement:
You could implement it with filament bias though, but I prefer fixed bias in my experience.
BTW: I had a monster 814SE Amp based on Michael Koster's design. An amazing amp which is now retired due to my 1 year old daughter
Andy,
One 4P1L represents about 200pF load to the driver. At 20kHz the charging current at maximum swing (around 26Vpeak) to push the 4P1L to maximum output, is about 650uA. You want at least x5 current driving capability so it will be 3.2mA. Just about there. For 2 output valves, clearly you will need double that (due to capacitance x2) so a follower with more than 6mA quiescent current is needed. 10mA is more than enough.
I'd be keen to listen to your 01a preamp when ready!
cheers
Ale
One 4P1L represents about 200pF load to the driver. At 20kHz the charging current at maximum swing (around 26Vpeak) to push the 4P1L to maximum output, is about 650uA. You want at least x5 current driving capability so it will be 3.2mA. Just about there. For 2 output valves, clearly you will need double that (due to capacitance x2) so a follower with more than 6mA quiescent current is needed. 10mA is more than enough.
I'd be keen to listen to your 01a preamp when ready!
cheers
Ale
I apologize for my delayed response.
I totally agree with you.
Fortunately, I have come across unforgettable acoustic experiences;
a) I have brought my first version of 811A SE amp for the10th MJ Diy amp contest held at Audio-technica Astrostudio in Tokyo on Oct, 5, 2002.
https://www.audio-technica.co.jp/corp/astrostudio/info.html
b) 8025A SE amp (12W) in ETF2008, Biezenmortel, in Nether land.
ETF2008
c) 4P1Lx3 PSE amp (8W) in the 28th Toyama craft Audio club exhibition.
http://www.diyaudio.com/forums/attachments/tubes-valves/522161-one-more-4p1l-se-.jpg
Hiro
I'm sure the "concert hall" experience is without equal. Best I've heard in a large scale system is the Apogee full-range ribbons. Beyond 4P1L territory....
For myself, I'm a professional musician, though don't play that much these days. So in a large room, I'm more used to live music. At home it's basically ambient music in a smaller room, while I'm on my computer or watching films or TV. All goes through my 4P1L SET. So my priorities are always timbre and transparency. Soundstage has nothing to do with it. Quality of sound does!
For myself, I'm a professional musician, though don't play that much these days. So in a large room, I'm more used to live music. At home it's basically ambient music in a smaller room, while I'm on my computer or watching films or TV. All goes through my 4P1L SET. So my priorities are always timbre and transparency. Soundstage has nothing to do with it. Quality of sound does!
Hi Hiro,
I have downloaded a copy of your presentation on "Single-ended triode Amplifiers using zero-bias transmitting tubes".
I could scratch my head forever if I had to pick one. All lovely amplifiers. Taking into account all factors I would choose the 22W 811 SE.
I`m with you that the "concert hall" is superior than that of my listening room.
It`s cool you listened PC or TV sound through your 4P1L SET.
I guess you use a small/medium size of high quality SP.
Thank you for your affirmative view to my amplifiers.
I do attempt lots of touch up jobs to them from time to time for optimization in the given listening room environment.
As a result almost all have been modified from original schematics.
At present, I like the 4P1Lx3 PSE at best.
I am going to return to the initial "Gubernator" project, but with 5 x 4P1L in parallel per channel. Stay tuned!
Good luck on your up coming "new Gubernator" project 5x4P1L.
Yes indeed - Anatoliy is in the Hall of Fame for 4P1L amps already.... More to come!
Hi Hiro,
if you say it is this good then I might build one for myself. I happen to have suitable output transformers (Lundahl LL1627/90mA 2.3K) and some 4P1L's that I bought when prices were really low. Also have the other signal tubes you used. I only need to make a good power transformer and chokes. Possibly at the end of August when I will have good time to work on it.
Already in good sized room like 7m x 5m incredible results can be achieved. However this requires the room has to be acoustically treated and a lot of work. Modern digital stuff is simply not as good and at some point the level of manipulation becomes arbitrary doesn't bring anywhere. Of course it can't be a normal living room. In smaller rooms like 3.5 x 5 the limitation occurs with larger ensembles but still incredibly good with small ensembles. You can get a soundstage bigger than the mere physical distance between the walls. However our brain will soon win over this "impossible" presentation and concentrate on the music.
Last edited:
Already in good sized room like 7m x 5m incredible results can be achieved. However this requires the room has to be acoustically treated and a lot of work. Modern digital stuff is simply not as good and at some point the level of manipulation becomes arbitrary doesn't bring anywhere. Of course it can't be a normal living room. In smaller rooms like 3.5 x 5 the limitation occurs with larger ensembles but still incredibly good with small ensembles. You can get a soundstage bigger than the mere physical distance between the walls. However our brain will soon win over this "impossible" presentation and concentrate on the music.
I'm sure all this is true. But in my case I'm not interested in soundstage. When I played in orchestras I was in the middle of everything in the bass section, and most of the orchestra was in front of me. When I play in jazz bands I can be anywhere on stage. Wherever I am, it's the instrumental timbre and the transparency I listen for. At home I may be listening half way down the room at 90* to the speakers. I may be listening in bed even. I rarely sit in a "listening position" because I'm usually doing other things at the same time as listening. Plus I'm rarely in the "listening position" at live events - I'm more likely to be on stage.
We all vary in how we listen and what we listen to. Soundstage means nothing to me, but everything to others. I can go into ecstasy over the tone of a Steinway piano, or a particular singer, or brushes on a snare, which is why I use DHTs and nothing else - I get that exquisite tone from them. Other listeners just hear a singer, some drums and a piano.