That is an amazing build, very nicely laid out. Its a similar spec to my 4P1L which takes up an entire 2U rack chassis. I also use mine to drive an F4 which makes a great combo 🙂
Radu - nice work on the PCBs. The only comment I would make straight away is about the parts used - the cathode resistors and the coupling caps. I know that miniaturisation is your goal, but I would want to use bigger parts. Resistors - for wirewound I would use at least vitreous enamel like for instance Welwyn W24 series. Or even something more exotic. They did sound better to me than the standard ceramic ones. And I would certainly use Russian Teflon coupling caps, FT-3 preferred very slightly to K72, but not much difference. These parts are not friendly to PCBs!!! If you use 0.22uF teflon caps they're huge, same if you parallel two 0.1uF. Although now I think of it you can use the FT-2 version which is rated 200v but good for more.
It's always a design issue when the ideal part won't fit on a PCB - I'm sure we've all been there. You can always add them off the board - the usual solution.
It's always a design issue when the ideal part won't fit on a PCB - I'm sure we've all been there. You can always add them off the board - the usual solution.
Andy,
Great day indeed. Your system sounds superb overall. I urge everyone to build the 4P1L PSE amp.
Here are some notes and pictures of a fantastic testing and listening session
DHT day | Bartola Valves
Cheers,
Ale
Great day indeed. Your system sounds superb overall. I urge everyone to build the 4P1L PSE amp.
Here are some notes and pictures of a fantastic testing and listening session
DHT day | Bartola Valves
Cheers,
Ale
I use a 4P1L "line stage" as the first stage in my 2-stage amp. I've been using some old parts I had - LL1660/18mA and Hammond 126C 15mA. Ale's tests showed that with PSE 4P1L you really have to be putting 25mA or 30mA through the 4P1L. I'll be getting something like a Hammond 126B 30mA to use as a plate choke with a FT-3 cap. That should make a difference. The "Daemon" amp actually works and sounds very good, though it did physically attack Ale's measurement equipment. I'll look inside and see if there's anything obviously out of order. We never figured out what the issue was.
Very enjoyable day. I can only repeat that Ale's 01A Gen 2 is remarkable, and anyone using a preamp should build it. Fortunately I have a big box full of 01As. He-he.....
Very enjoyable day. I can only repeat that Ale's 01A Gen 2 is remarkable, and anyone using a preamp should build it. Fortunately I have a big box full of 01As. He-he.....
Andy,
Yes indeed, PCBs are not friendly to many good components. I think there is enough room for a W24. The reason I biased the 4P1L at 5Vish was to reduce the amount of heat from the bias resistor ( in the picture is a non- inductive 8 ohm 20W that dissipates less than 3W). The coupling cap is another story... I found that the Mundorf Supreme is good anough for me in combination with the First Watt F4.
Cheers,
Radu
Yes indeed, PCBs are not friendly to many good components. I think there is enough room for a W24. The reason I biased the 4P1L at 5Vish was to reduce the amount of heat from the bias resistor ( in the picture is a non- inductive 8 ohm 20W that dissipates less than 3W). The coupling cap is another story... I found that the Mundorf Supreme is good anough for me in combination with the First Watt F4.
Cheers,
Radu
A question to Kevin: What Active Output Stage you refer to in the message below?
I quote your site:
"A selection of “modules” for the construction of a SE Line Stage using your choice of directly heated tubes such as the classic 26 or the Russian 4P1L and your choice of Lundahl line output transformers. This will be a low cost alternative to the excellent Active Output Stage Kit, but with sonics that put it in league with commercial equipment costing much more."
Br
zz
I quote your site:
"A selection of “modules” for the construction of a SE Line Stage using your choice of directly heated tubes such as the classic 26 or the Russian 4P1L and your choice of Lundahl line output transformers. This will be a low cost alternative to the excellent Active Output Stage Kit, but with sonics that put it in league with commercial equipment costing much more."
Br
zz
It is 1/Gm. If Gm in mA/V then the result is in kohms.
I've measured lots of 01As and Gm varies between 650 and 800 Micromhos ( or 0.65-0.8 mA/V). The output impedance will then be between 1.54-1.25kohms.
Best,
Radu
I've measured lots of 01As and Gm varies between 650 and 800 Micromhos ( or 0.65-0.8 mA/V). The output impedance will then be between 1.54-1.25kohms.
Best,
Radu
Are you sure? I would say it is 1/gm, indeed, but the gm of the Jfet, which is probably much higher, probably more than 10mA/V at 3mA. So output impedance would be 100 ohms or less.
Yes you're right Erik. It is 1/gm of the cathode follower. My wrong.
Actually I measured Zout in the past. Looked on my notes and I measured about 450 ohms ( using two different loads 500K and 10k). Still not 1/gm of the Jfet ( I am using BLs so the transconductance must be 5-6mA/V and up).
Actually I measured Zout in the past. Looked on my notes and I measured about 450 ohms ( using two different loads 500K and 10k). Still not 1/gm of the Jfet ( I am using BLs so the transconductance must be 5-6mA/V and up).
Ale, instead of a "gyrator" load have you tried using the mu output of a CCS? See lower left schematic in the attached figure vs upper left (similar to your arrangement, really a mu follower not a gyrator). Similar low impedance output but one less signal path cap needed.
I use the CCS mu output in my 26 preamp with great success. I use a circuit similar to yours in my 833 amps to drive the 833.
I use the CCS mu output in my 26 preamp with great success. I use a circuit similar to yours in my 833 amps to drive the 833.
Attachments
Hi Magz, indeed. The mu-follower output is what I've been using for years for both preamps (4P1L and 01a) as well as for drivers (e.g 46 DC-coupled into 814). As you experienced for sure in your outstanding amp and preamp, you get best drive and very low distortion.,I also prefer the sound of this topology, that has been my experience so far over the years.
Agreed. My question was more around the constant current (bottom row) versus constant voltage (top row) versions of the circuit. Have you tried the constant current version (without the cap) in your O1A preamp?
I did some few tests with CCS and didn't notice big differences. Mind you that was on my test amp in the workshop. I think is worth testing it if you can hack the PCB to allow both constant voltage and current modes
Can you describe your impressions compared to an transformer coupled amplifier stage?
When I was testing prospective driver tubes I found I got better objective performance from the Gyrator v's the ccs whilst using the same dn2540. I didn't compare using 2 mosfet in ccs, which I'd think would have improved the ccs performance.
I have made a couple of changes during last few days:
1. wired LL1671 from 4:1 to 2:1 - I'm not sure why but that was a very big difference in sound quality. Lots of air, holographic image.
2. added a diode to 3d grid - bass became better (controlled)
3. switched from Alps to Elma 04 24 switch - I feel that the switch added to resolution, but it could be just my imagination 🙂
When I turn the volume I got noisy clicks, especially when I leave it on playing for a while (some capacitance builds up?). How to handle this issue?
Also I still have some whistling buzz in the background (annoying). When I added another choke to heaters supply it was gone, but then I changed something (I don't remember what) it came back. So I think it is connected with fields, cabling, intercapacitances.
Anyway, even with this problems I can say that it is a very good pre-amp in terms of music quality. I am comparing with my SS Classe pre-amp which is of course more convenient and has better control of difficult passages with lots of instruments. There is also some lack of details in mid-bass but voice in 4P1L pre-amp is something fantastic. Also a difference in price!
1. wired LL1671 from 4:1 to 2:1 - I'm not sure why but that was a very big difference in sound quality. Lots of air, holographic image.
2. added a diode to 3d grid - bass became better (controlled)
3. switched from Alps to Elma 04 24 switch - I feel that the switch added to resolution, but it could be just my imagination 🙂
When I turn the volume I got noisy clicks, especially when I leave it on playing for a while (some capacitance builds up?). How to handle this issue?
Also I still have some whistling buzz in the background (annoying). When I added another choke to heaters supply it was gone, but then I changed something (I don't remember what) it came back. So I think it is connected with fields, cabling, intercapacitances.
Anyway, even with this problems I can say that it is a very good pre-amp in terms of music quality. I am comparing with my SS Classe pre-amp which is of course more convenient and has better control of difficult passages with lots of instruments. There is also some lack of details in mid-bass but voice in 4P1L pre-amp is something fantastic. Also a difference in price!
Wired as 2:1 means lower inductance, less bass response.
Probably that's why it sounds different.
Probably that's why it sounds different.