the 1:2 It must work but 1:1 will always be better and the input transformer certainly not or if so i will buy dozen of them 🙂
Are you suggesting that the input transformer be eliminated? What are your thoughts about eliminating it completely?
no , you need it to have some gain but even with a good model like the lundahl LL7903 with 1.8 ratio will sound worse than adding a tube gain stage .
some tube schematic are just for style with exotic name but in the real life can be a waste of money .
Are there any specific tubes that you would recommend to replace the input transformer and the driver 50 tube? Trying to kill 2 birds with 1 stone, as they say.
That’s quite a step-up at the input!
Signal level at the input ?
You can use eml-30a and a 1:1 it
Works very well
Signal level at the input ?
You can use eml-30a and a 1:1 it
Works very well
Last edited:
if want a mix retro:modern
or the mix with a small pentode WE717 , WE 310a , 6J7 , 6C6 " a la yamamoto" style to stay in the exotic name
if want a mix retro:modern
or the mix with a small pentode WE717 , WE 310a , 6J7 , 6C6 " a la yamamoto" style to stay in the exotic name
Juanitox, thanks for sharing the schematics. Do you have experience with either of these schematics? Preference? I have not done any A/B comparison of interstage tranny vs. cap/resistor. Any thoughts on this, as both of your schematics eliminate the interstage tranny.
i have tried interstage vs Choke loaded or RC coupling and prefer the later but the only IT i have tried is the lundahl 1660 , i don't think a PIO of mica coupling cap could ruin the sound a bad IT certainly will.
i have tried both type of driver didn't like the C3M " to good to be true "
i prefer another aproach with small vintage pentode WE310a EF37A 6J7 in pentode mode with a VR tube regulation on the screen , or DHT like 26 , 71a , 46 if you don't need gain or if you have a preamp with high gain. the 50 is a fantastic sounding tube , i will first try it with a small and cheap pentode ( except real WE310a) in pentode or Triode mode. 🙂
i have tried both type of driver didn't like the C3M " to good to be true "
i prefer another aproach with small vintage pentode WE310a EF37A 6J7 in pentode mode with a VR tube regulation on the screen , or DHT like 26 , 71a , 46 if you don't need gain or if you have a preamp with high gain. the 50 is a fantastic sounding tube , i will first try it with a small and cheap pentode ( except real WE310a) in pentode or Triode mode. 🙂
I have a schematic somewhere.. Will post when I find it. if you take away the input stepup from your schematic and change the it to 1:1 you're almost there...
I made a se 50 4 or5 years ago, settled on the EML 30a as driver it coupled to the 50 with the iso nc 20f interstage transformer.
Attached a measurement from the secondary of the interstage
(secondary open during measurement)
1 khz 77,3 Vrms out. The EML 30a is one hell of a driver tube...
also attached 1 khz, 116Vrms output
I made a se 50 4 or5 years ago, settled on the EML 30a as driver it coupled to the 50 with the iso nc 20f interstage transformer.
Attached a measurement from the secondary of the interstage
(secondary open during measurement)
1 khz 77,3 Vrms out. The EML 30a is one hell of a driver tube...
also attached 1 khz, 116Vrms output
Attachments
Do you have four type 50 tubes to use for this project? Do you have a solid state preamp or a tube power amp to drive the input transformer?
I'm trying to achieve the same kind of a result with a 300b output in another thread. If I read it correctly the 50 has a mu of 3.8. So like with the 300b, choices I faced were:
- increase the input voltage to 5 or 8 volts or whatever. In my case this comes from a AK4490 DAC and it could mean adding another opamp stage. Maybe clever stuff like a balanced out DAC and a balanced to SE input transformer.
- put a step-up transformer (SUT) on the input. It had better be good, though.
- add another tube stage. Thoriated is nice, like 10Y or 01A.
- use a directly heated pentode. I'm in the process of trying this with a 4P1L and getting a lot of help with calculations since I've never used a pentode stage. Hopefully I'll have some results later. Ale Moglia documents this in his Bartola blog.
- use an indirectly heated tube and do it with 2 stages. The usual solution. But it's not directly heated if you must have that as a priority.
- Use a higher gain output tube. A 4P1L has a gain of around 10 so that's already half way to a solution. I use a pair in PSE. But there's a few higher gain outputs, like the 211 for starters. Some are high voltage, though.
- use a higher gain directly heated input tube. This will probably be a rare beast and cost a lot of money, but if you don't care about that, the EML beckons.
- ???? probably forgot other solutions
I can't help at this stage I'm afraid, since I haven't found the answer. I rejected a 6e6P stage though it has fans because the 4P1L is just a superior tube for me. Battling with the 4P1L in pentode and not there yet. A LL1540 in 1:2 was a failure. If there's a Nirvana I haven't found it yet.
But in the meantime, 4P1L into PSE 4P1L has worked for me for years with filament bias. I sometimes use a 26 on the input. 26 smoother, 4P1L more detailed.
- increase the input voltage to 5 or 8 volts or whatever. In my case this comes from a AK4490 DAC and it could mean adding another opamp stage. Maybe clever stuff like a balanced out DAC and a balanced to SE input transformer.
- put a step-up transformer (SUT) on the input. It had better be good, though.
- add another tube stage. Thoriated is nice, like 10Y or 01A.
- use a directly heated pentode. I'm in the process of trying this with a 4P1L and getting a lot of help with calculations since I've never used a pentode stage. Hopefully I'll have some results later. Ale Moglia documents this in his Bartola blog.
- use an indirectly heated tube and do it with 2 stages. The usual solution. But it's not directly heated if you must have that as a priority.
- Use a higher gain output tube. A 4P1L has a gain of around 10 so that's already half way to a solution. I use a pair in PSE. But there's a few higher gain outputs, like the 211 for starters. Some are high voltage, though.
- use a higher gain directly heated input tube. This will probably be a rare beast and cost a lot of money, but if you don't care about that, the EML beckons.
- ???? probably forgot other solutions
I can't help at this stage I'm afraid, since I haven't found the answer. I rejected a 6e6P stage though it has fans because the 4P1L is just a superior tube for me. Battling with the 4P1L in pentode and not there yet. A LL1540 in 1:2 was a failure. If there's a Nirvana I haven't found it yet.
But in the meantime, 4P1L into PSE 4P1L has worked for me for years with filament bias. I sometimes use a 26 on the input. 26 smoother, 4P1L more detailed.
20B / 30A drives 50
I re-did the schematic (see below) using an EML 20B driver tube (mu = 20) or an EML 30A driver tube (mu=32). Eliminated the input transformer. I have a few questions:
1. What values should the interstage transformer have (Henry and Amps). Please let me know how you derived these values, as I am in the learning stage and would appreciate the help.
2. Is a grid Leak resistor necessary if using an interstage transformer?
What do you think???
I re-did the schematic (see below) using an EML 20B driver tube (mu = 20) or an EML 30A driver tube (mu=32). Eliminated the input transformer. I have a few questions:
1. What values should the interstage transformer have (Henry and Amps). Please let me know how you derived these values, as I am in the learning stage and would appreciate the help.
2. Is a grid Leak resistor necessary if using an interstage transformer?
What do you think???
Attachments
If you can find the schematic (with details), it would be very helpful. See me posted below with a new schematic 20B/30A --> 50.
Thanks!
EML20B should have a load impedance that's about 10K, so we would want an interstage transformer with ~100H of primary inductance that has been designed to function as a 10K IT.
The Hammond 126C looks to be up to the job, albeit at lower plate current than the book operating point that EML gives. The Sowter 8424 is also workable, and you could get some support from Brain regarding running it as a 10K unit.
The Hammond 126C looks to be up to the job, albeit at lower plate current than the book operating point that EML gives. The Sowter 8424 is also workable, and you could get some support from Brain regarding running it as a 10K unit.