What about my calculations for selected bias 28mA, Ih-600mA, -20V BIAS, 235Va?
B+ 235+20+5 OPT = 260V
Filament bias resistor 34R (dissipation 12W) so 50W
B+ 235+20+5 OPT = 260V
Filament bias resistor 34R (dissipation 12W) so 50W
For 650mA could I use these 30R 1% 50W non inductive resistors?
-19,5V is very near -20V...
I ask because I can't get 34R 50W non inductive resistors.
-19,5V is very near -20V...
I ask because I can't get 34R 50W non inductive resistors.
Attachments
Last edited:
DH tubes "likes" if the filament is starved, for example 4P1L 0.6A instead of 0.65A .... if you use it as preamp (no need essential current).
If you use it as power stage, I not recommend the filament current reduction.
-20V bias requires 18.95V (20-(2.1/2)) on filament bias resistor, so with 0.65A current the required filament bias resistor is 29R ... practically 30R (12.66W dissipation!). At least 50W resistor required, with heatsink!
Choosing 15R the dissipation only 7.2W. The resistors in #55 post are perfect, even without heatsink (only passive air flow -6..8 holes- required).
BTW, why 20V?
The assumed output power only few hundred mW, so practically any operating point (around 30mA anode current) between 10 and 20 can be used.
If you choose lower value, the filament bias dissipation will be much lower.
If you use for first stage gyrator loaded #26, as in #72 post schematic shows, even 190V B++ is enough.
+260V is totally unnecessary for power stage: wasting power, high heat production, expensive filament bias resistor, more expensive R.C. power supply.
Otherwise regulated power supply recommended for power stage (the first stage gyrator serves the #26 at any B++ voltage greater than 180V). The headphone amp generally requires clean power.
If you have higher HV raw supply that recommended, the regulator takes up the excess.
If you use it as power stage, I not recommend the filament current reduction.
-20V bias requires 18.95V (20-(2.1/2)) on filament bias resistor, so with 0.65A current the required filament bias resistor is 29R ... practically 30R (12.66W dissipation!). At least 50W resistor required, with heatsink!
Choosing 15R the dissipation only 7.2W. The resistors in #55 post are perfect, even without heatsink (only passive air flow -6..8 holes- required).
BTW, why 20V?
The assumed output power only few hundred mW, so practically any operating point (around 30mA anode current) between 10 and 20 can be used.
If you choose lower value, the filament bias dissipation will be much lower.
If you use for first stage gyrator loaded #26, as in #72 post schematic shows, even 190V B++ is enough.
+260V is totally unnecessary for power stage: wasting power, high heat production, expensive filament bias resistor, more expensive R.C. power supply.
Otherwise regulated power supply recommended for power stage (the first stage gyrator serves the #26 at any B++ voltage greater than 180V). The headphone amp generally requires clean power.
If you have higher HV raw supply that recommended, the regulator takes up the excess.
My idea it's to make the 4P1L in a separate box from the 26. What do you think?
This way I can use the 26 as preamp for the loudspeakers system.
This way I can use the 26 as preamp for the loudspeakers system.
Yes.
Be sure, that OPT tolerates anode current.
If you want to use first stage in separated box, C1-R5 (probably 470k) is will part of preamp and another 100k grid leak resistor needed for power amp.
BTW if you make "Line out" output from "Ug" and use switch to turn off power amp supplies (from +190V and R.C. transformer), the integrated amp can be used as preamplifier.
Be sure, that OPT tolerates anode current.
If you want to use first stage in separated box, C1-R5 (probably 470k) is will part of preamp and another 100k grid leak resistor needed for power amp.
BTW if you make "Line out" output from "Ug" and use switch to turn off power amp supplies (from +190V and R.C. transformer), the integrated amp can be used as preamplifier.
10.43+2.1+3.5= 16.03V raw supply output.
I would use at least 17-18V raw supply.
Do you have V9?
V4 requires more headroom than 3.5V. I usually use at least 5V headroom for V4.
If you have both versions, I would use V9 for #26 and V4 for 4P1L (there is not so important the better quality, than in preamp).
I would use at least 17-18V raw supply.
Do you have V9?
V4 requires more headroom than 3.5V. I usually use at least 5V headroom for V4.
If you have both versions, I would use V9 for #26 and V4 for 4P1L (there is not so important the better quality, than in preamp).
Thanks Bela, several questions more to be sure before make the headphone amp:
1-Filaments have to be connected parallel for 2.1V: pin 1 & 7 together and 8 filament center-tap?
2-Have I to connect 300R resistors between anode & pins 3 (2nd grid) & 4 (3rd grid)?
3-Pin 5 to GND?
I have R.C. V9 but is for 801A.
R.C. V8 is powering 26
TIA
1-Filaments have to be connected parallel for 2.1V: pin 1 & 7 together and 8 filament center-tap?
2-Have I to connect 300R resistors between anode & pins 3 (2nd grid) & 4 (3rd grid)?
3-Pin 5 to GND?
I have R.C. V9 but is for 801A.
R.C. V8 is powering 26
TIA
Just stumbled over this thread. Did you look at this?
https://www.diyaudio.com/community/threads/45-amp-build-direct-coupled.320418/post-5396135
https://www.diyaudio.com/community/threads/45-amp-build-direct-coupled.320418/post-5396135
45 is obsolete only Emission Labs still in production at 727€ the pair....the cheapest solid plate
https://www.dbtronictubes.es/index....emission-labs-matched-pair&controller=product
https://www.dbtronictubes.es/index....emission-labs-matched-pair&controller=product
Rod Coleman notes for his V4 regulator:
"Please choose R8 according to the amount of filament bias:
Examples:
R8 is 1K normal (no filament bias)
1K for filament bias of up to 7V
4.7K for 14V of filament bias (Regulator output = 16,1V, Supply input = 21,1V)
10K for 20V of filament bias (Regulator output = 22,1V, Supply input = 24,1V)
10K for 24V of filament bias (Regulator output = 26,1V, Supply input = 31,1V)
For 500mA to 650mA filament current (adjustment range) please set R1=3,3 and R2 = 3,3
All these resistors are in the kit!
Enjoy the 4P1L, excellent Russian Valve!
Rod"
Bela have I to use 4,7K for R8?
Last edited:
Yes.
The value is not extremely critical, between 4k2-4k7 still working.
"4.7K for 14V of filament bias (Regulator output = 16,1V, Supply input = 21,1V)"
As you can see Rod also recommend 5V headroom for v4.
The value is not extremely critical, between 4k2-4k7 still working.
"4.7K for 14V of filament bias (Regulator output = 16,1V, Supply input = 21,1V)"
As you can see Rod also recommend 5V headroom for v4.
4P1L is cheap within EU. Military surplus stocks sinking, but in former soviet block (mainly in BG and UA) are still quite significant. Selection (the standard deviation is quite large) is relatively easy due to the available quantity.
Access to #45 is difficult. Stocks the are long gone, can be accessed on an "extinction basis". The prices are in the sky.
If the 4P1L dead, changing it relatively simple, many of them available. Even months pass till suitable #45 can be bought somewhere.
Don't get me wrong, #45 more better, than 4P1L ... but Felipe is relatively beginner of tube collection.
We are collecting tubes for decades (when these was relatively cheap), so if hundred bucks tube gone, we replace it from the reserve. If no spare stock, the action will be painful and livelong.
Access to #45 is difficult. Stocks the are long gone, can be accessed on an "extinction basis". The prices are in the sky.
If the 4P1L dead, changing it relatively simple, many of them available. Even months pass till suitable #45 can be bought somewhere.
Don't get me wrong, #45 more better, than 4P1L ... but Felipe is relatively beginner of tube collection.
We are collecting tubes for decades (when these was relatively cheap), so if hundred bucks tube gone, we replace it from the reserve. If no spare stock, the action will be painful and livelong.