The 300b has it's own floating power supply ,hooked at a biasing potantial and anchored to the ground by a nonpolerized 2uf capacitor .
(The filament and its ac supply are simplified as cathode on the circuit, for clarity) .
Fixed bias is simple but it requires a grid circuit resistance less than 50kohms.To drive directly ,an ecc83 triode alone is not enough . I tried numerous known low output impedance circuits ,and I came up with this one . It is a cathode biased , but instead of bypassing with a capacitor I use a positive feedback current from the cathode follower to cancel all ac components on R5.
The 300b running at 350v/80ma has grid biased -70v. If loaded with habitual 2.5k ,the driver must provide a swing of +70v/-100v to have the plate swing symmetrically. The 310a ,the dedicated driver has this possibility ,but the ecc83 can deliver at best +70v/-90v. So what impedance loading can give best match? I tried betweeen 3k to 14k loads adjusting at best in each case , two outstanding loadings came out. The 7k and 3.5k. With 7kohms loading, a woofer (sn12b Paudio) started singing as a fullrange ,once the driver adjusted as shown , it delivered a clean white celestial sound. The high damping factor gives clean impulses at lows and high presence at low mids. If this loading suits best for high dynamic musics as classic or jazz , The 3.5kohms loading gave richer sound for low dynamic ones, as rock or pop .The adjustments for 3.5k needs 390k resistors instead of 330k ones and the output voltage is 240 volts instead of 180 adjusted by R5. Of course the bias of 300b becomes 310v.
I wanted both ,but one driver is to choose. I opted for 7k for its best and used an ultra linear output transformer Thai SAC silk S-660 (20W for SE 6.6kohms). Luckily an 8 ohms at it's secondary reflected 7K on the primary and 3.2k if it is powered at UL tap.
I always use doubler power supply along semiconductor diodes. see the thread
http://www.diyaudio.com/forums/power-supplies/291240-rectifier-tube-semiconductor.html
Start up problems :
If nothing is done at start up the 300b will draw 200ma for 5 seconds. To remedy this problem , a relay is activated for 4 seconds at start up that cuts the filament supply of the cathode follower .Now the 300b shaws 200ma to fall back immediately to 90ma then gradually to 80ma , allowing it's aging to be mesured .
This amplifier is singing heavenly since 4 years.
KOKORIANTZ
(The filament and its ac supply are simplified as cathode on the circuit, for clarity) .
Fixed bias is simple but it requires a grid circuit resistance less than 50kohms.To drive directly ,an ecc83 triode alone is not enough . I tried numerous known low output impedance circuits ,and I came up with this one . It is a cathode biased , but instead of bypassing with a capacitor I use a positive feedback current from the cathode follower to cancel all ac components on R5.
The 300b running at 350v/80ma has grid biased -70v. If loaded with habitual 2.5k ,the driver must provide a swing of +70v/-100v to have the plate swing symmetrically. The 310a ,the dedicated driver has this possibility ,but the ecc83 can deliver at best +70v/-90v. So what impedance loading can give best match? I tried betweeen 3k to 14k loads adjusting at best in each case , two outstanding loadings came out. The 7k and 3.5k. With 7kohms loading, a woofer (sn12b Paudio) started singing as a fullrange ,once the driver adjusted as shown , it delivered a clean white celestial sound. The high damping factor gives clean impulses at lows and high presence at low mids. If this loading suits best for high dynamic musics as classic or jazz , The 3.5kohms loading gave richer sound for low dynamic ones, as rock or pop .The adjustments for 3.5k needs 390k resistors instead of 330k ones and the output voltage is 240 volts instead of 180 adjusted by R5. Of course the bias of 300b becomes 310v.
I wanted both ,but one driver is to choose. I opted for 7k for its best and used an ultra linear output transformer Thai SAC silk S-660 (20W for SE 6.6kohms). Luckily an 8 ohms at it's secondary reflected 7K on the primary and 3.2k if it is powered at UL tap.
I always use doubler power supply along semiconductor diodes. see the thread
http://www.diyaudio.com/forums/power-supplies/291240-rectifier-tube-semiconductor.html
Start up problems :
If nothing is done at start up the 300b will draw 200ma for 5 seconds. To remedy this problem , a relay is activated for 4 seconds at start up that cuts the filament supply of the cathode follower .Now the 300b shaws 200ma to fall back immediately to 90ma then gradually to 80ma , allowing it's aging to be mesured .
This amplifier is singing heavenly since 4 years.
KOKORIANTZ
