Some time ago I prototyped a 6C33 PP amp.
The design is unusual in that the output plates provide local negative feedback to the screens of the LPT driver stage. With the right choice of pentode here, the triode non-linearity can be largely be corrected. In the end I went with Smoking-amp suggestion of using a pair of 6197’s as drivers which can cope with the output swing needed. For a totally valve solution the cathodes could be returned to the negative rail with a resistor.
The input stage is hum-buck allowing the input screen cable to float slightly avoiding hum loops with external equipment. The 12ax7 heaters are run off DC and the stage (in the corner) and input wiring kept away from the mains transformers and heater magnetic fields.
The schematics are for simulation rather than containing everything.
Servo bias is used for the output tubes using a PCB (about 90ma per tube). There’s a HT delay which adds some resistance on the HT transformer AC output for a few seconds. The warm-up is unusual as the LTP stage initially drives the OPT through the screens before the main tubes kick in.
I’ve now finished the stereo version, and are pleased with the sound. The wiring is an interesting combination of strip-board and tag-board.
The HT supplies and heaters are easily accommodated with a couple of off the shelf transformers.
250VA toroidal for the heaters wired for 12.6V and 300VA 230V-230V giving 300V HT. The negative rail is derived from this with a AC voltage inverter as its only a few milliamps.
6C33’s need a large signal amplitude to drive fully (grid needs to get to 0V). I thought running the driver off the same supply as the output would do. This was OK for some tubes but not others. In the end I added a small boost supply (a small toroidal mains + bridge rectifier and cap) to add 60V to the driver stage giving 360V. Not ideal.
I ended buying 6 output tubes as they are not matched and quite all over the place, as they are not intended for audio use. The ceramic tube sockets don’t hold the tubes that well, and you need to check they are fully home after moving the unit. They could do with a clamp over the top.
Distortion is about 0.3% @ 10W 1KHz mainly 3rd harmonic.
Output power is 100W rms per channel.
Noise and hum are not audible. Damping factor is very good.
Anyway I think its a nice use of some 6C33B and definitely worth a go.
The design is unusual in that the output plates provide local negative feedback to the screens of the LPT driver stage. With the right choice of pentode here, the triode non-linearity can be largely be corrected. In the end I went with Smoking-amp suggestion of using a pair of 6197’s as drivers which can cope with the output swing needed. For a totally valve solution the cathodes could be returned to the negative rail with a resistor.
The input stage is hum-buck allowing the input screen cable to float slightly avoiding hum loops with external equipment. The 12ax7 heaters are run off DC and the stage (in the corner) and input wiring kept away from the mains transformers and heater magnetic fields.
The schematics are for simulation rather than containing everything.
Servo bias is used for the output tubes using a PCB (about 90ma per tube). There’s a HT delay which adds some resistance on the HT transformer AC output for a few seconds. The warm-up is unusual as the LTP stage initially drives the OPT through the screens before the main tubes kick in.
I’ve now finished the stereo version, and are pleased with the sound. The wiring is an interesting combination of strip-board and tag-board.
The HT supplies and heaters are easily accommodated with a couple of off the shelf transformers.
250VA toroidal for the heaters wired for 12.6V and 300VA 230V-230V giving 300V HT. The negative rail is derived from this with a AC voltage inverter as its only a few milliamps.
6C33’s need a large signal amplitude to drive fully (grid needs to get to 0V). I thought running the driver off the same supply as the output would do. This was OK for some tubes but not others. In the end I added a small boost supply (a small toroidal mains + bridge rectifier and cap) to add 60V to the driver stage giving 360V. Not ideal.
I ended buying 6 output tubes as they are not matched and quite all over the place, as they are not intended for audio use. The ceramic tube sockets don’t hold the tubes that well, and you need to check they are fully home after moving the unit. They could do with a clamp over the top.
Distortion is about 0.3% @ 10W 1KHz mainly 3rd harmonic.
Output power is 100W rms per channel.
Noise and hum are not audible. Damping factor is very good.
Anyway I think its a nice use of some 6C33B and definitely worth a go.
