POWERTRON
Hi, I'm new to the group but couldn't resist responding to mention of the POWERTRON amplifier design. I built a version a while back and found it quite exceptional regarding transient response (excellent!) but had stability problems with the biasing of the first two stages(DC coupled). Drifting was a problem even over an extended operating period. Certainly a unique design. Has anyone else built one? Would be pleased to hear of any ideas regarding the drifting problem.
Hi, I'm new to the group but couldn't resist responding to mention of the POWERTRON amplifier design. I built a version a while back and found it quite exceptional regarding transient response (excellent!) but had stability problems with the biasing of the first two stages(DC coupled). Drifting was a problem even over an extended operating period. Certainly a unique design. Has anyone else built one? Would be pleased to hear of any ideas regarding the drifting problem.
My experience with the Menno Vandervenne VDV70/100
Pedrihno said:
"I was rather thinking about Menno's barebone design (chapter 11 of Menno's book) with EL34's (30 to 100 W), or some improvement of it, such as the project of Rudy Godmaire (http://www3.sympatico.ca/r.godmaire/), because it gives some freedom of choice on the valves, and the type of feedback (if any) and phase splitting. Or this, or a McIntosh-like design, as I said in my former thread."
I have built a pair of these as monoblocks:
They are NOT the unity coupled design but rather a very simple push pull design using a 12AU7 front end / phase splitter of a common cathode amplifier stage with a direct coupled cathodyne ("split load") phase splitter driver and parallel push pull output tubes.
I auditioned various EL34s and settled on JJ Electronics as being slightly better than Winged C Svetlana. I tried it it Ultralinear mode and Triode Mode. Ultralinear Mode has too high an output impedance to be acceptable and even triode mode is a little higher than I would like BUT gave very good sound with my nominally 6 Ohm speakers.
The best improvement I achieved was simply by substituting a JJ Electronics ECC99 for the 12AU7. I also got significant improvement and output impedance reduction by substituting 6550 biased at 75mA per tube in place of the EL34s (KT88 would also work well but physical space limitations meant that I could not fit them). Because of my cramped mechanical layout I also had temperature problems with the 6550 and am back running EL34s again. I placed a temperature probe against the output transformer at the point closest to the output tubes. With EL34s at 47mA bias per tube the output transformer case stabilised at 62 degrees C after 4 hours use. With the 6550s at 75mA bias per tube the temperature at the same point stabilised at 80 degrees C - just too high for my satisfaction. This temperature is due to the proximity of the output tubes NOT power dissipation in the transformer itself and so with a better layout (larger chassis to separate the output transformer from the output tubes by a larger distance) the 6550s or KT88 would be OK.
The fact that I could not use the 6550 was a disappointment because the cathodyne phase splitter is wonderful circuit - so long as the anode and cathode loads are the same - that means so long as the output stage does not transition from class A to class B (tube not driven to cut off). The 6550 at 75mA meant that it very rarely moved out of Class A so not only was the output improved but better performance was achieved from the phase splitter.
Despite the simplicity of the design it really does give a stunning sound and I use these "Menno Bricks" in preference to my 845 Single Ended Triode Amplifier. With the EL34s in Triode Mode I get around 46 Watts per channel (65W with the 6550s) with -3dB point at 10 Watts of over 200kHz.
I am considering rebuilding these on a larger chassis to enable me to use JJ KT88s OR rebuilding the front end / phase splitter as a ECC88 (or 6H30pi) cascode differential amp with a 6SN7 cathode follower on each of the push pull sides to allow say 5dB of global feedback. Less than 6dB of global feedback is generally OK. More than this temnds to close down the sound stage (the stereo image) although the amazing high frequency response of the Toroidal Output transformers MAY allow use of a little more feedback without the stereo imaging compromise (yet to be checked).
Cheers,
Ian
Pedrihno said:
"I was rather thinking about Menno's barebone design (chapter 11 of Menno's book) with EL34's (30 to 100 W), or some improvement of it, such as the project of Rudy Godmaire (http://www3.sympatico.ca/r.godmaire/), because it gives some freedom of choice on the valves, and the type of feedback (if any) and phase splitting. Or this, or a McIntosh-like design, as I said in my former thread."
I have built a pair of these as monoblocks:
They are NOT the unity coupled design but rather a very simple push pull design using a 12AU7 front end / phase splitter of a common cathode amplifier stage with a direct coupled cathodyne ("split load") phase splitter driver and parallel push pull output tubes.
I auditioned various EL34s and settled on JJ Electronics as being slightly better than Winged C Svetlana. I tried it it Ultralinear mode and Triode Mode. Ultralinear Mode has too high an output impedance to be acceptable and even triode mode is a little higher than I would like BUT gave very good sound with my nominally 6 Ohm speakers.
The best improvement I achieved was simply by substituting a JJ Electronics ECC99 for the 12AU7. I also got significant improvement and output impedance reduction by substituting 6550 biased at 75mA per tube in place of the EL34s (KT88 would also work well but physical space limitations meant that I could not fit them). Because of my cramped mechanical layout I also had temperature problems with the 6550 and am back running EL34s again. I placed a temperature probe against the output transformer at the point closest to the output tubes. With EL34s at 47mA bias per tube the output transformer case stabilised at 62 degrees C after 4 hours use. With the 6550s at 75mA bias per tube the temperature at the same point stabilised at 80 degrees C - just too high for my satisfaction. This temperature is due to the proximity of the output tubes NOT power dissipation in the transformer itself and so with a better layout (larger chassis to separate the output transformer from the output tubes by a larger distance) the 6550s or KT88 would be OK.
The fact that I could not use the 6550 was a disappointment because the cathodyne phase splitter is wonderful circuit - so long as the anode and cathode loads are the same - that means so long as the output stage does not transition from class A to class B (tube not driven to cut off). The 6550 at 75mA meant that it very rarely moved out of Class A so not only was the output improved but better performance was achieved from the phase splitter.
Despite the simplicity of the design it really does give a stunning sound and I use these "Menno Bricks" in preference to my 845 Single Ended Triode Amplifier. With the EL34s in Triode Mode I get around 46 Watts per channel (65W with the 6550s) with -3dB point at 10 Watts of over 200kHz.
I am considering rebuilding these on a larger chassis to enable me to use JJ KT88s OR rebuilding the front end / phase splitter as a ECC88 (or 6H30pi) cascode differential amp with a 6SN7 cathode follower on each of the push pull sides to allow say 5dB of global feedback. Less than 6dB of global feedback is generally OK. More than this temnds to close down the sound stage (the stereo image) although the amazing high frequency response of the Toroidal Output transformers MAY allow use of a little more feedback without the stereo imaging compromise (yet to be checked).
Cheers,
Ian
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.