The Rosenblit 6AS7 OTL
- Thermionic Idler
- Old Hand
- Posts: 1018
- Joined: Sun Apr 22, 2012 8:22 pm
- Location: Southsea
#46 Re: The Rosenblit 6AS7 OTL
Bringing this one back up to the top again after noting that my "Son of Beast" seemed to go down quite well at Owston - and also chatting with Paul Barker about the 6AS7 version and how good that sounded, piqued my interest.
I've been studying the circuit in 'Audio Reality' this morning versus the Glass Audio and it does look like hybridising the design to use the patented driver stage should be reasonably straightforward (with octal tubes of course). Modelling in Spice should help with this.
From my perspective, I've had two of the designs that use the patent - the T16 and my current SOB - and both have worked very reliably over the years so I would be loath to alter this aspect. ***Edit - though I have just found a Tubecad article suggesting an improvement: http://www.tubecad.com/april_may2001/page33.html
I was pondering this morning how to make it work with long balanced lines from the preamp, given there's a global feedback look from the output to the cathode of the input valve. The obvious way seems a bit inelegant - 'unbalance' the input with a transformer then split the phase again a stage later.
What about this: input + and - to a balanced pair of input valves. Do away with the phase splitter and either directly connect to the patent driver / buffer stage, or use a pair of cathode followers depending on what the Miller is like. Both the input valves will need global feedback input to their cathodes, but of course, one of the feedback lines has to be out of phase.
So here's my cunning plan - insert a high bandwidth opamp into the feedback loop just before the input valves, to split the phase. The opamp based pre that I have planned passes a perfect squarewave at 100kHz and has bandwith to 2Mhz with stupidly low distortion levels, so it ought to be fast enough I'd have thought?
Edit - might be better to connect the feedback loops to the cathodes of the buffer valves in the driver stage maybe? that would be simpler and would mean the opamp wouldn't be needed. But that would mean the feedback would no longer be 'global' and I'm unsure without doing more homework whether there would be a benefit in reduction of output impedance...
I've been studying the circuit in 'Audio Reality' this morning versus the Glass Audio and it does look like hybridising the design to use the patented driver stage should be reasonably straightforward (with octal tubes of course). Modelling in Spice should help with this.
From my perspective, I've had two of the designs that use the patent - the T16 and my current SOB - and both have worked very reliably over the years so I would be loath to alter this aspect. ***Edit - though I have just found a Tubecad article suggesting an improvement: http://www.tubecad.com/april_may2001/page33.html
I was pondering this morning how to make it work with long balanced lines from the preamp, given there's a global feedback look from the output to the cathode of the input valve. The obvious way seems a bit inelegant - 'unbalance' the input with a transformer then split the phase again a stage later.
What about this: input + and - to a balanced pair of input valves. Do away with the phase splitter and either directly connect to the patent driver / buffer stage, or use a pair of cathode followers depending on what the Miller is like. Both the input valves will need global feedback input to their cathodes, but of course, one of the feedback lines has to be out of phase.
So here's my cunning plan - insert a high bandwidth opamp into the feedback loop just before the input valves, to split the phase. The opamp based pre that I have planned passes a perfect squarewave at 100kHz and has bandwith to 2Mhz with stupidly low distortion levels, so it ought to be fast enough I'd have thought?
Edit - might be better to connect the feedback loops to the cathodes of the buffer valves in the driver stage maybe? that would be simpler and would mean the opamp wouldn't be needed. But that would mean the feedback would no longer be 'global' and I'm unsure without doing more homework whether there would be a benefit in reduction of output impedance...
Deck: Garrard 301 - Audio Origami PU7 - NW Analogue DH3S
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
- Paul Barker
- Social Sevices have been notified
- Posts: 8864
- Joined: Mon May 21, 2007 9:42 pm
#47 Re: The Rosenblit 6AS7 OTL
U lost me early on there.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
- Thermionic Idler
- Old Hand
- Posts: 1018
- Joined: Sun Apr 22, 2012 8:22 pm
- Location: Southsea
#48 Re: The Rosenblit 6AS7 OTL
When I get a chance I'll try and draw out what I mean in Spice, but I'll obviously need to be careful not to infringe any patents.
Deck: Garrard 301 - Audio Origami PU7 - NW Analogue DH3S
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
- andrew Ivimey
- Social Sevices have been notified
- Posts: 8307
- Joined: Mon Jun 11, 2007 8:33 am
- Location: Bedford
#49 Re: The Rosenblit 6AS7 OTL
That's got me thinking too.. I brought my 6AS7 Otl to Owsters when it was a little village pub up the A1. Phase splitting was an ECC81 as was the input valve. I can't 'member what feedback. The amp still works fabby well. It is a Rosenblit/bolt design I just can't remember much but I still have the design notes somewhere....
Philosophers have only interpreted the world - the point, however, is to change it. No it isn't ... maybe we should leave it alone for a while.
- Thermionic Idler
- Old Hand
- Posts: 1018
- Joined: Sun Apr 22, 2012 8:22 pm
- Location: Southsea
#50 Re: The Rosenblit 6AS7 OTL
Ok so having done a bit more reading and thinking, realised I don't need the op amp! Just apply the feedback to the cathode of the in-phase input valve.
And in fact the phase splitter valve can be kept as is, just turn it into a differential stage by applying the reverse and in phase signals to the two grids respectively.
I'll draw out what I mean later.
Edit: that would allow the tubecad mod to be applied too as it relies on the "phase splitter" being there.
And in fact the phase splitter valve can be kept as is, just turn it into a differential stage by applying the reverse and in phase signals to the two grids respectively.
I'll draw out what I mean later.
Edit: that would allow the tubecad mod to be applied too as it relies on the "phase splitter" being there.
Deck: Garrard 301 - Audio Origami PU7 - NW Analogue DH3S
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
- Thermionic Idler
- Old Hand
- Posts: 1018
- Joined: Sun Apr 22, 2012 8:22 pm
- Location: Southsea
#51 Re: The Rosenblit 6AS7 OTL
Excellent!andrew Ivimey wrote:That's got me thinking too.. I brought my 6AS7 Otl to Owsters when it was a little village pub up the A1. Phase splitting was an ECC81 as was the input valve. I can't 'member what feedback. The amp still works fabby well. It is a Rosenblit/bolt design I just can't remember much but I still have the design notes somewhere....
My plan would be to replace the input valve with a 6SF5GT - I have a stash of 23 of them which I got when I had the Fi 2a3 amps. Electrically same as 12AX7 I believe. And sub a 6SN7 for the 12AU7. That way -- all octal.
Deck: Garrard 301 - Audio Origami PU7 - NW Analogue DH3S
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
- andrew Ivimey
- Social Sevices have been notified
- Posts: 8307
- Joined: Mon Jun 11, 2007 8:33 am
- Location: Bedford
#52 Re: The Rosenblit 6AS7 OTL
I deliberately made an amp so I could swap ecc83 with 6sl7 and was delighted to find that the 6sl7 sounded better....
Philosophers have only interpreted the world - the point, however, is to change it. No it isn't ... maybe we should leave it alone for a while.
- Thermionic Idler
- Old Hand
- Posts: 1018
- Joined: Sun Apr 22, 2012 8:22 pm
- Location: Southsea
#53 Re: The Rosenblit 6AS7 OTL
Started drawing this out last night and realised I still haven't got this right - bear with me...
Deck: Garrard 301 - Audio Origami PU7 - NW Analogue DH3S
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
- Thermionic Idler
- Old Hand
- Posts: 1018
- Joined: Sun Apr 22, 2012 8:22 pm
- Location: Southsea
#54 Re: The Rosenblit 6AS7 OTL
Hmm - I don't think the Futterman-based designs can be made balanced in an elegant way - because quite a lot of the circuit is effectively single-ended. You've got that totem-pole output stage that needs the asymmetrical drive to ensure the gains balance properly, and how do you apply global feedback in one phase, to a dual phase input signal without unbalancing that too. It's beyond my ken.
Meanwhile, the Atma-sphere circlotron OTL circuits are rather elegantly balanced throughout, and available online. Lots of heat rejection though - over half a kilowatt for the M60's. Could reduce the number of output valves to reduce the heat but then the output impedance goes up. Though I did read an interesting article in Tubecad which suggested making speakers by hooking multiple drive units in series - so you'd get say a 32 ohm speaker instead of 8 ohm. That would be one possible solution to the output impedance issue.
Meanwhile, the Atma-sphere circlotron OTL circuits are rather elegantly balanced throughout, and available online. Lots of heat rejection though - over half a kilowatt for the M60's. Could reduce the number of output valves to reduce the heat but then the output impedance goes up. Though I did read an interesting article in Tubecad which suggested making speakers by hooking multiple drive units in series - so you'd get say a 32 ohm speaker instead of 8 ohm. That would be one possible solution to the output impedance issue.
Deck: Garrard 301 - Audio Origami PU7 - NW Analogue DH3S
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets
Phono: Pete Millett LR, Lundahl SUT, AMB σ22 PSU
Linestage: Bruno Putzeys Balanced Pre + Hans Polak relay mod, LCDuino control
Power amps: 300B PP 'Symmetric Reichert'
Speakers: FF225WK Big Mets