You would think this would be easy to find but hours of web searching has not turned up anything substancial.
I was planning to build a Lux MQ-36 6336A OTL amp clone. But after doing some initial research decided I didn't like the inherently unsymetrical nature of the futterman type OTL circuit.
So I now intend to build a basic Circlotron type OTL power amp using 2 6336As per channel.
I know all the drawbacks re parts and heat etc and don't mind.
CAN ANYONE HELP ME WITH A SCHEMATIC FOR A COMPLETE circlotron otl AMP USING 6336As?
John Corneille
PS I want to keep it triode all the way....preferably with 12AX7, 12AU7 inputs and parralled 12BH7A drivers....but anything can be modified I guess.
I was planning to build a Lux MQ-36 6336A OTL amp clone. But after doing some initial research decided I didn't like the inherently unsymetrical nature of the futterman type OTL circuit.
So I now intend to build a basic Circlotron type OTL power amp using 2 6336As per channel.
I know all the drawbacks re parts and heat etc and don't mind.
CAN ANYONE HELP ME WITH A SCHEMATIC FOR A COMPLETE circlotron otl AMP USING 6336As?
John Corneille
PS I want to keep it triode all the way....preferably with 12AX7, 12AU7 inputs and parralled 12BH7A drivers....but anything can be modified I guess.
Hi,
I can't help you with a circuit diagram of a circlotron although there are many available on the net that could be easily adapted to 6336s, example here http://members.aol.com/aria3/index.html
Can you explain what you mean by "the inherently unsymetrical nature of the Futterman type OTL circuit"?
Please do not compare the Futterman circuit to an ordinary SEPP, (Single Ended Push Pull) output stage. In a SEPP the upper tube works as a cathode follower and the lower tube as an ordinary grounded cathode amplifier and obviously this circuit is unsymmetrical. In a Futterman this is not so, in an original Futterman stage both tubes has the same output impedance as grounded cathode and the stage is symmetrical. A better variant is the so called inverted Futterman where both tubes works as cathode followers.
As an example how symmetrical a Futterman circuit is we can look at the amount of 2nd order distortion, in my OTL I have measured a 2nd order distortion of -55dB before applied feedback, this is a very good value and can not be much better even by a Circlotron circuit, (in my amp most of the 2nd order distortion comes from the driver stage and not the output stage which is even more linear).
My conclusion is that the most common argument for using a Circlotron in favor over a Futterman is not valid, both are symmetrical and gives low 2nd order distortion.
Refards Hans
I can't help you with a circuit diagram of a circlotron although there are many available on the net that could be easily adapted to 6336s, example here http://members.aol.com/aria3/index.html
Can you explain what you mean by "the inherently unsymetrical nature of the Futterman type OTL circuit"?
Please do not compare the Futterman circuit to an ordinary SEPP, (Single Ended Push Pull) output stage. In a SEPP the upper tube works as a cathode follower and the lower tube as an ordinary grounded cathode amplifier and obviously this circuit is unsymmetrical. In a Futterman this is not so, in an original Futterman stage both tubes has the same output impedance as grounded cathode and the stage is symmetrical. A better variant is the so called inverted Futterman where both tubes works as cathode followers.
As an example how symmetrical a Futterman circuit is we can look at the amount of 2nd order distortion, in my OTL I have measured a 2nd order distortion of -55dB before applied feedback, this is a very good value and can not be much better even by a Circlotron circuit, (in my amp most of the 2nd order distortion comes from the driver stage and not the output stage which is even more linear).
My conclusion is that the most common argument for using a Circlotron in favor over a Futterman is not valid, both are symmetrical and gives low 2nd order distortion.
Refards Hans
icarex69,
Andrea Cufioli has a great site for OTL:
http://www.bonavolta.ch/hobby/en/audio/6c33c.htm
Also:
http://www.tubetvr.com/otl.html
http://www.wdehaan.demon.nl/tubeamps/6336/
http://www.tubecad.com/july99/page7.html
Regards.
Alain.
Andrea Cufioli has a great site for OTL:
http://www.bonavolta.ch/hobby/en/audio/6c33c.htm
Also:
http://www.tubetvr.com/otl.html
http://www.wdehaan.demon.nl/tubeamps/6336/
http://www.tubecad.com/july99/page7.html
Regards.
Alain.
Thats OK, I just wanted to make sure that you made your choice based on facts rather than rumours and misunderstandings spread on the internet, (and even by some OTL vendors).
For choice of circuit I wouldn't be so fixed if it was originally intended for a 6336 or not, the important thing is that the circuit is sound and competently designed, the example I gave as a link seems to be one of the better ones around. If you want to use 6336 instead of 6C33C as it is designed for originally I would use 2 6336 for each 6C33C, that would get you close enough to get reasonable tube life.
Regards Hans
Here are two related pieces from the Wayback Machine that seem close to what you're looking for:
Scott Dorsey 6C33 Circlotron
Commentary by Ralph Karsten
Both articles were written by actual builders. The second piece suggests improvements to the first. Good luck and be sure to report your progress!
Scott Dorsey 6C33 Circlotron
Commentary by Ralph Karsten
Both articles were written by actual builders. The second piece suggests improvements to the first. Good luck and be sure to report your progress!
Before you build your circlotron amplifier you might want to read the article I wrote about 10 yrs ago detailing my experience doing the same. You may glean some useful information and there is a lot of information about the pitfalls I encountered.
The design in question was overly ambitious - I admit it. I was younger and more foolish..
And not successful! 
It was based on 6 x 6C33 in a circlotron configuration.
Some of the conclusions I reached:
Delay B+ to output stage until fully warmed up.
Bias needs to be independently adjustable for each bank of output tubes, if a small number of tubes are used, individual bias pots are preferable.
Output tubes must be very closely matched to assure good current sharing.
Use a small cathode resistor (1 ohm or less) on each output tube (section in case of dual triode) to allow you to measure and set quiescent current.
Fuse one supply rail and the speaker output. (Provides better VC protection in the event of a short.)
Size fuse for speaker output for average power, and no larger than needed not to blow under max spl conditions with your speaker.
Run your output tubes well within their rated dissipation, you aren't going to be able to run this thing in class A at meaningful output currents anyway.
You may not need global feedback depending on speakers. (Mine sounded far better without.)
Forget 4 ohm speakers, and seriously consider vintage 16 ohm drivers - output power will be much higher particularly with the small number of tubes you are planning.
Most designs for 6AS7/6080 will work just fine with about half the number of 6336..
Here's the link to my site:
http://www.kta-hifi.net/projects/amp_page/6c33amp/6c33.html
Note that the sound quality I got out of this thing when it was working was excellent. I over elaborated the design, and never got it to fulfill its promise. Shortly after this I discovered the joys of SE amplifiers and have since focused my time and effort in that direction.
Have Fun!
The design in question was overly ambitious - I admit it. I was younger and more foolish..
And not successful! It was based on 6 x 6C33 in a circlotron configuration.
Some of the conclusions I reached:
Delay B+ to output stage until fully warmed up.
Bias needs to be independently adjustable for each bank of output tubes, if a small number of tubes are used, individual bias pots are preferable.
Output tubes must be very closely matched to assure good current sharing.
Use a small cathode resistor (1 ohm or less) on each output tube (section in case of dual triode) to allow you to measure and set quiescent current.
Fuse one supply rail and the speaker output. (Provides better VC protection in the event of a short.)
Size fuse for speaker output for average power, and no larger than needed not to blow under max spl conditions with your speaker.
Run your output tubes well within their rated dissipation, you aren't going to be able to run this thing in class A at meaningful output currents anyway.
You may not need global feedback depending on speakers. (Mine sounded far better without.)
Forget 4 ohm speakers, and seriously consider vintage 16 ohm drivers - output power will be much higher particularly with the small number of tubes you are planning.
Most designs for 6AS7/6080 will work just fine with about half the number of 6336..
Here's the link to my site:
http://www.kta-hifi.net/projects/amp_page/6c33amp/6c33.html
Note that the sound quality I got out of this thing when it was working was excellent. I over elaborated the design, and never got it to fulfill its promise. Shortly after this I discovered the joys of SE amplifiers and have since focused my time and effort in that direction.
Have Fun!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.