• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

This circuit ok?

CarlyBoy said:
Aside from the reliability of the 6C33C

They're terrible, either avoid 'em or use 'em for what they were intended: series pass regulator.

I have some particular questions, but if the circuit isn't sound (sorry) I won't bother.

How low powered? You can definitely use 6AS7s instead. How many, that depends on both Po and Zl.

Cirucit looks OK except that the DC offset pot would be better in the cathode circuit, as opposed to the plate. The cathode has more control than the plate, and you would neither need such a big pot, nor expose yourself to dangerous voltages.

I would also like to see a DC offset servo on the finals. It doesn't take much DC offset to poof your speeks.

Another thing: this must run from a Lo-Z source, otherwise, your gNFB is going to become badly imbalanced, and that won't help the sonics a bit.
 

kevinkr

Administrator
Paid Member
I designed and built a somewhat more complex circlotron based on 6C33 about 10yrs ago. The key thing with 6C33 is to burn them in for about 24 hours at moderate plate current to stabilize their characteristics before use. The filaments should be allowed to warm up fully before B+ is applied, this will go a long way to preventing arcing and other problems.

There are a lot of people who use 6C33 successfully here on diyaudio and would vehemently disagree with Miles although my experience with them should make me very sympathetic to the viewpoint expressed.

Selection and burn in is one key to success with the 6C33, another is to make sure that the filament current does not cause contact failure.
The cheap Chinese sockets won't cut it, some surplus Russian mil-spec stuff works fine. One common approach is to use the European style terminal blocks removed from their plastic bodies to make strong physical connections to the filament pins - obviously if you do this you won't be able to just unplug them, but you won't have fireworks from failing filament pin contacts either. (I used forced air cooling around the sockets which kept them cool enough to avoid contact problems, and used good sockets. The chassis was basically a sealed plenum pressurized by a fan which blew air out around the sockets which were slightly recessed - this worked very well - this monoblock amplifier had 6 x 6C33 per channel.)

The 6C33 can sound very good indeed, I was unsuccessful partly because I was distracted by the allure of SE toward the end of the project. I gave the amps away and the fellow who inherited them has been able to get them to work.

Check out my site for the sordid tale of the 6C33 otl..
 

kevinkr

Administrator
Paid Member
Don't assume the 6AS7 is all that great either, just a different set of problems.

Many USA and Canadian made 6AS7G do not perform well or last long in OTL service. The current hot tip is to use Russian surplus. (I have some and they seem more consistent in section to section match than the US ones I had.)

The 6C33 is a practical choice if you are careful, a number of Dutch and German members of this forum are running circlotrons with these and have reported very good long term reliability. (up to 6yrs on the same set of outputs.)

Search here for otl, 6C33, etc. Make up your own mind.. Other potential choices are 6336, some tv sweep tubes are suitable, (ask George of tubelab fame) etc.

I ran mine with no feedback and it sounded very good driving 4 ohm maggies despite the relatively high output impedance. I didn't find dc drift on the output to be a problem - my issue was with arcing during warm up, excessive hum, and a very inadequate driver design. (No spice in those days.. )
 
CarlyBoy said:
Miles, how low powered you ask? Say 30W.

I've no experience with the 6C33C. Have done enough reading to be tentative, but also trying to keep an open mind. I'd rather stay with the 6AS7...

I've done some very preliminary design work on this since I'd like to do a OTL some day. 6AS7s are a good choice, and you can get 30W with eight sections per phase, and stick to the conservative side of the ratings. Here's a loadline for this.
 

Attachments

  • 6as7ga-loadline.png
    6as7ga-loadline.png
    77.4 KB · Views: 243

kevinkr

Administrator
Paid Member
I should think with the Quad 57 you wouldn't need more than 15 - 20 watts total.. ( A single pair of 6C33 will do that for you.) I apparently have a bee in my bonnet tonight... Probably should just ignore me.. :devilr:

Confession, I think I am going to revisit the 6C33 issue in a more modest form at some point. Maybe a pp cathode follower amp... ;)