Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: Contributors :: Advertise


It is currently Mar Tue 26, 2019 7:27 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 8 posts ] 

Which output tubes would you choose?
6BG6Gs (or 1614s, 807s, 5881s etc) for 30 watts  100%  [ 2 ]
6V6s for 20 watts  0%  [ 0 ]
Total votes : 2
Author Message
 Post subject: Preparing for a Circlotron amplifier build - 6V6 vs 6BG6G?
PostPosted: Mar Tue 05, 2019 8:07 pm 
Member

Joined: Apr Tue 17, 2012 7:58 pm
Posts: 60
Location: Metamora MI, 48455
Hello all,

I'm preparing for a Circlotron amplifier build as part of a "period correct" monophonic Hi-Fi setup. My driver of choice is the General Electric 1201-A, rated at 25 watts into 8 ohms.

Anyway, I'm on the fence about which output tube to use: 6V6s for 20 watts output, or 6BG6Gs for 30 watts output. Electro-Voice built Circlotrons using both tubes back between 1954 and 1956-1958 or so. The topology never really caught on due to the complexities of building two floating power supplies, but the output stage itself is more or less mathematically equivalent to a McIntosh using the same tubes at the same B+ (and therefore screen) voltages, and same load. Both exhibit 50% cathode feedback, both require identical drive voltages, both stages will bias similarly, etc. Most importantly, both offer exceptionally low output impedance, low distortion, while retaining the high efficiency of pentodes and allowing for Class-B operation without any considerable cross-over distortion.

The Circlotron is also a hell of a lot easier to implement when compared to winding the "Unity Coupled" output transformer the McIntosh requires.

The 6V6s require a 2 kohm impedance transformer, and -40 volts bias, with 420-430 volts on the plates. This is some serious spec busting, and I would worry about shorting and arcing in the outputs. Plate and screen dissipation actually fall well within limits because the outputs are biased so cold. They'll produce 20 RMS watts with ~1% to ~3% THD (only considering output stage and bootstrapped driver). The audiophiles are driving up the prices of decent 6V6GTs, and I don't trust the Russian and Chinese junk.

The 6BG6Gs require a 1000 to 1600 ohm impedance transformer, -50 volts bias, with 435 volts or so on the plates. This is also spec busting, technically, but being horizontal output tubes, and essentially 807s on a octal bases, I know they can withstand some degree of abuse from personal experience. Plate and screen dissipation are again within limits. In this configuration, the 6BG6Gs can produce 30 watts with ~1% to ~3% THD (only considering output stage and bootstrapped driver). Being predominantly a TV collector, I also have a supply of new old stock 6BG6Gs to last me my natural life and then some.

In favor of the 6V6s is the ability to use 6X4 rectifiers for the floating power supplies, simplifying the custom power transformer a bit. Only a single 6.3 volt filament winding would be needed. Power consumption, and heat generated by the amp, would also be slightly reduced. The con here is the potentially poor reliability of the output tubes.

In favor of the 6BG6Gs is the extra five watts of audio power, and potentially greater reliability. The cons here are the plate caps with 435 volts on them, which could be a shock hazard, the need for two additional 5 volt filament windings for a pair of 5Y3s, and additional power consumption and a hotter amplifier.

I'm not sure if the 10 watt difference in RMS power is really even that great of a disadvantage for the 6V6s or advantage for the 6BG6Gs. That translates to only a 1.8 dB difference in peak SPL, regardless of loudspeaker. A 1 dB difference is already on the limit of audibility.

There's no clear cut winner for me, maybe there is among the collective wisdom here. Which would you choose?

Attached is a schematic diagram of the setup, and the .asc file for simulation in LTSpice. This is a topology that deserves much greater attention in my opinion.


Attachments:
File comment: LTSpice schematic
6V6_Circlotron.zip [2.04 KiB]
Downloaded 3 times
File comment: Schematic representation of output stage and drivers
6V6.png
6V6.png [ 106.19 KiB | Viewed 257 times ]
Top
 Profile  
 
 Post subject: Re: Preparing for a Circlotron amplifier build - 6V6 vs 6BG6
PostPosted: Mar Tue 05, 2019 8:11 pm 
Member

Joined: Sep Tue 15, 2015 1:16 am
Posts: 378
Location: 18424 PA
Guitar amps have pushed the 6v6 specs for many years. If you want new robust 6v6's, try JJ's 6v6s, they are well known to take a lot of abuse.


Top
 Profile  
 
 Post subject: Re: Preparing for a Circlotron amplifier build - 6V6 vs 6BG6
PostPosted: Mar Wed 06, 2019 4:24 am 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 27332
Location: Detroit, MI USA
The design using the 6X4's is ridiculous, to say the least, and shows the designers had no clue what they were doing. The max rating for heater-cathode voltage is 450, so it's dangerous to run them so close to the limits with no safety margin, and not something that a responsible engineering team should have let out the door.

If you want to built it go for the 6BG6 design and use the later version with modified bias. Why not use solid state rectifiers so you don't need two 5 volt filament windings?

Hint: none of those designs have to run at that high of a B+ voltage to work well. A bit of power output will be lost at lower B+ but the tubes will thank you and it's easy to adjust the bias to match the B+ you end up with.

If I wanted to custom build a 25 watt or so amplifier, I'd copy a Williamson design using the 6BG6's, 807's, 6L6's or 5881's. I built dozens of them over the years and that was the most reliable design and easy to source parts for in years past.

_________________
Dennis

Experience is what you gain when the results aren't what you were expecting.


Top
 Profile  
 
 Post subject: Re: Preparing for a Circlotron amplifier build - 6V6 vs 6BG6
PostPosted: Mar Wed 06, 2019 8:17 pm 
Member

Joined: Apr Tue 17, 2012 7:58 pm
Posts: 60
Location: Metamora MI, 48455
Mr. Detrola wrote:
The design using the 6X4's is ridiculous, to say the least, and shows the designers had no clue what they were doing. The max rating for heater-cathode voltage is 450, so it's dangerous to run them so close to the limits with no safety margin, and not something that a responsible engineering team should have let out the door.

If you want to built it go for the 6BG6 design and use the later version with modified bias. Why not use solid state rectifiers so you don't need two 5 volt filament windings?

Hint: none of those designs have to run at that high of a B+ voltage to work well. A bit of power output will be lost at lower B+ but the tubes will thank you and it's easy to adjust the bias to match the B+ you end up with.

If I wanted to custom build a 25 watt or so amplifier, I'd copy a Williamson design using the 6BG6's, 807's, 6L6's or 5881's. I built dozens of them over the years and that was the most reliable design and easy to source parts for in years past.


I would tend to agree about the 6X4s. I hadn't considered the heater-cathode breakdown voltage, but you're right. That is way too close for comfort. I also think it was probably not a great idea to use a B+ voltage of 420 volts on a 6V6GT.

The EV 30 watt Circlotron seems to be to be a bit more sensible with regards to respecting the limitations of the tubes.

I know one doesn't necessarily need to run the valves so hard, but then running at lower voltages puts the tubes into a less linear part of the transfer characteristic, and also sacrifices power as you've said. 30 watts was a considerable amount of output power for an amplifier in the A1 to B1 region using 6L6 derivatives, and that's part of the appeal of the design. 50 watts was possible of course by swinging the grids positive with respect to the cathodes, but that has its own complications. 20 watts from a single pair of 6V6s was frankly, unheard of.

I don't have any real issues per se with the Williamson topology, but I have issues with the typical implementations. The original triode version has a tendency toward instability, and only produces about 10 watts before clipping. The Ultralinear versions only give about 20 watts at clipping, still require 400 volts or so on the plates, and are much more sensitive to imbalances between tubes, and can actually cause distortion to rise precipitously beyond that of even a normal pentode output stage should the imbalance become too large.

Both the traditional Williamson and the UL version also require very high quality, usually very expensive, output transformers.

The Circlotron has the advantages of lower output impedance, much greater tolerance to mismatching between parts, and doesn't require output transformers of nearly the same quality. But, there's no free lunch. You've gained those positive attributes by creating drive difficulties, and creating complexities in the power supplies. In particular, the capacitance between the two B+ secondaries must be held low. This is usually not too difficult to achieve however. Aside from all that, it would be fun to try something different for a change. Everyone and their brother has some Williamson variant; commercial Circlotrons are frightfully scarce, and the homemade examples are even more so.

As far as sourcing parts goes, Heyboer out near Grand Haven and Toroidy in Poland will wind whatever one wants for a reasonable cost.


Top
 Profile  
 
 Post subject: Re: Preparing for a Circlotron amplifier build - 6V6 vs 6BG6
PostPosted: Mar Thu 07, 2019 4:45 am 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 27332
Location: Detroit, MI USA
I was going to suggest Heyboer for winding an output transformer since they made OEM parts for quite a few companies and their reputation is among the best. If designing a custom transformer for your needs, why not spec it with some extra turns on the primary and tapped, so you could run it with either 6V6's or 6BG6's?

I was thinking two separate but identical power transformers, off the shelf ones ought to work fine, and would give you the needed isolation between the HV windings.

_________________
Dennis

Experience is what you gain when the results aren't what you were expecting.


Top
 Profile  
 
 Post subject: Re: Preparing for a Circlotron amplifier build - 6V6 vs 6BG6
PostPosted: Mar Thu 07, 2019 5:34 pm 
Member
User avatar

Joined: Jan Mon 18, 2010 2:13 am
Posts: 18020
Location: Dayton Ohio
I would go the 6L6-et al group myself.

Someday I want to build a stereo amp using 1625s, but it will only be 15-18 watts per channel.
I want to go easy on them so they last a while :)

-Steve

_________________
Radio Interests
-Zenith
-Sparton
-Pre-War FM
Consoles and floor models, the bigger, the better!


Top
 Profile  
 
 Post subject: Re: Preparing for a Circlotron amplifier build - 6V6 vs 6BG6
PostPosted: Mar Tue 12, 2019 9:25 pm 
Member

Joined: Apr Tue 17, 2012 7:58 pm
Posts: 60
Location: Metamora MI, 48455
azenithnut wrote:

I want to go easy on them so they last a while :)

-Steve


That's the beauty of running the outputs deep in AB-B territory like one would find in a Circlotron or McIntosh. The tubes are biased with little cathode current so that the tube life ~ the life-span of the heater, and the plates and screens remain comparatively cool, but without drawing so little cathode current that cathode poisoning results.

The tubes are beat up, but only on the 15 to 30 watt peaks, something that rarely happens while listening to actual music. With a typical, efficient loudspeaker, Harry F. Olson at RCA found that typical powers actually used in a living room environment hovered between one tenth of one watt and one watt, with occasional peaks of around 5 to perhaps 10 watts.

I guess one could look at the above as actually being an indictment of the McIntosh and Circlotron high power amplifiers. If typical listening is done at powers << 10 watts, why does one need 20 or 30 watts?


Top
 Profile  
 
 Post subject: Re: Preparing for a Circlotron amplifier build - 6V6 vs 6BG6
PostPosted: Mar Sat 16, 2019 12:21 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 8517
Location: Beaver Falls, PA. USA
In one word.... Headroom. If a loud peak comes a!ong, the amplifier can handle it without clipping. Also, the efficiency of the speaker system can make a huge difference, as well as the type of music; low frequency content requires more power.

_________________
Tim KA3JRT


Top
 Profile  
 
Post New Topic Post Reply  [ 8 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 5 guests



Search for:
Jump to:  




























Privacy Policy :: Powered by phpBB