Videokarma.org

Go Back   Videokarma.org TV - Video - Vintage Television & Radio Forums > General Off Topic Forums

We appreciate your help

in keeping this site going.
Reply
 
Thread Tools Display Modes
  #16  
Old 02-12-2018, 06:49 PM
benman94's Avatar
benman94 benman94 is offline
Resident Lunatic
 
Join Date: Jun 2011
Location: Detroit, MI
Posts: 1,190
Quote:
Originally Posted by Electronic M View Post
One of the RCA tube manuals does give a 6AS7 OTL schematic/parts list...

I went with a reverse Futterman circuit with some negative feedback in the driver stage. With proper adjustment, I was able to achieve almost no DC voltage across the speaker terminals (someday I may rework the bais adjustment circuit to require less finesse). After a year of heavy service, I rechecked the bias balance and tube emission and there did not seem to be a change. My amp is a 2/4 channel with the rear outputs AC drive path and feedback switched over to parallel with the front when in stereo mode. I was very conservative with DC quiescent current in the outputs...I expect output tube life to be close to heater life. My output rails are only +/-60V so there is not much more risk to the speakers in the event of a rail short than some transistor amps.
I built that amp during a summer break in college when buying the outputs at $5 a pop, the chassis sheet metal, wood, and pots (most everything else was from my junk box) was a great stretch of my budget. The power supply evolved with the amp...Eventually utilizing SS regulators for nearly all B+ and B- rails. I could not afford the right power trans so I engineered around the junk box transformers I had....That box has every valid internal mounting point utilized to it's utmost....Probably weighs as much as an RCA 630.
You've inspired me to try my hand at a Futterman style OTL. It would be nice to stop forking over so much money to Brian Sowter at Sowter Transformer...

With +60/-60 rail voltages I would feel a hell of a lot better about such a topology. The one 6AS7G amp. I'm familiar with uses +140/-140 volt rails. That makes me nervous and for good reason. It also calls for a potentially lethal hot chassis design. No power transformer, no output transformer, hell, no inductors at all! A bit scary to have something like that plugged right into the wall...
Reply With Quote
  #17  
Old 02-12-2018, 07:41 PM
Electronic M's Avatar
Electronic M Electronic M is offline
M is for Memory
 
Join Date: Jan 2011
Location: Pewaukee/Delafield Wi
Posts: 14,758
The output's purpose has usually been more focused on efficiently coupling the needed power to the speaker instead of raw signal gain....If it weren't then there would be no preamp/driver stage tubes in all the amps made, and all one would have is outputs.
__________________
Tom C.

Zenith: The quality stays in EVEN after the name falls off!
What I want. --> http://www.videokarma.org/showpost.p...62&postcount=4
Reply With Quote
  #18  
Old 02-12-2018, 07:57 PM
Electronic M's Avatar
Electronic M Electronic M is offline
M is for Memory
 
Join Date: Jan 2011
Location: Pewaukee/Delafield Wi
Posts: 14,758
Quote:
Originally Posted by benman94 View Post
You've inspired me to try my hand at a Futterman style OTL. It would be nice to stop forking over so much money to Brian Sowter at Sowter Transformer...

With +60/-60 rail voltages I would feel a hell of a lot better about such a topology. The one 6AS7G amp. I'm familiar with uses +140/-140 volt rails. That makes me nervous and for good reason. It also calls for a potentially lethal hot chassis design. No power transformer, no output transformer, hell, no inductors at all! A bit scary to have something like that plugged right into the wall...
If you swap the grid coupling caps and put them on the opposite output tube grids from the Futterman design (IIRC this was the called the Technics variant) you will have better output impedance match to low impedances. Back a few years when I was working on mine I found a site that proved it mathematically...Wish I still had the link.

OTLs are IMO the best of both worlds in amps; combining and balancing the redeeming traits of the characteristic sounds of tube and SS gear.
__________________
Tom C.

Zenith: The quality stays in EVEN after the name falls off!
What I want. --> http://www.videokarma.org/showpost.p...62&postcount=4
Reply With Quote
  #19  
Old 02-12-2018, 08:10 PM
maxhifi's Avatar
maxhifi maxhifi is offline
VideoKarma Member
 
Join Date: Dec 2010
Posts: 1,841
Maybe this has something to do with it - source: 1955 Radio Shack Catalog
Attached Images
File Type: jpg tube prices.jpg (120.2 KB, 19 views)
Reply With Quote
  #20  
Old 02-12-2018, 08:17 PM
benman94's Avatar
benman94 benman94 is offline
Resident Lunatic
 
Join Date: Jun 2011
Location: Detroit, MI
Posts: 1,190
Quote:
Originally Posted by maxhifi View Post
Maybe this has something to do with it - source: 1955 Radio Shack Catalog
Fair enough. It wasn't a cheap tube so in the case of damper service it was the bean counters that won out. But, for a push pull amp, you only need 1 6AS7G vs 2 6L6s. The cost in tubes is still comparable.
Reply With Quote
Audiokarma
  #21  
Old 02-12-2018, 08:23 PM
maxhifi's Avatar
maxhifi maxhifi is offline
VideoKarma Member
 
Join Date: Dec 2010
Posts: 1,841
Fair enough, but a pair of 6L6 was usually used in 20-30W amps, for the 10W class there was the 6V6.

And triodes are hard to drive, needing more stages. Triodes in general lost favor over time, even the 2A3 is found in hardly any hifi era products.
Reply With Quote
  #22  
Old 02-12-2018, 08:49 PM
benman94's Avatar
benman94 benman94 is offline
Resident Lunatic
 
Join Date: Jun 2011
Location: Detroit, MI
Posts: 1,190
Quote:
Originally Posted by maxhifi View Post
Fair enough, but a pair of 6L6 was usually used in 20-30W amps, for the 10W class there was the 6V6.

And triodes are hard to drive, needing more stages. Triodes in general lost favor over time, even the 2A3 is found in hardly any hifi era products.
The 2A3 and its variants were used, at least initially. The 6B4G was found in the Sun Radio amps, the 6A5G in the Peerless, the 2A3 in the Brook amplifiers. They weren't terribly common post-1950, but they had adherents. With the popularity of the Williamson, I would hazard a wild ass guess and say that a majority of Hi-Fi amps used between 1947 and say 1953 or 1954 used either triodes or triode strapped beam power tubes.

I'm not going to argue the merits of the triode vs pentode/beam power tube. That is an ongoing battle that has been beaten to death from the dawn of Hi-Fi. I think it is sufficient to say that if the demand of triode output stages didn't exist, hundreds of thousands of Williamsons, with their triode strapped KT66s, 807s, and 6L6s wouldn't have been built. You wouldn't have had articles in Audio Engineering magazine extolling the virtues of the triode strapped 6550s circa 1954. I don't fall into either camp. I think you can build an excellent sounding amp with a pair of 6V6s and a pile of garbage with a pair of 300Bs. The circuit matters far more than the output device of choice.

Sure the 6V6 could do 10 watts in a push pull pair... at 5% THD, with a shitty damping factor, and virtually requiring some sort of negative feedback. A push pull pair of 6L6s could get you ten cleaner watts without much effort. A 6AS7G or pair of 6B4Gs or 6A5Gs could have done even better yet, and with less effort...

There's a reason Paul Klipsch demonstrated his Klipschorn in the early years with a single Brook 12A and not a push pull 6F6 or 6V6 amplifier. It is very easy to build an excellent sounding push pull triode amp. It is much harder to achieve the same level of quality with pentodes/beam power tubes.
Reply With Quote
  #23  
Old 02-12-2018, 08:55 PM
maxhifi's Avatar
maxhifi maxhifi is offline
VideoKarma Member
 
Join Date: Dec 2010
Posts: 1,841
Ok true, but the culture of the time was to use as much negative feedback as would make the amplifier unstable, and then back it off a bit. Given this sort of treatment, beam power tubes formed the basis of high fidelity amps from the early 50s until the end of the tube era. Triodes were still around but no longer mainstream territory.

The triode revival of the 90s was another story. I do remember in the early 90s hifi world made a stereo amp with a 6080 (industrial 6AS7). There have been others too, and also the OTLs.

I run single ended 2A3 in one system, and push pull 807 in another. I like em both
Reply With Quote
  #24  
Old 02-12-2018, 09:16 PM
Electronic M's Avatar
Electronic M Electronic M is offline
M is for Memory
 
Join Date: Jan 2011
Location: Pewaukee/Delafield Wi
Posts: 14,758
I'm a fan of negative feedback myself, and built it into my OTL...It is not needed at lower volumes, but if I want to crank it to 11 the feedback adds some headroom before distortion.
__________________
Tom C.

Zenith: The quality stays in EVEN after the name falls off!
What I want. --> http://www.videokarma.org/showpost.p...62&postcount=4
Reply With Quote
  #25  
Old 02-12-2018, 09:19 PM
benman94's Avatar
benman94 benman94 is offline
Resident Lunatic
 
Join Date: Jun 2011
Location: Detroit, MI
Posts: 1,190
Quote:
Originally Posted by maxhifi View Post
Ok true, but the culture of the time was to use as much negative feedback as would make the amplifier unstable, and then back it off a bit. Given this sort of treatment, beam power tubes formed the basis of high fidelity amps from the early 50s until the end of the tube era. Triodes were still around but no longer mainstream territory.
This is the entire basis of the Williamson, which despite the beam power tube output valves, is in essence just another triode amplifier. If you're tying the screen grid to the plate, you have a triode.

I would say that post-1954, neither the straight beam power tube arrangement nor triode strapping was terribly popular in comparison to the distributed loading arrangement popularized by Hafler and Keros.

No matter what camp someone is in, commit to one and don't go the distributed loading route. It is quite simply a mixture of the worst characteristics of triodes with the worst characteristics of the beam power tube/pentode.
Reply With Quote
Audiokarma
  #26  
Old 02-12-2018, 09:34 PM
maxhifi's Avatar
maxhifi maxhifi is offline
VideoKarma Member
 
Join Date: Dec 2010
Posts: 1,841
The main advantage of ultralinear, so far as I can tell, aside from damping factor, is that you can use a screen voltage the same as the plate voltage, and avoid complexity in the power supply. Beam power tubes really work better when Vg2<Vp, but this of course asks for a more complex and expensive power supply, regulated screens being best.

I don't really like it either, but it does offer a lot of value for money, in terms of performance versus cost and complexity.
Reply With Quote
  #27  
Old 02-12-2018, 10:09 PM
benman94's Avatar
benman94 benman94 is offline
Resident Lunatic
 
Join Date: Jun 2011
Location: Detroit, MI
Posts: 1,190
Quote:
Originally Posted by maxhifi View Post
The main advantage of ultralinear, so far as I can tell, aside from damping factor, is that you can use a screen voltage the same as the plate voltage, and avoid complexity in the power supply. Beam power tubes really work better when Vg2<Vp, but this of course asks for a more complex and expensive power supply, regulated screens being best.

I don't really like it either, but it does offer a lot of value for money, in terms of performance versus cost and complexity.
There's a paper from Williamson in which he rips into the so-called ultralinear design and proves mathematically that under certain sets of conditions, it affords more distortion than a properly built beam power tube output stage with regulated screens, or a comparable output stage with triodes or triode strapped tubes.

You're 100% right, the UL connection was a cheap way to handle the problem of the screen grids. It also ignores the fact that a great number of output valves simply can't have the screen at the same voltage at the plate. About the craziest example I can think of is the 6146. 200 volts max on the screen, and 600 or 750 volts IIRC on the plate. Yikes! Good for 100 watts, so the entire neighbourhood can hear what you're listening to
Reply With Quote
  #28  
Old 02-12-2018, 10:11 PM
benman94's Avatar
benman94 benman94 is offline
Resident Lunatic
 
Join Date: Jun 2011
Location: Detroit, MI
Posts: 1,190
Quote:
Originally Posted by Electronic M View Post
I'm a fan of negative feedback myself, and built it into my OTL...It is not needed at lower volumes, but if I want to crank it to 11 the feedback adds some headroom before distortion.
There's a certain segment of the audiophoolery crowd that eschews feedback of any kind. I prefer to handle it in smaller local loops when possible only for reasons of absolute stability into any load, but still, anything that lowers distortion and widens bandwidth at the expense of gain, which is not exactly hard to come by, is a win-win in my book.
Reply With Quote
  #29  
Old 02-12-2018, 10:48 PM
maxhifi's Avatar
maxhifi maxhifi is offline
VideoKarma Member
 
Join Date: Dec 2010
Posts: 1,841
If you spend a lot of time playing with the curves, even the usual audio tubes benefit greatly from a lower screen voltage. And since pentode gain is largely determined by screen voltage, regulation pays big dividends.

My empirical experience supports your conclusion that pure pentode or triode is better than ultralinear.

As for negative feedback, it's a tool. It obviously is a very useful tool, but like any other tool it can be mis-applied. I think lots of approaches to tube amp design have merit.
Reply With Quote
  #30  
Old 02-13-2018, 09:45 AM
benman94's Avatar
benman94 benman94 is offline
Resident Lunatic
 
Join Date: Jun 2011
Location: Detroit, MI
Posts: 1,190
Here's the article I was looking for from Williamson:

http://www.keith-snook.info/wireless...perlatives.pdf

Cathode feedback windings are great, IF you can drive the output stage. In the case of the typical triode or triode strapped pentode/beam power tube, cathode feedback windings just won't work.

In the case of the pentode/beam power tube, this is certainly an approach that deserves more exploration. But again, it comes at a cost, higher driving voltage at the grids and a much more expensive and unusual OPT.
Reply With Quote
Audiokarma
Reply


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -5. The time now is 05:11 PM.



Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
©Copyright 2012 VideoKarma.org, All rights reserved.