That worked perfectly. 142V on the B+ rails.
That adds another transformer option. Might even be better for AB2 (switched) if a person wanted that option. With all the parts here I'll probably add a switch, not that I'll ever need it but to compare A2 to AB2.
That adds another transformer option. Might even be better for AB2 (switched) if a person wanted that option. With all the parts here I'll probably add a switch, not that I'll ever need it but to compare A2 to AB2.
Also the 6080 with two plates tied together has 26 watts of plate dissipation and a very healthy 2.5 amp filament. The less linear performing 7236 is 30 watts plate, 2.4 amps filament. The 6080 uses 180 volt plate, the 7236 150 volts plate.
The rare 7236 have no tubes for sale on eBay. 6080 has 127 new tubes for sale on eBay starting at $12 each.
The rare 7236 have no tubes for sale on eBay. 6080 has 127 new tubes for sale on eBay starting at $12 each.
Last edited:
The 6080 tubes seem to be variable as well. Some have 3 micas, some don't. At least some have the sink/radiator on top for the grids - good. I suspect that they are consistent with the design the manufacture used for their 6AS7GA tubes, RCA with additional cooling, GE without. Again, just because it says "RCA" doesn't mean it was manufactured by RCA.
eBay has many NOS RCA 6080 for sale. Tube dealers can easily provide matching RCA 6080 in NOS.
Ralph, All,
I am almost done. I had a bias issue. I failed to hook Bridge 1 (-) to the 600R resistor that that is the + half of the auto-bias out. This lead to a relatively slow cascade failure - which I prevented by watching and switching off the rails.
While I was goofing around with this I thought I would try some different tubes in the circuit (at that moment, for lack of a better idea). 6AS7GA, 6336A, 5628A and a 6AS7G. The 6AS7G was the slowest to go to full tilt with the missing bias connection. This probably says something about it's ability to survive where a GA would not. BTW the 5628A had to be shut down sooner than a 6AS7 and the 6336 was about the same. It looks like with this build (correctly) I could use any of these tubes but 4 of the larger tubes in place of the eight 6AS7s.
Should be playing by this afternoon. I am going to install/balance the RCA versions of the GA.
The switch I am going to use for feedback will arrive today. It will have 3 positions, off, 2meg and 1meg. Can anybody suggest using a setting other than the 3 I have listed?
I also turned the 37.4K resistor in the bias circuit into a 50K variable on top of a 30K, giving a swing between 30K and 80K. This allows for setting the initial bias of a group of 4 output tubes since I am using 2 inverters (1 adjustment per triode/inverter). The balance controls are also X4.
Russ
Important P.S. Remember my build is stereo.
I am almost done. I had a bias issue. I failed to hook Bridge 1 (-) to the 600R resistor that that is the + half of the auto-bias out. This lead to a relatively slow cascade failure - which I prevented by watching and switching off the rails.
While I was goofing around with this I thought I would try some different tubes in the circuit (at that moment, for lack of a better idea). 6AS7GA, 6336A, 5628A and a 6AS7G. The 6AS7G was the slowest to go to full tilt with the missing bias connection. This probably says something about it's ability to survive where a GA would not. BTW the 5628A had to be shut down sooner than a 6AS7 and the 6336 was about the same. It looks like with this build (correctly) I could use any of these tubes but 4 of the larger tubes in place of the eight 6AS7s.
Should be playing by this afternoon. I am going to install/balance the RCA versions of the GA.
The switch I am going to use for feedback will arrive today. It will have 3 positions, off, 2meg and 1meg. Can anybody suggest using a setting other than the 3 I have listed?
I also turned the 37.4K resistor in the bias circuit into a 50K variable on top of a 30K, giving a swing between 30K and 80K. This allows for setting the initial bias of a group of 4 output tubes since I am using 2 inverters (1 adjustment per triode/inverter). The balance controls are also X4.
Russ
Important P.S. Remember my build is stereo.
Last edited:
6080 and 7236 tubes have low intrinsic impedance . . .
But that tube's intrinsic impedance is much larger than the load . . . 4, 8, and 16 Ohm loudspeakers.
Making OTL work: global negative feedback which reduces the output tubes effective impedance to be far less than the tubes intrinsic impedance.
But that tube's intrinsic impedance is much larger than the load . . . 4, 8, and 16 Ohm loudspeakers.
Making OTL work: global negative feedback which reduces the output tubes effective impedance to be far less than the tubes intrinsic impedance.
The left channel is up and running. Idling at 480ma per + rail (X2). Only running my test speaker. Considering that, sounds good.
At the moment I have 0 feedback and that will remain an option while in operation. As I was going through the setup I was only running 2 tubes at a time with no feedback. No arcs or sparks on the GA tubes. ( this would have been the worst impedance situation ).
Has anybody ever run less than a 1meg resistor in the GFL? It won't take anything to try, but actually listening to it - well, that will take time. Looking for experienced opinions. I think that some of the Futtermans run 1K. That would be a lot of GFB.
Russ
At the moment I have 0 feedback and that will remain an option while in operation. As I was going through the setup I was only running 2 tubes at a time with no feedback. No arcs or sparks on the GA tubes. ( this would have been the worst impedance situation ).
Has anybody ever run less than a 1meg resistor in the GFL? It won't take anything to try, but actually listening to it - well, that will take time. Looking for experienced opinions. I think that some of the Futtermans run 1K. That would be a lot of GFB.
Russ
phlogiston,
Class A, 0.5 Amps x 0.5 Amps x 8 Ohms = 2 Watts peak, or 1 Watt RMS. Any more power than 1 Watt RMS is Class AB.
Multiply 0.5 Amps times the B+ Voltage.
That is a lot of quiescent power to get that first watt. I hope the "Power Usage Police" do not show up.
That would warm up my house in the wintertime.
If I designed and built OTL amplifiers,
I would probably end up with two things:
A fire
Another Hernia
All my amplifiers use output transformers; they are not much more efficient than OTL, whether it matters, or not.
I know OTL sounds good; but I choose not to go down that path.
Have Fun!
Class A, 0.5 Amps x 0.5 Amps x 8 Ohms = 2 Watts peak, or 1 Watt RMS. Any more power than 1 Watt RMS is Class AB.
Multiply 0.5 Amps times the B+ Voltage.
That is a lot of quiescent power to get that first watt. I hope the "Power Usage Police" do not show up.
That would warm up my house in the wintertime.
If I designed and built OTL amplifiers,
I would probably end up with two things:
A fire
Another Hernia
All my amplifiers use output transformers; they are not much more efficient than OTL, whether it matters, or not.
I know OTL sounds good; but I choose not to go down that path.
Have Fun!
Last edited:
FWIW my atmasphere M60 inspired monoblocks
have worked flawlessly for several years and still sound amazing. Thanks Ralph for your generosity!
Each has 10 Russian 6N13S’ and 1KW power x-formers. Luckily I stocked up on tubes and have a lifetime supply. They drive my Heresies with ease and zero feedback added.
have worked flawlessly for several years and still sound amazing. Thanks Ralph for your generosity!
Each has 10 Russian 6N13S’ and 1KW power x-formers. Luckily I stocked up on tubes and have a lifetime supply. They drive my Heresies with ease and zero feedback added.
That is good to hear. Mine is up and running now too. No fires or police - but maybe if I turn it up - - -. I am going to install the feedback loops as I described, with OFF being one of the options. I still have a lot of wire to dress. I'll probably use waxed string (old ATT).
Also, to answer my own question - in converting RCA to XLR, I hooked the shield (return) of the RCA to pin 3 of the XLR only. Pin 1 of the XLR still goes to ground but there is no connection to the RCA. This works fine on a short run and I have no noise - yet. A longer run of cable using the RCA could be a problem.
I cant tell any difference between a 1meg feedback loop and none, on my test speakers. It probably looks better on the scope but I hardly ever listen to the o-scope.
Driving it with my old Galaxy note 3 at full output I barely see any movement from the meters on the B+ rails. I assume that notable movement would indicate a transition to AB. I can drive it much harder - later - if the are no cops around ;-) (the meters are on the 4 + rails , switched and indicate idle current - 480ma)
I cant tell any difference between a 1meg feedback loop and none, on my test speakers. It probably looks better on the scope but I hardly ever listen to the o-scope.
Driving it with my old Galaxy note 3 at full output I barely see any movement from the meters on the B+ rails. I assume that notable movement would indicate a transition to AB. I can drive it much harder - later - if the are no cops around ;-) (the meters are on the 4 + rails , switched and indicate idle current - 480ma)
Last edited:
I’m running mine in balanced mode because my DAC and pre-amp are both balanced (Schitt).
Also using Zero autoformers on the outputs for a different set of speakers I occasionally use.
Also using Zero autoformers on the outputs for a different set of speakers I occasionally use.
I built a 6B4 pre for one of my Futtermans. My problem is that the cables crossing the room are RCAs, but I have an old S-video cable that might work. Just 1 more project.
Oh, and also, keeping with the topic, the 6AS7GAs installed and calibrated are RCAs, the one with a grid radiator (sink) on top.
During the build I ran the GEs (no radiator) and they were all laying on the horizontal, which is probably a bad idea in any case. There were no arcs or sparks and no pops through the speakers but they were all NOS so that could have developed eventually.
I have all of the + and - rails fused at 1 amp (X 8). If I really want to go AB they will probably need an upgrade.
Oh, and also, keeping with the topic, the 6AS7GAs installed and calibrated are RCAs, the one with a grid radiator (sink) on top.
During the build I ran the GEs (no radiator) and they were all laying on the horizontal, which is probably a bad idea in any case. There were no arcs or sparks and no pops through the speakers but they were all NOS so that could have developed eventually.
I have all of the + and - rails fused at 1 amp (X 8). If I really want to go AB they will probably need an upgrade.
wattsman,
Draw a 60 Ohm Load line across all the the plate curves of a 6080 triode.
Pick a quiescent plate to cathode voltage, and a quiescent plate current.
The middle of the load line needs to cross at the quiescent values.
Now . . .
Look at all the grid voltage steps (from plate curve to plate curve).
Are all the step sizes equal?
Then do the same for a 15 Ohm load line, look at step to step linearity.
Then do the same for a 240 Ohm load line, look at the step to step linearity.
What is your conclusion about load line impedance, versus linearity on a 6080 triode?
Wait a minute, I just used my last 60 Ohm speaker driver, and my last 240 Ohm loudspeaker driver . . . I guess I need to parallel lots of 6080 triodes, Right?
Draw a 60 Ohm Load line across all the the plate curves of a 6080 triode.
Pick a quiescent plate to cathode voltage, and a quiescent plate current.
The middle of the load line needs to cross at the quiescent values.
Now . . .
Look at all the grid voltage steps (from plate curve to plate curve).
Are all the step sizes equal?
Then do the same for a 15 Ohm load line, look at step to step linearity.
Then do the same for a 240 Ohm load line, look at the step to step linearity.
What is your conclusion about load line impedance, versus linearity on a 6080 triode?
Wait a minute, I just used my last 60 Ohm speaker driver, and my last 240 Ohm loudspeaker driver . . . I guess I need to parallel lots of 6080 triodes, Right?