Why not use current production 7591? They would be a good match to the existing transformers and are easy to drive.
I thought it was Portuguese, oh well... 
Anyway, interesting article that explains more. Campbell mentions sonic performance several times, which would explain the choice of finals and the inclusion of screen regulation, something you don't see very often (this makes the 3rd example seen in the wild: one was also a plate modulator, and the other a theatre amp that used 807s).
He does seem to be running the finals a bit thin, at PD= 6.75W. (The Le Renard runs the finals up to 17.5W) Maybe to accommodate 6BQ6GTAs? The 'GTA's will melt down in Le Renard, as these wimpier versions red plate badly, though the 'GAs and 'GTBs both work just fine, and neither show any trace of colour. Maybe he didn't realize that HD finals are rated more conservatively than audio finals?
From what little I could find out about the IST, it looks like a "10K" primary, and it's good to 300Hz. Hideous for fidelity, but OK considering the purpose. He does, however, run R1 (the shared cathode bias resistor) on the edge of poofage.
"At full output from an audio tone signal, the current will swing to about 165 mA...", but not for long, as R1 will be dissipating over 7W, and it's specced at 2W.
I'd like to see a 10W resistor there.
There are other claims made here that lead me to believe he doesn't really know what he's doing: "About 45V of bias is obtained with the 270R resistor..." -45V puts the 6BQ6 into plate current cutoff. If that's the case, then where does that 45V come from?
I= 45/270= 166.67mA
That current could only come from VR-150, which is specced for 40mA max. Also:
45^2/270= 7.5W, and it's a 2W resistor
So which poofs first: R1 or VR-150?
"If a battery of about 30V is used for bias.." Here, he contradicts himself. Given the stated bias current for the finals, 25mA, VGK= -30V will give you that. No way could he have measured 45V across R1.
Sloppy write-up?
Or blind squirrel
who found a nut?
Also, I really dislike that PS design. It doesn't include bleed-off resistors, and that's dangerous. If you power cycle that before the finals heat up, and 6BQ6s, with their enormous cathodes, take quite a bit of time before they're hot enough to pull current. All the small signal tubes in Le Renard are glowing brightly and the cathode current meter is still pegged at the zero stop.
Anyway, interesting article that explains more. Campbell mentions sonic performance several times, which would explain the choice of finals and the inclusion of screen regulation, something you don't see very often (this makes the 3rd example seen in the wild: one was also a plate modulator, and the other a theatre amp that used 807s).
He does seem to be running the finals a bit thin, at PD= 6.75W. (The Le Renard runs the finals up to 17.5W) Maybe to accommodate 6BQ6GTAs? The 'GTA's will melt down in Le Renard, as these wimpier versions red plate badly, though the 'GAs and 'GTBs both work just fine, and neither show any trace of colour. Maybe he didn't realize that HD finals are rated more conservatively than audio finals?
From what little I could find out about the IST, it looks like a "10K" primary, and it's good to 300Hz. Hideous for fidelity, but OK considering the purpose. He does, however, run R1 (the shared cathode bias resistor) on the edge of poofage.
"At full output from an audio tone signal, the current will swing to about 165 mA...", but not for long, as R1 will be dissipating over 7W, and it's specced at 2W.
I'd like to see a 10W resistor there.There are other claims made here that lead me to believe he doesn't really know what he's doing: "About 45V of bias is obtained with the 270R resistor..." -45V puts the 6BQ6 into plate current cutoff. If that's the case, then where does that 45V come from?
I= 45/270= 166.67mA
That current could only come from VR-150, which is specced for 40mA max. Also:
45^2/270= 7.5W, and it's a 2W resistor
So which poofs first: R1 or VR-150?
"If a battery of about 30V is used for bias.." Here, he contradicts himself. Given the stated bias current for the finals, 25mA, VGK= -30V will give you that. No way could he have measured 45V across R1.
Sloppy write-up?
Or blind squirrel
who found a nut?Also, I really dislike that PS design. It doesn't include bleed-off resistors, and that's dangerous. If you power cycle that before the finals heat up, and 6BQ6s, with their enormous cathodes, take quite a bit of time before they're hot enough to pull current. All the small signal tubes in Le Renard are glowing brightly and the cathode current meter is still pegged at the zero stop.
Last edited:
Hey tubelab,
In an earlier post you wrote this: "A modern 6L6GC can make about 30 watts in push pull triode connection, and does sound a bit different than the same tube in pentode mode."
Do you know what the plate to plate output (zout) would be in this case? Do you know of a schematic for this type of amp?
Kevin
In an earlier post you wrote this: "A modern 6L6GC can make about 30 watts in push pull triode connection, and does sound a bit different than the same tube in pentode mode."
Do you know what the plate to plate output (zout) would be in this case? Do you know of a schematic for this type of amp?
Kevin
Last edited:
Back in 2009 I was part of a globe spanning cooperative effort to design a 6L6GC amp. ChrisH built the amp and has used it happily ever since. I built a breadboard that has been used to test dozens of tubes from 45's to big TV and transmitter tubes....including 6L6GC's. The entire design and build is in a very long thread:
http://www.diyaudio.com/forums/tubes-valves/133034-6l6gc-ab2-amp.html?highlight=6L6GC+AB2
The first test with 6L6GC's are in post #127, and the test data is in post #144. Short story, 6600 ohms on 450 volts results in 41 watts in triode allowing for AB2 drive. Less power will result with lower B+ voltages.
His final schematic is in post #627, and somewhere else too.
http://www.diyaudio.com/forums/tubes-valves/133034-6l6gc-ab2-amp.html?highlight=6L6GC+AB2
The first test with 6L6GC's are in post #127, and the test data is in post #144. Short story, 6600 ohms on 450 volts results in 41 watts in triode allowing for AB2 drive. Less power will result with lower B+ voltages.
His final schematic is in post #627, and somewhere else too.
I think so, the 45V seems to be a typo, perhaps he meant 25V? Which is closer to the 30V battery bias, anyway, the cathode resistor value seems off as well becasuse none of the current/voltage makes sense if that was the case. Despite all that, at least the (translated) article put you onto the 6BQ6's, so perhaps it wasn't all bad...😀Sloppy write-up?
45V would have to be a typo. (It happens: I found a glaring one in Chemical Abstracts and that's a peer reviewed professional journal.) There's no way. Given the stated bias current, 50mA, that, plus 40mA will give V= 270(90E-3)= 24.3V -- and that would come close to giving a Q-point plate current of 25mA. He's pushing that VR-150 for all it's worth since the screens will pull 23mA/final at full power peaks. This is why I opted for an active voltage regulator for Le Renard, even though the screen voltage does coincide nicely with the voltage produced by the VR-150. For that, I used a VR-75 pulling ~7.0mA as a voltage reference, and let the series pass device handle the screen current demand. It would also seem that he's using a 2W resistor as a "fuse" of last resort in case something goes wrong, and the OPT is left unloaded. I really don't like the neglect to include bleed-off resistors.
Even though that modulator falls far short of high fidelity (open loop design, tiny, frequency limited IST, small coupling capacitors, and cathode bias) it'll work better than a helluvalot of other designs I've seen, and some schemos are just plain ridiculous.
G2 current?
Hey Everyone,
I've chosen a bias point of 58ma with -16.5 volts on the grid. This amp will run in triode mode. In all the 'convert XXX to triode' they mention using a 100 ohm resistor to connect the plate to G2. The data sheet doesn't give a dissipation spec for G2 in triode. And a 100ohm resistor wouldn't limit current anyway. Does this mean that in triode G2 doesn't flow any current?
Hey Everyone,
I've chosen a bias point of 58ma with -16.5 volts on the grid. This amp will run in triode mode. In all the 'convert XXX to triode' they mention using a 100 ohm resistor to connect the plate to G2. The data sheet doesn't give a dissipation spec for G2 in triode. And a 100ohm resistor wouldn't limit current anyway. Does this mean that in triode G2 doesn't flow any current?
?
G2 will always consume current as long as there is a positive voltage on it. That is usually the case for normal operation, triode, pentode, UL, cathode follower......
I would use at least a 1 watt resistor. 6L6 types work well with a 100 ohm resistor, it is there mostly to prevent oscillation. Some EL34's benefit from a higher value.
I found that I could even blow a 1 watt resistor by plugging my guitar into the HiFi amp and setting the preamp on KILL! All SSE's get a 2 watt resistor for that reason, but the extreme case was crummy EL34's in UL driven beyond clipping. I never know if someone out there will do crazy stuff with my amp designs.....there are a few people out there that do!
I'm currently breadboarding up an amp similar to the one I had years ago (the 6L6GC AB2 amp). I plan to test a bunch of different tubes, but 6L6GC will fit, so I will do some testing. What impedance are your OPT's?
That's another sticking point with the HK iron. I measured them and they are are 3650 for 16 ohm and 3800 for 4 ohm. Seems kinda low even for 7355. I may have to run the 8 ohm on the 4 ohm tap which would be about 7600. (My math skills are not the greatest I plan to finish it and see what it sounds like?)
Use the 82 ohm 2W
I posted some 6L6GC test data in the 6L6GC in AB2 thread. Post # 651
http://www.diyaudio.com/forums/tubes-valves/133034-6l6gc-ab2-amp-new-post.html
I posted some 6L6GC test data in the 6L6GC in AB2 thread. Post # 651
http://www.diyaudio.com/forums/tubes-valves/133034-6l6gc-ab2-amp-new-post.html
- Status
- Not open for further replies.