I have a PP transformer rated at 65 watts. Primary runs 3300 ohms. It's intended to run something like 6550 in Ultra Linear. But I'd like to try to find a triode. Things like 845 can run around 4500 ohms at 850 volts. Maybe the Russians have some big triode with a lower plate load? Thought ???
Triodes work best with high impedance loads.
65W as triode with low impedance, say 33% efficient, says you need 200W of dissipation, 100W per plate. BIG tubes (or many of them).
P-P 2A3 with 3K load makes 15 Watts at 300V. Scaling to 65W suggests a B+ like 625V.
P-P 300b with 4K load does 20 Watts at 350V supply. Scaling says 65W needs 630V, confirming the 2A3 extrapolation.
P-P 845 has an AB1 spec for 4600r load, 1000V B+, 75W. Re-loaded to 3300r it might be 55W out. The 1000V, and the 175V grid drive, are discouraging.
The legendary Russian big-bottle pass tube is rated 250V-300V max, and we need more.
Basically, when you get into these power levels, at such a low impedance, as Triode, you need such a huge cathode that it makes sense (commercially) to raise the impedance and voltage rather than put big bucks in cathodes and heater supplies. BIG triodes are more often aimed at >8K load/pair and >1000V supply.
I'm sure you could do something with six 6550/KT88 triode-wired.
65W as triode with low impedance, say 33% efficient, says you need 200W of dissipation, 100W per plate. BIG tubes (or many of them).
P-P 2A3 with 3K load makes 15 Watts at 300V. Scaling to 65W suggests a B+ like 625V.
P-P 300b with 4K load does 20 Watts at 350V supply. Scaling says 65W needs 630V, confirming the 2A3 extrapolation.
P-P 845 has an AB1 spec for 4600r load, 1000V B+, 75W. Re-loaded to 3300r it might be 55W out. The 1000V, and the 175V grid drive, are discouraging.
The legendary Russian big-bottle pass tube is rated 250V-300V max, and we need more.
Basically, when you get into these power levels, at such a low impedance, as Triode, you need such a huge cathode that it makes sense (commercially) to raise the impedance and voltage rather than put big bucks in cathodes and heater supplies. BIG triodes are more often aimed at >8K load/pair and >1000V supply.
I'm sure you could do something with six 6550/KT88 triode-wired.
6CB5A ( 26 Watt Pdiss Horiz. Sweep) has been used in triode mode to make SET amplifiers.
Rp in triode (@ 90 mA) would be around 432 Ohms. Triode class A power typically gets loaded at 3 to 5 times that per tube (SET). (3X for power, 5X for low distortion) Using 4X Rp and P-P class A (another 2X) that would come out to 3450 Ohms P to P.
6CB5 AMP BUILD.
6CB5 in triode PP?
VinylSavor: 6CB5A
Rp in triode (@ 90 mA) would be around 432 Ohms. Triode class A power typically gets loaded at 3 to 5 times that per tube (SET). (3X for power, 5X for low distortion) Using 4X Rp and P-P class A (another 2X) that would come out to 3450 Ohms P to P.
6CB5 AMP BUILD.
6CB5 in triode PP?
VinylSavor: 6CB5A
You might probably try with the 6C33C. Keeping plate dissipation at 30W or less you can run this up to 450V plate voltage. You might run at 400V/75 mA. With 3.3 plate-to-plate in class AB1 operation you should get around 70W or even a bit more. However you would also need a swing of 370-380V grid-to-grid to get such power because of the low gain of the 6C33C. Linearity not really 845-like though...
If the OPT has taps on the secondary, you might put an 8 ohm speaker on the 16 ohm tap. This will reflect 1650R instead of 3300R allowing you to use EL509 in triode with 300V B+, and 120ma idle. Screen tied to plate through 220R. Good for 68W, 10W class A limit. Even with the same parameters but 3300R loading, you'll get 40W, 22W of which is class A. 
This is crazy. 250 Watts just to light the filaments. However if my math is half right, a pair will put 65 Watts into 3300 Ohms, essentially in class A, with plates quite hot (for gettering) but far from melt-down.
And it is a very pretty tube. Four plates! Actually four 75W tubes built as one.
And one of the few BIG tubes rated for R-C grid drive (250K max Rg1).
Must be the "TL" version of 304. The TL barely has enough conductance to be a low-Z audio tube, the TH version is far too stiff.
I do not know if you can still buy a working 304TL. WWII burned up boatloads. Vacuum Tubes, Inc has it listed for $225 but "Out of stock". eBay lists some, of course, "no way to test".
And it is a very pretty tube. Four plates! Actually four 75W tubes built as one.
And one of the few BIG tubes rated for R-C grid drive (250K max Rg1).
Must be the "TL" version of 304. The TL barely has enough conductance to be a low-Z audio tube, the TH version is far too stiff.
I do not know if you can still buy a working 304TL. WWII burned up boatloads. Vacuum Tubes, Inc has it listed for $225 but "Out of stock". eBay lists some, of course, "no way to test".
Attachments
Attachments
Last edited:
The 304TL pair puts out 65 Watts with 850V B+. And takes 125 Watts of filament power per tube.
304TL (new) $200: 300R4 thru 885 – Fair Radio Sales
If you want something hard to drive (curves below):
(and only takes 60 Watts of filament power)
4D32 only takes 24 Watts of heater power. And is available at the Fair Radio link above for $24 each.
6KG6/EL509 only takes 12 Watts of heater power.
Attachments
Last edited:
Maybe 500 Horn type speakers in series-parallel.
Very large football stadium with HiFi PA.
In that scenario, I would suggest 500 pairs of 12HL7 tubes (distributed as 2 tubes to each of the 500 Horn speakers)
Cheaper than the 304TL solution too!! 12 x 12HL7 in parallel could replace each 304TL tube.
Triode curves:
(50 mA max DC per tube spec, curves: 10 mA/div Vert., 50V/div Horiz.)
Very large football stadium with HiFi PA.
In that scenario, I would suggest 500 pairs of 12HL7 tubes (distributed as 2 tubes to each of the 500 Horn speakers)
Cheaper than the 304TL solution too!! 12 x 12HL7 in parallel could replace each 304TL tube.
Triode curves:
(50 mA max DC per tube spec, curves: 10 mA/div Vert., 50V/div Horiz.)
Attachments
Last edited:
Hmmm.... well 6KG6 has an Rp of 380 Ohms at 100 mA in triode mode. (and 34 Watt Pdiss rating)
12HL7 has an Rp of 1430 Ohms at 25 mA in triode mode. (and 10 Watt Pdiss rating) and 21000 gm!
4 x 12HL7 in parallel would give 357 Ohms Rp at 100 mA total in triode. (and 40 Watt Pdiss) And 84000 gm! (5.5V peak drive signal) Not -too- absurd actually. (that GMI-83V needs 400V pk drive !!! now that IS absurd)
(were $1 each last year, but $4 this year)
12HL7 has an Rp of 1430 Ohms at 25 mA in triode mode. (and 10 Watt Pdiss rating) and 21000 gm!
4 x 12HL7 in parallel would give 357 Ohms Rp at 100 mA total in triode. (and 40 Watt Pdiss) And 84000 gm! (5.5V peak drive signal) Not -too- absurd actually. (that GMI-83V needs 400V pk drive !!! now that IS absurd)
(were $1 each last year, but $4 this year)
Last edited:
Just my attempt at Jocularity. In truth I just enjoy the fun of large triode amps. I wrote an article about the Altec 287W a few years back. I have a couple MI-200s I drive things like a Bozak Concert grand and I have a line up of Cerwin Vega cabinets that I outfitted with Vifa & Scanspeak drivers. Truthfully most the time the amps go into a dummy load.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.