If you’re not running them at full power that advantage is lost. And the EH quad I have is not compact by any stretch of the imagination. Maybe an eighth of an inch shorter than EL34’s and just as big around. Guts look more like 6L6’s. What they are pushing now isn’t even close. You can’t even get the normal 25W of power with 10k load, let alone 6.6k.
The big advantage was supposed to be their higher gm and decent linearity. Vg1 is less negative than anything else, and losing less voltage in a cathode resistor than other solutions which give the same power.
An effective work around which doesn’t require more drive? If you are running in pentode mode there is the 6JA5 (or more common 10JA5). You have to lower the screen way down to the 120-150 range, but when you do it biases up just like a 7868 and provides the same gain. Same pentode exists with a free 12AT7 type triode as 6LU8. Yeah, they can be run up at 450V on the plates if the screen in properly dialed in to keep plate dissipation in the 15-19 watt range. Obviously this won’t help you with triode or UL operation. Prices are creeping up since they have been discovered, but still compared to new production 7591 or even 6L6 they are a bargain. Has anyone tried to restore their old Fisher with these things? I have successfully recreated their power amp sections exactly (other than vg2) and re-used their OPTs in upgraded circuits.
The big advantage was supposed to be their higher gm and decent linearity. Vg1 is less negative than anything else, and losing less voltage in a cathode resistor than other solutions which give the same power.
An effective work around which doesn’t require more drive? If you are running in pentode mode there is the 6JA5 (or more common 10JA5). You have to lower the screen way down to the 120-150 range, but when you do it biases up just like a 7868 and provides the same gain. Same pentode exists with a free 12AT7 type triode as 6LU8. Yeah, they can be run up at 450V on the plates if the screen in properly dialed in to keep plate dissipation in the 15-19 watt range. Obviously this won’t help you with triode or UL operation. Prices are creeping up since they have been discovered, but still compared to new production 7591 or even 6L6 they are a bargain. Has anyone tried to restore their old Fisher with these things? I have successfully recreated their power amp sections exactly (other than vg2) and re-used their OPTs in upgraded circuits.
This is a interesting suggestion with a few issues to over come.
According to the data sheets it looks like if you regulated the screen of a KT77 to around 250V the anode current vs Vg1 and bias characteristics match up pretty well with a NOS 7591. S would be close as well it seems. May even allow the current driver circuit to work without much change.
Not going to work with a UL amplifier however.
Then there is the huge increase in filament current increase in a KT77 over as 7591 that would need dealing with.
The KT77 is also so tall over a 7591...receivers or integrated amps would be pretty steam punk looking when you were done making the holes.
I may pull out a few 6CA7s I have hanging around and run some plate curves of Ia VS Vg2/Vg1 to check the possibility of electrically replacing a 7591s with EL34, 6ca7, KT77.
I like it.
At 2 3/4" tall the 10JA5 is real short just like a NOS 7591. Socket not so much but sockets can change.
The 10JA5 is 1 1/2" in diameter vs 1 3/16" for a NOS 7591 so space may be tight with a few amps.
A screen voltage regulation circuit is not so much to add to most amplifiers if the drive circuits can stay almost the same.
Time to dig into the junk box and see if I have any in there.
Thanks for the suggestion.