Yes, this does make some sense.. Still I would expect the plate to grid spacing to be much less critical than say the cathode to grid spacing in terms of tube parameters. I really could also be quite wrong here.. 😀
Unfortunately the problem with moving the plate to the location where G2 formerly resided probably means some new tooling is required and this is no longer a trivial exercise.
It would be interesting to figure out what the parametric ramifications of just removing the screen and suppressor grids actually is - to what degree does this affect transconductance, rp and mu? Sadly the guys who would likely know this off the top of their heads are probably long gone..
Last edited:
If you just remove the g2 you get a beam triode. Like 6HS5 and similar. (so the experiment has been done) Mu is then up around 300 and the plate requires KV to get any current.
I agree with DF96 and Dave S that the new plate has to conform to the old g2. Which generally will make it too small for the dissipation rating unless plenty of fins get included.
If someone is going to "triodize" an audio tube like KT88, why not go for something much better like a 6LR6, 6HJ5 or 6LQ6. Then one can use a low Z OT too.
I agree with DF96 and Dave S that the new plate has to conform to the old g2. Which generally will make it too small for the dissipation rating unless plenty of fins get included.
If someone is going to "triodize" an audio tube like KT88, why not go for something much better like a 6LR6, 6HJ5 or 6LQ6. Then one can use a low Z OT too.
Last edited:
revenge on the pentode ?
here it comes ...... WHY would you want to maim the the most successful penthode ever made in its output class to make a simple triode . the reasons for suggesting such an barbaric act are beyond me . [/COLOR
edit . is this some kind of revenge ? from the USA . for the fact that at the time the EL34 was designed we still had the patent on the pentode ? and you barbarians had to do whit beam forming plates 😀
anyway . how can you even think you can make a plug and play el34 triode version ? this is also beyond me . as it wont work in a million years .
here it comes ...... WHY would you want to maim the the most successful penthode ever made in its output class to make a simple triode . the reasons for suggesting such an barbaric act are beyond me . [/COLOR
edit . is this some kind of revenge ? from the USA . for the fact that at the time the EL34 was designed we still had the patent on the pentode ? and you barbarians had to do whit beam forming plates 😀
anyway . how can you even think you can make a plug and play el34 triode version ? this is also beyond me . as it wont work in a million years .
Last edited:
Since they aren't made in Holland anymore (Philips held the original pentode patents) I'm not sure you should be seeing too much red over the issue. 😱 😀 All hypothetical and of the "what if?" category. 😀
Further plenty of pentodes were made here under technology cross-licensing agreements. Beam technology was often used in the high voltage regulator tubes commonly used in early color TV sets, oops I forgot Europe had no early color TV.. 😛😛 (I'm just teasing, and I know first hand how superior the performance of PAL was compared to NTSC, but it did come quite a few years later, having learned from our obvious mistakes..) 😀
I hadn't even considered what smoking-amp and Dave raised about the position of the plate relative to G2 and the need to move it. I think I understand now, I have noted the very close spacing of the plate to the cathode and grid in both the 6CK4 and 6AH4. The 300B has much wider spacings filament to grid and grid to plate and manages a much lower mu, but still probably closer than would be the case with a modified EL34.. This would make an entirely new tube for which there would be no discernible initial market. The triode connected EL34 sounds pretty good in my experience. I'm still not a fan of pentode sound despite recent dabblings with a pentode (6V6) SE amp..
Personally I'd rather spend my lucre on improved dhts, wherever they came from originally.. 😀
Last edited:
Well the biasing will likely be a problem as far as "plug n play" goes. More grid voltage drive required, Miller capacitance thru the roof.
Maybe could up the gm a bunch with a frame grid, then the drive required might be similar to the pentode. OH, wait, 8417, just triode that one.
Maybe could up the gm a bunch with a frame grid, then the drive required might be similar to the pentode. OH, wait, 8417, just triode that one.
Last edited:
Well granted that just removing G2 will make getting any plate current at reasonable voltages rather difficult. Just try grounding G2 and see what happens.
The 6HS5 and the likes uses a frame grid that is spaced one nano nerf from the cathode to get the high Mu. Yes I smashed one just to see whats going on inside and it does look like a sweep tube with G2 left out until you look at G1. Never saw a frame grid sweep thbe before. They use a "diffusion bonded cathode" to support the very close spacing. The one I killed had the "diffusion bonding" scattered all over the inside of the tube. These tubes fail from a K to G1 short quite often. The 8417 does too.
Can of worms, eh?
Ok, so put it on a can opener, dump out the worms, and voila! 😀
Empty can, no worms... problem solved!
Thanks, no applause please.
_-_-bear
Ok, so put it on a can opener, dump out the worms, and voila! 😀
Empty can, no worms... problem solved!
Thanks, no applause please.
_-_-bear
Anode to cathode spacing (for a triode) is just as important as grid to cathode spacing. The ratio between them helps set mu. For a tetrode/pentode g2 plays the role of the triode anode, and their anode is merely an electron collector.
Ok, let's reverse the process. Take a known DHT, be it a 300B or an 811a and change out the cathode for an indirectly heated one. What would that look like? Possible?
_-_-bear
_-_-bear
OK, ok. So what was the cathode/heater assembly like in that tube, and so why was the linearity not so good?
_-_-bear
And - isn't the 6AS7 usually used as a pass element vreg tube?
_-_-bear
And - isn't the 6AS7 usually used as a pass element vreg tube?
Not having seen a 6AS7 to inspect the internals, I can only make some guesses based on the curves (so take this with a grain of salt). The tilting over plate curves and increasing Mu with current indicate significant grid wire "island" effect. The main criterion for a regulator would likely be low plate resistance for low voltage drop, so the grid is probably widely spaced pitch-wise to get that, but closely spaced distance wise from the cathode to make up the lost gm. Linearity, not being a big concern for a regulator, got lost in the shuffle. Might still work good in P-P.
Last edited:
I am holding a 6AS7 in my hand here along with a 50 and an 841. Grid wire spacing is very wide in the 50, similar between 841 and 6AS7. Grid is very close to the cathode and to the plate, roughly in the middle for the 6AS7.
50 also has grid roughly spaced in the middle between filament and plate, but at much greater distances. 841 has about four times the spacing between plate and grid as it has between filament and grid. rp is very high and mu = 30. Still very linear.
This is all just eyeballed.
50 also has grid roughly spaced in the middle between filament and plate, but at much greater distances. 841 has about four times the spacing between plate and grid as it has between filament and grid. rp is very high and mu = 30. Still very linear.
This is all just eyeballed.
Starting with a clean sheet of paper, is it possible to make a IDHT that biases up like a triode-connected EL34 - that is, plug it into a fixed-bias amplifier with a EL34 socket, and have the tube conduct more or less the same current as an EL34?
That's what I'm curious about. No need to re-do the 6AS7 - that's a regulator tube optimized for maximizing current flow, not linearity.
That's what I'm curious about. No need to re-do the 6AS7 - that's a regulator tube optimized for maximizing current flow, not linearity.
One could do that in the context of an existing plate part by in essence making the heater "fatter" so that the cathode was effectively closer to the plate...?
But the EL34 is perhaps not such a good part for this because it is a beam tube, meaning that the electron stream is formed and more or less and concentrated...? (therefore the plate being of the wrong shape)
Although I have seen 811a used in RF linear service where very very nice holes were burned into the plate, more or less dead center... before that you see a nice red/orange spot. So much for an even "spread" of electrons hitting the plate?
I think the 6W4 (?) is a lower power 6V6 tube and has more or less a "plain old" rectangular plate shape... indirectly heated. Perhaps some clues are to be had in this tube...
_-_-bear
But the EL34 is perhaps not such a good part for this because it is a beam tube, meaning that the electron stream is formed and more or less and concentrated...? (therefore the plate being of the wrong shape)
Although I have seen 811a used in RF linear service where very very nice holes were burned into the plate, more or less dead center... before that you see a nice red/orange spot. So much for an even "spread" of electrons hitting the plate?
I think the 6W4 (?) is a lower power 6V6 tube and has more or less a "plain old" rectangular plate shape... indirectly heated. Perhaps some clues are to be had in this tube...
_-_-bear
Last edited:
Reminds me of the time when I thought I wanted to build my own super low impedance power tube out of thoriated tungsten welding rods. Directly heated of course.
- Status
- Not open for further replies.