Those Telefunken tubes do look great, I ought to try some at some point. ;)
To continue on my technical ramblings -that nobody seems to care about anyway lol- I've found a bit more information that confirms that my reasoning about heptodes isn't totally out there, and that at least a few of my ideas not only work, but have been used in various schematics.
From some random forum post found on google:
"I understand what you are saying, and in fact in researching the ways a heptode can be used, I found several example where it is connected as a triode and one where it is connected as a pentode.
All in all, since g5 is tied to the cathode internally, and g2 and g4 are also tied together internally, we have the following combinations:
As pentode G2 + G4 used as screen grid and:
1. G1 tied to cathode, G3 used as the control grid (*)
2. G3 tied to cathode, G1 used as the control grid
3. G1 + G3 tied together used as control grid
4. G3 tied to plate, G1 used as control grid
5. G3 tied to G2+G4, G1 used as control grid
1. G2+G4 and G3 tied to plate, G1 is the control grid
2. G2+G4 tied to plate, G1 tied to cathode, G3 is the control grid (*)
3. G2+G4 tied to plate, G3 tied to G1 and used as control grid (*)
4. G2+G4 and G3 tied together used as control grid, G1 tied to cathode
5. G2+G4 and G1 tied together used as control grid, G3 tied to cathode
6. G2+G4, G1 and G3 tied together used as control grid
7. G1 and G3 tied to cathode, G2+G4 used as control grid
That makes 5 different pentodes and 7 different triodes but not all are equally promising. The combinations marked with (*) I have seen in various schematic diagrams.
In particular, G2+G4 do seem to be the equivalent of a pentode screen grid, so probably pentodes 1-3 and triodes 1-3 are the useful ones."
So, basically -forgetting about the grounded anode idea for the time being- this guy says that two of my suggestions work.
- Strapping pin 6 to pin 5 (grid 2 and 4 to anode, the amp does on the EF95 setting anyway), chopping pin 1 off and somehow tying it to pin 2 (strapping grid 1 to the cathode, doable but difficult without an adapter) and wire-modding pin 7 to socket hole 1 (using grid 3 as the signal grid). EF95 setting.
- Or -much easier- strapping pin 6 to pin 5 (same as above, basic EF95 setting), and wire-modding pin 7 to pin 1 (tying grids 1 and 3 together).
Now, of these two options, number 2 is the obvious winner. Most of you are already using those heptodes/pentagrid converters with grid 3 (pin 7) floating and have reported that it works better that way. That is probably -as I wrote earlier- because grid 3 is intended, like grid 1, to be a control -signal- grid, and you don't want a signal grid without a signal tied to the anode sandwhiched between two screen grids.
So, instead of just leaving grid 3 floating -which despite everything else, can't be ideal- why not try and strap it with the good ol' wire-mod to pin 1 (with the same EF95 setting)? It makes perfect sense to be using the "signal" grids in a heptode for, well, signal purposes, and tying everything else to the anode.
Anyone feel like trying before I do it? ;)