I was going to say something about the circuit the tubes are ran in affecting
performance but figured it would just take up space. But since were here now.
--------------
Some may be aware of some or all of the following, so bear with me.
The driver circuit in the original 70 is, voltage gain pentode direct coupled to a
triode in a concertina phase splitter/driver arrangement. (all in 1 dual tube)
Now being direct coupled means the plate of the pentode section of the 7199's
are connected directly to the grid of the triode section, so the plate voltage of
the pentode section sets the whole circuitry in motion.
Ideally you want 80-100v at the plate of the pentode, then double the cathode
voltage of the triode at it's plate, screen of the pentode 10-15v less than plate.
---
Example: Say there is 90v at the plate of the pentode.
You'd want it's screen at somewhere around 78v or so.
The triode section would have roughly 95v at it's cathode.
So you want it's plate to be about 190v, (double the cathode)
---
So with dropping any of these tubes in here then stating this or that sound and
perform better then others, you really should be checking there operating range.
(some tubes may possibly even be running to hard for there own good/lifespan)
Now the ST70-2 ran 6GH8 tubes with all the exact same circuitry, I haven't been
able to find voltage readings for an driver section on the original or the second to
see how the tubes were actually ran, I can only assume the 6GH8 "worked" there.
--------------------
So if it were me I'd get some Holland 6BL8's, test say 5 of them in the circuit then
adjust the circuit values/voltage to have the tubes setup best as they could there.
(I'd also cut the plate and cathode resistors values on the splitters in half, say 24k)
(Also cut the grid resistors on the output tubes in half, then triple coupling caps to)
Don't just test 1 tube either to see how the circuit is running, test at least 5 of any
one type to get the fair assessment of the voltages at all the driver section layouts.