[Alex]

Alex:
The only way the current n the screen grid can reverse is the secondary
emission from the screen grid. For this to happen, the tube shall be a
tetrode with VpVg2 or a pentode with Vg3 Vg2. If g2 has a pronounced
and
consistent secondary emission, then Vg2 voltage will tend to rise (jump
because of the diodes) to about 20...50V below the plate (tetrode case) or
below g3 (which is nearly connected to plate in these odd circuits). Thus
g2
will be tracking plate with no g2 net current!!!!!

However I do not think one can dependably rely or reliably depend on the
secondary emission from g2.

And of course, if the beamformer plates or g3 is connected to the cathode,
this will not work at all. And majority of the output tubes do have g3 or
beamformers connected to cathode.

P. Turner:
I cannot observing evidence of g2 secondary emissions.

Patrick Turner.

Alex:
Which power tetrode did you use for the experiment? Or which pentode with g3
connected to plate? ( I do not have a tetrode to experiment.) Note that
there will be no secondary emission current in a conventional pentode or
beam tetrode with g3 or beamformer connected to cathode, so EL84 or 6V6 will
not exhibit any secondary emission from g2.

Generally I would think that secondary emission from g2 is possible (as from
the plate) but it is not consistent enough to build a circuit around it.