[Boatanchors] PA tube biasing?

[eugene at hertzmail.com]

Hi there,

Working on my harris RF-110A (urt-23) and had a question about biasing.
I am reading the procedure to properly set the bias of the two finals
(4CX1500B).  The instructions explain that the difference in idle plate
current of the two tubes should not exceed 40ma.  To assist in getting
the two plate currents to within 40ma of each other, there is a diode
voltage divider that allows you to change the screen voltage of each
tube from 250v to 320v. There are two jumpers with 6 positions to allow
each tube's screen voltages to increase or decrease based on the
position chosen.  The two tubes are currently set for the first position
for V1 and the next to last position for V2.

I have noticed that the two tubes are not identical, meaning they must
have been acquired at different times (one says eimac, one does not,
etc).  

My question is what causes the requirement of different screen voltages
to produce the same plate current? Is it simply differences in
manufacturing? Or does it indicate that one tube is older than the
other? If age is the cause, my concern is that with only one jumper
position left on V2 that I might be running out of life on the tube. I
suspect a 4CX1500B is very $$$.

So, is age the primary cause of the need to change screen voltages in
order to balance the tubes? Or is it mostly mfg differences in which
case, I could conceivably keep these screen voltages for a very long
time to come.

Thanks!
Eugene