I've worked on Macs in the past and usually have
grounded the tube first.
Someone told me that if the machine (at least this TRS-80) had been off
for 24 hours then there is little risk of getting a shock, so I didn't
bother grounding first. As it was there was no problem.
The charge is stored betwee nthe final anode (aquadag coating on the
inside of the 'flare') and a similar groudned coatign on the outisde.
This forms a capacitor with the glass envelope of the CRT as the
dielectric. This is the smoothing capacitor for the EHT supply, of course.
Now glass is a pprtty good insulator, and it's unlilely the capacitor
would self-disharge due to such leakage in a day oe so. But there are
other discharge paths. Some better monitors, paricualrly colour ones have
a bleeer resistor across the EHT supply. Often this is actually a
potential divvder used to measure the EHT voltage and thus control a
regulator circuit. In any case, it'll discharge the CRT in a few minutes
at most.
The otehr discharge path is via the electron beam. If, at turn-off, the
HET genrato suts fown first, there will be some reisidual beam current
(the cathode will eb hot enough to emit electrons, etc) and this will
obviously discahrge the CRT. I susepct that's what haappens in the M4,
sicne I am not aware of a bleeder resistor in the circuit. This will take
a minute or so at most.
I think leaving it overnight will do no more good than leaving it for 5
minutes.
However, it's always safest to discharge the CRT before working on it.
Beelder resitors can fail. Due to a fault there might be no beam current
at all. I anm not in favour of just shroting the anode conenctor to
chassi groudn, though. If there is a considerable residual charge, the
peak current will be quite high .This can damage the conenction between
the fianl anode and its conenctor inside the CRT. Yo uca nalso end up
with currents flwoing where you least espsct them,zapping semiconductors
all over the machine. No, I prefer to use a resistor or EHT meter for
this.
-tony