3 4 2 D O C U M E N T 3 3 8 O C T O B E R 1 9 2 4
apparatus. I draw in resting indoor air through a capacitor with the aid of a ventila-
tor. The capacitor has 50 compartments of surface 5 cm × 5 cm at 1-mm intervals
between the plates and a 50-to-200-volt potential difference.
[1]“insulated cover”
[2]“single-wire electrometer”
[3]“50 to 200 V”
[4]“switch”
[5]“par[tition wall]”
[6]“ventilator”
[7]“capacitor”
At the outset, I had focused on a special source of error. Fields exist in every room,
particularly in the proximity of the 200-volt battery. Since air is always somewhat
ionized, in the proximity of all conductors we must encounter air in which one kind
of ion has been repelled; so a surplus of ions that are oppositely charged to the con-
ductor is present. I therefore take 2 measurements at a time and in between reverse
all fields. That is what the switch is for, drawn between the source and capacitor
and the partition wall, which prevents the capacitor from receiving air from fields
that cannot be reversed. Likewise, I can switch the lighting circuit of the whole
housing from + and 0 to – and 0. The mentioned source of error should therefore
disappear. I can, of course, also let the ventilator run when the capacitor is not
charged. The ions picked up by the wind (which, incidentally, do not disturb my
arrangement
much)
can hence be measured separately.
The isolation also posed problems. I had not completely followed the basic tenet
that the insulation may not be under tension anywhere. When, e.g., the battery
stood isolated on paraffin, fields and corresponding currents or dielectric afteref-
fects and similar disturbances prevailed in the paraffin. Now I set these compo-
nents, isolated, on a metal plate that is connected to the electrometer wire (via the
battery). The metal plate itself is then well insulated. The same principle also ap-
plies to the switch and the capacitor, where it is even more important because there
the fields alternate.
[4]
[5]
[6]
[7]
[8]
[9]
[10]