D O C U M E N T 1 1 0 S E P T E M B E R 1 9 2 3 1 0 7
simpler than we thought. The reflective plane is far too poor to obtain a decent op-
tical image of the filament, even when the gyroscope isn’t rotating at all. The prob-
lems of production, since it boils down to
10–4
mm, are so great that currently only
Zeiss can do such a thing. That is why every compass should be sent to Zeiss to
have the surfaces polished. It appears that only in this way can this problem be
solved satisfactorily and efficiently.
Now to the problem of the additional rotation
twist.[4]
I have to confess that I am
converted to the standard nautical compass for the reversing motor, which yields an
energy translation of
104
and in the normal cases possesses sufficient sensitivity
with a nice robust design. Its contacts deliver sufficient energy to operate a Siemens
transmitter, if the number of revolutions translated by the Siemens transmitter are
reduced to .
The amplifier tube is being experimented with. It has to be used if a sensitivity
greater than about 0.2 degrees is needed. The amplifier tube is, of itself, a highly
finished instrument but has serious deficiencies for our purposes. The tubes avail-
able to us, which are a purchase from last year, can only yield about 1 watt at an
operating tension of 120 volts, hence far too low for driving the Siemens transmit-
ter. One should therefore have much larger tubes. The application of higher ten-
sions brings inconveniences. Furthermore, the tubes operate really advantageously
only when they are excited by direct current; only then can they convert alternating
current into alternating current of the same period and form. Direct current of 120
volts is not always available. In order to obtain really high translations, much great-
er tensions are even required. If aggregates of amplifier tubes are needed, then the
servicing after failures isn’t so simple, either. Each time a glow wire burns out, the
additional-twist device runs the risk of failing. By contrast, the reversing motor
seems not to offer any failure risks, so to speak, and its life expectancy is practically
infinite.
If larger precision in the additional twist is required, the amplifier tube can be
combined with the reversing motor. However, a beneficial substitute for the revers-
ing motor by the amplifier tube is quite unlikely.
The trials are ongoing, mainly regarding whether the amplifier tube can be prof-
itably run solely by alternating current.
Cordial regards to you and your wife from your
A. Einstein.
1
10
----- -