Fortress Technologies An E1T Timepiece, Руководство пользователя

Страница: 117 / 129

longer than the counting process it-self and
sets a limit to the maximum counting speed. A
suitable pulse for feeding into the x' electrode
of the E1T is shown in Fig. 5.9. The slope of the
leading edge of the pulse should not be less
than 2 x 10 7 V/sec and that of the trailing edge
should not be greater than 1.2 x 10 6 V/sec. If
the average amplitude of the pulse is to be
13.6V, the rise time should not therefore be
greater than 0.7 μ sec and the time of fall should
not be less than 11 μ sec.
The mechanism of the counting process can be considered to operate in the following way. If the
operating point is at a in Fig. 5.6 corresponding to an indication of zero, the anode and x"
potential is about 230V whilst the x' deflector electrode potential is about 156V. If a fast rising
positive going pulse of 14V is applied to the x' electrode (raising its potential to 170V), the
voltage of the x" electrode remains constant for a very small fraction of a second owing to the
stabilising effect of the capacitance C. The operating point is therefore momentarily moved to the
point c' on the characteristic of Fig. 5.8. The pulse then decays slowly so that the potentials of x'
and x" decrease at about the same rate. Thus the operating point on the characteristic of Fig. 5.8
at c' is transformed relatively slowly into the characteristic of Fig. 5.6, but the operating point now
has time to move along with the curve and finishes at c in Fig. 5.6.
It can be seen from Fig. 5.6 that the horizontal slot in g
4
(see Fig. 5.4) lifts up the low voltage part
of the anode current/anode voltage characteristic so that the height of each peak above the load
line is fairly constant. The rate at which the stray capacitance, C, can be discharged by the E1T
anode current during counting operations is dependent on the height of each peak of the
characteristic above the load line. A reasonable height for each peak is essential in high speed
counting circuits. This subject is more fully discussed in the section of this chapter which deals
with the design of an input circuit for 100 kc/s operation. The stabilising effect of C on the anode
potential should not be confused with the stabilising effect that R
a2
has on the position of the
beam. Advantage is taken of the latter effect (which is suppressed during the steep front of the
input pulse by the presence of C) for maintaining the beam at the correct position after it has
been displaced.
5.5 - FLYBACK CIRCUITS
When the tenth input pulse is received, the E1T tube must be reset from 'nine' to 'zero'. Normally
this resetting process is initiated by a pulse from the reset anode, al, which is connected to the
H.T. positive line via a 39k Ω resistor (as in Fig. 5.7). If the tube is initially indicating the digit 'nine'
and an additional input pulse is received, the beam will be deflected to strike the reset anode.
The current passing to this anode will cause a voltage drop across the 39k Ω resistor and a
negative pulse can therefore be obtained from the reset anode. The pulse may be used to trigger
a monostable multi-vibrator which is designed to provide suitable pulses to reset the tube and
also to trigger the next decade. Another method of obtaining a pulse to reset the E1T circuit does
not depend on the use of a reset anode. When the beam is deflected from position 'nine' onto
the reset anode, it leaves the g
4
electrode. This electrode is fed from the H.T. line via the resistor
R
g4
and its potential therefore rises as the current through the resistor falls. This rise in potential
can be used to render a triode conducting and the triode in turn provides a pulse to cut off the
E1T. The E1T itself may be reset by two basic methods. In the first method a negative pulse is
applied to the control grid, g
1
or a positive pulse to the cathode, k. This pulse should have an
Page 117 Version 1.0 Copyright Grahame Marsh/Nick Stock 2019
Fig 5.9 - An Input Pulse Suitable for the Operation of
the E1T