US3138762A - Discharge device having cathode voltage drop effecting control of opposite tube - Google Patents

Description

June 23, 1964 K. REINTJES 3,138,762

DISCHARGE DEVICE HAVING CATHODE VOLTAGE DROP EFFECTING CONTROL OF OPPOSITE TUBE Filed July 18, 1958 2 Sheets-Sheet 1 Fig.2

INVENTOR A Z Rein (/65 BY mmfiw m June 23, 1964 K. REINTJES 3,138,762

DISCHARGE DEVICE HAVING CATHODE VOLTAGE DROP EFFECTING CONTROL OF OPPOSITE TUBE Filed July 18, 1958 2 Sheets-Sheet 2 R07 R02 i T mA mA 01 r '02 C k F I .3 1 k .uk FT g k ik 1 BY m 8 41 5% Maw Aif ys.

United States Patent 3,138,762 DISCHARGE DEVICE HAVING CATHODE VOLT- lggglEDkOP EFFECTIN G CONTROL OF OPPOSITE The invention relates to an arrangement of circuit integration which is distinguished by the following characteristics:

(1) One tube is being controlled by the cathode circuit of a second tube and vice versa. The decrease in voltage on the cathode resistance of tube 1 is effective be tween grid and cathode of tube 2. Consequently, the control voltage between grid and cathode of tube 1 is being received by the cathode resistance of tube 2.

(2) The anode voltages of both tubes and especially their characteristics, the inverse amplification factor as Well as the cathode-anode resistances are equal electrically.

(3) The arrangement includes integration links (R-C or inductance combinations).

These characteristics, used as described in the specification, comprise an arrangement which allows an accurate integration with a minimum of switching elements.

Under the name of Ferroscope or Ferrograph oscillographs are being offered which are primarily used to test ferromagnetic substances. For instance, if hysteresis curves of an oscillograph tube should be shown, it is necessary to integrate the voltage of the secondary winding of the core to be tested. Since these secondary windings are usually very small, the integration elements to be used are equipped with multiple stage amplifiers. Further, the phases of this arrangement do not operate accurately and this has previously prevented a wider dispersion of ferrographs.

The described arrangement will achieve a certain degree of sensitivity without any further amplification which has previously only been achieved by multiple amplification. A further advantage of this arrangement is that the phases operate accurately.

These advantages, together with the symmetric construction and the symmetric output voltage, especially qualify the subject invention to be used in an electron ferroscope.

FIG. 1 shows an ideal family of anode current, anode volts characteristics.

FIG. 2 shows the basic arrangement without integration links as it results from the mathematical derivation.

FIGS. 3 and 4 show the connection of the basic arrangement to the R-C combinations which are used as integration links.

FIG. 5 shows the inductances which are used for integration (i.e. repeating coils, transformers), and which can only be used as integration links in the combination as shown in FIG. 2.

The basic arrangement must be symmetrical. It is not necessary that the integration links be arranged symmetrically. These links may also be coupled with only one tube.

To achieve an accurate integration, the cathode resistance has to be adjusted to a predetermined value. The idealized characteristics of FIG. 1 should be used to mathematically derive this value.

FIG. 2 shows two triodes R0 and R0 taking their anode supply from the two secondary windings of a transformer through full-wave rectifier circuits, the supply being smoothed by a shunt condenser. Each tube has an anode resistance, R and R respectively, and a cathode resistance R and R The two cathodes are joined through a resistance R and the grid of each tube is connected to the negative end of the cathode resistance of the other tube. By this means the grid voltage of each tube is determined by the potential drop in the cathode resistance of the other tube. There is also shown in the anode circuit of each tube a milli-amperemeter mA.

Calling the total anode current i with a numerical suffix indicating the tube referred to, the grid voltage 11,, and the supply voltage U it may be seen from FIG. 2 that there is a potential 1 a1- s1 on the grid of R02 and g2= a2- s2 on the grid of From the family of i,,/ U,, characteristics shown in FIG. 1 it is seen that where D is the reciprocal of the amplification factor R =Ri+Ra+Rs and Ri is the internal resistance of the tube.

Thus

which signifies mathematically that u can take any desired value.

From the expressions so far deduced there result the following:

If R ,=D.R grid voltages and therefore anode currents are in equilibrium in all positions:

If R D.R the one grid voltage tends to assume the stable value ri /=0, the other the stable value u i max. R which in FIG. 1 is u =8 v.;

If R D.R the potentials of both grids tend towards a point in the middle of the load line, marked R,,,+R in FIG. 1, and when R =R this stable point is exactly in the middle of the load line.

Exact equilibruim of the grid potentials can, his true, be achieved only with difiiculty, since even the least departure from (D.R R leads to a determinate value for u moreover the characteristics of the tubes depart from the ideal. However in measuring circuits the departures can be kept so small as to have no effect upon measured values. The less the range of measurement the greater the precision required from the circuit.

If a network such as shown in FIG. 2 is brought to an unstable condition (for instance by varying the resistances R i and u show a strong tendency to oscillate.

Though the drawings show triodes what holds good for triodes holds good in principle for pentodes. Pentodes are preferably to triodes if the characteristics of both are equally free of distortion, because the greater internal resistance of pentodes makes less significant variations in the data of the circuit.

In the circuit of FIG. 2 a combination of resistance and capacitance can be introduced in such fashion, illustrated in FIG. 3, as to oppose changes in bi and i,,. In FIG. 3 a condenser C in series with a resistance R is provided in shunt to each tube; the particular scheme of FIG. 3 shows only one resistance R common to both capacitanceresistance shunts. If i should diminish then (in the example shown in FIG. 3) i g will increase. Consequently the condensers C will be charged and discharged respectively through R,,. A current 'i will flow through R and the potential of the cathode R0 will become negative relatively to that of R0 This will also shift the grid potential u towards negative, and this will oppose the change of current. So there will be heavy damping. The magnitude of the feed-back voltage will be If now a pulse of voltage in the direction indicated is applied to R (for instance by discharging through it a condenser C as in FIG. 3) this will produce a positive pulse on the grid of R0 and a negative pulse on the grid of R0 The anode current of R0 increases and that of R0 diminishes. The voltage u produced by the change of current 5i, opposes the applied voltage. If the voltage is applied in the opposite sense, the sense of u is also changed. But since there is some charging of C there is a certain delay in the rise of the feed-back voltage u This is made apparent if R is too small by the occurrence of a pulse of current in the direction of the applied voltage. If the resistance is sufliciently increased, the current through R is reduced practically to zero, and if the conditions of equilibrium adjustment according to the invention are practically fulfilled it can be said that the result is:

and

e.dt RTornk l (9) that is, the change 51",, in i is proportional to the time integral of the input voltage.

The network of FIG. 3 is symmetrical. There are two resistance-capacitance circuits in it, each in parallel with a tube, but the resistance component R is common to both. The input voltage is in series with the resistance link R There are two different arrangements of the resistance R shown (s. FIGS. 3a and 315).

FIG. 5 shows a network in which the feed-back voltage M is produced by inductances.

The result is:

M46113: Jedi J edt Where M is the inductance of the circuit transformers to be measured.

The circuit shown in FIG. 5 which contains an inductance opposing a change in anode current, and in which the change in anode current is proportional to the time integral of the input voltage, is symmetrical. In the example of construction shown the primary windings are in the anode circuits and the secondary windings lie between the cathodes and the input terminals for the voltage to be integrated. It is, however, equally possible to couple the feed-back inductance with one tube only.

If the feed-back inductance of FIG. 5 is reversed it no longer damps but amplifies variations of anode current. The result is vigorous oscillation.

The arrangement according to FIGS. 3, 4 and 5 shows symmetric deflection voltages for oscillograph tubes on the anodes across the resistances.

I claim:

In a thermionic tube circuit the combination of a pair of sources of current supply and two tubes, each having an anode and a cathode and at least one control grid, and each having a resistance in its anode circuit connecting the anode to the positive pole of a separate one of the sources of supply, and each having another resistance in its cathode circuit connecting the cathode to the negative pole of a separate one of said sources, connections between the cathodes of said tubes and between the negative end of the cathode resistance of each tube and the grid of the other tube whereby the grid voltage of each tube is governed by the potential drop across the cathode resistance of the other tube, and capacitance means connected across at least one tube to produce negative feed-back.

and

References Cited in the file of this patent UNITED STATES PATENTS 2,441,579 Kenyon May 18, 1948 2,581,456 Swift Jan. 8, 1952 2,636,985 Weissman Apr. 28, 1953 2,675,471 Berry et al. Apr. 13, 1954 2,774,868 Havens Dec. 18, 1956 2,797,320 Beaufoy June 25, 1957 2,920,194 Geiger et al. Jan. 5, 1960 FOREIGN PATENTS 103,041 Sweden Nov. 18, 1941

Images