COMPARATOR USING A TUNNEL DIODE

E. M. Ryskin

UDC 621.317.799

As a result of converting a voltage into a quantity that is inversely proportional by means of an integrated operational amplifier [i] a time interval At ~ I/U is formed which is determined by the moment when two voltages are equal -- a linearly-varying voltage and a constant voltage. A comparator (CO) is used to compare these voltages and register the moment of their equality. In the converter mentioned it is possible to use any type of CO. However, the.familiar CO circuits [2] are sensitive to the back swing of the sawtooth voltage. In order to avoid false operations, either special devices are used in the circuit for suppressing the back swing or the measuring channel is blocked at this time, both of which complicate the circuit. It is possible to rule out the sensitivity of the CO to the back swing if the moment of equality of the two voltages is registered by a pulse shape having a polarity that differs for the forward and back swings. This condition can be realized in the CO circuit with a tunnel diode, a flipflop, and a switching device. In the proposed circuit [3] the dc makeup current does not flow through the diode so that the diode's operating point is returned to the origin (when the voltage on the diode U d and the current magnitude passing through it I d are equal to zero) by the back swing of the sawtooth voltage. The circuit of a CO using a tunnel diode is shown in Fig. 1 and curves explaining its operation are shown in Fig. 2. In order that the nonlinearity of the initial portion of the sawtooth voltage should not affect the CO's operation, a two-channel circuit is employed. The curves shown in Fig. 2 are for the operation of only one channel. The channels have different reference levels which are obtained by a suitable choice of the resistors R1 and R2 connected in series with the tunnel diodes TDI and TD2. The sawtooth voltage U(t) is fed to the input of the CO. The operating point of a diode is determined by the intersection of the volt-ampere characteristic of the diode (the lines i, 2, and 3 in Fig. 2) and the load line 4. A change in the voltage U(t) shifts the operating point along the tunnelling branch 1 of the volt--ampere characteristic. At the moment t = tl the condition U(t) =Uol = U o + IoR1 is fulfilled. The operating point of diode TDI jumps from A to B. The voltage on the diode abruptly changes by Ud~ = U B - U A. I t then increases slowly and begins to drop on the back swing. The operating point in this case is moved along the diffusion branch 3 of the volt-ampere characteristic, the voltage on the diode is reduced to Umi n i, after which the operating point jumps to the tunnelling branch 1 and is returned to the origin of coordinates. In this way a rectangular pulse is formed by the tunnel diode (point M in Fig. i).

u~

RT!

TD1

Fig. i Translated from Izmeritel'naya Tekhnika, No. 2, pp. 77-78, February,

1977.

This material is protected b y copyright registered in the name o f Plenum Publishing Corporation, 227 West 1 7th Street, N e w York, N. Y. 10011. N o part o f this pu blication may be reproduced, stored in a retrieval system, or transmitted, in any f o r m or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, w i t h o u t written permission o f the publisher. A copy o f this article is available f r o m the publisher f o r $7.50.

282

U

U~;n

Io

0

'J,:

0,5

lI t,

J

l

1

-"

TABLE 1

---

~'i___n__n Uirl.o o:,

I I

k'

I I

I 2 3 4 5 6 7 8 9 I0

I I

1 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1

Arc

-'Ja

I00 III 125 143 167 200 250 833 500 I000

100 111 125 144 168 202 254 338 511 1034

o" .o

6a

0 0 0

-U0,7 --1,6 +1,5 ,Juo

45,4

+0, I +0,25 4-0,4 --,0,45 +0,5 -60,75 4-O,9 " ~i,1 +1,5 + 3 , 15

Fig. 2 The second channel with R2 and TD2 operates similarly. The condition U(t) = U o 2 = U o 2 + Io=R2 is fulfilled for it at the moment t= t2. The voltage on the diode TD2 changes abruptly by AUd2. It is clear that if R2 > RI, then t2 > t~. The voltages Ud~ and Ud2, after being differentiated by the circuits CIR3 and C2R4 (point N in Fig. I), control the operation of a flipflop consisting of transistors T3 and T4. At the output of the flipflop a voltage pulse Uou t is formed which has a duration A t = t2--t~ inversely proportional to the rate of growth of the sawtooth voltage U(t). The flipflop is controlled by the silicon transistors T1 and T2 operating in a switching mode which are in the blocked condition. Positive pulses, which are formed during the forward swing of the sawtooth voltage, open the switches, are amplified, and reverse the flipflop. Negative pulses, which are formed during the back swing, do not change the condition of the switches and consequently do not alter the state of the flipflop. A temperature change has a great effect on the operation of a CO. As is well known [4 to 7], the parameters of tunnel diodes are nearly independent of the temperature (tenths and even hundredths of a percent per degree), and U o decreases but I o is increased with rising temperature. The temperature drift of the point A ( I = Io; U = U o) moves along a trajectory close to the load line so that the drift of the actuating voltage of the CO U(t) = Uo is less than the drift of point A on the volt-- ampere characteristic. Owing to the two-channel system the instability of the duration for the output pulse is less than the instability of the actuating voltage. The operation of the CO using tunnel diodes was studied in conjunction with an integrated operational amplifier [i]. In order to determine the linearity of converting an input voltage to a digital code, a pulsed voltage was applied through a calibrated attenuator to the input of the operational amplifier. The amplitude of the input pulses Uin.o was chosen so that the figure i00 flashed on the illuminated indicator board connected to the output of the CO. Then, by means of the calibrated attenuator the values 0.9, 0.8, ..., 0.i of the initial amplitude were set up, the conversion results were recorded, and they were compared with the calculated values of Ncalc= i00 Uin.o/Uin. The results of the test and also the conversion errors calculated by the method presented in [I] are shown in Table i. The discrepancy between 6actual and dcalc is not outside of the limits of accuracy for the calculation of the conversion error [i]. The results given in Table 1 were repeated for numerous measurements. LITERATURE CITED i.

E. M. Ryskin,

Izmer. Tekh.,

No.

6 (1973).

283

2.

A . A . Kuznetsov and O. A. Kuznetsov, Elements of High-Speed Analog-Digital Converters tin Russian], Energiya, Moscow (1966). E . M . Ryskin, Inventor's Certificate No. 410362, Byull. Izobret., No. 1 (1974). N . A . Belova et al., Tunnel Diodes [in Russian], Nauka, Moscow (1966). A . A . Vizel' (editor), Tunnel Diodes and Their Use in Switching Circuits and Apparatus for the Superhigh-Frequency Range [in Russian], Sov. Radio, Moscow (1965). V . I . Fistul' (editor), Tunnel Diodes [in Russian], Inostr. Lit., Moscow (1961). P . V . Konstantinov, in: Semiconductor Devices and Their Application [in Russian], (ed. by Ya. A. Fedotov), 21st edition, Sov. Radio, Moscow (1969). t

3. 4. 5. 6. 7.

DIODE MEASURING BRIDGES A. S. Buevich

UDC 621.317.733.025.088.6:536.531

Resistance transducers are used widely for the remote measurement of various physical parameters. When such a transducer is connected to a Wheatstone measuring circuit via a twoconductor line, the resistance of the conductors and the insulation resistance between them affects the measured results. The error in measuring the transducer resistance that is caused by the cable can be substantially decreased and in some cases practically eliminated by using diode bridge measuring circuits with ac [i]. Figure i shows the simplest tor diodes connected in each arm tion (Fig. la) makes it possible while the circuit with the series the cable's strands.

diode measuring bridges with rectifiers such as semiconduc(series or parallel). The circuit with the parallel connecto eliminate the effect of the cable's leakage resistance connection (Fig. ib) eliminates the resistance effect of

A dc measuring instrument 2 (Fig. la), which is shunted for ac by a capacity, is connected in series with a source of ac i. The measuring circuit has two arms -- transducer Rr, which is connected through a two-conductor cable having equivalent leakage resistance R1, and standard resistance R s (a resistance box). Connected in series with each arm of the bridge is a diode and the polarities of the diodes with respect to the ac source in the different arms are different. During the first half-period of the supply voltage a current flows through diode DI into the transducer R r. During the second half-period current flows through diode D2 into the standard resistance R s. For the two half-periods current flows into leakage resistance R 1. If the transducer's resistance equals the standard resistance, then the value of the current in the supply circuit during the first half-period equals the current value in the second half-period. Inasmuch as the directions of the current in the first and second half-periods are opposite, the average value of the current over a cycle in the supply circuit is equal to zero. Consequently, the measuring instrument gives no indication of a dc component in the supply circuit, which is the criterion for a bridge balance. The bridge balance is not disturbed by a change in the leakage resistance. If the transducer's resistance is not equal to the standard resistance, the bridge is unbalanced. In this case the absolute value of the supply-circuit current during the first half-period differs from the current value during the second half-period and the measuring instrument indicates a dc current component:

Translated from Izmeritel'naya Tekhnika, No. 2, pp. 79-80, February,

1977.

This material is protected b y copyright registered in the name o f Plenum Publishing Corporation, 227 West 1 7th Street, N e w York, N. Y. 10011. N o part o f this publication m a y be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, w i t h o u t written permission o f the publisher. A c o p y o f this article is available f r o m the publisher f o r $7.50.

284