FREQUENCY SHIFT OF A RELAXATION OSCILLATOR DUE TO AN EXTERNAL ELECTRODE* O.S. STOICAN INFLPR, P.O. Box MG-36, RO-077125, Bucharest-Magurele, Romania E-mail: email@example.com Received December 15, 2007 Experimental observation on the frequency variation of a relaxation oscillator due to an external electrode is reported. The experimental setup consists of a gas filled voltage regulator tube inserted into a relaxation oscillator circuit. The oscillations frequency for various operating conditions has been measured. The results are discussed. Key words: relaxation oscillator, gas filled voltage regulator, glow discharge. 1. INTRODUCTION The use of the properties of the plasma generated inside the glow lamps or gas filled electron tube as an unconventional detection method for various physical effects was reported in several papers. For example in ,  the experimental setups containing discharge tubes used to detect the external microwave field are presented. In this paper the effect of an external electrode on the electrical characteristics of a gas filled voltage regulator tube is described. The change of the discharge properties is observed by measuring the oscillations frequency of a relaxation oscillator circuit. A similar method has been described in order to detect the optogalvanic effect [3–5]. 2. OPERATING PRINCIPLE OF THE RELAXATION OSCILLATOR Generally, a relaxation oscillator consists of an appropriate non-linear switch K, a capacitor C, a resistor R and a direct-current power supply E. The electrical diagram of a typical relaxation oscillator is shown in Fig. 1 (Top). The Ri simulates the equivalent resistance of the switch K when it is closed (state “on”). The * Paper presented at the 8th International Balkan Workshop on Applied Physics, 5–7 July, 2007, Constanţa, Romania. Rom. Journ. Phys., Vol. 54, Nos. 3–4, P. 385–390, Bucharest, 2009 386 O.S. Stoican 2 equivalent resistance of the open switch K (state “off”) is supposed to be infinite. The basic requirements for a non-linear switch that may be used as a relaxation oscillator are as follows. The state of the switch depends on the voltage Uc across its terminals. If the voltage Uc>U2 then the switch state is on, while Uc<U1, with U1<U2, the switch state is off. The threshold voltages U2 and U1, respectively, are intrinsic parameters of the non-linear switch. The state of the switch when U1 <Uc<U2 can be either on or off depending on the previous state (hysteretic characteristic). The capacitor C is connected parallel to the non-linear switch and is charged through the series resistor R. When voltage Uc reaches value U2, the switch K is closed and the capacitor C is discharged suddenly through the switch K. When voltage Uc becomes equal to U1, the switch K is open, the discharge of the capacitor C is interrupted and the cycle restarts. The operating frequency f of the relaxation oscillations can be approximated by the relation: 1 E − U1 ≅ RC ln (1) f E − U2 The waveform of the voltage Uc for an ideal relaxation oscillator is shown in the Fig. 1 (Bottom). By analyzing the time variation of the voltage Uc, the parameters U2 and Fig. 1. – Electrical diagram of a relaxation oscillator (Top). Waveform of the voltage Uc across the terminals of the non linear switch K (Bottom). 3 Frequency shift of a relaxation oscillator 387 U1 can be estimated. Any change of these parameters determines a variation of the oscillations period which can be easily measured. Usually, these parameters are related to some properties of the device used as a non-linear switch. As a consequence the relaxation oscillator principle provides a sensitive tool to study such kinds of physical systems. 3. EXPERIMENTAL SETUP The experimental setup consists of a gas filled voltage regulator tube inserted into a relaxation oscillator circuit (Fig. 2). A gas filled voltage regulator tube is an electron tube electrically equivalent to a Zener diode. Basically, this component represents a dc electrical discharge tube operating in the glow discharge regime. For a dc electrical discharge tube the breakdown voltage is larger than the minimum voltage necessary to maintain the discharge. This feature allows the use of a dc discharge tube as a non-linear switch for a relaxation oscillator. In this case U1 and U2 represent minimum voltage necessary to maintain the discharge and discharge breakdown voltage, respectively. The voltage regulator tube RFT Str 90/40 (identical to Philips 90C1) was used in our experiments. Its electrodes are enclosed in a cylindrical glass envelope filled with a gas mixture. This kind of electron tube does not use a heater electrode. Accordingly to the datasheets  the nominal regulating voltage is 86–94V for discharge current varying in the range 1–40mA. Fig. 2. – The schematic diagram of the experimental setup. The electron tube has been placed inside a copper cylinder. The copper cylinder is made of 0.1 mm thick copper sheet, is 18.3 mm in diameter and 55 mm 388 O.S. Stoican 4 in length. Its inner diameter is equal to the electron tube outer diameter. The copper cylinder is sliding so that electronic tube can be partially covered. A supplementary resistor Ra is connected in series with the electron tube allowing the precise measurement of the oscillations period. A short voltage peak occurs across the terminals of the resistor Ra when the capacitor C is discharged through the electron tube. In this way the oscillations period may be accurate measured as the time interval between two consecutive voltage peaks. The values of the components R, C and Ra are 90.5kΩ, 470nF and 110Ω, respectively. The voltage E provided by a regulated power supply can be varied in the range 0–170V. The copper cylinder is connected or disconnected to the negative pole of the power supply using the switch S. 4. EXPERIMENTAL RESULTS The oscillation frequency of the relaxation oscillator has been measured in two cases. In the first one the copper cylinder is removed. In the second one the copper cylinder is connected to the ground (negative pole of the power supply). Fig. 3. – Waveform of the voltage Uc across the terminals of the voltage regulator tube when the cylinder is removed. In Fig. 3 and Fig. 4 are shown the waveforms of the voltage Uc in the two cases. In both cases the supply voltage E=125V. The waveforms have been recorded using a digital oscilloscope. As seen in Fig. 3 and Fig. 4, the oscillations frequency increases from 25.07 Hz to 27.79 Hz. This variation appears due to the decrease of threshold voltage U2. The frequency measurements have been verified using a digital frequency counter. The same results are obtained if the copper 5 Frequency shift of a relaxation oscillator 389 cylinder is disconnected from the negative pole of the power supply without the need to remove it. If the copper cylinder is connected to negative pole of the power supply but it is partially removed (~ 50%), the measured oscillations frequency was 26.63 Hz. The measurements on the frequency shift of the relaxation oscillator as a function of the supply voltage E are summarized in Table 1. The result shows that the frequency shift ∆f increases with operating frequency. Fig. 4. – Waveform of the voltage Uc across the terminals of the voltage regulator electron tube when the copper cylinder covers the whole glass envelope of the electron tube and it is connected to the negative pole of the power supply. Table 1 The operating frequency of the relaxation oscillator for various values of the supply voltage E Supply voltage Operating frequency Operating frequency Frequency shift when copper cylinder is when copper cylinder is removed connected to the negative pole of the power supply. E [V] f1[Hz] f2[Hz] ∆f=f2-f1 [Hz] 164.3 62.1 67.8 5.7 160 58.2 63.7 5.5 155 53.6 58.76 5.16 145 44.6 49 4.4 140 40 44.1 4.1 135 35.4 39.2 3.8 130 30.5 34.1 3.6 126 26.4 30 3.6 390 O.S. Stoican 6 5. CONCLUSION The experimental results suggest that the copper sheet acts as an additionally external electrode. When this electrode is proper biased the glow discharge breakdown voltage U2 in the gas filled electron tube decreases. In the usually conditions the relative variation ∆U2/U2 due to this electrode was found to be around 1.3%. This value is small and it is very difficult to be evidenced by measuring the voltage levels of the waveforms. Using a relaxation oscillator circuit the corresponding relative variation ∆f/f of the oscillation frequency is around 10%. This frequency shift can be easily measured using a digital frequency counter. The magnitude of the effect depends on the relative position electron tube-copper sheet and the relaxation oscillator operating frequency, respectively. These characteristics show that the external copper sheet is capacitively coupled with the internal electrodes of the electron tube. Further work will be devoted to evaluate more accurate the influence of the operating frequency on the effect. This work is done in the framework of the project PRET (2-Cex 06-11-7) and is supported by the National Authority for Scientific Research (ANCS)-Romania. REFERENCES 1. N.S. Kopeika, N.H. Fartiat, Video Detection of Millimeter Waves with Glow-Discharge Tubes, IEEE Transactions on Electron Devices, ED-22, 8, 534–548 (1975). 2. N.S. Kopeika et al., Commercial Glow-Discharge Tubes as Detectors of X-Band Radiation, IEEE Transactions on Microwave Theory and Technology, MTT-23, 843–846 (1975). 3. G.Y. Yan, K.I. Fujii, A.L. Schawlow, Relaxation-oscillator detection of optogalvanic spectra, Optics Letters, 15, 2, 142 (1990). 4. G.A. Petrucci, J.D. Winefordner, N. Omenetto, The relaxation oscillator as a resonance photon detector, Appl. Phys. B, 62, 5, 457–464 (1996). 5. D. Pavčić, D. Veža, Optovoltaic spectroscopy of a miniature neon discharge, Fizika A, 8, 3, 195– 204 (1999). 6. P. Mikoljczyk, Universal Vademecum, Electronic tubes and semiconductor elements, Panstwowe Wydawnictwa Techniczne, Warszawa, 1960, p. 1053.
Pages to are hidden for
"FREQUENCY SHIFT OF A RELAXATION OSCILLATOR DUE TO AN"Please download to view full document