Unijunction NE555 Capacitor Meter

Theory 

Since the readout is a dc meter, it is desirable that there be a linear relationship between the capacitance to be measured and the dc output of the measuring circuit. It so happens that the average dc value of a pulse waveform has a direct linear relationship to the duty cycle of the waveform. This is illustrated in fig. 1. 

Duty cycle is defined here as the ratio of pulse width (in units of time) to the time between the beginning of each pulse. 

In fig. 1, T I is the pulse width, and T2 is the period (reciprocal of frequency) of the waveform. The pulses in fig. 1 have a peak amplitude of 10 volts; therefore, when the duty cycle is 0.2, the dc value of the waveform i s 0.2 times 10 volts, or 2 volts. In like manner, when the duty cycle is 0.5, the dc value of the waveform is 5 volts, and when the duty cycle is 0.8, the dc value is 8 volts. A general expression for the dc value of the waveform may be written as :

dc value =(T1/T2)*Ep           (eq.1)

where Ep, is the peak amplitude of the pulse it; volts. If T2 and E, can be held constant, then the dc value will be directly proportional to T I , the pulse width. 

T2 can be held constant by making the pulse train have a constant frequency, and Ep can be made constant by regulating the power supply voltage to the pulse-forming circuit.

This constant-voltage requirement makes an ac-operated power supply preferable to battery operation. Now, if the width of the pulse ( T I ) can be made directly proportional to the value of the capacitor under test, the problem is solved. Fortunately, a monostable multivibrator (one-shot)can be designed so that its pulse width is directly proportional to its timing capacitor, and this completes the idea for the system. Fig. 2 shows a block diagram of how such a system may be built. The trigger source is simply a free-running pulse generator which has a constant frequency and produces a narrow negative output pulse. 

Each time a trigger pulse occurs, the one-shot multivibrator initiates an output pulse whose width is determined by the capacitor-under-test. The larger the capacitor, the wider the pulse. 

Since a dc meter reads the average value of the pulse waveform, the meter may be calibrated directly to read capacitance. Care must be taken, however, that the pulse width does not exceed the time between trigger pulses. Also, the frequency of the trigger source must be high enough to prevent jitter of the meter needle.

Circuit

In the complete schematic the trigger source consists of a programmable unijunction transistor, Q1, an inverter-amplifier, Q2, and their associated components. There is nothing particularly critical about this part of the circuit, but the output trigger pulse at the collector of Q2 should have a frequency pretty close to 500 Hz, and the pulse amplitude should be about 12 volts (the power supply voltage). 

The pulse width of the trigger is about one microsecond in this circuit. Any trigger circuit which will provide this output may be used. 

A Signetics NE555V timer IC is used as the one-shot multivibrator. According to the data sheet, the output pulse width of this device is given by:

PW = 1.1RC              (eq.2)

where PW is the pulse width (the same as T1 in eq. 1), R is the timing resistor (selected by the range switch) and C is the capacitance under test.

Substituting into eq. 1:

dc value = 1.1RCEp/T2             (eq.3)

 With the range switch in any one position, all of the terms on the right-hand side of eq. 3 are constant except C, the unknown capacitor; therefore, the dc value of the output voltage has a direct linear relationship to the capacitance being measured. 

The resistors associated with the range switch ( 1k through 10 meg) should have a tolerance of 5 percent or less. A 10k trimpot in series with the meter is used to make a one-time calibration; once adjusted, it needs no further attention. 

It was necessary to include a front panel zero-adjustment pot for the lower capacitance ranges. This is because the input capacitance of pins 6 and 7 of the NE555V and stray capacitance of circuit wiring is about 25 pF. This produces an output pulse even when there is no capacitor connected to the test terminals. 

Without the zero adjustment pot to buck out the voltage from this pulse, the meter will read 25 percent of full scale when the range switch is in the 100 pF position with no capacitor connected to the test terminals. The zero adjustment circuit must have a relatively low resistance so that variations in its setting will have no appreciable effect on calibration. 

Since the zero setting circuit consists of a 470-ohm resistor and a 100-ohm pot in series across the 12-volt supply, it draws about 20 mA. This is another reason battery operation was ruled out. 

The simple zener-regulated power supply provides 12 volts at up to 50 mA is needed to power the circuit.