The calibration of the instrument coefficient of a traditional pulse-type flow sensor is generally accomplished by a volumetric flow standard device and a counter and timer, as shown in Figure 1. The instrument coefficient is defined as the number of pulses emitted by the sensor when a unit volume of fluid passes through it, typically in units of 1/L (pulses per liter) or 1/m³ (pulses per cubic meter). According to the provisions of reference [1], to ensure the effectiveness of the flow meter's instrument coefficient, the absolute value of the relative error of the number of pulses output by the flow meter in a single calibration should generally not exceed 1/3 of the repeatability of the flow meter under test. Since the counting error of a typical counter is ±1 pulse, the counter should collect a sufficient number of pulses N within the calibration time interval (the time t between two control pulses in Figure 1) to achieve the required calibration accuracy. For some large-diameter flow sensors, their instrument coefficients are generally small (e.g., a 200mm turbine flow meter has an instrument coefficient of only 1.5L⁻¹; a vortex flow meter has an even lower instrument coefficient, only about 0.2/L⁻¹). For such flowmeters, collecting a sufficient number of pulses requires both a long calibration time and a large calibration device (a large standard container). Due to these limitations, the required number of counted pulses is often insufficient. Dual-time counting technology is a widely used pulse interpolation technique internationally. It allows for sufficient technical accuracy with a smaller calibration device within a shorter calibration time, and it was initially applied to miniature volumetric tube flow standard devices. Our developed "Pulse-type Flow Sensor Calibrator" is a dual-time method calibrator composed of a traditional counter and dual-time measurement and control technology. Experiments show that this calibrator is convenient and reliable, shortens calibration time, allows calibration of larger diameter flow sensors with a smaller standard container, and achieves higher technical accuracy than conventional calibration methods. 1. Dual-Time Counter Principle: Pulse interpolation technology is an effective way to increase the output signal resolution of the flowmeter in a piston calibration device, thereby reducing the size of the calibration device. To obtain a sufficient number of pulses when calibrating a flow meter, two approaches can be taken: first, increase the output signal resolution of the flow meter to obtain as many pulses as possible within a limited calibration time; second, increase the effective volume of the calibration device. Generally, the number of pulses output by a flow meter per unit volume of fluid is limited (such as the turbine flow meter and vortex flow meter mentioned above), and the effective volume of the calibration device cannot be made very large. Pulse interpolation technology solves this problem well, and there are several methods available, such as the dual-time method, the four-time method, and the phase-locked loop method. By using a "small volume" (effective volume of the device) to collect 500 pulses, the piston calibration device can achieve the same accuracy as a large-volume calibration device collecting 10,000 pulses. [align=center] Figure 1 Instrument coefficient calibration diagram[/align] The principle of the dual-time method is shown in Figure 2a. Under the condition that the flow pulse signal period is stable, the pulse interpolation number is: Under the condition that the stability of the flow standard device meets the standard regulations, the flow pulse signal period can be considered stable, so the pulse interpolation number obtained by formula (1) should be valid. Besides the dual-time method, the four-time method can also be used to determine the number of pulse interpolations. The four-time method measures four times t1 to t4, as shown in Figure 2b. The number of pulse interpolations is... This paper takes the dual-time method as an example to design a pulse-type flow sensor calibrator. 2 Hardware Design of the Calibrator The hardware principle block diagram of the pulse-type flow sensor calibrator is shown in Figure 3. [align=center] Figure 3 Hardware principle block diagram of the dual-time flow calibrator[/align] This calibrator does not use a microprocessor and has good reliability. The control signal can be selected using a single-pole double-throw switch K1 to select a very narrow pulse signal or a level signal. When using level signal control, switch K2 can be used to select high-level control or low-level control. When the control signal is a pulse signal (the first control signal in Figure 3), switch K1 selects pulse control. Assuming the Q output of the initial state trigger TR1 is low level L (it doesn't matter if the output is high level H), the high level H is fed back to the D terminal. Switch K2 selects high-level control or low-level control. When the control signal is a pulse signal (the first control signal in Figure 3), switch K1 selects pulse control. Assuming the initial state trigger TR1's Q output is low (L) (it doesn't matter if the output is high (H)), the high-level output H is fed back to the D terminal. Switch K2 selects high-level control (if the initial state trigger TR1's Q output is high (H), K2 can select low-level control). The inputs of NAND gates B and C, and the D terminal of trigger TR2 are all low. Therefore, gates B and C are closed, and the Q output of trigger TR2 will also be low under the influence of the flow pulse signal, thus closing gate E. Counters and timers T1 and T2 are both in the stopped state. The clear button can be used to return the counters and timers to their initial zero state, displaying all zeros. When the "start counting" control signal pulse (the first control pulse) arrives, since the D terminal of TR1 is high (H), the control pulse triggers TR1, causing its Q output to be high (H), and immediately opens NAND gates B and C, causing the counters and timers T1 to start counting and timing. At this time, NAND gate E is not yet open, but the D terminal of flip-flop TR2 is already at a high level. The first rising edge of the flow signal after the leading edge of the control signal triggers TR2, causing its Q terminal to output a high level and opening NAND gate E. Timer T2 also starts counting. When the "stop counting" control signal pulse (the second control pulse) arrives, TR1 is triggered again, causing its Q terminal to output a low level L, thereby immediately closing NAND gates B and C, and stopping the counter and timer T1 from counting and timing. However, NAND gate E is not immediately closed, but only the first rising edge of the flow signal after the leading edge of the "stop counting" control signal pulse can trigger TR2 to output a low level L, closing NAND gate E and stopping timer T2 from timing. Substituting the data obtained from timers T1 and T2 into equation (1) will yield a more accurate pulse interpolation number. When the control signal is a level signal (the second and third control signals in Figure 3), switch K1 selects level control, which is equivalent to bypassing flip-flop TR1 and directly controlling NAND gates B and C and the D terminal of flip-flop TR2. Switch K2 is selected to control either high or low level operation. The remaining actions are exactly the same as pulse control. 3. Specifications and Results 3.1 Test Instrument Specifications In addition to functioning as a controllable counter and timer, this test instrument also has the function of measuring signal frequency and period. It is not used for calibrating flow meters but can be used independently as an instrument for measuring signal frequency or period. Specific specifications are as follows: ① Timer: 6-digit LED display, resolution 1ms; ② Timer (including frequency and period): 8-digit LED display, maximum resolution is 1Hz frequency, 0.1 period, count ±1 pulse; ③ Measurement range: frequency 10Hz～100MHz, period 0.5～10s, count capacity 99,999,999, timer 1ms～999.999s; ④ Manual switching between T1 and T2 displays. 3.2 Test Results ① The periodic square wave signal output by the 51 series microprocessor was used as the standard for frequency and period measurement. The results are listed in Table 1. [align=center]Table 1 Verification results using periodic square wave as standard[/align] ② Verification using a standard frequency signal generator, results are listed in Table 2. [align=center]Table 2 Verification results using a standard frequency signal generator[/align] ③ This calibrator was used to verify the instrument coefficient of the turbine flow meter, and achieved good results.