Digital frequency meter using 3310 LCD

Frequency meter is one of the experimental instruments that we often use. This experiment uses a microcontroller, counting circuit and liquid crystal device to design a broadband frequency meter. It is expected to achieve accurate frequency measurement in the 10Hz-1.1G range. 2) Preamplifier and crossover circuit: Does anyone have a good circuit? (Note: This experiment is in the improvement stage, and this circuit diagram will be provided later). Design idea The measurement of frequency is actually counting the signal within 1S, and the count value is the signal frequency. There are usually two methods used to design a frequency meter using a microcontroller. 1) Use the counter that comes with the microcontroller to count the input pulses, or measure the period of the signal; 2) Use a counter external to the microcontroller to count the pulse signal, and the count value is then read by the microcontroller. . Since the frequency of the input clock of the counter of the microcontroller is usually only a fraction or even a few tenths of the system clock frequency, direct measurement of the signal frequency using the counter of the microcontroller is greatly restricted. This experimental circuit adopts mode 2, using a 74LS393 four-bit dual binary counter and the T1 counter of Atmega8 to form a 24-bit counter. The maximum count value is 16777215. If the input signal is divided by 64 by the MB501 frequency divider before measurement, the maximum measurement frequency under a fixed 1S time base is 1073.741760Mhz. In order to easily obtain an accurate 1-second measurement gate signal, we used the asynchronous real-time clock function of Atmega8 and used a 32.768Khz crystal oscillator to generate a 1-second timing signal from TC2. Measurement principle: The microcontroller opens the measurement gate, that is, PB1 outputs high level, and the TC2 timer starts at the same time. The 74LS393 starts to count the input pulses. Every time the 74LS393 counts up to 256, the T1 counter of the Atmega8 also counts up once. When the 1S timing arrives, the microcontroller generates an interrupt, PB1 outputs a low level to close the measurement gate, and then Atmega8 reads the count values ​​of 74LS393 and T1, and then sends them to the LCD for display. Since the measurement gate time of 1S is difficult to test under amateur conditions, the real-time clock is simultaneously displayed on the LCD in the experimental program to judge the accuracy of the 1S gate time. In the experiment, I used the precise time of the GPS satellite displayed on the CDMA mobile phone for comparison. The smallest unit of time displayed on the mobile phone is minutes. Once the minute value of the mobile phone jumps during measurement, the second value displayed on the LCD will be immediately recorded. In this case, after the frequency meter has been running for a period of time, the second value displayed on the LCD will be recorded several times. It can be accurately judged whether the asynchronous clock of the frequency meter is accurate. During the experiment, I let the frequency meter go about 10 decimal places, and the measured 1S clock was still very accurate. Experimental question: The measurement result of the experimental test of 8M active crystal oscillator is shown in the figure above, while the measurement result of the 4M ordinary crystal oscillator test is 3.999932MHz. It feels like the measurement is slightly on the low side. Experiment source program: app_using_3100_LCD_frequency_meter_c_code_1.rar