carpenter
Full Member level 6
35 years ago, we designed a measuring device that actually measured the relative permittivity of the measured object. It is not a laboratory meter, but a thing that becomes part of the measured object and is very difficult to repair or change. Therefore, a low price and a long service life (tens of years) are required
The key part looked like this
- Base clock source is quartz 4MHz
-on U2A the clock is divided by 2 and an inverse 2MHz clock is created on Q/
- U2B with U1C and U1B shortens the positive pulse to 35ms (when U1 is ALS) or 100ns (when U1 is HC), The circuit with U3A works the same way
- the signal from U2B passes through a 0,5 or 1m long wire in the measured environment, the same inverse signal from U3B does not
- both signals are then compared to U3B via inverters U4A and U4B
- outputs Q and Q/ from U3B are inverted by Q4A and Q4E, then converted to an analog voltage at the R16/C4 and R11/C3.
When the signals are in phase, at 5V logic, the voltage of U1 and U2 is theoretically equal to 2.5V. the more the signal passing through the measured environment is delayed, the higher the voltage U1 and the lower the voltage U2. If we apply the voltages U1 and U2 to the instrument amplifier, we obtain a voltage proportional to the delay of the signal, and when we measure the temperature, we know how we can relatively accurately measure the properties of the measured object, manifested as relative permittivity.
For a signal, it is not the pulse length that is critical, but the steepness of the eising edge and the frequency,2MHz is the lower limit of usability 4 to 10MHz would be more appropriate.
Nothing prevents me from using the same today, or am I in some faster 74 logic, if it exists?
I also think of using some small IC with programmable logic.Use MCU signal to generate PWM and PWMn.
The problem is that I can't fully see all possible problems, so I ask the more experienced people if it's worth digging into it, and if so, in what direction?
The key part looked like this
- Base clock source is quartz 4MHz
-on U2A the clock is divided by 2 and an inverse 2MHz clock is created on Q/
- U2B with U1C and U1B shortens the positive pulse to 35ms (when U1 is ALS) or 100ns (when U1 is HC), The circuit with U3A works the same way
- the signal from U2B passes through a 0,5 or 1m long wire in the measured environment, the same inverse signal from U3B does not
- both signals are then compared to U3B via inverters U4A and U4B
- outputs Q and Q/ from U3B are inverted by Q4A and Q4E, then converted to an analog voltage at the R16/C4 and R11/C3.
When the signals are in phase, at 5V logic, the voltage of U1 and U2 is theoretically equal to 2.5V. the more the signal passing through the measured environment is delayed, the higher the voltage U1 and the lower the voltage U2. If we apply the voltages U1 and U2 to the instrument amplifier, we obtain a voltage proportional to the delay of the signal, and when we measure the temperature, we know how we can relatively accurately measure the properties of the measured object, manifested as relative permittivity.
For a signal, it is not the pulse length that is critical, but the steepness of the eising edge and the frequency,2MHz is the lower limit of usability 4 to 10MHz would be more appropriate.
Nothing prevents me from using the same today, or am I in some faster 74 logic, if it exists?
I also think of using some small IC with programmable logic.Use MCU signal to generate PWM and PWMn.
The problem is that I can't fully see all possible problems, so I ask the more experienced people if it's worth digging into it, and if so, in what direction?