Regarding Digital Input Design

[cyber_dev]

Hi all,

I am designing a digital input circuit. It should accept the input state as logic high from 4V to 40V. I will also use the same circuit to measure the frequency of input signal (PWM). The frequency to be measured is up to 25kHz. The input will have a pull down or pull up resistor optionally. I need your suggestions regarding the circuits shown below.




For the first circuit with schotky diodes,

datasheet of diode

1-) What is the path for reverse leakage current of the diodes? The datasheet says that Ir = 100 nA max. @25°C and it is about 30 uA @125°C. Assume that there is a 100K pull down resistor at input. If no external signal is applied to input, what would be voltage at input for the worst case? Would the reverse current generate a voltage across the pull down resistor? If so how much?

2-) Replace the resistor (33K) (which is in series with diodes) with a resistor that have value of 1M. What would change? (The question is the same as above.)

For the second circuit with zener diode,

1-) The zener diode capacitance is given 450 pF @25°C in the datasheet. With a low impedance (<1K) source, can a 25 kHz signal's frequency be measured correctly even if @125°C temperature?

In general, which circuit above do you suggest me to use? Is there a need for modification? Thanks in advance.

Best Regards.

 

[yogece]

I interpret your requirements,
you want to design a digital interface circuit (3.3V TTL) the digital output should be logic 1(3.3V) for the input voltage range from 4V to 40V & <4V you want convert to logic 0(0V) and you want to ,measure the frequency of the digitized signal.

Donot use your circuit because you said the junction capacitance of zener diode is 450 pF @25°C .

why do you use pull-up and pull-down resistors in your circuit and 220PF capacitor at the output of 74HC14?


Here my suggestions for new design,

 

[cyber_dev]

Thanks for your reply.

Indeed any voltage above the schmitt trigger high input level should be interpreted as logic high. The 4 to 40V range is guranteed to be logic high.

I designed a similar circuit to your suggestion to measure frequency even in an amplitude level 50mV. But the circuit for digital input should be simpler and less complicated.

I have tried the first circuit, which is with shotky diodes, without pull down/up resistor. All the resistor were the value of 1M. It worked well even with a 30 kHz signal. But I am not sure how it behaves with a pull up/down resistor and at high temperature for the worst case. And another point to consider is how much it will degrade the performance of analog inputs when clamping occur. I have one power supply; AGND and DGND is separated with ferrite bead. AVDD and DVDD is also.

The capacitor at output of schmitt is used to remove the ringing edges. It works as expected, I have seen it on scope.

Need more suggestions...