hello... this may seems a litle strange, but i need to convert some input that is 4Khz and 10% duty cycle to a clean DC


as you can see in the diagram, the pulses are 'low', so i want the output of this circuit to be 0 when there are puses detected, and 5v otherwize.. it's kind of a simple data holding...

i want to keep it simple with capacitors, i know there are many many many ICs to do the job. Please don't suggest any.

precision is not a problem, specialy about the time at which the output will swing from a state to another.

If it works, is there any danger the capacitor sends destructive currents through the diode D1?

Again, if the idea is correct, what or how to calculate the value of R1 and C1.


I case you want to know, the input of this circuit is actualy the ouput of a proximity sensor...

thanks in advance...

the pulses are 'low', so i want the output of this circuit to be 0 when there are puses detected, and 5v otherwize.

It won't be 0V. More like 0.7-0.8V depending on R,C and duty cycle.

If it works, is there any danger the capacitor sends destructive currents through the diode D1?

Yes.... To minimize this, put a resistor in series with the input line. You also want to limit the max current that the proximity sensor will have to supply. The series resistor will do this.

Again, if the idea is correct, what or how to calculate the value of R1 and C1.

It depends on many things. The frequency, input and output impedances, acceptable risetime on output, and the minimum duty cycle you wish to detect.

Thanks a lot!

Originally Posted by kchriste

It won't be 0V. More like 0.7-0.8V depending on R,C and duty cycle.

any idea how to make it less than 0.7 v?

Originally Posted by kchriste

Yes.... To minimize this, put a resistor in series with the input line. You also want to limit the max current that the proximity sensor will have to supply. The series resistor will do this.

It depends on many things. The frequency, input and output impedances, acceptable risetime on output, and the minimum duty cycle you wish to detect.

does any body has one of the simulators and run this circuit and see the optimal values of R and C....? that would be very nice... i don't have any

I suggest you try using a 555 timer as a 'missing pulse detector' - this will give a logic output, one way with the pulses, and the other way without the pulses - which I presume is what you're after?.

Originally Posted by Nigel Goodwin

I suggest you try using a 555 timer as a 'missing pulse detector' - this will give a logic output, one way with the pulses, and the other way without the pulses - which I presume is what you're after?.

Exactly! how didn't i think about that before ??!!

thx