One Shot Circuit with Delay 1

[dherke]

Hello,
I have a project at work and would like some help by some more experienced people. The Circuit im working on is named computer data good line. The line is normally high(12V) and a group of negative pulses are put on this line with an interval of about 300 micro seconds. These pulse sometimes come out in groups of 3, and other times groups of 4. The pulses themselves are approximately 30 micro seconds width and separated by about 10 micro seconds.
I need to control the number of pulses by only having 1 pulse outputed on each group of pulses. The next group of pulses could be ignored also.
My question is : Would a one shot (monostable multivibrator) circuit be the best way to get only 1 pulse on the output. Im guessing i would need some sort of delay on the output as well to prevent the other pulses in the group from outputing. How would i implement this? using 555 timer maybe? would it be simpler just to use some flip flops? Thanks for any info.

 

[VEBill]

You might want to Google on 'Missing Pulse Detector' circuits (typically very simple with the 555). For your application, you'd simply have to turn the logic upside down.

 

[itsmoked]

A cmos 555 would work as VE1BLL and you've mentioned.

Is it the best solution? Depends on a lot of things. Quantity, specific other aspects of your situation. Since I don't know ALL the details I often suggest using something like a PIC processor.

In this case it could be a 6pin processor. Then you could handle any and all situations and generate a precise output depending on a small bit'O code.

 

[logbook]

An analog monostable using a 555 would work fine. Suppose you settle for one pulse per group of pulses. The last pulse of the train would be 120µs from the begining and the next train starts at 300µs. If you make a non-retriggerable monostable, which means the timing starts from a trigger pulse but further trigger pulses are ignored during the timing interval, you would aim for the mid-value (120+300)/2= 210µs. This allows maximum component variation without circuit malfunction (the caps woud typcially be ±10% and the resistors ±5% but you have ±40% range available).

 

[dherke]

I forgot to mention that i need to output only the first pulse in the group. Looks like a non-retriggerable monostable may be the way to go. If i go on the negative edge trigger, would this allow it to accept only the first pulse and then the timing would start from the first pulse. also on the output, i would like the pulse to look similar to the inputs first pulse(same pulse width and height). Is this possible while still ignoring the rest of the pulses in the group? I think a 555 timer may not be able to handle such a small pulse width.

I dont know if a missing pulse detector circuit would work because it would not be outputting the first pulse in the group.

I have a PIC programmer but have not written any programs for it. Is there a database on the internet of shareware PIC programs that engineers can use? Thanks.

 

[buzzp]

Try

http://www.geocities.com/SiliconValley/Way/5807/dat.html

AND

http://www.piclist.com/techref/piclist/index.htm

There was one site that had massive info including downloadable code. I can not recall the name as I have switched to Atmel. Good luck.