Announcements
IMPORTANT: Cypress Developer Community is transitioning on October 20th. To learn more and be prepared for this change, check out our latest announcement.
cancel
Showing results for 
Search instead for 
Did you mean: 

PSoC 5, 3 & 1 MCU

Anonymous
Not applicable

hello dear colleagues,

   

i need to have a simple  solution for detecting pulses ( f=100Hz width 800µs).

   

 think of a monoflop with retrigger, but do not know how to implement.

   

must detect the absence of pulses from zerocross detector.

   

if pulses are detectet, normal operation

   

if pulses are absent, failure condition ( to be detected with 1-bit status register).

   

 

   

any ideas how to get this done the PSoC  way ? (schematic)

   

 

   

Chip:CY8C3866LTI-067

   

Creator: 2.2

   

best regards Markus

0 Likes
7 Replies
Anonymous
Not applicable
        Hi EFG, How about One Shot Pulse Generator?   
0 Likes
Anonymous
Not applicable
          
      

I think PWM with One shot mode is suitable for you.

      

Schematic goes like this.

      

And Edge detector, Deboucer and Glitch filter(?) are also available.

      

I think. 

      

 

      

Schematic

      

WiPry_Analizer

      

 

     
    
   
0 Likes
ETRO_SSN583
Esteemed Contributor

The rates and pulse period sufficiently slow you could do this entirely in code.

   

You have not stipulated decision latency.

   

 

   

Your tasks are -

   

 

   

1) Detect missing pulse, no input

   

2) Detect pulse too short/long, period OK

   

3) Detect pulse OK, but period is not

   

 

   

  You could use a counter that is enabled by input, when input returns low

   

look at counter value to determine if runt pulse, or if pulse too long.

   

 

   

A secondary timer/counter to implement period checking, to make

   

sure no pulse case is handled, as well as out of spec period for

   

pulse.

   

 

   

Cocde or HW can be used, consider a LUT to implement a state machine

   

along with counters/timers to measure pulse parameters.

   

 

   

Regards, Dana.

0 Likes
ETRO_SSN583
Esteemed Contributor

This reference has a one shot approach, but keep in mind one

   

shots notorius for poor timing accuracy.

   

 

   

www.ti.com/general/docs/lit/getliterature.tsp

   

 

   

You could implement a one shot with comparator and reference

   

to get more accurate timing.

   

 

   

Regards, Dana.

0 Likes
ETRO_SSN583
Esteemed Contributor

If you decide to take the state machine approach -

   

 

   

www.cypress.com/

   

 

   

Regards, Dana.

0 Likes
Anonymous
Not applicable

hello dear colleagues

   

many thanks for your fast responses and all the inspiration.

   

i did the PWM single shot solution and it works very well.

   

it really helped me out with no additional software overhead.

   

 

   

best regards Markus

   

(EFG is the name of the company i work for ;.)

0 Likes
Anonymous
Not applicable
        Very Good! However I have one anxiety. Your signal_in and Clock_2 is wired OR. It may ok, that is a test wired, will never got collision.   
0 Likes