# Missing pulse detector

#### helloho

Joined Mar 10, 2011
6
Hi all,

I am designing a missing pulse detector.
below is the schematic.

The incoming signal is about 950Hz. The pulse width is about 600uS and the pause between pulse is 480uS.

As i know the width of this output pulse is defined by the values of R2 and C1. R2 and C1 must be chosen such that the output pulse width is slightly greater than the time between each incoming pulse.

So for my case, R2 and C1 must be adjusted to be more than 600uS?

anyone can suggest a suitable R1 an C1 for me.

Thanks

#### Wendy

Joined Mar 24, 2008
23,396

Interesting circuit. It is triggered when the pulse goes low, so if a pulse is not sent in around 1.05ms then it triggers, the period of the 950Hz signal. Figure this as the approximate time duration of the RC circuit.

T = 1.1 R C

So if the resistor is 10KΩ, C1 should be around 0.1µF. C1 could be around half that, but not reliably.

Here is the schematic I would use...

BTW, your math does not add. Given the pulse on (600µs) and pulse off (480µs) your period is 1080µs. The frequency for this setup is 926Hz.

Last edited:

#### SgtWookie

Joined Jul 17, 2007
22,230
It IS an interesting little circuit.

But, some things were bugging me about both of them, so decided to simulate it.

Both of the previously posted circuits have failure modes in that one won't detect a missing pulse if the input stays high (PNP version), and the other won't detect a missing pulse if the input stays low (NPN version).

The first version used a 1 MEG pot with no resistor in series with it. If the pot was accidently run all the way to one end, thus giving nearly 0 Ohms resistance, either the pot or the 555 timer would produce smoke.

The attached version triggers on the falling edge of the input signal via a cap; the signal needs to have a fairly short fall time. 1N4148/1N914 diodes clamp the trigger input to the power rails. The advantage these three addition components provide is that if the input signal gets "stuck" either high or low, the missing pulse failure will still be detected.

I also added a 1k Ohm pull-up resistor on the output. This allows the output to get reasonably close to the positive rail when using bjt 555 timers; otherwise you might only get Vcc-1.2v out of it.

#### Attachments

• 64.1 KB Views: 279

#### helloho

Joined Mar 10, 2011
6
Hi SgtWooki,

So you recommend R1=150K and C1=10nF for my case?

Thanks

#### Ron H

Joined Apr 14, 2005
7,061
Be aware that this circuit will probably not work with Vcc>≈12V, due to Vbe breakdown, which is not modeled in the standard models.
I simulated it with a 2N3906bd subcircuit that I made. It includes Vbe breakdown=6V. The circuit failed when Vcc rose above ≈11V.
If anyone is interested, I can post the subckts for 2N3904bd and 2N3906bd.

#### Wendy

Joined Mar 24, 2008
23,396
You can put a simple capacitor in front of the schematic I showed and it will work. I don't think it will have the breakdown problem Ron mentioned either.

Wookie made a good call, if R2 is adjusted to 0Ω the 555 is toast. With these kinds of circuits a second resistor to prevent this condition is pretty much expected.

#### helloho

Joined Mar 10, 2011
6
Hi Guys,

I had constructed the circuit suggested by Wookie and checked the output using a scope.

Above picture show when the beam is received. (high stage)

above picture show when beam is broken(missing pulse) = low stage

This picture is when i place the transmitter and receiver further apart.
I observe that the signal is more unstable. (more noise)
That will affect the performance?

Ps:ignore the voltage shows in the scope.

thanks

#### hgmjr

Joined Jan 28, 2005
9,027
Can you provide the entire schematic so that we can see what your transmitter and receiver circuits look like?

hgmjr

#### SgtWookie

Joined Jul 17, 2007
22,230
I did not show the required supply bypass capacitors across the 555 timer Vcc/GND pins in my schematic; they aren't needed in the simulation, but are required for real components. See this thread: Decoupling or Bypass Capacitors, Why?

You will need a minimum of a 0.1uF ceramic or metal poly film and a 1uF or larger aluminum electrolytic to keep the circuit stable.

It would help if you could display the input signal along with the 555 out signal.

If you keep getting the "glitches" after adding the 0.1uF and 1uF+ caps across the Vcc/GND pins of the 555, then you may need to increase the size of R2 somewhat. Your 10nF cap may be somewhat higher than 10nF.

Last edited: