[SOLVED] Detect breaks between TTL pulse trains

[hmalissa]

Hello,

I'm looking at a TTL trigger output that sends trains of 100ns pulses with a separation of down to 500us. After the pulse train has ended, there is a break of a up to a few seconds, then the next pulse train comes.
I want to detect the breaks between the pulse trains. Unfortunately the number of pulses in the pulse train is not know exactly, and it can vary from on train to the next, so counting pulses is not an option.
But since the separation of pulses within a train and the duration of the break between trains are on rather different time scales, I was wondering if there is an easy way of detecting the breaks. The separation between the pulses is 500us, or longer, depending on my experiment. And the duration of the break can be arbitrarily long (right now I've set it to 20 seconds, and I detect the breaks by looking at a scope).
Unfortunately I'm not very good at designing circuits, so I don't even know where to start. I would be very glad if someone could point out to me how such a circuit might look like.

Thank you very much!

 

[goldsmith]

I'm looking at a TTL trigger output that sends trains of 100ns pulses with a separation of down to 500us. After the pulse train has ended, there is a break of a up to a few seconds, then the next pulse train comes.

Hi Hamalisa
What is your idea about an MCU for this aim ? i think that may be suitable .

I want to detect the breaks between the pulse trains.

Have those pulse trains any discipline ? ( breaks ? ) or perhaps their changes are random ?
Best Wishes
Goldsmith

 

[bepobalote]

use a monostable which is triggered by the pulse train.
Pulse width of this monostable must be larger of 2 of your pulses.
You can find a similar circuit on some NE555 datasheets.
Just as an example, take a look at: https://www.diyelectronicsprojects.com/2012/07/missing-pulse-detector-circuit-using.html

 

[hmalissa]

use a monostable which is triggered by the pulse train.
Pulse width of this monostable must be larger of 2 of your pulses.
You can find a similar circuit on some NE555 datasheets.
Just as an example, take a look at: https://www.diyelectronicsprojects.com/2012/07/missing-pulse-detector-circuit-using.html

That's perfect, thanks a lot. That's exactly what I need. I can feed it to the PC using the parallel port.

- - - Updated - - -

What is your idea about an MCU for this aim ? i think that may be suitable .

I think that would be overkill. I'll try the 555-based solution first. But thanks a lot, anyway.

 

[goldsmith]

I think that would be overkill. I'll try the 555-based solution first. But thanks a lot, anyway.

Hi again
I thought you need high value of precision . but if you're not looking for high precision , action yes that method would be fine . ( 555 based solution ! )
Good luck
Goldsmith

 

[hmalissa]

Hi again
I thought you need high value of precision . but if you're not looking for high precision , action yes that method would be fine . ( 555 based solution ! )
Good luck
Goldsmith

Well, yes. But the timescales within the pulse train and in between the pulse trains are so different. I can set the timeout of the missing pulse detector much longer than the default shot repetition time, and still much shorter than the delay between the pulse trains. I think that should work.
Thanks a lot!