Time between two pulses

Forum for discussing PicoScope version 6 (non-automotive version)
Post Reply
Posts: 0
Joined: Mon Jun 12, 2017 7:27 am

Time between two pulses

Post by deme » Mon Jun 12, 2017 8:00 am

I would measure, with a 3406 oscilloscope, the times between pulses (Th≈250 nS) that can happen randomly in a time range from 10 minutes to 1 hour.
The event (on trigger) capture function of the Picoscope software would have been perfect if it would include the absolute timestamp of each captured frame.
As this is not possible (at least on a .mat file) I thought to use the LabVIEW SDK application to overcame this issue.
Since I’m not an expert in programming I would ask your opinion on the following practice:
The idea is to stream data frames of 10uS sampled at 125nS time interval,
to queue these frames and apply a software trigger to classify and save on file only the frames that contain the pulse.
At the same time, knowing the order in the queue and the frame dimension, it will be possible to record for each saved frame, the precise relative time from the beginning of the recording.
Can I have a feedback?
Thanks for help
Best regards

Posts: 2855
Joined: Tue May 31, 2011 3:43 pm
Location: St. Neots, Cambridgeshire

Re: Time between two pulses

Post by Hitesh » Fri Jun 16, 2017 8:02 am

Hi Demetrio,

Which specific PicoScope 3406 model are you using - 3406A/B/D or D MSO?

You might be able to use the rapid block mode to capture only the pulses as long as there is a few microseconds between the pulses depending on the trigger re-arm time of the scope at that sampling rate.

You could potentially set the scope to output a single cycle of a waveform when the trigger occurs, but you would have to use another device to measure the time between the cycles to give a relative time.


Software Dev. Engineer

Advanced User
Advanced User
Posts: 51
Joined: Mon Nov 26, 2012 9:16 am
Location: Netherlands

Re: Time between two pulses

Post by bennog » Fri Jun 16, 2017 6:04 pm

if you do streaming mode in the SDK you could do 10Ms/sec (this will give you a 100nS per sample)
And depending on the memory or how you write the software you can measure this with 100nS accuracy.
If they appear once every hour you need to do real-time data processing on the sampled data.
1 hour would generate about 72GB of data so be prepared.


Post Reply