Count Event Rate Below Threshold

Post discussions on projects you are working on
Post Reply
xrl
Newbie
Posts: 0
Joined: Wed Jul 11, 2018 11:14 am

Count Event Rate Below Threshold

Post by xrl »

Hi All,

I'm trying to measure the dark count rate of some photomultipliers using a picoscope 3206D. Typical signal is pulse above 6 mv with a width of 20 ns. Is there a way to set the picoscope with 10 ms/div, sample interval 1 ns and trigger set to none and just get the number of counts which go above the 6 mv threshold?

I have been reading out individual waveforms and using a separate analysis, but this takes a long time as the data files are rather large, it would be good just to get the net counts which went above 6 mv per waveform.

Thank you!

Martyn
Site Admin
Site Admin
Posts: 4491
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

Re: Count Event Rate Below Threshold

Post by Martyn »

Without seeing the actual waveform that is produced I think you need to look at Measurements where you will find Edge Count, Falling Edge Count and Rising Edge Count.

Give these a try and then if they do not give you the information you need post a psdata file and we can take a look at other options.
Martyn
Technical Support Manager

sbabic44
User
User
Posts: 3
Joined: Thu Nov 30, 2017 10:17 pm

Re: Count Event Rate Below Threshold

Post by sbabic44 »

This sounds like a perfect candidate for rapid acquisition mode + DeepMeasure...

Gerry
PICO STAFF
PICO STAFF
Posts: 1145
Joined: Mon Aug 11, 2014 11:14 am

Re: Count Event Rate Below Threshold

Post by Gerry »

I answered the same query on our help desk, so I'm posting a revised answer here, for the benefit of other forum users:

It is possible to set the Timebase to 10ms/div and the Number of Samples to maximum to give you a sample interval of 1ns, and use these values to capture 20ns wide pulses. As mentioned by Martyn, you can then use a rising edge count Measurement to count the pulses. If you only have 20ns pulse width pulses (i.e. you don't have to differentiate between 20ns pulses, and say 40ns pulses, for example) then you can isolate the ones larger than 6mV by setting the lower threshold for the measurement to 6mV, and setting the upper threshold to a value just above that. In order to count the edges measurements need to work on the data in a buffer, so you need to fill a buffer, or buffers by using a triggered capture.
The only issue with performing a Standard Measurement is that the number of edges that can be counted is limited, so if the dark count rate is more than a few hundred pulses for the Total Capture Time that you are using, you would have to reduce the Timebase setting, otherwise the rising edge count won’t return a value. You would then also need to use Rapid Trigger Mode, if you want to minimise the data loss between buffers. If you have a long time period over which you want to perform the count then this could mean having to flick through a large number of buffers of data (using the waveform overview) and noting the pulse count down for each one to sum at the end.

So, an alternative method would be to use Deep Measure, as mentioned by, sbabic44, to perform a set of measurements, and then isolate pulses over 6mV, by setting a threshold value so that you can just look for overshoots.
The only issue with this method is that as you can only re-arrange columns (you can’t count) you would need to export the data to a CSV file (or files) to be able to use Excel to perform the counting.

So the method you use is going to depend upon how many pulses you need to count in total, which will be determined by the amount of time over which you need to perform the capture along with the dark count rate over that time. To give you an idea of what to expect, I have created an example using pulses (a square waveform) from the Signal Generator, here: https://drive.google.com/file/d/1Bl1HnG ... sp=sharing. I filled the 512MS buffer of the PicoScope 3206D MSO by using 53 buffers to capturee waveforms in Rapid Trigger Mode. I then performed a Standard Rising Edge Count using the Channel signal level rulers to specify the upper and lower thresholds. Because there are a high proportion of pulses for the time period of one buffer, I have had to take the Measurement between Time rulers, so in this case the Timebase should be reduced and more waveforms captured. Even without reducing the Timebase, Deep Measure would be a better solution here, however if you have considerably less pulses to count, and can use just one 100ms buffer or a few buffers then Standard measurements would be the way to go.

Regards,

Gerry
Gerry
Technical Specialist

Post Reply