Oscilloscope Requirements for My Project

Which product is right for your exact requirements
Post Reply
Natan_T
Newbie
Posts: 0
Joined: Thu Dec 06, 2012 1:01 pm

Oscilloscope Requirements for My Project

Post by Natan_T »

I am a graduate student doing an experiment in quantum cryptology. I will describe the experimental setup and then say what it is I am looking for.

I have a pulsed laser, operating at 10kHz (that's a 100usec period). Each pulse is 4.1 nsec wide and is later attenuated to achieve (theoretically) one photon per pulse that travels through an optical system. There are three different single-photon detectors in the system, and in a given laser period, at most one of them can fire. The detectors' output is a 2V-high, 15nsec-wide pulse. If necessary, I have a trigger output from my signal generator available, meaning a 10kHz square wave of amplitude 2.2V.

What I need to know is ONLY when each detector fired - not what the pulse shape was, not how high the pulse was, etc. Just the time coordinate. I do not need this in real time, meaning that the data transfer time to the computer is probably unimportant. The idea is to operate a predetermined number of laser pulses, get the data from the scope, and then analyze the results "offline".

I intend to do the analysis in MATLAB, so please tell me if that causes a problem.

ziko
Advanced User
Advanced User
Posts: 1705
Joined: Fri Dec 01, 2006 10:03 am
Location: St Neots

Re: Oscilloscope Requirements for My Project

Post by ziko »

Using Matlab is not a problem really, you can write your own script or you can save your data in PicoScope 6 as a binary file.

In terms of hardware specification wise, you mention a 4.1ns wide pulse, is this going to be measured or is it just the 15ns pulse?

Kind regards
Ziko

Technical Specialist

Natan_T
Newbie
Posts: 0
Joined: Thu Dec 06, 2012 1:01 pm

Re: Oscilloscope Requirements for My Project

Post by Natan_T »

What will be measured is only the detector output, meaning the 15-nsec pulse. Again, I only care about the timestamp of the pulse.

I have attached a simple block diagram of my system that should clarify things.
Attachments
diagram.doc
(26.5 KiB) Downloaded 452 times

ziko
Advanced User
Advanced User
Posts: 1705
Joined: Fri Dec 01, 2006 10:03 am
Location: St Neots

Re: Oscilloscope Requirements for My Project

Post by ziko »

Ok a 15ns pulse means to just see a signal the square wave you would normally be looking at something around 250MHz bandwidth to give you a few harmonics for your fast edges. Since you don't need the fast edges you will need a bandwidth that's perhaps at least 50MHz at least since you are not too bothered about the rise time.

Having a bandwidth of say 50MHz will make the waveform look like a sine wave. So really you need a bandwidth that is above 50MHz. Now in terms of your sample rate, as a general rule of thumb you need to sample 3-5 times higher than your signal if it is a sine wave and 10 times for a square wave (this also depends on the interpolation the software uses to draw the waveform, we use sinx/x interpolation )

So your sample rate would need to be maybe 150-200 MS/s, this will need to be done for 3 channels which means you would need a 4 channel device that can deliver this for each channel.

The 3400 A/B are four channels, have bandwidths between 60MHz and 200MHz a sample rate of 1GS/s for one channel (approximately 250MS/s per channel).

http://www.picotech.com/high-performanc ... copes.html

Key difference between As and Bs is that the A have a signal generator, whilst the B devices have a signal generator and an arbitrary waveform generator, also the memory is double.

The question is what time base are you going to be looking at? As this will dictate the total time you can view all your data and therefore how much memory you need.

Hope this helps.
Ziko

Technical Specialist

Post Reply