Excess counts when photon counting

[chem_matt]

Hi Jaime,

Whoops, sorry, I've been having issues getting attachments to work properly on this site.  I've tried again, and also posted the oscilloscope trace on an external site:

http://www.usask.ca/chemistry/groups/paige/NI_correspondence/TDS1k2k_FP.png

The yellow channel shows the Labview generated pulse train, and the blue channel shows the pulses generated by the APD (obviously much shorter).  Based on this trace, I'd expect to count around ten edges over the course of a 1 second  interval.

In terms of software, I've just been using the Count Digital Events.vi in the "examples" or just running through the test panels in the Measurement and Automation explorer.

I'm not too sure how relevant this is, but just for fun, I routed a Labview generated pulse train from ctr0 out into A0 and measured it using the test panels. I get signals jumping from 0.00V to 5.0V as expected.  Doing the same for the APD shows jumps from ~ 0.14V to higher values (though you don't see the 4V high signal, presumably 'cause the pulses are too short).    Not sure what the source of this offset is, but I almost wonder if fluctuations above this are registering as counts.  This could relate to a grounding issue.

Anyways, thanks so much for your persistence, I really do appreciate it, and hope you have some further ideas!

- Matt.

[hoffiz]

Hi,

Indeed it seems to be a ground issue from your last response and test.  Can you please take a look at this article: Measuring Floating (No referenced) Sources, here you can use the table 1 diagram and follow it to make the same test with the “Analog Input”, using a differential measurement with a floating signal.

I guess the blue signal on the scope sceen shot is the APD, right?

I have attached a screen shoot with something we can try. I’m not completely sure if it will work since your APD signal is something that I cannot simulate in my end. But it won’t harm trying it on your end. Try adding the property node “Counter Input:Count Edges: Digital Filter:Minimum Pulse Width Property” that will allow you to specifies in seconds the minimum pulse width the filter recognizes.

I hope it helps

Message Edited by Jaime F on 02-11-2008 04:31 PM

[chem_matt]

Hi Jaime,

I reckon we've got it!  Our electronics guys put together a simple adjustable voltage divider and connected it to the detector.  Essentially we just reduce the size of both the background fluctuations and the TTL pulses.  If you tweak the trim pot on the voltage divider, you can adjust the voltages such that the noise doesn't trigger false counts but the real pulses are still TTL compliant and give real counts. 

It took some work, but everything's a go!  Again, thanks for all of your help - you're aces!

- Matt.