Making a coincidence circuit with ns resolution

[qBot]

I need a LV-compatible coincidence counter with ns resolution (coincidence window width of about 5 ns or less is preferable). The input is a 5 ns width (FWHM) 1.2 V pulses from an avalanche photodiodes generated at about 10 kHz rate. I would like to know if I'm wright or wrong stating the following. Coincidence window is determined by the interchannel skew which is tipically about hundreds of picoseconds (500 ps for PCI-5105 and 100 ps for PCI-5152, for instance), so ns coincidence resolution is easily achievable.

And I would also like to know the "integration time"of the above mentiones PCI cards or the equivalent term to find in specifications. By saying integration time I mean the time it takes an ADC to aquire a sample (accumulate one data point). For my application it is better to have this integration time as low as possible (5 ns ideally), to lower the noise coincidences...

[Dan_K]

Hello qBot,

I took a look at your application, and although I am not very familiar with coincidence counters I decided to take a step back and treat this like any other signal. Correct me if I am wrong, but if I am reading your requirements correctly it looks like you need a digitizer capable of recording multiple 1.2V 5 ns signals over a period of time. Taking a look at the specifications of the NI 5152, it should be sufficient for your application, as it can support 300 MHz bandwidth (5 ns is 200 MHz, so this should work) and it has less than 2 ns of rise time on the input.

You may, however, want to take a look at the NI 5153 or NI 5154, which provide 500 MHz and 1 GHz front end bandwidth, respectively. Either of these cards would give you a better step response, allowing for a better shape of your input signal, helping to filter out any erroneous signals.

If you have any questions about these devices please let me know and I will help you to find the best digitizer for your needs.