The use of double counters in photon counting to avoid dead-time

[benjathemin]

This is less of an issue with regards to the measurements and more of a "what's the idea behind it?" . In Hamamatsu's Photon Counting Unit, the C9945-13 series, they reference double counters to allow measurements with no dead-time. This wasn't so clear as I figured with the photon counting head and say a PCI6602 counter/timer card, Lab View and a setup similar to the one in this post attached:

https://forums.ni.com/t5/Counter-Timer/Photon-Counting-Time-Stamping-and-basic-help-with-how-DAQmx/t...

there isn't a clear need for 2 counters unless 1 counter is storing the number of events while the other facilitates the storage of the time tag. So my question boils down to what the use of the double counters is for with regards to dead-time in measurements?

[Kevin_Price · Accepted solution]

I'm not at all familiar with the Hamamatsu equipment, and don't know what they mean about "double-counters" and how it prevents "dead-time."   If you use an NI 6602, you should only need 1 counter to do lossless photon time-stamping as illustrated in the linked thread.  But bear in mind that high photon burst rates puts you at risk of a FIFO overflow error due to the very small FIFO on a 6602. 

Newer devices (6612 counter/timer, X-series MIO boards) made the counter FIFO's *much* bigger, and consequently support very much higher sample rates.  There would also be advantages of dedicated PCIe bandwidth vs smaller & shared PCI bandwidth, and (I think) 1 dedicated DMA channel per counter instead of 3 total for the whole device.

-Kevin P

[benjathemin]

thanks for the quick response.

will keep in mind the buffer size and the number of DMA channels when I further assess my setup.