Period Measurement Vs. Edge-counting

[Eyal_S.]

Hi all,

I used to have the pci-6602 and now I have the pci-6251. 

When I use 1 counter for period measurements the software crashed when a high frequency burst arrives. This seems to be a common problem when using an avalanche photon diode. This has been widely discussed in this forum and from my understanding it has to do with an on board FIFO. It is also my understanding that when using daq instead of daqmx the software does not crash. Unfortunately, Labview 8 does not support daq.

If instead of period measurements I count photons per sampling interval (which I use the 80 MHz clock for) this problem seems to be fixed.

I don't understand why counting photons per sampling interval would "fix" the problem. What happens to the high frequency bursts in this case? 

Thanks,

Eyal

[Broken_Arrow]

I feel your pain, I've had a similar issue. It is my understanging that the sampling period gives the system time to compute the signals since it has to gather the period first before knowing what's in it. When sampling a duty cycle, it's quasi-real time and there's no wiggle room for sudden changes that cause overruns.

BTW, you can use Traditional DAQ on 8, it's just not supported by NI, but it does work. Worked last week.

[Eyal_S.]

Now it seems like I can not even use a 10MHz clock for binning. I get the following error:

Error -200141 occurred at DAQmx Read (Counter 1D U32 1Chan NSamp).vi:1

Possible reason(s):

Data was overwritten before it could be read by the system.

If Data Transfer Mechanism is Interrupts, try using DMA. Otherwise, divide the input signal before taking the measurement.

I thought I used the 80 MHz clock a few years ago (on a different setup) to do the binning but maybe my memory is playing tricks on me.

I wonder what would be the best way to do time stamping. I tried to use period measurements with 1 counter after I filtered out bursts within 300ns  and the program still crashes every once in a while (which I can't accept).

I need to measure time stamps but it seems like I can't solve the problem of two quick pulses coming one after the other. 

Richard, did you ever try measuring the dead time of the card when using daq? Because I'm worried that even though the software won't crash I will miss a lot of data.

[Jordan F]

Hi Eyal!

According to the specifications of the 6251, the maximum source frequency is 80MHz.  This means that the counter can count pulses that are 12.5ns (1/80e-6) apart.  I have a few questions on what you are trying to accomplish:

What is the input frequency range to be measured?

Are you using a shipped example (or modified version) in LabVIEW?

What counter task are you doing? Period, event, etc.?

How many samples are you reading at a time?

In regards to your error (-200141), the article linked below explains a little more on how buffered counting works:

Difference Between Finite and Continuous Buffer Mode for Counter Operations?

 

Cheers!

Jordan Fink