Real Time Enforcement

[bbentelin]

Recently my lab switched to using a 2GHz sampler (PXI-5152) from a 100MHz sampler (PXI-5122).  The changes to the labview programming were trivial to make the 5152 work so I am intreagued by an anomoly in the data.

We have a radioactive source raining down particles through a diamond sample and onto a scintillator.  The scintillator triggers the acquisition of data from the board the diamond is attached to.  Previously we would get a 'false trigger' every once in a while, say 10% of the time.  Background particles would sometimes hit the scintillator to trigger it, but as the particle did not pass through the diamond, the signal out of the sample is almost non-existent. 

But after chaning to the 5152 I haven't seen any false triggers.  The program runs much slower when sampling at 2GHz, but I don't understand how this would affect the false triggers.  It shouldn't have acquired some filtering technique in the change.  As I said before the block diagram changes are trivial, just reasource name channel etc. except that I had to change the value on niScope Configure Horizontal Timing's enforce realtime to false.  I've never used that before, and I don't quite know what it means but it makes the program work.  Could this make an impact?

Thanks for your imput,

Ben

[bbentelin]

Sorry, my spelling mistakes are bothering me, here is the post again:

Recently my lab switched to using a 2GHz sampler (PXI-5152) from a 100MHz sampler (PXI-5122).  The changes to the labview programming were trivial to make the 5152 work so I am intrigued by an anomaly in the data.

We have a radioactive source raining down particles through a diamond sample and onto a scintillator.  The scintillator triggers the acquisition of data from the board the diamond is attached to.  Previously we would get a 'false trigger' every once in a while, say 10% of the time.  Background particles would sometimes hit the scintillator to trigger it, but as the particle did not pass through the diamond, the signal out of the sample is almost non-existent. 

But after chancing to the 5152 I haven't seen any false triggers.  The program runs much slower when sampling at 2GHz, but I don't understand how this would affect the false triggers.  It shouldn't have acquired some filtering technique in the change.  As I said before the block diagram changes are trivial, just resource name channel etc. except that I had to change the value on niScope Configure Horizontal Timing's enforce realtime to false.  I've never used that before, and I don't quite know what it means but it makes the program work.  Could this make an impact?

Thanks for your input,

Ben

[bbentelin]

Recently my lab switched to using a 2GHz sampler (PXI-5152) from a 100MHz sampler (PXI-5122).  The changes to the labview programming were trivial to make the 5152 work so I am intrigued by an anomaly in the data.

We have a radioactive source raining down particles through a diamond sample and onto a scintillator.  The scintillator triggers the acquisition of data from the board the diamond is attached to.  Previously we would get a 'false trigger' every once in a while, say 10% of the time.  Background particles would sometimes hit the scintillator to trigger it, but as the particle did not pass through the diamond, the signal out of the sample is almost non-existent. 

But after chancing to the 5152 I haven't seen any false triggers.  The program runs much slower when sampling at 2GHz, but I don't understand how this would affect the false triggers.  It shouldn't have acquired some filtering technique in the change.  As I said before the block diagram changes are trivial, just resource name channel etc. except that I had to change the value on niScope Configure Horizontal Timing's enforce realtime to false.  I've never used that before, and I don't quite know what it means but it makes the program work.  Could this make an impact?

Thanks for your input,

Ben

[bbentelin]

Can anyone answer this question:

I noticed a stark change in the resolution of the waveforms at the threshold where the Real Time Enforcement change has to happen.  In turning Real Time Enforcement to false does the acquisition no longer perform an acquisistion of all the data from one trigger, but average the data of several triggers into one waveform?  If this were the case then I wouldn't be seeing a false triggers because the false trigger's data is thrown in with the real data.  In the averaging process the false data is concealed.

Thanks,

Ben

[bbentelin]

Another question...

Why is real time enforcement a problem?  My computer's processor is 2.4GHz, and the module says that it is 2GHz, so why is there an error with the max realtime sampling rate?  Does the module say that it is 2GHz because it is able to do this RIS method of averaging to make it appear to have a 2GHz sampling rate, when really it can only maximally sample at 1GHz (the threshold at which Real Time Enforcement needs to be false)?

Thanks,

Ben

[Efrain_G]

Dear Ben,

In turning Real Time Enforcement to false, you will allow your acquisition to perfrom RIS (Random Interleaved Sampling). This technique is used to acquire periodic signals, where the sampling is slightly shifted in phase on every cycle. By repeating this process for multiple cycles, you will obtain samples in between samples. Therefore, you can reconstruct your periodic waveform very accurately since you will have a much larger number of samples available (Just as if you had sampled at a much higher rate).

[bbentelin]

Can you exlpain the process necessary to setup the 5152 for TIS at 2GHz?  Am I able to use the block diagram that I have right now with some minor tweaks or will it require more of an overhall of the initilization process?

[Efrain_G]

Ben,

There is a Knowledge Base in our website that answers this question:

http://digital.ni.com/public.nsf/allkb/C57D36F398EADB8C8625721200377BB1?OpenDocument