retriggerable wavetrain acquisition using a digitizer

[halcyon]

In performing retriggerable wave train Ai input acquisition using a digitizer with the Multiple-Record acquisition example, what is the limitaion to the frequency of the trigger used? From the literature, the rearming time for the hardware can be around 1 to 100us so this allows you to capture data if the triggers occur 100us apart (10kHz trigger).  If so, is there any way the digitizer can keep up with the occurance of trigger at 20k Hz performance retriggerable wave train Ai wave train acquisiton?

There is a way round this with a high speed DAQ card using the retriggerable wave train Ai acquisition utilising 2 counters to gate the generated sampling clock. This avoid the rearming of the software trigger. Is there an equivilent example in NI-SCOPE using a digitizer so that I can perform high frequency retriggerable wave train acquisiton? 

[RER]

Hello Halcyon,

Could I ask for some more information pertaining your hardware please? Specifically, which DAQ board and Digitizer are you using?

Also, I wish to enquire about the sampling frequency (and more specifically the amount of data you are trying to save for each record). The digitizer would be capable of keeping up with your 20kHz signal… depending on the size of the data you are trying to acquire!!

National Instruments sells high speed DAQ cards up to 1.25MHz. If you are willing to compromise on the amount of data acquired by each record, you may find it easier to carry out the acquisition using just the DAQ card – using the counters to generate the pulse train.

Thank you for your patience, and I look forward to hearing from you soon,

[halcyon]

Hello Rich,

I am using a 10M Sample/s board PCI 6115 DAQ card to perform retriggerable wavefrom aquisition with a TTL trigger signal with a frequency of 1k Hz. Using the Retriggerable example utilising counters to generate the sampling clock frequency works quite well and avoid the rearming of the trigger.

With a new sweep laser which sweeps at a constant rate of 20k Hz giving out a TTL trigger source, this required a faster sampling aquisition card to perform data acquisition. For a 20k Hz signal, the period is 40us. Assumming only acquiring within the 20us time slot and the requirement that the sampling pts has to be more than 1000 pts, the sampling rate of the DAQ card needs to have a rate of 50M Sample/s, therefore a PXI 5122 digitizer is bought (100 MSample/s).  

This takes me back to the question which is will this digitizer PXI 5122 capable of keeping up with the trigger of 20k Hz frequency?  Are there example programs in NI-Scope which I could use? The likely examples I have looked at are the 2 examples: 

niScope-EX Multi Record Fetch More Than Available Memory.vi

niScope-EX Multi Record.vi

Will this two example program capable of keeping up the the 20k Hz retriggerable signal?

Regards

ps how do I remove a listing in the labview forum board as I have duplicated this message and would like to delete it!

[RER]

Dear Halcyon,

Thank you for the swift reply! Because you have rather hefty system requirements, the test setup will require careful consideration. Here are some thoughts...

1.    You are acquiring 1k samples in 20us, which will require a sampling rate of 50MS/s. This is not a problem, as the 5122 is capable of acquiring data at 100MS/s.

2.    The digitizer has a rearming time of 8us, and you have specified 30us between acquisitions. The digitizer will be able to keep up with the triggering rate.

3.    Multi-record is possible with an advance trigger, and (as you have already mentioned) a good starting point is the example... niScope-EX Multi Record Fetch More Than Available Memory.vi. This should get you to the point of acquisition.

4.    The possible problem would be with committing the data to PC memory. You are acquiring 50MS/s with 14bit resolution, which will work out as 100MB/s. However, because you are only acquiring 1000 per 50us, the average data rate will be a far less...

100MB/s / (50us / 20us) = 40MB/s

(i.e. even though you are acquiring data at a very fast speed for 20us, you have the full 50us between triggers to stream the data to the PC)

...continued...

[RER]

...continued...

The difficulty is that, at 40MB/s, you are pushing the limits of most SATA hard-drives. You have a couple of option available to you...

i. First would be the use of a RAID array HDD. The following RAID array is capable of streaming a whopping 200MB/s...

https://www.ni.com/en-us/support/model.hdd-8263.html

ii. Secondly, depending on the number of retriggerable acquisitions you require (more specifically, the amount of data you need to save) and the size of the digitizer’s onboard memory (between 8-512MB), you may be able to save all the data to the digitizer’s buffer and move it across to your PC at your own pace.

iii. The third and final method is the use of a 3^rd party product called StreamStor. Although not as fast or versitile as some of the National Instruments RAID arrays, it is still a viable option.

https://forums.ni.com/t5/LabVIEW/Stream-data-to-disk-LabVIEW-RT-and-NI-PXI-5122/m-p/120716?requireLo...

Lastly (phew!) could I ask you please provide me with a link to the forum thread you created in error.

I hope this has been of some use to you. Best wishes and kind regards,

Rich

Ps. You may be interested in the following articles...

https://knowledge.ni.com/KnowledgeArticleDetails?id=kA00Z000000kHVNSA2&l=en-US

[halcyon]

Hello Rich,

Thank you ever so much for the advice and information. It has been a great help for what I need to do.

The  link to the redundant post is as follows:

https://forums.ni.com/t5/LabVIEW/limitation-of-performing-retriggerable-waveform-acquisition/m-p/702...

Thank you once again

Regards