Sampling at a higher rate and storing the data with NI USB 5133

[Anilvs77]

Hi,

I am using NI USB 5133 and Labview 7.1.1. I need to acquire data from a sensor at the rate of 10 MHz. I am using the same signal as a trigger (Analog edge) with a very small trigger level and a huge time out value. I have written a code in Labview and trying to get all the samples that are fetched. The problem is with the record length option in the horizontal time settings. When I work with the default options, only 1000 samples are stored whenever there is a trigger. I tried to increase the record length but limited by the system configuration to 5 Mega samples. Is there any way out by which I can record all the samples that are fetched, as it important for the post processing. Will I capture all the samples if I set record length same as sampling rate ?(Provided the system configuration allows this)

[dansch]

Hi Anilvs77,

I understand that you are trying to configure your USB digitizer for an acquisition at 10MHz sample rate, but are having trouble with the record length. Please correct me if I am wrong in this assumption, but it sounds like you are simply trying to acquire more samples than the memory can hold on this device. This device has 4 MB/channel of onboard memory, and so you cannot specify the record length to be greater than this amount. Because the USB-5133 does not directly support continuous acquisition, the only option for acquiring more than 4 MB is to initiate and fetch multiple times with a loop. There is a helpful KnowledgeBase article titled "NI USB-5132/5133 Acquisition Benchmarks" which explains this concept and provides benchmarks for this type of acquisition. Hope this helps,

[Anilvs77]

Hi Daniel,

Thanks for the response. The benchmark link was of great help....

Anil

[marcelappb]

Hi,

I have ni Usb 5133, and I need acquire a sample of 50MS/s. This sample has approximately duration of 1 second after trigger. The 5133 has a memory onboard of only 4M, then it would be necessary intantaneously to transfer the data to the memory of the PC, therefore with these 4M,I only acquire 80ms of the total sample. How I obtain to get this sample of the 1s without losing given?
I read about DMA, however he did not help.
I need help.
Sory my English.
Marcela.

[S_Hong]

Hi Marcela,

Unfortunately, at the rate you are acquiring your data, you will need to transfer data to your PC at a rate of at least 50MS/s * 1 Byte/Sample = 50MB/s. However, this is not possible with the USB-5133 according to some the benchmark tests we have performed. Therefore, one recommendation is to lower the sampling rate and keep fetching your data in a loop to offload some of the data from your on-board memory. If this is not possible, another solution might be to simply shorten your acquisition length.

If none of these options are possible, we may have to look into some of our other PCI or PXI High-Speed Digitzers. Because these devices use the PCI or PXI bus, you should not run into memory issues for your application.

Please let me know if you have any question.

[S_Hong]

Hi Marcela,

I just wanted to clarify that since there is no fetch pointer in our USB-513x digitizers you will not be able to run your acquisition continuously. So the only way to perform a "continuous" acquisition would be to have your initiate, fetch and stop in a loop. However, by doing so, you will be missing some points in between acquisitions. Therefore, as I mentioned in my previous post, you can either slow your sampling rate slow enough such that you can fit all of your data within the 4MB memory onboard memory or decrease your acquisition length by taking less samples.

S_Hong

National Instruments

Applications Engineer

[Jennifer O]

A 32MB/ch memory option for the USB-5132 and USB-5133 has been released, which increases the maximum record length by a factor of 8. 

Jennifer O.

[q-nust]

I have SonyVaio core i5@2.6 GHz with 8Gb of RAM so how much Samples per second I can fetch with NI USB 5133 4Mb/channel device?

[Nathan-P]

As you can see, you have posted a new question onto a thread that has been dead for a few years.  To get your questions to the right people, please post new questions into a new discussion form topic.  

That said, you are limited only by the bandwidth of the USB bus.  Here: https://www.ni.com/en/support/downloads/instrument-drivers/instrument-bus-performance---making-sense...  I have a Dell Precision T3500 and was able to get a Max of 16.7 MB/s streaming from a USB-5133.

-Nathan