Aggregate sampling rate and Multichannel Scanning Rate

[crimolvic]

What does this both parameters mean?

for example, for the NI PXI 6220:

The Aggregate sampling rate is 250 k
and the Multichannel scanning rate is 500 k

I need to know this, because I need to calculate the maximum SS/H sampling rate and I need to know how do I determine the MSTm, by using the 250k or by using the 500k


thanking you in advance

CJMV

[Otis]

The maximum sampling rate for the PXI-622X Devices is 250kS/s.

If you need any additional characteristic information then please look at the Data Sheet for the PXI-6220.

[crimolvic]

OK, and what does the multichannel scanning rate mean?

[Otis]

Hello crimolvic,

I think what the confusion here is what is the difference between the maximum rate and the per channel rate for a Data Acquisition (DAQ) Board.

Our DAQ boards are given performance ratings for the maximum number of samples that the Analog to Digital Converter (ADC) can sample at a given frequency.

For an M-Series board there is 1 ADC on which all of the signals are multiplexed. That means that if you have 5 channels being sampled by a 250kS/s board, then the maximum per channel rate you could obtain would be 50kS/s. If you divide the maximum sampling rate by the number of channels it will give you the maximum per channel rate.

I hope this addresses the confusion about the different stated sampling rates and that you now will be able to better find out the proper rate for your system.

Regards,

[ksyz]

Hi, I am using a NI 6224 DAQ card, the maximum sampling rate is 250KS/s. If i used 16 differential channels, the maximum sampling for each channel is 250KS divide by 16 or 32?

Also, if i am going to measure just a dc voltage, what sampling rate and sample data should be? Are there any limitation when setting these parameters when measuring DC voltage?


Thanks in advance!

ksyz

[Otis]

Hi ksyz,

You can very easily find out the max rate programmatically by using this Property Node:

You can find more details about how to do it here:
KnowledgeBase: Can I Programmatically Determine the Maximum Sampling Rate of My DAQ Card?

Even when using differential measurements, you are still just looking at up to 16 channels.  You can see why here:

As you can see because this is an instrumentation amplifier (source) it is actually just measuring the difference between Channel N+ and Channel N-.  So you are still only taking 16 measurements.

To determine the maximum rate, simply take maximum sampling rate divided by the number of channels.

For a DC signal, you don't really need any sampling rate.  Simply take one sample and you'll determine the DC value.  Or, take many samples and average them for a more precise measurement.  If you are trying to find the rate at which you should sample, then follow the Nyquist Theorem (see Bandwidth, Sample Rate, and Nyquist Theorem).

Regards,

[Ayush]

Hello,

I also have a question about the maximum sampling frequency for a NI 6363 DAQ Card. The specification says that the card supports a single channel sampling rate of 2MS/s, however, for a multi-channel sceanrio it reduces to 1MS/s (aggregate). I wonder if this reduction is related to the use of two connector-blocks (supported by the NI 6363) or already happens if several channels are used for one connected block. It's absolutely clear to me that the sample rate per channel can be calculated by dividing the maximum sampling rate by the number of channel, however,  I dont know which one of the two sampling rates I should use.

From the previous posts and reading other threads in forums I could not find any answer.

In my current set-up I use one connector block with one differential sensor input and I use Matlab for the data accquistion. If I increase the sampling rate above 1MS/s Matlab throws an error. This is why I assume that 1MS/s is the maximum sampling rate, but shouldnt it be 2MS/s??

Thank you for answering my question.

[Seth_B.]

The reason that the 1 channel rate is 2MS/s and the multi-channel aggregate rate is 1MS/s is that the multiplexer on the board has settling time associated with it.  When reading one channel, we don't have to switch the multiplexer and don't have to wait for it to settle.  As soon as we add any other AI channel, we have to switch the multiplexer and are then at the mercy of it's settling time, which only allows for a maximum of 1MS/s aggregate.

On one channel, you should be able to sample at 2MS/s.  With multiple channels, your aggregate rate will be 1MS/s.  What is the error that you are receiving?

[skyclimber69]

Sorry Guys I'm still perplexed about the meaning of "Aggregate" in the context of sampling. I'm currently reviewing the NI-9025 and it also quotes 250 kS/s aggregate sampling rate... what is the actual of meaning/definition of Aggregate Rate or 250 kS/s Aggregate?

Cheers

David

[DanCee]

Hi David,

I understand your question and im surprised the guys here havent answered it clearly. Perhaps they dont know.

But please can someone LET ME KNOW if the following explanation of aggregate is wrong: Lets suppose i have a PXI 6259 card. It boasts 

"16-Bit, 1 MS/s (Multichannel), 1.25 MS/s (1-Channel), 32 Analog Inputs" in its description and in the specification its says the 1MS/S Multichannel is "Aggregate".

This means that when more than one channel is sampled the divider used is 1MS. So for 16 difference inputs the fastest acquisition speed is 1000000/16 = 62.5kHz. And for two channels it will be 500kHz etc. If you are using 32 single ended channels then its 1000000/32 = 31.25Khz.

So by simply using the internal multiplexer we reduce the aggregate to 1MS from 1.25MS (or 4uS).

This is how i understand it although its not clearly defined anywhere on the NI site as far as i can read, again can a member CONFIRM this please as its important.