limited sampling rate

[imperial-aero]

Hi,

I've got a VI that reads in analog inputs. The DAQ device I'm using is a PCI 6229. I believe that the max. sampling rate of this device is 250kS/s. So, if I were were to take inputs from 16 channels I thought I could enter a sampling rate of 15kS/s (since 250/16 > 15) . However, when I look at my output data, I can find that the sampling rate was actually 14925 S/s not 15000 S/s which I entered.

When I enter a sampling rate of, say, 14kS/s the data shows that it's actually 14kS/s. So I'm wondering whether the maximum limit of my sampling rate is around 14.9kS/s and not 15kS/s.

Cheers.

[BenjaminR]

Hi,

To be able to help you I’d like to know if you have some error messages? Have you tried a sample rate of 250kS/s on 1 channel? Do you use your 16 input in referenced or differential?

[imperial-aero]

It didn't produce any error messages when I ran it with 16 input channels and 15kS/s sampling rate (all my input channels were at RSE). I'm just confused as to why the sampling rate on the output file is shown to be 14.9kS/s. I just ran it with 1 channel and 250kS/s sampling rate and the output file did show that the sampling rate was in fact 250kS/s. I've attached my VI for you to have a look.

Probably different to my original question; I've used an example shipped with NI for some analog output, and I'm not sure what a particular function of it used for. I've attached a picture of it here as well (the part which I don't understand circled). 

[johnsold]

Many multiplexed devices need to allow for settling time for the mux and the amplifier in addition to the sampling rate divided by the number of channels. I did not look up the specifications on your device, but I suspect that this may be the issue.

The time per sample at the nominal rate is 1/(250000/16) = 64 us. The period at the actual rate of 14925 samples/second = 67 us. Three microseconds for switching and settling is not unreasonable.

Lynn

[BenjaminR]

Hi,

I’m ok with johnsold for the nominal rate but normally the sample rate (here 250kS/s) consider the switching time. This is why I’m not sure about johnsold explanation. I will look for to know why you have just 14925S/s using 16 channels and not the 15625S/s.

Moreover, the cluster entry is the cluster that has the elements that you want to replace. At least one element of the cluster of entry must have a label that it is bound. You should always be wired terminal cluster of entry. I can’t say more about it because I don’t have all the VI.