Can higher sampling rates be achieved using multiple cards synchronised using the RTSI bus

[parsec]

Hi, I require a sampling rate higher than 20Ms/s at a reasonable bit depth (higher than 8 bit). We have many spare cards lying around here, but the fastest cards are only capable of 10Ms/s.

There are a few cards that we can buy, but anything capable of a bit depth higher than 8 bit is quite expensive.

Is it possible to stagger data acquisition events using the RTSI bus to allow multiple cards to acquire the same signal at the sum of the cards' sampling rates. (assuming they are both capable of the same sampling rates).

Intuitively this could be done with a line delay (monostable) on the RTSI bus to stagger the data acquisition events. Is this plausible? Is there a cleaner way of doing this?

thanks,

Rehan.

[Raajit_Lall]

Hello Rehan,

The solution you are suggesting is definitely an option. Something else that you can also try is acquiring data for the first card on the rising edge of a clock signal and then acquire a data point using the second card on the falling edge of the clock signal. This way you would essentially be doubling the sampling rate using two cards of same sampling rate and you can share clock signals over the RTSI line.

If you look at the example Cont Acq&Graph Voltage-Int Clk, located in Help->Find Examples, if you are using LabVIEW, you will see the DAQmx Timing.vi on which you can specify whether you want to use a rising or falling edge. So, you can use this method on the two different cards and use rising edge for one and falling edge for the other.

Thank you,

Raajit