A faster alternative to an AT MIO 16 DE10?

[Luis Villa]

I am currently using an AT MIO 16 DE 10 DAQ card I am using 24 digital and 8
analog channels and the two GPCTRs. This card will go up to 100ksamples/s.
This sample rate is divided between the number of channels I use for an
acquisition. I want to use all 16 analog inputs in this card. Is there a
functionally equivalent card that will give faster sample rates?

If the events I am sampling have a frequency of 5kHz am I right to assume I
need at least to sample at 10kHz per channel? This means I need a card that
will sample at least at 160ksamples/s? Would I be better of oversampling 4
times? Is there a formula to determine optimum sampling rate?

Does multiplexing mean that if I sample each channel at 1000 samples
per
second by the time I read channel 15 I will be reading it 16 ms after I read
channel 0? Is there a way around this delay?

Thanks

Luis

[Chris Matthews]

Nyquist theorem says that you have to sample at least 2x the frequency of the incoming signal to prevent aliasing. However, if you aren't just trying to avoid aliasing, but also want an accurate representation of the waveform, you must sample at MUCH higher than that. There isn't an equation for that, it just depends on the input signal and what shape it is. Usually for normal waveforms like sinusoidal, 10 times over sampling is recommended.

As for multiplexing, it depends on your board. For the one you have, yes, if you read 15 channels at 1000 S/s then the board is running at 15 kHz and channel 15 is read roughly 1 ms away in time fom channel 1. If you need simultaneous sampling, NI provides simultaneous sampling DAQ boards(6110E, 6111E) and a connector bl
ock to do simultaneous sample & hold (SC-2040).

As for recommending hardware, it sounds like you will need to go to two cards. NI doesn't have a faster MIO board that has 24 digital lines. You can get a 24 line DIO board (6503) really cheap though. Then for your MIO board, it will depend on if you need simultaneous sampling. The 6110E has simultaneous sampling and is really fast (5MS/s), so it would probably be best.

In the future, you should submit your questions like this to the Measurement Devices->Multifunction I/O section of Developer's Exchange instead, or call National Instruments so our sales representatives can help you with your hardware selection.

Best Regards,

Chris Matthews
Measurement Studio Support Manager