Synchronized sampling from PCI-mounted card and PXI-mounted card

[einarsto]

Hi, everyone

My latest problem concerns sampling. I need to acquire samples from around 20 different (differential) channels and my PCI-card (NI-6224) has only 16 differential inputs. I do, however, have another card (NI-6229) mounted in our PXI-system. This PXI-system seems to be connected to the computer with a network cable. When i try to perform synchronized sampling from these cards with the PCI-card as master I get an error message saying that these cards cannot cooperate. There must be a way to do this...?

I'm using DAQmx and labview 8

Einar

[Jimmie_A.]

Hello!

Just to verify that I understood you correctly: you have a PCI board mounted in your desktop and then another board mounted in a PXI chassis and that you would like to synchronize these?

If you would have two PCI boards you would use a RTSI cable mounted between the boards and by using that you would be able to synchronize the acquisition that is performed on the two boards.

If you would have two PXI boards you would use the backplane of the chassis in order to get the synchronization.

I would recommend you to get another PXI DAQ board since I cannot figure out how you would be able to achieve what you would like to do in an easy and convenient manner.

Regards,

Jimmie A.

Applications Engineer, National Instruments

[Kevin_Price]

This isn't as convenient, but it's a whole lot cheaper than another DAQ board:

Use the DAQmx Export Signal property node to send your master sampling clock out to a valid PFI pin.  Configure your slave AI to get its sampling clock from one of its valid PFI pins.  Run a physical wire between those PFI pins on the two boards' terminal blocks.

Note: "valid PFI pin" is device-dependent.  I think that MAX will give you info on possible signal routes for your devices.

-Kevin P.

[einarsto]

Thanks to both of you for good answers! One small question,  Jimmy: Wouldn't it be cheaper to buy an extra PCI-card instead? ..or is there a specific reason why you recommend buying an extra PXI-card?

I will do some homework on PFI-pins. If I can make this work I definitely deserve a raise!

Regards,

Einar

[einarsto]

Hello again!

I've tried to modify one of the synchronized sampling examples in LV8 in order to realize the trigger communication. I have (of course) a couple of questions.

1) Will the attached vi work once i connect the physical wire? I haven't tried yet for reasons that are explained below.

2) Both my cards are connected to 2 BNC terminal blocks each (BNC-2110). All these four blocks have 9 contacts called PFI 1-9. How do I know which ones that correspond to the right PFI-source?

3) After the sampling: Where will the data collected by the PXI-system be stored? On the PXI-controller harddrive or on the host computer? A bit confused.

Bear with me

Regards, Einar

[einarsto]

Oooops. Wrong file. Try this instead:

Einar

[Jimmie_A.]

Hello!

It is just a matter of taste I guess but I think PXI chassis and synchronization is more convenient since you don’t have to care about physical wires or a RTSI cable.

Data will be stored at your host since it is the Windows application and LabVIEW with its DAQmx Read functions that actually transfer the data from the buffer created by DAQmx to LabVIEWs memory space in RAM.

[einarsto]

Hi!

We have ordered an extra PXI-card, but it would be interesting to know if it's technically possible to do it the original way (with one card in the host computer and one in the chassis). Is it, do you think?

So the data is stored at the host computer?? I f.ex. shared variables were absolutely necessary because data was stored at the pxi-controller and had to be transferred during execution.

regards,

Einar

Message Edited by einarsto on 04-24-2006 07:47 AM