Multiple Cards - One Souce of Input

[artaxerxes]

I an considering installing PCI 6601s in 6 PCs used in fMRI experiments.  Generally one of these or perhaps 2 at any given time will be active and receiving an external trigger signal from an fMRI Scanner.  Whilst one USb device might be suitable - it would not give relaible timing information (unless you know better!)

Previousy we had Paralllel Port inputs whihc we could easily plug between the PCs to transmit the signal.  But with 6x I would need 6 breakout cables.  Is there an easier way to plug these into one breakout box?

Regards,

[John_P1]

Hi artaxerxes,

What sort of timing information are you hoping to measure?  USB DAQ devices usually have a buffer on them, so if you are looking to timestamp relative events or something of that nature then the non-determinism of the bus transfers usually isn't much of a factor.

If you provide more information about what you need to measure I should be able to give you a better idea of what hardware would be suitable. 

The 6601 uses a 68-pin SCSI connector so the cables should also be fairly interchangeable.  It seems that you could install a 6601 in each of your PCs and use a single breakout box connected to whatever you needed to measure.  You could swap the cable and breakout box as necessary between the 6601s--if I understand correctly it sounds like you were doing something similar previously over the parallel port.  However, I'm not sure I understand correctly so please correct me if I have misunderstood.  Anyway, I'm not necessarily advocating this option as USB might actually be a possibility for your application (depending on the details).

Best Regards,

[artaxerxes]

We currently have 6 parallel ports, one each on a front panel breakout of the 6 rack mounted stimulus PCs. We were happy enough with the users swapping the parallel cable as necessary between the fMRI and the individual PCs... we basically only need 1bit with interrupt from these parallel port for syncing with the fMRI scanner pulses. 

Your counter card will allow old MATLAB and NeuroBS Presentation code to remain aware of the fMRI signal - with a bit of a re-write.  Psychophysics experimentat stimulus based on this signal means we can't use chaper USB solutions either.  It seems that the obvious solution is 6 cards, 6 cables and one break out box. 6 cables are necessary as we don't want the users going behind the rack to plug and unplug.  Hence me asking if you have a mutiple entry breakoutbox. It does seem massive expense though just to recover a 1bit signal!

We could alleviate the requirement for 6 cables if there was a way to break out at the front of the PC.  Do you do a front panel (CDrom blanking plate, floppy blanking plate) terminal for the PCI card.  I could always route an extension cable to a blanking plate but an premade one would be more elegant!

Thanks for your attention

[artaxerxes]

Having a look around I see I could just make a custom cable to take 2-3 lines from each card to BNC and then route them to a big box of BNC connectors.  This alleviates some of the cost!  6 x £300 is close enough to the Quattech PCIe ExpressCard or PCI Cardbus + Parallel Port cards that we had looked at.

Ta muchly.