Multiple PFI external signal input problems on USB 6210 (multiplex issue?)

[JohnDoe-]

Hi,

I'm using a USB 6210 board and a self-written program in a third party software to make timestamps for photon arrivals from a 50/50 split signal so I can cross correlate it.  When I configure the board to take data from one photon detector (only one input connected), it works fine.  However, when I try to use the other counter to process the signal from a second detector, I get very wrong data. 

After alot of troubleshooting, it seems that when I have only one photon detector hooked up to pfi0 and a count edges task configured and executed looking at both the pfi0 and pfi1 inputs, I start getting data from the first detector.  I know this is happening because it occurs when the second detector isn't hooked up, but when I hook a wire into pfi1 and touch the other end to any conducting surface (even my fingers!). 

I've done some reading and it sounds like this is a "multiplex" problem that occurs with high sampling rates, but I've only read about it happening with analog scanning.  I noticed that the 6210 board is configured in such a way that it will "share electrons" on inputs that are right next to each other because of the fast switching of the multiplextor system (or something to that effect).  I assume this is happening because, for some reason, the signal being pumped into pfi0 is partially available on the rest of the inputs, so that when I try to count off of one of them, if I ground the wire to something (detector, metal or fingers), it starts counting that signal.  I've tried hooking up the first detector to pfi0 and grounding pfi 1 and trying to take data from the second signal hooked up to pfi2 or 3, but the same problem occurs. 

I just need to know if theres any way to shut off the multiplexing, I really don't think I need it.  I only need to take data through two inputs so I can cross correlate, perhaps even use a third input for a trigger to synchronize both detectors.  The next thing I was going to try was to see if I could ground pfi1 and set up a dummy operation that would collect data from that input that I wasn't going to use just to distract the board in a way....so it wouldn't cause problems for the second detector input hooked up to pfi2.  So far I haven't figured out a way to do that since the board only has two counters and I need both of them for real data.

I greatly appreciate any help!! 

[Aaron_W.]

John,

There is no possibility that this is a Ghosting issue. Since the Counters are not multiplexed it is not possible that you are seeing the Ghosting from one channel to another. You might however be seeing some crosstalk between the lines. You might be able to help minimize this problem by turning on the Digital Filter for the device. 

If you look at the documentation for the USB-6210 it gives details about the Digital Filter. Try and enable it and see if you have the same problem.

http://www.ni.com/pdf/manuals/371931f.pdf#page=95

[JohnDoe-]

Thanks for your response.

I can't seem to figure out how to enable digital filtering.  I'm using a third party software program (Igor) and not Labview.  I've recently purchased a second USB 6210 in hopes of eliminating the cross talk.  I now have one detector connected to one board while the other detector connected to the other board.  The boards have no common connection other than to the computer I'm running the software on.  Somehow, I'm still getting cross talk between boards.  I've also noticed that the board seems to accept really low voltages when it counts or triggers.  That is, if I touch a wire with my finger, it takes it as a trigger pulse.  Am I to assume that this can be fixed with filtering?  Is there some sort of amplitude filtering I could enable so it wont pick up noise as much? 

Also, is there any way to enable the filtering in MAX, or some way to set up a counter operation with a pulse train that will function as a filter.

[Aaron_W.]

John,

It is strange that you are seeing such behavior with your USB-6210, could you provide a connection diagram of how you have your connections made between your 6210 and your photon detector. This sounds like it might be a grounding problem of some kind. As far as the ability for you to enable digital filtering with the device, you will have to check with the makers of the software and the driver to see if there is a property in their driver that will allow you to enable this property on the 6210. Let me know if you can produce a connection diagram. Thanks. 

[JohnDoe-]

I've gone through a few different connection methods to try to get rid of the problem.  The one I used most frequently that produces the aforementioned results involves connecting the photon detectors to a BNC 2121 accessory connector block thats actually made for a PCI 6602 and then wire jumping from that block to the USB 6210 board.  The connection diagram for this set up is attached.  Basically if I have one detector hooked up to the board, it registers count for that, but if I execute a counting operation for the other board, simply touching the stripped 18AWG wire causes it to start counting as well.  The count rate that is registered makes it appear as though the boards are crosstalking somehow.  The diagram I have attached does not include my trigger set up, as I'm currently just trying to get the boards to count two different signals at the same time without any ghosting. 

I had a feeling it might be a grounding issue, so I've recently tried stripping a BNC wire and hooking the detectors up to the board directly without using the BNC connector block as a mediator, but I haven't tested this configuration enough yet to see if it works or not.  Right now I'm trying to see if giving all the connections a common ground will help.

Thanks for all your help!

[AnalogKid2DigitalMan]

I looked at your diagram but did not see any connections from the photon detectors to the 6210 digital ground. You need a ground reference connection.

-AK2DM

[JohnDoe-]

Thanks for your solution to one of my problems!  I grounded both the boards to the BNC connector block and it fixed the crosstalk problem between boards.  When I tell both counters to count but leave one detector unplugged, I only get counts from one board.

Now for the more tricky part.  I am trying to implement a trigger pulse that will start both board's counting operations simultaneously.  I've now got a trigger set up on my PCI 6602 board that produces a pulse out of PFI36.  I'm using a BNC splitter to split that signal, jumping them to each USB 6210 board's pfi3 ports through the BNC connector block much like I was using it to jump the detectors into the board. 

It seems that I am still getting crosstalk within the same board as the counting starts even though I never apply the actual trigger pulse.  That is, I tell the board to wait for a trigger pulse, but it starts counting without my say so.  I assume this might be a grounding issue also, though I've used up all of my digital ground lines on the 6210 since it seems the more i ground to the block, the less ghosting I get.  Should I try to ground the same DGND port multiple times onto the connector block?  Any other suggestions?

[Aaron_W.]

John,

Have you been able to scope your digital trigger line to see the noise or do you think the problem is with how your software is configured?