c-FP-DI-304 intermittent changes

[tst]

I'm relaying this for someone else, so hopefully I'm not getting anything wrong:

In several different instances, we've seen a situation where the inputs on a c-FP-DI-304 card are changing randomly. Basically, they momentarily go on even if what they're connected to doesn't output a signal.

My understanding was that this particular card does not have any filters to protect against noise and that in the cases we've run into this, eliminating noise in the system made the problem go away.

My question is whether anyone is aware of any other things which can cause such behavior or of alternative solutions? Replacing the cards with the DI-301 cards is apparently impractical as they only have 16 channels as opposed to 32 in a 304. We do have a software filter, but that's not a good long-term solution.

[Drew_T]

tst,

As far as noise on digital lines goes, the following are some resources you may want to review.  However, your best bet is probably to protect against noise by using shielded, twisted pair wiring.

In a TTL digital circuit, the margin of error in signals is only 0.3V, which can cause problems in a noisy environment.  Using 24V Digital Logic can help in this case by improving the margin of error.  More about the subject can be found here: http://zone.ni.com/devzone/cda/pub/p/id/262#toc4

It sounds like your noise may be random, but in the case that it is caused during switching, you may look into a debounce filter of some sort: http://digital.ni.com/public.nsf/allkb/14A90E5759875BBF86256DD400034FDD?OpenDocument

Finally, if the noise is of a certain frequency, you can implement signal conditioning using a discrete filter to remove noise on the line before it enters the device.  This will probably not be the source of your error or help very much, but is something you can look into.

[tst]

Some of these have already been implemented - the cards already use 24V digital logic, the cables are all shielded and grounded, etc. Others are not necessarily practical at the moment (such as adding filters or replacing a lot of cables).

That said, I'm told the problem has been resolved - The system has several dozen motors with frequency converters and while it was specified that their power cables should be isolated from the other cables in the system (as well as being shielded and grounded), apparently it wasn't done properly. I was told that they were now moved further away from the DI cables and that this was enough to solve the issue. Hopefully that's really the case.