Removing digital spikes

[flycast]

I know there is quite a bit in LabView on signal processing. I imagine that someone here knows how to do this.

I have about 12 digital signals that are coming from a NI 9425 DI unit. There is something in our system that is creating transient spikes in that digital data. These digital spikes are always false and short lived (0.003 seconds or so) from 0 - 1 and then back to 0. They are often accompanied by a pneumatic solenoid cycling from open to closed or vice versa. We never have any actual valve signals that are that short. Here is an example (red arrows):

My wiring guy has not been able to determine the source yet and eliminate it.

Is there a filter of some kind that would work to remove these false spikes without changing the genuine transitions from high to low?

-or-

Should I write my own using some kind of shift register or something in a loop?

[santo_13]

Given your 9425 DI is fast (7uS), a 3ms high signal could be still valid data. Since this could be still valid data but with respect to your application it is not valid, you would have to implement filtering at the application level to filter these short pulses.

Since you say this glitch occurs at every pneumatic solenoid cycling, it must be the inductive kickback from the coils causing this issue. You can solve this by adding flyback diodes across any inductive load.