That is why I used the word "maybe"!
If Woodhouse is adding a 4049 just to convert the 12V external square wave to 5V, an opto will be a perfect choice because then the ext signal and circuit ground will remain separate, which is a good practise anyway, it eliminates inadvertant(system level) ground loops. Further it will eliminate the need for the 12V supply and 4049. Of course the opto should be preferably schmidt opto like 6N137.
If the 12V is also being used to power the square wave signal source (not mentioned anywhere) then the cause of glitching involves the transducer too, in which case as Kai also has pointed out, a ground loop may exist. As temporary solution an opto can reduce the number of places the ground track must be broken and rewired to get the existing PCB working, particularly if it is a ground plane. In the long run, new artwork, among other things like decaps etc is a superior solution, because a board with a ground loop and decoupling problem is most likely to have this issue elsewhere too and revision would be good thing.