Does 'Snicko' really work? - Sound generated by object passing without touching (Those that don't understand cricket, please ignore this context, I will get to the physics...)
England are playing Pakistan at Lords and a decision has once again been overturned based on evidence from the 'snickometer'. (see over 1.4 ) It's always bothered me slightly that there seems to be a possibility that what this detects is not the ball hitting the bat (or glove in this case) but could just be the two passing very close.
When two objects collide in air, e.g. a ball hitting a bat, there is typically a noise and a deviation in the motion of one or both of the objects. In cricket, these collisions can be very fine as the ball passes the edge of the bat and detecting them is important to the game itself. The primary detection devices for the ball hitting the bat are the eyes and ears of the umpire, but in recent years the teams have been able to ask for decisions made in this way to be reviewed by an umpire with technological assistance. One of these pieces of technology, the 'snickometer', or 'Snicko' detects and enhances the sound of the ball supposedly hitting the bat. My doubt is whether this is fully accurate or whether Skicko fails to distinguish between contact and a very-near-miss.
My thoughts run something like:


*

*The sound we wish to detect is caused by the vibration in the surface of a bat and/or ball when they collide.

*Sound, as a pressure wave, could also be caused by the compression of air as the ball approaches very closely to the edge of the bat, and rarefaction as it leaves the vicinity of the bat.

*How much do these sounds differ and how can they be differentiated by a sound detection device?


The first thing that springs to mind is that the vibrations in the case of the bat and ball colliding are plural, and thus will show for longer in the detection device. A perfectly spherical ball passing a bat would only cause a single rise-fall-return change in pressure. However, a cricket ball is not a perfect sphere, it has a raised, stitched seam; this creates miniature ridges that would vary the compression between the ball and bat-edge as it passed.
So, is my knowledge and understanding sound? (pun unintended but left anyway)
If it is, how would we, or how does Snicko, differentiate between 'ball hitting bat' and 'ball passing very closely past bat'?
 A: The snickometer filters out background noises first of all. The microphones near stumps pick up the noise, which is then amplified by using some sort of resonance amplifier.
Now the ball passing the bat will more or less cause no appreciable change in sound at all. Pressure waves caused by the ball will not cause the same impact as contact force will. Firstly the area exposed to pressure wave will be more than that for the contact force and as a result impact will not be that much, and therefore the oscillations won't be significant. Secondly the time of impact is very less for contact forces and as result a large impulse is generated, which increases the amount of oscillations significantly.
A more confusing sound will be like the bat-pad touch, or bat hitting the ground. Theoretically there are two ways to determine this.


*

*Analyse the waveform. Take as templates the characteristic data of previous bat-ball contacts and compare the now bat-ball contact to the previous data. Technically this is not feasible, as it will take a large amount of time, although the third unpire might look at it for some intuition.

*See the time period of contact. This is more realistic. Since 
a) the bat and the ball are rigid solid objects (the Young modulus is more than pad and is immovable as compared to round where dust usually flies off on contact), and
b) the ball travels at a fast speed (as compared to the speed of bat hitting pad or bat hitting ground), 
the time period of contact is very small compared to other contacts where both these conditions are violated, and directly by seeing how long the deviating waveform lasts, the umpire can easily understand whether the ball has touched the bat and something else has not happened.
A: physics is one thing, real life testing is another. I am sure you can get faint sounds from ball passing bat. I know, because I was once given out to one when I am pretty sure I never touched it. And Joe Root has just had a decision overturned where he didnt touch it but there was a faint deflection on snicko. Not a big spike like a true snick, bit still a deflection.
