Common mode noise and balanced detectors

[matheyrichs]

I'm working on a laser interferometry system that uses a pair of balanced photodetectors to record the signal. Each detector is on a separate port of a beamsplitter cube, and as a result receives (in theory) the SAME signal. The rationale for using the balanced detectors is that "they reject common-mode noise on the signal".

What exactly is this common-mode noise in the first place? If it is something present at the detector in each signal, then why does that specific piece of the signal get filtered? It seems to me if this is really as simple as subtracting what's common at each detector, then the desired signal would also be lost (seeing that it is the same at each point...).

Can anyone help me understand how a balanced detector scheme works? Thanks!

 

[dlgoff]

...if both inputs fluctuate by the same amount (while remaining constant relative to each other), this change will have no bearing on the output.

http://en.wikipedia.org/wiki/Common-mode_rejection_ratio"

You may have seen this wikipedia discription of Optical Hybrids. But I think it will answer your question about balanced detectors.

http://en.wikipedia.org/wiki/Optical_hybrid"

Regards
Don