P3D Re De-polarising filter?

[Peter Homer <P.J.Homer@xxxxxxxxxxxxx>]

>Thanks for the insights.  Just to clarify; what exactly does a quarter-wave
>retarder do to the polarised light?  And if I can get hold of one, that
>sandwiched between the lens and a standard linear polariser should do the
>trick?

It consists of crystaline or crystal like material so has two directions in
which it can transmit light at right angles to each other but has two
different refractive indices for these directions (Birefringent). When it
is orientated with these directions at 45 dergrees to the plane of plane
polarised light it forces it to split into two equal intensity  planes at
right angle to each other. When these emerge from the retarder the two
planes resolve according to vectors, if there is no difference in
refractive index as with a cubic crystal neither wave is retarded relative
to the other and they resolve back into a single plane poarisation
bisecting the 90 degree angle between them. Which means that they return to
the original single plane polarisation and there is no detectable change.
The same would be the case if retarded by any whole number of wavelengths,
but if the retardation difference is one half wavelength then the direction
of the one vector relative to the other has reversed and so the plane
polarisation that they resolve into is at right angles to the original one.
Where the difference is one quarter wavelength they are out of step and the
direction of one relative to the other keeps changing along the wave train
so the resultant direction rotates and you have a helical wave. Which can
have either a right or left handed pitch depending on which of the plane
polarisations is retarded to which. Looking towards the light source this
means you have either a left or right hand circular polarisation.

I think it should work if you sandwich it between the polarised output of
your projector and the final polar at 45 degrees to the projector
polarisations.

P.J.Homer