PBS beam combining two NUBM38s

[Light superglue]

For me waveplate is only important when I combine 2 single beams (single- or multimode) of perfect shape and want to keep the shape of the combined beam so.

But for pretty loosy beam-packs of these NUBM3x MDPs presence of costly waveplates would not do much of improvement.

 

[Junkers]

Well time to finally add some closure to this thread. I've opted to combine the 2 NUBM38s using a prism. I know this is a potentially lossy method but I've ordered an AR coated prism and am willing to tolerate some losses for the sake of keeping the form factor small. The manufacturer states that when used as a 180 degree reflector (incident beams normal to the surface) reflective losses should be as low as 0.25% per interface. I've measured the loss with my current prism at around 8-9% which seems odd to me, I was expecting it to be higher. I know that K9 glass typically has a loss rate of around 8 - 10% with a standard lens so this leads me to wonder if this prism already has some sort of coating on it. I don't think it does as it only cost me a few dollars, though it did come from an optics manufacturer on clearance.

This final picture is of the prism I am currently using. I am not sure if this 'window' effect is a sign it may be coated or just an artifact from manufacturing.

 

[kecked]

Very eloquent solution. One optic! I don’t include the collimation as every setup needs it. Thanks for sharing.