One of the problems of using a new led-optic combination where no specific data is available, is working out the correct distance between the led mounting face and the lens mounting face.
Having printed a quick fixed lens holder, I realised an adjustable holder might be better, and so printed this one (right) with three M3 screws in the top to retain the lens and three in the bottom as adjustable jacks.
BTW, M3 screws have an 0.5mm pitch, which means they are their own micrometer adjuster – 125μm for quarter of a turn, for example. To adjust the bed of my Ender 2 3d printer, I printed round knobs, each with seven bumps around their peripheries, to slip over the heads of the M4 bolts, to get 100μm movement per bump.
The earlier fixed lens holder implement the 16.0mm spacing that I deduced from Gaggione’s document for a similar led (which printed at 15.95-16.05mm – well done Prusa Mini).
This left four small flares in the beam pattern, one emerging at 45° from each of the four corners of the rectangle. It was seeing these flares that made me think that 16.0mm was not quite right – I will try to get a beam photo when I am better set-up, but it is tough for a camera to pick out faint flares when the middle of the beam is so bright.
The adjustable lens holder allows spacings down to ~15.6mm to be explored before the rather fat back of the optic hits the (filed down) screw heads holding the star pcb down, or the lens comes to rest on the solder holding the wires in place – LLC15 collimators are really best used with emitters on unobstructed pcbs, and there is no reason it should be otherwise.
Anyway, looks like 15.7 to 15.9mm is fine, and 16.0 is pretty good too – the corner flares are omni-present, it seems. Lateral alignment needs to be just right as well, suggesting an adjustment mechanism in the final design.