DIY Scienscope fiberoptic light

aidanmoore · Post by **aidanmoore** » Sat Dec 01, 2018 4:11 pm

Hi,

It's difficult to get into the port to do an accurate measurement, but the fly eye lens appears to be hexagonal with parallel edges about 38.5mm apart.

aidanmoore · Post by **aidanmoore** » Sat Dec 01, 2018 4:17 pm

I count 21 seperate lenses across 38.5mm, but the spacing between lenses is not consistent. I'd ballpark another 6 lens diameters in the spaces and guess
38500/27 =1426um. But that's pretty rough measuring.

Saul · Post by **Saul** » Sat Dec 01, 2018 4:40 pm

aidanmoore wrote:I count 21 seperate lenses across 38.5mm, but the spacing between lenses is not consistent. I'd ballpark another 6 lens diameters in the spaces and guess
38500/27 =1426um. But that's pretty rough measuring.

Thank you very much ! It is more accurate than I expected

According my calculations from the Nikon website picture and approx diameter, I got 1.7mm. Pretty close

Is this plastic or glass ?

enricosavazzi · Post by **enricosavazzi** » Thu Jan 03, 2019 2:28 pm

aidanmoore wrote:I count 21 seperate lenses across 38.5mm, but the spacing between lenses is not consistent. I'd ballpark another 6 lens diameters in the spaces and guess
38500/27 =1426um. But that's pretty rough measuring.

For what is worth, the white paper at the following link states (on page 5) that the minimum and maximum number of "channels" (=lenslets)
across the width or height of a fly eye lens is approximately 7 and 11, respectively, and explains that the problem with a large number of
lenslets is that the edges between lenslets are not infinitely sharp and cause scattering. So the number for the Nikon fly eye lens is higher than
these recommendations, but again, maybe scattering is not a big problem in this particular use, or the edges are sharper than in ordinary cast-glass lenses.

https://customers.zemax.com/ZMXLLC/medi ... f?ext=.pdf

Another thing is that this white paper makes an example with a double lenslet sheet for better uniformity, but in the Nikon illuminator there
seems to be only one (there is an example of this arrangement earlier in the paper).

Interesting also that an odd number of lenselets across the array is said to be better than an even one to avoid a central darker area.