Transmission spectra of several lenses

Started Apr 6, 2018 | Discussions thread
ForumParentFirstPrevious
Flat view
just Tony
just Tony Veteran Member • Posts: 4,088
Transmission spectra of several lenses
7

Roger Cicala stole my thunder a few days ago:

https://www.lensrentals.com/blog/2018/03/a-geek-of-many-colors-adventures-in-spectrometry/

In a case of convergent evolution, it turns out that I'm interested in the spectra of lenses too. Starting with my DIY reflection mode spectrophotometer that I built for the purpose of creating printer/paper profiles...

https://www.dpreview.com/forums/post/60459110

...which delivered some good results...

https://www.dpreview.com/forums/post/60793207

... I made the joining pieces necessary for collecting spectral measurements in transmission mode:

Fiber optic in (from the lab illuminator), fiber optic out (a super inexpensive Toslink cable).

Roger, you might recognize that lens.

Inside that PK-13 extension tube I anchored the aperture actuator lever, which let me measure G lenses wide open.

I'm using a purple filter to suppress intensity in the green region (where halogen lamps have their maximum) in order to get better signal to noise ratio in the near-UV and near-IR:

The light isn't super bright after leaving that diffusion disk; nope, it doesn't cook the lenses. They get warmer from outdoor use in the daytime.

The brains of the operation is a high end microcontroller in the Arduino universe. The color touch screen is a great interface. I like colored charts just as much as Roger does.    Total system cost is somewhere around $500. The biggest expense was the spectrometer module I got through GroupGets.

Three types of readings are taken: my "white" light reference spectrum (just don't put any lenses in the path), my "dark" reference spectrum, and the spectrum from white light passing through the lens. The data is processed into % Transmittance according to:

(Lens - Dark) / (White - Dark) = %Transmittance  (do that at each measured wavelength)

I was surprised that f # was not a significant variable. Due to the nature of the optical path I set up, the most significant one was focal length. Longer FL spreads the light out more. The 24mm f/1.8G's transmitted light was more intense than the no-lens arrangement, a big surprise. I had to stop it down to keep it in range. The other big surprise was how very low the signal was in the case of the 300mm. I had to collect a huge set of readings on that one and it's still noisy.

Here's the data of the lenses I have on hand, or recently had here. Several of the more impressive lenses are rentals. First, here's validation that nothing quirky is going on with the data handling:

No lenses in place.

One of those lenses would not be great for IR. The oldest lens is from 1977.

The 300mm gave me only 5% intensity compared to the white light reference, far less than the other lenses.

And wrapping it up is a lens that is considered unusual today, but it made sense in the film era: a fisheye lens with a set of internal filters that can be brought into or out of the optical path using a selector lever. That's not a bad place to install filters on a fisheye lens.

A variable spectrum lens.

 just Tony's gear list:just Tony's gear list
Sony RX100 VA Nikon Z7 Fujifilm GFX 100S
ForumParentFirstPrevious
Flat view
ForumParentFirstPrevious
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow