Filamentus Headlampus with new CFI60 (and nasty CA)

Images taken in a controlled environment or with a posed subject. All subject types.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

abpho
Posts: 1524
Joined: Wed Aug 17, 2011 7:11 pm
Location: Earth

Filamentus Headlampus with new CFI60 (and nasty CA)

Post by abpho »

Picked up a new Nikon CFI 10x from Nikon Instruments in Toronto. The Mitutoyo is still way outside my spending limit. I also changed my headlamps today and decided this would be a good time to test my new lens.

The results are okay, except for chromatic aberration being the worst I have ever seen it. I tried two different diffusers for the LED lamps that I have. This is by far my best results to far.

Image

Any suggestions? I will try more later. A hood and maybe mask the rear of the objective/adapters to cut down on reflections. Both of which have not done much good for me in the past.

Here is an editing versions. Slightly different crop. The left most coil in this image is the right most coil in the above image.
Image

Thanks.
I'm in Canada! Isn't that weird?

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Yeah, shiny metallic surfaces sure don't make the Nikon CFI's look good.

To tame the CA, it helps to use very diffuse lighting. Don't think in terms of diffusing the lamps; instead think about surrounding the subject in a white cloud that's illuminated by the lamps. The more opportunity the subject has to see a bright spot of illumination, the more opportunity it has blow out highlights and throw purple light into nearby dark areas. I shot my filaments by wrapping tissue paper around the whole thing.

Also for this type of subject I've found it helps to carefully adjust that pesky DMap slider so as to make all space between the wires go black in preview. Those bright facets tend to produce pretty sharp edges even in their expanded OOF renditions. I suspect that mostly what we're seeing in the space between the wires is edges of OOF blobs.

There's a small area of what looks like focused detail that puzzles me. It's located near the right side of the first image, just below & left of center in the second. Maybe the stack went back a little farther it needed to, and picked up some real detail back there?

--Rik

canonian
Posts: 891
Joined: Tue Aug 31, 2010 4:00 am
Location: Rotterdam, Netherlands
Contact:

Post by canonian »

In this case with a subject consisting mostly of silvery tones, you can slightly reduce the purple fringing by toning down the saturation to almost colorless.
It's cheating, but with subjects like these it helps.
The first winding of the coil (second image) is blown out and shows a lot of ghosting on the right.
Maybe the angle of one of the LED lamps, or underexposing could help a bit?

maciek
Posts: 49
Joined: Thu Jan 03, 2013 9:42 am
Location: Poland Masovian Region

Post by maciek »

Recently I did some stacking of filaments so allow me to post a picture which might help to assess if the setup is appropriate.
Here is what I consider the adequate picture with smooth OOF transitions - with no sharp edges. I use styrofoam diffuser - two layers in front of light source if one layer is not enough and no direct light at all.

http://www.photomacrography.net/forum/u ... _800_1.jpg

Maciek

g4lab
Posts: 1437
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

That is not chromatic aberration. It is really there. It is spectral dispersion of the big blue peak that is contained in "white" LED lighting. The lens is fine.
Take the same picture with quartz halogen or xenon arc lighting(including flash)or solar daylight and you will not have that blue chromatism.

This is why I dislike LED lighting. Also they cause a lot of metamerism too.
They call them Daylight but they are not even close. A shame too because they have so many other wonderful characteristics.

The big blue peak around 460 nm is a very obvious characteristic of the LED spectrum. My Leica microscope rep was showing me what was otherwise a really nice value in a stereo scope. The MZ series which sells for $2400 new.
Had a very nice image. I brought out my uncapped integrated circuit chip which has instead of gold bonding wires, little chrome colored bands. There was a blue image next to each one. Totally not there with fiber optic quartz halogen lighting.
Last edited by g4lab on Sun Mar 24, 2013 12:18 pm, edited 2 times in total.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

g4lab wrote:That is not chromatic aberration. It is really there. It is spectral dispersion of the big blue peak that is contained in "white" LED lighting. The lens is fine.
Hhmm... So what causes the dispersion, if not CA?

I agree that the big blue peak can make CA look worse, by putting more energy into a band that is not handled well by the lens. You make a good point that the illumination spectrum can affect how CA appears in the final image. I do not recall seeing this tested or demonstrated. It would be a good project for someone who has both this lens and an apochromat, and some spare time.

--Rik

Charles Krebs
Posts: 5865
Joined: Tue Aug 01, 2006 8:02 pm
Location: Issaquah, WA USA
Contact:

Post by Charles Krebs »

Not really the same subject matter, but I've shot darkfield images (using tungsten as well as flash) where this same amorphous blue "dispersion" is clearly seen in photographs using an achromat objective, but nearly completely gone when switching to a good apochromat objective. Under identical conditions the apo clearly eliminated (or at least greatly diminished) this error.

There is no question about the blue spectral peak (at about 450nm) in most LEDs, especially the "daylight" or "6000K" ones that many seem to like. It would not be surprising that a light that puts (as Rik said) "more energy into a band that is not handled well by the lens" would make things even worse.

So I think both things... light characteristics and lens correction are "at play" here.

g4lab
Posts: 1437
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

I think it's a combination of the point nature of the LED light source, the high reflectivity and also slit shaped nature of the subject matter and the apparent distance to the source.

If the lens has that much CA you should see it with any source.

It is my opinion that it is showing you what is there. I can't prove it because I have just sold my only LED ringlight. A Volpi. I just don't care for them.

I am happy to have my opinion corrected if it is wrong. :)

Post Reply Previous topicNext topic