Confused about NA, DOF & step size
Moderators: rjlittlefield, ChrisR, Chris S., Pau
-
- Posts: 88
- Joined: Mon May 11, 2015 9:21 pm
- Location: Hocking County, Ohio , USA
Confused about NA, DOF & step size
I have been using the Zerene DOF tables (Table 2-C) to determine my step size. I guess I'm not understanding it correctly. I calculated the DOF for NA 0.28 @ 0.00718 mm, using the Zerene chart, which I think can be written as 7.18 µm. I'm using 0.005 mm step size for my Motic APO ELWD 10x NA 0.28 objective. In a post I read earlier today, the DOF for the mitty 10x NA 0.28 was stated as 1.7 µm. This started me to question my step size so I found the specs for my Motic objective & it lists a DOF of 3.5 µm. (0.0035 mm). Why is the DOF in the Zerene chart twice this ? I'm totally confused now. Maybe I just don't understand the metric system?
Mitty say 3.5um at Edmunds.
Nikon say more iirc.
We've seen this discrepancy before, Rik dug it out, I'll look for his post.
It depends on a few things of course, (+sensor) but unless you can see banding, I don't think you'd gain a lot of resolution with smaller steps than 5 micron.
The discrepancy is mostly the way Mitutoyo vs Nikon use diffraction vs more classical methods for working it out.
I'll look for it.
.
.
.
.. This is I think the one I sought:
http://www.photomacrography.net/forum/v ... 917#141917
Nikon say more iirc.
We've seen this discrepancy before, Rik dug it out, I'll look for his post.
It depends on a few things of course, (+sensor) but unless you can see banding, I don't think you'd gain a lot of resolution with smaller steps than 5 micron.
The discrepancy is mostly the way Mitutoyo vs Nikon use diffraction vs more classical methods for working it out.
I'll look for it.
.
.
.
.. This is I think the one I sought:
http://www.photomacrography.net/forum/v ... 917#141917
Chris R
Another Chris here. I agree with what Rik wrote at the link ChrisR pointed out, but let me approach seeming discrepancy in a simple way:
First, DOF is a subjective concept--different photographers will make varying decisions about what is good enough. My own preferred increments are about half the size of the ones recommended in the Zerene Stacker FAQ. In truth, a wide range of increments will work.
Second, you want to leave some overlap between images. Motic's DOF number probably doesn't account for overlap; Zerene Stacker's FAQ definitely does.
Third, DOF is different at different wavelengths (colors of light). DOF is typically specified for 550 nm, which is green. But most subjects have colors other than green. Violet light has the the shortest wavelength, 400 nm. If you do DOF calculations using 400 nm instead of 550, you get a smaller DOF as your answer. The convention to calculate based on the wavelength of green light rather than violet has always struck me as odd; I want to use a number that works for my pickiest color of light, not the color in the center of the spectrum.
--Chris S.
Edited to fix important typo in violet wavelength.
First, DOF is a subjective concept--different photographers will make varying decisions about what is good enough. My own preferred increments are about half the size of the ones recommended in the Zerene Stacker FAQ. In truth, a wide range of increments will work.
Second, you want to leave some overlap between images. Motic's DOF number probably doesn't account for overlap; Zerene Stacker's FAQ definitely does.
Third, DOF is different at different wavelengths (colors of light). DOF is typically specified for 550 nm, which is green. But most subjects have colors other than green. Violet light has the the shortest wavelength, 400 nm. If you do DOF calculations using 400 nm instead of 550, you get a smaller DOF as your answer. The convention to calculate based on the wavelength of green light rather than violet has always struck me as odd; I want to use a number that works for my pickiest color of light, not the color in the center of the spectrum.
--Chris S.
Edited to fix important typo in violet wavelength.
Last edited by Chris S. on Mon Aug 08, 2016 7:42 pm, edited 1 time in total.
-
- Posts: 88
- Joined: Mon May 11, 2015 9:21 pm
- Location: Hocking County, Ohio , USA
- rjlittlefield
- Site Admin
- Posts: 23625
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
It's mentioned in the linked post, but to be more explicit: one big source of discrepancy is that Mitutoyo quotes DOF using a "single sided" measure = maximum deviation from perfect focus, while Nikon and most others use a "double sided" measure = total distance from front to back of the acceptably focused slab. Of course those are different by a quick factor of 2.
That's the reason why Mitutoyo says 3.5 microns at Edmunds for their 10X NA 0.28, while the formula that I use for Table 2-C at http://zerenesystems.com/cms/stacker/do ... romicrodof would give 7.015 microns. It's actually the same basic formula under the covers, just interpreted two different ways: lambda/(NA^2) for the two-sided view, versus half that for the one-sided view.
--Rik
That's the reason why Mitutoyo says 3.5 microns at Edmunds for their 10X NA 0.28, while the formula that I use for Table 2-C at http://zerenesystems.com/cms/stacker/do ... romicrodof would give 7.015 microns. It's actually the same basic formula under the covers, just interpreted two different ways: lambda/(NA^2) for the two-sided view, versus half that for the one-sided view.
--Rik
-
- Posts: 88
- Joined: Mon May 11, 2015 9:21 pm
- Location: Hocking County, Ohio , USA