8000ED Image Circle

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

ray_parkhurst
Posts: 3413
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

8000ED Image Circle

Post by ray_parkhurst »

I've been working on the Sensor-Pan system, and have added a simple X-stage to allow me to move the subject so as to keep it centered on the sensor. This has a similar effect as using a wafer target with near exact copies at regular intervals, but I can use my preferred subjects and can move arbitrary distances. Lighting is still challenging since the subject moves more than the lens, so it's still not possible to keep optimum lighting on subject.

I decided my first subject with the system would be the Nikon Coolscan 8000ED lens. My experience with this lens is that it has a large enough image circle and low enough distortion for use in the APS-C Subject-Pan system, but I had no data on quality with larger image circles, which I am now more interested in due to the Sensor-Pan system.

I shot the subject, a 1954-D RPM #1 Lincoln Cent (A CherryPicker's Guide variety), at 1:1. I suppose I should have shot at 1.2x but I don't expect much degradation at 1x and that's where I wanted to use it.

I initially tried shooting with a 2x teleconverter, which I reasoned should work very well since I would only be looking at the center of the image. However, I saw some unexpected rapid degradation from the center, a flattening, and then rapid degradation toward the outside of the IC. The 2x magnification (combined with 2x downsizing) does improve the system resolution due to reduction of the sensor-related aberrations, so the effect may be real and just more visible that without the 2x TC, but I am publishing straight 1:1 images first. I upgraded my lighting after doing the 2xTC images, and will likely re-do them with the improved lighting a bit later.

I'll give my conclusions first, then share the images...

I believe the 8000ED is an f2.8 lens, so at 1:1 I am at EA5.6. This is much larger than I generally shoot with my HRT2i, and the result is false color generation. I usually use my 95PN at f3.3 and 2:1 for EA9.9, and false colors are significantly reduced but not completely gone. With the 2xTC at EA11.2 the false colors are gone. This is relevant because at the center of the IC the false colors were strong, but as I moved away from center they gradually reduced, and eventually disappeared. I don't know the name of this phenomenon, but certainly as you get closer to the edge of the IC, aperture reduces, and I believe this is what I'm seeing with these images. To put some numbers to this, I see strong false colors out to 20mm from the center (40mm IC), but by 25mm they have reduced to what I am used to at ~f9.9. At 30mm they are further reduced, and are completely gone by 35mm. Correlated with the false color reduction is a drop in sharpness.

I also see constant magnification out to 10mm from center (20mm IC), then gradual drop until 25mm, a bigger drop at 30mm, and even bigger at 35mm. I believe this would produce "pincushion" distortion, though the effect is fairly small.

I had to make very small focus adjustments to maintain critical focus on the secondary MM area. I did not note the amount but it was on order of 10um for each 5mm from center, so perhaps 70um total. I was a bit surprised that the lens did not have a flatter field, but indeed my system may be slightly off. I will check this to be sure the focus shift is with the lens and not the system.

So my conclusion is that this lens is nearly perfect for use in shooting US Cents, which are 19mm diameter, on a FF sensor at a bit above 1:1. It is probably also OK for larger coins using a MF sensor or Sensor-Pan APS-C, for instance shooting a US Dollar (38mm diameter). I don't think I'd plan to go beyond 40mm IC with it as the EA reduction is causing visible de-sharpening at 50mm and beyond.

Images taken with HRT2i, no sharpening.



100% Crops of the Date and MM showing false colors and geometry. Measurement noted is Image Circle diameter, not distance from center:
Img_0014_1a1.JPG
Img_0013_1a1.JPG
Img_0012_1a1.JPG
Img_0011_1a1.JPG
Img_0010_1a1.JPG
Img_0009_1a1.JPG
Img_0008_1a1.JPG
Img_0007_1a1.JPG


200% Crops of the secondary MM area showing false colors. Measurement noted is Image Circle diameter, not distance from center:
Img_0014_2a.jpg
Img_0013_2a.jpg
Img_0012_2a.jpg
Img_0011_2a.jpg
Img_0010_2a.jpg
Img_0009_2a.jpg
Img_0008_2a.jpg
Img_0007_2a.jpg

ray_parkhurst
Posts: 3413
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: 8000ED Image Circle

Post by ray_parkhurst »

The system settings for the 8000ED and 105PN are very close so I went ahead and shot the 105PN at 1:1. I will only show the 200% detail shots on a small divot on Lincoln's cheek. I went out to 90mm to see how far it would go. 60mm still looks good, with a little degradation at 70mm, consistent with the 62mm IC spec. Required zero focus compensation vs position except at 80mm and 90mm.

f3.3 (the optimum per my tests), resulting in less false color than on the 8000ED. I suspect the 8000ED would benefit from a little stopping-down.
HRT2i.
Lighting still not fully optimized (will it ever be?).
Img_00111a.jpg
Img_00091a.jpg
Img_00081a.jpg
Img_00071a.jpg
Img_00061a.jpg
Img_00051a.jpg
Img_00041a.jpg
Img_00031a.jpg
Img_00021a.jpg
Img_00011a.jpg

Lou Jost
Posts: 5943
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: 8000ED Image Circle

Post by Lou Jost »

Nice test! Many exotic lenses have bigger image circles than we typically use. We should really take advantage of that, like you are doing.

Olaf G
Posts: 18
Joined: Wed Jan 09, 2019 11:37 am
Location: Germany

Re: 8000ED Image Circle

Post by Olaf G »

The 8000ED projects 63.5mm (specs according to Nikon) on a 57.5mm sensor (measurement done by Enrico Savazzi), i.e. the lens is designed to cover 57.5mm at a magnification of approx. 0.9x. BTW, the dot on the lens is facing towards the sensor, this is how the lens is mounted in the scanner. Your test shows very well the behaviour that can be expected from the 8000ED.

Edit: according to Marco Cavina's discussion of the different prototypes of the 8000ED the lens was designed to cover 55mm (if I understand it correctly).

ray_parkhurst
Posts: 3413
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: 8000ED Image Circle

Post by ray_parkhurst »

A lot depends on how you define "Image Circle". Indeed a bit beyond where I tested, the lens begins to "vignette", ie the image goes dark. Some would say this is the edge of IC, but my (loose and variable) criteria is that the edge of the IC is the point where the sharpness and/or CA have noticeably degraded from the main part of the IC. So from that perspective I'd probably say the 8000ED has a 70mm IC.

Olaf G
Posts: 18
Joined: Wed Jan 09, 2019 11:37 am
Location: Germany

Re: 8000ED Image Circle

Post by Olaf G »

ray_parkhurst wrote:
Fri Jan 29, 2021 5:17 pm
A lot depends on how you define "Image Circle". Indeed a bit beyond where I tested, the lens begins to "vignette", ie the image goes dark. Some would say this is the edge of IC, but my (loose and variable) criteria is that the edge of the IC is the point where the sharpness and/or CA have noticeably degraded from the main part of the IC. So from that perspective I'd probably say the 8000ED has a 70mm IC.
Yes, I agree. What I wanted to say is that the lens is optimized for an IC of 55mm but the usable IC might be larger.
Maybe I do some testing this afternoon with my 8000ED and a Mirex shift setup...

Post Reply Previous topicNext topic