Image 1. This image shows the center of the field at 100 percent enlargement with a Nikkor 400mm f/5.6 ED IF Lens.

This is a lens test of a Nikkor 400mm f/5.6 ED IF AIS manual-focus lens. The star field is of Alcor and Mizar (the two brightest stars in the image) in the handle of the Big Dipper.

It is pretty easy to see that the brightest stars in the image are not round. This is not because of poor guiding or tracking, or poor focus. It is because of aberrations in the lens.

The image was shot with a Canon EOS 20Da DSLR camera body (using a Fotodiox adapter) at ISO 800 with 30 seconds of exposure time, unguided on an Astro-Physics 600E mount with periodic error correction enabled.

The image was focused visually with the 20Da's live focus feature. Seeing was pretty good this night and I thought I could see the first diffraction right at focus at high magnification with the live focus output to a CRT television.

Although the results are not included here, focus was also tested on both sides of the indicated visual focus, but visual focus produced the best-focused results.

The image was shot as a JPEG so some sharpening has been applied in the camera.

Image 2. This image shows the center of the field at 500 percent enlargement.

Image 2 is exactly the same image as Image 1, but enlarged 500 percent to better show the shape of the stars in the image.

It is easy to see that the stellar profile is a misshapen "arrowhead" instead of circular. The results are actually quite similar to those obtained with high-end Canon L series telephoto lenses, such as the 300mm f/2.8 test and 400mm f/2.8 test.

Also notice the bright red/magenta fringe around some of the stars. This is due to chromatic aberrations in the lens, that despite using ED lens elements, that are still present. It is possible that these aberrations are aggravated by the increased red sensitivity of the CMOS sensor with the modified long wavelength filter in the Canon 20Da.

Image 3. Nikkor 400mm f/5.6 ED Lens working at f/8, 100 percent enlargement.

Image 3 shows the center of the field at 500 percent enlargement with the lens stopped down to f/8. The diffraction spikes are caused by the aperture blades of the diaphragm in the lens.

Image 4. Nikkor 400mm f/5.6 ED Lens working at f/8, 500 percent enlargement.

Image 4 is the same area in Image 3 but enlarged 500 percent.

Note that despite the performance improving, the arrowhead shape is still present. Stopping the lens down any further, to f/11, would reduce the usefulness of the lens to the point where it is not worth using.

Thoughts and Observations

Initially I thought the misshapen stars in the Canon 300mm f/2.8 L USM IS, 400mm f/2.8 L USM IS and 70-200mm f/2.8 USM L IS lenses could possibly have been caused by the Image Stabilization (IS) element not being centered, even though the IS was turned off for astrophotography. This test with the Nikon 400mm f/5.6 ED lens makes me seriously doubt that. This Nikkor lens does not have image stabilization, or vibration reduction as Nikon calls it.

So, what other possibilities could it be?

One is that the low-pass, long-wavelength filter is somehow involved with the Canon 20Da camera. Perhaps there is something in the design or manufacturing that introduces some kind of wedge into the image. Perhaps it is camera specific to this one unit... maybe the filter is tilted or something. But I produced these same arrowhead results with a Canon EOS 1D Mark II body on earlier tests, and the arrowheads are not present on stars shot through my Astro-Physics' refractor. Also, going back through my archives, I see the same results with Nikon Digital SLR bodies and Nikkor ED lenses.

This leads me to conclude it is definitely a lens problem. I think that current digital sensors are high enough resolution to be capable of resolving them. These lens designs were made during the days when film was the primary recording medium. Combined with the incredibly difficult test of a star field, we are just seeing things that film usually did not record. I say "usually" because in going back through my files, I did see some examples of something very similar on some Ektachrome 200 images as can be seen below.

Image 5. Nikkor 300mm f/2.8 ED Lens working at f/2.8, 500 percent enlargement.

Image 5, seen above, shows a 500 percent enlargement of a star field near the center of the field in a shot of the North America nebula taken with a Nikkor 300mm f/2.8 ED lens on unhypered Kodak Ektachrome 200 transparency film with the lens working wide open at f/2.8. This film is very good for fine grain and high resolution.

In this image shot on film, we can see much of the same lens behavior we saw for the digital sensors. The bright core of the star, while not misshapen like an arrow head, is nevertheless miscentered, and surrounded by a larger diffuse halo around the bright core of the star.