The tube lens in its assembly:
Here is the tube lens assembly mounted on the Bratcam. Since the assembly has a Chris Hejnar Arca Swiss compatible rail for a base—as do my bellows and macro lenses—any of these can be interchanged quickly with the quick-release clamp:
Parts labeled (the lettering is kept consistent with the earlier post—so it may seem a bit strange, here):
All stock numbers and prices are Edmund Optics, US dollars as of February, 2011, unless otherwise indicated:
A.) Mitutoyo MT-1 tube lens: (Stock number NT54-774, $628). This lens has a focal length of 200mm, and when used with a Mitutoyo (or, I presume, Nikon) infinite objective, provides the magnification rated for the objective. A Mitutoyo MT-2 lens is also available, with a focal length of 400mm, which doubles the objective’s magnification. These are not to be confused with the MT-4/MT-40 tube lenses, which are described as “designed for end-user integration,” and appear to be of more limited application.
B.) MT-1/MT-2 C-mount adapter: (Stock number NT58-329, $159). This is a two-part adapter designed specifically for the MT-1 or MT-2 tube lenses. The tube lens is mounted inside this adapter (see photos later in post) and provides standard C-mount threads on both sides. Note that this adapter has a slightly larger diameter than a T-tube, so the T-tube ring mounts (parts I) can not be placed around it.
C.) Mitutoyo to C-mount 10mm adapter: (Stock number NT55-743, $33). This is a short tube with, on one end, a male C-mount thread to screw into part B, and on the other end, a female M26 x 0.706 mm (26 TPI) thread to accept a Mitutoyo objective. Users of Nikon infinite objectives should note that while this adapter may work for them, the Nikon M26 thread pitch is slightly different—0.75mm.
D.) Male T-Mount to Female C-mount adapter: (Stock number NT58-753, $39). This is for the camera side of the tube lens—bumps the thread size up to T-mount, which permits a wider tube for convenient mounting and to avoid vignetting. The junction between parts D and E is not visible in the photo, as it is covered by the ring mount (part I).
E.) T-Mount Extension tube 100mm: (Stock Number NT52-296, $33). Provides appropriate additional tube length and convenient support.
F.) 35-60mm T-mount Fine Focus Tube: (Stock number NT52-300, $71.00). I have this cranked out a few millimeters so that the tube lens (without objective) focuses at infinity. There is a set screw on this tube that can be tightened to lock the tube after adjustment. I find that it must be locked down for rigidity. Since the Nikon flange focal distance is 46.5mm vs. Canon’s 44mm, a Canon shooter would want to crank this part out 2.5 additional mm. Other camera types would of course differ.
G.) T-mount to Nikon F-mount adapter: Free to me, because I have several, but about $15 online. Canon or other brand shooters would obviously use their appropriate camera mount, here.
H.) Arca-Swiss rail from Chris Hejnar (not an Edmund part—about $40). This came from my box of rails, and had ¼-20 inch threaded holes, which Don Wilson, of Wilson’s Welding & Fabricating, bored out and countersunk in order to accept flat head screws. These screws go from the underside of the rail, up through the rail and into the T-mount rings (parts “I”). (For those who wonder why I had the threads bored out, be aware that directly attaching two parts having threaded holes, via a threaded screw, doesn’t work very well—the two parts will tighten to an orientation that is almost surely not the one you want. So the screw must be freely turnable in the base rail.)
I.) T-Mount 48mm Ring Mount with ¼-20 tapped hole: (Stock number NT52-304, $65). There are two of these in the rig.
Pictures of the Mitutoyo tube lens and holder are scarce, so here are some of mine.
1 Parts B (left) and A (right)—that is, the C-adapter and tube lens. The brass rings partly visible inside the C-adapter are shims for use with the MT-2 tube lens, and not used in this assembly.
2 Same as above, seen from the side.
3 Here with the end of the C-adapter removed, exposing three screws that will hold the tube lens.
4 Here with the tube lens screwed into the end of the C-adapter.
Below is an Edmund Optics Mitutoyo filter holder (Stock number NT56-993, $69). This holder has Mitutoyo M26 threads on both ends, and so goes nicely between the tube lens assembly and objective. In it I placed a 25.4mm diameter Edmund Optics Techspec High extinction Linear Glass Polarizing Filter (Stock number NT47-316, $135). It has a high extinction factor and is very color neutral—and so works well for cross-polarization. As an alternative to this expensive filter, Edmund Optics stock number NT54-926 might be a usable alternative at much lower cost—$23.00—but the specifications are not as good. It would also be possible to mount a polarizing filter in a short C-mount tube with appropriate retaining rings, and place this tube between parts A and C.
Edit: For this filter holder, the combination of the inner threads and the retaining ring that screws into them leaves a 3mm gap at the tightest adjustment. Despite the fact that Edmund optics descibes this filter holder as "compatible" with polarizing filters," the polarizing filter is only 1mm thick. This left the polarizer a bit loose, with 2mm of "jiggle room." While this did not seem to present an optical problem in test shots, it still bothered me, as I don't like having optical elements move randomly in an assembly. However, the addition of a 0.49$ USD hardware-store O-ring has nicely cured this issue. The O-ring has an outer diameter of 26mm and a thickness of 2mm. (A similar 1-inch O-ring was surprisingly a bit too loose.) After the addition of the O-ring, the polarizing filter is now very snug. Thanks to Rob T. at Edmund Optics tech support for helping me work this out.
Here is the assembly with the filter holder and my longest infinite objective, the Mitutoyo Plan Apo 50x. Despite the fact that the lens is cantilevered pretty far out into space, it is steady in test stacks. That said, I’ll probably add additional support, if only because it bothers me to have things sticking out into the air unsupported.
How well does it work? My quick answer is “very well”—but I have not done rigorous testing. After a number of trial stacks, I have not found it wanting. Before, I was mostly using the Nikon 200mm f/4 ED-IF micro lens to decollimate the Mitutoyo objectives. While the Nikon lens did work for Mitutoyo objectives from 2x through 20x, my sense was that it didn’t equal David Millard’s apo lenses intended for large format. That said, I doubt we can truly compare results by different photographers with different subjects and lighting. Interestingly, when using the Nikon micro lens for decollimation, I was never able to get a remotely sharp image with my pickiest objective, the Mitutoyo Plan Apo 50x. With the Mitutoyo tube lens, getting decent results with the 50x is easy.
Also, I think an important test for a decollimation lens is how it performs with the objective spatially separated from it. This spreads out the light waves a bit, which tends to use less optimal portions of the decollimation lens. This is important if one wishes to use a filter, beam splitter, or other apparatus between objective and tube lens. With the Nikon lens used to decollimate, I could not get useful images with a polarizing filter adapter in place. But with this tube lens setup, I can.
Here is a quick test shot with the Mitutoyo Plan Apo 50x (208 images @ 1 micron, Zerene Stacker PMax). Subject is Uraneus ripheus; for amusement, I wanted the shot to include both shiny scales and dark scales, which challenges the dynamic range of my sensor and workflow.
Another quick shot—same objective, same subject, different spot—and this time with cross-polarization. While it may appear a bit less sharp than the above image, I don’t think it really is—the effect seems to be caused more by the harsher direct lighting necessitated by x-pol. (146 images @ 1 micron, Zerene Stacker PMax.) This and the above shot could definitely use some retouching, but these were meant to be very quick demonstrations. Also, I realized after the fact that I did not flock the inside of the filter holder—something I should definitely do. This might have introduced some flare.
A very important note: This assembly desperately—desperately—required flocking. I used Protostar material for this, flocking every possible surface. This made an immense difference in performance.
Things I may change:
1. I purchased, and may add, a C-Mount 30mm Ring mount with ¼-20 tapped hole: (Stock number NT52-930, $65). This is similar to the ring mounts for the T-tubes (parts I), but for smaller C-tubes. I would have to add a short C-tube extension between parts A and C, as part C is not wide enough to occupy the ring mount without interfering with the objective. I don’t think I really need this, but as mentioned above, I have some discomfort in having the objective on such a long cantilever; that said, if it’s working well at 50x, I probably don’t need it.
2. I think I’ll add an additional ring mount (part I) around part F, or just drill an additional hole in the Hejnar rail and move one of the existing ring mounts. Again, this is probably not necessary, but one can never have too much support.
3. Add a second Hejnar rail to the side of the assembly for vertical camera orientation. Vertical orientation is quite possible now, by slightly loosening the screws that tighten the ring mounts and turning the assembly. But I’m not sure I like that approach—a second Hejnar rail would add rigidity and be quicker.
4. Make a beam splitter to go between objective and tube lens for axial lighting.
5. Make a drop-in filter holder to replace the filter holder I described above. I’d like to be able to rotate a polarizer without changing the position of the threading, and would like to be able to drop in a blue filter after focusing (this would allow me to use short-wavelength visible light for monochrome images, but have the full range of light for focusing).
--Chris
-edited to correct typos (longest infinite objective, not finite), and others
I knew you wouldn't wait too long to do this! 
. 
