Building a macro adapter for the Fuji W1 3d camera.

My description of a prototype prismatic adapter for macro photography with the Fuji 3d camera inspired at least one reader to implement the idea in a practical device. Paul Turvill sent me a series of emails detailing his construction (with acrylic and front surface mirrors), and has granted permission to put it here, along with early results.

All of the stereo pairs are presented for cross-viewing.

     


Mr. Simanek,

After reading your excellent review and the accompanying article on building and using a macro attachment for the Fuji FinePix W1 camera, I decided to purchase the camera and attempt to create a macro adapter of my own.

I've attached a series of photos illustrating the result of my efforts.

Fuji_01 - Fuji_03 are overall views of the adapter, constructed mostly of 0.080" thick styrene sheet, and using 22x29mm first surface mirrors from Edmund Optics. At $7.20 each in quantities of six, the mirrors are much more affordable than optical prisms. For added portability, I also constructed a snug-fitting cover of 0.080" and 0.030" styrene sheet to protect the mirrors from unnecessary exposure to dust and other elements while the unit is being transported.

Fuji_04 is a cross-stereo view of the adapter, with its cover installed. The tripod screw-and-socket was salvaged from an old camera case, and not only holds the W1 snugly in place, but also provides for using the camera and adapter combination on a tripod (a virtual requirement, since the adapter effectively defeats the camera's built-in flash!). This stereo shot, obviously, was made without the adapter.

Fuji_05 (also cross-stereo) is the first shot I made with the completed adapter: the workbench at which the adapter was created.

Thanks again for the camera review and the adapter article.

Regards,
Paul Turvill

Donald,

Thanks for your kind words.

So far, I can find very little fault with the Fuji W1. Admittedly it is a "niche" product, but I'm mystified by Fuji's apparent reluctance to aggressively market it in North America. Could they really have been so put off by a couple of bad (mostly uninformed!) press reviews? I too am (mostly) retired, and yes, my interest in photography, both 2D and 3D goes back many years. My sister and I once had a huge collection of Viewmaster reels, and how I wished I could have afforded one of the handful of stereo film cameras when they were first marketed in the 1950's. My current interest in stereo work is more in the area of things like the Holmes Stereoscope, and I've made a number of nice 3.5x7" cards using various techniques (usually a slide bar) with one or another of my current digital cameras. So the "all-in-one" Fuji W1 was a natural for me. Even if Fuji never come out with the "W2" that some of the critics are asking for, I will not have missed out.

As far as my work so far with my adapter, the 22x29mm mirrors require the camera to be set at or near its full zoom capability. This has the advantage of allowing macro-style photos to be taken with the camera at some distance from the subject; I've had pretty good luck down to about 18" or so. The one small inconvenience I'm finding at these distances isn't so much with the W1's autofocus ability, but rather with the auto-parallax feature. At 2-3 feet, autoparallax works fine, but when working a bit closer, I find I usually have to prefocus and then manually adjust the camera's parallax to get reasonable image alignment. Not difficult, but it is an additional step.

In my opinion it is not necessary to fiddle with the camera parallax controls. I leave that set on "auto" and let StereoPhotoMaker take care of alignment and stereo window position. However, I think it is important to ensure that the prisms preserve nearly parallel axes, without convergence. As far as I can determine the camera's "converge control" does not modify lens axis convergence but only controls the placement of the "stereo window" frame. This would be of some importance if you intended to view the pictures without further "processing", but StereoPhotoMaker software accomplishes the same thing and lets you control the results.—D.E.S.
As you surmised, I do use Stereo Photo Maker (SPM) for my stereo work. However, I was fortunate enough to achieve what appears to be excellent alignment of the four mirrors in my adapter, and the angular and vertical adjustments applied by SPM are actually very small. When I undertook the design of the adapter, I decided to try a rather simple visual alignment method: I first cut a styrene "base plate" to the correct size (24x106mm) and then scribed its two centerlines directly on the material. Next I laid out and scribed lines representing the four 45-degree mirror faces, with their centers at 12mm and 38.5mm from the centerline, respectively, and finally a set of lines representing the optical path from the 24mm spaced entry "windows" to the exits, spaced at 77mm to match the camera's lens spacing. Next, I scribed a set of lines parallel to the mirror faces but offset by the 1.6mm thickness of the mirrors; this gave me the placement for the 22x29mm styrene rectangles which serve as the mirror supports.

The next part is where a bit of "cheating" took place. One at a time, I glued the outer (77mm) mirror supports in place and with double-stick tape applied a mirror; then before the glue (Testors styrene cement) set up hard, I adjusted the position of the mirror so that the scribe lines representing the optical path were correctly reflected in the mirror -- that is, they appeared to be continuous into the mirror and perfectly straight and level. I was pleaseed with how easy it was to spot even very small deviations using this method. The "advertised" setup time for the Testors cement is 90 seconds, but I found that I had perhaps 5 minutes or so before I could no longer make adjustments. Once satisfied with the alignment, I blocked the mirror supports in place, and let everything sit until the cement fully hardened.

For the two center mirrors, I first glued the two supports together to form a right angle, carefully adjusted to be exactly 90 degrees, with both mounting surfaces exactly vertical. Then, with both mirrors mounted, I first test fitted the assembly in place, and then glued and adjusted in the same manner as the first two mirrors. With the baseplate and its 4 mirrors sitting on the bench, I used a laser pointer to verify the alignment -- that is, to be sure that two parallel beams entering at 24mm apart resulted in parallel beams leaving at 77mm apart. Again, I was pleased (and more than a little bit surprised!) with the result.

Finally, I made another 24x106mm piece, and glued it to the top of the mirror assembly. The original "base" actually became the top of the unit, because this effectively hides the scribe lines from casual view. Once the "mirror box" was done, attaching the endplates and base was a pretty straightforward process. The bottom camera mounting plate actually became a lamination of three 0.080" sheets of styrene, partly because of the salvaged tripod screw I was using, and partly to add enough rigidity to keep the camera correcly aligned with the mirrors.

I'm kicking myself now, because in my fervor to build this thing and try it out, I neglected to photographically document its design and construction. I'm not sure how soon I may undertake a second effort, but if and when I do, I'll try to remember to keep a camera handy to document each step. I spent the better part of two days making this unit, so I'm not sure that manufacturing them would be a viable option. During my second career (after my first retirement) I tried my hand at software development, and found that the few sales I made never compensated for the hours of "support" expected of me, and the frustration of dealing with those who simply copied my ideas. I finally ended up putting the ideas and basic principles into the public domain and letting end users deal with the problems. Hobbies are supposed to be fun, after all.

Regards,
Paul

P.S. —As for the "public domain" thing, if you wish to publish any of my ideas, photos, etc., on the Web or elsewhere, please feel free to do so. I don't mind "credit where credit is due," but of course, neither can I accept responsibilities for the successes or failures of others making use of the materials.

Donald,

Some of your comments got me to thinking...I'll probably be ordering parts & materials for as many as five more units similar to my first one. The major additions will probably be a 58mm adapter ring (compatible with my Canon EOS lenses) to accommodate closeup lenses, and an off-camera flash arrangement, using a low-cost "remote" sensor and hot shoe extension cord. The latter will take a bit of experimentation, but it should allow quite a bit of flexibility in lighting.

My current motivation in the 3D area has been in recording scale models: my HO-scale (1:87) model railroad and my wife's 1:12 and 1:24 scale dollhouses and furnishings. Some early "adapter enabled" RR views are attached. Note that the adapter permits the HO models to be shot with a "p.d." of "only" about 7 feet, as opposed to the 22 feet represented by the W1's unaided lens spacing. The 7-scale-foot spacing still creates some stereo exaggeration (which I like) but not to the extreme provided by the 22-scale-foot spacing (which I find a bit disconcerting). These were all shot using normal room lighting with the camera & adapter on a tripod and 2-second self-timer. The "noise" from the high ISO setting is fairly evident, so I'm hoping to produce better work once I get into some better lighting situations (we've a lot of rain and virtually no outdoor shooting opportunities here for a couple of weeks, so I haven't had much chance to play with good old fashioned sunlight).

Regards,
Paul

Sawmill.

Sawmill.

Icehouse.

Mine Head.

Donald,

I've made a couple of modifications to my original adapter.

First is the addition of a 58mm filter adapter ring, made from the back half of a low-priced "stack cap" set. I cut a rectangular opening to more-or-less exactly fit the opening for the two front mirrors, and then mounted it with a couple of 2-56 flathead machine screws tapped into a couple of small styrene blocks glued to the original housing. I'va attached a couple of test macro shots using +2 and +4 close up lenses from Tiffen.

I've also added an accessory shoe bracket designed to mount an optical remote flash trigger, which in turn will trigger an external flash unit. Unfortunately, the Fuji "Super Smart" flash logic has so far OUTsmarted my efforts. The Fuji system appears to produce a brief flash pulse a short time prior to actually triggering the shutters; then, a few milliseconds later it triggers both the shutters and the synchronized flash. My "slave" flash triggers on the early pulse, and can't recycle quickly enough to mimic the flash that's actually synchronized with the camera's shutters. As a result, I'm getting a nice, bright off-camera flash just before the shutters operate, which, of course, does absolutely nothing useful. Somehow I have to "desensitize" the optical trigger to the first pulse, and still allow it to "see" the synched pulse. I'll keep working on this issue, time permitting.

Photos attached.

The 58mm adapter ring in place in front of the mirrors.

Felt-tip marker pen. Taken with a 2 diopter lens.

Electrical plug prongs, taken with a 4 diopter lens.

Regards,
Paul


Closing note by DES.

Next we must tackle the flash issue. Unfortunately the Fuji's flash is located between the lenses and any macro adapter of this design blocks its light. One solution would be an external slave flash triggered by the Fuji's flash. The Fuji 3d uses TTL (through the lens) metering with one pre-flash to determine the flash exposure for the on-camera fill flash, and has no way to shut it off. Also, older "hot-shoe slave" adapters are notoriously incompatible with certain digital cameras and certain flash units. There are inexpensive units, the Seagull SYK-5 flash slave units "for digital" which may have quality issues. Several models of Wein adapters, such as the HS-YLD are sold, again with reported compatability issues. I have tested the Neewer brand slave flash, which can be had for less than $10 from Amazon.com. It has a slide switch that can be set to ignore 1, 2 or 3 pre-flashes. It works reliably with the Fuji 3d, and should work with most digital cameras. It is not a high output flash, but for macro photography the camera's own flash provides sufficient light, and with a little more tinkering we can actually use it with this adapter. See a macro adapter with flash.

Return to the the 3d and illusions page.
Return to Donald Simanek's front page.