5×7 film holders

walnut 5×7 holders

July 13-14

Saturday the 13th. Weather is still too bright and dry, so retreating into the cool, dark shop once more. Had some  walnut left from the camera so decided to make a dozen new 5×7 film holders. While Fidelity plastic holders are relatively inexpensive, they sort of feel like it. I’m tired of the odd and inconsistent way they mask film rebates, of darkslides binding in the slots, of the film gate springing closed when loading and scuffing the film, of the bulging and warping in weird spots, of the extra dust plastic attracts. It’s not terribly exciting making holders in a small un-exotic format like 5×7,  but it’s considerably more tedious than making big ones. Parts need to be thinner, tolerances are tighter. But I love the format, and this project diary will give me the chance to make notes in real time, instead of trying to remember every little trick or setup after I’m done.

An untested plan

Spent most of the morning measuring a Fidelity holder and plotting and scheming in CAD. The most time-consuming part was trying to find places to note the swarm of measurements. Dimensions are a little different that classic Fidelity.   The film loading gate is less wide, allowing more a more even rebate reveal on each end of the film, and the side rebates are thinner as well. This allows for a little more film to be exposed overall. Also, the thickness of the holder is a little thinner as well, since I’m using 1/32 for the septum instead of the 1/16″ in the Fidelity. Only 1/32 difference, but adds up to 3/16″ in a stack of 6 holders.. doesn’t sound like much, but 6 Fidelity holders are a little too snug in the pouch, hard to grab one without accidentally pulling the darkslide out of the holder next to it.  The T-dimension and rib placement are the same though, will be able to use plastic or wood holders with the same back and ground glass location.

1:1 print to scale parts from. Calipers and indexed drill bits are a good way to double check increments in 1/32″ of an inch, and to make sure printout is to scale. I wasted a lot of time on all these measurement call-outs, should have just printed a  simple reference rule to confirm the scale!

I set up an impromptu joiner on the router table, really saves a *lot* of time over truing and squaring edges with a handplane. 3 pieces of tape give enough offset to true up an edge with a couple passes. My router table top is truly shameful, but it’s flat where it counts.

Not much to show yet, but most of the parts have been milled to thickness. Just need to be slotted, trimmed, joined, sanded, glued, baffled, finished, assembled…

Micro mill set up for slotting /slitting.  1/32″ slitting blade in arbor. The feather-board keeps the workpiece tight to the slitting saw and the table. Not pretty after 3 different batches of holders, but still functional.

Moire test. From left: Baffle rails, film gate rail, film gate stock (each stick represents at least 3 pieces), and the holder sides cut to approximate length.

Sunday the 14th. Got the slotting done today. Must be 3 linear miles of slots to cut. Some setup details I always seem to forget: indexed drill bits would well for slotting setups as well. Can use the bit as a spacer to set the slitting saw blade height and final depth of cut. After setting the height, make sure to lock down the mill’s depth lock collar too, not just the Z axis- I always forget to do this. Prepare extra stock in each profile for test cuts. Don’t slit too deep in each pass or the blade with heat up and warp, especially the 1/32 blade. For the film recess in the center slot around the septum, switch to the 1/16 blade and re-center- the 1/32 blade is jut too thin and will deflect if you try to move the blade to pare off the 1/64″ for the recess on each side of the septum slot. Bring the dimensioned parts inside at night- the grain raises dramatically overnight on freshly milled surfaces. And I can’t repeat this enough- make plenty of extra stock for test cuts.
I was a little worried these profiles would be too thin, but the walnut is holding the details extremely well. The stock I had left over was quarter-sawn, so no tensions are being released when milling or slotting- the slots aren’t rolling the thin profiles in or out with pent-up tension. Still not sure I will pull this project off- some of the parts are alarmingly thin in real life. Once they’re put together it’ll be more than strong enough, it’s just the stresses of milling all the joints that worry me.
The CAD drawing has been a big help, no errors so far or missed dimensions. Easy to check sizes and slot locations by just placing the part on the plan.

July 21

Joinery and assembly steps (updated):

1. cut bridle mortises for baffle ends on styles
2. haunch cuts for baffle ends on styles
3. cheek cuts for baffle ends on styles
4. bridle tenon cuts on baffle rail
5. cut septums
6. bridle mortise cuts for film gate rail
7. haunch cuts for film loading relief (try #8 before #7 next time, need to have wood there to support the cheek mortise cuts, other wise the cut will blow out the grain)
8. Cheek mortise cuts for film gate flap
   
9. rip darkslide step profile on styles
10. haunched lap joint cuts on film gate
11. dado cuts in film baffle for end light trap (3/4″ bit)
12. rib lock bevel cuts (move fence with 3/4 bit for dado, then bevel rib with v-groove bit)
13. fit darkslides
14. cut slot in baffle rail for v-spring, fit baffle springs
15. mill darkslide pulls (not rendered on plan)
16. bevel top edge of spring baffle rail for darkslide pull clearance
17. drill pilot holes for darkslide locks
18. Glue-up
19. Rabbet shallow recess in film gate for tape hinge (not rendered on plan)
20. Surface prep (I like to gently scrape with a razor blade. Don’t trust sandpaper- don’t want to ruin the tolerances)
21. Finish (rattle-can lacquer works quite well, and is easy to touch up without have to strip down to bare wood)

Tenon jig for slotting bridle mortise. Parts are ganged together to save time. I goofed a bit on the plan- my saw blade isn’t thin enough for a 1/16 slot mortise.  And the slitting blade on my tiny mill isn’t nearly big enough to cut to the depth required. After finally getting the blade perfectly centered, the slot ended up being 3/32. Slot width isn’t critical though, just wanted to make sure I had enough depth on the baffle rail for the shoulder cuts that help keep the frame assembly square and strong.

Slot cut. There’s really no ideal order to cut the joints on the baffle end. Because it’s slotted, then relieved on three sides, the parts are thin and want to deflect out of parallel when the stresses of the bit/blade hit the workpiece. I ended up cutting a bunch of tiny scrap pieces to fit in all of the slots to temporarily reinforce the parts during the following cuts. Pain in the ass, but worth it- parts stayed parallel and no blown out grain at all.

The cheek cuts are made on a router table. I used a 3/4″ mortise bit and a supplemental fence with an over-sized push block to help stabilize the part when cutting. I set the fence  to have the bit right at the shoulder line and made several passes at each depth. Then raised the bit after each pass. Much cleaner than nibbling at the depth on the table saw, and less ridges to clean off with a handplane. Plus, the shoulders line up perfectly around the part.

Temporary slot-filler piece in place; pencil and #2 screw for scale.

Shoulder and haunch cuts for the baffle rail. All of 1/32″ deep on each cheek, and the haunch is 5/16″ tall.

The frames half-completed. Really didn’t get a lot done today. Cutting thin delicate parts can be stressful.. Still, the worst is over. I’ll deal with the film-gate end and the septums next weekend. The sides are still too long… I briefly thought about making these into 5×8 holders instead of 5×7, but lately I’ve been really enjoying squat formats after years of pano work.

After dinner decided to try and get more done. Garolite, 36″x48. Just enough for all 12 holders, darkslides and septums, but had to plan the cuts in CAD to make sure I cut it in the right order.

5 @ 5.25″ rips / 4 darkslides and 1 septum each. The balance left enough for the rest of the septums, 8 @ 5.25″ crosscuts. A standard triple-chip-grind rip blade works well on the the garolite. Just remember to clamp some facing to the fence so the sheet doesn’t get wedged in the gap underneath. My 80 tooth crosscut blade on the chop saw (alternating top bevel) didn’t work so well, fairly chipped the edges. But it’s dull as a tire;  I’ve been using it on aluminum and everything else.  It still works well if I support every cut in an MDF sandwich. But that’s getting old so I’ll pick up a new blade this week.

My 1/8″ bit was the exact width of the film gate rail stock, so attached a block for extra mass with carpet tape, really helped keep vibration and bit chatter down.

Slot mortises for film gate rail. I don’t have a chisel thin enough to square up the ends of the mortise so I ease the edges of the ends of the film gate rail instead.

Septum and film gate rail test fit. The baffle is left wide- the extra width will support the dark slide locks/ l-hooks.

Film load test. Perfect, but wish I would have tried cutting the mortises for the film flap before making the film loading relief cuts.

A suggestion for next time on the cut order- try #8 first- the wood removed at the #7 step made the wood blowout when I cut the open mortises for the film gate flap. It won’t blow out doing a climb cut on one side of the holder, because the bit is climbing into the wood, forcing the workpiece into the fence.  But the rotation of the bit on the other side wants to split the grain. I had assumed I’d be able to flip the part and feed the other side in from the opposite direction, dumb. I made tiny little  fill pieces to support the cut and cut  score lines on layout to make sure the grain wouldn’t split and ruin an entire holder(s).  These mirror / reverse cuts are can be a little difficult to plan out correctly.  Not sure this change in order wouldn’t just transfer the blow-out problem to the flap mortises, but might be worth trying on a piece of scrap next time.
Enough for today.

July 27

Worked on the darkslide pulls and film loading flaps. The pulls can be a bit of a hassle.  I haven’t really been happy with the design of the pulls on my previous holders, but can’t really think of anything better. This time though I laminated maple to the walnut to indicate exposed/unexposed instead of staining or painting the alternate side.

1/16″ strips of walnut and maple. Clamping caul is made out of MDF to locate the thin strips for glue up. They want to curl when the glue hits so the tape holds them in place

Solid wood backing to distribute clamping pressure

A kerf cut into some fir to to hold with parts for planing the edges flush. The two-tone shavings are pretty cool.

Pull stock was slitted with the 1/32″ blade on center, cut to length, then fit with a small rosewood grip. I added a small groove to the pulls and rounded them over after gluing them in place. With a little modification, the planing jig also helped with locating the grip on the pulls uniformly.

Flap end mortises. Also shown is the new profile with the extra depth off the darkslide grooves ripped away. You can either continue the film loading relief  section through the top of the holder style, or rip that relief’s  depth off the whole style. I usually just rip it off, it always seems easier. Plus, as thin as the parts are, the less groove depth the less fragile it seems to be.

 

 Film loading flap stepped baffle.

Assembled

Getting close- just need to do the baffle springs and velvet, then glue up and lacquer. I milled the light trap profile in the baffle rails using a 3/4″ bit for most of the dado and finishing up the riblock with a V-groove bit, just need to cut them to length and layout the screw locations. I did a mock up of the spring baffle today- cut a through-slot near the center of the baffle rail for a V-spring to fit in, so one spring works for both sides. A velvet strip is glued to the leading edge of each side of the baffle rail and a small portion of each side of the spring to lock the spring in place   [updated baffle details below].  I’m out of brass shim stock for the springs so will have to finish up next weekend.
Still need to mill the shallow rabbet so the tape hinge will flush up with the surface of the holder where it wraps around. The tape used for the hinges (Filmoplast-T bookbinding tape from Talas online) is only about 1/64″  thick, but having if flush out with the surface will make sure the T-dimension is consistent across the entire surface, and protect the tape edges  from fraying as well. The holders are 15/32 thick, so with  1 3/16″ wide  tape the rabbets will need to be 5/32 wide on each gate.

Part count before the final push.

August 3-4

Finishing up. Scraped and sanded the parts, cut the nerve-wracking baffle slots in all of the rails, glued together the frames, and lacquered the parts. Turning out well. Stil not wild about this type of walnut- very ‘thready’. Still, the wood is light and stable, no denying that. These are weighing in quite a bit lighter than the fidelity holders. Loaded they are around 6.5 oz each, while the Fidelities are around 10 oz each. A pleasant bonus I hadn’t expected at all.

Laundry day. Spring clips help take the stress out of glue up on several levels. They are very light so they don’t warp the frame, and are surprisingly strong. And simple to place very quickly before the glue flashes over. I was able to glue up all 12 in around an hour. The line I normally used for sizing paper make a great place to keep them ‘flat’ during curing. Top picture- sacrificed a drafting triangle so I could check for square during glue up.

Finishing- septums masked off.  Bottom picture: scraps of garolite were used when spraying the darkslide pulls and film loading gates; they are so light they tend to jump off the table with the lacquer spray hits them. The binder clips again worked well for gluing the pulls to the slides.

1st baffle attempt

I ended up folding the baffles a bit differently- glued the velvet to the spring’s spine so there’s no leading edge for the darkslide to pry loose. The metal strip is folded it half, with each leading edge doubled over for rigidity. On each side, the spring is folded again in half, up towards the top (bottom picture). This series of folds gives a compound spring action to the baffle to make up for irregularity in the spring. The whole assembly floats, so it has to be sized exactly right. Too tight and the spring won’t function, too loose and light streams in. Update- redid the baffle slightly. The crimp of the metal’s edge was allowing ‘crinkles’ to develop in irregular places along the fold, defeating the spring action. No leaks yet, but was worried they would get worse over time. So I removed the velvet temporarily,  cut away the extra fold of the brass shim leaving a single layer, then wrapped a layer of BK-5 cloth around the whole spring. It’s glued to the brass along the spine, wrapped around the brass to cover the entire shim with a little extra room so the brass won’t cut into the fabric, then taped at the seam in the V-crease. After it’s installed in the slot, the velvet goes back in like before. The BK-5 cloth is thin,  light-tight, very difficult to to crease because of the rubber coating,  and so it forms a springy, light-tight, weatherstripping-style bulb on the folded edge that makes up for any irregularity in the brass spring. Also, I glued an additional strip of felt to the leading edge of the baffle rail on each side of the holder, just before the spring, to scatter light a little more before it even gets to the spring.

Bottom picture: The safety hooks were a close fit. When I beveled the edge of the baffle rib for clearance for the darkslide pulls, the pilot holes were compromised a bit. Sort of a nice V profile to it, and still plenty of thread in the wood, so whatever.

Final weight with film, under 6.5 oz. The plastic fidelity holders average about 10 ounces, depending on the vintage. So these are quite a bit lighter- 12 holders weigh around 2.5 lbs less. Nice side benefit I wasn’t expecting.

Ok. Just tested for light leaks and registration so I’m done, finally. Except I need something for memo strips. And the darkslides are a little tight in the slots but whatever. 3 tedious, fussy projects in a row and I’m officially tired of the shop.

Resources

Forgot to add a list of sources for materials-

• Film gate hinge tape (Filmoplast T 3cm x 10 meters from Talas) http://apps.webcreate.com/ecom/catalog/product_specific.cfm?ClientID=15&ProductID=21116
• Brass shim stock. .002″ x 6″ x 20′. (I also tried using SS shim stock for the trap springs but it tended to tear once folded. ) http://www.mcmaster.com/#catalog/120/3790/=qszvwi
• Darkslide locks (simply .08″ zinc-plated L-hooks available at any hardware store but cheaper in bulk from McMaster. Wish I could have found some with a slightly thinner wire diameter. ) http://www.mcmaster.com/#catalog/120/2981/=qszurr
• Flathead screws for the light baffle rails (#1 x 3/8 stainless steel sheet metal screws) – http://www.mcmaster.com/#catalog/120/2988/=qt005q
• Garolite XX grade from McMaster as well. Black, 1/32″ thick. (Each batch should be tested for opacity. I actually check each dark slide individually.)  http://www.mcmaster.com/#catalog/120/3637/=qszv77
• BK-5 cloth from Thor labs for the light trap fabric. http://www.thorlabs.us/thorproduct.cfm?partnumber=BK5
• I also used thin strips of low-pile velvet as a light trap seal for the flap, and to keep the film from sliding under the flap as the holders get jostled around in the back pack. You guessed it, McMaster- http://www.mcmaster.com/#fabric/=qt01cb
• For the misleadingly simple memo strip, it’s been sort of an ongoing struggle to find the a decent surface that will allow 1000s of cycles of write/erase. I tried using a ‘Dymo’ printed strip with the holder number (and a bunch of extra spaces so it would print out longer for notes), but it’s not very durable and the adhesive is so tacky it ends up pulling the wood fibers up on removal. And it’s too glossy to write on with a pencil. I also tried some special dry-erase strips, but you have to use a specific pen so that didn’t seem too practical. I also tried what I  I’ve used before-  white nylon “V-Flex” weather-stripping that has an adhesive strip on one side. Once the V fold is cut off at the crease it’s a decent dry-erase surface, at least for a pencil. It does help to scuff it a little with some Emory cloth to give it a matte surface. Have not tried ink on it. http://www.amazon.com/M-D-Building-Products-3525-Weatherstrip/dp/B000CSGEIC. But this isn’t ideal- very difficult to write on since it’s so smooth, the adhesive tends to delaminate from the strip completely, just leaving a bare strip of adhesive on the holder, and when it does hold the adhesive has a lot of creep- the strips can slide around on the holder over time when jostled against nearby holders. Pretty ugly.
  In the end I decided to use melamine foil- the ‘self-edge’ lamination strips used to hide the core of shelves and plywood. The 13/16″ size works well. Helps to sand the edges before installing to get rid of some of the extra glue so it doesn’t squeeze out and make a mess. The melamine surface is the closest thing I’ve found to the traditional Fidelity memo strip. Very easy to write on with pencil, doesn’t smear off accidentally before you can unload the holders for development, and comes with a choice of pressure-sensitive adhesive (PSA), or hotglue/iron-on. Since regular PSA has a nasty habit of pulling up the wood fibers of this walnut on removal, I decided to use the hot glue version, which is reversible with heat. (You can also get it without glue if you want to use something else). I unloaded the holders before I ironed the strips on, but with a hot iron the glue melts quick before the heat spreads to the rest of the holder.

I’m not sure how much wood went into this project, but including the test scraps and the waste truing the parts up, I wouldn’t be surprised if it’s 2-3 board feet or total. Having a large chunk of 8/4 stock would have helped- you can re-saw as you see fit for the most stable grain, and of course the color matching is easier, if that matters. These holders came from scraps from several different boards from the camera project, the I had to get more when I realized I was short. but I lucked out in the stability of the wood, and how well it holds details and thin profiles. Film holders are such an intensive project I frequently hesitated and fretted and wished I would have gotten a different wood to start with- cherry is probably the idea wood for holders, and camera projects in general. Relatively light, stable, and tight grained. But despite my hand-wringing, this walnut did turn out well for this project, and it’s kind of nice that they match the camera, and the very ridiculously light weight of the wood is great, so I’m glad I followed my most frequent advice. Dude: Shut the fuck up already and get on with it.

 

July 2014

I get asked this, so after using these heavily for almost a year there are a few things I would do differently. I’d get some real fingerstock spring material for the light trap baffles instead of bending the shim stock-which just isn’t true spring temper. Also I’d make the haunches on the baffle end of the styles a little deeper, or maybe key the spring into the haunch so this fit isn’t so critical to keep light from getting into the darkslide slots. On the flap end I might make the rail with more of an astragal profile as an extra precaution against light leaks. But these things are just a minor overkill issues, no problems with light leaks, and they’re still a joy to use. Every time I pick up a plastic Fidelity holder, I remember why I made these in the first place. It was definitely worth the effort.