Press "Enter" to skip to content

Improved 5×7, now with real five and seven

I was so appalled by the state of my shop at the end of the last project that I decided to clean it up and repair a lot of the worn out tables, jigs, and machinery so it would be all ready to go whenever I’m ready for a new project. I dissembled and adjusted the mill, organized the bits and end mills and router bits, tuned up the drill press and made a decent table for it, rebuilt my table saw/router table outfeed table, trued up all the fences and fixtures, replaced the workbench top, and generally organized and cleaned throughout. The shop still isn’t fit for visitors or even a heavily cropped jpeg tour, but it will be much less annoying to spend time in it now. 

Problem is, now I want to spend time in it.  I really need to remodel the other bathroom in the house, re-roof the garage, but those don’t sound like much fun at all. So for now going I’m just going to make a new 5×7,  and wallow in some displaced productivity.

I’m a little puzzled by this because I’m pretty happy with the 5×7 I have. The only thing that’s always bugged me is that it’s slightly too snug in my backpack at 11″ square (including hardware and knobs). I’ve been recycling a set of homemade bellows since I made a 5×12 to 5×7 conversion back about 6 years ago, and this dictated the size of the standards when I finally made a dedicated 5×7. Since I can’t get decent bellows cloth anymore, I’m just going to order a set of custom bellows so I can make the camera whatever size I want.

Mahogany offcuts from some old project. Should be plenty for the camera. There’s some wicked interlocked grain in this stock though, as well as some hidden tensions. I have to rough mill everything pretty generously and leave enough stock to be able to dress everything flat and straight.

I always start with the back. Size of the rear standard will be 9″ square. This will shave off 1″ in each dimension, maybe more depending on the final hardware. I finally bought a dado set for the table saw, and made a new miter slot sled and box joint jig, and some new zero-clearance blade inserts while I was at it. For the joints I always test on some offcuts. For gluing up, I make a inset panel out of scrap melamine that fits inside snugly to make sure everything stays square. Mahogany is open-grained so I mask off to keep glue from filling and staining the pores. For that reason I make the finger joints a little long and handplane them flush later because the end grain really drinks up the glue.


Storyboard for the ground glass frame and film holder location and dimensions. This also provides the dimensions for the stock used on the back.

T-dimension. Indexed drill bits are a handy way to check depth to the film plane against a straightedge. I roll it all over the film plane to make sure there are no high/low spots. A 13/64th bit was just right.

Milling the ground glass mortises on the micro mill. I made a new ‘tooling plate’ to attach to the x-y table out of MDO, which makes it easier to attach jigs and fixtures for cutting wood parts. It’s oversized so it helps keeps sawdust out of the ways. Despite having a new router table top, these small joint are much easier to locate and cut on the mill where the bit is up top and every dimension can be fine tuned by leadscrew. To simplify joinery on this ground glass frame I planned the stock thickness to be twice the T-dimension for the film holder, plus 5/32″ (1/8″ bit for the mortise plus 1/32 slop) so I could flip the parts without moving the fence, make mirror cuts so the mortise would be perfectly centered. To do this you have to climb cut for the mirror passes (upper photo), so a good featherboard prevents the bit from pulling the work away from the fence.  I did cut the T-depth rabbet first on the router though. The styles of the frame will have offset shoulders on the tenon to fit the stub mortises.

September 2nd

Not much progress. The mahogany sprung so much during re-sawing that I figured I should leave the rough-cut stock for a week or so to make sure it’s stable before bringing everything down to final dimensions. It’s a little odd to see so much movement in such small pieces. I laminated the back from 2 layers of 3/16″ stock, overlapping the corners. I’d ripped some 1/2″ pieces to dress down to 3/8″ for the back, but  the parts bowed so much during re-sawing that by the time I planed one surface flat they were too thin to cut the lap joints out of one piece.

In the meantime I made a few shop cabinets to further organize things, but I also went ahead and glued up the rest of the parts for the rear standard I’d already fitted- the ground glass frame and the back. They’re still true after a week, so will probably continue with the rest of the parts over the holiday weekend.  I also went ahead and made the bellows frames so I could get the bellows ordered. I’d heard good things about a bellows maker on Ebay- ecbuyonline2008- but coming from Hong Kong they might take a while to arrive. The surface of the rear frame parts will be more narrow that I’m used to at 11/16″,  so I increased the thickness from 1/4″ to 3/8″ and used some well-seasoned and really stable quarter-sawn cherry. Yeah it seems a waste since the frames are just going to get painted black, but needed something a little more dense and rigid than mahogany or poplar and I’m out of maple.



Clamping up the ground glass frame. A raised platform helps provide room for the clamps. The clamping bed is the same size as the frame to help spread the clamping pressure. A scrap of plywood inset into the glass rabbet keeps everything square, tape around the edges keeps squeeze out from sticking. The back itself was glued up in mostly the same way, except I glued the laminations together first (picture below).

Fine tuning the OD of the back. I always make parts slightly oversized in case a glue-up goes bad. A simple shooting board helps to both square-up the frame and keep all side square while adjusting the fit.  I finally took the time to tune up my chisels and planes, lapping and sharpening them on a piece of float glass with a slurry of aluminum oxide grit and oil down to 5 micron- the same grit I use for grinding focusing screens. The planes leaves a lovely glass finish on the interlocked grain of this mahogany, makes ribbons out of end grain. After it’s sized, a light trap rabbet is cut around the perimeter so it nests into the rear standard frame.

This is kind of an obsessive detail, but for the film holder trips I sized the back opening so the rib lock would be an open rabbet on the top edge of the frame. That way I could install a piece of phenolic to close the light trap. The rib lock is so close to the inside edge I always worry about it splintering off in use, especially mahogany which doesn’t hold thin profiles well. This hasn’t happened yet on any of my other cameras, and probably never will, but it’s one less thing to worry about.

Now that the holder is registered to the back I can start working on the spring design. I’m thinking about using spring wire instead of leaf springs. It’s lower profile than feeler gauge, spring temper, stainless steel, and available in a lot of sizes. I ordered some 1′ lengths of  0.102″, 0.08″, and 0.063″ diameter stock from McMaster-Carr, hopefully one is suitable.


Lens stage. Rabbets are cut into each side to provide light traps for the bellows frame and the lens board. The bellows frame is thicker than the lens board, so the rabbet depths aren’t equal.  Stub mortises join the corners.

September 10th

Waiting on making the camera bed, still trying to decide if this will be a fold-up camera, or a wide-angle type camera that doesn’t fold. There are benefits to both types, but for now working out the details of the movements and that may dictate the style. I’ve pretty much decided on movements. Front: Rise, fall, base tilt, and swing; no axis tilt or shift. Rear: rise, base tilt, shift, and swing. I always put front axis tilt on every camera I make but rarely use it.


  • Quicker to set up.
  • Potentially lighter and stronger, at the expense of movements of course.
  • Wider range of movements are possible, especially front rise, since the standard doesn’t have to nest within the confines of the rear standard.
  • I have some shorter gear racks available, so wouldn’t have to buy more.
  • I already have a perfectly decent folding 5×7, not sure I would be able to improve on it much other than slightly reducing the overall size.
  • Since it won’t fold, base tilts will require much less range of motion, so they could be stronger. Center tilts would be stronger still, since the base of each standard would be fixed and rigid, but I use base tilts a lot to plumb the standards when using drop bed camera angles to increase rise and fall. Also, center tilts on the back have to be pretty modest, or the standards will block a film holder from being inserted into the back in the horizontal position.


  • More compact. This is a huge consideration, alone it almost out-pencils everything else. I’d rather use a small, heavy, primitive piece of shit than a bulky marvel of precision engineering.
  • Long bed extensions are generally simpler to design and build.

I rarely use long lenses so currently leaning towards a non-folder. l’ll start working on the rear movements first since that will dictate the width of the bed, but wait on the base tilt for now in case I eventually decide on a folder.

I had rear rise on my 5×12 but never used it because the design I came up with was pretty terrible considering how heavy and wide the back was on that camera. But after using rear rise on my Arca Swiss I realize how nice it is for fine tuning the framing of a shot. Rear rise indirectly provides additional front fall as well.

Quick mock up for rear tilt and rise out of scrap parts. The rise was a total failure, but the base tilt works extremely well, surprisingly strong. I can’t budge it when locked down. The tilt pivot is a press-fit 5/16″ bronze flanged bearing, which helps prevent slop and makes it stronger. I won’t make the radial slot so long in the final part, even the 45° marks on the part would be overkill for a non-folder. Nice to be able to eliminate the angle braces. I’ll probably use something similar on the front standard as well.
There are a few things that bug me about this tilt design. The locking knobs aren’t anywhere close to where you’d naturally hold on to the standard to adjust the tilt. With stay braces, the knobs are located at a much better leverage point for fine adjustments. On the design above, it will probably be awkward to loosen the knobs and precisely adjust the tilt since they are so close to the tilt pivot. Also, due to this proximity, there’s  a chance that they very act of tightening the knobs will change the angle of the tilt. Maybe some Belleville washers would keep enough tension on the parts when loosened to prevent these problems.

I mocked up a simple slot and pin style of rise out of scrap aluminum (photo above), where a knob tightens the movement down and an inline pin bushing an inch away keeps the two parts from twisting, but the pin and knob need to be spaced apart fairly wide to prevent the leverage from twisting the parts, which would greatly increase the length and weight of the parts by the time some actual space for the rise is factored in. Better to have a key way such as a dovetail to keep the parts aligned. The X-feed on my mill is only 6″,  a lot of tedious setup, moving, and re-indicating the part in needed to slot anything longer. The rear rise parts won’t be over 6″ long, but I’ve always wanted to try milling longer aluminum stock on my router table, so I decided to try it with some WD40 for cutting fluid and really light passes. That way I can cut shorter pieces out of the longer stock, and not have to mill individual pieces.

Tail slot was cut first. I used a 3/4″ dado bit to plow out the center, then switched to a 1/2″ 14° dovetail bit for the final passes.  Worked well, only took about 1/32 off at a time and lightly wet the part with WD40 before each pass. I also turned off the router and cleaned the router table before each pass. It took a while, but still quicker than cranking the X table feed on the micro mill. It’s a little tricky controlling the depth of cut on a router table unless you have a cast iron router table, even aluminum router table inserts tend to flex, so I made a push block out of cherry that spans the width and length of the part and helps keep even pressure on the 3/16″ aluminum bar stock.

Parts cut in two and ends squared up on the mill, one for each side of the rear standard.

Overall the router worked pretty well. The tolerances in the slot are within 10 thousandths of an inch. Not good enough to resist lifting forces, especially with such a low angle dovetail, but decent enough to eliminate radial twisting in rise and fall movements. The rise dovetails will be captive within the mounting brackets, and having one on each side will eliminate any lifting out of the ways.

The pin was not too hard to fit to the tail, I have a micro feed dial on my router fence that allows me to dial in passes at a few thousandths of an inch. I left enough depth between the pin and tail to install a UHMW film bearing between the parts. Both parts were mirror cut- after each pass, the part was flipped 180° for the next pass to keep everything centered. Frequent spring-passes help too- running another pass without changing the fence or bit depth. Deburring the parts between passes on the fitting cuts is important as well.

Aluminum dovetails ways are tricky to get just right. Too tight they bind and gall, too loose and the entire point of having a dovetail is lost. The rear standard will have a lot of leverage over the shift movement, so I’ll take the time to machine that on the mill with a proper 45° or 60° dovetail cutter. The extra angle on the end mill together with the tighter tolerances possible on the mill should help prevent the standard from rocking in the key way. I might also gib the way  by cutting a little relief groove one one side of the tail slot, then slitting a narrow spring kerf next to it on the shoulder of the slot, and then thread in some 2/56 set screws along the edge of the part to fine tune the fit as needed.


Gib detail for rear shift. Angles are drawn at 45° for the dovetail, but should probably be no more than 60° to reduce stress. Metalworking bits seem to measure backwards from 90°, woodworking router bits  measure from 0° up.

September 17th

Settled on a non-folder, so started on the bed extensions. The front slide will be a double extension, should get close to 16″ with all of the extensions combined. Going to use 1/4″ precision ground plate for the top deck that carries the front standard. I squared up the plate stock on the micro mill- barely had enough X and Y travel to square up this 4″x 6″ part.

Cutting lightening holes. Blue sharpie helps the layout lines stand out. The part was clamped to a sacrificial scrap of melamine.

The micro mill is trammed pretty well, but the finish isn’t the greatest.  I used the shooting board and a 1-2-3 setup block faced with a diamond stone to keep the deck square and parallel while polishing out the mill marks.

The nesting part of the front extension is mahogany with a recess cut in to align the plate. A rabbet is cut on the plate flush with the wood, which will be capped with a guide bar.

Delrin extension guides and rear extension slides

I wanted to try out some different materials for this camera, especially for the bearing surfaces. Most of the better-quality composites are pretty expensive to experiment with, so just got a little UWMW and Delrin to try out. The UHWM stock is very slick but slightly spongy. The Delrin seems just as slippery, but more rigid and seems less likely to compress, so for the extension guide blocks I’m using that. I’ve never worked with it before, so tried a shallow test cut on the router. It doesn’t exactly melt it, but the surface does get a little rough, even at low speeds. The mill leaves a really nice finish, but it deflects easily so the part needs to be fully supported. Also, it throws up a pretty serious burr but that cleans up well with a  hand plane.  I rough cut to length with a chop saw, then squared all the long faces on the mill before cutting the grooves.   This required a lot of cranking so made a new drill adapter for the mill hand wheels that I can take it on and off easily, and can use it for either X or Y feed.  But even with power feed this was pretty tedious, so for the grooves I rough cut those on the router table with a 3/16″ down-cut spiral bit, the cleaned up the slots on the mill.

September 23

This is all progressing pretty slowly. I’ve been organizing the shop as much as working on the camera. I finally sorted all my screws and small parts, which took an entire day. Filled 7 part bins, those small cases with about 20 compartments in each. Tired of ordering more parts simply because I can find them in the chaos. Then I build a cabinet to put the parts in, as well as to hold tools. I also rebuilt the micro mill cabinet to reduce the vibration a bit. Then I sharpened the thickness planer blades and aligned the feed rollers with the bed.  I realized I was getting a .005″ taper side to side when running stock through it due to the uneven feed pressure.  I’m spending a lot more time keeping the shop clean and organized, which seems to include me sitting on a stool wondering what to do next.

Picking up the back again.  Got the wire spring, but didn’t inspire much confidence. Decided to go with the torsion spring design again.

Spring rails for back, also locates the film holder. I did as much milling as I could on a single piece of wood, then ripped it down the middle to make two parts. The recess in the middle is for the spring housing, 3/8″ ( 1/4″ ID) aluminum channel. The stub tenons locate the channel, I cut those with a chisel after ripping the part in two. The long stopped rabbet on the inside edge gives clearance for the spring arms, the flats at the ends and full width strip on bottom will guide the film holder in and be the light trap as well. Still haven’t decided if I will screw or glue these to the back. Screws would be a lot better for finishing and maintenance, but might also be a potential light leak unless glued down.

Once the parts were organized, I found some parts left over from the last project. The 10-32 screw are the pivots for the spring arms, which thread through the aluminum channel with the the sleeve bearings inside, which are the mandrels for the torsion springs. Mandrels must be sized so the springs don’t bind when under load, the ID of torsion springs reduce when compressed. The spring pictured are a decent fit inside the aluminum channel.

Radial slot for the spring arm. The spring arm has a pin next to the pivot that goes into the slot and engages the torsion spring. Found a shortcut for hand-milling these radial slots, just use the drill bit that was used for boring the pivot hole and clamp it in the vise.  I fit a scrap of hardwood into the channel to keep it from deforming during milling and drilling. [Yes everything in my unheated shop is rusted out, despite keeping surfaces oiled up through the winter. I don’t spend a lot of money on  tools for this very reason. No reason to buy a rotary table and let that get all pitted as well.]

The spring arms are cut to the same length, tapped for the pivot (10-32) and spring post (6-32), then slotted on the opposite end for attachment to the ground glass frame. I chamfered each slot for a #6 flat head screw. Only one pair of arms really needs to be slotted, but I’d rather cut 4 short slots than 2 long ones due to the thinness of the material. When the spring arms are in the raised position it shortens the distance between the attachment points by about 1″, so I made each slot about 5/8″. The countersinks were cut first with a V-groove router bit, then slotted with and end mill. Stop blocks help for all these steps, but I had to keep reminding myself not to layout 4 identical parts, 2 need to be left-handed and 2 right-handed, the layouts aren’t on center so each feature needs two setup steps. Finally I just marked each arm with blue sharpie and laid out each feature just to be sure.

Afterwards the parts were ganged together and the ends shaped with a file. Then a slight taper was cut atop each arm.  The taper is mostly to ensure that the arms don’t stick up on the ends once the springs are pre-loaded and attached to the ground glass frame, but they also look a little less blocky with a taper. Once everything is installed on the back,  the wood rails will get cut to final length, planed flush with the spring housing channel, and also planed to follow the final taper both front and back.