5x4 Pinhole Camera Scratch Build

Messages
549
Edit My Images
Yes
Decided I need to get my finger out if I'm to have a new camera for Pinhole Day, so made a start this evening. The plan is a 5x4 made from wood, with a pinhole to image plane distance of around 100mm.

Some bits of wood...

5x4 Pinhole 01.jpg

It's all 12mm thick oak, from a couple of offcuts. The sides, top and bottom are cut from a bit 95mm wide, and the front from a bit 145mm wide. The rebates are all 6x6mm, apart from the front which is 12mm wide and 4mm deep. The plain edges that can be seen on the bits here will butt against the rebated surface on the front - the idea is to try and get the front panel a bit further from the film plane. There are two tripod bushes, made using tee nuts with the correct thread. The visible one is in the left-hand side panel, and there's another in the base. There will be internal covers over these to prevent light leaking in.

The bits loosely put together to get an idea of how it will look...

5x4 Pinhole 02.jpg

More 6x6mm rebates on the top, bottom and left-hand edges to make light traps around the film holder, with the right-hand side cut flush. The little ridge on the film holder lands inboard of this part, so the thin strip will be used to extend the flush surface inwards, and a groove cut for the ridge to light-trap the fourth edge.

The front has 'left' written on it to ensure it's oriented correctly when glued up because the centre of the image is slightly offset (6mm to the right). The centre has been marked, but no drilling yet for the pinhole.
 
Don't forget to cover the insides of those T-nuts!
 
Don't forget to cover the insides of those T-nuts!
Already done...

5x4 Pinhole 03.jpg

...along with the groove for the ridge on the film holder and the opening at the front. Not glued up yet - it's clamped together so that I can cut the ends of the groove in the top and bottom panels, which was done with everything in-situ using a small 3mm wide chisel. The main part of the groove was made by cutting a thin slice off lengthwise (on a bandsaw), cutting a rebate with a shoulder plane, and then gluing the thin strip on to make the groove. The resulting groove is about 3mm wide and 2mm deep. The film holder fit is pretty good, about 1mm left to right (constrained by the groove), and maybe 2mm up and down. I'm planning to add some velvet light trap around the rebates which should reduce this and give a more snug fit.

The opening is chamfered on this side just in case there's some vignetting. It doesn't look like there will be, but it depends on where the plane of the pinhole ends up, so better to chamfer now than find out I'm losing bits off the corners once it;s finished.

There's a rebate on the other side of the hole...

5x4 Pinhole 04.jpg

It's 7mm deep and the diameters are 35mm and 25mm.

With the film holder in place and the dark slide open, the distance from the film plane to the rebate is bang on 95mm. The pinhole plane will be a bit forward of this - I reckon the distance to that will be at least 103mm, probably a bit more (unless I change my mind about what I do in this area).

A bit of sanding to do on the internal surfaces, and then I'll do the gluing and finishing. Painting the far edges and surfaces beyond the tripod bush covers will be tricky if it's fully assembled, so I'll probably partially paint the inside, then glue, then sand the outside before doing the rest of the internal painting and the outer finish.

Coming along nicely.
 
Last edited:
That should "cover" the potential problem!

What size hole will you be using?
 
The pinhole will be 0.4mm. According to the Mr Pinhole calculator, the optimum for 100mm is 0.422, but that's a bit hard to make accurately with a drill, and having the hole slightly under size has less impact on sharpness than having it slightly over size. It's looking like the pinhole distance will be 103.75mm, give or take, which equates to f/259 effective aperture. That's close enough to f/256 as makes no difference and lands on a whole stop increment, which will make working out exposures straightforward.
 
Last edited:
Gluing and painting has started.

I decided to glue the sides together and do most of the black painting with the front off...

5x4 Pinhole 05.jpg

A bit of space is left around the edges at the front which will be painted over once the front is glued on, which means I can cover up any glue squeeze-out. It also keeps me clear of the surfaces to be glued without having to mask them off. Similar idea with the back - paint close to the edges but leave space for now. I'll be putting shellac on the rebates where the film holder goes because it'll be a better surface for the velvet light trap to stick to, and a better one to peel it off from if that becomes necessary. (It's also an easier surface to wipe off any wayward black paint if I think I can do the last bit of that without masking, and find that I can't...)

A second coat will be needed, which will need to wait for a day.

In other news, a cheapie UV filter and a lens cap arrived in the post...

5x4 Pinhole 06.jpg

Below them is the remains of the Industar-61 that got robbed to make a loupe for use with large format cameras.
 
Front is now glued on, outside sanded, and a coat of shellac applied...

5x4 Pinhole 07.jpg

The edges and corners are all rounded a little to make it a bit better ergonomically.

5x4 Pinhole 08.jpg

Once the shellac finishing is done (probably a light sand and a second coat), I'll do the rest of the matt black.
 
Looking good.(y)
 
Aye, woodwork always takes a step forward when the finish goes on. :)

Light sand and a second coat of shellac is done, and I decided to mask the film holder rebate before doing the rest of the matt black...

5x4 Pinhole 09.jpg

...which was just as well. I now need to leave that for at least a day to dry.

Before doing this, I had a look through the front with a film holder in place, and it looks like the light trapping is pretty good without any velvet ribbon added. I may still add the velvet, but I'll wait until things are a bit more complete first. Although the plan at present is to use some elastic to retain the film holder, I'm having thoughts of adding some sort of sprung back later that'll let me slide the holder in and out from the side, and that would work better without the velvet.

The next thing is the pinhole. The idea is to glue the front bezel from the Industar-61 to the front of the camera, and then take advantage of the 40.5mm thread. The diameter of the part of the bezel that goes into the hole is 33mm, but the nearest cutter I have is 35mm, so the diameter of the bezel was built up with a couple of layers of stiff card that's about 0.4mm thick...

5x4 Pinhole 10.jpg

A fiddly job done using medium viscosity superglue. That takes it to a diameter of 34.75mm, and is a nice fit with a little bit of slack. The reason for this is mainly to ensure that the pinhole is more or less centred without having to mess about with estimating the mid point. It'll be glued into the camera using epoxy for its gap filling ability.

The bezel allows two things to happen - hold the pinhole in place and be able to fit the pinch-style lens cap (which will be the shutter). Originally, I was planning to use just the retaining ring from the cheapie UV filter, but turns out that 40.5mm threaded parts come in slightly different diameters, and the ring doesn't really grip the thread in the bezel well enough (it kind of half fell in when I tried to screw it on). Plan B is to fit the pinhole into the filter ring instead, and then fit that to the bezel (the main ring screws on just fine). What I don't want to do is glue the pinhole directly to the camera because making it removable gives me the option to experiment with a zone plate at some point in the future. If I have to use the whole filter ring, the distance from the image plane to the pinhole will increase a little, so I'll look at the numbers to check that things will still be okay (should be, I like to check).

It's getting there - just a handful of relatively small tasks left, now.
 
I took the masking tape off last night once the black paint had mostly dried to matt (but still not cured). There were a couple of tiny bits where the paint had gotten under the tape, and these were scraped off using the tip of a scalpel blade. Overall, the transition at the edges between matt black and amber shellac looks nice and tidy.

It's now time to make the pinhole.

Start with a bit of 0.05mm stainless steel shim that's definitely bigger than the filter ring...

5x4 Pinhole 11.jpg

Mark a centre on the tape, stick it down onto a bit of MDF, and drill the hole...

5x4 Pinhole 12.jpg

The drill bit I used for this is a cheapo that isn't very well made. I had already broken the 0.4mm one when I tried to drill by hand using a pin vice, meaning I had no option but to go to 0.45mm. I tried the above using a bit of aluminium drinks can to see how things felt, and found that the hole was fluted or drawn - the drill didn't really cut into the metal so much as spin a hole into it. A close examination of the tip revealed that it's very badly made, with poor cutting edges which were also off centre. I did a trial in a little bit of the stainless shim to see how the drawn shape looked and reckoned I could deal with it, so drilled the proper piece...

5x4 Pinhole 13.jpg

On the left is the top side of the sheet after drilling - you can see that it dips in a little. In the middle is the underside, where the drawn shape is a bit clearer. This side was ground on a very fine ceramic sharpening stone to get the face back to flat, which can be seen in the right-hand image.

If anything, this should result in a slightly better pinhole because the thickness tapers a little towards the hole. This means that the apparent opening for rays coming in at an angle is slightly bigger, which means less fall-off in the corners of the image.
 
The next thing was to make the bit of shim into a circle...

5x4 Pinhole 14.jpg

A circle was marked as a guide, and then black marker used to cover the area before working around it with a pair of dividers. The point with the really narrow tip went into the hole and the other was used to do the scoring. The point in the hole is 0.5mm diameter, but tapers, and the MDF surface was hard enough to prevent the point going in too deep and enlarging the hole.

The shape was then cut out using sacrificial scissors (every workshop should have sacrificial scissors), and a bit of fettling of the edge on the sharpening stone got it down to something that fits snuggly into the filter ring...

5x4 Pinhole 15.jpg

With that done, I fitted the assembly into the Industar-61 bezel and measured from the inside face of that to the shim, and got 12mm. That gets added to the 95mm for the distance from the image plane to the bottom of the rebate in the front of the camera (which the inner end of the I-61 bezel nearly touches), to give a distance from image plane to pinhole of 107mm.

To find out whether this is in the ballpark or not, the pinhole has to be measured. It will work, but I like to know how close, or not, it is to optimum, and to know what the effective aperture is so that I have an idea of how much to bias my exposures from specific readings in the field (ie, how much extra to give it).

Measurement is done by scanning the pinhole at a known resolution, counting the pixels, and crunching some numbers. I do this graphically, like this...

Pinhole Measurement Scan.jpg

This is a screen-grab from Photoshop. The scan is zoomed in, and I make single-pixel marks every second pixel, with indicators at every 10 pixels to make counting easier. Clearly, the edge is somewhat blurred, and I'm not sure if it can be scanned sharper - this was done with the pinhole on the glass bed of my V700, and I used the 10x8 transparency setting (which assumes the neg is on the glass rather than in a holder and above the glass). Scan resolution was 2400 dpi.

The calculation to get from pixels to millimeters is: ( Pixels / Resolution ) x 25.4

It looks like my pinhole is 40 pixels in the white area, and maybe getting towards 42 when it's started to darken (I'm guessing that the scanner is seeing the effects of diffraction at the edge of the hole, so I'm tempted to think that the physical limit is somewhere fairly close to the white area.)

Crunching the numbers for three pixel counts, we get the following diameters in millimeters...

40 - 0.423
41 - 0.434
42 - 0.445

According to the calculator on Mr Pinhole, the optimum diameter for 107mm is 0.436, so I think this is easily close enough. For my range of sizes, I get f numbers of 253, 247, and 241. I'm happy to call that f/256 and be done with it - no exposure biasing needed unless I feel like it.

I'm pleased with how this has turned out, even though it entailed a bit of luck. It's much more accurate than I envisaged it would be, so it'll be interesting to see what the images are like.
 
Last edited:
It was time to have a closer look at the light trapping now that the finishing was done. I set it up on the tripod with two elastic loops holding the film holder in place, darkened room, torch, much peering and maneuvering of the torch. All I could find was some shine from the left-hand side (where the end stop is for the film holder). No amount of faffing about revealed any other leaks.

Only one bit of velvet light trap to do, then...

5x4 Pinhole 16.jpg

This was a bit convoluted. Start with a bit of card, and stick a bit of double sided tape onto it with the backing paper still attached. Then stick another bit of double sided tape onto that and remove its backing paper. Then stick the velvet ribbon onto the upper bit of tape, aligned neatly to one edge. Mark some lines on the card in line with the edge of the ribbon, extend those about 5.5mm down to make a guide for placing a ruler, then trim with a scalpel. This produced a bit of ribbon with tape attached of the right width to fit in the vertical part of the rebate. The original woven edge is uppermost, and the cut edge, which is slightly prone to fraying, down in the corner. A bit of pressing, a going-over with the hoover to get the loose bits out, and check for leaks again - none found.

Time to fit the Industar-61 bezel. I painted the face with the engraving matt black, and bunged it in with some Araldite Rapid...

5x4 Pinhole 17.jpg

There's some glue spooge inside which I didn't see because of the angle the light was coming from, but has been picked up by the camera's flash. Once it's cured, I'll put some matt black paint on it (it was already past its open time and I wasn't able to wipe it off).

Nearly there...
 
Last edited:
Time to cut up some bits of brass...

5x4 Pinhole 18.jpg

1/8" rod, cut to a smidge over 9mm long and with one end rounded over.

Next, add some masking tape to the camera body and draw on some lines centred on the pinhole...

5x4 Pinhole 19.jpg

Use these as a positioning guide for a couple of paper triangles...

5x4 Pinhole 20.jpg

The brass rods are viewfinder pegs, and the triangles (there's another on the right-hand side) are the horizontal and vertical fields of view of the camera. The idea is to work out where to place the pegs visually, particularly on the top because the rear ones have to miss the 6mm deep rebate for the film holder, but not come in so far along the sides that they get in the way of the elastic straps. After a bit of messing about, 11mm was deemed to be the number. On the side, I felt this (minus 6mm (because it's the open end that the film holder sits in) was a bit close so I set those at 8mm from the surface the holder mates with (so 14mm total). The fronts were all set at the same 11mm I had used at the start.
 
Last edited:
I then used a depth gauge and the sides of the paper triangles to make some marks for drilling...

5x4 Pinhole 21.jpg

...and drilled the holes to a depth of 6mm (with an end stop on the pillar drill).

A drop of superglue went into each hole, followed by a peg...

5x4 Pinhole 22.jpg

I did this with the masking tape still in place, on the basis that some glue was bound to come out as the pegs went in, so better to have it on the tape than on the camera. Unfortunately, some of the tape has gotten stuck, and wasn't really coming off when picking it at it with fine tweezers. I'll let it set and see if it comes off easier when it's brittle. If I have to, I can always disguise any messy bits with a touch of shellac.

There are two pegs at the front (and the apexes of the triangles overhang) because the pinhole is forward of the front panel - the two sighting lines cross each other at a virtual point beyond the body of the camera. It would be possible to do it with three pegs set out in a smaller triangle that still has the correct angles, but there would be parallax error at close distances because a single front peg means the cross-over point would be behind the pinhole and the sighting lines would be inboard of the real field of view angles. Since a pinhole camera has vast depth of field, close-ups are certainly an option, so having accurate sighting lines is the sensible choice.

Not much left to do - tidy up the bits at the bases of the pegs, and try a little experiment.
 
I've had success with drilling very small holes in sheet brass and aluminium by clamping the metal between small pads of wood. I use smallish squares of the metal (1 1/2" or so) rather than a larger circle. the smaller squares can then be stuck over a larger hole in a larger circle (or square).
 
I was planning to sandwich it between a couple of bits of thin MDF, but seemed to forget when I actually started drilling. That said, I was pressing down either side with a finger, so it was still decently flat. It seems to have turned out well in the end.

The experiment was to see if I could chemically blacken the stainless shim material, but I had no luck with what I have lying about and my limited knowledge and research. Need to look into this more, but I decided to knock it on the head for now. Some matt black was applied to the glue spooge behind the bezel, and that now looks fine. Leaving the bits of supergluey masking tape around the sighting pegs to harden worked - much easier to chip little bits off with a scalpel. There were still some traces, so I put tiny little touches of shellac on anyway. So, with those done, I think I can say that the camera is complete for now.

The Excelsior Mk1 is born. :)

5x4 Pinhole 23.jpg

5x4 Pinhole 24.jpg

Cost of materials was probably no more than £10 (and nearly £6 of that was the filter and lens cap). Quite a lot of time involved, though - it's a good job I don't have to pay myself for doing this sort of thing.

Next thing to do is cut up some paper to 5x4 and try some exposure tests...
 
Well done, great project. Congratulations and looking forward to seeing your results.(y)

The camera looks very professional .
 
Lens is a bit slow and soft... ;)
 
Well, it appears to work...

Pinhole Images 01.jpg

Old Ilford MGIV glossy (probably at least 10 years old), no filter used, developed in Ilford Multigrade developer at 1+9. No contrast adjustment in post - just a bit of unsharp mask.

The top row and the first on the bottom row are a series of bracketed exposures in 1 stop increments. The third from left on the top was rated at 6asa, as were the other three on the bottom. At f/256, the exposure was 60 seconds, and conditions were sunny with a slight haze in the sky. Not quite f/16 for sunny 16, more like f/11 or f/11.5.

Quite a big shift in density between the bracketed shots, and from the negs, it looks like the unfiltered paper is too contrasty for the conditions - filtering it to bring it down a grade would probably work.

Same image converted to greyscale, flipped horizontally and inverted...

Pinhole Images 02.jpg

Again, no contrast adjustment in post, just a bit of unsharp mask. If anything, I like the shadow detail in bracket 4 (top left here, 1 stop over) but prefer the other parts in bracket No3 (nominal 6asa). I'll see how things look when I scan them. It's interesting to note that the bottom left shot here looks to be lower contrast - this was shot directly into the sun, with the sun in frame.

No sign of any light leaks, which is excellent to see. The two middle shots had the sun over the right shoulder, fairly high in the sky, and these would have been the most vulnerable to leaks. (The bracketed shots had it on the left, which should be the most benign.)
 
Last edited:
The Excelsior Mk1 in action.

The setup for the bracketed shots...

Pinhole Action 01.jpg

...and the view from behind the camera....

Pinhole Action 02.jpg

I took these two, and a third, while in the midst of making the heliographic impression upon the photo-sensitive paper within the camera.


Set up vertically for the shot into the sun...

Pinhole Action 03.jpg

Forgot to dangle the bag for this one. It was a bit breezy, so I'll need to see if there's any hint of camera movement in the shot when I scan it.
 
So when does mass production start??:) cos I want one!

I'd have to have coloured straps I think …..Red, white and blue for France oops I mean England!:LOL:
 
I've been looking at scans of the test shots, and comparing these to the two shots I did on the Marion tailboard camera, which were also made on Ilford MGIV with the same development. The only difference is that the Marion shots were made using a yellow filter (B+W 022, which seems to be a bright yellow, or close to a lemon yellow in painter-speak). Conditions were pretty similar for both the Marion and pinhole shots, with the former maybe having a bit less subject contrast.

I originally felt that the Marion negs were lacking in contrast, but after some messing about with both sets of scans in Photoshop, it's much easier to work with the levels and shadows/highlights adjusters to get an image with a nice tonal range on the shots with the filter. With the two pinhole shots that were close (nominal 6asa and 1 stop over exposed), are both just a bit too much of a rescue job. They're kind of okay, but it becomes clear that the shadows or highlights are being compromised too much when compared to the shots through the yellow filter. Even if there's an optimum exposure that reduces the compromise at one end or the other (rate at 4.5asa, perhaps), it's still going to be tricky to get it right if the subject contrast is fairly high. It's easy to get precise exposures when you're into 1 or 2 minutes, because a few seconds error is tiny, but metering has to be precise to be sure of determining correct exposure first. Being half a stop out would immediately place the highlights or shadows in the range that's hard to deal with.

So, I think the plan will be to use a yellow filter if the subject contrast warrants it, and no filter if it's flat and overcast. Something I still need to think about is filter factor. With the Marion, I gave 2 stops extra exposure for a filter that only needed one, on the basis that the paper might be lacking sensitivity due to its age. Looking at those negs again, the shadows are noticeably grey compared to the white border on the paper, while there's less difference between the light leaks and the highlights. Given this, I think it'll be fine to rate the paper at 6asa and add 1 stop if I'm using the yellow filter.
 
Last edited:
Back
Top