Chroma - Lasercut Acrylic 4x5 Field Camera

Half of the base (the other piece is 6mm thick) with the focussing rail fitted;

IMG_7780.JPG

Half of the upright fitted, showing the revolving section fitted in place.

IMG_7783.JPG

Part of the lens board and fittings to hold it in place in the front standard.

IMG_7786.JPG

Part of the rear standard insert which will have a piece of 2mm Acrylic bonded to it to provide the friction fit for the DDS holder.

IMG_7788.JPG

It all looks a bit disjointed at the moment but I'm happy with the size/fit so far.
 
Last edited:
Are Sketchup's pseudo circular straight edges tangential (ie outside) or connecting points on the circle (ie inside)? Could make a difference to the fit of parts. I'd be guessing the latter, inside the circle, so sockets will get bigger with sanding, and male parts smaller... could maybe leave things a bit loose?

(I'm really nervous about the potential for deliberate misinterpretation here!) ;)

Erm, good question....basically, rather than being a smooth circle, the Sketchup models export as multiple short straight edges due to the amount of edges configured as standard. I'd increased the number to around 200 (roughly) but once it's cut I can see that it's not completely smooth. The other software in using is a dedicated desktop publishing application so draws several hundred edges as default instead so creates a perfectly smooth circle.

As it is, the part still rotates smoothly enough so won't be a problem but is something I'll correct for the next cut.
 
Usually the ends of the segments lay on the circle.

And all software will describe a curve as short segments, both on the screen, and at export. It's just a matter of degree. Often you can increase the number until it looks like a smooth curve or circle. Even with many hundreds of segments, PCs these days, don't have a problem with the extra amount of data.
 
Last edited:
Usually the ends of the segments lay on the circle.

And all software will describe a curve as short segments, both on the screen, and at export. It's just a matter of degree. Often you can increase the number until it looks like a smooth curve or circle. Even with many hundreds of segments, PCs these days, don't have a problem with the extra amount of data.

Yep, that's exactly the issue/description :0). I guess I could set a much higher number of edges in Sketchup but I think the DTP software (Artcam) works better with the laser cutting software so I'm best using that.
 
There's plenty of CAD software that will export a circle as a circle. AutoCAD and similar tools can export a circle to a DXF file, and other tools will read it as a circle - there's no faceting involved. Same for generic 3D files like STEP and IGES. Also, gcode has commands for describing circles and arcs. It's entirely possible to limit any faceting to the actual XY stepping of the machine doing the cutting, and that is often extremely fine - ultimately determined by the size of the circle and XY positional resolution of the machine.
 
True. But as soon as the shape will be used, either on the computer screen or by a cutting machine, it's back to the little segments again. The cutting machine, as you said, will have it's own segment discretisation.
 
I updated the models last night to add another layer for a few pieces that need to be engraved with 1mm rebates rather than just a basic through cut. Hopefully they will be cut today so I'll find out.
 
Are the wood/MDF pieces staying or are they just to try out until plastic parts are made?


Steve.
 
Just to give an idea of size, this is where the DDS will mount to the rear standard.
Can I suggest a small change to the interlocking tabs? Make the stems thicker to increase strength. And the flanges shorter so there is less leverage to snap them, yet they'll still perform the same job. The opposing part (between the tabs) is thicker, so it could go a bit smaller without weakening too much.

thickness.JPG
See picture showing various degrees of thickness:
 
Last edited:
Can I suggest a small change to the interlocking tabs? Make the stems thicker to increase strength. And the flanges shorter so there is less leverage to snap them, yet they'll still perform the same job. The opposing part (between the tabs) is thicker, so it could go a bit smaller without weakening too much.

View attachment 100337
See picture showing various degrees of thickness:

Thanks for the tip, I'll test the rigidity once it's all assembled. Just to add, this is only the two central 3mm sheets on the rear standard that hold the rotating ring for the back. There are another two sheets of 6mm parts that need to be cut and will be sandwiched either side of this so that should add to the strength.
 
Thanks for the tip, I'll test the rigidity once it's all assembled. Just to add, this is only the two central 3mm sheets on the rear standard that hold the rotating ring for the back. There are another two sheets of 6mm parts that need to be cut and will be sandwiched either side of this so that should add to the strength.

Am I right in thinking that the part of the base shown above with the focussing rail goes between the narrower gaps in the upright part (ie, between the tabs that stick out rather than the stems)? And that there's another part of the base with wider bits that go between the stems of the upright? And, are all of these parts to be glued together?
 
Am I right in thinking that the part of the base shown above with the focussing rail goes between the narrower gaps in the upright part (ie, between the tabs that stick out rather than the stems)? And that there's another part of the base with wider bits that go between the stems of the upright? And, are all of these parts to be glued together?

Yes, there is another 6mm layer to be added to the base (below the focusing rail) along with two more 6mm pieces either side of the current rear standard. Once they're all assembled I'm planning on their interlocking fingers being a tight enough fit to stay in place with friction ao it can be disassembled. If I'm not happy with the strength of that, I'll add a bracket at either side to lock them at 90 degrees.

Existing cameras like the Intrepid use more traditional metal brackets to hinge the front/rear standards to the base and allow the camera to fold down for transport. My intention is to simplify the design to the point that it can be quickly and cheaply machined from sheet material with no extra metalwork required.
 
I'd be careful with the idea of the interlocking fingers having a good friction fit, especially if it's to be assembled and disassembled. Regarding fit, each part needs to have a 'perfect' size that will allow the whole to work - anything outside of that will either result in slop or be a very tight fit such that assembly will be difficult in the first place. All components have tolerances, and the sizing needs to be spot on to get friction. Even a handful of microns out can change it from friction to sliding, or to some small amount of damage or surface wear if tight. Aside from that, no matter how snug it is to begin with, there will be wear during assembly and disassembly that will eventually turn the fit into a sliding one in any case, at which point, there is a risk of slop (once you have slop, wear reduces markedly).

Whether or not slop is an issue is down to the size of the gaps between the surfaces, and in what direction the slop might be a concern. With a sliding fit, the back could move away from the base, causing the film plane to be in an unpredictable position. Similar thing with unintended swing. There is also a risk of tilt, which would feel like the rear standard wobbling at the top when in use. The movements might be slight and not have an adverse effect on the images, but the user's perception of it might be that it's not as well made as it could be.

For what it's worth, in my line of work (precision mechanical design), I would never use a simple friction fit for something that's going to be assembled and disassembled as part of its normal usage. Always a sliding fit that's secured with some mechanical thing, like screws, a flexible clip or sprung latch/catch/pin. The only time I'd use a friction fit is for something that's captive and not normally removed - like a dowel in one part that has a sliding fit in the mating part (only used for precision alignment of the two parts, not for holding them together - the friction fit at one end of the dowel is to stop it falling out, and slightly improves accuracy). With that, the dowel needs a pin punch and a hammer to take it out of the part it's captive in, which would be rare and might need something like Loctite to get a dowel back in that that will stay captive (a bit of inny-outy of a press-fit captive dowel can turn the fit into a sliding one).

I would be doubly cautious of trying to make a sliding fit from acrylic. As engineering plastics go, it's not the greatest. It's not particularly hard wearing and can be prone to impact damage at the edges. I virtually never use it, and even then, it's for things like safety screens in machine housings rather than a mechanical component. That said, the laminate thicknesses you're using don't seem too bad and I think the joint itself would be robust enough. However, I would definitely be looking at some sort of mechanical method to hold the parts together. I'd maybe consider something along the lines of captive threaded studs pointing down from the bottom edge of the rear standard, which go into open-ended slots in the base, the two clamped together using female thumbscrews (which can be left on the studs). I'd also maybe buffer the interface between the thumbscrews and the acrylic base with something like nylon washers to reduce risk of crush damage to the acrylic. (Lots of ways to do this, but that strikes me as easy to make - two tapped holes in the upright and glue in two studs - and still quick to assemble in the field.)
 
Last edited:
I'd be careful with the idea of the interlocking fingers having a good friction fit, especially if it's to be assembled and disassembled. Regarding fit, each part needs to have a 'perfect' size that will allow the whole to work - anything outside of that will either result in slop or be a very tight fit such that assembly will be difficult in the first place. All components have tolerances, and the sizing needs to be spot on to get friction. Even a handful of microns out can change it from friction to sliding, or to some small amount of damage or surface wear if tight. Aside from that, no matter how snug it is to begin with, there will be wear during assembly and disassembly that will eventually turn the fit into a sliding one in any case, at which point, there is a risk of slop (once you have slop, wear reduces markedly).

Whether or not slop is an issue is down to the size of the gaps between the surfaces, and in what direction the slop might be a concern. With a sliding fit, the back could move away from the base, causing the film plane to be in an unpredictable position. Similar thing with unintended swing. There is also a risk of tilt, which would feel like the rear standard wobbling at the top when in use. The movements might be slight and not have an adverse effect on the images, but the user's perception of it might be that it's not as well made as it could be.

For what it's worth, in my line of work (precision mechanical design), I would never use a simple friction fit for something that's going to be assembled and disassembled as part of its normal usage. Always a sliding fit that's secured with some mechanical thing, like screws, a flexible clip or sprung latch/catch/pin. The only time I'd use a friction fit is for something that's captive and not normally removed - like a dowel in one part that has a sliding fit in the mating part (only used for precision alignment of the two parts, not for holding them together - the friction fit at one end of the dowel is to stop it falling out, and slightly improves accuracy). With that, the dowel needs a pin punch and a hammer to take it out of the part it's captive in, which would be rare and might need something like Loctite to get a dowel back in that that will stay captive (a bit of inny-outy of a press-fit captive dowel can turn the fit into a sliding one).

I would be doubly cautious of trying to make a sliding fit from acrylic. As engineering plastics go, it's not the greatest. It's not particularly hard wearing and can be prone to impact damage at the edges. I virtually never use it, and even then, it's for things like safety screens in machine housings rather than a mechanical component. That said, the laminate thicknesses you're using don't seem too bad and I think the joint itself would be robust enough. However, I would definitely be looking at some sort of mechanical method to hold the parts together. I'd maybe consider something along the lines of captive threaded studs pointing down from the bottom edge of the rear standard, which go into open-ended slots in the base, the two clamped together using female thumbscrews (which can be left on the studs). I'd also maybe buffer the interface between the thumbscrews and the acrylic base with something like nylon washers to reduce risk of crush damage to the acrylic. (Lots of ways to do this, but that strikes me as easy to make - two tapped holes in the upright and glue in two studs - and still quick to assemble in the field.)

Thanks for the advice. I agree that even with a perfect snug fit (all parts are machined to 0.1mm tolerances) there's always the risk of movement/flex. I want to keep a good balance between quality and cost but not to the point that it results in a poor camera. Looking at my pieces, I think a pair of simple 90 degree angle brackets attached to the forward face of the rear standard will allow for a pair of bolts to be fitted through the base into either captive nuts or thumbscrews below the base.
 
Last edited:
Thanks for the advice. I agree that even with a perfect snug fit (all parts are machined to 0.1mm tolerances) there's always the risk of movement/flex.

In my world, that's big. :) A 4mm dowel in a 4.005mm hole will give a sliding fit.

To be honest, I'd be doubtful of getting a good friction fit with tolerances of +/- 0.1mm. I don't know if that's related to the width of the cutting beam or the XY position tolerance of the machine, but I suspect it would result in some faces having friction and some having gaps - and no means of predicting where the friction and gaps will end up.


I want to keep a good balance between quality and cost but not the point that it results in a poor camera. Looking at my pieces, I think a simple 90 degree angle bracket attached to the forward face of the rear standard will allow for a pair of bolts to be fitted through the base into either captive nuts or thumbscrews below the base.

That could work, although it means making angle brackets and the fixing holes for them. Also adds an angular tolerance that may or may not be better than the angle you'd get with the cut edges of the acrylic Captive studs into slots that are part of the general cutting of the parts means less stuff to make and less hole drilling and tapping, but will be subject to the angular tolerance created by the cut edges (whether that's due to tilt in the beam, or as a result of laminating layers that have their square edges protruding by different amounts due to positional tolerance of the beam).

0.1 is quite big and is amplified if that tolerance is applied to a relatively thin part that extends perpendicularly. For example, if the upright was 10mm thick and 100mm high, a 0.1mm difference in height from one side to the other of the 10mm edge results in a tilt of 0.57°, which would cause the top of the upright to deviate from vertical by 1mm. If the tolerance is +/- 0.1mm and if you have a worst case where one side of the edge is +0.1 and the other is -0.1, then the angle doubles for the given thickness, as does the deviation of the top of the upright. If the plate is thicker than 10mm, the angle reduces. If the upright is taller, the deviation increases. It's a case of deciding whether the calculated deviation for a given assembly and tolerance is likely to have an adverse effect. (Certainly, I think clamping the two together would be a good thing in terms of user experience - slop of +/-2mm at the top of the upright would be noticeable. I'm tempted to think that it wouldn't be especially bad, optically if it was stable, given how much tilting and waggling goes on in LF anyway.)
 
In my world, that's big. :) A 4mm dowel in a 4.005mm hole will give a sliding fit.

To be honest, I'd be doubtful of getting a good friction fit with tolerances of +/- 0.1mm. I don't know if that's related to the width of the cutting beam or the XY position tolerance of the machine, but I suspect it would result in some faces having friction and some having gaps - and no means of predicting where the friction and gaps will end up.



That could work, although it means making angle brackets and the fixing holes for them. Also adds an angular tolerance that may or may not be better than the angle you'd get with the cut edges of the acrylic Captive studs into slots that are part of the general cutting of the parts means less stuff to make and less hole drilling and tapping, but will be subject to the angular tolerance created by the cut edges (whether that's due to tilt in the beam, or as a result of laminating layers that have their square edges protruding by different amounts due to positional tolerance of the beam).

0.1 is quite big and is amplified if that tolerance is applied to a relatively thin part that extends perpendicularly. For example, if the upright was 10mm thick and 100mm high, a 0.1mm difference in height from one side to the other of the 10mm edge results in a tilt of 0.57°, which would cause the top of the upright to deviate from vertical by 1mm. If the tolerance is +/- 0.1mm and if you have a worst case where one side of the edge is +0.1 and the other is -0.1, then the angle doubles for the given thickness, as does the deviation of the top of the upright. If the plate is thicker than 10mm, the angle reduces. If the upright is taller, the deviation increases. It's a case of deciding whether the calculated deviation for a given assembly and tolerance is likely to have an adverse effect. (Certainly, I think clamping the two together would be a good thing in terms of user experience - slop of +/-2mm at the top of the upright would be noticeable. I'm tempted to think that it wouldn't be especially bad, optically if it was stable, given how much tilting and waggling goes on in LF anyway.)

I think I'm going to have to assemble all of the parts first then go from there. I've just checked with my friend and his laser cutter works on a 0.01mm tolerance rather than 0.1mm so better although I do agree that there is always risk of play, especially on my current version that's cut from softer MDF.

I've been looking at ways to draw up a bracket without complicating the design or the manufacture of parts and think I may have come up with something. I can't fit the bracket on the rear face up the upright because there's the rotating back attached to that which means I either need to fit the bracket to the front face, which has minimal width available due to the bellows and would also look untidy or fit it somewhere else.

If I extend the three stems on rear standard further down past the base, I can fit another piece of material below the base and then bolt though that. Probably easier to show a picture than explain so I'll draw it up quickly.
 
0.01 is much better. That should easily reduce unwanted tilt of the upright to negligible amounts. Joining the thick edge of the upright directly to the rear of the horizontal surface of the base should get very good precision with regard to angle.
 
Here's an updated design with the 90 degree bracket built in to the rear standard.

RearBracket.jpg

There would be a pair of captive nuts embedded into the lower half of the base

RearBracketBelow.jpg

Then knurled bolts would go through the bracket into the nuts to secure the rear standard (along with the friction fit of the parts themselves between the rear and base.

RearBracketBelowBolts.jpg

The only weakness I can see is that the rear standard is bonded to the bracket plate at the corner. If someone put enough force on the standard it could break that bond. However, as the pieces are simple to produce, as well as being separate from the main body, it would be possible to just swap out the rear standard.

I've just cut a piece of 3mm sheet to test the shape. I'd make the proper plate out of 6mm for strength and prevent flex.

IMG_7794.JPG
 
I will need to adjust the shape of the plate slightly once I've got the camera assembled and I can see where the centre of gravity is. I'll need to fit a tripod thread somewhere on the base so that will determine where my bracket fits up to and its' shape. I also want to keep the bracket as short as possible whilst still being effective so that when the camera is broken apart for transport, it's as small as possible. I don't want a massive bracket plate sticking up when the rear standard is sat down in a bag!
 
Agreed about the weak point. I would say that the plate can extend less than is shown there - the primary resistance to leverage is between the right-angle join and the two screws. The material beyond those isn't doing much. Also, the holes could be open-ended slots (open towards the front). That way, the screws can stay in place when disassembled. Just slide the two parts together and nip them up when assembling. Less likely to get dropped that way, as well as easier to use in cold weather.

Is there a reason for the two 3mm layers at the back of the rear standard? Could it be done with two 6mm layers in the middle and a 3mm on either side?
 
Last edited:
Agreed about the weak point. I would say that the plate can extend less than is shown there - the primary resistance to leverage is between the right-angle join and the two screws. The material beyond those isn't doing much. Also, the holes could be open-ended slots (open towards the front). That way, the screws can stay in place when disassembled. Just slide the two parts together and nip them up when assembling. Less likely to get dropped that way, as well as easier to use in cold weather.

Is there a reason for the two 3mm layers at the back of the rear standard? Could it be done with two 6mm layers in the middle and a 3mm on either side?

Sorry, I missed your post earlier while I was uploading the updated bracket plate design. Good idea about the open slots, anything to keep the camera as simple as possible is a bonus.

The reason for the mix of 3/6mm layers is so that the parts can all be machined as straight cuts without any routing being required. The two central 3mm layers retain the rotating stepped disk, the 6mm layer in front of those provided another anchor to the base and secures the rotating disk. The rear section is made up primarily of a 6mm pieces with a 3mm sheet embedded and a 2mm piece on top of that to provide the stepped mounting face for the film holder. Behind that 6mm sheet is another 6mm acting as a surround to help secure the film holder.

Sounds more complex than it is! My original design used only 6mm sheets but needed a lot of routing to create the rotating back mechanism and holder. By doing this I can cut the parts quickly from fewer sheets of material.
 
Right, I've updated my models with the new bracket and removed a couple of parts that were originally designed with rebates. Removing the rebates makes the model much more simple and quick to produce. Hopefully I'll have the the rest of the parts cut tomorrow so can start assembling it all properly over the weekend.
 
Yes, much easier with a picture. :)

Are the layers to be glued together on the two parts? If there are no fixings or other means of mechanical alignment, how are the rotating disk and DDS holder aligned?

Yes, the DDS holder layer (6mm sheet) will be bonded and pinned to the rotating disk itself. All other layers will be bonded together to form a solid rear standard. I'm embedding 4 x 6mm diameter magnets into the rotating DDS holder layer and the rear face of the rotating disk layer so the back is free to rotate then retains its position for landscape/portrait orientation.
 
Yes, the DDS holder layer (6mm sheet) will be bonded and pinned to the rotating disk itself. All other layers will be bonded together to form a solid rear standard. I'm embedding 4 x 6mm diameter magnets into the rotating DDS holder layer and the rear face of the rotating disk layer so the back is free to rotate then retains its position for landscape/portrait orientation.

Pins are good. I like the idea with the magnets (neodymium, I presume). I've found Cermag to be very good...

http://www.cermagmagnets.co.uk/

Lots of choice and excellent prices.
 
Pins are good. I like the idea with the magnets (neodymium, I presume). I've found Cermag to be very good...

http://www.cermagmagnets.co.uk/

Lots of choice and excellent prices.

Thanks for the link. I bought a bag of 25 6mmx6mm neodymium magnets from Spider Magnets via ebay last night for £4.99. I've bought from them before for previous projects and they normally deliver quickly. I wanted 6mm length so they can be fully embedded into both layers to give a stronger fit. I believe Intrepid use magnets to secure their rotating backs as well so I'm sure they will be strong enough.
 
Yes, that volume of material should be fine. I recently used some 10dia x 2mm ones for something, and they needed a decent amount of force to sheer them apart, and they're just a tad less volume than 6x6.
 
If you build it they will come...... :D
 
It's starting to get quite exciting isn't it.
 
Bit OT but relevant to you, Steve, I think. I recently came across a blog/magazine called Don't Take Pictures. Not sure what to make of it, yet, a bit arty and much more serious than PetaPixel/Phoblographer etc... but their latest article is "Some Assembly Required", mentions a pair of handmade Aussie panoramic 6*14 cameras (now there's an idea, Steve... only one really needed, perhaps). Anyway, at the end of the article is this message:

"Have you made or modified your own photographic equipment? Let us know at info@donttakepictures.com"
 
Bit OT but relevant to you, Steve, I think. I recently came across a blog/magazine called Don't Take Pictures. Not sure what to make of it, yet, a bit arty and much more serious than PetaPixel/Phoblographer etc... but their latest article is "Some Assembly Required", mentions a pair of handmade Aussie panoramic 6*14 cameras (now there's an idea, Steve... only one really needed, perhaps). Anyway, at the end of the article is this message:

"Have you made or modified your own photographic equipment? Let us know at info@donttakepictures.com"

Thanks Chris. I'll have a proper look at that site later.
 
I've just assembled the camera to see how heavy it is. With everything, including the focusing rail, but without the bellows, it's 1015g so I'm happy with that. It's slightly heavier than the Intrepid which is listed as 900g but I've still got options to lighten it further by removing some of the material from the bed and back. I'll wait until it's all assembled with the bolts etc and weigh it again before I think about reducing the weight. The rear 6mm part could be cut down considerably as it's basically creating a frame around the DDS to give it some extra grip.

Whilst it's slightly heavier than the Intrepid, I also think I could assemble one for lower cost meaning it could be cheaper to buy :0)
 
Last edited:
Back
Top