So you want to make a 3D print from a cyanotype solar print. Maybe you saw the documentation of my "Sculpting Shadows" project. Maybe you got here through some creative Googling. Regardless, this page is for you. There are probably other ways to do this, but this is how I did it. I'd love if you let me know what you learn through your own experimentation!
The basic concept here is that you'll be using a solar print as a depth map to create a 3D model and then 3D printing it. For those unfamiliar with depth maps, essentially software interprets the luminance data of a pixel (how bright it is) as depth information. Depth maps can be used for a variety of applications, but in this case the lightest parts of the image become the more raised parts of the 3D model, and the darker parts become the more recessed parts.
​​​​​​​There are about a million tutorials online about how to make cyanotype solar prints, so I'm not gonna reinvent the wheel here and describe that whole process. 
The areas that are more shaded by our objects stay white, and the areas that the sun hits become a darker blue. Note that the solar print that results from three-dimensional objects like the rambutans above have some faint midtones that follow their curves, because though they cast hard shadows, some light leaks in from the sides. The closer an object gets to the solar paper, the more light it blocks. This effect will make a big difference in how these prints translate to 3D models.
There are a few things to keep in mind when making solar prints destined for the third dimension:
1) Objects that are more three-dimensional will lead to more three-dimensional prints. Duh? Super flat things that create a stark contract between the exposed and unexposed areas of your print will lead to a 3D print that is essentially two levels: the flat background and the raised foreground. This may be what you want for your project, but if you want 3D prints that have more curves and bumps and ridges and ravines, pick objects that don't lie flat on your print and therefore allow more light to leak in around their sides to create midtones in the solar print.
2) If you're hunting those sweet midtones that give your 3D print the curves you want, you might also want to expose your solar prints for longer than you might otherwise. This is a little bit me theorizing more than hard fact or experience, but I suspect going a little long on the exposures allows more time for weaker light to leak into the sharp shadows of your objects and create subtler gradients. 
3) Look out for small particles. When I did this I was using organic matter (plants, seedpods, etc.), and it was easy for a little piece of matter to flake off onto my solar print during exposure. This would create a bright dot which might not look like much in the 2D print but would be a distracting spike on the 3D print.
4) Pick the right solar paper. I used two different papers, PAPER 1 and PAPER 2. PAPER 2 was larger and had a more pronounced texture. This texture carried through to the 3D print. I actually like to way this particular 3D print ended up looking, but in general I try to avoid textures so that the structure that comes through in the 3D print is more authentically representative of the shadows cast by my objects. Also, as a side note, I found that it was harder to get a distinct, high contrast print with PAPER 2, and this might have been because the plexiglass included with the paper was blocking UV rays or something? Dunno, more experimentation required.
5) For best results, let the solar print fully develop. I found that my prints continued to deepen in color for at least 24 hours after exposure. You want the maximum dynamic range (difference between the image's darkest and lightest bits) possible to create a 3D model with the greatest fidelity, so that extra deepening is useful. That being said, I made most of the 3D models on this page from prints that had not fully developed yet because I didn't realize they hadn't fully developed yet, and they look pretty good.  
I go into more detail about solar printing techniques I have tried and show how they translated to 3D on the "Sculpting Shadows" page
A flatbed scanner is probably ideal for capturing an image of your solar print, but I was in the field for this project and therefore used a camera for all the examples on this page. The important factors to consider at this stage are distortion and visual noise. If you are using a flatbed scanner you don't have to worry about either of these, and can move along to the next step.
If you want your eventual 3D print to honor the geometry of your solar print, you want to minimize distortion. (You might be into the distortion, and that's a cool idea--you can skip this section, you rebel.) Distortion can come into play in this step if you use a camera with a wide angle lens and photograph your solar print at close range. A phone camera has a pretty wide angle lens, so if you are using your phone camera I would recommend not taking your photo at close range. Pay attention to the edges of the paper as displayed on your screen--notice when they bow out and are no longer parallel to the edges of the phone screen. Back up until they are parallel. You can crop out the space around the print later. If using a non phone camera, use the longest lens you have or zoom in all the way.
Noise is the real killer here. Subtle variations in brightness from pixel to pixel can create a very "fuzzy" 3D model. You might be able to find a way to use this effect for an interesting model, but I haven't. So you want to minimize the grain in your photo. How do you do this? Photograph your solar print in bright, even light. If you know about ISO settings in your camera, set that ISO as low as it will go. Use a flash if you like, so long as it produces an even light across the image (look out for light falloff around the edges of the image, particularly if you are photographing at close range). 
Some disclaimers: I used Photoshop. Photoshop costs money for people who don't pirate it. If you develop a methodology based in some free software, please let me know. I'll update this How To and give you credit. ​​​​​​​I want as many people as possible to have access to this process.
Also, this guide is written assuming very little Photoshop knowledge. If you still get hung up on anything, please contact me so I can change the guide--you probably aren't the only one!
Open up your image in Photoshop and crop it down so that only solar print remains. If you used tape or tacks or something to keep your solar print in place during exposure, be sure to crop out the resulting white spots. Or not, if you're into that, whatever I don't judge. 

You're eventually going to want to make some adjustments to the image before converting to 3D, but I encourage you to go ahead and do a trial 3D conversion to get a sense of what kinds of adjustments need to be made. To do this, go to Window > 3D in the menu bar. In the 3D tab, select "Mesh from Depth Map," and "Solid Extrusion" from the pulldown menu (experiment with the other options sometime to see what else you can come up with!). Click "Create," and if you are asked to switch to the 3D workspace, go ahead and do that. Hey, your solar print is now 3D! It's probably a bit deeper than you want. Let's adjust the dimensions of the model to approximate what you want your 3D print to eventually look like.
First, I like to change the texture of the model from the default (the image you converted it from) to something more neutral so I can better see the actual shape of the model. To do this, in the 3D tab select "Background" (or whatever the layer you converted to 3D was called--it will be nested under "Depth Map"). Then in the Properties tab open the pulldown menu with the image of a sphere and choose a more neutral texture. I am partial to the simple blue.​​​​​​​
Now turn the model to its side so you can see how deep it is. Do this by selecting "Depth Map" in the 3D window or clicking the model itself, then clicking and dragging to adjust the model's angle. Be patient with Photoshop, it does this 3D stuff very slowly. I find the default depth is always much deeper than I want--the deeper it is, the more surface variations are exaggerated. To change the depth, first select "Depth Map" in the 3D tab. Then in the Properties tab, click the little 3D box icon, which will bring up the Coordinates window. The third column of numbers shows the dimensions of the model, with Z being the depth. Don't worry about X and Y--you'll be setting your 3D print size later, what you're messing with now is just the dimensions relative to each other. Click the cube icon at the top of this column and uncheck "Uniform Scaling" if it is selected. This will allow you to change the depth without affecting the other dimensions. Try out some different numbers in the Z field until you arrive at something you're happy with, then hit Enter. ​​​​​​​
Zoom in and turn it around and explore your model. It's probably fuzzy! It's probably too deep! It's probably too pointy in some spots you don't want it to be! Let's undo the conversion (Ctrl+Z a few times or click back a few steps in the History window) and make some adjustments to address some of these issues. Yeah, I know, you've done so much hard work, but I recommend going back and forth like this throughout the process so you can see the effects of your adjustments.
Let's address the fuzziness first. This is caused by camera noise and other fine textures that result in brightness variations between pixels in your image. In 3D, these variations cause the depth of the model to jump up and down from pixel-to-pixel. The main tool I use to get a smoother surface is the Median filter, which reduces the differences between adjacent pixels by assigning each pixel a value equal to the median value of the pixels around it. (Don't
worry too much about how it works if that's not interesting to you. I used the Median filter for years without knowing exactly what it was doing.) To get to this filter, in the menu bar select Filter > Noise > Median. The "Radius" setting tells the filter how wide a net to cast to calculate a median value for each pixel. In simpler terms: higher number equals smoother. The setting you'll want to go with depends on the size and detail of your image, but you're looking for a sweet spot where you lose harsh pixel-to-pixel variation but maintain the details of your shapes. Experiment! Try one setting, convert to 3D, notice the effects, undo and try a different setting. I found that I could often go higher than I thought and still maintain good shapes. (For the curious, the images I used for this project are about 4500x3000 pixels and I usually used a Radius value of 10.) 
For some of my models I also used noise reduction settings in Adobe Lightroom or Photoshop, but I wouldn't rely on them completely--I found the Median filter solves more problems than just noise.
If you test out your 3D conversion again at this point, you might notice that the model is smoother overall but still has some unwanted spikes or dips. Look at your 2D image for any light or dark specks that stand out. The light ones are pretty common, and could be the result of flecks of matter falling onto your solar print during exposure. Regardless, if you see some distracting spots you can remove them really easily with the Spot Healing tool in Photoshop.
Remember that the brighter a pixel is, the deeper the model will be in that spot. Think about how thick or thin you want the thinnest part of your model to be. The darker you adjust the dark areas in your image to be, the thinner they will be in 3D. Similarly, the brighter you make the whites in your image, the thicker those parts of the model will be. Anything that is absolute black will actually drop out of the 3D model once we leave Photoshop, so if you want to make some 3D prints that aren't square, or have holes in them, that's how to do it. If you make your whites very bright, you will see a plateau effect--the thickest parts of your model will flatten out and will lack the bumpy texture of the rest. 
To adjust your levels, in the menu bar select Image > Adjustments > Levels. Under "Input Levels," you'll see a bumpy histogram of your image, with three little gray triangles along the bottom: a dark one, a medium one, and a light one. If you drag the dark one to the right, it will make your darks darker (thinner); the light one to the left will make your lights lighter (thicker); and the medium one back and forth will adjust the brightness/thickness of your medium tones. ​​​​​​​
If you want to make your darks lighter (thicker) or your whites darker (thinner), adjust the triangles along the "Output Levels" line. A common problem I ran into when making these models is that the part of an object that was touching the solar paper during exposure would create a very bright white area, and then falloff quickly into darker blue. This would create spikes or thin raised ridges that I thought were incongruous with the whole. For these models, before conversion to 3D I would adjust the white level to make it darker while trying to keep the rest of the image at about the same levels.
There are a lot of ways to adjust the levels in your image, this is just one. If you want finer control than the Levels tool gives you, play around with the Curves tool (Image > Adjustments > Curves) and see if you can tame that beast.​​​​​
Once you have made all the adjustments you need to make and it looks good when you convert to 3D, you're ready to export the model out of Photoshop. While viewing your model in 3D, select 3D > Export 3D Layer in the menu bar. In the window that pops up, select "STL" in the "3D File Format" dropdown menu, hit OK, and pick a name and a destination for the file. As far as I have been able to tell, the dimensions here don't make any difference for our process.
I'm not going to go into too much detail about how to prepare an .STL file for 3D printing. If you have a 3D printer, you are already familiar with the process. If you're using someone else's, they can show you the ropes or you can refer to one of the many guides available on the internet. Try to import your .STL file directly into Slicer or whatever 3D printing software you're using. Slice it up and keep a close eye out for anything hinky. Sometimes these models just work, in which case print away and rejoice! Sometimes, though, the models that result from this process have some problems that 3D printing software doesn't quite know how to fix.
To solve these issues, I use the free software Meshmixer. To get your .STL file into Meshmixer, click the big import button. Marvel at how much smoother this program handles 3D than Photoshop! Once you're done marveling, click the "Analysis" icon in the left toolbar, then "Inspector" at the top of the menu that pops up. You'll see some colored spheres floating around your model. Each one of these marks a problem with the geometry of your model. Try clicking "Auto Repair All" and if that doesn't do anything weird to your model and you load it up in Slicer or whatever and it doesn't look hinky anymore than rejoice! 
If "Auto Repair All" does something weird, try choosing different options from the "Hole Fill Mode." If you click the spheres one at a time it will fix them using the method selected. If your image had a lot of pure black areas because you want them to drop out of the model, you will have A LOT of these spheres floating around. If it doesn't load well into Slicer and "Auto Repair All" messes it up, good luck. If you come up with a foolproof way to print models like this, please let me know.

Back to Top