A variety of red and black glass objects are shown on a white background. In the foreground, there are two black spiral-patterned earrings. To the left is a red and black shape with three points on the top. On the right, a deformed glass sheet is shown bent over concentric red and black glass rings. In the center top is a red glass vase with a roughly-textured exterior.

Paste Extrusion For 3D Printing Glass And Eggshells

In contrast to the success of their molten-plastic cousins, paste extrusion 3D printers have never really attained much popularity. This is shame because, as the [Hand and Machine] research group at the University of New Mexico demonstrate, you can use them to print with some really interesting materials, including glass and eggshell. Links to the respective research papers are here: glass and eggshells, with presentations in the supplemental materials.

To print with glass, the researchers created a clay-like paste out of glass frit, methyl cellulose and xanthan gum as shear-thinning binders, and water. They used a vacuum chamber to remove bubbles, then extruded the paste from a clay 3D printer. After letting the resulting parts dry, they fired them in a kiln at approximately 750 ℃ to burn away the binder and sinter the frit. This introduced some shrinkage, but it was controllable enough to at least make decorative parts, and it might be predictable enough to make functional parts after some post-processing.

Path generation for the printer was an interesting problem; the printer couldn’t start and stop extrusion quickly, so [Hand and Machine] developed a custom slicer to generate tool paths that minimize material leakage. To avoid glass walls collapsing during firing, they also wrote another slicer to maintain constant wall thicknesses.

The process for printing with eggshell was similar: the researchers ground eggshells into a powder, mixed this with water, methyl cellulose and xanthan gum, and printed with the resulting paste. After drying, the parts didn’t need any additional processing. The major advantage of these parts is their biodegradability, as the researchers demonstrated by printing a biodegradable pot for plants. To be honest, we don’t think that this will be as useful an innovation for hackers as the glass could be, but it does demonstrate the abilities of paste extrusion.

The same team has previously used a paste printer to 3D print in metal. If you don’t have a paste printer, it’s also possible to print glass using a laser cutter, or you could always make your own paste extruder.

3D Print Glass, Using Accessible Techniques

When seeing a story from MIT’s Lincoln Labs that promises 3D printing glass, our first reaction was that it might use some rare or novel chemicals, and certainly a super-high-tech printer. Perhaps it was some form of high-temperature laser sintering, unlikely to be within the reach of mere mortals. How wrong we were, because these boffins have developed a way to 3D print a glass-like material using easy-to-source materials and commonly available equipment.

The print medium is sodium silicate solution, commonly known as waterglass, mixed with silica and other inorganic nanoparticles. It’s referred to as an ink, and it appears to be printed using a technique very similar to the FDM printers we all know. The real magic comes in the curing process, though, because instead of being fired in a special furnace, these models are heated to 200 Celsius in an oil bath. They can then be solvent cleaned and are ready for use. The result may not be the fine crystal glass you may be expecting, but we can certainly see plenty of uses for it should it be turned into a commercial product. Certainly more convenient than sintering with a laser cutter.

Illustrated Kristina with an IBM Model M keyboard floating between her hands.

Keebin’ With Kristina: The One With The Grasshopper Typewriter

Do you consider your keyboard to be a fragile thing? Meet the glass keyboard by [BranchNo9329], which even has a glass PCB. At least, I think the whole thing is glass.

The back side of an all-glass keyboard. Yeah.
Image via [BranchNo9329] via reddit
There are so frustratingly few details that this might as well have been a centerfold, but I thought you all should see it just the same. What we do have are several pictures and a couple of really short videos, so dive in.

I can tell you that [BranchNo2939] chose a glass substrate mainly due to curiosity about its durability compared with FR4. And that the copper circuitry was applied with physical vapor deposition (PVD) technology.

Apparently one of [BranchNo2939]’s friends is researching the bonding of copper on to glass panels, so they thought they’d give a keyboard a go. Right now the thing is incomplete — apparently there’s going to be RGB. Because of course there’s going to be RGB. Continue reading “Keebin’ With Kristina: The One With The Grasshopper Typewriter”

The “Unbreakable” Beer Glasses Of East Germany

We like drinking out of glass. In many ways, it’s an ideal material for the job. It’s hard-wearing, and inert in most respects. It doesn’t interact with the beverages you put in it, and it’s easy to clean. The only problem is that it’s rather easy to break. Despite its major weakness, glass still reigns supreme over plastic and metal alternatives.

But what if you could make glassware that didn’t break? Surely, that would be a supreme product that would quickly take over the entire market. As it turns out, an East German glassworks developed just that. Only, the product didn’t survive, and we lumber on with easily-shattered glasses to this day. This is the story of Superfest.

Continue reading “The “Unbreakable” Beer Glasses Of East Germany”

How To Make Conductive Tin Oxide Coatings On Glass

Glass! It’s, uh, not very conductive. And sometimes we like that! But other times, we want glass to be conductive. In that case, you might want to give the glass a very fine coating of tin oxide. [Vik Olliver] has been working on just that, in hopes he can make a conductive spot on a glass printing bed in order to use it with a conductive probe.

[Vik’s] first attempt involved using tin chloride, produced by dissolving some tin in a beaker of hydrochloric acid. A droplet of this fluid was then dropped on a glass slide that was heated with a blowtorch. The result was a big ugly white splotch. Not at all tidy, but it did create a conductive layer on the glass. Just a thick, messy one. Further attempts refined the methodology, and [Vik] was eventually able to coat a 1″ square with a reasonably clear coating that measured an edge-to-edge resistance around 8 megaohms.

If you’re aware of better, easier, ways to put a conductive coating on glass, share them below! We’ve seen similar DIY attempts at this before, too. If you’ve been cooking up your own interesting home chemistry experiments (safely!?) do let us know!

The Strangest Way To Stick PLA To Glass? With A Laser And A Bit Of Foil

Ever needed a strong yet adhesive-free way to really stick PLA to glass? Neither have we, but nevertheless there’s a way to use aluminum foil and an IR fiber laser to get a solid bond with a little laser welding between the dissimilar materials.

A piece of sacrificial aluminum foil bonds the PLA to glass with a form of laser welding, with precise control and very little heat to dissipate.

It turns out that aluminum can be joined to glass by using a pulsed laser process, and PLA can be joined to aluminum with a continuous wave laser process. Researchers put them together, and managed to reliably do both at once with a single industrial laser.

By putting a sacrificial sheet of thin aluminum foil between 3D printed PLA and glass, then sending the laser through the glass into the aluminum, researchers were able to bond it all together in an adhesive-free manner with precise control, and very little heat to dissipate. No surface treatment of any kind required. The bond is at least as strong as any adhesive-based solution, so there’s no compromising on strength.

When it comes to fabrication, having to apply and manage adhesives is one of the least-preferable options for sticking two things together, so there’s value in the idea of something like this.

Still, it’s certainly a niche application and we’ll likely stick to good old superglue, but we honestly didn’t know laser welding could bond aluminum to glass or to PLA, let along both at once like this.

UC Berkeley Prints Glass Nanoparticles

In a recent video, [Joel] of 3D Printing Nerd interviews a researcher at University of California, Berkeley about their work with glass 3D printing technology. A resin is impregnated with tiny glass nanoparticles and produces green parts. An oven burns away the resin and then another heating step produces the actual silica glass part. You can see a video about the process below.

As you might expect with glass, the temperatures are toasty. The first burn is at 1100 C and the fusing burn is at 1300 C. The nanoparticles are about 40 nanometers across. The resulting parts are tiny with very small feature sizes. The technology to do this has been around for a few years, and the University continues researching this form of computed axial lithograph (CAL) 3D printing. These parts are so small that it uses an adaptation called microCAL that produces much smaller parts at high precision. However, the equipment available today won’t produce very large objects. The video talks about the uses for some of these small glass items.

We wonder how much the firings in the ovens change the tiny tolerances. They obviously work, so either they account for that or it doesn’t shrink much.

If you want your own 3D printed glass, a laser system might be more practical. If you just want transparent plastic, your FDM printer can do that. Really.

Continue reading “UC Berkeley Prints Glass Nanoparticles”