So while doing some research I stumbled upon a wiki page on reprap from a few years back where the user was creating a glass nozzle to replace the brass and PTFE assembly.1

Does anyone know the theory behind this? Glass is a great insulator so I could see how that would be beneficial for the heat break part but I can't see how it is appropriate for the nozzle as this is normally brass which is a good conductor.

Surely the glass takes much more energy to heat up?

On a side note I've seen similar projects using ceramic instead.

| improve this question | | | | |
  • 2
    $\begingroup$ Welcome to 3dPrinting.SE! While I don't know for sure, I'd suggest the reason for it is glass has great wear properties. You're right in that it has great insulative properties, which means to me it wouldn't transfer heat into the filament very well ... which is probably why it has never really been used too much (if at all). Just my speculation, though, as I don't know for sure. $\endgroup$ – Pᴀᴜʟsᴛᴇʀ2 Nov 4 '19 at 15:21

First off, this is not a glass nozzle, it is a whole hotend design. A super simplistic one.

Glass is, like ceramics, not a good thermal conductor but has a quite good thermal resistance - it only melts at about 1600 °C, which means you will never have to fight melting or warping of the filament path itself at all - the heater copper wire will melt at about 1084 °C, so way before the glass, and most plastics that are printable start to decompose at less than 400°C.

Construction-wise, this design has some benefits:

  • Due to the design and material properties, this hotend doesn't need cooling fins and a "coldend" is not needed at all.
  • The whole hotend being one solid piece makes it pretty much a "plug and play" item and prevents leaks.
  • Glass is extremely abrasive resistant. This means a glass nozzle could be used for stuff like carbon fiber filament very long.
  • Glass can be molten, repaired and modified with fairly simple equipment, e.g. a burner and some skill.
  • Glass could be easily cleaned up to medical and food-grade machine ratings. The simplicity of the hotend assembly could make it autoclavable as a whole piece.

It has some downsides though:

  • Glass is brittle and does not take lateral forces and sharp impacts kindly. In other words: Handle with extreme care.
  • Due to the glass being an insulator, the inside of the hotend will have a lower temperature than the outside.
    • A fairly thin-walled meltzone could mitigate this problem to some degree at the downside of making it even more prone to breaking
    • The insulating behavior means, that the meltzone has to directly feed into the nozzle with as little unheated area as possible to prevent the molten plastic from solidifying inside the nozzle again.
  • The skill needed to create a properly sized nozzle from glass is tremendous.
| improve this answer | | | | |
  • $\begingroup$ Many glasses have their glass transition between 500 and 800 °C, they get soft far below the melting point (quartz glass being the exception). $\endgroup$ – cbeleites unhappy with SX Nov 5 '19 at 21:50
  • $\begingroup$ @cbeleitessupportsMonica most copper wires also start to destroy themselves well below the melting point. Plus, softening of the glass would be of no issiue: no plastic is shaped at 500 to 800 °C but the highest ones I know are worked at 300-400 °C. $\endgroup$ – Trish Nov 5 '19 at 22:04

Ceramic I can understand - very strong, great thermal range capability. Glass not so much - you'd need some seriously careful annealing at least.
In either case the material is much harder than brass, or even steel, so you could presumably use tougher tools to unclog, etc. as needed. If you're using materials loaded with wood or metallic particles, the glass/ceramic tip will be less likely to degrade than brass.

BTW, glass being a thermal insulator means it may take longer to heat up, but the energy required is probably less . The specific heat of glass is on the order of .84 J/gm-K . Compare with brass at 0.38 , but keep in mind the rate at which brass will shed heat into the air vs. glass. In either case the energy is tiny compared with the thermal mass of the hotend assembly.

| improve this answer | | | | |
  • $\begingroup$ Is being an insulator possibly beneficial to keeping the molten material at temperature as it exits? I would think metal nozzles are somewhat below the hotend temperature at their tips. $\endgroup$ – R.. GitHub STOP HELPING ICE Nov 4 '19 at 15:56
  • $\begingroup$ @R.. they are a little, but the thermal flow in the nozzle is much higher. think about it like this: A ceramic plate can be heated on one spot and be brought to glow there, and a couple centimeters over, it is almost room temperature, as the flow through it is bad. If the same plate was metal, it would be much hotter there. I expect that a glass or ceramic nozzle would be quite colder than the heating zone. $\endgroup$ – Trish Nov 4 '19 at 20:24
  • 1
    $\begingroup$ @Trish: With enough time and energy, an insulator can be properly heated across its body. But, more importantly (and extending R's point), the higher the heat flux of the material, the less fickle it is in terms of shedding heat to variation in circumstances (e.g. drafts). I'd wager a guess that a more consistent nozzle temperature leads to a more consistent flow out of the nozzle, leading to a better overall printed product. $\endgroup$ – Flater Nov 5 '19 at 20:41
  • $\begingroup$ In contrast to the page linked in the question, I'd probably want to go for quartz glass - that would avoid a whole lot of trouble wrt. tensions due to temperature changes. It also has a much higher glass transition temperature than "normal" glasses. Of course, correspondingly higher temperatures are needed to draw the capillary. $\endgroup$ – cbeleites unhappy with SX Nov 5 '19 at 21:56

The glass nozzle might be beneficial with abrasive filaments with metal particles that would wear down the brass. As glass can be reasonably easily shaped into a nozzle with standard DIY-tools it might be an interesting project.

A commercial nozzle for filaments with extreme abrasive properties is the Olsson Ruby nozzle. These are designed to hold an actual ruby in it, though ruby is probably not feasible for a DIY-nozzle.

| improve this answer | | | | |
  • $\begingroup$ I believe the ruby nozzles were running about $100US. Not something I'd purchase for intermittent use, that's for sure. $\endgroup$ – Pᴀᴜʟsᴛᴇʀ2 Nov 5 '19 at 2:25
  • $\begingroup$ I can order an Olsson Ruby for a little shy of 100€ per piece for the base version and about 170€ for the high temp version. $\endgroup$ – Trish Nov 5 '19 at 11:19

The glass nozzle was an attempt to be able to see the plastic filament melting and to see what is happening during retraction. I also though to use a glass nozzle to heat the plastic to just under melting point then use a LED laser (10 Watts thereabout) to heat just the tip of the plastic through the nozzle to bring it to the final temp where it would melt. Turns out it is unneeded, and lasers waste a lot of energy as heat.

| improve this answer | | | | |

One shortcoming would be that when it comes back to lay down a new line next to an existing line, I would think that it would need to be able to melt the previously printed plastic, especially any bumps and strings.
High thermal conductivity for good heat flow seems important.

| improve this answer | | | | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.