A review of melt extrusion additive manufacturing processes

This appears to be the first scientific paper on FDM hot ends.
If you find a copy of it, please PM me

A review of melt extrusion additive manufacturing processes: I. Process design and modeling

Findings - Fused deposition modeling and similar processes are among the most widely used rapid prototyping processes with growing application in finished part manufacturing. Key elements of typical processes are described; including the material feed mechanism, liquefier and print nozzle, and the build surface and environment, along with approaches to part finishing. Approaches to estimating the motor torque and power required achieve a desired filament feed rate are presented. Models of required heat flux, shear on the melt, and pressure drop in the liquefier are reviewed. Upon leaving the print nozzle, die swelling and bead cooling are considered. Approaches to modeling the spreading of a deposited road of material as well as the bonding of polymer roads to one another are also reviewed.

Originality/value - To date, no other systematic review of process design and modeling research related to melt extrusion additive manufacturing has been published. Understanding and improving process models will be key to improving system process controls as well as enabling the development of advanced engineering material feedstocks for fused deposition processes.
[www.emeraldinsight.com]

Edited 1 time(s). Last edit at 05/01/2014 02:44PM by A2.

Hey, A2, thanks for all these articles you post. But just wondering, do you already have your own printer? You're very active here, but I read in another post that you don't have your own printer yet.

Wouldn't it be nice to move on from the theoretical side to the practical practice of working with a 3D-printer? Or was I mistaken and do you in fact have a printer in your possession?

Kind regards,
Marinus

I don't have a printer, I'm waiting for the worm gear GDR-Simpson to hit the streets, and I'm in no rush.
So during the interim, I'm designing a DLP-SLS printer.

Demigod of Knowledge, A2 doesn't have a printer?
Interesting, but actually not all that surprising. More than enough know to see the reasons in waiting for better stuff.
My first days all I saw was something that would quickly become mainstream, that technologies would develop to allow anyone at some point to just get a 3D printer and make whatever they want with minimal effort. I wanted to jump in as quickly as possible before just anyone could do it. But now that I see things, that's either not going to happen, or it's going to be a lot furthur down the road than the media likes to toute. Do I regret my decision? No, I've learned a lot, it's been a path that's held together my engineering and creativity niche.

Dang! I've been looking for something like this for the last couple of weeks. I'd love to have a more theoretical/quantitative model for whats going on in the FDM process.

Unfortunatly it looks like this issue isn't out yet. It's marked as an "EarlyCite" article which means what we're seeing is just a preview. (For more info on that website's "EarlyCite" thing see this link: [www.emeraldgrouppublishing.com] ).

It looks like Rapid Prototyping Journal releases an issue every two months though, and since this is issue three we can probably expect this article to be published by the end of June at the latest. Based on their latest published issue, we should be able to purchase the article for $32 once it's out. (Though if you're a University student/have access to a University Library you should be able to access it for free.)

Edited 1 time(s). Last edit at 05/01/2014 01:50PM by rm446.

It's out now! I got myself a copy and it does contain some complex/detailed formulas for things like pressure drop across the liquifier, temperature distributions, etc. A good chunk is how different processes affect mechanical strength. Over all I'd say its very useful if you're very interested in theoretical scientific approaches to what's going on and not afraid of dimensionless equations.

Also some interesting tidbits I did not realize, like apparently for two layers to bond, their interface must be above the glass transition temp of the filament. That explains why PLA prints so nicely, its low glass transition temp of 60°C! Basically the hot layer coming from the extruder heads up the layer below it so this is achieved. If your plastic conducts heat in such a way that this doesn't happen, the layers won't adhere (and thus things like Ultem have to be FDM printed in an oven).

thanks for pulling A2's thread back up. missed it originally. now I have plenty of bathroom reading to do

Tks rm446!

I've spent many years designing fusion welding machines. Two things to ensure a proper fusion weld are required, one, the resin needs to be above the glass transition temperature, two an oxide layer forms over the melted surface, so you need to push the plastic into the opposing surface enough to crack that oxide skin until a non oxide front presents it's self.

If you want to take it to the next level of robust fusion welding, you need to retract or minimally stretch the bond, this helps to align the spherulites (spaghetti like plastic chain of molecules). Polymer chains are disorganized, and by giving them a little tug you orientate the spherulites into laminar strands which increases the bond strength. This is commonly done with plastic living hinges, just after the part is ejected from an injection mold, and still warm, the hinge is bent, this orientates the spherulites, which give the living hinge a long useful life.

Living hinge
[en.wikipedia.org]

Interesting A2, I was not aware of the oxidize layer thing. I did notice this paper mentioned the layer to layer bond strength analysis assumed "perfect wetting" which I'm guessing means it ignores imperfect contact due to things like the oxidization layer you mention.

I wonder with spherulites though, are those present in significant quantities on FDM filament? I remember that spherulites are essentially how polymer crystals organize themselves, but I thought FDM materials were amorphous or at least had very low degree of crystallinity. This makes sense based on what I've read on polymers (crystallization adds complexity to the melt process), but then again I've never seen anything that outright says what degree of crystallinity of my PLA fillament is, so I could be wrong on this.

You are correct, the plastic needs to be either crystalline or semi crystalline for this to work.

There are amorphous and crystalline PLA plastics. The travel speed of nozzle vs. the extrusion rate can be used to stretch the spherulites.

Drying Parameter PLA:
Amorphous :113°F (45°C)
Crystalline : 176°F (80°C)
[www.natureworksllc.com]

Crystalline and semi-crystalline materials are particularly prone to thermal shrinkage; amorphous materials tend to shrink less. Shrinkage is in the direction of flow. Knowing this possibly you could compare the shrinkage of two different resins to try and determine if it's amorphous or crystalline.

I have a copy the paper.Additionally, please keep the paper for your personal use only.

Edited 1 time(s). Last edit at 10/20/2014 06:46AM by wtent.

@wtent,

Thank you!