Today we’re going to look at the process of 3D printing thermoform molds for packaging. A few months ago, I set off on a journey to create an incredible demonstration part of the Origin One technology with a threaded sample part.
Pretty quickly, though, I found myself designing and printing soft jaws for the assembly of the parts (tapping and press fit inserts), and dressing up a common ¼-20 hex bolt using our Stratasys F370 to give it an easy to grip thumb screw cap by inserting the bolts into a print and embedding them into the part.
With so many great components to this project, I knew that it would be great if I could compile them all into a kit that we could show off at events and bring into customer meetings, but this kit was missing one vital component: professional packaging.
Sure, I could always print packaging trays directly to keep all of the parts in this kit safe, but that would cost an arm and a leg to accomplish for my target of about 75 total kits. I thought through my options and disqualified the ones that weren’t a good fit. I never like the look of pick and pack foam trays, and our quantity doesn’t justify a customized manufactured foam tray either. Luckily, I have one ace up my sleeve that leverages our 3D Printers particularly well in prototype design and end use production: Vacuum Thermoforming!
For those unfamiliar with vacuum thermoforming, it’s the process of taking a thin sheet of a thermoplastic (often PETG, ABS, or Polycarbonate, but can include other higher end materials as well), heating it up for a period of time until it is pliable, stretching that over a form (usually either a part or a mold), and then applying vacuum to pull that plastic into shape. Once the plastic is pulled, it cools to a rigid solid form within seconds, and since many thermoform sheet materials cost between $1-2 each, this makes it a VERY viable option for the quantities of kits that I’m looking to make. For our project, after testing different materials and thicknesses, we settled on PETG sheets (low cost, rigid, food safe) at .5mm thick, which end up costing $1.74 each in pre-cut sheet sizes that fit our vacuum thermoformer.
It’s possible to make thermoform molds out of any 3D printing technology. Considering that I have access to the top five out there, I had plenty of options to make my molds. PolyJet, Stereolithography, and even Digital Light Projection technologies would all make incredibly smooth parts, and Selective Absorption Fusion would be my lowest cost option but hands down the best technology for the job is Fused Deposition Modeling! Why is that? Because using FDM technology I can make my part hollow with the click of a button, and even make the entire underside of my mold an open face, which leads to the MOST vacuum pull I can possibly achieve out of those five technologies, and by a very large margin.
This is actually critical in Thermoforming, and is one of the key things that has to be taken into account when thermoform mold designers create their molds. Usually, you’re looking to achieve a very smooth surface finish, but you must add a ton of pin holes through your mold in order to achieve a decent vacuum. Using FDM technology, we don’t have to work hard at all to get the best vacuum pull possible, since our parts are hollow, and have some minor porosity on the surface layers that allow air to pass through those “solid” surfaces surprisingly well too.
There are a few other reasons I had for picking FDM outside of the incredible vacuum that we can get using this technology:
- It’s Fast – I’ve never designed a vacuum thermoform mold before this project, but I’ve printed many in the past for our customers, so I need to ensure that I can “Fail Fast” and learn quickly from my design mistakes so that this project doesn’t drag on for months on end! In a perfect world, I get it right on the first try and we’re all done, but I’m a realist and I know that the best way to power through this project is to design, print, test, and repeat. Using a relatively coarse resolution, we can create new molds to run overnight, or in some scenarios even same day following a morning test and redesign, making this Rapid Tooling through FDM an incredible asset to the project!
- It’s Inexpensive – If I was using machined prototype molds, hand carved wood molds (this is legitimately one of the traditional methods for this packaging technology), or even resin based 3D printing technologies, each one of these prototype molds would not only take much longer between iterations, but also cost a ton of money to produce! While most of my iterations varied in size and total material usage, they all tended to be in the $75-150 range using lower cost thermoplastics. Even our final production molds, which were made out of Polycarbonate, hit a total price of $204 for both the tray and cap molds, which at 75 kits (keeping in mind two PETG sheets per kit) brings our total cost of packaging thermoform trays to $6.19 per box! The more kits that I make, the lower the cost per unit gets as well, since I’m dividing the cost of the molds into more packages, so if I was making 300 kits, the cost per box (again, both the cap and the tray) would only be $4.16 each!
- Tons of Materials – I mentioned above that our final molds are made in Polycarbonate, which we’ve used because it has a high heat deflection temperature (won’t melt on us after a few pulls), it’s very rigid (won’t warp on us at any point in time), and it is capable of being printed at higher resolutions (smoother final molds with no sanding or other post-finishing – this mold is straight off of the printer and onto the thermoformer!). With that said, Polycarbonate is a bit more expensive than some of our other materials, so when running our prototype iterations, it made far more sense to run them in materials like ABS and ASA, whatever we had loaded from other projects, as these materials can also survive a handful of test cycles without issue.
What Printer Did We Use?
The Stratasys Fortus 450mc is the perfect printer for this project! First off, it runs Polycarbonate great, it has a fantastic build capacity of 16” x 14” x 16” (far exceeds our mold constraints of 8” x 8” x 1.5-3”), and it is capable of tolerances that are typical of most machining applications of +- 0.005” on the x/y axis, +- 1 layer on the Z axis. On the Fortus 450mc, we’re capable of printing Polycarbonate at the layer heights of a fast and coarse 0.013” all the way down to a slower but fine and smooth 0.005”.
For our final molds though, we dialed in on the resolution of 0.007”, which prints relatively fast, is quite smooth and high detail, but is also the finest resolution that we can use the breakaway support material with, which is perfect for thermoforming since we don’t need any supports outside of the base of the parts!
We also chose the Fortus 450mc for the software; the 450mc can use the easy to learn and use GrabCAD Print, but for this project we actually want to do some very advance part modification in removing the lower cap layers from the part for maximum vacuum, so Insight is the best software tool for the job!
Becoming a Mold Designer
I mentioned it earlier, but my only experience prior to this project working with thermoform molds was in printing them for other people, so I knew that I had my work cut out for me. What I knew for sure going into this was pretty much limited to:
- Positive draft angles are your friend,
- Oversize your cavities to ensure that your parts fit into the sheet after the pull
- Don’t make your molds too tall or large, as this will add unnecessary stress on the sheet during the pull.
Let’s go on a step-by-step journey of each of my mold iterations, as I learned quite a bit from each one of these prototype runs, and I think it can dramatically help in accelerating future mold design with a Polycarbonate FDM part:
REV I – This was the single most aggressive mold attempt that I went for! This one was setup to be a clamshell design that would fit into an 8” x 6” x 4” box, with a living hinge in the middle. On a positive note, I love how the hinge worked, but unfortunately the existence of the hinge led to the back half of the mold being very floppy due to the weight of the FDM and Origin Parts in the tray.
Also, because of the size of this one part design, the sheet was being pulled over a LONG distance, which made it both difficult to get pull depth in my tray cavity, and also hard to get any detail. My major takeaway from this iteration was that I need to simplify things down a bit, and make it a two part mold; one pull for the tray, a second pull for the cap.
My secondary takeaway was that the sidewalls of this iteration don’t appear to have any real structural strength, so the addition of sidewall ribbing seemed important.
REV II – Given that I was shifting to a two sheet package, my focus went to producing the tray itself, and given what I had learned from the cavity on the REV I, I wanted REV II to put as little stress as possible on the sheet, so I went from a concave mold design to a convex one with tall sidewalls.
As it turns out, this actually led to even worse pulls, as it became incredibly stressful for the sheet to pull over a 4 inch wall and down 3 inches into a cavity, even with the smaller footprint. My takeaway from this iteration was that the concave cavities just seem to work a bit better, so let’s try that again as a single part.
REV III – This was interesting; I would consider REV III my first real success! Solid pull and good structural strength, the only issues that I saw immediately were that my oversized cavities were still a bit too small. Next iteration: dial in the cavity sizes, but overall keep this design. There was only one hiccup here…
REV IV A and B – I was told that we were no longer using an 8” x 6” x 4” box, we would instead be using an 8” x 8” x 3” box, which is drastically different! Using everything that I learned up to this point, I started from scratch to ensure that I would have a clean model that would enable me to easily make the Cap mold off of it.
Within this revision, there were a few minor modifications from the beginning to the end, which mostly comprised a shift from a 2.5” tall tray down to a 1.5” tall tray, but in general I kept the ribbing on the sidewall and the top, moved the thumb screw cavities since we’re working with a larger space, added an “L” and an “R” to identify which soft jaw sat where, and made the thumb track in the middle larger to make it easier to take parts out.
REV IV B would end up being the final iteration for the tray, and from there now it was time to make the cap!
The Cap mold was much easier to design, since I was able to build it off of the backbone of the final tray model, as it would effectively have the same X/Y dimensions as the tray but span the entire 3” Z axis of the part. The tray sits down in the bottom of the box, the cap goes right on top of it, close the box up and everything is immobilized and safe!
The Cap only had one revision, which was split into an A and B, as outside of minor tweaks on size, the only major changes were the addition of a bridge between each full height column, and slots in the lower two columns to allow the tray to sit flush on top of the cap inside of the box when on display. The latter was a change that I was particularly excited with because it doesn’t affect the structural integrity of the sheet OR the difficulty to pull it; it’s strictly an upgrade.
The Thermoform Mold Forming Process
We’ve got an entry level Vacuum Thermoformer in our print lab, and after quite a bit of testing, what I found worked the best were some very low cost PETG Sheets that were 297mm x 420mm and 0.5mm thick. In order to heat up these sheets enough, we warm up the Thermoformer to between 150-180C, raise the sheet to just below the heaters, wait approximately 75 seconds, and at that point the sheet is pliable and will be sagging down enough to pull.
For good release, I do like to coat the surface of the thermoform mold with a food safe mold release agent, but some mold designs with more dramatic draft angles may release easier than those that I designed (all draft angles on my molds were at the very least 8 degrees). With that, all you do is turn off the heat, lower the sheet, and hit the vacuum button for roughly 10 seconds, and then you can pop the sheet off of the mold and you’ve got a successful vacuum thermoformed packaging tray!
What Did It Take to Make These Molds?
I mentioned a bit earlier what the total cost for these final Polycarbonate molds were, but breaking it down per part:
Part Tray Mold, printed at .007” Layer Height, Sparse Double Density with the base layers removed using Insight Software for MAXIMUM vacuum pull through the mold – 16.124 cubic inches of Polycarbonate, 0.531 cubic inches of Breakaway Support, total cost of $87.32 of combined material, and a print time of 14 hours and 13 minutes.
Cap Mold, printed at .007” Layer Height, Sparse Double Density with the base layers removed using Insight Software for MAXIMUM vacuum pull through the mold – 21.704 cubic inches of Polycarbonate, 0.531 cubic inches of Breakaway Support, total cost of $116.61 of combined material, and a print time of 18 hours and 32 minutes.
Wrapping Things Up
Like a well-oiled machine, I was able to mold the Trays and Caps for all 75 demonstration part kits, and after a bit of time cutting away the excess sheet material, the kits were all fully assembled and ready to send out! Designing these Polycarbonate trays was a real treat for me, and going through the iterations and final mold using the Stratasys Fortus 450mc was a breeze.
The Polycarbonate material works perfectly for the final molds, and running the earlier iterations in slightly lower cost thermoplastics was great as well. You may be thinking to yourself that this was an incredibly expensive venture, but an entry-level low-end thermoforming unit, like the one that we’re using here, is only about one thousand dollars. Now that we finally have one in our shop, you can count on more projects utilizing it in the future!
Download the free FDM Thermoforming Design Guide below or reach out to one of our experts to learn more!