Not long ago a donor dropped several spools of ABS filament off at Fox.Build for use in the 3D printers. We weren’t sure of the history of the filament, and since we didn’t immediately get it sealed against the summer humidity we were concerned it might have picked up moisture somewhere along the way. Knowing moisture in the filament can wreak havoc during the printing process (the water can boil within the plastic and cause gas bubbles and all sorts of nastiness) it seemed like the best option was to try to dry it just to be safe. Internet research revealed that a standard desiccant (aka a “do not eat”) would be sufficient to keep it dry once it was dry, but would not be enough to get it dry in the first place.
There are two methods commonly used to dry filament; 1) Bake it in an oven, and 2) use a vacuum chamber to extract the moisture. Oven baking didn’t seem that exciting, and no one really wants their next pizza to taste like ABS, so building a vacuum chamber it was! The first step was to get a vacuum pump. Supposedly you can get one from an old dehumidifier, but my two dehumidifiers broke down after I had already bought a standalone vacuum pump on eBay. Here’s the pump:
The next step was to find some sort of a suitable container that could be sealed with the filament inside. I had a bunch of 5 gallon buckets sitting around, and they could hold a lot of filament, so that seemed a likely choice. That didn’t work out quite like planned!
The ideal vacuum chamber might have been an old pressure cooker, but I didn’t have one. I looked around the house for a couple days before spotting an unused oil tank from a race car. It was big enough, sturdy (made from aluminum), and had lots of fittings I could use to connect the vacuum. Here’s what it looked like when I took it off the shelf:
It was pretty nasty inside:
But once it was cleaned up it fit a spool of ABS perfectly!
The next step was to create a lid. I had an old piece of Lexan, and I liked the idea of being able to look into the chamber, so I cut a Lexan circle to fit on top of the tank:
When I put the lid on the tank I could not pull a vacuum – the lid didn’t seal against the o-ring that was originally part of the oil tank design. Part of the issue was the lip of the tank was not perfectly flat. The original tank lid was 3/8″ thick aluminum and was bolted down all the way around, causing lots of squish on the o-ring and overcoming the lack of flatness. The Lexan couldn’t do that by itself, and I didn’t want to undo 20 bolts every time I put something in or took something out. So, I put black silicone RTV all around the o-ring and built it up a little, then set the lid on with wax paper and a heavy chunk of steel. This formed a gasket that conformed to the lid once the RTV had hardened.
The next step was to test it. After the RTV had hardened I removed the lid and wax paper, found an old power steering hose, and connected the vacuum pump to the tank. When I turned on the pump you could hear it seal and see the Lexan bow down into the tank! It sealed! The final step was to mount the tank on something to stabilize it (I did everything prior to this with the tank in a vise) and dry some filament! Here’s the final version:
You can see the lid bowed in by the vacuum – so it is definitely working! I dried several spools of ABS with this (several more to go) and have been testing them in the SolidDoodle printer that was also donated to Fox.Build. So far there have not been any issues I can attribute to moisture in the filament! The next project (or should I say another one of many?) is to tune up the SolidDoodle so it prints as well as our Dremel.