Dry Filament is Good Filament !
OK, that is simplistic, but for our unfilled Nylon 6 filaments root cause analysis over the last two years shows that drying accounts for over 90% of technical enquiries. Most of the rest are relatively simple temperature or printer compatibility enquiries.
So lets get right into the weeds on this……
We’re heating up plastics in ovens so good ventilation and a fire control kit (that you know how to use) are assumed throughout this tip sheet. Everyone should know how to address small fires, that goes double for anyone who uses ‘hot’ tools in their work or play. If you don’t, then your local fire department will have instructional courses.
Protect youself and your loved ones by knowing what to do, even if it is just getting out and staying out.
Now that’s out of the way we can crack on.
Nylon 6 is hygroscopic
Nylon 6 is hygroscopic in its ‘pure’ unfilled, uncompounded form.
That means it will suck moisture out of the air as well as from surfaces like sweaty hands. If you touch our filament when it is fresh and ready to print you can feel that happening on your finger tips – a slight stickiness as it wicks your finger dry. This sensation is especially obvious if you plunge your hands into a bag of our raw material in micro-pellet form. They come out feeling as if they have been rubbed in talcum powder.
If a filament maker is claiming a non-hygroscopic Nylon 6 or a zero warp Nylon 6, then it has stuff other than Nylon 6 in it alongside the ‘pure’ polymer. That’s not necessarily a problem, but it’d be nice to know whats in there to enable informed choice because every planned addition has an impact. If it didn’t, it wouldn’t be in there.
Fishy Filaments currently only supplies un-filled Nylon 6, but that might not always be the case. If we do make additions to the filament in the future it will be for either performance or aesthetic reasons, not just as economic padding. Right now we’re working on a carbon-fibre filled version (for performance) and a crushed oyster shell material (for aesthetics) that is unlikely to be distributed as anything but a company-specific ‘special’.
So – Fishy Filaments unfilled 3DP products are hygroscopic. In other words they attract and absorb moisture from the air or through contact and will retain that moisture until it is actively removed.
How to tell if filament is wet or dry
There is that slight stickiness as described above, and of course if you have access to a piece of lab kit called a moisture balance you can tell to the 0.01% by weight of water present (this is the one we use).
But a more ‘3DP normal’ way would be just to try a short section to see how the hotend deals with it and how the molten plastic looks, sounds and feels once cool.
Our nylon should be translucent to transparent once cool depending on its thickness, and almost completely transparent when molten at 250C+. As it cools you should see the transparent nature tend towards translucent as the melt passes down through the glass transition point.
In the following vid of our lab’s test extruder you can see filament being produced and the change in appearance it goes through as it cools.
Once solid it should be smooth, almost slick, as Nylon 6 is a low friction material.
If you see bubbles, the molten bead kinking as it falls under gravity, hear a hissing, popping or snapping then the filament is certainly wet and unusable in its current state.
If you don’t see or feel those really obvious signs but the print, once it hits the bed never seems to stick, warps and peels quickly or has a rough feel to it then you’ve probably dried it, but just not enough. Those bubbles can be really small and just look like a cloudiness, but they are still there. Any bubble is a void filled with steam that will contract on cooling because nylon will reabsorb the steam/moisure in the bubbles massively increasing the shinkage factor of the material So it doesn’t really matter if the bubbles are small or large what impacts the resistance to catastrophic warping is their presence or absence in the melt.
How to dry our filament
You can remove moisture from our filaments by using
Heating to 80C for an extended period at ambient pressures will dry our material. More on the ‘ambient pressure’ coment later.
But beware !
Too hot and dyes in the nylon will thermally decompose, first darkening then going on to a full black. Experiments show that 120C seems to be a critical point at which the dye will certainly oxidise, but that it takes time for this oxidation to permeate through the filament.
The amount of time to oxidise is a function of how high the temperature is and oxygen availability, so a melt pool in an active hotend melt chamber doesn’t have time to oxidise while printing, but if you leave the filament in the hotend overnight to cool and reheat the next day the exposed filament will suffer degradation. You might even get a plug forming as the material that has been heated will re-absorb moisture as soon as exposed to it at temperatures under ~50C.
If you don’t clean your hotend between jobs or use a silicone boot to protect it, small amounts of nylon will accumulate and go black for sure. Its not difficult to clean off when hot but, as always, take care.
Note; as the filament darkens at temperatures below 190C it is only the dye that is affected. The polymer is fine. Above 190C and the polymer itself will be impacted too.
Too cool (under 50C) and the moisture that has penetrated into the molecular structure of the nylon will not be driven out, and no matter how long you dry it you will never completely dry the filament. You might get it into a usable window but you won’t optimise performance.
A dessicant gel dryer on its own is not enough to dry our filament completely
There appears to be a minimum threshold of about 50C at which this molecular dessication starts to really kick off for our Nylon 6. This 50C figure is an inflextion point between temperatures at which our filament absorbs water and those at which it dries.
Between 50C and 80C the ‘time to dry’ falls, but not in a linear manner.
Above 80C the time to dry does not appreciably drop any further, so hotter isn’t better here.
Its a really difficult study to do outside a well equipped lab but our filament dries at 80C far quicker than at 70C which is far, far quicker than 60C is far, far, far quicker than 50C, at which point time to dry tends to infinity 😉
So how long should I dry for then ? What is an extended period ?
Its really tough to say definitively because local humidity and the type of dryer being used will impact drying times.
What we recommend is that you invest in a means of heating the filament to 80C at a Relative Humidity of 20% H2O for 5 hours or more, and that this level of technology will allow you to define your own local drying schema appropriate to your needs. Typically that means overnight at 80C because rH 20% is quite low for a shop without a specialist drying setup.
If you can’t achieve those parameters reliably then your drying methodology will struggle to deliver a consistent filament to the print head.
What does that mean in the Real World ?
Domestic ovens that you use for food are NOT recommended.
You are looking for a steady, diffuse, but relatively low heat that most domestic ovens don’t provide anyway, and you don’t want your grilled cheese smelling of plastic do you ? So separate food and filament, and keep them separate.
We recommend a small oven dedicated to filament prep, that has a cut-off timer and a temperature control.
We’ve seen solid solutions in stone-based pizza ovens and even modified toaster ovens, but be careful to shield your filament from direct radiant heat or it will deform or discolour. Ceramic tiles have been shown to be great baffles in ovens that have exposed elements, but please no open flames ! And be careful to put a tray under the filament just in case everything goes wrong.
Our nylon has a V0 rating for flamability under the UL94 rating system, which means it is ‘self-extinguishing’ but even so, better safe than sorry.
With a good system in place the basic hardware to dry and maintain your nylon filaments shouldn’t cost more than $100.
Of course you could go all-in as our print agency partners, J-Supplied, have done here with an Apium F300 filament dryer connected to a Raise 3D Pro2, a Craftbot FLOW IDEX with a hood and a Createbot F430 with a fully enclosed and heated chamber.
This is a top-of-the-range heat-only solution – Ask for Mike if you want to know more
Of course you can go further, faster, bigger, stronger if you have the space, skills and cash.
Vacuum dryers are like a TARDIS for nylon filaments
In the plastics industry vacuum dryers are well established, but they are only just coming to town in the 3D printing world. There are good reasons for that; the compressors needed to provide a reduced air pressure can be noisy and energy hungry, and there are risks associated with the home use of both high and low pressure vessels. And for many desktop printer users its just another thing that looks like a waste of money.
Unless your time has a well defined value.
We expect innovation in this space as time pressure to prepare more advanced filaments grows, either in terms of extended preparation times and their impacts on project delivery or directly through higher employment costs.
We believe that the current range of ambient pressure filament preparation and delivery cabinets (such as the Ultimaker Pro Bundle) are soon to be supplimented with a range of more technical/industrial solutions specifically targeting 3DP (such as the Apium F300 and the Mass Portal FD5) and in the future these will include vacuum drying.
Home cook vacuum dessication systems are already out there;
Check out Charlie Patterson’s home build vacuum dryer and its off-the-shelf equivalent from Vision Miner. They are pretty much identical in terms of hardware and if you have the confidence to home build Charlie tells us his setup cost under £250 in hardware.
Vision Miner’s ready to build kit isn’t much more at US$350, but of course if you are outside the US couriers costs will need to be added.
Vacuum drying works by reducing the partial pressure of water around the filament, effectively sucking the moisture out of the nylon, where heat alone is more akin to evaporation.
Put the two together and you are looking good !
The harder the vacuum, the quicker the drying, but the longer it takes to get to that lower air pressure and the bigger the kit needed. So there is a balence in how much it is worth investing in this tech.
If an organisation needs to vacuum dry more than 3 or 4 spools at a time then they should really be speaking to an equipment manufacturer to define their operational and safety needs in detail. Both Charlie and Vision Miner can fit around 5kg in their pressure vessels but only one or two in their ovens, so a little sizing optimisation is needed there if you have an industrial workflow to achieve.
Vision Miner’s website says they can get a spool of PEEK print-ready in 4 hours including heating before vacuum drying.
We haven’t tested this but see no reason to challenge it. Charlie says his works really well on our filament !
One last thing – Relative Humidity is, like, ‘relative’
rH or relative humidity is a measure of the amount of water present in the air. Its an incredibly variable environmental parameter at a small scale. It can vary with time, temperature and pressure, so that an enclosed space that provided rH 25% yesterday might only provide rH 30% today because there is a new weather system in town. And for our filament that could shorten the printable life of well prepared spool by hours.
So the upshot is that external environmental conditions can cause print fails in hygroscopic materials because they impact rH, whether that be in the printer enclosure, in the filament storage caddy in the bowden tube or just in the room. A blast of cold, wet air across the print bed can set in train a cascade failure.
The aim of ‘in-print’ drying to to avoid or eliminate exposure to variations in rH.
It is relatively straightforward; bowden tubes between all drying equipment, air tight fits, printer hoods or enclosures, maintaining the filament at 50C, not opening the door on the printer ‘just to check’ and the like are all achievable and relatively cheaply. In-print solutions include kit like the PrintDry, hacks of fruit dessicators or an air tight caddy with a liberal amount of silica gel dessicant.
Choose your price point on the basis of your expected print duration and end-quality criticality.
So lets put all that together
If you only have an ambient temperature and pressure dryer i.e. one with a dessicant gel only, you will probably fail with our filament more than you succeed.
to 50C- filament will absorb moisture until it reaches an equilibrium with the relative humidity
50-190C – nylon in the filament will dry safely, but above 120C the dye will start to oxidise and darken with time (hours)
190C+ – nylon will start to degrade if kept at these temperatures for longer times (hours)
Ideally you need to dry our nylons at a steady heat of 80C and relative humidity of 20% H2O or lower.
At ambient pressures this will, in most cases, take at least 12 hours (overnight). It can take longer, especially if you are using a part-consumed spool that has been exposed to moisture during a previous job(s).
If you have a vacuum dryer, placing a hot spool into the chamber and removing the air will accelerate the drying process dramatically, but it has to be hot (80C all the way through) otherwise the effect is much diminished.
Once you have the filament prepared correctly don’t blow all that hard work by just sticking the spool on a hook in the open air. You need to maintain its dryness in order to optimise your print quality. This is where the lower temperature, lower tech drying solutions come in and work really well.
And thats all folks !
Happy printing !