I'm really looking forward to seeing if you can make this work. I print materials that need to be extremely dry and I've just been practicing vigilant drybox discipline, but a more forgiving solution would be excellent.
Essentially, you're preheating the filament prior to feeding into the print head. Might consider options to increase the filament exposure to the heated air ... but take care not to overheat the filament.
Please do. I hate filament boxes. Limits spool sizes I can use, and for some filamenta even causes clogs or breaks because some filaments don't like to be arched through the PTFE tube and whatnot. Really hate it. If you can achieve the same result without dry box, it would be awesome. Maybe play with silica inside the tin cans
I would 100% love to see a version 2 of this inline filament dryer system, especially one that can do 4+ filaments at a time. I've been needing to print TPU, and something like this will be super cool!
"Then I did what I needed to do in the first place, research" - me, every time I start a new project haha. Great video, insane drying set up, and really interesting results!
The best way to remove surface moisture from the filament is to create a vacuum, use a vacuum pump in a container and remove the atmosphere, the moisture will sublimate off the filament. You don't have to unwind and rewind the filament. you can cycle the process once or twice. I am buying a vacuum pump for chemistry experiments to change the boiling points of whatever fluid I am working with.
Built such a system myself, but found that you need to raise the temperature for it to be effective. Putting an old 3D-Printer Hotbed into the chamber is not difficult, but passing the power inside was challenging
@@timw7406 Vacuum is a good heat insulator. Maybe, It's better to repurpose old Dryer Machine to preheat and predry. And use vacuum as a finishing touch.
@@vasiliynkudryavtsevAgreed. VisionMiner has videos on this. Dry in repurposed bench top oven. Then a couple of minutes in vacuum chamber after. If you had the budget, a lab vacuum oven would be pretty slick.
@@vasiliynkudryavtsev Because the vacuum is such a good insulator you can use a small amount of radiant heat to achieve lots of heating. An entire spool might optimistically radiate 2-5 watts of its heat away when heated to 100c. A radiant heat source thus only needs about 2-5 watts of output to heat hitting the spool in order to get it to that temp. A common choice is incandescent lights shining thru the walls for cheap at home methods.
Retired now, I spent 2 years developing dehydration test system for our motor Stators, moisture in stators in AC not good. our measuring system used DC heat on the stator wires, and a vacuum and dry nitrogen purge. the real good part was the dry Nitrogen purge, i would apply about 15psi dry nitrogen to the stator holding cavity, pull vacuum, the nitrogen would absorb moisture and became the medium for me to capture it in a Liquid nitrogen bath. we would weigh test tube before and after to get the weight of the water pulled from stators. this was a 4 hr test cycle, not piratical for 3d printing. Just letting you know how a measuring system works... perhaps it might trigger a thought for you. Me i keep all my filament in dehydration containers all time. just because i know how hard it is to dehydrate. your idea super good - should help lots of us.
I would like to see a version 2 of this. One idea I had was to make your "in line" drier into the shape of a toroid so you could loop the same section of filament another time or more. Kind of like how a beer line chiller system works. just a quick idea. not fully thought out. Thanks for the great vid. subscribed!
I had the same thought! Maybe the torus is filled with silica gel and the air circulates sucking the moisture from the filament and it gets pulled into the silica gel. When the system is idle it gets hotter inside to dry out the silica gel.
This is related to a classic problem given in thermodynamics classes. The question is how hot does the oven need to be to cook the potato in ten minutes. The answer is that it can’t be done because you’d set the potato on fire at the needed temperature. With filament it’s probably the glass transition temperature that puts a limit on how fast you can dry it out.
I guess you could make the can longer so it has more time to dry out? Like at some point the plastic will be too wet for it to be possible but it could probably still be good for certain moisture levels.
@@MrNoipethat'd make a pretty good experiment tbh. I guess it would only work well with materials with a high glass transition point though (if at all)
Five meters of 5mm ID tubing (like what you'd use for a bowdon tube, only larger) with a tee fitting at one end blowing hot dry air into it. The filament goes into the other end of the tube, against the flow of hot dry air, and comes through a seal on the other side of the tee and into the printer. The only issue is you'd need somewhat higher static pressure, maybe 1bar or so. Moisture migrating through a material pretty much acts like heat and can be fairly well modeled with thermodynamic equations. To get maximum moisture out of the material you want the highest possible difference in moisture content, highest possible temperature without damaging the filament, and the longest possible contact time. It's a heat exchanger except you're exchanging water instead of heat. Turn it into a very long and narrow counterflow unit and you'll probably get excellent results.
This is a bit of a departure from your form factor suggestion, but I was wondering whether, in order to increase the moisture content differential, silica drying beads could somehow be added in line, or before injecting the hot air into the drying chamber, or even both. That should provide an overall dryer environment, and the optimal volume of beads could be perhaps calculated to optimize the whole thing. Am I making some sense here ? Thanks :)
@@crepuscule47 Drying the hot air on the way in would be ideal for this, especially for those of us who live in swamps :) Higher pressure air, through a drying tube (silica, probably), then heat it, then through the counter flow tube seems like the order of operations to me.
The filament needs a lot more time in the drying zone. I was going to do this but set the project aside a year or two ago. Unless you want a super long tube, you need a system of pulleys so the filament can run back and forth across the hot dry air zone several times. You probably also need active drive for the pulleys to avoid overloading the extruder.
@@mirag3304 You can do that but you still need either a system of pulleys or coupled input and output filament hobs so the same length of filament remains in the zone rather than pull from the extruder trying to tighten it down to a point (open core) or to tension itself against the core (toroid) just like it'd do on a spool.
No idea how well this translates but I’ll share anyway. I used to work with giant coils of steel wire. When prepping it to go into the machine it was wrapped onto a a second drum a few times to help with feeding and getting it all lined up. Something like that inside the cans would add a lot of surface area without needing to double back via pulls or something else as other have suggested. My thought is another tube inside you can corkscrew the filament on giving a much more even and long drying time.
That sounds like a great solution. Instead of a long skinny soup can, use something large diameter and short length like a cookie tin. Put a short spool inside on bearings so it turns easily, and loop the filament around several times. I think the filament would try to migrate sideways along the spool as it turns, though. How did the industrial machines deal with that?
@@dekutree64If it's anything like industrial commercial software: They probably didn't deal with that. They could've easily left it janky while resting on their laurels due to patents and other ways to prevent competition. Then, they just charge an outlandishly large price for their shit.
Although other people may have invented this, this is an ingenious idea and I think that the cost of what you invented compared to the $2000 dollar price tag is ridiculous. I will be making something similar, or taking some inspiration to make something amazing just like this!
@@sanctusletum8522 Then wrap the outside and/or inside with heat insulating material, ceramic paint? Your usual house wall heat insulation material? The usual heat materials used for a forge? So many ideas.
Kudos! You hit the nail on the head. Just like you, I hate drying a whole roll of filament for umpteen hours only to forget what it was that I needed the filament for when the cycle is done. 🥴 Keep up the research and development and I’ll keep watching. Good luck! 👍
In Poland we make filament dryers from cheap vegetable/fruit/mushroom dryers, since first printers became popular, as drying the forest mushrooms is very popular here. They have airflow, 250W heater driven by thermostat and only need replacement of drying sieves with spool holder.
Yep these have been staple filament dryer for years. I think the commercial ones are inspired by them. Do the commercial ones not have airflow? That's a pretty fundamental oversight.
@@sligit Commercial ones resemble commercial heaters for heat-shaped shoe medical insoles. They are safe, look nice and you are sure that using them you will not be sued for damaging patient's shoe. The fact they don't work at all (insole remains thermally untreated) is the very least concern there.
What would be cool is a drybox with dedicated "in-line" dryer section. Basically use one heater and fan to pre-dry the complete spool, but have the end of the filament moved over some rollers inside the box before it leaves for the printer/extruder to improve drying of the filament from all sides. What i see a bit of a challenge with all the longer in-line dryer options is, that it should still be easy and fast to change the filament spools. So you would have to design it in a way that gives you good access to the full filament path. I'd love you to explore this topic more for sure!
I just got the stupidest idea. Reverse bowden (or halfway bowden or smth) where you just blow hot air through the PTFE tube the filament goes through. That could also work as the exhaust for the drybox, at least partially. But you could also do this from the printer going out.
A v2 that worked with the AMS would solve filament changes handily since it would be in between the AMS (and therefore the spools and also the automated filament feeder) and the printer.
@@Litl_Skitl The only problem with a long bowden tube is probably the drag you create. Even if you use a tube for 3mm filament with 1.75mm filament, you will need to curve that tube around and you create friction. Maybe a very stron extruder does not care much, but i'm pretty sure you would see a difference in print results just because of that. And with flex filaments, it would get even worse as you have long stretch path, thinning out the filament at the extruder gears making it hard to grab and prone to slipping.
The Sunlu S4 you showed has fans and vents. Airflow. It also has humidity sensors that will keep drying as the humidity rises. Also most dryers allow you to print directly from the machine, while it's drying.
You missed the part when he explained that even with fans the airflow hits only the outer layers of filament on a spool. Inside layers are still a wet noodle.
@@DmitrySholokhovno they aren't. It only takes some time until the heat reaches the core of the spool. But you don't print from the core. You people are searching for a problem which doesn't exist. Use a filament dryer while printing and it's all good. Let it warm up 10-20 minutes beforehand. That's usually the time to slice, upload and warm up the printer anyways. Most comments of people don't even seem to have a dryer and start to think this guy invented something everybody needs.
@@REDxFROG thank you for saying this. After watching the video, I was still scratching my head thinking this isn't needed. It worked when he took a drybox of filament and printed with it... of course it did. It probably would have printed the same without his tin can contraption. If you store filament this way too, you can print whenever you want without waiting for it to dry.
Thanks for the video. I almost patented an inline dryer for filaments 12 years ago, did the calculations about needed drying lengths, designed simple setups etc. As to the why we dont see this more often: In the 3D printing industry, somehow the knowledge about how to properly prepare material for extrusion seems to not have transferred from regular extrusion processes like injection molding. Every TDS for granules used in conventional extrusion has at least a drying recommendation, most even have a specified maximum residual moisture content for processing. Stratasys went around this in providing (well) sealed containers, which mostly were used quick enough so the users did not often encounter this issue. Then after the patents expired and reprap started using welding wires for printing, nobody really thought about this. Since more and more people are into 3D printing now with materials more prone to issues with moisture (PLA and ABS are not as problematic as TPU or PA), the topic gets more attention. I still see posts almost daily about users boasting they never dried any filament and that it's plain useless. As a material scientist myself, statements like this make my head hurt. That's like saying you don't ever need to clean a surface before using glue on it. Of course you always can be lucky, but it's just good practice to stick to proven and meaningful recommendations.
Lots of comments on this video as well about not needing to dry filament, i guess they leave in a dry area and or store filament properly. Maybe you want to share some advice for v2 from your experience?
This is brilliant. I would try using copper pipes which can be bent into a spiral shape to have a much longer length for faster printing while taking up less space
@@robogoofers9131 an 8 feet long insulation sleeve is about $2 at the hardware store. Depending on the design, you can have external heating instead of forced air through the long tubes and take advantage of the conductivity. Lots of options.
I was just thinking the same thing...Except instead of copper, you can use rings/guides throughout the can, leaving most of the filament exposed. Think of when running cat 5 in a drop ceiling, you just put hangers every so often to just help keep the cable suspended.
4:30 i think they might be using molecular sieves just based on appearance, but i’ve got no idea what else they’re using in the machine in order for them to justify that $2000 price tag
Science! I guess they just want to be payed for the efforts of R&D they put into making the device. And i could see the machines being assembled by hand and not really optimized well for fast and easy assembly. It's a niche market with low volume, especially for a "professional" solution. Going for a cheap low profit margin solution was not theyr goal, as the big brands already cover that market with the cheaper filament dry boxes.
I bet the drywise uses a mostly closed loop where the warm air is recycled through the molecular drying sieves instead of continuously heating cool air. Great video, excellent first POC, and love your humor. Looking forward to part 2!
I think this is REALLY interesting, and it seems you have proven it really hasn't been done in a feasible manner. I think there is a product here, just waiting to be invented and marketed. You got this, man! DO IT!
This looks very interesting. I wouldn't necessarily go with tin cans, but rather a tube with relative big diameter (>6mm) and blow hot (60c-70c) air into this tube, either from the printer side, or from the dry box side. Allowing the tube to be open at the other end so the hot air and the humidity are free to egress the tube after drying the filament. There are plenty of PTFE tubes that can do this which are easy to buy from China, or you can use a PVC pipe, it might even work with your local garden hose as well - depending on the material it is made of.
I was thinking the exact same thing, and given PTFE has a high temp resistance, you could wrap some sort of heating coil around the tube to keep the flow of air warm
PTFE is overkill, simple nylon or even most silicone tubes would be cheaper and have more than enough temperature resistance for anything short of PEEK
This is amazing. I would greatly appreciate seeing this turned into a finished project. Maintaining correct filament moisture levels is critical to good quality 3d printing. An inline filament dryer that has a reasonable cost and is effective would be a significant improvement.
You can't remove the moisture from filament, that has absorbed moisture over time of weeks and months. One easier way is to loadup the filament in a vacuum drier(used for degassing epoxy resins). You also need to keep some dessicant inside, and keep the container warm(85C if you put the reel on pedestals). You could dry any filament this way, in a matter of an hour or two at the most. You can also refresh those brittle pla, by putting it in 70C water for 15 mins, and doing the above.
@@smoothbraindetainer If you know how 3d filaments are made, extruded filament are drawn in water, before wound into a spool. There needs to be a min. moisture needed to keep filament from turning brittle.
@@smoothbraindetainer A brief dunk in hot water, will not affect tolerances, as long as the filament is dried later. Also, the filament is not stretched during the water dip. Have done this to my verbatim brand filaments with no issues.
If you have to use a drybox, then the advantages of the inline system completely disappear. You would need to run the inline dryer longer and use more power than a conventional dryer. In a conventional dryer you would just dry it and store it in a dry box.
yeah when he put it in the drybox, is it a dry filament or an already wet one? atleast if it's a wet one then it means you can remove the need for initial drying time
Interesting idea! Some ideas for improments: (1) Put heater elements and fans pushing air into opposite direction to the filament movement. That way the filament hits the hottest air the last before exiting the tube. The hotter the air, the less the humidity of the incoming air and you need to dryest possible air at the end of the process. When filament is really moist in the incoming feed, any amount of air movement will already improve things so it would be wasteful to use the dryest possible air in this end. (2) Feed the air from the fan through lots of dry silica gel before heating the air. That way you start with as dry air as possible before entering the heater element which makes the air even dryer. (3) Setup the system to pass the filament through the same tube multiple times. The amount of water that evaporates from the filament is probable not enough to fill the volume of the tube so when you have lots of airflow in the tube, the moist air from the filament will probably stay close to the filament and another looped strand closeby can be located in the same can. As an alternative to 3, you could try using just a loop of plastic tube to work as a container for moving air. In that case, you could have a coil of tube for the outcoming air where the moist filament goes first (e.g. 2 meter loop of 20 mm diameter plastic tube), followed by the area with the heater element. The plastic tube must be resistant enough to take the output air from the heater element but it doesn't need to be resistant enough to hold the actual heater element. As you have to keep air temps low enough not to melt your filament, the plastic tube for the hot air should be strong enough anyway. I would recommend following setup: (a) Vertical tube-like container for the silica gel, with a bottom cap that has lots of holes to allow airflow into the silica gel. I think some kind of metal grid might be the best option here. (b) A fan attached to the silical gel container pulling air from the silica gel. (c) A duct for the forced air to the heater element (d) A duct from the heater element to filament (e) The duct connected to the plastic hose that contains the incoming moist filament. The fan must be strong enough to pull enough air through the filament and still have enough positive pressure to push air through the long plastic pipe fast enough. And you must have dry silica gel in the container but you can have lots of silica gel in closed containers for fast use when you want to use moist filament. Some experimentation would be needed figure out if you want to contact the silica gel with cool or hot air, but I would guess that the correct solution is to contact it with cool air which is why I explained the above setup. The above setup results in zero heat sources (including the fan) before the silica gel but it makes things harder for fan performance. If fan doesn't increase temps too much, having the fan to push enough forced air into the whole setup would be much easier to arrange.
I would definitely like to see version 2, I live in a very humid part of the country and if some cans and a little extra electricity can fix the issue without waiting for the dry box to work I am all for it!!! as for multiple drying, I think you would still only need to really dry one at a time since you only print with one filament at a time..
I have an idea to further your research. How about making a heated container box for several filament spools. That box would allow output for whatever filament is chosen. It would feature a little fan and a heater. Both heater and fan would run 24/7. The chosen filament would slowly exit the box toward the printer, guaranteeing an always present dry material. The fan would only supply a very little flow, just enough to slowly renew the air via some small opening, but not enough to bring an undesired extra humidity. The temperature could reach well over 70C, making sure that the relative humidity becomes very low. The box could easily contain many spools, like 5 or 6. It could be made out of plywood, an easily available material that exhibits decent heat insulation characteristics. The inside could be lined up with an insulation lining. Just a thought.
I really like this project. I think you have the starts of a really good idea and I think you are more than capable of making something that performs at least 80% of what the $2000 dryer would at a fraction of the cost.
I've used the Thordsen. It works well enough after a sunlu dry box but lacks performance on fully saturated nylons. It would benefit from having actual airflow to improve the release of water. If you can make something a little better than the Thordsen then you would be doing good things for the community. The heater control element for their unit can be used with virtually anything else. When purchased from Aliexpress it ends up costing about 40 USD for just the control unit. So potentially you could take the Thordsen and reassemble it into something better.
I like your idea. I was going to make a big box dryer that would keep all my filament dry all the time using a 200 watt electric heater and a PID controler to control the temperature. The PID controler comes with a SSR and a K coupler and I also ordered a panel mount 5 amp circuit breaker to maker sure it was a good setup. Now, I think I'm going to use some EMT conduit and two rubber stoppers available from Home Depot to make a dryer like yours. Excellent idea.
I think this concept has a lot of potential! I think the easiest way to go about it is by using a slightly leaky system. This would let dry, hot air suck the moisture out of the filament and transfer it to the environment. Ideally you'd have a heating element at the bottom so the hot air goes up, and the filament going down. You could have a bunch of 2-5mm holes in the system that double as entry holes for the system, and a separate cap for the exit that does not let much air through past the filament (to reduce condensation there). The side would be a good place for something like a hair dryer to push air into the system. Having the heating be outside of the container would also reduce issues with heat creep and filament softening along the path. Coincidentally, I think increased heat creep and jams due to filament arriving at the extruder pre heated might be why this hasn't taken off yet. I have to say though, your tpu looks like it is stored in a shower or something. Mine only gets that bad if I leave it out of its box for months. Do you live in a very humid climate?
I'm very excited to see you explore this idea further. I understand that your goal is speed, but I have concerns about adding a possible point of failure to the printing loop. What i might propose as an in-between step is a reel to reel system: two dry boxes connected by the wind tunnel, a motor turns one spool and pulls the filament along, even if it only matched the current 40mm/sec speed you could dry a full spool in about two and a half hours, and by separating the drying from the printing I'd bet the speed could be dramatically increased.
Maaaan ! If you make some serious version 2 capable doing this in some compact small format, I am gonna be the first one to support you on patreon if you share project and let us tinkerers have the honor to use it afterwards ! Great Job !!
Even though I fully agree that current filament dryers are a pretty dumb solution when you think of it and inline filament drying would make a ton of more sense, my guess for it never taking off is the ease of use of current filament dryers compared to an inline filament dryer... A "traditional" filament dryer is pretty much "set and forget" and while the idea of an inline filament dryer sounds very cool in theory, you already touched on the many headaches it can introduce since you now have to factor in print speed and a whole lot of other factors making it a hassle to set up reliably. Having said that, I'm really looking forward to see the next iteration of this project!
I have an addition for you. When you showed the design of the filament spool and that it wont alow air flow to dry the filament due to it's solid sides. you can buy reusable spools with exposed sides or just make your own using mesh, this whould alow direct air flow from a large 12" fan from the side. ofcourse this requires you to transfer the filament to the new spool.
Awesome idea. I would love to see this developed into a reliable open source solution. I can see this becoming the standard in filament drying. Perhaps it could be made physically smaller by coiling some filament inside the dryer, perhaps around a small wheel/which like thing.
I'm a former analytical instrumentation specialist... I've read few comments below. My best bet would be the 1 feet diameter drum surrounded by few coil of filament. The drum is laying on the side with a bottom flange and a bearing (no motorization required). The drum is in a slightly larger drum box. You need proper fan + heater + thermocouple on Arduino or even simple mechanical thermal switch. Filament goes into the drum, spinn few turn and get out. The more resident time you need, the more turn you need. I've subscribed to see the next step!
thats what i do with my sunlu 4s, i'll dry it a few hours before i start printing & leave it on during printing. that way it is contantly preheating & drying the filiment while printing. it helps a lot with ABS i seemed to get less warping. this seems like a DIY version that would probably work pretty well. Keep it up Brother!
Yes please! I'm so totally going to make a can dryer now and have some fun on my own, though I doubt I'll have anywhere near as much expertise as you do. I'll be watching anxiously!
I wonder if you could take a piece of spiral-cut PTFE and run that down the length of the can, then use some sort of funnel to ensure most of the air gets directed into the PTFE spiral alongside the filament, so it's guided down the center but can breath outwards all around it. Also you might be interested in the membrane based electric dehumidifiers. "mdl series dehumidifier" is a good example you can search.
I found a simple solution. I use a large Rubbermaid type of container that holds all the filament. Maybe 10 spools. I fill up an old filament box with dessicant that I cut out lots of holes in. About once a week I throw the dessicant in the oven at just over boiling water temperature for about an hour. This keeps it around 24% in the box and I don't worry about it on the machine because it's not there long enough to absorb water. I don't leave it on the machine when not printing. This is simple and it works for me.
YES! Well Done!!! This entire past year, I have been shouting from the hilltops that filament drying is THE most important next place to put our collective energies! Sorting out this challenge for pro users and major hurdle for those new to the hobby before the constant epic print failures frustrate them away is vital for the future of 3d printing!!! Try Activated Alumina granules next .... they can absorb Wayyyyyy more water and have a healthier lifespan than silica gel beads and the color changing dye used in them. You do need a higher temp to drive off the moisture and "recharge" the granules but this ensures even better performance over a longer span of time .... depending on your humidity level. I also recommend getting a dehumidifier for any closed room that you have your printers in.... so other off-the shelf technologies can easily be added right now for home users. I know dehumidifiers aren't cheap... but compared to destroying and wasting 2 rolls of filament in cost they are worth it..... get the biggest one you can so you don't have to empty it as often.... set a phone reminder as often as it takes to go empty it. ....and if you don't have dry boxes, or a dry room, enclose everything you can (entire printer and at the very least your filament, and add a cheap house fan for PLA cooling inside the enclosure ( so get/build the biggest one you can or get the smallest fan you can fit).....this will avoid the Creality X1 problems seen with excessive heat requiring the top be taken off to vent. 👍👍
That Creality dryer at 1:02 has a blower that blows the hot air directly over the spool opening and the air flows continuously around spool in the direction that the filament is wrapped (why it's shaped the way it is with the rounded lid). I have one and I have notice it does dry quicker then most options and once you use it as you print the deeper layers get dried as they get closer to the surface. I have been very happy with it so far.
A quick search into this and there's two approaches that seem common. One is vacuum which was mentioned by other people. However, microwave drying is also apparently a good method to remove water from plastic since microwaves are tuned well to excite water molecules specifically. The main need of any method is to not introduce too much heat or you can degrade the plastic. We microwave plastic containers all the time, but we never really think about moisture content in the plastics. We also use many plastics in food use that don't really absorb any moisture. One would need to test if there is sufficient moisture in a filament spool to cause heat problems, aka melting or degradation of the plastic molecules where we might see a much weaker part.
For those interested storing spools you've dried can easily be done with a home vacuum sealer and their plastic bags works really well - on the cheap you can also use a dry bag and a vacuum cleaner with some silica gel for storage. Remove the air you remove the moisture buildup and using a rare compound or color isn't going to be several day job drying it over and over.
There are a few ceramic heater cores that are in a cylinder shape for 3d printing. A few could be added in series. It's might be better because you could more precisely control temperature and be way closer to the filament, but you would need some electronics. Something like the CHC kit 24v
It may have been suggested, but when you come close to a working version, try to test how much air-flow is needed for optimal results. you may find that you need less than you needed and therefore smaller fans, less energy. GREAT work. You have my subscription! Russ from Ocala, Florida, USA
I am very interested to see this idea explored further. Inline filament drying seems like a no brainer compared with the other solutions on the market!
The physics of this problem are both interesting and complicated. The reason heating up the air works is because hot air has a higher capacity to hold moisture than cold air. This is why we frequently track humidity as relative humidity. I don't fully understand the physics, but the water molecules in the filament are more likely to sublimate into a material such as hot air when the relative humidity is low. I think your best results with this method would come from passing the air through a desiccant chamber on the cold side (relative humidity is higher and therefore the moisture is more likely to be pulled out by the silica) before it passes through the heater, which will further lower the relative humidity.
For the commercial inline dryers, they are probably using Zeolite, it’s used as a moisture absorber in industrial applications. It can be regenerated by heating it up, just like Silica Gel.
I came up with this idea as well about 3 weeks ago. The problem is: It is basically impossible to know the correct temperature while printing because of the varying speed of the filament itself. I found a guy on the Voron-Discord who did the same thing with two V6-Hotends slapped together, but he said he still dries it conventionally. I ended up engineering a new filament dryer because I am using an Industrial spool. I guess, if you know the density of the material, the speed of the fillament being consumed and a fixed temperature, it might work but at this point, it might just be easier to dry it conventionally and that's the reason for the other thing costing 2100$ in total I guess. The smaller part probably just measures the filament speed and the larger part dries it.
Your English is phenomenal and your wife is amazing/cute for correcting you when she does hahaha. No worries I have to do the same for my partner as they're from Colombia.
This is definitely something I find interesting. I also had the same though about in line filament drying. They do it for pneumatic systems, especially for painting, so there likely is a way to achieve a similar result with filament.
Someone probably already suggested this, but you could create a longer drying path by having it wrap around a free spinning spool inside the heated chamber. This can keep the chamber relatively small , and I’m imagining that the spool has a perforated center core that lets airflow reach the inside of the wrapped filament. I would think only a few wraps around could create decent amounts of exposed length so you don’t have overlapping layers on the spool like you would with a full spool.
Cool idea man, thank you for sharing. Maybe a few reasons why this is not popular, first we don’t live in very humid environments. Secondly we keep it in a drybox and insert it into a hotend with a reverse bowden. Secondly our printers are additionally enclosed. I have printed with almost all filaments you can think of, very hygroscopic filaments like TPU or water solubles and have never experienced such extreme stringing or water logging.
I like your drier. If you could make it so it circulated the air internally and contain desiccant it might work better still. Your device looks like it pulls fresh air from outside, but like you said that's a 60% humidity, which sounds less than ideal.
I have a $39 stack dehydrator with a cardboard outer sleeve. I do a few spools at a time then store them in sealed containers with moisture capture sachets. Works perfectly. I do like your novel approach. 👍🏻
One thing I'd suggest looking into, though it'd be more complex (and something I've been toying with as an idea) is figuring out how to integrate a silica desiccant regeneration cycle. The idea is to run cool air through silica, or some other desiccant, warm the air (dropping relative humidity in the process), blow that over/around the filament, then have a second system that somehow blows warm air over another patch of the desiccant. Something like a rotating plate of desiccant so that "dry" desiccant is exposed to the incoming air to dry the filament, and "wet" desiccant is constantly dried, with a slow motor to keep the whole system rotating. This isn't a new idea; it's how plastic pellet driers work in injection molding on an industrial scale, though those often dry 200+ pounds of plastic continuously.
Yes, this makes so much more sense than drying the entire roll of filament... I never understood who dries an entire roll of filament for 8 hours. In the US, sure, electricity is basically free over there for some reason, but in most other countries it's just crazy to run an oven for 8 hours just to dry some filament... Would love to see a more "proper" V2.
I like the idea, it's the implementation and other things you mentioned. It worked, at 40mm/s but what happens at say 200mm/s or throw in an AMS. It would be easier to make the AMS a dryer, some other company did this, can't remember which one though. This would solve most issues. I say most because most filament is dry out of the sealed bag but every once in a while, it's like it sat on a shelf for a week before it was sealed. Also, the most hydroscopic filements are all CF of GF or some sort of blend. Since you can actually see the carbon fiber in them, it's easy to see how they leave more pockets of air for moisture to absorb into. The higher the amount of CF there is, the more hydroscopic it is. While filaments like TPU and ABS are hydroscopic, they don't come close to any CF blend for that reason. Most recommended 24 hours of drying at 80°C for PA-CF. TPU doesn't need those temperatures or time so that would be the real test. Personally, I recently got the Polydryer by Polylute and it's great. It has a separate dock from the storage container, comes with reusable dececant, has a hydrometer in it, and already has mods on printables for various things like drying to boxes at once. The storage boxes, while not cheap (25 US),, are reasonably priced and have excellent build quality and comes with a hydrometer, reusable dececant and spiol. The box sits on the dock, hot air is blown up the back, the front port sucks it in which is where the dececant is. You can remove the box and seal it air tight with some plastic caps with rubber around them. The dececant and hydrometer stay in the box. Now you can simply store that box and put another one on it (extra storage boxes are sold separately). It's the nearest advantage to come to dryer boxes in a while which isn't saying much but still, more versatile than anything else on the market. The fact they are allowing community mods is also a plus. With CF there pretty much has to be a dryer box. If you can get CF working on this for a reasonable cost then that's when more interest in this will happen. Out of all the rolls of TPU I have bought there was one with major issues due to moisture, others printed fine for the first print then I did dry them and store them in ziplock bags with reusable dececant but once I get 3 or 4 of those dryer boxes, that's really all I need as I mostly print PLA which can't sit out for months.
You could probably use one can if you put a small spool inside the can and wrap the filament around it say 8 or 10 times. As the filament is drawn out, it'd always keep however many loops you have on the spool.
Very good! With a PTC heater and the same material I'm making a chamber heater. I Have no needs in instant dehydrator, for the moment (I have a chamber for the mounted filament, so when I become to design I chose the filament and meanwhile it dehydrate, and the for the others I use vacuum bags) but I have a suggestion for you: if you give a longer path to the filament with a pulley system (like the grouped ones for sails, for example), you can optimise your idea, but only for flexible, obviously, or if you use a flexible extensible exhaust tube, you can achieve larger radius curve so you can work with non flexible too. For silcagel: with a little precision scale (0.01g) you can make bags with fine mesh tissue (new silcagel pre-weighted), and with a microwave oven you (250g=max 450w) you can renew perfectly the silcagel in about 15 minutes, checking the bags original weight. Is important to cut the session in parts (I use 5/10min) and in between, you take off the bag and moving it, you leave evaporate the moisture and cool the silcagel, until you can hold it in your hand. Is important that the power of microwave never can make the bag so hot that you can't hod it in your hand for less than one/two seconds: going over the 100C° degrees can only degrade the silcagel.
interesting idea sir. there is one thing though, you introduce several new points of potential failure with this compared to the regular way of drying the filament when you get it instock. store it in a drybox for later use. this way you dont have to wait to print anything :) regular pp storage boxes with doorseal tape and a hygrometer. add some silica in a box with a fan and it is just a squeaky clean and cheap solution.
This is very interesting! i've been using silica for years, but kept them in a sealed box with the filaments in it. Never thought about "in-line" drying, i'll definitely try it
One thing that I have learned about TPU is that slower print speeds will exacerbate moisture issues. I used to have really bad surface finish with my TPU prints until I upped the print speed significantly.
In my mind, I envision a single piece copper tube with a diameter close 2-3mm in which the filament goes through. A 3d printed adaptar is printed at the end that makes a Y, one side of the Y allows the filament to go outside to the printer, the other end goes to a much bigger opening that connects to a 120mm fan that creates a vacuum. . On the other side of the tube, you have the same Y: Just that instead of a connection to a fan, you're breathing dry air from a box full of silica gel (Similar to the first commercial product that you showed. . Then, you can wrap heating elements around the copper tube (Like the ones used in motorcycle handle grips) and finally and insulation around all of that that would keep the copper tube heated at a reasonable 60-70 degrees without too much use of energy. This feels like such a cool project!
Store your filaments in a pumped out vacuum bag with indicator silica. Don't leave filament in an environment with a RH% higher than you want your filament to absorb. There are many mods that can be done to the basic Sunlu dryer to recirculate the heated through indicator silica. If your're printing in an environment with a high RH%, stick the filament in a filament dryer while your printing, or at least on a roller in a food box containing indicator silica with the outlet hole just large enough for the filament to pass through, and keep the device you're using close to the 3D printer while you print.
Try using the drying system or even just the desiccant cartridge from a tractor/trailer. It has to constantly dry the compressed air used for braking , actuators, etc . The parts can be sourced for cheap. Or just store your filament spools in white rice. Thanks for the video!!!
I just added a fan to my old filament dryer and my new filament dryer has a built-in fan. Running this removes the moisture fast. It doesn't matter that only outer spool layers are dried because the moisture is removed so quickly using the fan and heat together. I wish the BambuLab AMS also had a dryer built in, besides the desiccant.
I have been doing this for years. But I use a very large sealed chamber I place the whole spool inside. I use a fan that heat the chamber and produces violent air flow. The filament is dried as it comes off the spool and the moisture is deposited into aluminum oxide cans that I recharge each time I place a new spool in the chamber.
Hey there 3d printing ideas, I work with metal filaments and run my own small 3d printing/ design engineering consulting gig. I have worked with metal filaments and they have a variety of problems of their own and I would like to share what I have done to improve my filaments. I personally would increase the heat and airflow. in an industrial injection molding operation a large part of it is drying the pellets before feeding them into the injection tube. The drying time varies based on a number of variables. So in your case you'd have to account for water accumulation in the filament, air humidity, the speed of the filament going through the nozzle and as well as what ways you can pull the moisture out of the filament. Additionally, how wet the filament can be to print "normally". Nylon gets stronger after it gets wet, but you can't print with it when its wet. If you're someone that wants to speed print using "wet" filament, you'd have to have a vacuum chamber set up to pull the moisture out of the air then draw it out of the filament towards the outer edges of the filament itself, then you can push it through a filament dryer with desiccant. The name of the game is to make the air surrounding the filament drier than the filament itself that way the water pulls out. Its simple physics after that, the hotter it is the faster the moisture will dry, the more air flow you have the easier it will be to pull it from the filament. add a desiccant and now you have even more things pulling moisture out. The vacuum chamber is really overkill for basic and noncritical components but it ensures you get the most moisture out of your filament as fast as possible prior to printing and then you can precondition it by warming it like you've done. You just have to extend the warming time to match conditions where you're printing. so a "wetter" filament will require a longer heating chamber. A more humid climate will require a longer heating chamber. Prewarmed filament will also print easier as the hotend doesn't have to bring a filament from room temp to melting, rather 100c to 220c for pla for example. Hope that gives you some tips and tricks. I would also suggest utilizing nichrome or kenthal, it allows you to control the temperature more and you can wrap it around a glass tube that you can run the filament through alternating the hot portion with an exposed air portion. In my tests it worked good but that was my v1 I feel like putting it close to the filament AND adding airflow in the can with the desiccant would be interesting to see. I just leave my desiccant in the box.
If you decide you want to really knock this up the S tier, I would recommend putting your filament on a spool and holder into a vacuum chamber and feed it through a series of silicone wipes. Now you've got molecularly-dry filament. Getting a 12CFM pump is pretty cheap now days and will empty a large chamber in minutes.
Contra flow drying, increasing the length of the drying path and insulating "the cans" so the air stays hot. Also the diameter of the tube does hot have to be the diameter of the can. A diameter of about 25mm should be adequate. A recuperative heat exchnger could also be used to recover energy and elevate the drying temperature.
Thanks guys for the lovely comments, i guess ill have to make V2...
Yeah.. you'll have to 😂
I'm really looking forward to seeing if you can make this work. I print materials that need to be extremely dry and I've just been practicing vigilant drybox discipline, but a more forgiving solution would be excellent.
Yes give me more!
Essentially, you're preheating the filament prior to feeding into the print head.
Might consider options to increase the filament exposure to the heated air ... but take care not to overheat the filament.
Please do. I hate filament boxes. Limits spool sizes I can use, and for some filamenta even causes clogs or breaks because some filaments don't like to be arched through the PTFE tube and whatnot. Really hate it. If you can achieve the same result without dry box, it would be awesome. Maybe play with silica inside the tin cans
The “You are a genius”… “Thank you hunny” is so sweet ❤❤❤
It was absolutely wholesome and adorable.
that's one clear green flag
I would 100% love to see a version 2 of this inline filament dryer system, especially one that can do 4+ filaments at a time. I've been needing to print TPU, and something like this will be super cool!
Drywyse is the only working in-line dryer on the market. Yes, it's for industrial use mainly due to pricetag
If you heat TPU as you print it it stretches too much and print fail
@@user-uf8nn6he3e Literally the first 30 seconds of this video showed TPU being inline dried and printing well.
@@bosstowndynamics5488 not my experience printing a functional job
I’m with you, but only if he promises to never touch the shotgun mic again
"Then I did what I needed to do in the first place, research" - me, every time I start a new project haha. Great video, insane drying set up, and really interesting results!
As they say, a month in the laboratory can often save a full hour in the library.
For those that were wondering, hes using a 300w PTC heater.
I had to dive into his printables text to find the wattage.
"I want to print now" Love this. Keep up the good work!
So relatable, the part about forgetting it as well 😆
@@spacerabbit1619yessss this is the major reason I don't print something.. god damn drying
100% yes, I would love to see a version 2! This is a MUCH better solution than filament dryers!
That's an opinion.
@@REDxFROG and you're a frog
The best way to remove surface moisture from the filament is to create a vacuum, use a vacuum pump in a container and remove the atmosphere, the moisture will sublimate off the filament. You don't have to unwind and rewind the filament. you can cycle the process once or twice. I am buying a vacuum pump for chemistry experiments to change the boiling points of whatever fluid I am working with.
I think CNC Kitchen tested this with good results.
Built such a system myself, but found that you need to raise the temperature for it to be effective.
Putting an old 3D-Printer Hotbed into the chamber is not difficult, but passing the power inside was challenging
@@timw7406 Vacuum is a good heat insulator. Maybe, It's better to repurpose old Dryer Machine to preheat and predry. And use vacuum as a finishing touch.
@@vasiliynkudryavtsevAgreed. VisionMiner has videos on this. Dry in repurposed bench top oven. Then a couple of minutes in vacuum chamber after. If you had the budget, a lab vacuum oven would be pretty slick.
@@vasiliynkudryavtsev Because the vacuum is such a good insulator you can use a small amount of radiant heat to achieve lots of heating. An entire spool might optimistically radiate 2-5 watts of its heat away when heated to 100c.
A radiant heat source thus only needs about 2-5 watts of output to heat hitting the spool in order to get it to that temp.
A common choice is incandescent lights shining thru the walls for cheap at home methods.
I also live in an area with around 65% humidity basically all the time and would love to see this continue development. Subscribed!
Retired now, I spent 2 years developing dehydration test system for our motor Stators, moisture in stators in AC not good. our measuring system used DC heat on the stator wires, and a vacuum and dry nitrogen purge. the real good part was the dry Nitrogen purge, i would apply about 15psi dry nitrogen to the stator holding cavity, pull vacuum, the nitrogen would absorb moisture and became the medium for me to capture it in a Liquid nitrogen bath. we would weigh test tube before and after to get the weight of the water pulled from stators. this was a 4 hr test cycle, not piratical for 3d printing. Just letting you know how a measuring system works... perhaps it might trigger a thought for you. Me i keep all my filament in dehydration containers all time. just because i know how hard it is to dehydrate. your idea super good - should help lots of us.
YES, we're all interested in it, please do a follow up!
I would like to see a version 2 of this. One idea I had was to make your "in line" drier into the shape of a toroid so you could loop the same section of filament another time or more. Kind of like how a beer line chiller system works. just a quick idea. not fully thought out. Thanks for the great vid. subscribed!
I had the same thought! Maybe the torus is filled with silica gel and the air circulates sucking the moisture from the filament and it gets pulled into the silica gel. When the system is idle it gets hotter inside to dry out the silica gel.
This is related to a classic problem given in thermodynamics classes. The question is how hot does the oven need to be to cook the potato in ten minutes. The answer is that it can’t be done because you’d set the potato on fire at the needed temperature.
With filament it’s probably the glass transition temperature that puts a limit on how fast you can dry it out.
I guess you could make the can longer so it has more time to dry out? Like at some point the plastic will be too wet for it to be possible but it could probably still be good for certain moisture levels.
Depended on how big is the potato
You can def cook a potato in a microwave oven in 10 minutes
@@TomCrockett-bl1gp yes, but you can't microwave filament (can you?)
@@MrNoipethat'd make a pretty good experiment tbh. I guess it would only work well with materials with a high glass transition point though (if at all)
Five meters of 5mm ID tubing (like what you'd use for a bowdon tube, only larger) with a tee fitting at one end blowing hot dry air into it. The filament goes into the other end of the tube, against the flow of hot dry air, and comes through a seal on the other side of the tee and into the printer. The only issue is you'd need somewhat higher static pressure, maybe 1bar or so.
Moisture migrating through a material pretty much acts like heat and can be fairly well modeled with thermodynamic equations. To get maximum moisture out of the material you want the highest possible difference in moisture content, highest possible temperature without damaging the filament, and the longest possible contact time. It's a heat exchanger except you're exchanging water instead of heat. Turn it into a very long and narrow counterflow unit and you'll probably get excellent results.
This is a bit of a departure from your form factor suggestion, but I was wondering whether, in order to increase the moisture content differential, silica drying beads could somehow be added in line, or before injecting the hot air into the drying chamber, or even both. That should provide an overall dryer environment, and the optimal volume of beads could be perhaps calculated to optimize the whole thing. Am I making some sense here ? Thanks :)
@@crepuscule47 Drying the hot air on the way in would be ideal for this, especially for those of us who live in swamps :) Higher pressure air, through a drying tube (silica, probably), then heat it, then through the counter flow tube seems like the order of operations to me.
If you could cool the air down first and condense the water you could then heat it up and the relative humidity drops.
The filament needs a lot more time in the drying zone. I was going to do this but set the project aside a year or two ago. Unless you want a super long tube, you need a system of pulleys so the filament can run back and forth across the hot dry air zone several times. You probably also need active drive for the pulleys to avoid overloading the extruder.
Yeah, and still it's possible to overload the extruder even with direct drive
why not loop it like a beer line chiller? shape the drying zone like a toroid and loop the filament a couple times inside.
Maybe a copper coil?
@@mirag3304 You can do that but you still need either a system of pulleys or coupled input and output filament hobs so the same length of filament remains in the zone rather than pull from the extruder trying to tighten it down to a point (open core) or to tension itself against the core (toroid) just like it'd do on a spool.
Why not implement the drying zone into a filament buffer? Could be an idea
No idea how well this translates but I’ll share anyway. I used to work with giant coils of steel wire. When prepping it to go into the machine it was wrapped onto a a second drum a few times to help with feeding and getting it all lined up. Something like that inside the cans would add a lot of surface area without needing to double back via pulls or something else as other have suggested. My thought is another tube inside you can corkscrew the filament on giving a much more even and long drying time.
That way only one side of filament would be exposed to air though
@@JGnLAU8OAWF6 spool made of mesh
That sounds like a great solution. Instead of a long skinny soup can, use something large diameter and short length like a cookie tin. Put a short spool inside on bearings so it turns easily, and loop the filament around several times. I think the filament would try to migrate sideways along the spool as it turns, though. How did the industrial machines deal with that?
@@dekutree64If it's anything like industrial commercial software: They probably didn't deal with that. They could've easily left it janky while resting on their laurels due to patents and other ways to prevent competition. Then, they just charge an outlandishly large price for their shit.
Friction force will be huge problem with TPU
Although other people may have invented this, this is an ingenious idea and I think that the cost of what you invented compared to the $2000 dollar price tag is ridiculous. I will be making something similar, or taking some inspiration to make something amazing just like this!
Would love to see v2! Could so get behind a printable system. Maybe using standard PVC piping instead of the cans for more safety and less jank.
Aluminum air duct. Like dryer ducting. Flexible, easily ventable, and heat resistant.
Hahaha pvc, more safety and heating don't go in pair
@@sanctusletum8522 Then wrap the outside and/or inside with heat insulating material, ceramic paint? Your usual house wall heat insulation material? The usual heat materials used for a forge? So many ideas.
PVC not going to cut it, the temp go's to 110C
@@3DPI67Absolutely - no idea what the thermal limitations are of PVC. Just sharing idea ✨vibes✨😂
Kudos! You hit the nail on the head. Just like you, I hate drying a whole roll of filament for umpteen hours only to forget what it was that I needed the filament for when the cycle is done. 🥴 Keep up the research and development and I’ll keep watching. Good luck! 👍
In Poland we make filament dryers from cheap vegetable/fruit/mushroom dryers, since first printers became popular, as drying the forest mushrooms is very popular here. They have airflow, 250W heater driven by thermostat and only need replacement of drying sieves with spool holder.
That's what I use here in Florida, where we have weeks of 90% humidity in the summer.
How much?
Yep I use a fruit dryer in the UK, got it on Amazon, works great
Yep these have been staple filament dryer for years. I think the commercial ones are inspired by them. Do the commercial ones not have airflow? That's a pretty fundamental oversight.
@@sligit Commercial ones resemble commercial heaters for heat-shaped shoe medical insoles. They are safe, look nice and you are sure that using them you will not be sued for damaging patient's shoe. The fact they don't work at all (insole remains thermally untreated) is the very least concern there.
What would be cool is a drybox with dedicated "in-line" dryer section. Basically use one heater and fan to pre-dry the complete spool, but have the end of the filament moved over some rollers inside the box before it leaves for the printer/extruder to improve drying of the filament from all sides.
What i see a bit of a challenge with all the longer in-line dryer options is, that it should still be easy and fast to change the filament spools. So you would have to design it in a way that gives you good access to the full filament path.
I'd love you to explore this topic more for sure!
I just got the stupidest idea.
Reverse bowden (or halfway bowden or smth) where you just blow hot air through the PTFE tube the filament goes through.
That could also work as the exhaust for the drybox, at least partially.
But you could also do this from the printer going out.
A v2 that worked with the AMS would solve filament changes handily since it would be in between the AMS (and therefore the spools and also the automated filament feeder) and the printer.
@@Litl_Skitl The only problem with a long bowden tube is probably the drag you create. Even if you use a tube for 3mm filament with 1.75mm filament, you will need to curve that tube around and you create friction. Maybe a very stron extruder does not care much, but i'm pretty sure you would see a difference in print results just because of that. And with flex filaments, it would get even worse as you have long stretch path, thinning out the filament at the extruder gears making it hard to grab and prone to slipping.
@@ProtonOne11 Reverse bowden at least shouldn't have that problem. Neither should a bit out of the drybox have that.
Yeah, Bowden works for some things not really great for stretchy, flexible, compressible stuff.
Thats part of the issue with some inline dryers.
The Sunlu S4 you showed has fans and vents. Airflow. It also has humidity sensors that will keep drying as the humidity rises.
Also most dryers allow you to print directly from the machine, while it's drying.
It also has a storage mode keeping the filament dry and ready to go on standby
You missed the part when he explained that even with fans the airflow hits only the outer layers of filament on a spool. Inside layers are still a wet noodle.
The S2 new version also has at least one fan
@@DmitrySholokhovno they aren't. It only takes some time until the heat reaches the core of the spool. But you don't print from the core. You people are searching for a problem which doesn't exist.
Use a filament dryer while printing and it's all good. Let it warm up 10-20 minutes beforehand. That's usually the time to slice, upload and warm up the printer anyways.
Most comments of people don't even seem to have a dryer and start to think this guy invented something everybody needs.
@@REDxFROG thank you for saying this. After watching the video, I was still scratching my head thinking this isn't needed. It worked when he took a drybox of filament and printed with it... of course it did. It probably would have printed the same without his tin can contraption. If you store filament this way too, you can print whenever you want without waiting for it to dry.
Thanks for the video. I almost patented an inline dryer for filaments 12 years ago, did the calculations about needed drying lengths, designed simple setups etc.
As to the why we dont see this more often: In the 3D printing industry, somehow the knowledge about how to properly prepare material for extrusion seems to not have transferred from regular extrusion processes like injection molding. Every TDS for granules used in conventional extrusion has at least a drying recommendation, most even have a specified maximum residual moisture content for processing. Stratasys went around this in providing (well) sealed containers, which mostly were used quick enough so the users did not often encounter this issue. Then after the patents expired and reprap started using welding wires for printing, nobody really thought about this. Since more and more people are into 3D printing now with materials more prone to issues with moisture (PLA and ABS are not as problematic as TPU or PA), the topic gets more attention. I still see posts almost daily about users boasting they never dried any filament and that it's plain useless. As a material scientist myself, statements like this make my head hurt. That's like saying you don't ever need to clean a surface before using glue on it. Of course you always can be lucky, but it's just good practice to stick to proven and meaningful recommendations.
Lots of comments on this video as well about not needing to dry filament, i guess they leave in a dry area and or store filament properly. Maybe you want to share some advice for v2 from your experience?
I been thinking about this for ages, just never have time! Glad someone picked this up.
This is brilliant. I would try using copper pipes which can be bent into a spiral shape to have a much longer length for faster printing while taking up less space
copper would pull out all the heat from the air.
@@robogoofers9131 an 8 feet long insulation sleeve is about $2 at the hardware store. Depending on the design, you can have external heating instead of forced air through the long tubes and take advantage of the conductivity. Lots of options.
I was just thinking the same thing...Except instead of copper, you can use rings/guides throughout the can, leaving most of the filament exposed. Think of when running cat 5 in a drop ceiling, you just put hangers every so often to just help keep the cable suspended.
4:30 i think they might be using molecular sieves just based on appearance, but i’ve got no idea what else they’re using in the machine in order for them to justify that $2000 price tag
Science!
I guess they just want to be payed for the efforts of R&D they put into making the device. And i could see the machines being assembled by hand and not really optimized well for fast and easy assembly. It's a niche market with low volume, especially for a "professional" solution. Going for a cheap low profit margin solution was not theyr goal, as the big brands already cover that market with the cheaper filament dry boxes.
Looks like an MS to me too. Does seem like a pretty hardcore solution.
They also installed a patent pending marketing system.
most likely. the air is being pumped through them to super dry the air. id be interested to see someone buy a vacuum drier and see how that works
I bet the drywise uses a mostly closed loop where the warm air is recycled through the molecular drying sieves instead of continuously heating cool air. Great video, excellent first POC, and love your humor. Looking forward to part 2!
I think this is REALLY interesting, and it seems you have proven it really hasn't been done in a feasible manner. I think there is a product here, just waiting to be invented and marketed. You got this, man! DO IT!
Your video confirms that behind every great man there is a great woman.
This looks very interesting. I wouldn't necessarily go with tin cans, but rather a tube with relative big diameter (>6mm) and blow hot (60c-70c) air into this tube, either from the printer side, or from the dry box side. Allowing the tube to be open at the other end so the hot air and the humidity are free to egress the tube after drying the filament. There are plenty of PTFE tubes that can do this which are easy to buy from China, or you can use a PVC pipe, it might even work with your local garden hose as well - depending on the material it is made of.
I was thinking the exact same thing, and given PTFE has a high temp resistance, you could wrap some sort of heating coil around the tube to keep the flow of air warm
Might want to go from printer side so that the end always has the driest air....
It must be cool to live at places with free energy
PTFE is overkill, simple nylon or even most silicone tubes would be cheaper and have more than enough temperature resistance for anything short of PEEK
Why a big diameter tube?
This is a great idea! This is my first video I've found of yours. Excited to see more!
This is amazing.
I would greatly appreciate seeing this turned into a finished project. Maintaining correct filament moisture levels is critical to good quality 3d printing. An inline filament dryer that has a reasonable cost and is effective would be a significant improvement.
You can't remove the moisture from filament, that has absorbed moisture over time of weeks and months. One easier way is to loadup the filament in a vacuum drier(used for degassing epoxy resins). You also need to keep some dessicant inside, and keep the container warm(85C if you put the reel on pedestals). You could dry any filament this way, in a matter of an hour or two at the most. You can also refresh those brittle pla, by putting it in 70C water for 15 mins, and doing the above.
Boiling? You're gonna destroy your tolerances like that. 50° C max
@@smoothbraindetainer If you know how 3d filaments are made, extruded filament are drawn in water, before wound into a spool. There needs to be a min. moisture needed to keep filament from turning brittle.
@@aware2action the water is not the problem. Did you read my comment?
@@smoothbraindetainer A brief dunk in hot water, will not affect tolerances, as long as the filament is dried later. Also, the filament is not stretched during the water dip. Have done this to my verbatim brand filaments with no issues.
@@aware2action nice editing of your original comment
If you have to use a drybox, then the advantages of the inline system completely disappear. You would need to run the inline dryer longer and use more power than a conventional dryer. In a conventional dryer you would just dry it and store it in a dry box.
yeah when he put it in the drybox, is it a dry filament or an already wet one? atleast if it's a wet one then it means you can remove the need for initial drying time
Pretty sure he's using that to hold the spool, nothing more.
This would be a very nice and cheap way to dry filament, especially when its like 60%-70% humidity in Germany right now. Would love to see more :)
Nice to see someone thinking out of the dry box. You're absolutely right about air circulation. Great work! Keep going!
The whole "nobody doing this" thing could easily be explained by companies registering patents. If there was one that is.
Like with a lot of other 3D printing ideas
@@kendokaaa and with a lot of other ideas, good or otherwise
Interesting idea! Some ideas for improments:
(1) Put heater elements and fans pushing air into opposite direction to the filament movement. That way the filament hits the hottest air the last before exiting the tube. The hotter the air, the less the humidity of the incoming air and you need to dryest possible air at the end of the process. When filament is really moist in the incoming feed, any amount of air movement will already improve things so it would be wasteful to use the dryest possible air in this end.
(2) Feed the air from the fan through lots of dry silica gel before heating the air. That way you start with as dry air as possible before entering the heater element which makes the air even dryer.
(3) Setup the system to pass the filament through the same tube multiple times. The amount of water that evaporates from the filament is probable not enough to fill the volume of the tube so when you have lots of airflow in the tube, the moist air from the filament will probably stay close to the filament and another looped strand closeby can be located in the same can.
As an alternative to 3, you could try using just a loop of plastic tube to work as a container for moving air. In that case, you could have a coil of tube for the outcoming air where the moist filament goes first (e.g. 2 meter loop of 20 mm diameter plastic tube), followed by the area with the heater element. The plastic tube must be resistant enough to take the output air from the heater element but it doesn't need to be resistant enough to hold the actual heater element. As you have to keep air temps low enough not to melt your filament, the plastic tube for the hot air should be strong enough anyway.
I would recommend following setup:
(a) Vertical tube-like container for the silica gel, with a bottom cap that has lots of holes to allow airflow into the silica gel. I think some kind of metal grid might be the best option here.
(b) A fan attached to the silical gel container pulling air from the silica gel.
(c) A duct for the forced air to the heater element
(d) A duct from the heater element to filament
(e) The duct connected to the plastic hose that contains the incoming moist filament.
The fan must be strong enough to pull enough air through the filament and still have enough positive pressure to push air through the long plastic pipe fast enough.
And you must have dry silica gel in the container but you can have lots of silica gel in closed containers for fast use when you want to use moist filament.
Some experimentation would be needed figure out if you want to contact the silica gel with cool or hot air, but I would guess that the correct solution is to contact it with cool air which is why I explained the above setup. The above setup results in zero heat sources (including the fan) before the silica gel but it makes things harder for fan performance. If fan doesn't increase temps too much, having the fan to push enough forced air into the whole setup would be much easier to arrange.
I too am impatient and forget why I was drying a certain filament! 😆 More testing and iterations please!
I would definitely like to see version 2, I live in a very humid part of the country and if some cans and a little extra electricity can fix the issue without waiting for the dry box to work I am all for it!!! as for multiple drying, I think you would still only need to really dry one at a time since you only print with one filament at a time..
I have an idea to further your research. How about making a heated container box for several filament spools. That box would allow output for whatever filament is chosen. It would feature a little fan and a heater. Both heater and fan would run 24/7. The chosen filament would slowly exit the box toward the printer, guaranteeing an always present dry material. The fan would only supply a very little flow, just enough to slowly renew the air via some small opening, but not enough to bring an undesired extra humidity. The temperature could reach well over 70C, making sure that the relative humidity becomes very low. The box could easily contain many spools, like 5 or 6. It could be made out of plywood, an easily available material that exhibits decent heat insulation characteristics. The inside could be lined up with an insulation lining. Just a thought.
Very interested please keep it up! Looking forward to all your work regarding filament drying.
I really like this project. I think you have the starts of a really good idea and I think you are more than capable of making something that performs at least 80% of what the $2000 dryer would at a fraction of the cost.
I've used the Thordsen. It works well enough after a sunlu dry box but lacks performance on fully saturated nylons.
It would benefit from having actual airflow to improve the release of water.
If you can make something a little better than the Thordsen then you would be doing good things for the community.
The heater control element for their unit can be used with virtually anything else. When purchased from Aliexpress it ends up costing about 40 USD for just the control unit. So potentially you could take the Thordsen and reassemble it into something better.
I like your idea. I was going to make a big box dryer that would keep all my filament dry all the time using a 200 watt electric heater and a PID controler to control the temperature. The PID controler comes with a SSR and a K coupler and I also ordered a panel mount 5 amp circuit breaker to maker sure it was a good setup. Now, I think I'm going to use some EMT conduit and two rubber stoppers available from Home Depot to make a dryer like yours. Excellent idea.
I think this concept has a lot of potential! I think the easiest way to go about it is by using a slightly leaky system. This would let dry, hot air suck the moisture out of the filament and transfer it to the environment. Ideally you'd have a heating element at the bottom so the hot air goes up, and the filament going down. You could have a bunch of 2-5mm holes in the system that double as entry holes for the system, and a separate cap for the exit that does not let much air through past the filament (to reduce condensation there). The side would be a good place for something like a hair dryer to push air into the system.
Having the heating be outside of the container would also reduce issues with heat creep and filament softening along the path.
Coincidentally, I think increased heat creep and jams due to filament arriving at the extruder pre heated might be why this hasn't taken off yet.
I have to say though, your tpu looks like it is stored in a shower or something. Mine only gets that bad if I leave it out of its box for months. Do you live in a very humid climate?
I'm very excited to see you explore this idea further. I understand that your goal is speed, but I have concerns about adding a possible point of failure to the printing loop. What i might propose as an in-between step is a reel to reel system: two dry boxes connected by the wind tunnel, a motor turns one spool and pulls the filament along, even if it only matched the current 40mm/sec speed you could dry a full spool in about two and a half hours, and by separating the drying from the printing I'd bet the speed could be dramatically increased.
Maaaan ! If you make some serious version 2 capable doing this in some compact small format, I am gonna be the first one to support you on patreon if you share project and let us tinkerers have the honor to use it afterwards ! Great Job !!
Even though I fully agree that current filament dryers are a pretty dumb solution when you think of it and inline filament drying would make a ton of more sense, my guess for it never taking off is the ease of use of current filament dryers compared to an inline filament dryer...
A "traditional" filament dryer is pretty much "set and forget" and while the idea of an inline filament dryer sounds very cool in theory, you already touched on the many headaches it can introduce since you now have to factor in print speed and a whole lot of other factors making it a hassle to set up reliably.
Having said that, I'm really looking forward to see the next iteration of this project!
I have an addition for you. When you showed the design of the filament spool and that it wont alow air flow to dry the filament due to it's solid sides. you can buy reusable spools with exposed sides or just make your own using mesh, this whould alow direct air flow from a large 12" fan from the side. ofcourse this requires you to transfer the filament to the new spool.
Awesome idea. I would love to see this developed into a reliable open source solution. I can see this becoming the standard in filament drying.
Perhaps it could be made physically smaller by coiling some filament inside the dryer, perhaps around a small wheel/which like thing.
I'm a former analytical instrumentation specialist... I've read few comments below.
My best bet would be the 1 feet diameter drum surrounded by few coil of filament. The drum is laying on the side with a bottom flange and a bearing (no motorization required). The drum is in a slightly larger drum box. You need proper fan + heater + thermocouple on Arduino or even simple mechanical thermal switch.
Filament goes into the drum, spinn few turn and get out. The more resident time you need, the more turn you need.
I've subscribed to see the next step!
thats what i do with my sunlu 4s, i'll dry it a few hours before i start printing & leave it on during printing. that way it is contantly preheating & drying the filiment while printing. it helps a lot with ABS i seemed to get less warping. this seems like a DIY version that would probably work pretty well. Keep it up Brother!
Yes please! I'm so totally going to make a can dryer now and have some fun on my own, though I doubt I'll have anywhere near as much expertise as you do. I'll be watching anxiously!
Very interesting! I'm interested in seeing V2 and V3! Don't stop working on this!
I wonder if you could take a piece of spiral-cut PTFE and run that down the length of the can, then use some sort of funnel to ensure most of the air gets directed into the PTFE spiral alongside the filament, so it's guided down the center but can breath outwards all around it. Also you might be interested in the membrane based electric dehumidifiers. "mdl series dehumidifier" is a good example you can search.
I found a simple solution. I use a large Rubbermaid type of container that holds all the filament. Maybe 10 spools. I fill up an old filament box with dessicant that I cut out lots of holes in. About once a week I throw the dessicant in the oven at just over boiling water temperature for about an hour. This keeps it around 24% in the box and I don't worry about it on the machine because it's not there long enough to absorb water. I don't leave it on the machine when not printing. This is simple and it works for me.
YES! Well Done!!! This entire past year, I have been shouting from the hilltops that filament drying is THE most important next place to put our collective energies! Sorting out this challenge for pro users and major hurdle for those new to the hobby before the constant epic print failures frustrate them away is vital for the future of 3d printing!!!
Try Activated Alumina granules next .... they can absorb Wayyyyyy more water and have a healthier lifespan than silica gel beads and the color changing dye used in them.
You do need a higher temp to drive off the moisture and "recharge" the granules but this ensures even better performance over a longer span of time .... depending on your humidity level.
I also recommend getting a dehumidifier for any closed room that you have your printers in.... so other off-the shelf technologies can easily be added right now for home users. I know dehumidifiers aren't cheap... but compared to destroying and wasting 2 rolls of filament in cost they are worth it..... get the biggest one you can so you don't have to empty it as often.... set a phone reminder as often as it takes to go empty it.
....and if you don't have dry boxes, or a dry room, enclose everything you can (entire printer and at the very least your filament, and add a cheap house fan for PLA cooling inside the enclosure ( so get/build the biggest one you can or get the smallest fan you can fit).....this will avoid the Creality X1 problems seen with excessive heat requiring the top be taken off to vent.
👍👍
That Creality dryer at 1:02 has a blower that blows the hot air directly over the spool opening and the air flows continuously around spool in the direction that the filament is wrapped (why it's shaped the way it is with the rounded lid). I have one and I have notice it does dry quicker then most options and once you use it as you print the deeper layers get dried as they get closer to the surface. I have been very happy with it so far.
Awesome idea and I want to see a version 2. I also have a few ideas on how to make it work more efficiently.
A quick search into this and there's two approaches that seem common. One is vacuum which was mentioned by other people. However, microwave drying is also apparently a good method to remove water from plastic since microwaves are tuned well to excite water molecules specifically. The main need of any method is to not introduce too much heat or you can degrade the plastic. We microwave plastic containers all the time, but we never really think about moisture content in the plastics. We also use many plastics in food use that don't really absorb any moisture. One would need to test if there is sufficient moisture in a filament spool to cause heat problems, aka melting or degradation of the plastic molecules where we might see a much weaker part.
Very interested in a fast inline dryer.
Great work on the video, loved it.
For those interested storing spools you've dried can easily be done with a home vacuum sealer and their plastic bags works really well - on the cheap you can also use a dry bag and a vacuum cleaner with some silica gel for storage. Remove the air you remove the moisture buildup and using a rare compound or color isn't going to be several day job drying it over and over.
There are a few ceramic heater cores that are in a cylinder shape for 3d printing. A few could be added in series. It's might be better because you could more precisely control temperature and be way closer to the filament, but you would need some electronics.
Something like the CHC kit 24v
You should do a Kickstarter for this! A low-cost in line filament drier would be a great solution for a lot of people
It may have been suggested, but when you come close to a working version, try to test how much air-flow is needed for optimal results. you may find that you need less than you needed and therefore smaller fans, less energy. GREAT work. You have my subscription!
Russ from Ocala, Florida, USA
Ill need to come up with a testing rig for all this things first
I am very interested to see this idea explored further. Inline filament drying seems like a no brainer compared with the other solutions on the market!
Yes please keep exploring that idea, I love it!
The physics of this problem are both interesting and complicated. The reason heating up the air works is because hot air has a higher capacity to hold moisture than cold air. This is why we frequently track humidity as relative humidity. I don't fully understand the physics, but the water molecules in the filament are more likely to sublimate into a material such as hot air when the relative humidity is low. I think your best results with this method would come from passing the air through a desiccant chamber on the cold side (relative humidity is higher and therefore the moisture is more likely to be pulled out by the silica) before it passes through the heater, which will further lower the relative humidity.
For the commercial inline dryers, they are probably using Zeolite, it’s used as a moisture absorber in industrial applications. It can be regenerated by heating it up, just like Silica Gel.
I came up with this idea as well about 3 weeks ago. The problem is: It is basically impossible to know the correct temperature while printing because of the varying speed of the filament itself. I found a guy on the Voron-Discord who did the same thing with two V6-Hotends slapped together, but he said he still dries it conventionally. I ended up engineering a new filament dryer because I am using an Industrial spool.
I guess, if you know the density of the material, the speed of the fillament being consumed and a fixed temperature, it might work but at this point, it might just be easier to dry it conventionally and that's the reason for the other thing costing 2100$ in total I guess. The smaller part probably just measures the filament speed and the larger part dries it.
This is a great idea! Can’t wait to see how you continue to develop it. Great job on the video.
Your English is phenomenal and your wife is amazing/cute for correcting you when she does hahaha. No worries I have to do the same for my partner as they're from Colombia.
This is definitely something I find interesting. I also had the same though about in line filament drying. They do it for pneumatic systems, especially for painting, so there likely is a way to achieve a similar result with filament.
Would love to see V2!! 2k+ seems pretty wild for something like that when your at home so love to see projects like this.
This is super interesting would love to see a follow up!
Someone probably already suggested this, but you could create a longer drying path by having it wrap around a free spinning spool inside the heated chamber. This can keep the chamber relatively small , and I’m imagining that the spool has a perforated center core that lets airflow reach the inside of the wrapped filament. I would think only a few wraps around could create decent amounts of exposed length so you don’t have overlapping layers on the spool like you would with a full spool.
Cool idea man, thank you for sharing.
Maybe a few reasons why this is not popular, first we don’t live in very humid environments. Secondly we keep it in a drybox and insert it into a hotend with a reverse bowden. Secondly our printers are additionally enclosed.
I have printed with almost all filaments you can think of, very hygroscopic filaments like TPU or water solubles and have never experienced such extreme stringing or water logging.
I like your drier. If you could make it so it circulated the air internally and contain desiccant it might work better still. Your device looks like it pulls fresh air from outside, but like you said that's a 60% humidity, which sounds less than ideal.
I have a $39 stack dehydrator with a cardboard outer sleeve. I do a few spools at a time then store them in sealed containers with moisture capture sachets. Works perfectly. I do like your novel approach. 👍🏻
One thing I'd suggest looking into, though it'd be more complex (and something I've been toying with as an idea) is figuring out how to integrate a silica desiccant regeneration cycle. The idea is to run cool air through silica, or some other desiccant, warm the air (dropping relative humidity in the process), blow that over/around the filament, then have a second system that somehow blows warm air over another patch of the desiccant. Something like a rotating plate of desiccant so that "dry" desiccant is exposed to the incoming air to dry the filament, and "wet" desiccant is constantly dried, with a slow motor to keep the whole system rotating.
This isn't a new idea; it's how plastic pellet driers work in injection molding on an industrial scale, though those often dry 200+ pounds of plastic continuously.
Yes, this makes so much more sense than drying the entire roll of filament... I never understood who dries an entire roll of filament for 8 hours. In the US, sure, electricity is basically free over there for some reason, but in most other countries it's just crazy to run an oven for 8 hours just to dry some filament... Would love to see a more "proper" V2.
I like the idea, it's the implementation and other things you mentioned. It worked, at 40mm/s but what happens at say 200mm/s or throw in an AMS. It would be easier to make the AMS a dryer, some other company did this, can't remember which one though.
This would solve most issues. I say most because most filament is dry out of the sealed bag but every once in a while, it's like it sat on a shelf for a week before it was sealed. Also, the most hydroscopic filements are all CF of GF or some sort of blend.
Since you can actually see the carbon fiber in them, it's easy to see how they leave more pockets of air for moisture to absorb into. The higher the amount of CF there is, the more hydroscopic it is. While filaments like TPU and ABS are hydroscopic, they don't come close to any CF blend for that reason. Most recommended 24 hours of drying at 80°C for PA-CF. TPU doesn't need those temperatures or time so that would be the real test.
Personally, I recently got the Polydryer by Polylute and it's great. It has a separate dock from the storage container, comes with reusable dececant, has a hydrometer in it, and already has mods on printables for various things like drying to boxes at once. The storage boxes, while not cheap (25 US),, are reasonably priced and have excellent build quality and comes with a hydrometer, reusable dececant and spiol. The box sits on the dock, hot air is blown up the back, the front port sucks it in which is where the dececant is. You can remove the box and seal it air tight with some plastic caps with rubber around them. The dececant and hydrometer stay in the box. Now you can simply store that box and put another one on it (extra storage boxes are sold separately). It's the nearest advantage to come to dryer boxes in a while which isn't saying much but still, more versatile than anything else on the market. The fact they are allowing community mods is also a plus.
With CF there pretty much has to be a dryer box. If you can get CF working on this for a reasonable cost then that's when more interest in this will happen.
Out of all the rolls of TPU I have bought there was one with major issues due to moisture, others printed fine for the first print then I did dry them and store them in ziplock bags with reusable dececant but once I get 3 or 4 of those dryer boxes, that's really all I need as I mostly print PLA which can't sit out for months.
You could probably use one can if you put a small spool inside the can and wrap the filament around it say 8 or 10 times. As the filament is drawn out, it'd always keep however many loops you have on the spool.
Very good! With a PTC heater and the same material I'm making a chamber heater. I Have no needs in instant dehydrator, for the moment (I have a chamber for the mounted filament, so when I become to design I chose the filament and meanwhile it dehydrate, and the for the others I use vacuum bags) but I have a suggestion for you: if you give a longer path to the filament with a pulley system (like the grouped ones for sails, for example), you can optimise your idea, but only for flexible, obviously, or if you use a flexible extensible exhaust tube, you can achieve larger radius curve so you can work with non flexible too. For silcagel: with a little precision scale (0.01g) you can make bags with fine mesh tissue (new silcagel pre-weighted), and with a microwave oven you (250g=max 450w) you can renew perfectly the silcagel in about 15 minutes, checking the bags original weight. Is important to cut the session in parts (I use 5/10min) and in between, you take off the bag and moving it, you leave evaporate the moisture and cool the silcagel, until you can hold it in your hand. Is important that the power of microwave never can make the bag so hot that you can't hod it in your hand for less than one/two seconds: going over the 100C° degrees can only degrade the silcagel.
Damn an inline dryer is a very cleaver idea!! Love it man
interesting idea sir.
there is one thing though, you introduce several new points of potential failure with this compared to the regular way of drying the filament when you get it instock. store it in a drybox for later use. this way you dont have to wait to print anything :) regular pp storage boxes with doorseal tape and a hygrometer. add some silica in a box with a fan and it is just a squeaky clean and cheap solution.
Great job, Sir, - and yes, I'm keen for more of your research and probably also some results from version 2. Thanks.
This is very interesting! i've been using silica for years, but kept them in a sealed box with the filaments in it. Never thought about "in-line" drying, i'll definitely try it
This is quite neat! I look forward to seeing more. 😊
One thing that I have learned about TPU is that slower print speeds will exacerbate moisture issues. I used to have really bad surface finish with my TPU prints until I upped the print speed significantly.
In my mind, I envision a single piece copper tube with a diameter close 2-3mm in which the filament goes through. A 3d printed adaptar is printed at the end that makes a Y, one side of the Y allows the filament to go outside to the printer, the other end goes to a much bigger opening that connects to a 120mm fan that creates a vacuum.
.
On the other side of the tube, you have the same Y: Just that instead of a connection to a fan, you're breathing dry air from a box full of silica gel (Similar to the first commercial product that you showed.
.
Then, you can wrap heating elements around the copper tube (Like the ones used in motorcycle handle grips) and finally and insulation around all of that that would keep the copper tube heated at a reasonable 60-70 degrees without too much use of energy.
This feels like such a cool project!
Store your filaments in a pumped out vacuum bag with indicator silica. Don't leave filament in an environment with a RH% higher than you want your filament to absorb. There are many mods that can be done to the basic Sunlu dryer to recirculate the heated through indicator silica. If your're printing in an environment with a high RH%, stick the filament in a filament dryer while your printing, or at least on a roller in a food box containing indicator silica with the outlet hole just large enough for the filament to pass through, and keep the device you're using close to the 3D printer while you print.
That is amazing. It would be awesome to see version 2. Please keep your great ideas coming.
Try using the drying system or even just the desiccant cartridge from a tractor/trailer. It has to constantly dry the compressed air used for braking , actuators, etc . The parts can be sourced for cheap.
Or just store your filament spools in white rice. Thanks for the video!!!
I just added a fan to my old filament dryer and my new filament dryer has a built-in fan. Running this removes the moisture fast. It doesn't matter that only outer spool layers are dried because the moisture is removed so quickly using the fan and heat together. I wish the BambuLab AMS also had a dryer built in, besides the desiccant.
I have been doing this for years. But I use a very large sealed chamber I place the whole spool inside. I use a fan that heat the chamber and produces violent air flow. The filament is dried as it comes off the spool and the moisture is deposited into aluminum oxide cans that I recharge each time I place a new spool in the chamber.
Hey there 3d printing ideas, I work with metal filaments and run my own small 3d printing/ design engineering consulting gig. I have worked with metal filaments and they have a variety of problems of their own and I would like to share what I have done to improve my filaments. I personally would increase the heat and airflow. in an industrial injection molding operation a large part of it is drying the pellets before feeding them into the injection tube. The drying time varies based on a number of variables. So in your case you'd have to account for water accumulation in the filament, air humidity, the speed of the filament going through the nozzle and as well as what ways you can pull the moisture out of the filament. Additionally, how wet the filament can be to print "normally". Nylon gets stronger after it gets wet, but you can't print with it when its wet. If you're someone that wants to speed print using "wet" filament, you'd have to have a vacuum chamber set up to pull the moisture out of the air then draw it out of the filament towards the outer edges of the filament itself, then you can push it through a filament dryer with desiccant. The name of the game is to make the air surrounding the filament drier than the filament itself that way the water pulls out. Its simple physics after that, the hotter it is the faster the moisture will dry, the more air flow you have the easier it will be to pull it from the filament. add a desiccant and now you have even more things pulling moisture out. The vacuum chamber is really overkill for basic and noncritical components but it ensures you get the most moisture out of your filament as fast as possible prior to printing and then you can precondition it by warming it like you've done. You just have to extend the warming time to match conditions where you're printing. so a "wetter" filament will require a longer heating chamber. A more humid climate will require a longer heating chamber. Prewarmed filament will also print easier as the hotend doesn't have to bring a filament from room temp to melting, rather 100c to 220c for pla for example. Hope that gives you some tips and tricks.
I would also suggest utilizing nichrome or kenthal, it allows you to control the temperature more and you can wrap it around a glass tube that you can run the filament through alternating the hot portion with an exposed air portion. In my tests it worked good but that was my v1 I feel like putting it close to the filament AND adding airflow in the can with the desiccant would be interesting to see. I just leave my desiccant in the box.
If you decide you want to really knock this up the S tier, I would recommend putting your filament on a spool and holder into a vacuum chamber and feed it through a series of silicone wipes. Now you've got molecularly-dry filament. Getting a 12CFM pump is pretty cheap now days and will empty a large chamber in minutes.
Contra flow drying, increasing the length of the drying path and insulating "the cans" so the air stays hot. Also the diameter of the tube does hot have to be the diameter of the can. A diameter of about 25mm should be adequate. A recuperative heat exchnger could also be used to recover energy and elevate the drying temperature.