Solid-State Dehumidifier: The Ultimate Filament Dry Box 

I used to work on industry 3D printer dryboxes, and wasn't aware of these solid state dehumidifiers. Incredible technology, I wish i knew about this a few years ago. This was a good treatment of the filament moisture load problem. I used to perform calculations and experiments to determine diffusion times of moisture into and out of filaments. It's a multi scale problem, since you have diffusion at the filament level (small dia cylinder, 1D radial problem), then you also have the fact that the spools have large thicknesses, so this becomes a more complicated problem. There is also the diffusion and convection/advection through any gaps, through the filament tube, the 1d transport along that, the forcing from variations in pressure over a given day which effectively pump on and pressurize the volume with some frequency, etc. I would recommend doing calculations for absolute humidity, instead of relative humidity, to understand the desiccant moisture load of a given volume. It is straightforward to calculate how much moisture a dessicant can take up, and then relate this to how many times you can open the door to the drybox (because immediately, the air is exchanged), and evaluate how long desiccant should last at steady state. I did so much math and so many experiments... Molecular sieve is the best... 4A is incredible stuff. Also, how long a spool is in the air matters a lot for how much moisture it picks up. If you know the diffusion rate of water vapor in nylon for example, you can estimate how deep and how much water becomes trapped. Diffusion occurs radially inward and outward, because the spool starts dry, so the gradient drops at the surface when you place in dry environment, but some of the water still works its way inward since there is still low concentration there. If you catch it fast enough though, moisture can be pulled out very quickly. Happy to share more details if needed, but it quickly rises to needing to build a numerical diffusion model, which isn't too bad if you have an engineer or physicist friend. There are analytical solutions for special cases (the mathematics of diffusion, j. crank...its a complicated topic), but a numerical solution can help accommodate a range of forcing functions like opening and closing drybox doors, having varying environmental temperature and pressure, and having some steady leak rates to estimate how long dessicant should last.

I see CNC Kitchen, I WATCH. THIS is the content that RU-vid needs. Experiments that push things forward = GOLD. Other content creators would do well to take note of this.

I've been using Rosahl's M-3J1R membrane in my AMS for about a year. I placed it in the back of the base though, and put a 20mm fan drawing air through a hole where the desiccant would normally sit. Since they don't draw much power, I am just tapping into the power feed for the AMS and stepping down the voltage. It keeps the humidity in the AMS at 10-15% RHM, but does take a few days to come back down if I put rolls of hygroscopic filament that have been sitting out into the AMS.

Been watching Big Clive for what must be over a decade now. His channel has given me so many ideas when working with electronics. Only just noticed he has only recently made the 1m subscriber milestone after all this time. His channel deserves more love, he rocks.

How does it compare to 3A molecular sieve? I can pull moisture below 1%, with no heat, using zeolite. I look forward to that video

I had the exact same thought when I watched that bigclive video, thank you for fulfilling that desire to test it!

Missed the opportunity to mention your deburring tools when making the cutout. What I would do differently is mount the thing at the top of the lid. Water vapour is lighter than air and is the most concentrated at the top of the container.

Convection for the win. An interesting technology. My level of failed prints dropped enormously by storing filament in air tight containers, dry before use, then place the filament in a dry box while printing. Anything you can do to protect the material from moisture is a win!

Maybe combining a bag of desiccants and the device would combine the best of both worlds, it can absorb a lot of moisture in a short time, and over longer timespans dry the desiccants out again..

You might be able to clean up the plots by plotting the dew point or moisture content of the air (mg/m3) instead of RH%. This isn't perfect since the temperature shifts the equilibrium a bit with the polymers but better captures the actual moisture in the air.

❤ @12:46 😂😂 nice view

Relative humidity is a function of temperature and absolute humidity as hotter air can hold more moisture than cold. The absolute humidity was likely dropping smoothly in the full dry box test without fluctuations.

I almost never have issues with PLA or PETG in my area. I have a 12x12 shed with electricity, insulation, and drywall where all of my printers except for my Prusa (it's quiet enough to be inside) live. I am in a high desert, so the maximum humidity we ever really see is ~50% and indoors it is much lower. There are months where you have to use a humidifier in your bedroom while you sleep to avoid nosebleeds. I genuinely have empathy for how much people seem to struggle with keeping their filament dry. I've only had issues with it ONCE. Last year we had a VERY wet winter, and one of the windows in the shed had blown a seal and I didn't notice it. My PETG was sputtering and popping. I just dealt with the reduced quality and when summer came, it dried back out.

I'm glad to see you revisit this amazing technology! Some thoughts: 1. I noticed the potential for oxygen buildup right away. Fortunately, there are *many* materials that work as oxygen absorbers that are relatively cheap. For larger scales, 'oxygen concentrators' could be used to capture the extra. 2. For more utility, the SSDH could be built into a movable module that mounts to the _outside_ of a container and alerts you when the humidity has dropped sufficiently. At that point, you can seal the container and move the SSDH to another one. That way, you could prep multiple containers for long-term storage/printing directly from the container/etc. without needing to but a separate SSDH for each container. 3. I can see this being *INCREDIBLY* helpful and economical for those who--for example--use 5kg/10kg spools of things like nylon or large bins of pellets. (Post active drying, of course!) 4. I can now see the attractiveness of metal spools, as they _can't_ absorb moisture. 5. As high vacuum can help pull out moisture even faster, can you experiment with using a microcontroller with an induction burner to heat a vacuum vessel? (I'm assuming that the seal will be made of silicone.) Cheers!

Love your content. You don't just show off useless prints you made. Really love the effort in all your videos.

I've had the equilibrium debate a few times and people insist that a dry box will not dry filament, only keep it dry. My argument has been that the whole system must, according to Newton, attempt to reach equilibrium and therefore if you have patience, dessicant in a dry box MUST dry the filament too.

Connecting this to vacuum pump might be good idea. Getting even near perfect vacuum incredibly hard but can draw most moisture out without heating and can work faster. Getting partial vacuum but that air is very dry could speed up whole process compared to just dry air

I am very much looking forward to the dessicant comparison video. I did a lot of research when i was looking for a good way to store my large collection. I ended up settling on activated alumina. I believe I made that decision based on price, how low it can bring humidity, and ease of recharging. I can't wait to see the results of your testing!

After watching all sort of channels, the only one I still watch regularly is yours. The way you explain everything, the overall quality and ideas you come up with are just top-notch!

Every time I see a new Cnc kitchen vid I get excited

Nice video, one thing to add: Desicant is well capable of archiving humidities of 0-1%, but only when regenerated. In order to regenerate the material heat in the oven for 3h at 130 - 150°C. It is necessary to shake the desicant several times during regeneration or the regeneration will fail. For drying polymers you need to keep the humidity below 5%. This requires you to regenerate the desicant a lot. Usually you place your parts directly into the drybox from the packaging. Also: Any cardboard packaging has a hillariously high water absorption.

Interesting 👍😊 Be aware: Moist air is lighter (‼️) than dry air and therefore rises to the top. Oxygen is heavier than dry air and sinks to the bottom. So together… without continuous movement of the air inside the box, the placement of dehumidifier may be very important for the outcome.

This is next level . Exactly what we need in the community . Thank you !

I think the real place for this is in addition to desiccant. The desiccant gets it fairly dry, and the membrane gets it even dryer, and also recharges the desiccant over time. You can also add a door to the outside of the housing to prevent moisture getting in when it's off.

Well dang, I was getting excited, even though I knew it would be a fat price tag, but that is a bit much. Still, if this could work in a large filament storage bin I could see it being worth picking up a single unit for the purpose of storage and just pulling out my filament when needed. Great video. Lovely testing.

I would love to know if oven “reactivated” silica performs as well as fresh new silica or if it slowly gets worse.

Today I finished building a PID controlled dehumidifier using an EliteGourmet food dehydrator and an SSR. I haven't seen ANYONE on the internet with even a remotely similar idea...until this was uploaded. I can't wait to watch this to see if Stefan has thought of something I havent!

Of all the content on RU-vid, I am most excited to see a new video from this channel. I love the experiments and the pathfinding you do for the entire 3D printing community.

I am glad you addressed the increased oxygen concentration in the AMS. Higher O2 concentration and organic polymers is not a good combination. Did you try putting an O2 sensor into the AMS?

Im gonna be honest, i dont have a 3d printer or will ever own one, but these videos are just entertaining

Extremely thorough testing, thank you for your hard work sir

Yeeees!!!! I bought my husband a bambulab printer. I was looking for a dehumidifier as a following gift, but I could only find ones that were passive(silica gel) or didn't seem good enough for their cost. I was considering modifying a food dehumidifier. I'm excited for this video!!!

Hi Stefan, maybe you can try using a fan to circulate the air so I dry my silica and filaments by closing the hot bed of my printer and it greatly speeds up the drying times. We are next to the Rio de la Plata and the humidity is almost 90% so it is something that we have to do often the humidity kills us a big hug from Temperley, Buenos Aires, Argentina

Sylica is reusable, you need put it in oven for some time.

Worth every penny. I really don't mind paying the early adopter tax, as long as it pays off eventually.

Oh man you have the greatest ideas and make the coolest experiments - very thorough and educating! Never stop being you! This is a very interesting project... and I just bought an extra dehumidifier so now I have 2 - but I think I want to replicate what you have done here, for a more effective dry box for storage.

How can I join a class action lawsuit against square space for false advertising? I cant be the only one to find it extremely difficult, expensive, and frustrating.

Shoutouts to BigClive! That video’s where I learned about these devices as well!

I live in Colorado so I do not care one bit about drying my filament, but still love to watch your videos. The humidity here routinely is in the 5% range 😂. And we have a whole house humidifier rather than a dehumidifier.

Love your content. You're contributions to this field and this community are huge and you a true pioneer in 3d printing and manufacturing additives

I was about to make this one years ago equally when I saw BigClive... but when I looked around for the part found it expensive for my budget and I also lacked the required resources to make this properly with a transparent container... glad to see that you've made one even if right now most of the scene moved from PLA moisture problems...

Well done, Stefan!

iv never heard of such a device. It seems like quite a promising method for this kind of thing.

This is great information. I hope to build some filament storage modules in the future. I have already designed a filament holder to put in my planner boxes. It holder has a space in front to place test prints and a place for a information tag.

I've honestly thought about doing this for years, but when these first came out you couldnt source them and the one i read about was tiny and designed for like trail cameras or something small. Time to revisit this!

I'm so glad you made this video. I live n a Coastal Community where the humidity 60 - 100% humidity. It's a serious issue and for me this solution seems to be the only solution despite the initial cost. Constantly heating up the filament degrades the filament no question. Silica beads are a huge pain in the a$s to manage plus somewhat expensive. To me, despite the initial cost, this is the best solution outside of a professional drying cabinet. So now I'm weighing the electricity cost of a professional cabinet or using something like this with the filament changer.nope I've convinced myself. You raise a great question about durability though especially with the topic of microscopic plastics on the rise. Since there is movement in the filament changer there will be tiny bits of debris. If you were to add a fan to the panel it would need a pre filter to prevent premature degradation.

Amazing work, mate!

when you release a video it's always super interesting, amazing work as always.

For about 6 years, I've been using an Eva-dry E-333 Renewable dehumidifier in my filament storage box and it works great. For drying filament like nylon and PETG (which I very rarely need to do), i use a food dehydrator.

Very interesting video and perhaps something to watch for in the future for this kind of drying. Currently i have been using the Slice Engineering Desiccant and it works very well.

i saw the same Big Clive video and thought about doing this, i just wasn't able to find the solid state dehumidifiers. kept looking for the wrong thing, like electric dehumidifier. Great job and thanks for everything you do for the community.

As always solid data and quick presentation with positive attitude ... Thanks for engineering level of content ....

Knew exactly how this was going to end, since I loved that bigclive video. It gave me some project ideas as well, only for them to be immediately buried as soon as I saw the price.

Awesome research, thanks! 👍

Excellent idea and video! Shared it on the Bambu Lab forum.

This is a great work Yet, it’s the best I’ve seen

big clive to the rescue!

What is missing is a test with filament from the dehydrator and then print from a closed box with silca. I have had very good experiences with it.

Very nice! Let's hope those dehumidifiers will drop in price over time!

Even if not for drying filament, it can have other applications. So good to know about this cool product. 👍

I love your videos so much. You discover so many interesting and sometimes useful things. This looks really cool. You should release a dry box product and bring per unit prices down.

Seriously, this is some good stuff.

Didn't know these existed, great to know! I think the power usage might relagate these to on-printer active use dry boxes.

I mainly print PLA, PETG and a bit of TPU. I have a heating drying box, but I really rarely have to use it and get perfect prints off of my Prusa MK4 and MK3S+ 😊

Very cool. I would still just go with a solution like dry tote: recharging those packs take 3 minutes in a microwave at 80% duty. Assuming a 1200W microwave, it takes as much power as running the solid state dehumidifier for 15 hours. I only need to charge the bags once a month, but normally charge once a week for reliability. That is 10% of the power needed and the bags are relatively easy to replace.

THank you for finding this. I am going to build a dry box using them.

Very nice work Stephan.

Now I want one of these for my repbox.

Great video! To extend the dry time, you could cover just the dehumidifier with an airtight lid when you turn it off. Might prevent it from reabsorbing moisture.

Interesting - I think it's cool. I didn't know about these devices, and hope we will see boxes with them come to market.

Yes I did find this video interesting. Thanks for sharing.

Thanks for sharing Stefan! I’m always interested in emerging technologies how they can solve everyday problems. It would be cool to see how we can use old technologies made more efficient/effective using modern methods implemented in an atypical way to improve all various facets of 3d printing

For the money of those dehumidifers you could just buy and constantly use normal dryers like the Sunlu S4. But this video clarifies a lot and is extremely helpful! Thank you very much!

finaly! You are the king of 3D printerz! Danke schön ❤ Ich liebe Sie 😂

Great Video! I saw that same Big Clive video and wondered the same thing, but yeah, could never get past the price of those things. Really great execution on the build and good to see that it could actually work! Maybe next try a Peltier to make a solid state condenser that can collect and drain to outside the box? Would be interesting to see how this cheaper solid state alternative stacks up.

I made a box in which I keep an incandescent light bulb on (when needed) with a slotted shelf on the bottom on top of a water container. My desiccant media of choice is Calcium Chloride (CaCl2), after it becomes a brine solution, I just boil it off. I believe you can also do the same with silica gel, putting it inside an oven in a baking tray, but if I'm not mistaken, the amount of water per volume or weight that CaCl2 can absorb is higher than that of Silica Gel. And honestly, I can't remember the last time I turned the lightbulb on, I think it isn't needed.

Great video. Cant wait to see the next on on Desicant!

10% RH is an average winter day here in Colorado. Many hygrometers bottom out at 15% lol. I actually run a steam humidifier to get humidity up to 30-35% in my house so we don’t dry out, spark and crack through the winter! Helped my kids’ eczema when they were younger too.

Excellent overview and experiments. If I were implementing something like this, I would probably run a small amount of forced convection on both sides of the membrane.

btw, the plastic your boxes are made of is moisture-permeable too. So if you want ultimate low humidity, use metal or glass (covering a plastic box with aluminium tape seems to help quite a bit)

Super video, danke für dein Research. Cheers, Heiko

Holy Cow! Plumbers Crack at 12:47. That was hilarious! 😂😂🤣🤣 Great video, thank you Stefan!

I would like to see a room dehumidifier adapted to dry a closed system of spool storage with only water exiting and air entering at the same volume.

YESSS this is so perfect! i hope they implement this in V2 of the AMS eventually. it is the future!

Talk about timing! I've been working on a way to directly measure relative moisture content in filament. The big drawback is that its readings can only be compared between two samples of a given filament: it can say "sample A has 0.1% more moisture than sample B," but it always needs a per-spool baseline. The design is surprisingly simple, and is sensitive to fractional percentage changes in moisture content using under 3m of filament. The main testing routine has been cutting samples, drying them using the heated bed technique and/or soaking them in water over time. The sensor is verifying out nicely, but the results are basically "Common filaments just don't get that wet." This mostly matches what Thomas Sandlanderer found in his video: storage doesn't matter much for the most common materials. Nylon is great at absorbing moisture, so it's a good test subject, but only 2 out of 12 spools of PLA seem like they absorb meaningful amounts of moisture. PETG picks up a bit, while ASA has stayed surprisingly dry in my garage. I'm still figuring out if this is worth developing into a product, but once that's sorted out, I'll be very happy to share the details. (If it's a product, there are a few patents I should get. But once those patents are out there, anyone who does electronics can build their own in an afternoon for ~$25. And if it isn't a product, I'll release all my documentation on the web.)

You got me thinking of TEC based dehumidification.

Wieder ein super interessantes Video. 👍

I use 100% silica cat litter as my dry box desiccant, it is dirt cheap and it works OK I think. Hopefully it is in your upcoming video as well :)

Electrolysis also creates ozone (O3) when used in the open air like these work. The devices are intriguing, but may need venting for long term use depending upon your circumstances. Mixed with a heating and ventillation system, they would probably work very well, but the price would need to significantly drop to around 10% of their current price point.

Excellent stuff.

Great vid!

Really cool technology, hopefully this can trickle through the industry.

I've been designing my own dehumidifier for quite a while now and I think this could definitely be an interesting technique to look at. Thanks for sharing Stefan!

I just frankensteined a dehydrator, 60 litre sealable storage tub and added a couple of closable vents, it turns on for around 2 hours a day and all my most hygroscopic filament is consistently at between 8-12% humidity in the box. I haven't noticed any degradation in the filament storing it this way, I have 2 year old petg and nylon in there that I used a couple of days ago that printed beautifully. I have measured the temp of the box and at different levels it reached different max temps PETG goes in the 55c area and nylon goes in the hotspot which reaches just under 70c. I also have around 500g of desiccant in the box which keeps the moisture under control between cycles.

These things are neat! I absolutly can see lots of applications. If O2 build up is a concern maybe add some iron fillings! Its a very visual and easy to read oxygen getter! Although for the price I think the best choice for hobbiest will remain a well sealed steel box and a few packets of the forbiddin snack labeled "Do Not Eat"

I just got my P1S and AMS today, after it being backordered. Now I know what I want to do when I get a second AMS.... :D

Hahahaha. I was just looking for a dryer today. Yours is so much cooler! 😃

Really interesting ide, while i have never worked with anything that is particularly hygroscopic, i always just store my pla /petg in the original bag with the original silica. and have never had an issue, i even added some hydrometers in them and they kept their humidity after a year of just sitting in storage. (after having been used for like a week.)

When I ran a lab with several printers in it, I used to keep at least the nylon spools in vacuum garment bags with small silica packs; they had so little air inside and they worked well as humidity barriers so the packs never needed to be replaced while I was there (just under three years). This is even better because I could run them off energy harvesting of some type (the noise you hear in your office light ballasts is collectable 🙂) so they could be self contained and portable without having anything extra inside with the materials.

Nice info. I'm curious if you just toss your silica, or if you reuse it? I'd love to see a comparative on reusable silica, to see if it maintains its properties after multiple heat cycles.