Don't forget to share this video on Facebook, Reddit, Twitter and other social media! *Timestamps for the impatient* Introduction: 1:20 Why dry filament: 2:40 Damaging filament by heat: 4:58 Vacuum drying: 5:55 Moisture pickup in open air: 7:30 Food dehydrator drying: 8:00 Storing filament in vacuum + desiccant: 9:55 Do spools dry equally: 10:25 Drying with desiccant: 12:00 Drying in a circulation oven: 12:45 Summary/TLDW: 13:57
Damn the summary goes straight to an ad. I'd appreciate it if you'd have a summary with data points that are actually useful. I don't mind the ad, but not having the data there is frustrating. (Yes, I get it, free thing and you need to make money.) Today I happen to be in a hurry, and I know your videos take about ten minutes to get to a conclusion.
You should heat up the material before you put it in the vacuum chamber to get the best result. Look at this video. m.ruclips.net/video/lnVjxuCc4Wc/видео.html
Being a HVAC tech with years of experience dealing with vacuum pumps and pulling water out of HVAC systems installers forgot to keep capped during installation, the fact that a crappy single stage vacuum pump is being used isn't helping. A 2 stage vacuum pump that can be run overnight without overheating is pretty much required if you want to get a significant amount of moisture out of a enclosed system, as is the ability to maintain high vacuum for the duration. Also its generally a good idea to have some kind of water trap, either a cold finger or a tube packed with a chemical dessicant, or be ready to change your compressor oil often as the water extracted from the vacuum chamber will end up in the pump oil and damage the pump over time if the water isn't removed from the oil. If done right, with proper equipment, a vacuum drying setup should be able to drop the moisture content in the filament way lower than a food desiccator can in the same amount of time in most climates that aren't a done dry desert. I also understand that this testing is being done "on a budget" but its kind of misleading to say the dehydrator is simply better when half the issue here is that the vacuum setup is just too cheap to do its job well.
I use 2 stage rotary vane and turbo pumps regularly in my lab, and drying solids is always done under red pressure rather than dessicants, it's considerably faster and produces much dryer material. A dessicator might take days to achieve what just an hour of a few micro bar vacuum manages. Although as mentioned, overnight drying on a vac line gives best results, especially for bulk materials.
I demand a recount! I've had way better results with 12 hours of vacuum pump running than 24 hours in a food dehydrator. I always run mine with indicating desiccant visible through the lid so I can clearly see when it is "ready" to come out. My experience was with an even worse looking cheap pump that I put good oil into. It gets hot, but seems to work. I just change the oil once it starts looking milky.
@@TrevJonez As frager56 mentioned above, a water trap will save you a lot of oil changes: use a long clear tube full of indicator desiccant, and you can watch the color change, then you know when to stop and bake the desiccant.
I fully agree with your assessment but I think Stefan was catering to the 3D printing community that might not be able to afford a good two stage vacuum pump for merely drying out their filament. I also concur with having a desiccant or other means to absorb the shed moisture and keep it from getting sucked back into the filament once the vacuum is released.
I'm sure you realize this already but you have done the 3D printing community such a service with all of your videos providing real test data and results over all these years. Just amazing!
I have the same food dehydrator, got it at Aldi. I bypassed the factory temp controls and added a PID controller and a temp sensor. it can go higher and is very stable in comparison to the factory temp control which was swinging around 10-15 degrees..
What are the health concerns over silica with indicator dyes? Edit: For those wondering, the indicator dye used in silica beads will either be cobalt chloride (blue) or methyl violet (orange). Both of these dyes are toxic and carcinogenic, though methyl violet is regarded as less dangerous than cobalt chloride. Using small amounts of methyl violet indicator in sealed packets seems to be considered food-safe, but if the indicator is loose like how you buy it in bulk, you could inhale or ingest dust from the beads, thus consuming trace amounts of the dye.
@@tildey6661 you are correct, but I don't believe that is what he is saying. He states his opinion that the risk is minimal or non existant unless you have some excessive physical contact, such as eating it, with the compound in its salt form. The last part of his statement indicates that it is indeed toxic "as long as you do not eat a lot of it". What it being in a salt form, or any toxic compound being a salt for that matter, has to do with the risk it imposes I'm not really sure. Perhaps he is insinuating that under normal circumstances it is not a volatile compound and your not going to encounter it in the air, i.e. not being a volatile substance, not going to be in the air, not going to sublime, turn into a gas, no vapors, thus your not likely to come into contact with it under normal circumstances. That theory is a long shot and seriously flawed however, i'm not going to look into the vapor pressure and vapor temperature to begin to decipher what his logic was and its based on my assumption that he has limited education and training in chemistry. I could see someone without a strong understanding of chemistry thinking that there is no exposure to the compound unless you physically touch it.
Nobody is asking about what looks like a E3D toolchanger printer next to the vacuum chamber at 2:23? I'm eager to get a review, it looks so promising! As usual, great content Stefan!
Great Video, and a very great job with info for all of us 3D printer people! Really enjoy watching your videos. I see how the vacuum pump would work and do a great job. I am new to 3D printing. My wife got me my 3D printer for my Birthday last November. Here in Tennessee where I live, we are blessed with many rivers and man made lakes! I live in Middle Tennessee almost at the geographic center of the state! Humidity is a major problem here. My Grand Mothers, both of them, taught me to put rice in the salt shacker to absorb the moisture so the salt would flow. Rice is very hygroscopic (absorbs water from the air) and will pull it away from just about anything. So I put rice in a 3 inch by 4 inch mesh bag (you can buy online for $7.00 or so for 100) and put the rice bag and filament in a one gallon zip lock freezer bag. Zip it shut and the rice will absorb any moisture left in the sealed bag. Put one bag of rice in with a spool that has one half or less filament. I put two bags of rice in the gallon zip lock bags that have filament >= one half a spool. It cost almost nothing, and it makes working with the spools of filament very easy. Just open the bag and switch filaments. When you have a filament dryer, dehydrator, or a vacuum pump, you can only work with one spool at a time. There is nothing wrong with that, but drying them all with rice can be done all at once! I know I have only been a 3D printing guy for about 9 months now. At this moment I have 29 spools in bags with rice drying 😆!!! Again, thank you so very much for all that you do!!!!
You cannot dry stuff (completely) in a static vacuum. Unless you leave the pump running continuously the moisture will reach an equilibrium point within the chamber. When you leave the pump running constantly no equilibrium can be reached as the moisture is constantly being removed. To protect the oil in the pumps in labs you generally use two traps in series. The first bubbles the output of the chamber through sulfuric acid, which takes all the moisture out. The second trap bubbles the output from the sulfuric trap through mineral oil, which traps any fumes coming out of the first trap. Generally these two fluids only need to be changed once a year. Same with the pump oil. You will be amazed at the amount of water that will come off with only limited heating. Generally, proper vacuum drying will move about 1-1/2 times the moisture that air (hot) drying will.
@@termivan Generally you just make a line on the container (preferably clear) at the height of the fresh sulphuric. Once it doubles in volume it is time to change it. Usually that works out for us to be about once a year, but that will depend on how much you use it.
@@bradquinn4161 excellent, I see that my job (assembly shop) has about a 500g chamber of dessicant in line to the industrial pump, I am considering all these options to help prevent oil contamination in the future, where might I be able to shop for such accessories?
@@goguyted We used to use two of the used (and clean) acid bottles(2 litre?) with some of the red vacuum sealer. The bottles are shatter proof due to shipping regulations. If you have a local lab they MAY just give you a couple.
Many years ago I used to manage a thermoplastic molding workshop where we heated plastic sheets enough to be formed on a mold using vacuum. Some plastics are hygroscopic meaning they absorb moisture such as PETG and polycarbonate (lexan) being the worst. In order to mold PETG or polycarbonate sheets thicker than
You could consider this post a "peer review" of your findings. TLDR: I found results consistent with your findings. I've been running theses "drying" tests for over 6 months now and I don't think drying water is the issue. I've used ABS. ABS+, PETG, PLA and PLA+ in my tests, sorry, don't have the "exotic" filaments you have. I have not seen even ABS absorb much water when submerged for 7 days, that was above a .2% change, (2g of weight using a 1kg spool.) What I think needs to be tested, is what chemically happens to filament as it ages, and how applying heat to the filament changes the filament. My thinking is that heating the filament causes some gassing, chemicals in the filament to be released due to decomposition of the filament. Since I don't have the ability to sample the gases, this will only continue to be just a "guess". - My method of testing is to weigh the filament as soon as it is unsealed, dry the filament at 135F for 5:30 hours, re-weigh, submerge the filament for 7 days 100% under water, remove the filament from the water, spin dry the filament in open air for a few minutes to remove any extra water that may be trapped in the spool, weigh the filament, re-dry the filament at 135F for 5:30 hours, re-weight the filament, and record my findings. If all I see is a .2% change in the filament, then I'd think water is not the issue. Does anyone else have any test data to share? So far this all has been subjective data about drying and there is still lots of research to be made - WITH this said, I still recommend drying where there is a filament issue, what ever heating the filament does, it seems to improve print quality. I'd like to see more of these types of tests, however, I'm not sure if you can get the test equipment you need, or if there' s a "market" from your watchers. My guess is not everyone else is a nerd. Thanks again for a great video.
There is one major problem with your test method... liquid water has a very strong cohesive force and does not like plastic in the first place... it is not as likely to penetrate as the individual molecules of water in a high humidity environment. Your water was also likely to be room temperature (or lower) which will also affects absorption. As an example, drop ramen in cold water and hot water and see the difference in time it takes for the ramen to get "soggy". Humidity is measured while water is in its gaseous state, *Not* when it has condensed to a liquid... the temperature that it condenses is the dew point. A far more appropriate test would be to "steam" your filament for an hour and measure the weight change, then try drying it.
@@edwardpaulsen1074 Ed, Thanks for the reply, I was hoping someone would reply to "check" my work, however, you have only brought more questions - I like that! Since my numbers are very close to the ones posted in the video, how do you think steaming would be a "real world" test? These tests are to determine the normal absorption of water, during normal use, not try and force as much water into the filament as you can. To go back to your Ramen example, we are trying to measure water absorption with it sitting on the counter, not being boiled in water - We leave our filament out in room temperature air, not in a hot sauna. I'm sure I could be missing your point. Thanks.
meance perhaps he is saying to get maximum water absorption, use steam. It would be interesting to see how different the values would be. I live in the southern US, so high heat and humidity are a factor here.
@@smportis Think about what you say - Steam is normally created at 218 F, you your thinking heating filament PAST the temperature that most filament is extruded at is what he's saying? An open container of water will normalize at the temperature the humidity will be in the area - room temperature is 80F, the humidity is 80F. Any other tests are not valid because you are not simulating the normal environment filament will be exposed to. - Even the water submersion is an valid test, unless you store your filament in the rain.
@@smportis Soory it's early in my day, so I should have used C instead of Water boils at 100C/218F, and filament is extruded ate 212C, TWICE what water boils at.
Great video with a lot of detail. I usually store my filament rolls (i use mostly PLA) in the plastic bags they came with. I add a perforated transparent film-container filled with dessicant to each bag and close them with ordinary bag clamps. Benefits are: less potentially humid air in the bags compared to plastic boxes), doesn't require much space, is relatively air-tight, costs next to nothing.
Thanks for posting this. I dry wood frequently in a vacuum kiln I made. The two key factors are the amount of vacuum, and heat but not too much. Your little vacuum pump should be able to deliver adequate vacuum. I automate my kiln with a simple mechanical timer turning the pump in every 90 minutes for about twenty minutes. I also put a one way valve in-line because the pump doesn’t maintain vacuum by itself very well when turned off. I use heat mats that don’t get warmer than 40C. When the temp in the material reaches about 36C the water really starts to come out. Because the heat transfer is by conduction the maximum thickness for best results is about 100mm. I can dry wet wood at 55%+ moisture to nearly 0% in 3 days. This is 120 year old technology for wood. A German friend has found a 25W motorcycle seat heater to use in a vacuum chamber similar to yours. Sandwiching the plastic rolls between heating pads will give you the best success and speed. You could also try just putting it in a sunny window. However, considering all the messing around, in the end your cheap desiccant in a decently closed plastic container seems to me to be the best option for most plastics.
Do you think heating the filament in a food dehydrator to say 40-50c then stick it in the vacume the way to do it then? Mabye I can screw in one of those water heater elements in the center.
@@warlockd I imagine you’d get a bit of a head start but you still need to keep the stuff warm, about 40C I’d guess, over a day or so under 26.5mmHg vacuum at least - more if you are closer to sea level.
One concern I have is people potentially not understanding the amount of desiccant required to actually reduce moisture. Despite your having shown and mentioned it, IMHO a lot of people just won't pick this up. I'd love to see a comparison data showing smaller quantities of desiccant to help absolutely clarify the "sufficient" quantity.
What if the filament attracts moisture more readily than the desiccant? Wet desiccant will hydrate the filament. I will vacuum bag with desiccant after the filament has been dried overnight.
@@jamesgreen4965We actually do use a lot of dessicants aside from silica gel (zeolites, activated alumina, molecular sieves, calcium chloride, sulfuric acid, clay...) you can even find polymer dessicants intended for injection molding (which could theoretically be turned into filament). Which dessicant is best depends on the use case. Sometimes loose beads, dust, or sulfuric acid can cause issues. Sometimes you need an indicator for replacement. Sometimes you need extremely low humidity (some types of silica gel aren't particularly good at low humidity). Sometimes you want a dessicant wheel which regenerates (releases water) very easily. Sometimes you want a dessicant that works at high temperatures, or very low temperatures.
Stefan, I’m just beyond stoked that I came across your channel tonight. I’m building my FIRST 3D printer and I’ve been immersed in everything for some time. However, I was curious, months ago, re moisture. As a kid, I’m 55 now, I worked in injection molding and blow molding and I was very tuned into our dryer technology at the time and the role drying played in the process. This video was priceless. Thank you, so very much.
I use vacuum sealed reusable bags for storage with desiccant packs, then dry in a food dehydrator for several hours before printing. Works like a charm!
Stefan nice video with good information. I got great results from a vacuum drying system that I built for drying nylon and carbon fiber PC but also found it was great for ABS, PETG etc. I noticed that extruding smells are also greatly reduced using dried filament. My vacuum pump is a small heavy scientific 50L/minute two stage oil sealed model. The ultimate vacuum (no flow) is 0.1 or 10 pascals if the ballast valve is open. When the ballast valve is open, the bleeding air removes condensates from the oil. Have a smile at my $50 stove top pressure cooker vacuum chamber that is big enough for 2 spools (fresco brand but the cooker is sold under many names). I only had to invert one threaded pressure safety relief valve so it would seal under vacuum. I designed 3 printed parts, and using a refrigerant diaphragm valve that seals to the lid with O-rings I was done. A removable flexible stainless-steel natural gas appliance hose connects the valve to the vacuum pump with another printed part. Unfortunately, I only have a 1gr resolution scale to measure any weight change. Instead as a dry indication test, I print clear filament at the maximum recommended extrusion temperature then look for bubbles in the purge line using a x10 magnifying glass, and with PC I don't get any in the first 12 hours after vacuum drying. With a non-scientific vacuum pump which produces less vacuum, I would feed a very small amount ballast air directly into the vacuum chamber but only while pumping to reduce the chamber humidity and the pump oil moisture. -Peter
Dear CNC Kitchen, I am using vacuum storage bags with silica gel to store my filament and I am mostly happy with it. I can also put my spool in the bag over my heat bed to warm the spool to combine the effects. It is a cheap solution.
Just lately ordered my first 3D printer and already your channel is my go-to spot for technical information about 3D printing. Keep up the good woork, as you yourself said "at least a bit scientific" content is much appreciated! :)
2. FYI Stefan - I emailed Taulman a couple of months back asking them for their advice on best vacuum drying practise esp for taulman 910 (a nylon) and other of their functional filaments explaining I was using a small vacuum pot like you, with a PID controlled heated mat underneath the vacuum pot and filament, with loads of dry desiccant inside the vacuum pot - and this was their reply - “With a vacuum unit about 8 hr's max.....usually around 4 at 40C and a good vacuum. Max temp should be about 60C....sometimes you can make the line a bit "oblong" as it's wound tight in some spools. As you're doing, the key is desiccants in the dryer....Most don't realize this is the main KEY as otherwise, you're just using hot and humid air and it won't really dry well”.
What's the issue with indicator? Haven't heard that before. NOT saying there isn't an issue, just asking for more info. First time I've come across this.
The blue-pink indicator is cobalt-based. The yellow-blue indicator should be fine. I wouldn't be too concerned about either if you don't plan to eat it ;) EDIT: Turns out yellow indicator has some variants, those with methylviolet are "concerning". www.giebel-adsorber.de/images/unternehmen/neue-produkte/GIEBEL-FilTec_Unbedenklichkeit-Silicagele_DE.pdf
The orange one is safe, the blue one contained cobalt chloride and has been banned in the EU since 1998. The orange one is also more expensive. www.agmcontainer.com/blog/desiccant/indicating-silica-gel/
Blue silica gel uses cobalt(ii)chloride which is carcinogenic. Orange silica gel uses methyl violet which is toxic, but not in the concentrations found in commercial silica gel. Concerns with methyl violet are more likely to do with water contamination due to improper disposal
Yep. The main concern here is probably contaminating your kitchen's oven, although the plastics themselves are probably outgassing more harmful stuff. Ideally you want a separate oven that is not used for anything food-related.
That was a lot of work and time, nicely done, thank you. I didn't see an ad before the video...not monetized? 489k subs and no ads, if that's right, I'm super impressed. I respect that. If I got that right, and I'll watch more from you to see, that's worth a sub "just because". Again, nicely done video and thank you.
I've been waiting for this one! Great analysis. I'd been wondering for some time if vacuum drying was worthwhile. Thanks so much for doing the footwork!
This is a fantastic video! I've seen all kinds of recommendations about drying, but they were all qualitative. One said, "You can't do any better than vacuum drying!", but clearly shown in your tests that's not true.
I successfully use a vacuum chamber to dry my filament. I use a cheap single stage vacuum pump left running all night but i didn't have much success until I wrapped the chamber in a heating pad to raise the vapor pressure of the water. With this method i usually get about a tablespoon of water out of a roll of wet nylon (it ends up in the pump's oil!).
I've had pretty good luck using vacuum to dry filament...mainly nylon and PETG. I use a hotplate and put the filament on a fire brick in the "chamber" (just a big aluminum stock pot), then preheat until the chamber walls are around 100c with the lid on, then I pull vacuum for and hour. I found the preheat crucial because once the vacuum is pulled very little heat is transferred to the filament. I also experimented with a large cup of desiccant in the chamber...but it comes out at least as dry as when it started. The hour process takes about 1.5 hours and has served me well, ymmv but it works for me.
Wow! What an incredible amount of work! I'm so happy to have access to someone like you that puts all these tests together and publishes the results in concise and fun ways. :D
Thank you Stefan! I have been toying with the idea of using a similar vacuum system as a dryer for about a year now, but never got around to buying one. Your video just saved me a whole lot of money. Thank you so much for producing such a detailed comparison of the two methods. I really appreciate it.
12:59 I think the english translation for circulation oven, is fan forced oven. Also most materials only need to be dried at 55C and not 70C. If that's not working well because you live in a humid environment, put your dehydrator in a bigger box (close to air tight) with desiccant (kitty litter works 😉).
Great video. I was able to use this to teach my son Ethan how convection dehydrators work. In addition to the fact that heating the air reduces the relative humidity, the water in the material will be heated as well, and higher temperature water evaporates more easily (quickly).
Have you tried combining them? Vision miner is selling a kit that involves an oven and a vacuum pump/pot with instructions to bake, then vacuum and agitate for maximum affect. Awfully intriguing, but I don't have anything that will print peek - which is their target audience for the several hundred dollar setup. Seems like it should work for lesser materials, but I don't have a vacuum pump/pot to try it out with. Thanks for all you do!
I have a mountain of old (1 year+) filament that I was trying to dump, now its all 'like new' thanks to a decent drying system! I find that an hour or so in the dryer with PETG improves print quality, even if there are no visible bubbles prior to drying. It shines better in vase mode...
Few things on vacuum drying, based on my experiences in a professional enviroment: 1. heat speeds the process up significantly, we use a vacuum oven for filament and also dessicant at work 2. the pump needs to extract the moisture constantl, ideally with some gas ballast( dry nitrogen) 3. rotary vane pumps emit oil everywhere, also towards the suction side, therefore an oil trap is generally a good idea, and an exhaust filter foe your own safety. 4. the oil can take up the moisture you trying to pump, gas ballast helps here and regular oil changes 5. with a good pressure gauge you see when the moisture gets less 6. at work we get so much water out the system, that we need to collect it from the exhaust of the pump In case of interest, we use an ancient Heraeus vacuum oven and a oil free and dry scroll pump with gas ballast. Drying a role of really moist filament easily yields 15-20 ml of water, this of course does not include the amount taken up by the ballast gas We do a ton of
Hut ab für den Aufwand! Solche Experimente dauern einfach echt lang wenn man sie ordentlich machen will. Deine Videos sind immer ein Vergnügen und extrem hilfreich.
Short term, I typically store filaments on the printer or in a cheap dryer that I got off of Amazon. For long term storage I have a vacuum sealer that I use to repack the spools in their boxes under vacuum with some desiccant in the middle.
Printing from my bedroom. Printer and all my filament is about a foot from the open window... had blowout issues all the time. Got a tiny office dehumidifier and slapped all the reels into a closed plastic storage tub with it. I just left it in there for a couple days before printing again, no problems since. I'll build a custom proper reel storage / dehumidifier when I move and have a dedicated workspace.
Pretty interessting video! Especially the fact of the dry box. Many people said it only helps for containing the hydrated filament, but now I know it actually dehydrates it a fair amount (not as good as oven or dehydrator, but good enough I guess). Thanks for the long tests and sharing the results!
It's not binary. It works on materials that don't greedily absorb moisture, but very hygroscopic materials may need to be heated. The heat excites the polymers(they get soft and wiggly), which allows deep moisture to migrate through to the surface. PET(not PETG) is an example of a material that needs heat to dry properly.
ho man, I'm so happy you did it and report it this way ! It's a 2 years old project you have confirmed ! I'm sure vacuum drying should be better, in fact you may give a try to it with a 2 staged pump ... going only 0,025 Pa. This should dramatically improve drying AND drying time, consuming really low energy compared to dehydrator solution. Thanks for making these test this scientifically compared ! really a huge help !
Thanks for these tests Stefan! I'll be sure to point any of my 3D-printing colleague to this video next time I get a weird look for saying we should always be drying our filament spools before using them to optimize printing quality and general printed parts toughness.
I made a dryer out of an old crock pot cooker. The ceramic dish got broken. So I took and made a 3d printed mold that fit 2 of my largest spools in it, filled the crock pot up with concrete. Next I drilled a 1/8" hole in the concrete and embedded a thermostat. It works really well, and to was made of stuff I had laying around the shop. In the future I wasn't to add an Arduino to control the temperature and have presents for each type filament.
Great work. may I suggest molecular sieve media instead of silica desiccant. It is more hygroscopic and so reduces moisture levels further. It also does not give up moisture as much. I see in your data that the filament was able to pull some moisture our of your silica media when dried and then stored there. If you elevate temperature of the silica it even gives off more moisture yet... Note when you regenerate the molecular sieve media you do have to get it quite hot to break the bonds and release the moisture...
What about microwave drying? Thats something I've tried with old TPU parts that tend to get stiffer after months of exposure to moist air. They got more flexible again immediately (after they cooled to room temp of course). That would be a good meeting to examine.
Since a microwave works by heating the water molecules, that could work... but I would worry about the uneven heating going over temperature and deforming some parts. Microwave ovens have turntables because of that uneven heating issue. I wouldn't recommend it for that alone, much less any additional out-gassing of the plastic in a place you cook food... unless you want to dedicate a microwave to that (along with the aforementioned issues)
I have an older gas oven with a constant pilot light. I keep my most often used rolls in the oven and when im going to use an other roll i just put it in the oven for 24hrs. I dont turn the oven just let the pilot do the work. In addition i keep my other rolls in a non working mini fridge and that helps to keep them pretty dry
Thanks Stefan! Great timing, I just ordered my dehydrator yesterday and I wanted to see a video like this for years... Thank you for demystifying dehydrating of filament! You the man!!
Got some in the description, though you can't get the "First Austria" outside of Germany. Just choose one with temperature control and plenty of space.
Again: very great video! I personally don't like to warm up my materials where I also warm up my food afterwards... For PVA I bought a Polymaker Polybox Edition 2 now and set my PVA inside of it... It uses 2x100g Bags of Silica gel. In the Manual it says that after 2 hours the humidity would drop below 15%... Now after 3 Days its still at 28%... Above 20% the manual tells to change the desiccant.... So the Silicagel was not even close to be able to dehydrate the moisture out of the PVA that it soaked in within the last year... I dried the silica gel now in an unused microwave for 5 min in defrost mode... When you take the bag out afterwards you really can feel that the bag is still a bit wet. Don't directly put the silica-bags in again - let it sit for a while to dry... else you'll have a humidity in your box of sometimes 50 to 80% ^^ Will do that now again till I dried out the materials!
Vacuum desiccators are usually used with a desiccant. I have a couple pounds of silica gel in my vacuum chamber and it works great. I'd like to see a comparison with a setup like this, as it's rather inefficient to run a dehydrator for twenty four hours and vacuum drying seems the better option, especially when you consider that a vacuum setup is a useful tool in general. Mine didn't cost very much either.
What about heating the filament in the food dehydrator then (while still warm) vacuum pumping? Vision Miner was using this method in a video from the 3DPrintingNerd.
@@CNCKitchen I think you should have done the test with warm/hot filament. You need energy in the vacuum to facilitate evaporation. If you put in "cold" spools, there's not enough energy to dislodge the water molecules. Ideally, you'd have a heating pad inside (or at least on the bottom of the chamber on the outside) that imparts energy into the spools while under vacuum to facilitate the evaporation. Without this added energy, the water is not going to go anywhere.
Hey there, do you remember my last comment about these pesky audible ads? Have to admit that your ad you created in this video isn't bugging me as much as they normally do. Good job. And, as always really nice video and a big thanks for all the hard work you put into your videos, really appreciate it. Thx! ❤
Stefan, the correct process to dissect any filament is with a vacuum oven. I mean; you have to combine both things obligatorily for the process to work correctly; First, the oven must heat the filament for a few hours (time and temperature depending on the material of the filament). After the filament "sweats" all the moisture, the vacuum must only be generated after that. Depending on the type of oven used, it is also useless to generate the vacuum and the temperature simultaneously, because if the heating system is outside the vacuum bowl, the vacuum is a thermal insulator and it would not conduct the temperature well to the filament roll either. Therefore it is necessary to heat first, and then to generate a vacuum.
I wondered about vacuum, I used to use it when i woodturned, cut fresh wood, microwave it on lower power until the wood sizzles then cool it in a vacuum pot to recompress the wood and force the moister out. Works well.
I'm not sure if this applies here, but a problem in some materials is that they have as high an affinity for water as desiccant does. Low hydrogen stick welding electrodes are like that. To keep those dry, I double bag my material. Put a little bit of desiccant in a mylar bag with the filament and heat seal the bag. (Impulse sealers are inexpensive.) Then keep the bagged filament in a plastic box with a large amount of desiccant. The mylar bag is an excellent barrier against water diffusion, and the desiccant in the plastic box will keep the relative humidity in the box low, further reducing diffusion. That setup keeps even welding electrodes dry, that normally have to be stored in an oven or metal cans.
Hey Stefan if you want to completely dry the filaments using vacuum pump you need to freeze the filaments, then put them in vacuum pump when water freezes it separates from the materials and when unfreezing at low pressure it goes from solid to gas directly also a totally dry desiccant should be present in the chamber while material us unfreezing
Stefan, your setup for vacuum drying is not even in vacuum (vacuum gauge @6:15 in video). You have a vacuum leak in your setup. Equal amounts of molecules being replaced in container as your pump evacuating…. This is an error in measurement.
I store all my spools in plastic bags (with the desiccant packets they come with) whenever they're not on the printer. This helps them to be protected not only from moisture (not too important for PLA, but other materials can suffer a bit), but also from dust, since I'm in a bit more dusty area than usual. I had success drying Nylon and recycled PET in the oven set to ~70°C, keeping them there for several hours or even overnight. They printed just smooth after that. I never had to dry anything else though :)
So in college when doing microelectronic fabrication our chambers removed moisture while under vacuum by having a coil of tubing we ran liquid nitrogen through. That way the moisture bouncing around in the near vacuum had somewhere to freeze too. It was explained to me that even under a very high vacuum moisture won't necessarily leave.
You're a champ! This is exactly the info I needed. I'm sitting on about 40 rolls of filament and while I'm almost constantly printing I need to make sure some of my more prized filaments are in proper shape. Thanks!
You should do an experiment using white rice dried in a dehydrator at 125 deg in place of desiccant. 125 drying temperature was determined by a manufacturer as the best temperature for drying rice. I have used this for storage in the past, and the rice can be redried periodically. You will not get fumes off of it like you will some desicant.
have to say 2 things: 1) GOOD JOB, very well done, congratulations. 2) Also a good selection of materials and a very good plot of the results, I was aiming for Petg, and now I know what I will do. 3) For the first time, i really want to test audible XD
Vacuum chamber plus desiccant is the standard that have been used in Chemical research since forever, laboratory glassware vendors sell a device that looks like a glass pressure cooker that has the brim of the cover ground pretty flat and you put appropriate lubricant so that you can slide it in and out and it has a connector for the vacuum pump, if you use heat + desiccant + vacuum you going to get as dry as it can get.
you can also weigh the desiccant to determine its moisture content. i remember seeing a garage dryer that was hung from a spring and when it touched the floor it was full, and needed refreshing. probably need to make some cotton bags with weights written on them for the desiccant to make it easier to use.
I indeed had lots of trouble printing PETG, a lot of prints came out very weak and brittle. At the time I was very disappointed by this because I mostly switched over to PETG to make mechanical parts. However, after reading about this and having it in a descendant bag for some time, the print quality much improved and now prints 'proper' parts.
The problem when vacuuming alone, is the temperature drop since you take out the higher level energy molecule and leaving the lower energy molecule to cumulate, which then requires to succion even more vacuum. The solution is to do both. I use to have access to a cole parmer vacuum oven, and these thing works very well, you put it about 4 hours and print quality were as good as new. I guess leaving the vacuum pot in the sun should also work for petg or abs.
My best results of drying filaments are in a dehydrator for vegetables and fruits. It cost low money, uses a little of electric power and helps drying my filaments in 1, maximal 2 hours. I mostly print PETG and Nylon with cf or gf. For my dehydrator I printed a drying chamber in which I can dry 3 rolls of filament at a time.
Crazy, I'm doing this kind of study with Ultem 9085 at work right now. From my testing vacuum plus a little heat (50C) is just about as good as you can get, it drives off of the moisture in 48 hours.
@@CNCKitchen Quite well. I haven't had time to run the filament in the oven for more than two hours, but even that made a clear difference. I'm thinking of buying a dedicated drier. We'll see
The vacuum chamber will not only boil water but it will also freeze it. Pre-heating wet filament only helps initially until the temperature drops during pressure reduction. Adding heat such as pad heater on the outside of the vacuum chamber can help some even though heat does not transfer well in a vacuum. In industrial freeze drying (lyophilization), heat and dry gas bleed (nitrogen) are used to speed up the process of sublimation. Controlling pressure at 0.5 - 1 mBar & 25 - 30 Deg C would be much more efficient than just letting the vacuum pump run and achieve it's minimum pressure. An efficient vacuum drying system is not practical for home use but a simple vacuum drying system is still a tool that can be used to remove moisture from filament without the negative effects of heat alone.
I also bought a set of vacuum pump and a pot. To dry it the filament also needs to be warm. But the critical point is when you float the pot after vacuum to open the lid it soaks all the ambient air with moisture back in. To make it professionally you had to do the floating after vacuum with dry nitrogen. At last I keep my non pla filaments in boxes like yours with silica gel pearls. On top I placed in each box a wemosD1 mini with a a humidity probe which tells me what happened inside without opening the box unnecessary.
I used to work in extrusion plant, the rule was a minimum of 2H @ 90 C° for the TPU we were using (hot air blown for bottom to the top on plastic pellets)
Ready Player One? Yeah no - but not why I watch your channel and you gotta pay bills. High quality videos, thanks Stefan! I was wondering the same. In the lab we used dessicant and vacuum to dry products of solvents.
Did you try a heat lamp on the filament in the vacuum chamber? It gives the water the energy it needs to change phases easier. With out it it freezes like a sublimator in space and inhibits evaporation
@@karoo_bushman6880 i think maybe if your doing it commercially it might be a good idea. For hobby use i would stick with the traditional heated dryers they are about $50 and do the job petty good. I run a esun Dryer i got on amazon and haven't had any issues. I think the advantage of the heated dryer is also that if your running a long print on a humid day the heated dryer can keep running while using filament.
I kept filamnets (PLA) in new garbage bags, in one of the bags I forget to put little silica bag. I didnt use that filament for about a year. After that same problem. The filament broke every 30-50 minutes. I had to babysit the printer that afternoon.
I store my filament in my kallax shelf from ikea. I installed a door and made it air tight, installed a moisture sensor and put a tub of silica gel inside. In each of the cabinets i can store five 1kg spools.
Don't forget to share this video on Facebook, Reddit, Twitter and other social media!
*Timestamps for the impatient*
Introduction: 1:20
Why dry filament: 2:40
Damaging filament by heat: 4:58
Vacuum drying: 5:55
Moisture pickup in open air: 7:30
Food dehydrator drying: 8:00
Storing filament in vacuum + desiccant: 9:55
Do spools dry equally: 10:25
Drying with desiccant: 12:00
Drying in a circulation oven: 12:45
Summary/TLDW: 13:57
Add your timestamps to the description and they show up on the progress bar now its a cool new youtube feature
Damn the summary goes straight to an ad. I'd appreciate it if you'd have a summary with data points that are actually useful. I don't mind the ad, but not having the data there is frustrating. (Yes, I get it, free thing and you need to make money.) Today I happen to be in a hurry, and I know your videos take about ten minutes to get to a conclusion.
@@avaviel Click the TLDW timestamp... Then you only have to watch the conclusion to get the requisite information.
You should heat up the material before you put it in the vacuum chamber to get the best result. Look at this video. m.ruclips.net/video/lnVjxuCc4Wc/видео.html
Or use an desiccant dryer with dewpoint lower than -20c m.ruclips.net/video/gKFJPzi0JYg/видео.html
Being a HVAC tech with years of experience dealing with vacuum pumps and pulling water out of HVAC systems installers forgot to keep capped during installation, the fact that a crappy single stage vacuum pump is being used isn't helping.
A 2 stage vacuum pump that can be run overnight without overheating is pretty much required if you want to get a significant amount of moisture out of a enclosed system, as is the ability to maintain high vacuum for the duration.
Also its generally a good idea to have some kind of water trap, either a cold finger or a tube packed with a chemical dessicant, or be ready to change your compressor oil often as the water extracted from the vacuum chamber will end up in the pump oil and damage the pump over time if the water isn't removed from the oil.
If done right, with proper equipment, a vacuum drying setup should be able to drop the moisture content in the filament way lower than a food desiccator can in the same amount of time in most climates that aren't a done dry desert.
I also understand that this testing is being done "on a budget" but its kind of misleading to say the dehydrator is simply better when half the issue here is that the vacuum setup is just too cheap to do its job well.
This..
I use 2 stage rotary vane and turbo pumps regularly in my lab, and drying solids is always done under red pressure rather than dessicants, it's considerably faster and produces much dryer material. A dessicator might take days to achieve what just an hour of a few micro bar vacuum manages. Although as mentioned, overnight drying on a vac line gives best results, especially for bulk materials.
I demand a recount!
I've had way better results with 12 hours of vacuum pump running than 24 hours in a food dehydrator. I always run mine with indicating desiccant visible through the lid so I can clearly see when it is "ready" to come out.
My experience was with an even worse looking cheap pump that I put good oil into. It gets hot, but seems to work. I just change the oil once it starts looking milky.
@@TrevJonez As frager56 mentioned above, a water trap will save you a lot of oil changes: use a long clear tube full of indicator desiccant, and you can watch the color change, then you know when to stop and bake the desiccant.
I fully agree with your assessment but I think Stefan was catering to the 3D printing community that might not be able to afford a good two stage vacuum pump for merely drying out their filament. I also concur with having a desiccant or other means to absorb the shed moisture and keep it from getting sucked back into the filament once the vacuum is released.
I'm sure you realize this already but you have done the 3D printing community such a service with all of your videos providing real test data and results over all these years. Just amazing!
Absolutely!
I have the same food dehydrator, got it at Aldi. I bypassed the factory temp controls and added a PID controller and a temp sensor. it can go higher and is very stable in comparison to the factory temp control which was swinging around 10-15 degrees..
Just curious, how did you do this?
@@littlecube141 same how???
How??
@@nxxxxzn ruclips.net/video/RNV8F5v13Vc/видео.html
@@nxxxxzn not my video but similar concept can be applied to any heating system. The controller is cheap on eBay or Amazon
What are the health concerns over silica with indicator dyes?
Edit: For those wondering, the indicator dye used in silica beads will either be cobalt chloride (blue) or methyl violet (orange). Both of these dyes are toxic and carcinogenic, though methyl violet is regarded as less dangerous than cobalt chloride.
Using small amounts of methyl violet indicator in sealed packets seems to be considered food-safe, but if the indicator is loose like how you buy it in bulk, you could inhale or ingest dust from the beads, thus consuming trace amounts of the dye.
Thank you, that was Exactly my question (I use blue indicator) I will be throwing it away immediately and buying non color indicator.
Cobalchlorid is a salt… there is nothing to worry as long as you do not eat a lot of it.
@@DrMarcArnoldBach whether or not a compound is a salt is not an indication of toxicity… Cyanide comes in many salts, for example. As does lead
@@tildey6661 you are correct, but I don't believe that is what he is saying. He states his opinion that the risk is minimal or non existant unless you have some excessive physical contact, such as eating it, with the compound in its salt form. The last part of his statement indicates that it is indeed toxic "as long as you do not eat a lot of it".
What it being in a salt form, or any toxic compound being a salt for that matter, has to do with the risk it imposes I'm not really sure. Perhaps he is insinuating that under normal circumstances it is not a volatile compound and your not going to encounter it in the air, i.e. not being a volatile substance, not going to be in the air, not going to sublime, turn into a gas, no vapors, thus your not likely to come into contact with it under normal circumstances. That theory is a long shot and seriously flawed however, i'm not going to look into the vapor pressure and vapor temperature to begin to decipher what his logic was and its based on my assumption that he has limited education and training in chemistry. I could see someone without a strong understanding of chemistry thinking that there is no exposure to the compound unless you physically touch it.
The orange-green ones I bought say they do not contain cobalt chloride. Now whether what they actually contain is toxic or not.. no idea.
Your scientific pursuit of all your topics is greatly appreciated!
Question: "Vacuum or Dehydrator?"
Short answer: "Yes."
Long answer: "Use both. Freeze-driers are a good example of this."
Nobody is asking about what looks like a E3D toolchanger printer next to the vacuum chamber at 2:23? I'm eager to get a review, it looks so promising! As usual, great content Stefan!
Coming soon!
@@CNCKitchen Can't wait! I'll see you in the next one 😁
@npgoalkeeper _ wow, good to know the internet police are on the job.....
Great Video, and a very great job with info for all of us 3D printer people! Really enjoy watching your videos.
I see how the vacuum pump would work and do a great job. I am new to 3D printing. My wife got me my 3D printer for my Birthday last November.
Here in Tennessee where I live, we are blessed with many rivers and man made lakes! I live in Middle Tennessee almost at the geographic center of the state! Humidity is a major problem here. My Grand Mothers, both of them, taught me to put rice in the salt shacker to absorb the moisture so the salt would flow. Rice is very hygroscopic (absorbs water from the air) and will pull it away from just about anything. So I put rice in a 3 inch by 4 inch mesh bag (you can buy online for $7.00 or so for 100) and put the rice bag and filament in a one gallon zip lock freezer bag. Zip it shut and the rice will absorb any moisture left in the sealed bag. Put one bag of rice in with a spool that has one half or less filament. I put two bags of rice in the gallon zip lock bags that have filament >= one half a spool. It cost almost nothing, and it makes working with the spools of filament very easy. Just open the bag and switch filaments. When you have a filament dryer, dehydrator, or a vacuum pump, you can only work with one spool at a time. There is nothing wrong with that, but drying them all with rice can be done all at once!
I know I have only been a 3D printing guy for about 9 months now. At this moment I have 29 spools in bags with rice drying 😆!!!
Again, thank you so very much for all that you do!!!!
You cannot dry stuff (completely) in a static vacuum. Unless you leave the pump running continuously the moisture will reach an equilibrium point within the chamber. When you leave the pump running constantly no equilibrium can be reached as the moisture is constantly being removed. To protect the oil in the pumps in labs you generally use two traps in series. The first bubbles the output of the chamber through sulfuric acid, which takes all the moisture out. The second trap bubbles the output from the sulfuric trap through mineral oil, which traps any fumes coming out of the first trap. Generally these two fluids only need to be changed once a year. Same with the pump oil. You will be amazed at the amount of water that will come off with only limited heating. Generally, proper vacuum drying will move about 1-1/2 times the moisture that air (hot) drying will.
how do you know when to replace the sulfiric acid
@@termivan Generally you just make a line on the container (preferably clear) at the height of the fresh sulphuric. Once it doubles in volume it is time to change it. Usually that works out for us to be about once a year, but that will depend on how much you use it.
@@bradquinn4161 excellent, I see that my job (assembly shop) has about a 500g chamber of dessicant in line to the industrial pump, I am considering all these options to help prevent oil contamination in the future, where might I be able to shop for such accessories?
@@goguyted We used to use two of the used (and clean) acid bottles(2 litre?) with some of the red vacuum sealer. The bottles are shatter proof due to shipping regulations. If you have a local lab they MAY just give you a couple.
Many years ago I used to manage a thermoplastic molding workshop where we heated plastic sheets enough to be formed on a mold using vacuum. Some plastics are hygroscopic meaning they absorb moisture such as PETG and polycarbonate (lexan) being the worst. In order to mold PETG or polycarbonate sheets thicker than
Or you could always use an oil-free vacuum pump
@@AU12 Yes, depends on the degree of vacuum required.
You could consider this post a "peer review" of your findings. TLDR: I found results consistent with your findings.
I've been running theses "drying" tests for over 6 months now and I don't think drying water is the issue. I've used ABS. ABS+, PETG, PLA and PLA+ in my tests, sorry, don't have the "exotic" filaments you have. I have not seen even ABS absorb much water when submerged for 7 days, that was above a .2% change, (2g of weight using a 1kg spool.) What I think needs to be tested, is what chemically happens to filament as it ages, and how applying heat to the filament changes the filament. My thinking is that heating the filament causes some gassing, chemicals in the filament to be released due to decomposition of the filament. Since I don't have the ability to sample the gases, this will only continue to be just a "guess".
- My method of testing is to weigh the filament as soon as it is unsealed, dry the filament at 135F for 5:30 hours, re-weigh, submerge the filament for 7 days 100% under water, remove the filament from the water, spin dry the filament in open air for a few minutes to remove any extra water that may be trapped in the spool, weigh the filament, re-dry the filament at 135F for 5:30 hours, re-weight the filament, and record my findings. If all I see is a .2% change in the filament, then I'd think water is not the issue.
Does anyone else have any test data to share? So far this all has been subjective data about drying and there is still lots of research to be made - WITH this said, I still recommend drying where there is a filament issue, what ever heating the filament does, it seems to improve print quality.
I'd like to see more of these types of tests, however, I'm not sure if you can get the test equipment you need, or if there' s a "market" from your watchers. My guess is not everyone else is a nerd.
Thanks again for a great video.
There is one major problem with your test method... liquid water has a very strong cohesive force and does not like plastic in the first place... it is not as likely to penetrate as the individual molecules of water in a high humidity environment. Your water was also likely to be room temperature (or lower) which will also affects absorption. As an example, drop ramen in cold water and hot water and see the difference in time it takes for the ramen to get "soggy". Humidity is measured while water is in its gaseous state, *Not* when it has condensed to a liquid... the temperature that it condenses is the dew point. A far more appropriate test would be to "steam" your filament for an hour and measure the weight change, then try drying it.
@@edwardpaulsen1074
Ed, Thanks for the reply, I was hoping someone would reply to "check" my work, however, you have only brought more questions - I like that! Since my numbers are very close to the ones posted in the video, how do you think steaming would be a "real world" test? These tests are to determine the normal absorption of water, during normal use, not try and force as much water into the filament as you can. To go back to your Ramen example, we are trying to measure water absorption with it sitting on the counter, not being boiled in water - We leave our filament out in room temperature air, not in a hot sauna.
I'm sure I could be missing your point.
Thanks.
meance perhaps he is saying to get maximum water absorption, use steam. It would be interesting to see how different the values would be. I live in the southern US, so high heat and humidity are a factor here.
@@smportis Think about what you say - Steam is normally created at 218 F, you your thinking heating filament PAST the temperature that most filament is extruded at is what he's saying?
An open container of water will normalize at the temperature the humidity will be in the area - room temperature is 80F, the humidity is 80F. Any other tests are not valid because you are not simulating the normal environment filament will be exposed to. - Even the water submersion is an valid test, unless you store your filament in the rain.
@@smportis Soory it's early in my day, so I should have used C instead of Water boils at 100C/218F, and filament is extruded ate 212C, TWICE what water boils at.
Great video with a lot of detail. I usually store my filament rolls (i use mostly PLA) in the plastic bags they came with. I add a perforated transparent film-container filled with dessicant to each bag and close them with ordinary bag clamps. Benefits are: less potentially humid air in the bags compared to plastic boxes), doesn't require much space, is relatively air-tight, costs next to nothing.
Thank you for mentioning the indicating silica gel health concern, I didn’t know about that!
Thanks for posting this. I dry wood frequently in a vacuum kiln I made. The two key factors are the amount of vacuum, and heat but not too much. Your little vacuum pump should be able to deliver adequate vacuum. I automate my kiln with a simple mechanical timer turning the pump in every 90 minutes for about twenty minutes. I also put a one way valve in-line because the pump doesn’t maintain vacuum by itself very well when turned off. I use heat mats that don’t get warmer than 40C. When the temp in the material reaches about 36C the water really starts to come out. Because the heat transfer is by conduction the maximum thickness for best results is about 100mm. I can dry wet wood at 55%+ moisture to nearly 0% in 3 days. This is 120 year old technology for wood. A German friend has found a 25W motorcycle seat heater to use in a vacuum chamber similar to yours. Sandwiching the plastic rolls between heating pads will give you the best success and speed. You could also try just putting it in a sunny window. However, considering all the messing around, in the end your cheap desiccant in a decently closed plastic container seems to me to be the best option for most plastics.
Do you think heating the filament in a food dehydrator to say 40-50c then stick it in the vacume the way to do it then? Mabye I can screw in one of those water heater elements in the center.
@@warlockd I imagine you’d get a bit of a head start but you still need to keep the stuff warm, about 40C I’d guess, over a day or so under 26.5mmHg vacuum at least - more if you are closer to sea level.
One concern I have is people potentially not understanding the amount of desiccant required to actually reduce moisture. Despite your having shown and mentioned it, IMHO a lot of people just won't pick this up. I'd love to see a comparison data showing smaller quantities of desiccant to help absolutely clarify the "sufficient" quantity.
What if the filament attracts moisture more readily than the desiccant? Wet desiccant will hydrate the filament.
I will vacuum bag with desiccant after the filament has been dried overnight.
@@White000Crow If there was a filament that attracted moisture better than desiccant, we would be using it as desiccant instead of silica gel.
@@jamesgreen4965We actually do use a lot of dessicants aside from silica gel (zeolites, activated alumina, molecular sieves, calcium chloride, sulfuric acid, clay...) you can even find polymer dessicants intended for injection molding (which could theoretically be turned into filament).
Which dessicant is best depends on the use case. Sometimes loose beads, dust, or sulfuric acid can cause issues. Sometimes you need an indicator for replacement. Sometimes you need extremely low humidity (some types of silica gel aren't particularly good at low humidity). Sometimes you want a dessicant wheel which regenerates (releases water) very easily. Sometimes you want a dessicant that works at high temperatures, or very low temperatures.
@@White000Crow that's not a realistic scenario. You should definitely use desiccant while drying
@@coledavidson5630 I use filament dryers now.
Stefan, I’m just beyond stoked that I came across your channel tonight. I’m building my FIRST 3D printer and I’ve been immersed in everything for some time. However, I was curious, months ago, re moisture. As a kid, I’m 55 now, I worked in injection molding and blow molding and I was very tuned into our dryer technology at the time and the role drying played in the process. This video was priceless. Thank you, so very much.
Thanks!
I use vacuum sealed reusable bags for storage with desiccant packs, then dry in a food dehydrator for several hours before printing. Works like a charm!
Wide size of bags do you use for vacuum sealing? It's hard time find a sealer that will seal past 12" for under $400
@@jordanl7489 ebay. 15€ for 50pieces disposable bags, but if you seal and cut well, you can reuse up to 3-5 times. 35x35 for me are good enough.
@@jordanl7489 You can use the bags which are normaly used for vocuum storing textiles. They are cheap and you can use a normal vacuumcleaner.
Stefan, vielen Dank daß Du Dir diese Arbeit für uns machst!
Stefan nice video with good information. I got great results from a vacuum drying system that I built for drying nylon and carbon fiber PC but also found it was great for ABS, PETG etc. I noticed that extruding smells are
also greatly reduced using dried filament. My vacuum pump is a small heavy scientific 50L/minute two stage oil sealed model. The ultimate vacuum
(no flow) is 0.1 or 10 pascals if the ballast valve is open. When the
ballast valve is open, the bleeding air removes condensates from the oil.
Have a smile at my $50 stove top pressure cooker vacuum chamber that is big enough for 2 spools (fresco brand but the cooker is sold under many names). I only had to invert one threaded pressure safety relief valve so it would seal under vacuum. I designed 3 printed parts, and using a refrigerant
diaphragm valve that seals to the lid with O-rings I was done. A removable
flexible stainless-steel natural gas appliance hose connects the valve to
the vacuum pump with another printed part. Unfortunately, I only have a 1gr
resolution scale to measure any weight change. Instead as a dry indication test, I print clear filament at the maximum recommended extrusion temperature then look for bubbles in the purge line using a x10 magnifying glass, and with PC I don't get any in the first 12 hours after vacuum drying.
With a non-scientific vacuum pump which produces less vacuum, I would feed a very small amount ballast air directly into the vacuum chamber but only while pumping to reduce the chamber humidity and the pump oil moisture.
-Peter
Dear CNC Kitchen, I am using vacuum storage bags with silica gel to store my filament and I am mostly happy with it. I can also put my spool in the bag over my heat bed to warm the spool to combine the effects. It is a cheap solution.
Could you explain more about the concerns with 12:32? All my desiccant is indicator changing but I also have a digital read out as well.
Sir, I bow down before your patience, rigorousness and persistence. I can't thank you enough for the wealth of information you provide to us.
Thanks
I just found your channel and now that I have a 3D FDM prinder, I really appreciate all the work and the data. Thanks.
Just lately ordered my first 3D printer and already your channel is my go-to spot for technical information about 3D printing. Keep up the good woork, as you yourself said "at least a bit scientific" content is much appreciated! :)
2. FYI Stefan - I emailed Taulman a couple of months back asking them for their advice on best vacuum drying practise esp for taulman 910 (a nylon) and other of their functional filaments explaining I was using a small vacuum pot like you, with a PID controlled heated mat underneath the vacuum pot and filament, with loads of dry desiccant inside the vacuum pot - and this was their reply - “With a vacuum unit about 8 hr's max.....usually around 4 at 40C and a good vacuum.
Max temp should be about 60C....sometimes you can make the line a bit "oblong" as it's wound tight in some spools.
As you're doing, the key is desiccants in the dryer....Most don't realize this is the main KEY as otherwise, you're just using hot and humid air and it won't really dry well”.
I live in a very humid region (40%+ ambient). This is very helpful information.
What's the issue with indicator? Haven't heard that before. NOT saying there isn't an issue, just asking for more info. First time I've come across this.
The blue-pink indicator is cobalt-based. The yellow-blue indicator should be fine. I wouldn't be too concerned about either if you don't plan to eat it ;)
EDIT: Turns out yellow indicator has some variants, those with methylviolet are "concerning". www.giebel-adsorber.de/images/unternehmen/neue-produkte/GIEBEL-FilTec_Unbedenklichkeit-Silicagele_DE.pdf
The orange one is safe, the blue one contained cobalt chloride and has been banned in the EU since 1998. The orange one is also more expensive. www.agmcontainer.com/blog/desiccant/indicating-silica-gel/
Blue silica gel uses cobalt(ii)chloride which is carcinogenic. Orange silica gel uses methyl violet which is toxic, but not in the concentrations found in commercial silica gel. Concerns with methyl violet are more likely to do with water contamination due to improper disposal
Yep. The main concern here is probably contaminating your kitchen's oven, although the plastics themselves are probably outgassing more harmful stuff. Ideally you want a separate oven that is not used for anything food-related.
@@Basement-Science Or you can dry it in your 35$ dedicated dehydrator that also dries filament. That works too :)
Danke!
That was a lot of work and time, nicely done, thank you. I didn't see an ad before the video...not monetized? 489k subs and no ads, if that's right, I'm super impressed. I respect that. If I got that right, and I'll watch more from you to see, that's worth a sub "just because". Again, nicely done video and thank you.
Some of the most comprehensive testing regarding 3d printing come this channel. Hands down. Thank you for all your hard work.
I've been waiting for this one! Great analysis. I'd been wondering for some time if vacuum drying was worthwhile. Thanks so much for doing the footwork!
Glad it was helpful!
This is a fantastic video! I've seen all kinds of recommendations about drying, but they were all qualitative. One said, "You can't do any better than vacuum drying!", but clearly shown in your tests that's not true.
I successfully use a vacuum chamber to dry my filament. I use a cheap single stage vacuum pump left running all night but i didn't have much success until I wrapped the chamber in a heating pad to raise the vapor pressure of the water. With this method i usually get about a tablespoon of water out of a roll of wet nylon (it ends up in the pump's oil!).
The amount of detail is impressive. Every video is like CSI or better FSI FilamentSceneInvestigation 😉
Fun to watch and aducational af 👍
I've had pretty good luck using vacuum to dry filament...mainly nylon and PETG.
I use a hotplate and put the filament on a fire brick in the "chamber" (just a big aluminum stock pot), then preheat until the chamber walls are around 100c with the lid on, then I pull vacuum for and hour. I found the preheat crucial because once the vacuum is pulled very little heat is transferred to the filament.
I also experimented with a large cup of desiccant in the chamber...but it comes out at least as dry as when it started.
The hour process takes about 1.5 hours and has served me well, ymmv but it works for me.
Wow! What an incredible amount of work! I'm so happy to have access to someone like you that puts all these tests together and publishes the results in concise and fun ways. :D
Thank you Stefan! I have been toying with the idea of using a similar vacuum system as a dryer for about a year now, but never got around to buying one.
Your video just saved me a whole lot of money. Thank you so much for producing such a detailed comparison of the two methods. I really appreciate it.
12:59 I think the english translation for circulation oven, is fan forced oven. Also most materials only need to be dried at 55C and not 70C. If that's not working well because you live in a humid environment, put your dehydrator in a bigger box (close to air tight) with desiccant (kitty litter works 😉).
Convection Oven
Great video. I was able to use this to teach my son Ethan how convection dehydrators work. In addition to the fact that heating the air reduces the relative humidity, the water in the material will be heated as well, and higher temperature water evaporates more easily (quickly).
Have you tried combining them? Vision miner is selling a kit that involves an oven and a vacuum pump/pot with instructions to bake, then vacuum and agitate for maximum affect. Awfully intriguing, but I don't have anything that will print peek - which is their target audience for the several hundred dollar setup. Seems like it should work for lesser materials, but I don't have a vacuum pump/pot to try it out with. Thanks for all you do!
I have a mountain of old (1 year+) filament that I was trying to dump, now its all 'like new' thanks to a decent drying system!
I find that an hour or so in the dryer with PETG improves print quality, even if there are no visible bubbles prior to drying.
It shines better in vase mode...
Few things on vacuum drying, based on my experiences in a professional enviroment:
1. heat speeds the process up significantly, we use a vacuum oven for filament and also dessicant at work
2. the pump needs to extract the moisture constantl, ideally with some gas ballast( dry nitrogen)
3. rotary vane pumps emit oil everywhere, also towards the suction side, therefore an oil trap is generally a good idea, and an exhaust filter foe your own safety.
4. the oil can take up the moisture you trying to pump, gas ballast helps here and regular oil changes
5. with a good pressure gauge you see when the moisture gets less
6. at work we get so much water out the system, that we need to collect it from the exhaust of the pump
In case of interest, we use an ancient Heraeus vacuum oven and a oil free and dry scroll pump with gas ballast. Drying a role of really moist filament easily yields 15-20 ml of water, this of course does not include the amount taken up by the ballast gas
We do a ton of
For how long do you leave the pump running?
Hut ab für den Aufwand! Solche Experimente dauern einfach echt lang wenn man sie ordentlich machen will. Deine Videos sind immer ein Vergnügen und extrem hilfreich.
Short term, I typically store filaments on the printer or in a cheap dryer that I got off of Amazon. For long term storage I have a vacuum sealer that I use to repack the spools in their boxes under vacuum with some desiccant in the middle.
I learn a lot with each of your videos. Thanks from El Salvador.
Thank you for spending your time and knowledge to clarify questions about filament drying. Live long and prosper!
Printing from my bedroom. Printer and all my filament is about a foot from the open window... had blowout issues all the time.
Got a tiny office dehumidifier and slapped all the reels into a closed plastic storage tub with it. I just left it in there for a couple days before printing again, no problems since. I'll build a custom proper reel storage / dehumidifier when I move and have a dedicated workspace.
Pretty interessting video! Especially the fact of the dry box. Many people said it only helps for containing the hydrated filament, but now I know it actually dehydrates it a fair amount (not as good as oven or dehydrator, but good enough I guess). Thanks for the long tests and sharing the results!
It's not binary. It works on materials that don't greedily absorb moisture, but very hygroscopic materials may need to be heated. The heat excites the polymers(they get soft and wiggly), which allows deep moisture to migrate through to the surface. PET(not PETG) is an example of a material that needs heat to dry properly.
ho man, I'm so happy you did it and report it this way ! It's a 2 years old project you have confirmed !
I'm sure vacuum drying should be better, in fact you may give a try to it with a 2 staged pump ... going only 0,025 Pa.
This should dramatically improve drying AND drying time, consuming really low energy compared to dehydrator solution.
Thanks for making these test this scientifically compared ! really a huge help !
Thanks for these tests Stefan! I'll be sure to point any of my 3D-printing colleague to this video next time I get a weird look for saying we should always be drying our filament spools before using them to optimize printing quality and general printed parts toughness.
I made a dryer out of an old crock pot cooker. The ceramic dish got broken. So I took and made a 3d printed mold that fit 2 of my largest spools in it, filled the crock pot up with concrete. Next I drilled a 1/8" hole in the concrete and embedded a thermostat. It works really well, and to was made of stuff I had laying around the shop. In the future I wasn't to add an Arduino to control the temperature and have presents for each type filament.
Great work. may I suggest molecular sieve media instead of silica desiccant. It is more hygroscopic and so reduces moisture levels further. It also does not give up moisture as much. I see in your data that the filament was able to pull some moisture our of your silica media when dried and then stored there. If you elevate temperature of the silica it even gives off more moisture yet... Note when you regenerate the molecular sieve media you do have to get it quite hot to break the bonds and release the moisture...
great video. I love the idea of weighing. I was surprised how much water had been absorbed.
cant imagen the effort behind these videos man!! thanks a lot
amazing dedication and time spent. Incredible Well Done
Thank you for the excellent investigation!
What about microwave drying? Thats something I've tried with old TPU parts that tend to get stiffer after months of exposure to moist air. They got more flexible again immediately (after they cooled to room temp of course). That would be a good meeting to examine.
In the laboratory, vacuum microwave ovens are used to dry out certain types of material. I always keep all my materials in boxes with desiccant
Since a microwave works by heating the water molecules, that could work... but I would worry about the uneven heating going over temperature and deforming some parts. Microwave ovens have turntables because of that uneven heating issue. I wouldn't recommend it for that alone, much less any additional out-gassing of the plastic in a place you cook food... unless you want to dedicate a microwave to that (along with the aforementioned issues)
That works really well with bronzefill filament.
@@xKatjaxPurrsx Any metallic filament to tell the truth, kkkk
I have an older gas oven with a constant pilot light. I keep my most often used rolls in the oven and when im going to use an other roll i just put it in the oven for 24hrs. I dont turn the oven just let the pilot do the work. In addition i keep my other rolls in a non working mini fridge and that helps to keep them pretty dry
Thanks Stefan! Great timing, I just ordered my dehydrator yesterday and I wanted to see a video like this for years...
Thank you for demystifying dehydrating of filament! You the man!!
I've been looking for the right dehydrator, mind providing a link to the one you chose?
Got some in the description, though you can't get the "First Austria" outside of Germany. Just choose one with temperature control and plenty of space.
Again: very great video!
I personally don't like to warm up my materials where I also warm up my food afterwards...
For PVA I bought a Polymaker Polybox Edition 2 now and set my PVA inside of it...
It uses 2x100g Bags of Silica gel.
In the Manual it says that after 2 hours the humidity would drop below 15%...
Now after 3 Days its still at 28%...
Above 20% the manual tells to change the desiccant....
So the Silicagel was not even close to be able to dehydrate the moisture out of the PVA that it soaked in within the last year...
I dried the silica gel now in an unused microwave for 5 min in defrost mode... When you take the bag out afterwards you really can feel that the bag is still a bit wet.
Don't directly put the silica-bags in again - let it sit for a while to dry... else you'll have a humidity in your box of sometimes 50 to 80% ^^
Will do that now again till I dried out the materials!
Vacuum desiccators are usually used with a desiccant. I have a couple pounds of silica gel in my vacuum chamber and it works great. I'd like to see a comparison with a setup like this, as it's rather inefficient to run a dehydrator for twenty four hours and vacuum drying seems the better option, especially when you consider that a vacuum setup is a useful tool in general. Mine didn't cost very much either.
What about heating the filament in the food dehydrator then (while still warm) vacuum pumping? Vision Miner was using this method in a video from the 3DPrintingNerd.
I usually do that before storing it, though I didn't really notice that it helped in any way.
@@CNCKitchen I think you should have done the test with warm/hot filament. You need energy in the vacuum to facilitate evaporation. If you put in "cold" spools, there's not enough energy to dislodge the water molecules. Ideally, you'd have a heating pad inside (or at least on the bottom of the chamber on the outside) that imparts energy into the spools while under vacuum to facilitate the evaporation. Without this added energy, the water is not going to go anywhere.
Hey there, do you remember my last comment about these pesky audible ads? Have to admit that your ad you created in this video isn't bugging me as much as they normally do. Good job. And, as always really nice video and a big thanks for all the hard work you put into your videos, really appreciate it. Thx! ❤
Stefan, the correct process to dissect any filament is with a vacuum oven. I mean; you have to combine both things obligatorily for the process to work correctly;
First, the oven must heat the filament for a few hours (time and temperature depending on the material of the filament). After the filament "sweats" all the moisture, the vacuum must only be generated after that. Depending on the type of oven used, it is also useless to generate the vacuum and the temperature simultaneously, because if the heating system is outside the vacuum bowl, the vacuum is a thermal insulator and it would not conduct the temperature well to the filament roll either. Therefore it is necessary to heat first, and then to generate a vacuum.
I wondered about vacuum, I used to use it when i woodturned, cut fresh wood, microwave it on lower power until the wood sizzles then cool it in a vacuum pot to recompress the wood and force the moister out. Works well.
I'm not sure if this applies here, but a problem in some materials is that they have as high an affinity for water as desiccant does. Low hydrogen stick welding electrodes are like that. To keep those dry, I double bag my material. Put a little bit of desiccant in a mylar bag with the filament and heat seal the bag. (Impulse sealers are inexpensive.) Then keep the bagged filament in a plastic box with a large amount of desiccant. The mylar bag is an excellent barrier against water diffusion, and the desiccant in the plastic box will keep the relative humidity in the box low, further reducing diffusion. That setup keeps even welding electrodes dry, that normally have to be stored in an oven or metal cans.
Hey Stefan
if you want to completely dry the filaments using vacuum pump you need to freeze the filaments, then put them in vacuum pump
when water freezes it separates from the materials and when unfreezing at low pressure it goes from solid to gas directly
also a totally dry desiccant should be present in the chamber while material us unfreezing
Stefan, your setup for vacuum drying is not even in vacuum (vacuum gauge @6:15 in video). You have a vacuum leak in your setup. Equal amounts of molecules being replaced in container as your pump evacuating…. This is an error in measurement.
I store all my spools in plastic bags (with the desiccant packets they come with) whenever they're not on the printer. This helps them to be protected not only from moisture (not too important for PLA, but other materials can suffer a bit), but also from dust, since I'm in a bit more dusty area than usual.
I had success drying Nylon and recycled PET in the oven set to ~70°C, keeping them there for several hours or even overnight. They printed just smooth after that. I never had to dry anything else though :)
The best review on the need to store your filament properly that I have ever seen.. Keep up the good work!!
So in college when doing microelectronic fabrication our chambers removed moisture while under vacuum by having a coil of tubing we ran liquid nitrogen through. That way the moisture bouncing around in the near vacuum had somewhere to freeze too. It was explained to me that even under a very high vacuum moisture won't necessarily leave.
This basically answered all of my questions, what a great video!
not mine. What about a heated vacuum? what about heating the filament then doing the vacuum?
You're a champ! This is exactly the info I needed. I'm sitting on about 40 rolls of filament and while I'm almost constantly printing I need to make sure some of my more prized filaments are in proper shape. Thanks!
You should do an experiment using white rice dried in a dehydrator at 125 deg in place of desiccant. 125 drying temperature was determined by a manufacturer as the best temperature for drying rice. I have used this for storage in the past, and the rice can be redried periodically. You will not get fumes off of it like you will some desicant.
have to say 2 things:
1) GOOD JOB, very well done, congratulations.
2) Also a good selection of materials and a very good plot of the results, I was aiming for Petg, and now I know what I will do.
3) For the first time, i really want to test audible XD
This video is great. Thanks so much for digging this out and sharing the results!
Vacuum chamber plus desiccant is the standard that have been used in Chemical research since forever, laboratory glassware vendors sell a device that looks like a glass pressure cooker that has the brim of the cover ground pretty flat and you put appropriate lubricant so that you can slide it in and out and it has a connector for the vacuum pump, if you use heat + desiccant + vacuum you going to get as dry as it can get.
you can also weigh the desiccant to determine its moisture content. i remember seeing a garage dryer that was hung from a spring and when it touched the floor it was full, and needed refreshing. probably need to make some cotton bags with weights written on them for the desiccant to make it easier to use.
I indeed had lots of trouble printing PETG, a lot of prints came out very weak and brittle. At the time I was very disappointed by this because I mostly switched over to PETG to make mechanical parts. However, after reading about this and having it in a descendant bag for some time, the print quality much improved and now prints 'proper' parts.
What a great resource! Thank you for creating and executing these reference experiments!
I wonder why you didn't try silica in vacuum chamber? That and maybe 50C heat would work wonders in few hours?
The problem when vacuuming alone, is the temperature drop since you take out the higher level energy molecule and leaving the lower energy molecule to cumulate, which then requires to succion even more vacuum. The solution is to do both.
I use to have access to a cole parmer vacuum oven, and these thing works very well, you put it about 4 hours and print quality were as good as new. I guess leaving the vacuum pot in the sun should also work for petg or abs.
My best results of drying filaments are in a dehydrator for vegetables and fruits. It cost low money, uses a little of electric power and helps drying my filaments in 1, maximal 2 hours. I mostly print PETG and Nylon with cf or gf.
For my dehydrator I printed a drying chamber in which I can dry 3 rolls of filament at a time.
sooooo long waiting for this video - u rock
Crazy, I'm doing this kind of study with Ultem 9085 at work right now. From my testing vacuum plus a little heat (50C) is just about as good as you can get, it drives off of the moisture in 48 hours.
Perfect timing as I was just drying mine. Awesome!
Perfect! How did it work out?
@@CNCKitchen Quite well. I haven't had time to run the filament in the oven for more than two hours, but even that made a clear difference. I'm thinking of buying a dedicated drier. We'll see
The vacuum chamber will not only boil water but it will also freeze it. Pre-heating wet filament only helps initially until the temperature drops during pressure reduction. Adding heat such as pad heater on the outside of the vacuum chamber can help some even though heat does not transfer well in a vacuum.
In industrial freeze drying (lyophilization), heat and dry gas bleed (nitrogen) are used to speed up the process of sublimation. Controlling pressure at 0.5 - 1 mBar & 25 - 30 Deg C would be much more efficient than just letting the vacuum pump run and achieve it's minimum pressure.
An efficient vacuum drying system is not practical for home use but a simple vacuum drying system is still a tool that can be used to remove moisture from filament without the negative effects of heat alone.
I also bought a set of vacuum pump and a pot. To dry it the filament also needs to be warm. But the critical point is when you float the pot after vacuum to open the lid it soaks all the ambient air with moisture back in. To make it professionally you had to do the floating after vacuum with dry nitrogen.
At last I keep my non pla filaments in boxes like yours with silica gel pearls. On top I placed in each box a wemosD1 mini with a a humidity probe which tells me what happened inside without opening the box unnecessary.
Thanks for all the time you spent to investigate! Really appreciate!
I used to work in extrusion plant, the rule was a minimum of 2H @ 90 C° for the TPU we were using (hot air blown for bottom to the top on plastic pellets)
Bringing industrial process into the home! Thats what this hobby is all about
Ready Player One? Yeah no - but not why I watch your channel and you gotta pay bills.
High quality videos, thanks Stefan! I was wondering the same. In the lab we used dessicant and vacuum to dry products of solvents.
I have been only using PLA and PETG. I store it in a vacuum bag with desiccant. When printing I put it just on the rack on the machine.
Thank you so much for those lots of works and showing it out...
Did you try a heat lamp on the filament in the vacuum chamber? It gives the water the energy it needs to change phases easier. With out it it freezes like a sublimator in space and inhibits evaporation
Intriguing! Would you say a vacuum drying oven would be a good solution? I think one can be had with pump for under 1000 bucks..
@@karoo_bushman6880 i think maybe if your doing it commercially it might be a good idea. For hobby use i would stick with the traditional heated dryers they are about $50 and do the job petty good. I run a esun Dryer i got on amazon and haven't had any issues. I think the advantage of the heated dryer is also that if your running a long print on a humid day the heated dryer can keep running while using filament.
Left in my room unboxed in ambient air, I have PLA filament snap when coming off the spool on its way to the extruder when printing!
I kept filamnets (PLA) in new garbage bags, in one of the bags I forget to put little silica bag. I didnt use that filament for about a year. After that same problem. The filament broke every 30-50 minutes. I had to babysit the printer that afternoon.
I dry my filaments in food dessicator for some hours after each use and store it in clothes vacuum bags with dessicant. Perfect!
I store my filament in my kallax shelf from ikea. I installed a door and made it air tight, installed a moisture sensor and put a tub of silica gel inside. In each of the cabinets i can store five 1kg spools.
You are awesome. This was the best treatment of the material I have seen. Thank you.