You don't want to trap the heat in, the heat does nothing except mobilise the moisture, if it's trapped it may just work deeper into the spool...
You need a gentle *flow* of warm air to remove the water. Think chimney.
You can either snip out the bottom of a couple of trays to make room for a filament roll or do what I did and print a tall spacer. A translucent PETG spacer has worked OK for me so far...
Either:
dryerinsert This file fits a Mini, print three and glue together to make a spacer.
dryerinsert1 This file fits a i3Mk3, print two and glue together to make a spacer.
Cheerio,
RE:
@burrfree
Many roads lead to Rome. If this worked, feedback would be nice. I wouldn't set the print bed temperature too high though (maybe 50 degrees C. for PETG).
Good success!
Karl
Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen.
RE: Using heated bed to dry filament
You don't want to trap the heat in, the heat does nothing except mobilise the moisture, if it's trapped it may just work deeper into the spool...
You need a gentle *flow* of warm air to remove the water. Think chimney.
....
Agree completely.
I just can't understand spending 500€+ on printer, ~20€/spool for filaments, and then not being ready to pay 50€ for actually working dehydrating solution.
[Mini+] [MK3S+BEAR]
RE: Using heated bed to dry filament
Poor mans dryer!
it works from what i can tell. I created a precision vent on front top right of the enclosure to ensure a calibrated exchange of air.
I wrote a program that displays the time left and shuts down when its done.
I have some flex material in there now that i tried printing with and was popping like crazy. tried this dryer thing for about an hour and a half and seen a bit of an improvement in “less popping”, so now im in for the 3.5 hour long haul. 
RE: Using heated bed to dry filament
Well, while this does work, the box doesn't retain heat very well. I lined the inside with tin foil to help reflect the heat and keep it inside, but its poor at best. Bed heated to 100C, inside temp of box was around 60C. Would be better I think if I insulated it better. But heat loss aside, I was able to dry my flex filament with no deformation of filament using 100C Bed and the box over the top for 3.5 hours.
Printing with the newly dried filament is like night and day. So it does work, but its probably not ideal unless you can trap the heat in better so it can reach the right temp.
RE:
I think you've actually got this precisely backwards - lining the box with foil will REDUCE the dehydrating effect, regardless of whether it increases the temperature (which I bet it will).
The purpose of a dehydrator is threefold: 1) increase temperature of the air to reduce the relative humidity of that air, 2) increase temperature of the wet thing so the vapor pressure of its moisture goes up/the water evaporates more readily 3) transport that liberated moisture somewhere else.
You're focusing on 1 and 2 - get the temperature up, while negatively impacting 3. If you heat up a wet sponge in a sealed box, moisture WILL come out of it, but pretty quickly it'll saturate the air and the sponge will stop drying. The best you can hope for is that it dews up on the walls of the container, and doesn't re-evaporate from there. On the other hand, get some airflow in there to exchange the high-dew-point air with fresh low-dew-point air, and that moisture has somewhere to go, so the cycle can continue until the sponge is dry.
Cardboard is actually a GREAT material to use for a dehydrating container, because it's somewhat insulating, but also extremely moisture permeable. Rather than exchanging moist air at all, the moisture just absorbs into the slightly colder walls of the cardboard container, then evaporates into the low-dew-point surroundings on the other side. Putting some tinfoil on the inside will prevent a little radiative heat transfer, yes, but it'll also stop ALL moisture transfer. They actually use paper as the medium of choice in ERVs (energy recovery ventilators) for modern HVAC systems: paper is the divider between cold dry external air and hot moist internal air, because it allows both heat and moisture transfer while blocking the gas itself.
This is a really common point of design confusion, to the extent that every filament dryer on the market except printdry gets it totally wrong, and performance is crappy as a result. PrintDry (or a modded dehydrator) works awesome, EXCEPT it's designed for dehydrating much wetter material at much lower temperature than the plastic needs, so airflow is much higher than it could be and I believe overall it's a lot less power efficient than it could be.
I'm working on designing a better DIY dehydrator right now, and trying to compare performance to the "cut up filament box over your heated bed" technique, so I'm in the process of doing a bunch of detailed data-heavy experiments at the moment. I'll share some data when I'm done collecting it 🙂
RE: Using heated bed to dry filament
@alexw I for one would be very interested in you results.
I don't understand how filament dryers work, the wet filament is heated, the surrounding slightly wetter air is exhausted and new slightly dryer air in sucked in. Wouldn't is be better to wait until air is saturated before exhausting it?
I like your cardboard box idea perhaps with some small paper panels.
RE: Using heated bed to dry filament
Wouldn't is be better to wait until air is saturated before exhausting it?
Better in the sense of energy efficiency, but that's not really a goal of the food dehydrator.
If you had a hurricane of 140mph, 150F 3%RH air hitting a spool, that'd dry it just as well as a trickle of 1mph 150F 3%RH air hitting it, since not much moisture is coming out. But it'd cost you a crap ton more to heat and blow the 140mph air.
The food dehydrator is designed for 1) do a good job dehydrating extremely wet things like fruit and raw meat 2) be cheap 3) be reliable 4) be cleanable. With how wet meat and fruit are, you could probably dehydrate them with JUST a fan, though it wouldn't be as reliable as adding in some heat.
Imagine a food dehydrator made of cardboard. It'd fall apart instantly, and god help you cleaning raw meat juice off of it. And ditto things like adding fan speed control - that'd require at LEAST extra knobs the end user wouldn't know what to do with, and probably active closed-loop control to make sure the chamber conditions are still "dehydrating" and not "dangerously bacteria-culturing" where as "hit it hot and fast" gets the job done safely if not efficiently.
On the other hand, energy recovery ventilators: you can be virtually guaranteed that you've always got a source of dry-enough air to prevent the paper exchange membranes from getting waterlogged, and I bet soaking them in a little fungicide during manufacture takes care of any mold issue. No big deal to make them of paper.
Anyway, I'll let you know how my cardboard dehydrator experiment goes 🙂
RE: Using heated bed to dry filament
Exactly this! I've been thinking of writing a primer on humidity physics as it relates to FDM filament. Here's a short version:
- The parameter that causes water vapor to penetrate or be extracted from polymer filament is Partial Vapor Pressure of Water. If the PVPW outside of the filament is higher than the PVPW inside the filament, water vapor will seep in. If PVPW is lower outside than inside, the filament will dry out at some rate. The greater the PVPW difference inside vs. outside, the faster water will move in or out of the filament.
- PVPW is directly related to Relative Humidity. Higher RH causes water vapor to move into filament. Lower RH causes water vapor to leave filament. We use RH rather than PVPW to describe the permeability of water vapor because it is easily measured.
- Air has a higher capacity for carrying water vapor at higher temperatures, & RH is the percentage of that capacity being used at a given temperature. Therefore, if you heat a given amount of air without introducing more water, you will be lowering its RH.
- This is the principle on which hair-dryers & food-dehydrators work: pass a flow of air over some heating elements so that its temperature rises. In response, the RH of the warm air flow will be dramatically lower. Pass the warm air flow over the material to be dried & water will be removed relatively quickly. This also requires continual flow of new cool air over the heating elements, & exhaust of the warm moist air away from the material being dried.
- By comparison, dry-storage containers work more slowly at room temperature by a different principle. They require a desiccant material, usually silica-gel, that absorbs water vapor more readily than FDM filament. They are also sealed against air-exchange because continual air exchange will overload the capacity of the desiccant. This is sometimes called passive drying & takes much longer than active drying. For that reason, desiccant dry-boxes are usually relegated to storing filament that is already dry.
I think you've actually got this precisely backwards - lining the box with foil will REDUCE the dehydrating effect, regardless of whether it increases the temperature (which I bet it will).
The purpose of a dehydrator is threefold: 1) increase temperature of the air to reduce the relative humidity of that air, 2) increase temperature of the wet thing so the vapor pressure of its moisture goes up/the water evaporates more readily 3) transport that liberated moisture somewhere else.
You're focusing on 1 and 2 - get the temperature up, while negatively impacting 3. If you heat up a wet sponge in a sealed box, moisture WILL come out of it, but pretty quickly it'll saturate the air and the sponge will stop drying. The best you can hope for is that it dews up on the walls of the container, and doesn't re-evaporate from there. On the other hand, get some airflow in there to exchange the high-dew-point air with fresh low-dew-point air, and that moisture has somewhere to go, so the cycle can continue until the sponge is dry.
Cardboard is actually a GREAT material to use for a dehydrating container, because it's somewhat insulating, but also extremely moisture permeable. Rather than exchanging moist air at all, the moisture just absorbs into the slightly colder walls of the cardboard container, then evaporates into the low-dew-point surroundings on the other side. Putting some tinfoil on the inside will prevent a little radiative heat transfer, yes, but it'll also stop ALL moisture transfer. They actually use paper as the medium of choice in ERVs (energy recovery ventilators) for modern HVAC systems: paper is the divider between cold dry external air and hot moist internal air, because it allows both heat and moisture transfer while blocking the gas itself.
This is a really common point of design confusion, to the extent that every filament dryer on the market except printdry gets it totally wrong, and performance is crappy as a result. PrintDry (or a modded dehydrator) works awesome, EXCEPT it's designed for dehydrating much wetter material at much lower temperature than the plastic needs, so airflow is much higher than it could be and I believe overall it's a lot less power efficient than it could be.
I'm working on designing a better DIY dehydrator right now, and trying to compare performance to the "cut up filament box over your heated bed" technique, so I'm in the process of doing a bunch of detailed data-heavy experiments at the moment. I'll share some data when I'm done collecting it 🙂
Whatever you find to do with your hands, do with all your might!
RE: Using heated bed to dry filament
Mind sharing your g-code for that experiment?
I'm running into the 30min safety timeout if you just turn on the heater from octoprint.
RE: Using heated bed to dry filament
I haven't had time yet to write up the details, but long story short, I think I've actually proved both the theory I outlined, and that in practice, your foil-lining approach is still better anyway.
I did a run with a saturated sponge and the cardboard box, no foil, 100C bed, measuring internal and external temp/humidity. I then re-ran the experiment with the box foil lined.
In both cases, my sponge was saturated with the same mass of water, and was more or less fully dried by the end of the experiment (3h). In the foil-lined case, the ambient humidity in the box got MUCH higher, and dew formed on the foil on the inside, but the temperature ALSO got higher than in the non-foil case, by 10 or 15C IIRC. Based on the inflection point of the humidity curve, I suspect (though didn't confirm) that the foil-lined experiment actually dried the sponge out faster, suggesting the increased temperature had more of an impact than the reduced moisture ventilation.
What I can almost certainly say is that the moisture load of 30g in a sponge is WAY higher than 2-3g in "wet" filament, so the ever-so-slightly drier the air could be will certainly matter less than the increase in temperature from foil-lining.
In either case, the 5mm or so gap between the bed and the box was plenty to let enough moisture exit, even in the foil lined case, and even with a wildly overzealous moisture load to dehydrate.
RE:
alexw I'm really interested by your posts, I've been toying with buying a dryer but anything at a sensible price has reviews that say it does'nt work. I don't understand a lot of the terminology that you use but
In either case, the 5mm or so gap between the bed and the box was plenty to let enough moisture exit, even in the foil lined case, and even with a wildly overzealous moisture load to dehydrate.
surely, hot air rising and all that wouldn't it be better t0 provide for exhaust from the top? Possibly even active to provide the flow of air we appear to be looking for?
RE: Using heated bed to dry filament
Interesting concept. I think a food dehydrator is cheap enough. I will applaud you endevoir.
--------------------
Chuck H
3D Printer Review Blog
RE: Using heated bed to dry filament
alexw I'm really interested by your posts, I've been toying with buying a dryer but anything at a sensible price has reviews that say it does'nt work. I don't understand a lot of the terminology that you use but
In either case, the 5mm or so gap between the bed and the box was plenty to let enough moisture exit, even in the foil lined case, and even with a wildly overzealous moisture load to dehydrate.
surely, hot air rising and all that wouldn't it be better t0 provide for exhaust from the top? Possibly even active to provide the flow of air we appear to be looking for?
It would definitely improve airflow, but at the cost of steady state temperature. I believe my experiments show that for this low moisture load, temp is more important.
Interesting concept. I think a food dehydrator is cheap enough. I will applaud you endevoir.
I definitely agree a food dehydrator is cheap enough. I think it's also rather inefficient for the task, though, since it blows quite a bit more air than necessary for the moisture load of a few grams in a roll of filament.
I did measure the print bed at around 100W average holding 100C bed temp, ~60C in the "chamber". The dehydrator took around double that for similar temps.
RE: Using heated bed to dry filament
Just one more question. Size? is it best to keep as close as possible to the diameter of the filament spool or go large to catch as much of the heated bed as possible?
RE: Using heated bed to dry filament
Just one more question. Size? is it best to keep as close as possible to the diameter of the filament spool or go large to catch as much of the heated bed as possible?
I don't think that matters much. Larger is probably better for airflow inside and capturing more heat from the bed, but will also lose heat faster (more surface area). I was just using a cut up filament box, so quite close to spool size.
RE: Using heated bed to dry filament
You should leave some space for air circulation. That's what will actually remove the moisture, not heating alone.
Just one more question. Size? is it best to keep as close as possible to the diameter of the filament spool or go large to catch as much of the heated bed as possible?
Whatever you find to do with your hands, do with all your might!
RE: Using heated bed to dry filament
My (extensive) experiments suggest it doesn't actually matter, and not that much air exchange is actually required to get the job done in this scenario.
RE: Using heated bed to dry filament
@alexw
I saw the brief description of your extensive experiments. If I may, I'd like to respectfully point out a potential problem. I don't think a saturated sponge is a good indicator for removing water in vapor form from filament. Heating the water converts it from liquid to vapor, but doesn't remove it from the surrounding area or the filament. It's not a good comparison. Liquid water acts physically different from vapor.
My (extensive) experiments suggest it doesn't actually matter, and not that much air exchange is actually required to get the job done in this scenario.
Whatever you find to do with your hands, do with all your might!
RE: Using heated bed to dry filament
If I may, I'd like to respectfully point out a potential problem. I don't think a saturated sponge is a good indicator for removing water in vapor form from filament.
I agree that the sponge experiment is not an analog of filament dehydration. I believe it is, in fact, a much more demanding case. The fact that temperature appeared more important a process variable than peak ambient humidity for such an unrealistically high moisture load suggests that, for a moisture load 1/10 as great, it would matter even less.
With specific numbers: RH peaked 30% at 65C ambient in my "no foil on the box" test, and 45% at 76C in my "foil lined box" test. Those correspond to absolute humidities of 48g/m^3 and 108g/m^3 of water vapor according to https://www.processsensing.com/en-us/humidity-calculator/.
The filament box used is .004 cubic meters, not accounting for the spool itself taking up volume inside. In other words, the "moisture in the box" for the no-foil test was .192g, and for the foil test was .432g. Recall, I liberated 30g of water from the sponge. In other words, the water got out of the box almost completely, regardless of the lack of forced ventilation, in both cases - probably simply due to vapor pressure of water inside the box vs outside, and natural diffusion since the volume wasn't pressure-sealed.
Heating the water converts it from liquid to vapor, but doesn't remove it from the surrounding area or the filament. It's not a good comparison. Liquid water acts physically different from vapor.
This statement seems self-contradictory to me and I'm not sure how to address it. I think I've provided sufficient explanation though of 1) how the water vapor would naturally exit even without forced air exchange and 2) that, in fact, it does.