Target humidity?  

Re: Target humidity?

I put a little battery powered device based on the htu21d sensor in my dryboxes. This sensor is able to measure very low humidity level, unlike a lot of them. I use ~250g of 4A molecular sieves for a 20qt box, and I replace the dessicant once the humidity raises above the minimum value that the sensor can detect, so, basically 0.

It's probably not necessary to keep the humidity *that* low, but I get around 2-3 months before I have to replace the dessicant, which seems pretty reasonable to me, so I figure I might as well try to keep it as dry as possible.

I'm a big fan of the molecular sieves over something like silica gel. It adsorbs moisure much faster, and keeps the moisture at a lower level for longer. It can't adsorb as much moisture at full saturation, but I always replace them before they get to full saturation anyway. Plus, it's also easy to regenerate, just throw it in the oven at 350F for a couple of hours. I never had much luck regenerating silica gel.

I can get a 1 gallon pail of these on amazon for like $40.

Re: Target humidity?

Anything above 0 will allow filament to absorb moisture (technically speaking). If you can't get a good dry box below 19% then your filament is probably on the wet side and is taking a while to dry. In addition to letting the dry box go as usual, throw in some new large desiccant packs to give the dry box a boost to see if you can get a lower number.

Also - always the chance a real dehumidifier will work better: At $50, might be worth considering finding a way to use something like this. Thermo-Electric based and parts could be scuttled and built into a custom built dryer enclosure.
https://www.amazon.com/Pro-Breeze-Electric-Dehumidifier-Portable/dp/B01DC5PPWM

ps: Other molecular sieves are used as desiccants; examples include activated charcoal and silica gel.

Re: Target humidity?

Thanks for the info.

Currently I'm using a "sealed" box with one of these in it: https://www.amazon.com/gp/product/B000H0XFCS

I haven't had to regenerate it yet.

It could be that the box isn't as sealed as it is supposed to be, but when there have been big swings in the ambient humidity (been running in the 25-30% range most of the winter, but we could open the house for a couple days last week and it jumped to 50%) the box has stayed stable at 19%. After I've been in the box for a spool it will read higher but drops back down to 19% pretty quick, so I know the desiccant is doing something 😉

The sensor I'm using was a cheap pack of 4. While they all seem to be in sync, they are off a few degrees and percent from other units I have (so I say 19%, but it could be as low as 16%). I'll look into getting something a bit more trustworthy to see if maybe I'm hitting some undocumented limit of the devices I have.

Re: Target humidity?

Yeah, I tried a few of those cheap humidity sensors, but didn't have much luck getting them to work in very low humidity environments. Most will only report down to 10% or so, and I didn't find them to be particularly accurate at such low humidities. I think they're primarily intended for reporting normal levels of ambient humidity, not for a dry box, where the goal is to get the humidity as low as possible.

I ended up making my own network-enabled sensor with a cheap esp32 board, an 3200mAH 18650 battery, and a htu21d sensor. I got the components from aliexpress, for a total of around $12 or so per sensor. Using esp's proprietary "espnow" AP-less protocol, I'm getting something like 6 month battery life on them, taking and reporting a measurement once every 10 minutes. If you go this route, Just be sure to get an ESP32 board with a low standby current. They vary quite a bit in this regard.

I would still definitely recommend trying some molecular sieves though. That made a big difference for me, in being able to get the humidity down and keep in down. e.g. https://www.amazon.com/Molecular-Sieve-4A-Gal-Lb/dp/B06XGQJ5N2 - although I'm sure you can find it cheaper in smaller quantities. If you use 250g of dessicant per drybox, the 1 gallon pail is enough for 10 boxes, which sounds like way more than you need 🙂

Re: Target humidity?

You can find some info on the performance of various types of dessicants at https://www.sorbentsystems.com/desiccants_charts.html . In particular, take a look at the table they have, and take note of the "Adsorptive Capacity at low H20 Concentrations" and "Rate of Adsorption" categories.

Re: Target humidity?

Being in high humidity deep south coastal USA, I quickly learned that humidity management is key to successful printing. It is frequently 60 - 80% humidity inside and I have to dry all my filaments and store in boxes with silica desiccant after a day of printing. For example, I just cracked open a new box of Amazon Basics PETG today and printed all day with it. The last prints exhibit noticeably more stringing than the first prints. When done for the day, I will dry in the filament dryer overnight and put in a small container with desiccant packs. I have multiple humidity indicators, electronic and the cards with color indicators on it. These are not scientific by any means, but they get the job done good enough.

I find that after drying in a filament dryer and enclosed in a container with desiccant, it indicates between 10 - 20 % indicated humidity while stored which provides optimal printing the next time I need to use. I have stored some filament for months like this and they are good to print with the next time (prusament with printer was stored this way for a few months - like 3) while I was using PETG and it was fine the next time I used it. I did dry it after use and returned to the sealed desiccant box.

My vote is anything less than 30% humidity storage will get you good results if the filament is dried before storage. Desiccant will not do a good job of removing moisture, it just maintains it at a constant level. A filament dryer is $30 on amazon and a godsend to the humidity challenged....

I use ~250g of 4A molecular sieves for a 20qt box, and I replace the dessicant once the humidity raises above the minimum value that the sensor can detect, so, basically 0.
....
I'm a big fan of the molecular sieves over something like silica gel. It adsorbs moisure much faster, and keeps the moisture at a lower level for longer. It can't adsorb as much moisture at full saturation, but I always replace them before they get to full saturation anyway. Plus, it's also easy to regenerate, just throw it in the oven at 350F for a couple of hours. I never had much luck regenerating silica gel.

I like your suggestion. Is there anything to be concerned with health-wise or other with molecular sieves? I read an MSDS on one I Googled after your post and it seems pretty inert and safe but I have no experience with them so I would like to know more about them....

Re: Target humidity?

I'm no expert, but there's no significant health concerns I'm aware of. Just avoid creating/breathing dust, for the most part.

You can actually dry filament with the molecular sieves, but it's significantly slower than with a filament drier. I always have to dry PETG filament before using it. Even new filament from a sealed bag with a silica gel pack is too wet, at least for my printer with a large nozzle (1.2mm). It will still bubble like crazy. So I'll take the filament out of the sealed bag and put it in a drybox, and the humidity in the drybox will raise a bit, and then slowly lower back down to the minimum value after about 3 days or so. Or, if I need it more urgently than that, I throw it in a cheap food dehydrator that I modified for filament drying.

I would recommend keeping your filament in a drybox while printing. I added a hole to one of my dryboxes with some plastic parts to hold a ptfe tube in place, place it next to my printer and feed the filament through the ptfe tube to the printer.

With the extra holes, I do have to replace the dessicant in that box more frequently, but I'm also opening it a lot to switch out filaments, etc. which also lets in a new batch of moisture each time.

Re: Target humidity?

There is no magic to molecular sieves. Silica Gel, even dry charcoal, are molecular sieves, too. The fancy name just means the method by which water is removed from the air. Some materials are better sieves than others. It's my experience the types that combine with the moisture and dissolve generally are a bit better at fast moisture removal (one brand is Dri-z-Air); but they are more expensive over time because they need to be replaced. Sieves like silica gels that absorb water - and can be regenerated with heat - loose efficacy over time as they absorb (less affected surface area), so a large enough 'sink' of gel is needed for the volume to be controlled over a period of time.

I use these, one pouch per spool. And they can be heated and refreshed. At the time, I paid $12 for a box of 10 packs...
https://www.amazon.com/gp/product/B002V3CHKA

And here's one of the soluble varieties.
http://www.drizair.com/

Re: Target humidity?

"Molecular sieve" is the name commonly given to synthetic zeolite, which has some advantageous properties over other desiccants for this particular application. It's not particularly relevant that other desiccants could be considered molecular sieves, because they're not commonly referred to as molecular sieves.

In particular, If you're looking to source this particular type of desiccant, "molecular sieve" is likely the best search term. e.g. if I search amazon for "synthetic zeolite", or even "zeolite", I get a few relevant items, but a lot of irrelevant stuff too. If I search for "molecular sieve", I get many more relevant results.

And if you want to talk technicalities, silica gel, and other similar desiccants don't absorb moisture, they adsorb it.

Regardless of any of the above, my point is that the material commonly referred to as "molecular sieve", also known as "synthetic zeolite", among other names, works as a better desiccant than the material known as "silica gel" for keeping a dry box dry. It adsorbs moisture from the air more quickly, and it can adsorb more moisture at low humidity levels than silica gel.

No, there is no "magic". It just has better properties for this particular application.

Re: Target humidity?

Nothing like marketing hype... but hey, it works. Did you know rocks also qualify as molecular sieves? They "adsorb" water over time just like the other materials I've mentioned. Wait, zeolite is a rock! 😯

Your choice of zeolite is also great for plants. It stores moisture for those dry spells. You can buy it at places like Home Depot for a lot less ($1 a pound) than what you pay for your special "molecular sieve" material. All you need to do is to dry it before using it as a desiccant.

Re: Target humidity?

Yes, if "marketing hype" equals quantifiable data about the performance characteristics of a particular dessicant, I do love that ;).

If you have the performance data for some of that "Horse Stall Refresher" from home depot, I'd love to take a look. I certainly wouldn't rule out the possibility that it would have similar performance characteristics. But often, you get what you pay for 😉

Re: Target humidity?

Just having fun now. How does a company make rock? One possibility is they crush it and them form nuggets from the powders, like sintering, and glues are probably too expensive and would defeat the purpose; but I suspect they do no such thing. More than likely the folks selling "premium" zeolite are managing diameters to optimize surface area for their specific application. Much like folk selling diatomaceous earth as filter media. On the other end, silicas can be sized to be better at initial draw down or to preserve dry performance. One chart can't really compare all the facts. That's be like saying one printer is better than another with one chart.

Re: Target humidity?

Just having fun now. How does a company make rock? One possibility is they crush it and them form nuggets from the powders, like sintering, and glues are probably too expensive and would defeat the purpose; but I suspect they do no such thing. More than likely the folks selling "premium" zeolite are managing diameters to optimize surface area for their specific application. Much like folk selling diatomaceous earth as filter media. On the other end, silicas can be sized to be better at initial draw down or to preserve dry performance. One chart can't really compare all the facts. That's be like saying one printer is better than another with one chart.

Well, from the zeolite wikipedia page:
There are over 200 synthetic zeolites that have been synthesized by a process of slow crystallization of a silica-alumina gel in the presence of alkalis and organic templates. Many more such structures could theoretically be made.[14] In addition to variations in structures, zeolites can also be made with a variety of other atoms in them to make them chemically interesting and active. Some examples of the so-called heteroatoms that have been incorporated include germanium, iron, gallium, boron, zinc, tin, and titanium.[15] One of the important processes used to carry out zeolite synthesis is sol-gel processing. The product properties depend on reaction mixture composition, pH of the system, operating temperature, pre-reaction 'seeding' time, reaction time as well as the templates used. In sol-gel process, other elements (metals, metal oxides) can be easily incorporated. The silicalite sol formed by the hydrothermal method is very stable. The ease of scaling up this process makes it a favorite route for zeolite synthesis.

Re: Target humidity?

Sure, but I actually saw a noticeable difference when I switched from silica gel to these 4A molecular sieves that I'm using now. The silica gel always seemed to have a hard time keeping the humidity down at the minimum value my sensors can report. Which is nominally 0% relative humidity, but I have a hard time believing that. Probably a few percent or so. While the molecular sieves keep it pegged at that minimum value for months. Only rising briefly if I open it to take out/put in filament, or rising by a few percentage points for a few days if I add in a new roll that hasn't been dried yet.

I am a bit curious how the much cheaper "natural" zeolite would perform though. Maybe I'll do some experiments at some point :). Although, I think I prefer the larger bead form of the molecular sieves over the smaller sand-like consistency of the natural zeolite that I saw. I imagine it's easier to deal with the beads - less dust, easier to handle and move around when drying them, easier to keep them in a perforated contained for better airflow, etc.

Your objections seemed to mostly be about the cost, but it's actually not that much more expensive than silica gel. And if you compare it on a "per roll of filament" basis, it's actually somewhat better than the packs of dessicant that you had mentioned you use. At $12 for a pack of 10, that's $1.2 per roll. While the 6 lb pail of molecular sieves I had bought is enough for 10 20 quart containers, which hold around 5 rolls of filament each, so 50 rolls total. I think I paid around $37 at the time, although it looks like they're up to $45 now. So that's a bit less than $1 per roll in either case.

My other complaint with the silica gel in particular is that they seem to be a bit tricky to regenerate. I have to admit I only tried a few times, but they never survived my attempts at regeneration. The molecular sieves are definitely more robust mechanically. Less prone to breaking/shattering, and I never had any issues regenerating them. Although they do require higher temperatures, and thus a higher cost, to regenerate. I'm honestly not sure how much gas is required to keep my oven at 350F for a few hours though. I guess I should watch my gas meter when I regenerate them sometime.

Re: Target humidity?

Well, from the zeolite wikipedia page:
There are over 200 synthetic zeolites that have been synthesized by a process of slow crystallization [snip] The ease of scaling up this process makes it a favorite route for zeolite synthesis.

Shame on you for bringing Wiki Facts into a discussion. And yes, people make diamonds these days, too.

On the serious side, silicas and zeolites all share one thing: they can be optimized for various tasks. It would be nice to find reliable data that portrays the efficacy of these various materials. Something that actually does a fair job removing moisture at low humidity may be worth a few extra bucks. Then again, not letting the filament get too wet in the first place is as effective as a perfect dewetter, and a lot cheaper.

Re: Target humidity?

My vote is anything less than 30% humidity storage will get you good results if the filament is dried before storage.

Agree. A peltier/thermoelectric-style dehumidifier is easy to stick into a plastic filament storage bucket or into a chamber of some sort. Mine sits between 20-25%, consistently below 30%. I have a ESP32s with humidity sensors in the relevant locations.

Re: Target humidity?

I'm honestly not sure how much gas is required to keep my oven at 350F for a few hours though.

350f is a bit on the "way too hot" side for drying desiccants. Try something closer to 225f. The idea is to vaporize the H2O, not the material you are drying. A convection oven does a better job, too. Simple radiant ovens don't recirculate the air enough, my thinking is moisture laden air tends to clump (a pressure density thing - think cloud forming over a lake or mountain peak).

ps: Suggesting 350f would be a cool marketing trick to convince people one product is superior: especially if the competitor melts at 325... lol. I did once find a product where the packets were sealed with a hot glue; even 225f was too hot. Then there's the notion zeolite isn't really a sexy name, you see it when shopping for plant soil right along side steer manure (eeww), and molecular sieve sounds so much more scientific and is definitely worth more than some innocuous igneous rock. :geek:

Re: Target humidity?

It depends on the dessicant. Molecular sieves "hold onto" the water much more strongly than something like silica gel. And they also have a much higher resistance to heat. 350F is actually on the low side for them. See, e.g. https://www.bio.umass.edu/microscopy/Molecular_Sieve_Drying.html , where they recommend 250C (480F).

Also see the last graph at https://www.sorbentsystems.com/desiccants_charts.html . 350F is right around the "bend" in the equilibrium capacity for these, where heating further gives diminishing returns. At 225F they would still retain a significant amount of moisture.

Re: Target humidity?

But yes, if I had access to a convection oven, I would definitely use that instead 🙂