I like the idea using the airbrush compressor! Knocking out initial water in the compressor reservoir to help dry the air before the desiccant will absolutely make the desiccant last longer. You can probably run higher flow rates without shortening the life of the desiccant, too. I think @NeverDie may have beaten you to the compressor idea though.
You’ll want to keep the system pressure low, though, depending on the reservoir / desiccant container. I used glass for mine and intentionally kept pressures low to not create an explosion hazard. You would want to make certain you can’t overpressure your system. I would not recommend a glass reservoir with a compressor front end. Plastic bottle all the way. With glass, not too many pounds of overpressure could send sharp glass shrapnel across a room.
Drying the air to be used in a filament dryer takes the guesswork out of drying and gives consistently low filament moisture contents. So far I’ve still only printed PLA, ASA, and PETG HF after drying but all print beautifully. Bambu discontinued their line of Silk PLA and went to Silk PLA+ apparently over difficulties people were having with it. The original Silk PLA printed beautifully after drying.
If you try drying using dry air, please report back how it goes for you. It’s a bit of bother and expense to set up for a dry air purge. The results are what matters though. If someone took away my air dryer, I’d absolutely build a replacement immediately. It’s now an integral part of my filament handling. And AMS desiccant lasts basically forever when you use “dry” filament.
I did try it with an airbrush compressor, but to my initial surprise it failed to separate out water in any meaningful way. I think the reason may be that if you’re using it to feed a lot of CFM into your filament dryer, the airbrush compressor and its storage tank tend to heat up pretty hot. Hot air holds more moisture, and so it fails for that reason. To compensate, I put a heavy blower fan on it to cool it down more, but still no improvement… So, the other factor that may have undermined it is that airbrush compressors don’t compress the air all that much, as least relative to a more typical shop compressor. And lastly–as probably biggest nail int he coffin–I was bleeding off the air at a slow rate, so the net effect was there wasn’t the big temperature drop that one gets with with a shop tool, most of which consume a lot more CFM’s. If I had run it at high CFM to get a temperature drop, then the poor little airbrush compressor would be adding so much heat that it might overcome the benefit.
So… if you can tolerate the noise and want to try it anyway, maybe go with a really big air compressor and run it at the higher CFM’s that it’s designed to handle.
Thinking about it again with a fresh mind: what might have worked would be sending pulses of high CFM air into the drying chamber rather than running at either high CFM continually or extremely low CFM to just bleed in a low-flow of air. That, and add an extra blower to cool your airbrush compressor down between pulses. It didn’t occur to me at the time, but that might conceivably work.
Anyway, the aquarium pump and desiccant stack approach still seems like the winner overall. Maybe dry the air using a dorm room freezer if you want to pre-condition it before sending it through the desiccant. Much less noise that way, but it’s a lot more complex setup, and you still need to eventually dry the desiccant anyway, just not as often. So, for that kind of benefit, just use lots more desiccant, and then you get almost the same benefit wiwthout an increase in complexity.
Technically it is a small shop compressor (2 gallon tank) that I will be using for my airbrush. I will need to choke it down a LOT to make it work fine. If I am willing to steal it from a pile of aquarium parts, I have a very nice needle valve that can do that (it was to turn 15psi of CO2 to about 4 bubbles per second)-which is probably a fraction of what the aquarium air pump does. At that rate, my 20lb CO2 tank lasts well over a year). But anyway, to speed up the project, I have ordered an inline water filter housing (2” x 10”). It is already designed for pressure, and at roughly $10 hard to make cheaper (it is also clear polycarbonate). To feed it in the meantime, found a usb powered air pump (again, less than $10), that comes with some silicone tubing. All I need yet is some 1/4” fittings for the water filter housing. I will play around with powering it via the air compressor - for research purposes I also got a set of quick disconnects so I can still hook the air compressor to it. But honestly, the USB air pump will be much easier to deal with than a 37lb compressor.
If it would truly last that long, then I think you may be on to something. How much does it cost to fill your co2 tank?
According to google filling a scuba tank costs $7-12. Offsetting the hassle would be getting extremely dry air… It’s a more complex setup, but one could make an argument for it. Out of laziness though I think I might still gravitate to the aquarium pump plus desiccant solution, if only because 1. it’s so darn easy and 2. based on current experience seems to work well enough.
I’d encourage someone to try the tank approach though and report back, especially if they already had the equipment. The reason: the science is sound and sometimes hidden benefits don’t become apparent unless/until you actually try something, and so far nobody has tried it. For instance, one might expect the drying to go faster if the make-up air is bone dry. How much faster? Possibly somebody smart could calculate that answer using only theory, but it might just be easier to simply try it and then know for sure.
Also, drying a highly hygroscopic filament like nylon might benefit from the bone dry air. How much more benefit? Again, hard to know without doing the experiment. If the incremental benefit is big enough, then $12 per year to fill a tank is peanuts, especially if its an expensive filament and if the extra drying avoids even one botched print. At least, that’s how I imagine an “in favor of” argument might go. For instance, PPS filament can easily cost >$200 for kilogram, so spending $12 to greatly reduce the odds of a print failure would be easy to justify.
Do you mean to shorten drying time or to improve the eventual dryness you end up with? Water is the essential plasticizer in nylon, so too dry is definitely a thing.
Anyone using high pressure sources, please be careful.
Using a needle valve, the pressure can still increase to whatever the upstream pressure is. That upstream pressure needs to be kept at safe levels with a regulator which I’m sure is involved, but it needs to be set low enough that a blockage of any kind doesn’t cause pressures to exceed whatever stuff in line can handle.
With higher pressure sources, columns that can handle those kinds of pressures are important. Definitely no glass unless specifically designed for the possible pressures.
I am really enjoying that, with PETG HF, I don’t have to worry about outer surface gloss levels and that I don’t have to slow the outer walls right down for a part to look consistent top to bottom.
And the colour selection is great.
However, I’m finding drying very tricky.
The first 2 rolls of black, and 6 other colours I tried - I dried for about 6 hours at 65 deg C and they printed perfectly.
Then I thought: great, I’ll order 11 more rolls of black for my project.
Opened 3 from the new batch, dried them 8 hours each, they all printed with bubbles/zits.
Dried them another 8 hours each, this time at 70 deg C, less bubbles.
Dried them yet another 8 hours each, they seemed to print OK, but as the print continued, about 100 grams in, bubbles appear again.
I check that the dryers are reaching temp with a thermal gun, looks OK.
Dried them yet another 8 hours each, same issue - printing OK but bubbles appear.
Dried them yet another 8 hours each, printed OK but there wasn’t much left on the roll.
I’m finding that it is very difficult to dry the filament near the middle, especially on a particularly damp batch. If I dried these spools for 40 hours from the beginning, I think the inside of the roll still wouldn’t print right. 7 more spools in black to go from this batch, I hope the other colours don’t have this issue.
I’m in the same boat, although only 6 rolls of PETG-CF stock. As a test, I’ve got one opened, half-consumed and re-packaged in vacbag with desiccant for about 2 months. UK ambient T & RH.
Gonna see if it’s useable after drying in my little FixDry for say 10hrs at 65Deg C.
Successfully printing new spools from the FixDry, which seems good enough for this and to maintain it for a couple of weeks.
For drying, the BBL TDS states 65Deg C for 8 hrs in a “blast oven”. That’s pretty onerous to dry the material properly. I might try buy a cheap second hand table top grill / half-oven with fans, but not sure it’s worth all the hassle as the printed material properties are not as good as basic PETG.
It’s been covered extensively on the filament drying threads, so I’m not going to repeat it all here. However, the short of it is: if you’re using time rather than weight to judge how dry it is, you’re doing it wrong. You need to weigh it multiple times throughout the drying process, and it’s not dry until it asymptotes.
And, yes, the humidity in the make-up air makes a difference, which is why the only right way to do it is with desiccated air. The evidence is so clear that I’m not even going to waste time debating it anymore. If you haven’t tried it, I think you’ll be amazed at what a difference it makes. I hear lots of people who say they don’t need it and refuse to try it, but I have yet to hear anyone say they tried it but have since decided it wasn’t worth it. i.e. everyone who tries it is sold on it.
The one short-cut that has emerged was when @MZip noticed that when drying the same type of filament he could identify the RH% at which it was dry. That makes it easier by saving you from having to weigh it repeatedly.
Desiccated air won’t make a meaningful difference when trying to get super damp PETG just to a usable state.
I’m not trying to get the last gram of moisture out of Nylon, it’s trying to move the saturated air from inside the spool out of the spool. Through the tight windings then through the tiny holes on the sides of the spool. The outer layer dries fine.
Also it’s 0 degrees C out where I am, the absolute humidity is so low that when the air is heated to 65, it’s less than 5% relative humidity. Granted the absolute humidity is higher inside, still doesn’t make much difference.
When I dry a spool, it’s not unusual for it to take literally days before it finally asymptotes. Why so long? To your point, maybe it’s the tight windings. Some people have proposed that maybe a hard vacuum could help suck it out of the crevices to speed things along, and I think it’s worth trying to see whether it does. But then again, if I’m understanding the Law of Partial Pressures, dry enough make-up air should be just as good as a vacuum, right?
Fair enough. And the rest of the year… what do you do then? Or do you time it to do batch drying during the winter?
As sure as clockwork somebody who lives in the Atacama desert who will pop up to say they don’t need it…
and I believe it. So one explanation would be that the people who try it are the ones who are desperate enough that they really do need it, and that’s why they are subsequently sold on it. If that’s the case, I’m certainly at peace with it. Eventually, life circumstance will sort you out one way or the other.
It gets complicated but I would argue desiccated air does indeed make a meaningful difference. The lower the humidity in the filament dryer, the faster and more thoroughly filament will dry.
What I’ve found is I can make my desiccant last longer by opening the filament dryer door with a prop until the RH in the drying chamber starts coming down. Then the dry air makes a big difference. During the initial spike you get when the spool is heating up, dry air is not as much a factor just because it gets swamped from all the water.
I use desiccated air and my filament dries to my target RH% in the dryer much faster than without dry air. Without dry air it may never reach my target RH. When I was doing my tests, ambient humidity was fairly high (around 40-50%) and filament would asymptote out and stop losing moisture before it was dry enough to print trouble free.
And about the moisture getting out of the center of a spool, the windings do slow the water movement some but not as much as you might expect. Even with a perfectly wrapped spool, the surface defects in the filament - cracks, pits, etc - are huge compared to the size of a water molecule. Millions of times bigger. Water molecules are 18pm on the long axis. Defects in filament are measured in microns if you check the micrographs.
Also, have you verified you are actually drying the filament by weight? If you’re seeing water effects then it’s a safe bet you have a fair amount of water. After drying, if you aren’t down by 2 to maybe 3 grams of water, your drying may be inefficient.
You can test water content in a spool by sealing it in a gallon ziplock freezer bag or poly cereal box with a hygrometer. The spool moisture will equilibrate with the air in the container and after 10-12 hours the RH you read will give you an idea of how much water content you have. It’s not the moisture content but is related to moisture content and that’s why you can get a go/no go idea. All of my spools peg the hygrometers in their poly boxes at the lowest reading they can display - 10%. They are cheap round hygrometers from Amazon but they give consistent results. I tried a spool of PETG HF straight from the shipping bag and the reading was like 42% RH IIRC. After drying it printed great and the reading was 10% when I stuck it in a poly container.
And looking at your other post, you’re going for multiple 8 hour drys and still getting poor results. I go 8 hours or so with dry air and get great results. I run with stock settings from Bambu and no moisture problems.
You really might be surprised what dry air can do for drying especially if you are having issues. Not everyone needs dry air but your experience seems to say you would.
It’s easy and cheap to try it out. A medium-sized aquarium pump, a container of desiccant beads, and some tubing and fittings and you can cobble up a test. You need to get the dry air into your filament dryer and it needs to be sealed up well enough to get the dry air to the filament. Sounds odd but the Sunlu S2 has a base held on by two screws that has a nice gap all the way around it so I used silicone seal to close the gap.
Also, dry air is more dense than humid so you want to bring the dry air in at the bottom. I think most dryers have holes near their tops to pass filament through but they are great for letting the moist air out.
I tried just using the aquarium pump with no desiccant in the line and it dried not that different from propping the door open. After the weights stopped changing (IIRC in the 30% RH range in the dryer), I put the desiccant back in line and RH started dropping again. I now just dry everything to a displayed RH of 19%. It works well for my printing. And that’s 19% RH hot - using the S2 RH display. When the spools cool down in the poly containers, the hygrometers always peg at 10% on their display.
The whole basis for the crazy things I say are in the thread “filament drying: preliminary results”. We covered a lot of stuff and there were some side discussions along the way but you can see how this was tested and verified and the actual numbers I may not remember so well. And big thanks to @NeverDie, @johnfcooley, and @IslandBill for their contributions.
Did you find that the magic RH% for indicating end-of-drying was general for all PLA’s, or differed by brand or even different models within the same brand? And at the model level, was it a different RH% for different filament colors, or the same RH% for all colors of that filament model? IIRC, you processed a lot of spools, but maybe it’s still too soon to generalize. If so, whatever distinctions you’ve noticed thus far would be interesting.
I started looking into that with PETG HF but using my setup and sensors, when I tried stopping at 24% RH I still had slight water issues so abandoned it and decided I’d just dry everything to 19% indicated.
The plan was to also try stopping at 24-25% with PLA since it’s a little more forgiving, but never tested it. I had gotten what I wanted so just moved on. I’d bet I don’t need to go to 19% with PLA but looks like it might be needed with PETG HF. But not enough info to really say.
Sorry not more helpful on that. I think it could save lots of time and make desiccant last longer though.
Have dried about 50 spools now and drying to 19% on almost all of them, that has been sufficient across the recently discontinued Bambu silks, PLA, matte PLA, PETG HF, and ASA.
Hi MZip and NeverDie, thanks for sharing your experience, and I will with-hold my reservations about how worth-while it is going to desiccated air, and I see that once you have the equipement set up, it’s not a big deal to run it.
If I were to do it, I would do an initial dry to get most of the moisture out, use a double dryer and put a spool in, and use the other space to put a massive load of desiccant in there in a container with adequate airflow, tape up the holes, and just let it recirculate via the internal fan - no tubes or pumps. Thoughts?
The intention of my previous previous post was more to point out the big batch variation in Bambu PETG HF - as most of my PETG HF prints perfect after a few hours of drying (maybe they were dry enough when I got them), but the last shipment of PETG HF black is just difficult to dry (with conventional methods).
It sounds good but the way you dry desiccant is to heat it. With desiccant in the filament dryer, it can’t dry the air as well because it gets heated too. It will work but you have to keep the desiccant especially dry.
Since it will become less effective in the dryer as it picks up water you just need to keep it fresh. I haven’t seen any numbers to say how much desiccant will dry how many spools though. That would still need to be explored.
The other thing is all the desiccant will be exposed to water at the same time and all will get the same moisture content. This kind of dovetails with the paragraph above but the consequence is the maximum level of dry you can reach will get wetter and wetter with each spool. The only point of the graph is that the more water the silica gel picks up, it raises the ultimate humidity level you can reach.
But it may be practical. You can calculate how it will work by whatever weight of beads you use and figure about 2 or maybe 3 grams of water weight if it’s really wet per spool.
I didn’t run those calculations but they could be favorable. The advantage or not is set by your ambient humidity. The method you propose only subjects the desiccant to the water carried by the filament. The way I do it, the desiccant sees the water in the air at ambient humidity. Just doing rough calculations I think they are probably fairly close depending on how much water you can tolerate in your beads at temperature and still hit the level of dryness you need to hit. For reference, with 800g of Wise Dry orange beads, I can dry 30+ spools of filament before needing to change it out and regenerate it. More if I don’t mind waiting a little longer as breakthrough starts interfering with drying times.
The advantage of a drying column is the gradient that sets up in it. The water mostly gets dumped at the inlet leaving still virgin beads at the outlet and a range of water levels in between. What that does is keep drying the air more and more as it moves to the exit and even though the column may be old and mostly used up, the dry beads near the exit can still take moisture out. It’s a cool look if you have indicating beads.
If you run into trouble with the desiccant in the filament dryer (and that may work well) a desiccant column could be placed outside your dryer to keep the desiccant cool and use a pump to recirculate air through the dryer and column. That gives you more water capacity and the benefit of the gradient for lowest exit air humidity.
You are totally right. It may work but it’s just not the proper way. I’ll comment on one of the other filament drying threads if I go down this rabbit hole, not to hijack the PETG HF thread
