My setup for regenerating molecular sieve

Bambu Filament and Accesories Accesories
1

Hi,
I wanted to share my setup for regenerating molecular sieve. I’m still convinced, that it is far superior to silica gel as a dessicant. The main downside is, that it needs 250 - 300 °C for regenerating it. So far, I used our household oven for that, but it is really a pita to handle a 300°C hot baking tray, trying to pour all those tiny hot balls into a pot for cooling down without burning yourself and spreading the balls all over the place.

A friend told me that he is successfully using an empty paint can with a desoldering hot air station, that blows into a hole at the bottom of the can. Those stations go up to 450°C, easily sufficient for the molecular sieve. That sounded much more practical apart for the can getting very hot too. So I set out for a better isolated jar. I found that there are french press coffee makers, that are double wall stainless steel for isolation. Sounds perfect. I found a cheap thin walled stainless steal tube in the local hardware store, that is part of a spring loaded rod for hanging a shower curtain.

As I already had a desoldering station and a good multimeter for temperature monitoring, the setup was not very expensive at around 43€, including 2 high temperature thermo couple sensors.

A first dry run was ok, but isolation of the french press was not as good as I expected. At 300 °C inside, the surface reached 160°C on the outside. Far too much for being safe to touch. So I created an additional isolation sleeve from kapton tape and ceramic isolation padding, that added another 15€.

Now, outside surface temperature stays well below 50°C

Drying of the molecular sieve works great. The tube guides the hot air down to the bottom. The first sensor measures the temperature there. The second sensor is located close to the spout, which acts as outlet for the air. When that reaches about 270°C, drying is finished and I can easily pour the molecular sieve into a cooking pot where it can cool down. I then store it in wide mouth bottles.

These are the parts, my setup consists of:

PXL_20260201_150436122
PXL_20260201_1504361221920×1446 298 KB

And it looks like that when assembled:

PXL_20260201_150810590
PXL_20260201_1508105901920×1446 281 KB

3 Likes
2

How has this been going for you? Do you still like/use the solution?

3

Yes I’m using it regularly and it works great for me. I have ~ 1kg spare molecular sieve. when that is used up, I have 1kg saturated sieve , which I dry in two rounds.
Its surprising how much temperature gets lost from the hot air nozzle to the bottom of the tube, around 90°C. So I usually set it 420 °C so 330 °c reach at the bottom. When drying is finished, the top sensor settles at around 260 °C.

1 Like
4

FYI i used sieves for grad school.

What variant are you using? 3A/4A?

420 is a bit too high for 3A. Generally would recommend heating under vacuum at a lower temp. You can use a pressure gauge to identify when all the water is gone (vac drops)

Some users have found that microwaving followed by vacuum is very fast and effective as well

5

420°C is the temperature I set on the tool. Only 330°C arrives at the bottom of the tube where it enters the can.
I mostly use 13X, as this has the highest capacity but I also have some 3A. I use slightly lower temperatures for the 3A.
I know that vacuum + temperature is the best, but I haven’t seen yet a cost and space efficient solution suitable for home use. When you research online, one can get the impression that molecular sieve seems to be prohibitively difficult to dry because you need advanced lab equipment, at least a vacuum pump and chamber are the bare minimum.
That is why I posted this simpler setup, which has been working great for me. If you already have a desoldering station and a multimeter, then the invest is really low and you just have to store a coffee maker somewhere :wink:
Drying takes some time (~2 hours per load), but I achieve 1-2% below the stated capacity from the datasheet which I find very satisfactory for such a simple setup.

6

Most blast ovens can cover that range. Around $300, but general purpose, so good for drying filaments like PPS, annealing, etc. Lay the molecular sieve beads on a sieve and slide it in.

7

Sure, but if all you want is to dry your sieve, then a blast oven is really expensive and takes up a lot of space.