The degree to which it evaporates will be a function of the difference between ambient humidity in the container, and it's equilibrium value of 43.5%RH, and also the size of the container. In my cabinet, which gets opened 2 or 4 times in a day, I've topped off the container water 3 times since I began using it in October. At no time was there a total loss of fluid. I'd estimate I replaced about 150mL of fresh water for a 6 week average interval. Each time I've added water, I've used our reverse osmosis tap, and I've knocked salt concretions off the filter with it, back down into the bucket. Each time I've done that, the hygrometer reading has risen to slightly over equilibrium for a few days afterward and then normalized.
I would be shocked, and maybe a little scared if anything could grow in that bucket. It's intensely hyperosmolar and more than a little alkaline. In fact, when I got the humidifier filter that I put in there to facilitate evaporation, I got one that had a mesh of thin metal, probably aluminum, around the outside. Well... THAT'S GONE!!... it dissolved!!! I presume I now have an additional salt of aluminum oxide in there. It doesn't appear to have changed the equilibrium value on the hygrometer so I haven't replaced the solution, but I would say it underscores that the harmlessness of "food grade" potassium carbonate is relative. This amount/concentration is probably a little on the corrosive side and I wash my hands after handling it. I do plan to replace the solution at some point, but for now it's educational to just watch the hygrometer and add some water now and then when the level falls noticeably.
Making up the solution at the front end deserves some mention, too.
Before I undertook this project, I did some experiments with different hygrometers, closed containers, and salt solutions. I wanted to learn about how the hygrometers worked, and what the response times were for those meters and for the salt solutions. I tried table salt, potassium carbonate, and pure water. I also got a "calibration kit" for hygrometers which is supposed to be a 75% standard (which is the equilibrium value for table salt, so save your money). Each time I made the "saturated solution" by putting a mound of salt in a little bowl, then moistening it with enough water to make it all look wet, but not enough water that the salt could dissolve. Each of these were MUCH more salt than water. I'd then put the hygrometer and the salt solution into a mason jar and watch it for a couple days. I tried a dial-based hygrometer, a couple digitals by Thermo-Pro and a Caliber IV. All worked just fine. Interestingly the dial worked pretty fast compared to my expectations for a cheapo thingie, and the accuracy wasn't horrible compared to the others. They're all within a couple of percentage points in the range I'm using them for. The dial was more off at 75% than the others, and still more off at 100% which is probably more important to know for cigars than guitars since we're not pushing those levels.
An interesting finding from that experience, though, was mixing K2CO3. It gets gummy and warm when wet. I figured that to make up a large batch of solution capable of keeping a whole cabinet appropriately humidified, I was going to need to use the whole kilo that I had purchased and make a maximal batch of solution.
I have a few lab supplies at home already, (saltwater reef tank supplies) and I used a 1 liter Pyrex Erlenmeyer flask and a magnetic mixer to dissolve the K2CO3. I started with about 400mL of water, and slowly added scoops of K2C03 while stirring. It got quite warm during dissolution, not enough to be uncomfortable to touch but probably the upper limits of comfortable. The literature said that the solubility was something around 115g/100mL, but unfortunately that's not exactly the number that I needed. This number tells you that 100mL of saturated solution is DENSE. 100 mL contains 115 grams of salt *AND* some water. How much water? ... no clue, but it's not 100mL because that 115 grams of salt takes up a fair amount of space. The number I really wanted was "how much water can I dissolve a kilo of K2CO3 into, in order to achieve a saturated solution?"
I would say the answer is only slightly more than 400mL.
I kept adding salt, creating a cloudy suspension of particles in water, and then letting it stir till it dissolved/cleared, repeating and repeating, letting it cool a bit between additions. Finally, when almost the whole kilo was in there, it wouldn't clear with mixing. At that point I had almost a liter of solution. The flask was quite full. I let it settle and there was some undissolved salt at the bottom proving that it was a saturated solution. I then decanted it into my bucket and threw the remaining grams of powder in for good measure.
So I got lucky but I almost didn't. If I had started with much more water, I would have been able to dissolve the whole kilogram of salt and still not achieve saturation. So, for my next batch, that's going to be my ratio again. I'm going to slowly mix a kilogram of K2CO3 into 400mL of water. I'm going to use a container that should hold substantially more than a liter and is heat resistant. I'll probably order a dedicated 2 liter Pyrex for the job, which may also become the actual holding tank. I may also omit the use of the evaporative filter because I don't think evaporation needs to happen quickly and I'm not certain how much salt is getting sequestered.
... sorry about the nerd-out. It's probably a doctor thing.
"If I have seen further it is by standing on the shoulders of giants." -Sir Isaac Newton
Armin Hanika 56PF