Sorry in advanced for the misspellings and grammar mistakes....
Heating Epsom Salt and Alum over a stove till they melt and then sticking the electrodes in makes a battery but not just any battery a AC battery. This battery outputs a small AC voltage when dried showing this could mean that its a true crystal battery. I've showed this video to others but all they care about is the galvanic batteries. This cell is not galvanic and it produces AC.
The Elmer's Glue, Salt substitute, and Epsom salt cells are still amazing cells. They can be shorted out for weeks and come back to their original voltage just fine and do more work. The one guy who replicated this cell has stated that he can run a joule thief pancake coil made by lidmotor off the cell for 4 hours and then let it rest for 1 hour and its able to do it again for 4 hours and so on.
I did find something out that is interesting, the Epsom salts contains water. Or better the Epsom salts also absorb water from the air very easily. Is this a bad thing? I don't know, because if water was their then the voltage would be higher on the cells but they're not they're lower. Also the water in the cell seems to be pushed out as the cell dries, the cell develops a wet spot around the glue on the paper. Also if water was in the cell a galvanic reaction would happened and shorting the cells out would make them weaker and the plates would corrode but they don't. I doing a test on the glue itself, I placed it on paper to let it dry and see if it develops a wet spot around it. If it doesn't develop a wet spot then the water has to come from the salts.
I did make a cell with only glue and salt substitute in it and the cell does work. The cell has a little less than normal voltage in it where normal would be 1.450 this cell has 1.400 in it. I did make a Glue and Epsom salt cell too, but its still wet unlike the salt substitute and glue cell which has fully dried and harden. This could mean that the Epsom salt cell is trying to push out the water it contained. What I find most interesting is that the glue and salt substitute cell even though its got less voltage it holds its voltage better as if it has more power in it. The glue and salt substitute cell is able to power the LCD clock just fine. This makes me want to build the salt substitute and glue cell and test it to see if it recharges itself like the other cells.
I took a copper tube and wrapped notebook paper around it and then wrapped the magnesium ribbon around that. I made a batch of distilled water which contain Epsom salt and salt substitute. Then dump the cell in that till it was good and wet and than sat the cell in a sealed container connected the LCD Clock. The clock is on at full power and working perfect the clock was set at 7:06PM. Its now 7:30 and its still working perfect. I know to the common person I just made a galvanic cell but from earlier testing I found that adding those two salts lowered the voltage which is the opposite of table salt and what you might expect. This test is just to see if the magnesium will corrode but from early test it might not or might really slowly corrode. Also from the Glue cells that contained those salts it should last a long time. I also know from doing this before that the cell should still provide power even when dry as I have do this before on smaller scale and the cell is fully dry and still giving voltage.
Stove Dry Crystal Cell
In search of a dry crystal cell (A crystal cell that did not need water added to it for it to work) I've found that mixing Alum, Borax, and salt substitute and cooking it made a good cell. Using copper and magnesium I have a really storng cell and connecting it to a joule theif like circuit helps to make it even better when lighting a LED.
I have found that a cell of only Alum cooked on the stove with the copper and magnesium doesn't produce the same result of the cell above. The power is much lower, so the borax and salt substitute are needed.
I think I might know a better way to make the cells. First place alum down and stick electrodes into it. Let alum cook til it boils or becomes a liquid and then once a liquid add the borax and salt substitute.
I've been the only one who made the dry cells. At first I was bashed for it as other contested that water was still in the crystal, but I prove them wrong with the simple experiment of sticking my probes in the crystal of Epsom salt and showing that even though it has water in the crystal the electrodes never touch the water thus no voltage. Then I was ignored, people still felt un easy about it but they slowly started to make it.
I'm working on my next idea, a true crystal battery or a literal crystal battery. I'm worried about sharing this one. These new cells have copper and aluminum instead of copper and magnesium because the magnesium would melt just sitting in the water. But I make ½ cup of boiling water and add 2 of ½ tablespoons of alum in the boiling water and stir. As the water cools and evaporates crystals are formed. But this process is taking forever. I think if I just dunk the electrodes which are wrapped in string so that water is absorbed and then hang it up it should speed things up a bit.
These literal crystal batteries are much harder than I expected they would be. These cells are hard to make and I think I'm doing it all wrong. As I'm learning is that the water is used to dissolve the salts and as the water evaporates the crystals start growing. So water leaving is the reason why you get crystal growth. My problem is that I'm not allowing the water to easy evaporate, I keep them in cups I need to put them in flat plates maybe with a black bottom and put it in the sun. The more spread out the water is the faster the evaporation. It also may help to have a fan lightly blow the flat plate of water water help with the evaporation.
So far I do have some crystal growth on the aluminum and copper electrodes and I have allows the cell to dry and it does produce a voltage. The voltage is around 250mV which is small but this tells me a lot. One thing it tells me is that this low voltage is not due to a galvanic reaction because a galvanic reaction would be closer to a volt. But the best news is that the cell is holding its voltage even with the meter hooked up to it, and as it is drying the voltage seems to be increasing too. Having the the crystal grow around the electrodes seems to be working great so far in the small test. Having 250mV holding steady is amazing in this small tests.
Using just a Epsom salt crystal may not work. The glue cell with just epsom salt didn't work well so I expect that growing the crystal around the electrodes might not work that well. It was when you doped the cells that it started working. Mixing Epsom salt , salt substitute, and Borax and see if one crystal can grow from that batch of three, that cell should power things.