Repeat your experiment using just distilled water (so that you know that you don't have any electrolytic chemistry going on to speak of) and see if there is a marked difference.

 

It doesn't conduct at all with only distilled water. open circuit. 2mohm or higher resistance reading. and minimal to no useful stored charge. as if the water is not even there. like its in open air

With 4 layers worth of napkins stacked for the separator.. and not sandwiched completely flat. it doesn't break down anymore. and still has reasonable leakage. and even more capacitance.
in a sandwich bag.
Its also completely flooded with far more than "enough" electrolyte so every part of the plates is covered by electrolyte solution.
leakage current is only a couple milliamps.
Its actually got enough capacitance I get a small spark off it if I short it after charging it Don't even need to flash an LED!

Seems all I was missing was an effective and better separator/insulator material to keep the plates apart and not touching.

I imagine if I made a much larger one I could get tens of thousands of microfarads easily. or more.

Leakage is about 0.6mA or 600uA at 16v! new record.

 

I decided to try forming it both ways with -16v and +16v
After doing this for a while and switching it back and forth. it now works much much better with far lower leakage current. Almost nil leakage current!
Charged to 16v and then watching the voltage with my multimeter it takes several seconds for it to drop to 13v! and it goes down very slowly from there.
I'm not sure if the capacitance has been reduced (maybe? since both electrodes have thicker oxide layer now)
but it works so much better its surprising even.

 

Aside from personal education, is there any benefit to making such a capacitor??
There have been RF projects producing non-polarized capacitors or high voltage RF circuits, but they use plasic or glass dielectric mmaterials that work as soon as the capacitor is completed.

 

its just fun experiment project DIY for the experience and stuff.
high voltage stuff is way too scary for me and I don't have a meter capable of measuring such types of capacitors. Nor do I have an oscilloscope for that either. so i'm not making any leyden jars or anything of the sort with thousands of volts. Not for me.
the capacitance value of those is also extremely miniscule. not nearly enough to even flash an LED.

Ive managed to get my sandwich-bag aluminum foil capacitor to make a small spark when charged up to 16v even after a few seconds of being taken off of the power supply!
And for some reason putting a reverse-form charge on it to about 5v seems to drastically reduce the leakage current. Does that cause a decrease in capacitance? due to oxide layer on both electrodes or something?
How does the voltage its formed/charged up to affect how much capacitance it has? Or is that more a result of the distance between the plates or something?
I'd make one in rolled up style with scuffed up surface for more capacitance but it would be quite a thick roll. and its really tricky to do it that way.

I think stacking maybe 3 or 4 layers of paper towel would be easier for making rolled up one. Take 3 or 4 sections of paper towel and stack them for good insulation properties and minimizing dielectric breakdown. then make a rolled up capacitor out of that.
Must leave a large amount of extra paper insulator on all sides to prevent any breakdown from occurring.
Can still wrap it fairly tight with 3 or 4 layers. just prevents any possibility of the plates coming too close together or imperfections in the paper causing holes for the foil to conduct through.

Unless there's a better option for an insulator/separator material.

 

Turns out im doing it wrong this whole time. I need to form the capacitors VERY slowly. otherwise the electrodes completely annihilate and get destroyed.
SLOW and steady is the solution! The slower the better.

For tightly rolled up ones it actually worked and I got a working high capacitance value capacitor. it took forever to form at the low current needed. it holds voltage very well and does not break down when used! it made it up to 10v but took a very long time.
the higher I go the slower i need to form it.

All of my issues were from rushing the forming process!
if the forming current gets too high something changes with the way it works. and it causes instant damage and destruction of the foil electrodes

As long as I do it right. I can tell its working because when I stop charging it. the voltage decreases very slowly and gradually on its own. and does not snap down to 0.5v instantly.
if forming breaks down from too much current the voltage instantly falls to 0.5v or something very low and its ruined and need to start over.

I think the problem is too high current = explodes the oxide layer off. causing the exposed bare metal underneath to draw lots of current. blowing more material off. and repeat until destruction.
Doing it slowly at low current prevents that from happening. keeping the oxide layer intact and protecting the bare metal from being exposed.

 

Congratulations on making your electrolytic capacitor!

You might to to estimate what your capacitance is, would be interesting to know.

 

I really need to just get a capacitance tester! I should of gotten one ages ago to be honest. already have a multimeter but its one of the cheap walmart ones. no capacitance feature.
I'm not sure what the capacitance would be on this. but the strips are about 1.5 to 2 inches width and about a foot long. rolled up into a somewhat loose roll. Not tightly rolled. Just somewhat rolled up the best I could do to fit in a small cup full of water.
it really does seem to help and make a big difference forming very slowly. As slow as possible is best.

Also forming it both ways seems to drastically reduce leakage current. Not sure why it helps. but it does for some reason! Creating a bipolar capacitor.
Its not just a small difference either. its ENORMOUS
The gain/improvement is 10 fold at least in leakage current rate.

EditL update! Ive managed to get it to exceed 10v while forming and not break down! I hope to get it to at least 10v when i'm finished! that will be a new high score for most capacitance and voltage for my homemade capacitor project!

 

Is it normal for it to take more and more current the higher voltage I want to form the capacitor to?
Although its still in the milliamps range the amount required increases gradually as I want to push the voltage higher.
Am I approaching the dielectric breakdown or something?
Or is there too many impurities and stuff in my materials.
it's not too bad. about 1 second per volt drop initially. and slows down a little over time as the voltage gets lower.

its only taking 1-4mA to form at each voltage step. even less than I thought. at 13v

 

How would boric acid work? I have some of that as an option for electrolyte.
What about borax? are they similar?

Is there an ideal concentration?

 

I started over again but this time with even lower forming current. just a few microamps or something very small
The leakage current is miniscule after forming the capacitor with the very low current!
it holds voltage for minutes without discharging by itself too quickly.

 

Similar to commercially available electrolytic capacitors. The performance is proportional to the effort in producing the capacitor. More time and effort gives better performance but raises the cost.
Now I am wondering about some of those oil filled electrolytic capacitors. Or are they a totally different species? I have not studied them at all, aside from the basic data sheets.

 

You don’t need a capacitance meter to measure capacitance.
You can get an estimate of the capacitance from RC time constant using an oscilloscope or DMM.

Time Constant = R x C

Time constant is the time in seconds it takes to charge the capacitor C farads via a resistor R ohms to 63% of the charging voltage.

It is also the time it takes to discharge the capacitor through a resistor R to 37% of the initial voltage.

 

I don't have many options for resistors or anything. my meter is not very accurate.

I'm wondering whats really the ideal forming current range? is there a minimum that you need to overcome some "resistance of forming oxide" or something?
it seems like lower is always better. to a point at least. too low and the leakage current overcomes the charging/forming current and the voltage you reach is really low...

 

There is, at least in theory, some energy level required to drive a chemical reaction in a given direction. At least, that is for reactions that require energy. Those reactions that release energy are the opposite.
In the graphic form of the reaction expression that is the symbol over the arrow. Many reactions are equilibrium types, which does make it confusing.
So the short answer is "yes", below some point the reaction would not happen, at least it would not be any faster due to adding electrical energy. So there is room for experimenting in that area, if you have lots of time..
While some of those very old capacitors definitely dried out, my impression is that current production electrolytic caps are different.

 

I don't have many options for resistors or anything. my meter is not very accurate.

I'm wondering whats really the ideal forming current range? is there a minimum that you need to overcome some "resistance of forming oxide" or something?
it seems like lower is always better. to a point at least. too low and the leakage current overcomes the charging/forming current and the voltage you reach is really low...

Welcome to the world of reality. Theory can only take you so far -- at some point, it's time to shoot the engineers and begin production. Or, in this case, realize that there comes a point where you have to do the grunt work of trying lots of things to see what does and doesn't work best (or at all). You've already been doing that to some degree. If you were actually wanting to make capacitors, you would try many different chemistries and many different conditions -- different concentrations, different temperatures, different forming current profiles. You'd map out what improves and what degrades the performance and let those inform even further experiments to optimize the process. Since this is something that others have definitely done on a commercial scale, you'd likely have begun with an extensive literature search and consulting with or hiring people that already have expertise in this area. Since you are just doing this for fun and your own satisfaction/education, you can greatly abbreviate all of those.

 

I don't want to accidentally create a dangerous substance or something toxic poisonous or explosive though so I still want at least a starting point thats reasonable and safe enough.
I don't want to be working with any deadly chemicals like sulfuric acid or some other kind of chemical that gives off gas or anything.

baking soda and water is pretty safe but not very effective. it works best if you just have open sheets of foil in a bath of liquid. with no insulator separator between them. not sure why its so much harder if they are rolled up or have a paper separator between them.
The capacitance doesn't seem to change by a large amount with spacing. but the breakdown voltage gets exponentially lower (only 1-3 volts) even if you try to form the capacitor slowly. it disintegrates the foil without forming an oxide layer.

I even tried a completely neutral separator material such as several layers of fine mesh nylon netting to separate them and make sure theres nothing leaching out of the paper towels or napkins or anything ive tried to use. The problem still occurs.
How do I solve this breakdown problem if my foil sheets are stacked close or rolled up tightly.
Is water and baking soda just not suitable with the foil close together?

 

Started over and stepped up the voltage only about 200mv at a time this time. with a resistor to limit the current to only a couple milliamps. and let it fall naturally on it's own.
Forming at about 200mv at a time only takes 0.2mA or 200uA of current.
Currently its at about 4.14v
its now a rolled up capacitor made of 2 sheets of foil 2 layers of toilet tissue 4 squares. and two lengths of foil. with about a half an inch of extra space all the way around so theres no shorts. Not rolled up too tightly. but rolled up enough that its all together.
its forming slowly but steadily. current falls to maybe 100uA before raise up the voltage a bit.

Just baking soda and distilled water like before.
Goal is a 16v capacitor if I can make it that far!
I'm not quite sure how long forming "should" take.. and what current would be the ideal for forming optimally. but it seems to be working well so far!
At this rate it may take many hours or even a couple days to reach 16v. but I hope its worth it for the result in the end!
The current decreases somewhat linearly over time. As the current decreases. it slows down gradually over time. though it still decreases at an acceptable rate. not concerning.

 

@realflow300
May be this information will helpful:

Building an electrolytic capacitor
https://www.kendeil.com/Portals/0/Tech-Academy/BUILDING_AN_ELECTROLYTIC_CAPACITOR.pdf

Surface Treatment Technology for Aluminum Electrode Foils in Aluminum Electrolytic Capacitors
https://www.chemi-con.co.jp/en/faq/detail.php?id=alFoilSurfTreate

Capacitors by R. P. Deshpande
https://ia803404.us.archive.org/26/....-deshpande/Capacitors-by-R.-P.-Deshpande.pdf

ADDED:
Never use baking soda and other alkali, because they simple dissolve aluminum oxide!
Form oxide layer, using AC voltage.

What can I use to make my capacitor other than baking soda?

"For example, this is obtained through anodization in an electrolyte such as sulfuric acid, chromic acid, or oxalic acid."
(From sources, linked above). Read, please.

 

What can I use to make my capacitor other than baking soda?
How does ac work to form an oxide layer compared to dc? I thought it was similar to reforming old capacitors. does that not work for some reason with DC?
after some forming and testing at a terminal voltage of about 3.3v it only requires a current of 7 to 7.5uA to overcome leakage current. although the current is still dropping from there. so the leakage current is still decreasing and dropping even further