Homemade aluminium foil capacitor - why doesn't it hold a charge ?

Thread Starter

Sally S

Joined Jun 29, 2017
18
Hi there,

I'm interested by capacitors and super capacitors and so i decided to make my own.
This is the textbook, school science project variety of capacitor, with A5 sheets of aluminium foil
separated by photocopy paper (as the dielectric).

This has been interesting. I've been able to get my capacitor to slowly release it's
charge over a period of 70 seconds (after the battery is disconnected)
before the display on my multi-meter registers zero volts.

Commercially manufactured capacitors can hold a charge practically indefinitely.
What i would like to know is:
Why does my homemade capacitor not hold a charge:
(It starts to discharge the very moment i disconnect the battery).

Both (aluminium) sheets are surrounded on both sides by paper.
It's not clear to me how, or why, my capacitor is discharging.

Thank you to anyone who can offer some suggestions or solutions

Sally :)
 

crutschow

Joined Mar 14, 2008
27,945
If you have the multimeter connected, then the capacitor is discharging through the multimeter input resistance (typically 10 megohms).

If you want to monitor the voltage without significantly discharging the capacitor, you will need a very high input impedance buffer circuit, such as the plus input of a MOSFET input op amp connected as a gain-of-one buffer circuit, or connected to the gate of a MOSFET source follower.
 

Thread Starter

Sally S

Joined Jun 29, 2017
18
If you have the multimeter connected, then the capacitor is discharging through the multimeter input resistance (typically 10 megohms).

If you want to monitor the voltage without significantly discharging the capacitor, you will need a very high input impedance buffer circuit, such as the plus input of a MOSFET input op amp connected as a gain-of-one buffer circuit, or connected to the gate of a MOSFET source follower.
Interesting.... 10 Mega Ohms...
Thank you crutschow :)
 
@Alec_t nailed it. Paper is very much affected by moisture. In the lab, I set up, I was capable of measuring currents to about 2E-12 Amps (2 pA) and applied voltages to +-100 Volts DC in an environmental chamber that had temperature control from -80 C to 200 C. Humidity was not instrumented, but could be reduced using a dry Nitrogen purge.

One day, I was bored, I actually measured the resistance of a piece of paper. Humidity definitely affects it.

You should try Kapton as your dielectric. http://www.ebay.com/itm/2-Square-Fe...hash=item4b1a24f8d5:m:mZ-2wFwCd9WntVjfPgMzzCg
 

MrAl

Joined Jun 17, 2014
8,504
Hi,

Yes dielectric and SIZE itself are two important aspects.

What you need to do is consult with some simple capacitor formulas to estimate the size of your capacitor, which is measured in Farads.

For example, if you have a 1 Farad cap charged up to 10 volts and with a 1 megohm meter connected to it, the time to discharge down to about 4 volts is about 278 hours. If you have a 1 microfarad cap charged to 10v and a 1 megohm meter, the time to discharge down to about 4 volts is just 1 second. So we went from more than 11 days to discharge down to just 1 second to discharge, and all because we had different size capacitors.

Some of the cheaper meters only have 1 megohm resistance while some have 10 megohm or close to that. Analog meters are much lower and could be as low as 20k or even lower.
 

Aleph(0)

Joined Mar 14, 2015
597
@Sally S

I say use use something with decent relative permittivity for dielectric (which for just low voltage can be thin sheet of taut latex) and measure DIY cap parameters with proper LCR meter so you're seeing big picture:)

So that won't be supercap (which is practically impossible to DIY) but it will have a lot more capacitance and a lot less DC leakage than with just paper dielectric:)!
 

crutschow

Joined Mar 14, 2008
27,945
Plastic wrap, as Externet suggested, would probably work best.
It's very thin, has a high dielectric constant, and should have a very high resistance.
 

OBW0549

Joined Mar 2, 2015
3,566
Has plastic wrap been tried instead of paper as dielectric ?
Yup.

I decided to do some science and took two pieces of aluminum foil separated by a sheet of laser printer paper, with an 8" x 10.7" area of foil overlap, and got 11.7 nF capacitance. Swapping a sheet of GLAD Cling-Wrap for the paper, I got 9.2 nF.

Rather strange results, as I would have expected the Cling-Wrap to yield a lot more capacitance, not less.

Go figure. :confused:o_O
 

Kermit2

Joined Feb 5, 2010
4,162
Paper is porous so it is a poor choice for this.
If it must be paper, try brown paper sacks which are thicker and apply medical grade mineral oil (from the pharmacy) to increase the dielectric rating if needed.
 

OBW0549

Joined Mar 2, 2015
3,566
My best guess -- and it is a really big guess -- is that you have trapped air between the Cling-Wrap and the foil.
That could be, although the foil sandwich is resting on a flat, smooth table top; there is a 1/4" thick pad of note paper on top of the sandwich; and on top of the notepad are 30 lbs. of lead weights spread out evenly over the surface. There may be a small amount of air trapped between the Cling-Wrap and the foil, but if there is it isn't much.
 

RichardO

Joined May 4, 2013
2,271
That could be, although the foil sandwich is resting on a flat, smooth table top; there is a 1/4" thick pad of note paper on top of the sandwich; and on top of the notepad are 30 lbs. of lead weights spread out evenly over the surface. There may be a small amount of air trapped between the Cling-Wrap and the foil, but if there is it isn't much.
Well, so much for my crystal ball. I may have to send it in for repair. :rolleyes:
 

OBW0549

Joined Mar 2, 2015
3,566
Well, so much for my crystal ball. I may have to send it in for repair. :rolleyes:
Don't be in too much of a rush to get a RMA on that thing: I made another measurement, this time with a 1/8" layer of soft plastic foam between the bottom of the notepad and the topside of the foil sandwich (to better distribute the force from the overlying lead weights), and got the capacitance up to 19.6 nF. So trapped air may indeed be a factor.
 
Top