Making a (not very
good) Capacitor
Revision History
07 October 2009 Originally drafted
09 October 2009 Added plastic wrap insulated capacitor
Recently reading Dummer & Nordenberg's Fixed
and Variable
Capacitor
book lead me to wonder how good (or, more likely, bad) a quickly assembled home
made capacitor would be.
A capacitor is simple enough in theory; two plates
separated by a dielectric. I made the plates from two foil wrappers from Bliss
chocolates.
|
|
Plates made from the foil wrapper of Bliss chocolates.
|
|
I carefully unwrapped two foils and smoothed them out with a rubber roller on a
granite flat. The resulting foil is approximately 3.0 x 2.6 inches.
|
|
Unwrapped and smoothed, the foil is approximately 3
inches x 2-5/8 inches.
|
 |
I then cut two strips of rather thick paper (0.005 inches) of a size to permit
the foils to overlap over most of their width, with projections from each end.
The photo below shows the concept. To show the arrangement,
I've spread the foils and dielectric paper out.
|
|
Foil and paper arrangement (spread out to show details)
|
 |
I then folded the capacitor several times.
|
|
Capacitor being folded.
|
 |
Completed capacitor. The foil extending from each end are the connection points
to the plates.
|
|
Completed capacitor
|
 |
At 1 KHz, the capacitor measures 474 pF. This is not too far from the low end of
my calculated value expected of 550-1100 pF. (The main reason for this
uncertainty in computed value is the dielectric constant of paper is usually
quoted as 1.5 to 3. This yields a 2:1 spread on computed values.)
The dissipation factor is on the poor to terrible side; 0.15.
I made a second capacitor with the same foil size, but
with thinner paper, approximately 0.003 inches thick. (20 lb copier paper.) It
measures 930 pF with a dissipation factor of 0.18. The higher capacitance is due
to reducing the dielectric thickness, smoother foil (improved rolling) and
possible difference in paper dielectric constant.
There are several ways one might improve this design:
- Use flatter foil. This will increase the capacitance
and also reduce the chance of dielectric breakdown due to voltage gradients
around the wrinkles.
- Use thinner paper. Cigarette paper, for example, is
around 0.001 inches thick. However, Dummer & Nordenberg say that paper of
this thickness will have a fair number of holes and hence is should be used
in a double layer. Two layers of 0.001 inch paper will increase the capacitance and probably improve
the dissipation factor compared with a single layer of either 0.003 or 0.005
inch paper.
- Coat the paper with silicon oil.
- Wrap it more tightly and use heat shrink tubing
as a surround.
- Or, use a dielectric other than paper, such as mylar
film.
|
|
Food Wrap Dielectric As an
experiment suggested by Dave, G3TJP, I made a capacitor with vinyl food wrap
dielectric. It uses the same chocolate candy foil plate as the earlier paper
dielectric units.
|
|
Upper capacitor uses vinyl food wrap dielectric. Bottom
uses paper.
|
 |
|
I measured the food wrap at 0.0013 inches thick. It's
difficult to cleanly cut, as it has enough static electric charge to cling
together. It also is flexible enough to allow the finished capacitor to be more
tightly folded, which also increases capacitance.
The result is 2434 pF with a dissipation factor of 0.0261 at 1 KHz. Both the
capacitance and dissipation factor are considerably improved over either paper
dielectric capacitor.
|
|
General Radio 1658 DigiBridge display of the food wrap
capacitor
|
 |
It's not in the same league as a store bought capacitor, but it's interesting
that with almost no care in assembly, a usable capacitor can be constructed.
Of course, there are a few minor details required, such as
a leads and a waterproof housing. A waterproof housing could be supplied with a
dip into liquid tape (suggested by G3TJP) or it could go into a length of
phenolic tubing, sealed with beeswax for a genuine antique touch. Leads are
slightly more difficult—traditionally they were crimped onto the foil, as
soldering to thin aluminum foil is tricky at best. This requires a crimp sleeve
that could be made from a short length of thin copper or brass tubing. |
|
|
|
|
|
|
|
|
|
|
