Clearly a decrease in thickness of the dielectric increases capacitance, but how about the metal plates on a parallel plate capacitor? If you increase or decrease the thickness too much will you see a increase/decrease in capacitance? Is there an optimal thickness of the metal plates?
\$\begingroup\$
\$\endgroup\$
2
-
1\$\begingroup\$ The simplest idea about gradually increasing the thickness isn't hard to gather. It provides more volume for the charges to spread out within and a little bit more surface area for a few more excess charges to accumulate across. It's a sea of charges either way, though. This may support greater currents due to the cross-sectional area but there is little other change. But most of the effect of the dielectric on capacitance happens between the plates. And the round trip path integral of the electric field (the round trip potential difference) must still sum to zero. Not much change, 1st order. \$\endgroup\$– jonkCommented Jul 9, 2020 at 4:04
-
1\$\begingroup\$ This is experimentally supported by olden-day air gap variable capacitors used to tune AM radios, where the plate thickness wasn't a consideration (however the gaps were) in their construction. You could buiild them with thicker or thinner plates, based upon your desire for sturdiness to bumps and dings, but the main design issue was the spacing. \$\endgroup\$– jonkCommented Jul 9, 2020 at 4:10
Add a comment
|
1 Answer
\$\begingroup\$
\$\endgroup\$
3
Think of metal film capacitors which literally have a metal film vapor deposited onto the dielectric. The less metal thickness the less the waste in mass and bulk and metal. It only needs to be thick enough to have full conductivity. Adding thickness just adds mass and bulk with no gain, so optimal thickness is to be as thin as possible.
Note that metal plates need to be thick enough to hold their own weight and shape, as in old style air-gap adjustable capacitors. The plates were about 5 mils thick.
Note that high-energy capacitors for arc simulation will use a thick dielectric with metal foil, soaked in a light oil as a coolant and to prevent internal arcing. These capacitors can be the size of a shoe-box to the size of a filing cabinet, so you just accept their size and weight, normally not portable at all.
What does affect capacitance is the thickness of the dielectric, so the thinner the better, but it must be thick enough to block/handle the rated voltage. More metal (and dielectric) in terms of windings also increases capacitance.
I am sure you have noticed that for a given voltage, more capacitance means a larger capacitor. If the voltage is fixed but you increase capacitance, you have a larger capacitor. For this reason capacitor manufactures offer electrolytics in both tall and skinny, as well as short and fat styles. Most designs make use of some compromise between too much height or too much width, then they chose parts that fit that style.
This is where the compactness of todays capacitors pay off with smaller foot-prints on the board.
-
\$\begingroup\$ What constitutes "thick enough to have full conductivity" i.e. at what thickness do you lose full conductivity? \$\endgroup\$– gstudentCommented Jul 9, 2020 at 3:25
-
\$\begingroup\$ Depends on the metal deposited by vapor. Many atoms thick to be sure, but this is where lab equipment determines minimum thickness based on ohms ^2 cm. Also this vapor deposit has to be equal over all conductive surfaces to have consistent product accuracy. \$\endgroup\$– user105652Commented Jul 9, 2020 at 3:30
-
1\$\begingroup\$ @gstudent Very thin films (for example less than tens of nanometers for copper) often do not conduct well due to interface effects, so you typically need to be at least that thick. Depending on the power handling characteristics, you might want thicker as well so that there is less resistive heating. \$\endgroup\$ Commented Jul 9, 2020 at 3:44