[Prev][Next][Index][Thread]

Re: Capacitance of a coil

Hi Carlos,

	My program (E-Tesla) should be able to do this.  However, bare coils are
rather hard for the program at this point.  Long thin coils fall on the
grid poorley which causes accuracy problems.  I am now working on version 3
which I hope will be much more capable in this area.  Also such coils are
very sensitive to the surroundings both in real life and when modeled.

Bare Coils (without a terminal) are rather interesting.  The voltage along
them seems to have a sine distribution which makes hand calculations rather
messy.  They store charge disproportionatelly along their length.  The area
around the bottom of the coil is not contributing much to the overall
effect.  A field map is most usefull here to get an idea of what is going
on.  Interestingly, the program always seems to predict a lower value for
the capacitance.  This may be real but the implications of what it would be
telling us is unknown.  It may be that the low currents near the top of the
coil are causing the inductance to appear lower than what a meter would
measure.  There are still some unknowns here.  The program seems to
indicate all is not as simple as we would like it to be :-)).

		Terry

At 10:25 PM 1/16/99 -0800, you wrote:
>John Couture wrote:
>
>>   Is the capacitance the same for a coil with the same R and L whether it
>> has one turn or 1000 turns? I am having difficulty with this. My research
>> and limited tests indicates the Medhurst equation and equations that do not
>> include the number of turns or length of wire should be used with caution.
>
>Remember that this capacitance is a first-order approximation for a
>transmission line effect, and so is valid at DC too. Only the general
>shape of the conductive surface is significant.
>The same formula probably works for a hollow thin metallic cylinder 
>with the lower end close to ground, or for a cylinder made of a wire
>mesh. A coil with a reasonable number of turns is an approximation of a
>continuous conductive surface with this form. Of course, with one
>turn only the formula would give a large error.
>
>(While making the comment above I was imagining what would be the
>effect of the distance to ground in the capacitance of a hollow
>cylinder with constant voltage to ground. If the idea is correct, 
>it must reach a limit when the cylinder touches the ground, close
>to the value given by Medhurst's formula, and not go to infinity, 
>as it may appear that is what would happen.
>Would Terry's simulator be able to verify this?)
>
>Antonio Carlos M. de Queiroz
>
>
>
References: