[Prev][Next][Index][Thread]
Re: Home brew capacitor
Ian,
Congratulations on the new addition in July, and the new job.
You're right - making a plate capacitor DOES take time, and is not free.
However, you'll end up with a robust, solid performer. One of the nice
things about polypropylene is its higher dielectric constant (3.1)
versus LDPE (2 - 2.3). The use of 30 2-mil layers to get 60 mils of
total dielectric thickness is a good idea provided you break the
capacitors in slowly and orient the plates vertically. Or are you using
a vacuum pump to extract the entrapped air?
Also, I think Perspex is the same thing as Plexiglas (Polymethyl
Methacrylate - PMMA). My plastics supplier's chemical resistance tables
indicate that PMMA is slowly attacked by transformer or mineral oil,
which is why I opted to use PVC seperator plates instead. However, I've
never heard of any coilers having oil leakage failures when using
Plexiglas top plates on the rolled caps...
What kind of a container did you put your caps in, and are you planning
on driving your coil and the caps to the 15 KVRMS design limit (after
break-in)?
- Bert --
Tesla List wrote:
>
> >From i_hopley-at-wintermute.co.ukSun Sep 22 15:24:38 1996
> Date: Sun, 22 Sep 1996 14:52:59 +0100
> From: Ian Hopley <i_hopley-at-wintermute.co.uk>
> To: tesla-at-pupman-dot-com
> Subject: Home brew capacitor
>
> Hi guys,
>
> Been quite a while since i posted here. Due to the introduction of one 8 1/2
> lb Glen Douglas Hopley on the 21st Of July and a simultaneous change of job.
> I have been rather occupied and not able to do much work on my coil.
>
> After the recent discussion on capacitors i thought i would give a quick
> description of the one i am building at the moment. The design came about
> after much studying of posts from Richard Quick and Richard Hull and many
> other members of the group and hopefully incorporates all the major features
> of a good homebrew cap.
>
> The main elements are A4 Polypropelene document wallets and heavy duty
> aluminium cooking foil. The wallets have a thickness of 2 Mils and after
> trimming the punched hole edges off and cutting in half give a folded
> rectangle of 8 1/4 X 5 7/8 inch. A sheet of aluminium foil was then cut to
> be 1/4 inch inside the edges of the plastic and was slipped inside allowing
> about 1 1/2 hanging out of one of the short sides.
>
> ___ ____
> | ______________________________________________ |
> | | Film | |
> g | -------------------------------------------------------: u | |
> | :
> i | | | :
> d | | | :
> e | | | Foil :
> | | | :
> a | | | :
> n | | | :
> g | ------------------------------------------------------:
> l | ___________________________________________ | e
>
> s ^
> Folded edge of film
>
> Two of these plates were stacked with the overhanging edge of foil at
> opposite sides, giving a double thickness of film and tested to destruction
> under oil, (see previous post). It was determined that the plates failed at
> aprox. 3.5KV. A safty margin of X3 was decided on so the cap was assumed to
> have a working voltage of 1KV. A max. working voltage of 15KV was decided on
> for the cap. at a size of 0.01mfd which allows finished units to be
> paralleled up in 0.01 steps as coil input power is increased.
>
> This Decreed that 15 X 0.15ufd caps needed to be built and connected in
> series to achive the desired cap.
>
> Each cap was built up with 52 plates stacked with the extended end hanging
> out of each short side alternativly. To help buid the cap. a jig was built
> using four lenghts of aluminium angle having an inside edge leghth of 1/4
> inch. These were fastened to a board so that the corners of the plastic
> sheet sat in to the inside corners of the angle and the foil was then spaced
> inthe film by the edges of the angle.
>
> The method of layering was as follows:-
>
> 1. A sheet of perspex, the same dims. As the film was laid on the
> building board, to give the cap strength.
>
> 2. The 52 plates were then stacked alternativly on top of the perspex.
>
> 3. Another sheet of perspex was laid on top of the first cap then the
> next cap was laid up on top as before. The first plate was arranged that
> when the caps were bused together the plates either side of the
> separator were connected together so that no electrical field was placed
> across the sheet.
>
> This was carried on untill all fifteen caps were stacked, then whole block
> was bound using tywraps and lifted out of the jig.
> The busing was done by folding the exposed edges of the foil on each
> adjacent pair of caps. and crimping together using mole grips.
> To enshure a constant pressure on the joint plastic spines from document
> folders were slipped over the joint to hold it firm.
>
> The input and output connections were made by folding a strip of 1 1/4 inch
> copper strip over the folded foils of the end capacitors, and drilling and
> bolting through, giving a copper tape lead out of 5/8 inch. Wide. The
> connections are brought out of cap tank using 1/4 Brass studs.
>
> unfortunatly a photograph could save a thousand words here but i dont have
> access to a scaner anymore.
>
> The entire finished cap measures 5 7/8 x 8 1/4 x 5 3/4 inch outside of its
> tank. To finish it, it requires parralel resistors connecting across each
> section to enshure even voltage distribution, than finaly dunking in a
> cotainer of oil.
>
> I have given it a quick (2 sec) burst of sparking with no oil at about 5 KV
> with no primary connected and judging by the noise, brilliance of sparks
> this is going to be a vast improvment on my old salt water cap.
>
> If anybody considers building such a cap please note the following:-
>
> The number of plates was determined by measurement with an LCR meter and
> will change with the thickness of film.
>
> Dont underestimate, and at the same time dont be put off by the amount of
> time required to build one of these caps. The jig greatly assisted the
> speed and accuracy that the cap could be built, but 780 plates and films
> takes a long time to prepare. I used a paper guillotine to do all the
> cutting. Spending about an hour or two at night when i could, it took me
> about 5 weeks to layer up the cap. Hopefully it will be worth it when
> finished as i bought enough materials two build two caps.
>
> Total cost per cap so far? About 35 U.K. Pounds
> *******************************************
> Ian Hopley ----> i_hopley-at-wintermute.co.uk
> Aberdeen
> Scotland Callsign M M 1 A B A
> *******************************************