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

Re: New Cap Type

Hi Reinhard
             Your idea is used in some commercial cap designs as a 
way of grading the voltage across the dielectric layers as you have 
done.

Malcolm

> Original Poster: RWB355-at-aol-dot-com=20
> 
> Hello All,
> 
>  I=B4m not sure if I "invented" something new, but I thought I would let =
> you
> know:
> 
> My cap design looks somewhat different from the rest of the poly design t=
> ypes
> I have seen. In the process of building and testing different homemade ca=
> ps I
> found out something very interesting. If you build a poly cap the convent=
> ional
> way you need lots of PE foil in order to prevent your cap from dying a hi=
> gh
> voltage death.=20
> 
> I stack my caps this way: 2 pieces of PE foil start the cap pack and 2 pi=
> eces
> end the  cap stack. In between this I alternate aluminium and PE, but I d=
> on=B4t
> interconnect any of the AL plates. I just connect the lower and the upper
> plate of the "cap". My idea behind this is very simple: If you do this, y=
> ou
> are really connecting multiple caps in series (i.e AL PE AL being one cap=
> ).
> This way I use only one piece of PE foil, making the whole thing easier t=
> o
> roll because the AL sticks nicely to the PE and you don=B4t have PE on PE
> slipping all around. I say my PE foil (6.7 mils thick) will hold up to 50=
> 0
> volts AC with ease. So e.g. if I want a 10kv cap I would need 20 PE foils=
> . I
> build one of these prototypes and here comes the killer: I fired it DRY i=
> n a
> TC tank. It took it with ease. After about 5 mins of running it was barel=
> y
> warm and NO corona or arcs between the plates even though I was putting 1=
> 5kv
> (50% overrated) into it and used NO oil.=20
> 
> The thermal and electrical stresses are greatly reduced, compared to the
> normal poly design. The only negative thing about my design is that you n=
> eed a
> lot of these to make your needed capacitance. However, I would be able to=
>  fit
> at least 5 or 6 of  these 2nf (protype case) caps in a single 4" PP pipe =
> and
> connected them in parallel for 10-12 nf. My pickup plates are placed in t=
> he
> middle of the top and bottom plate. My idea behind this is simple, too. A=
> s I
> take the current from the middle of the plate, the inductance of the end
> plates is smaller and the current doesn=B4t have to follow a single long =
> path to
> the discharge. So my cap should be able to discharge quicker than a
> conventional rolled design. Which translates to a better bang for the mon=
> ey.
> 
>  Besides, it=B4s easier and faster to add (they can be built in about 15 =
> minutes
> time)  or subtract a cap or two to change my capacitance (for tuning), th=
> an if
> I had one or two big poly rolls.
> 
> Has anybody else tried this or have comments?
> 
> coiler greets from Germany,
> Reinhard
> 
> 
>