Re: How could a pulse cap operate in TC?

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: "Mike" <induction@xxxxxxxxxxx>

Hi Grishka,
                 Terry, Malcolm and others all made good points in reply. 
From your question of what is used at high power RF, I would add, IF I had 
no budget and if I wanted a nice CW rating of 250 amps RMS at 20 kV in a 
water cooled ceramic design that commercially runs in induction heaters, 
with the ability to run at several times the typical TC frequency range, I 
would use the Draloric water cooled cap in combinations to reach my design 
frequency. But this is far from cheap and as Terry has pointed out rating 
plus total capacity and RMS amps, the MMC is a lot of value.
We sell .0076, 20 kV, 250 amp RMS caps in a range of $1,700.00 US. We buy 
them from Vishay and stock for clients or use them in rebuilding big heaters.
That is fine for a commercial, run all day and night mill with maybe 1000 
CW amps or more in the tank / work coil circuit
but most likely the average TC person is not going to pay that, only a 
company in emergency failure would.
But if you want to see some neat caps and spend virtual dollars on them, 
http://www.vishay.com/capacitors/ceramic-rf-power/ , like a virtual wish book.
Mike

----- Original Message ----- From: "Tesla list" <tesla@xxxxxxxxxx>
To: <tesla@xxxxxxxxxx>
Sent: Wednesday, July 13, 2005 2:28 PM
Subject: How could a pulse cap operate in TC?


> Original poster: Grishka <ghome@xxxxxxxxx>
>
> Hello everybody,
> The capacitor is the most critical part in a SGTC, because it should withstand
> great power levels at rather high frequencies. Many factors are involved 
> here -
> maximum speed of voltage increase - i. e. dV/dt, dielectric loses in a cap,
> level of its reactive power & some others. Many people doing rather nicy 
> TCs are
> completely careless in choosing right capacitor (MMC) and its parameters 
> in operation.
> I`ve seen many coilers using MMC and saying that "it`s working - so it`s 
> OK". But
> even using a MMC we should calculate some very important details.
> I remember Terry Fritz wrote here:
>
> http://www.pupman.com/listarchives/2002/July/msg01473.html
>
> "MMC caps have a giant advantage in that they are cheap and easy to test 
> under "deadly" conditions to
> determine exactly how "much they can take". Expensive commercial caps 
> can't be tested with such wild
> abandon and we have to trust the typically "scarce" information the 
> manufactures give us."
>
> What is this "information"? What exactly parameters are the most criticle 
> and important
> for us? Now I`ve got a question about types of caps you use - reading some 
> articles I found
> the Maxwell caps to be the most reliable - if I`m not mistaken they were 
> designed to operate
> at laser applications. In such applications capacitor works in a pulse way 
> with fast discharge
> into low-resistance load with frequencies not more than 100..200 Hz.
>
> Now questions:
>
> -How could a Maxwell cap work in a TC, where frequencies are hundreds of KHz?
> -Does it overheat a lot? What are the practical values for the dielectric 
> loses in your caps?
>
> Maybe people, who works with high-power RF applications can give some 
> advices about using capacitors
> in these devices?
>
>
>
>
> --
> Best regards,
>  Grishka                          mailto:ghome@xxxxxxxxx