Re: DeltaV/DeltaT vs capacitor roll length.

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: Bert Hickman <bert.hickman-at-aquila-dot-net> 

Tesla list wrote:
>Original poster: Finn Hammer <f-h-at-c.dk>
>All,
>On Kreso`s pictures of the Maxwell 37667 caps, it is obvious that the 
>individual capacitor rolls are as long as the inside of the case, so they 
>are about 5" long.
>On the spec. sheets that I have seen of (What we now call MMC caps) the 
>dV/dT is dependant on the length of the cap. The small types have a better 
>rating than the bigger types in longer cases.
>I am wondering how Maxwell manages to produce a cap with superiour dV/dC 
>and still from individual caps that are 3 times as long as the ones we use 
>for MMC`s.
>Cheers, Finn Hammer
>
>.

Finn,

With metalized polypropylene pulse capacitors, it's not so much the length 
of the cap as it is the surface area at the ends of the cap, since the end 
cap area governs the allowable current density for the cap. The sprayed-on 
end contacts are made from a low melting temperature metal, typically zinc. 
Unfortunately, the electrical contact between the sprayed metal and the 
metallization layer on the dielectric is difficult to control and is the 
weak link for this style of capacitor. Under high pulse or RMS currents, 
marginal contact points heat up. Heating of the capacitor body can also 
induce physical shrinkage problems within the PP dielectric, which can also 
contribute to progressive detachment. Rapid dv/dt conditions can also 
create mechanical stress due to slight dimensional changes within the 
capacitor winding through Coulomb attraction, particularly in capacitors 
with insufficient winding tension. If any of these problems cause the 
effective contact area to further decrease, heating can intensify in the 
areas still making contact, leading to progressive deterioration, lead 
detachment, and internal arcing.

This phenomenon is not a problem with foil-film construction since ohmic 
contact to the foil can be made considerably more robust. The best of both 
worlds is to use foil electrodes at each end combined with self-healing 
metalized film floating layer(s). This appears to be the construction used 
in the Maxwell 37667 cap and in many high current snubber caps.

Best regards,

-- Bert --
--