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

Re: Cross-linked caps



>Original Poster: wwl-at-netcomuk.co.uk (Mike Harrison) 
>
>>1) If they are cross-linked, I believe the voltage across each individual
>>cap will be more evenly distributed, as the capacitance variation across
>>the individual units will averaged out.  If one of the caps did in fact
>>have an especially low value and was not cross-linked, it would see a
>>disproportionally high voltage across it and be most likely to blow.
>
>I would think it would be a good idea to use a resistor across each
>cap to ensure the voltage is evenly distributed over all the caps - if
>you're running close to the limit, one cap failing would probably
>'unzip' all the rest. Obviously a cross-linked array requires fewer
>resistors. Cross-linking should also reduce the effects of
>part-to-part capacitance/esr variation, again reducing the risk of a
>serial failure due to uneven voltage distribution. 

I don't think resistors serve to equilize the voltage across the series
caps, although they may serve other purposes.

The impedance of the resistors in parallel with each cap stage would
normally be on the order of several megohms.  If it was much lower, it
would be wasting too much energy.  With the resistor impedance in the
megohm range, it will have absolutely no effect towards equilizing the
voltage  at the tank resonant frequency, as the capacitor's impedance
will be orders of magnitude lower.  I believe the benefits of the
resistors are:
1) They prevent any DC bias from building up on the caps, which when
   combined with the AC tank oscillations, may exceed the cap's rating.
   It's not clear if or how such a DC bias might accumulate, possibly
   from coronal rectification?
2) They ensure that any remaining charge on the caps after power is shut
   off is safely dissipated and the cap is safe to handle.


>Original Poster: "Reinhard Walter Buchner" <rw.buchner-at-verbund-dot-net>=20
>
>This idea of crosslinking originated on our german Tesla list.
>It was my suggestion and here are some of my thoughts
>behind this scheme:
>
>Advantages:
>----------------
>a.) You only need a single string=B4s worth of equalization resisitors.
>b.) Cross linking equalizes voltage distribution (as Gary already
>mentioned).
>c.) Current per cap is also equalized.
>d.) ESR should be lower.
>e.) If you DO cross link, the maximum number of caps that can blow
>     is "X". X being the number of caps in one string. If they short,
>     the matrix linking will protect all caps "underneath" the shorted
>     cap. If it open circuits, the capacitance is simply reduced by
>     one cap.
>f.) Using heavy bus bars (flat copper strips) to cross link, should
>    allow higher peak currents (more bang for the buck).

I don't think that the ESR would be affected by cross-linking.  Also, I
don't believe that heavy bus bars for the links are are needed.  If all
of the caps were 100% identical, no current at all would would flow
through the links, and if the caps exhibit only minor differences, the
link currents should be small.

>Disadvantages:
>------------------
>a.) Fine tuning is much more difficult (since you need to remove all
>     connections on the string).
>b.) If one cap in a string shorts, automaticly all caps (since one
>     complete "set" of caps is "removed" from the string. I.e. a 15
>     cap string will effectivly be reduced to a 14 cap string) see a
>     higher voltage per cap. This might lead to quicker failure, if
>     you can=B4t / don=B4t switch off fast enough.
>
>
>So, the question whether to cross link or not, is not that easy to
>answer (and I, for my part, haven=B4t decided on this yet). My
>present "play around with" setup DOES NOT use cross linking.
>(on the other hand, I am not using equalization resistors or
>safety gaps, either) I haven=B4t gotten around to trying the
>crosslinking (to see if it will increase performance) setup.
>
>Either way around, I am sold on the MMC idea and WILL and
>DO use this for my 8" coil setup. I am very pleased with the
>results (41+" from a 7.5kV/75mA NST isn=B4t bad). As a matter of
>fact, the MMC has given me the best results from all the caps I
>have built so far.

EXCELLENT!!!

>Very important (I think): Look for metal foil MKP (polypropelene)
>and NOT metalized plastic foil. The metalized plastic foil caps FAIL
>almost immediately. Use caps of high quality. Usually the better
>caps are larger in physical size (physical size vs. capacitance),
>than the El-Cheapo brand.
>
>Coiler greets from germany,
>Reinhard


Regards, Gary Lau
Waltham, MA USA