Re: MMC resister problem

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

Original poster: "Dave Leddon by way of Terry Fritz <twftesla-at-qwest-dot-net>" <leddon-at-attbi-dot-com>

Couldn't you "discharge" an MMC by applying a relative low voltage (say 6.3
VAC) across the capacitor array?  It would seem to me that the highest
stored voltage would then be only  8.9 volts.

Dave

At 12:33 PM 6/3/02 -0600, you wrote:
>Original poster: "Jim Lux by way of Terry Fritz <twftesla-at-qwest-dot-net>"
<jimlux-at-earthlink-dot-net>
>
>Having a design that by its very nature leaves charge and stored energy in
>the system, and is impossible to discharge externally, makes me a bit
>nervous.  If you were potting the caps permanently in epoxy or something it
>would be different (although, still dicey..)
>
>One of the beauties of a MMC is that if something goes awry and you DO lose
>a cap, you can replace just the one.  Therefore, serviceability of the
>arrangement is desired.   The plastic cover over the cap connections seems
>a "removeable service panel". If it's serviceable, it's accessible, and you
>should have separate bleeders. Alternately, you could be prepared to spend
>a lot of time with a pile of suitable clip leads, a carefully thought out
>sequence of applying the jumpers, and some discipline to follow your
>procedure, so you can jumper out all the caps one by one.
>
>I recognize that commercial energy storage caps (e.g. from Maxwell) are
>built up from series/parallel combinations of smaller caps, and, do not
>have internal bleeder resistors, however, they are not intended to be
>serviced (they're in sealed steel or plastic tanks). If I were dismantling
>one of these things, I'd be real careful during the process.
>
>
>A typical MMC component cap stores less than a joule, so it's not likely to
>kill you, but one should be careful of unforseen situations where a failed
>(open) cap (or a broken solder joint, more likely) causes the remaining
>caps in the string to retain charge, or where parallel strings can store a
>very significant amount of charge. While the individual cap may only store
>1/2 a joule, an MMC made up of 100 could store 50 Joules, which CAN kill
>you, or, less dramatically, cause a significant muscle contraction leading
>to mechanical damage or burn a nice little hole in your finger.
>
>There's a reason for those design rules in the NEC and the "good
>engineering practice" of separate bleeders for each element in a capacitor
>string.
>
>
>
>Tesla list wrote:
>> 
>> Original poster: "Justin Hays by way of Terry Fritz <twftesla-at-qwest-dot-net>"
><pyrotrons2000-at-yahoo-dot-com>
>> 
>> Hi All,
>> 
>> (snip from Jim Lux)
>> > A single resistor across a series combination of capacitors does
>> > not provide a safe bleeder function.  It is entirely possible that
>> > there may be significant charge on the capacitors.  "Gedanken
>> > Experiment":   two capacitors, equal values, charged to 1kV.  Hook
>> > them in series, with polarities opposite.  Voltage across series
>> > string is zero volts, so putting any resistor across the entire
>> > string won't result in any current flow, yet the capacitors are
>> > still charged.
>> >
>> > For a MMC, the only safe approach is to put resistors across each
>> > capacitor.
>> 
>> There is more than one safe approach, which I mentioned in my
>> previous post. Again, aside from using resistors across all
>> capacitors, you can do the following:
>> 
>> 1). Use a single HV resistor across the entire MMC
>> 2). Shield the open connections with plastic so there is a physical

>> barrier.
>> 
>> Doing both of these will keep you safe. Not one or the other, but
>> both. If you omit the plastic, you could take a shock from the
>> residual charge on the inner cap's. If you keep the plastic but omit
>> the main resistor, you could take a shock from other coil components
>> that are connected to the MMC...(that is, if and only if the
>> transformer secondary winding goes open circuit,,,in conventional TC
>> setup).
>> 
>> Take care,
>> 
>> Justin Hays
>> KC5PNP
>> Email: justin-at-hvguy-dot-com
>> Website: www.hvguy-dot-com
>