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

RE: The MMC debate continues . . . . . .



Original poster: "Lau, Gary" <Gary.Lau@xxxxxx>

In a controlled experiment that I performed, I found that the presence
of a residual charge is strictly dependant on the particular type of
capacitor used.  It's more than just dielectric material, cap package
type, or geometry of the component arrangement.  Below is a re-post of
the experimental results.

In a nutshell, the CD 942C caps that we most commonly use, did NOT
exhibit residual charges.  At least not for me, on that particular day.
But I would still argue passionately in favor of using bleeder resistors
for any cap, MMC or otherwise.  On one occasion and prior to using a
bleeder resistor, I attempted to determine the optimum primary tap by
dragging the tap clip across the primary turns while the coil was
running.  Violent arcing occurred between the clip and the primary turns
and I immediately abandoned that foolish idea.  I shut off the power and
picked up the disconnected tap lead to connect it to the primary.
ZAPPPP!!!  By powering off while the tap was disconnected, the normal
safety discharge path through the NST was defeated, and the charged cap
had nowhere to discharge but my hand.  Yes, the shock was due to my
having an unconventional configuration and not thinking it all the way
through, but who among us would analyze every thing every time to such
an extent?

The mechanism behind why some caps do and others don't develop residual,
asymmetric charges remains unclear, but the wisdom behind using such a
simple safety device is irrefutable.

Regards, Gary Lau
MA, USA


> Original poster: "resonance" <resonance@xxxxxxxxxxxx>
>
> I have played with AC powered TCs for nearly 45 years now and I have
> never ever had a shock from a small, medium, or large TC that runs on
> AC.  I recall Ed Wingate has issued warnings where he has had bad
> experiences with dual cap "equi-drive" systems, but I have not used
> this type of drive so really can't comment on it.
>
> Some of the VTTCs use rectification and Terry's Piranha MOT coil uses
> rectification with the MMCs so you would certainly always use
> bleeders on this design.
>
> Terry F. has also reported some storaged charges on MMC caps so I
> certainly don't want to discourage anyone from using the bleeders,
> but I have never found any charge on some of the smaller MMC style
> commercial coils we produce.  Both Matt and I have never received a
> shock from an AC MMC coil nor have any of my approximately 60
> customers of MMC small coils.  Many of these units are used in
> science museums 7 days per week and sometimes up to three
> performances per day.  The operators and techs put the coil away and
> usually clean them at least once per day and so far, no reported
> shocks from cap storage.  These are all strictly NST and AC powered
coils.
>
> Best to error on the side of caution if you have any doubts.
>
> I would suggest using a small ground probe with an NE-2 neon bult to
> "test" the caps and check them after use in your particular coil
> design.  If after dozens of operations, and no stored charge found, I
> think it would be safe to discontinue such tests.
>
> Be safe!
>
> Dr. Resonance


Gary's post from 1-19-2004:

>I ran my mini coil with 4 Geek Group CD942C 0.1uF/2000V caps in series
>with no bleeder resistors.  After running maybe 10 seconds, I unhooked
one
>end of the string, and using a DC multimeter and one of the passive
990Meg
>HV probes, attempted to read the residual charge on each cap.  I saw
nothing.
>
>I replaced the CD caps with a string of 4 BC Components 383-MMKP
>0.1uF/1600V caps.  Ran the coil, tried to measure a residual charge and

>saw none.  What the...  I KNOW I had seen residual charges enough to
make
>a good -snap- when I shorted them in the past...
>
>I replaced the 383-MMKP string with a string of 4 BC Components
376-MMKP
>0.047uF/1600V caps.  These are the ones that I first used when I saw
the
>residual charges.  I retuned and ran the coil for a few seconds,
>disconnected one end of the string, and sure enough, there was the
charge,
>plain as day!  I tried several iterations, and each time the charge on
>each cap varied, I think the highest value I saw over 3 iterations was
>about 610V.
>
>The two different BC caps have identical case sizes and were wired with

>identical geometries.  Clearly some types of caps are more prone to
this
>effect than others.  For others who have seen the residual charges,
which
>type of caps were they?
>
>Gary Lau
>MA, USA