Re: MMC vs Maxwell comparison

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

Original poster: "Terry Fritz" <twftesla-at-qwest-dot-net>

Hi John,

These are the measured dissipation factors for your type of Arcotronics
0.047 1500VDC caps.

fkHz Theta
50   0.00074 
100  0.00190 
150  0.00290 
200  0.00370 
250  0.00460 
300  0.00540 
350  0.00650 
400  0.00740 

For a 10 x 3 configuration, one would multiply these values by 3.33.  Very
good numbers!  

66.67 Mohms of drain resistance dissipates about 1.2 watts (9000 VAC) which
is actually about 0.5% of the coil's power (~225 watts).  Your testing may
actually be sensitive enough to see that!

I am a little surprised at your mention of how "quickly" the MMC heated up.
 I assume the temperature was less than 10C.  Your configuration should be
able to easily handle the current so I am a little surprised if they got
"very" warm.  I would guess they would heat to about 4C after a very long
run.  Your coil probably runs about 5 amps RMS in the primary.  You
mentioned a couple of minutes of intermittent run time so maybe the 4C is
just about what you are seeing.  That would confirm that I sort of know
what I am doing ;-))  When you feel an MMC, you are right there on the guts
of the capacitor itself so you feel the heat easily.  I think the Maxwell
will have more volume (thicker dielectrics since shorts are "bad") so the
heat gets spread out more.  The large case and oil also absorbs a lot of
heat before they start to get warm.

Probably the leads, dissipation, and drain resistors do have more loss than
the Maxwell.  10 Geek Group caps would be about the equivalent for $30.  If
we can get to 99% of a Maxwell cap for $30 (I don't know what Maxwell's
cost) I guess we do real good :-))  It's interesting that MMCs were
inspired to find a primary cap that would stand up in situations that
failed Maxwell caps.  Of course, we know far more about primary currents
now which was the key there.  One other giant advantage is that MMCs can
survive short over voltage events.  Maxwells get "messy" with over voltage! 

Wonderful testing you did here!!  I am surprised you could see such a fine
difference.  The results sound just right and would agree with the numbers
involved.  Glad you "finally" got an MMC ;-))

Great test!!

Cheers,

	Terry


At 05:50 PM 2/20/2002 -0500, you wrote:
>Hello all,
>
>I finally built an MMC using thirty RS Arcotronics 0.047uF, 1500VDC
>caps.  I used three strings, each with 10 caps, and the total combo 
>measures 0.0142uF.  I placed two one-half watt 10Meg resistors in 
>series across each cap.  These MMC's are reported to be the good type,
>and have the desireable polypropylene dielectric.  These are
>yellow and tubular, but I don't have the specs handy.  These are the
>type used by Richie Burnett and Alan Sharpe in England.  I believe
>that Malcolm is using the same type also.  The number on the 
>package is 114-474.
>
>I did the test using my new TT-32 TC which gives 33" sparks using a 
>0.0164uF Maxwell cap, and a 9/25 NST with sync gap.  I decided to
>replace the Maxwell with a different Maxwell measuring 0.0147uF, 
>to make the comparison with the MMC more comparable.  After 
>retuning, this gave the same 33" spark.  I adjusted the coil to be 32" 
>from the wall, and the sparks hit the wall about every 10 seconds 
>or so.  Then I replaced the 0.0147uF Maxwell with the MMC, and 
>tried retuning and rephasing slightly for best spark output.  It turned
>out that the same tune point and phase point seemed adequate for
>both capacitors.  I couldn't find any spark length advantage by 
>changing the tuning as I switched capacitors.  I did however do the
>tests for each cap with a variety of tune and phase positions.
>The spark was similar, but was slightly weaker using the MMC, 
>and hit the wall only about every 20 seconds or less.  The MMC 
>got somewhat warm within a couple of minutes of intermittent run
>time.  As a verification, I re-installed the Maxwell, and the sparks
>were again stronger than the MMC sparks.  I was somewhat 
>surprised that the MMC sparks were weaker, because I really 
>didn't expect to be able to see any difference in spark length.  
>I was also surprised by how quickly the MMC's heated up.
>
>Next, I moved the coil 1/4" closer to the wall, and this caused the 
>MMC sparks to hit the wall as often as the Maxwell sparks.  This
>suggests that the Maxwell sparks are about 1/4" longer than the
>MMC sparks.  I don't know if other MMC brands would perform any 
>differently.  It may seem difficult to detect a 1/4" difference in spark
>length, but the MMC consistently underperformed the Maxwell
>through all the test runs.
>
>Despite this difference (less than 1% in spark length), I find the 
>MMC's to be convenient, small and lightweight, flexible, and low
>priced for small coils, and the performance is close to the Maxwells. 
>Therefore the MMC's are a good choice for many coilers. 
>
>Measuring spark lengths to within less than 1% is difficult, so 
>perhaps other folks will do their own comparisons of this sort 
>and report their results.  I reported the full specs of this Tesla
>coil a couple of weeks ago.
>
>Cheers,
>John
>