Original poster: "Terry Fritz" <twftesla-at-qwest-dot-net>
Hi Bart,
We have done a lot of work figuring out how hard MMC caps can be run. The
voltages ratings of caps can be met and moderately exceeded without
problems. I have run EMMC caps well over voltage on all may coils since
day one without a failure. In fact, "I" have never run an MMC cap below
it's ratings ;-))
Geek group caps have been tested to 5700 volts before breakdown and then
they self heal and are good to go again. Panasonics go to 3600 volts. I
really don't feel the voltage is a problem at all with MMC caps. The real
area of concern is current and internal heating.
I watch MMC failures and see no reason for concern. Most MMC problems are
due to people not following the "tips" at:
http://hot-streamer-dot-com/TeslaCoils/MMCInfo/mmcinfo.htm
I used to send a printed copy with each MMC cap order in the good o'l
days. I also sent along 10Meg resistors for free that were also trouble
free over the years.
I understand your concern about specs, but we try to push MMCs to the
limits because it saves so much money and the caps still work fine. Of
course, one can over rate the MMC easily if that is what is wanted.
Most users of Geek groups caps follow the charts made by Mark Broker that
are based of my calculations and work with MMCs. Considering the number of
caps out there, I think the failure rates are excellent and those charts do
run the caps right up to their voltage ratings.
So I understand your concern, but we did our homework first and field data
is fine. Cap manufacturers back way off on "AC" ratings due to ion
inception inside the cap and the damage it does. However, they are
thinking of caps that have to run 24-7-365. MMCs are very short term
intermittent duty devices where ion damage is just not a factor. We have
torn MMCs apart after long use looking for ion damage and it is very light.
"I" would run Geek caps to 4000Vpeak in an EEMMC design and I bet they
would work just fine ;-)) I also bet I know a few things about poly caps
that most of the manufactures don't know :-)) However, Cornell Dubilier
has done a few tricks to extend the art even further!
Cheers,
Terry
At 07:01 PM 11/26/2002 -0800, you wrote:
>Hi Terry,
>
>I'm not as confident that running a 12kV cap bank at 12kVp is a good idea.
>Sure, manufacturers derate their caps, but they always have. Regardless,
>the user should still design in their own derating (MMC or not),
>especially in TC service. I think if we all ran on the edge like this, we
>would be seeing a high rate of failure.
>
>Take care,
>Bart
>
>
>Tesla list wrote:
>
>>Original poster: "Terry Fritz" <twftesla-at-qwest-dot-net>
>>At 9000 VAC, the peak voltage is 12728 volts. Six caps are rated at
>>12000 volts. So you are a little over. However, I would not worry about
>>it at all. Six caps per string should be fine. If you are using 10M
>>resistors, they are at their power rating (0.45 watts) so no problem
>>there either. However, the resistors may get a little warm but spacing
>>them should keep them from heating the caps. If you wanted to be real
>>safe, you could use two resistors in series per cap.
>>
>>Cheers,
>>
>> Terry
>
>