[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: New MMC has MASSIVE issues
Original poster: "Gerry Reynolds" <gerryreynolds@xxxxxxxxxxxxx>
Hi JT,
Three things come to mind.
1. Use polypropylene dialectric (others have said that) CDE 942C
series caps are excellant.
2. Spacing is important. The outer foil is just beneath the surface
of the cap and if too close to some else, it could arc thru. Look at
these links to get an idea of one way to insure no intercomponent breakdown.
http://hot-streamer.com/temp/GerryReynolds/15KV_120ma_SRSG/tc01_stackable_cap_cage.jpg
http://hot-streamer.com/temp/GerryReynolds/15KV_120ma_SRSG/tc02_stackable_cap_cage.jpg
http://hot-streamer.com/temp/GerryReynolds/15KV_120ma_SRSG/tc13_capacitor_banks_56nf.jpg
Note in the first two pictures, there are no bleed
resisters. However, they are present in the third picture.
3. You say that your cap is "matched" to your NST. I'm not sure what
you mean but every NST will series resonant with a particular cap
size. If this is the cap size you are using then this is a way to
get extreamely high voltages. Theoretically, at resonance, the
output voltage could become infinite. Practically, the resistance in
the NST will limit the resonant rise to the circuits Q * rated out.
Q= sqrt(L/C) / R where R and L are the effective R and L of the NST
and C is the Cp value you use.
Q could easily be 5 or more meaning that a 12KV NST could resonate up
to 60KV or more, if your safety gaps are set too wide and something
else does die first (like the NST or cap). This is one reason why
LTR operation is recommended where, for static gaps,
Cp=1.6*Cres. Cres is the resonant cap value for your NST and is
calculated by the following:
Cres (nf) = 10^9 / (2*pi*line_freq*XL)
where pi = 3.14159
XL = Vs_oc/Is_ss (inductive reactance)
if you are using a 12KV 30ma NST then XL=12000/.03 = 400000
ohms Tell us again what NST you are using and what is your Cp value
you end up with. Also when designing an MMC, Peak current, RMS
current, and Voltage ratings need attention to. Many on this list
will help you with the MMC design once you have selected an
appropriate capacitor.
Gerry R.
Original poster: "JT Bowles" <jasotb@xxxxxxxxxxx>
Hi, remember me saying i am creating a new MMC, rated double my previous cap?
I did that!, and it is closer to matching my NST also.
BUT HERES THE ISSUE:
*If I look carefully I can see dielectric breakdown on 2 of 12
capacitors, while in use
*The Mylar heats up fast
*My MMC randomly sparks from one capacitor to another. Sometimes it
wont do this at all.
HOW IS THIS POSSIBLE! when I had an 18KV rated MMC, and it gradually
broke down. But when i inserted my new MMC, rated 36KV, the MMC
began to degrade faster than my old one.
HOW IS THIS POSSIBLE!!!!!?
Wow, i just want a working MMC.
Any solid ideas as to why this is happening, terry? It really is
very dissappointing.