[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [TCML] Voltage - Gap
On 1/31/12 12:06 AM, jhowson4@xxxxxxxxxxx wrote:
Ah yes I see the posts you are talking about back in the archives.
I am going to spend a little bit of time reading threw this
material!
So for the 2000Vdc cap, and the heat dissipated for the standard .15
2Kv would be .9W, seems reasonable. I don't really see how this is
all that useful to me yet, but i am sure after reading thew the
material you helped me find, I will learn what to do with it.
Actually, dissipating almost a watt in something made mostly of plastic
might be pushing a bit. I guess they're quite a bit bigger than, say, a
2W resistor, but thinking in terms of stuff you have around the house
that dissipates a watt in a small area.
One way to estimate the power dissipating capability of the capacitor is
to look and see if there is a derating curve for the case temperature.
Going back to what you said earlier, with before you work too hard
modeling fine details, do a simple subset and see if it makes a
significant difference. You might find that in an RMS current,
heating sense that it doesn't make much difference between one big
pulse at Imax and 3 successive pulses at Imax/3, because the duty
factor is lower at the former. (Yes, heating goes as the square of
current, so that's clearly not true in this specific example, but
it's what I was able to come up with before my coffee as a direction
to look)
Its definitely something I need to consider, does this really
matter? Well I am having a good time learning about all this stuff in
the process, so why not try and figure it all out. Plus if i do i
might be able to correlate some interesting things and add to the
general knowledge base for everyone to learn from.
You make a good point. However, I find that it's nice to do a couple
cases by hand that sort of bracket what you're interested in. Not only
does that make sure you have the conceptual understanding, but it also
lets you spend your time effectively. You might be trying to model
something where there isn't any good data to model from and you wind up
chasing what is essentially experimental uncertainty in the raw data.
Maybe another term is to do a sensitivity analysis.
another interesting thing about my coil in particular, Because I am
basing it on Steves DC design, but optimizing the parameters, I can
pretty much assume that the tank cap will be fully charged on what
ever time scale I want it to be charged on. meaning that I can pick
my BPS to be what ever I want and still achieve the maximum bang
energy, assuming I have the current to push the system of course. I
am designing for 600bps because that's the maximum my gap will be
able to push. Hence my heavy interest in the RMS current, because
that means a huge difference in price of the MMC. Its will all be too
expensive for me anyway and I will be making my own poly caps in the
end. But its been an interesting and knowledgeable exercise thus far,
so I might as well keep going. FOR SCIENCEEEEEE!!
I don't know how much you consider your time is worth, but it's unlikely
that making your own poly caps will work as well as either surplus pulse
caps or a MMC. Making the rolled poly cap isn't all that hard, it is
somewhat time consuming. Although, scrounging appropriate materials is
harder than it seems. It's not like you go down to the supermarket and
buy aluminum foil and cling wrap. The archives are full of stories
about where to get the right plastic in the right thicknesses, and where
to find aluminum sheet.
And then there's the whole "how do you attach a wire to the aluminum"
question. After all, you don't want a nice low impedance, low ESR cap,
with high AC resistance wiring connecting it.
And the oil fill and pumpout/degassing process. Did I mention the
inevitable leaking oil?
This all said, making your own capacitors is a satisfying project: you
DO get to say "I did it myself", and some 15-20 years later you can
write this post<grin>
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla