why do you need the needle anyway? perhaps there is a better way of doing whatever you hope to accomplish
Another cool effect: charge a big cap, then short-circuit it by putting some steel wool between the two legs... Awesome AND reuseable!
Hm I'm basically trying to expand my horizon 
finding out what is possible... getting to know the limits of different applications, at the moment I'm playing around with different extremes.
In the end I might build a mini-exhibit featuring a miniature jacobs ladder (high V low A) glowing needle (high A low V) a laser... stuff..
So yes it has to be a needle , if you want people to be interested in your project, using recognisable elements is important.
Can anyone enlighten me on how to convert Joules to glowing needle degrees?
And another question: From several high current appliations I've seen people always use some kind of charging resistor. To prevent overheating I guesss, but is there any formula I can calculate such a thing with?
P.s. 1000°C isn't necessary but a nice red-orange glow would do. Also this is not inteded for continual use just one "flare" (I will call it: JANIO: Jazzars Awesome Needle Instant Oxidizer^^)
Can anyone enlighten me on how to convert Joules to glowing needle degrees?
I think you need to know the specific heat capacity of the substance you're trying to heat.
Can anyone enlighten me on how to convert Joules to glowing needle degrees?
Here is the equation you need:
Q = mC(T2-T1)
note: (T2-T1) is generally written as deltaT
Q = Heat energy(Joules)
m = mass(grams or Kg depending on which val used for 'C')
C = specific heat capacity(Joules/(gram*degreeCelsius) or Kg...
So you want to find 'Q'. You need to know the 'C' of your needle. I would just look for specific heat of steel. And you need to know its mass. As for the change in temp... I don't know what temperature makes things glow so you will have to guess. I think its completely possible, but figuring out how to discharge the capacitor(s) in an effective way might be challenging. Be sure to post some pictures of this when all said and done.
Well, I think it's possible to get a filament to glow given a small single-cell source. I've seen these:
And apparently this isn't a joke:
Hmm specific heat capacity of steel is 0.46 J/g*K that means I need 0.46 Joules of energy to heat one gram steel by one Kalvin
So from 25°C to 1000° that makes (average needle about 7.5g):
(0.46 J/g*K) * (7.5g) * (975K) = 3363.75J
seems a bit high...
@ halley ...wow just.. wow...
@ P_Wood Thanks
(average needle about 7.5g)
What sort of needle is that heavy? A sailmaker's?
seems a bit high...
Seems about right. Though, depending on the amount of time it takes, heat may radiate from the needle into the air before it ever gets to glow.
Question: What is the resistance of this needle? 
What sort of needle is that heavy? A sailmaker's?
lol ;D
Different metals have different ranges at which they melt. I used to be diy forging with different types of materials. The propane fired forge I built could get up to 2500 degrees f or even more If I redesigned it. If good at it you can tell about what temp the metal is by the color of it being heated.
Nicads will do it, just one would be enough @ 1.2v, nicads give all they've got in a short space of time, nickel metal hydride on the other hand are made for duration not high drain, so imagine six 1.2v 2400mah nicads in parallel, thats 14 amps for one hour, you can work out the rest, another impressive sight is 6 x 12v 40 watt solar panels wired in series, i used my multimeter probes to short the whole bank, it created a high frequency plasma stream well over an inch long, actually the metal probe tips got so hot they melted and fell off spoiling my fun
actually a pack of one hundered needles is that heavy caugh
avg. needle = 0.075g
E = 33.64J
With this potent formula: E = 0.5 * U² * C
we know that we want to have energy in the range of about 60 Joules.
At 9V I'd need ~1.5C.
How do get this transformed into Farad? Can I just say:
(1.5 A*s) / (9V) = 0.167F ?
although I'm not sure with the 9V do I have to divide by 9 because I said I'd use 9V in the first formula?
How do get this transformed into Farad?
Not sure exactly what you did there. This is what I know:
W = 1/2CV2
C = W/1/2*V2
C = 33.64J / (0.5*(9V)2) = 0.8306F
or C = 830,600uF
Yes you're right... thats what you get from mixing variables...
I actually thought the C was coulomb instead of capacitance..
guess I should get some sleep.
it created a high frequency plasma stream well over an inch long
What would cause a DC system to oscillate like that?
It is one thing knowing how much energy it takes but another in getting that energy to be dissipated in the needle and not in the wiring from the power source to the needle. The problem is that as the needle has a very low resistance the wiring's resistance has to be substantially lower. This will be a practical difficulty. You will probably have to use things like large jumper leads to connect to the needle. Even then I suspect that most of the resistance will be in the connections. It is the place of the greatest resistance that will be the place of the greatest heat dissipation.
What would cause a DC system to oscillate like that?
Lack of understanding of what was going on. 
@Groove
@Grumpy mike
your right it is my lack of understanding so if you know please explain, the only way i can describe it was a huge purple stream but not with hard edges and not in a straight line, more like a stream from a plasma globe and it had a very high pitched hissing sound, im sure it was very high temperature plasma because 1. it melted my probes in seconds and 2. the longer the arc stretched out (over an inch) there was a most deffinate peak in the middle where the superheated air was rising, the strange thing was that it was only about 112volt when i measured the output and the panels are rated at 40 watts
I think I can solve that with using multiple smaller capacitors like 16 or 25, all with individual wires that lead to one bigger piece of metal, and than the needle.