The Maxwell P285 ultracaps are only rated to an absolute max of 2000A. Also, their ESR is 0.22 milli-ohms. I calculated the resistance of each of the copper feed rods to be about 0.11 milli-ohms (copper, 1cm diameter, 500cm long). I presume the droplet shorts the gap because 5V is not enough voltage to spark across a gap of any reasonable size. Presume that the molten silver droplet has a resistance of 0.05 milli-ohms. The total drop-connected circuit would have a resistance of about (0.22+0.11+0.05+0.11) = 0.49 milli-ohms. With a 5V source, this would yield a current of 10200A, about 5x the maximum rated current for the ultracapacitor. Most of this energy is going into the capacitor ESR and the copper bars; only 10% goes into the droplet. So, the droplet is getting (10200A)^2 x 0.05 milli-ohm = 520W until it evaporates and the heat carries away enough silver vapor that the plasma discharge ceases. If I estimate the silver droplet to be at 2mm in diameter and molten at 962C, then it is 407 micro-mole/drop. Presume it gets vaporized and heated to 2500C, then the drop needs 25.4 J/mole-C to heat and 254 kJ/mole to vaporize. So this drop needs 119 J to vaporize and heat to ~2500K (good estimate, Dave). At 5200W delivered, this would take 0.023 seconds to deliver and he would only be able to do about 40 drops a second if you included a gap between drops.
Note that this would correspond to (10200)^2 x (0.22E-3) = 22.9 kW x 0.023s = 526 J/drop deposited as heat into the 4 capacitors, so 132J per capacitor per drop. At 40 drops per second, *each* capacitor would dissipate 5264 kW. These capacitors are going to get hot! Max temperature for the caps is 65C. So, at 40 drops/second and 119J/drop, this would be 4760W of continuous input into silver (not counting the power to melt the silver) and 46.6kW input into the system. The 5V on a 3400 Farad capacitor stores 42500 J. So each shot of 119J would decrease the voltage by only 0.265V which is only a 5% ripple. I think the bigger deal would be the 10x of rated absolute maximum current and high heat as far as the life of the capacitor would be concerned. If you look at the chart on slide 49, where shows the measured spectrum, he also says that the power integrated over 4pi steradians is 527kW (sounds a little like a Lugano measurement, but lets take it initially at face value). If that is true that 527kW of radiant energy is produced, with the estimated input of 46.6kW, that would mean he is realizing a HEAT COP of about 11. Now, if you figure that 50% of that heat goes into heating the tungsten plate that is converting the heat energy into a blackbody suitable for the solar cells (optimistic), and the solar cells are 30% efficient, then you get a net electrical COP of 1.7. So, if you hooked the electrical back to the input to get it to self-sustain, you get a remaining electrical output of 0.7 x 46.6kW, which is 33kW of net electrical power output from the self-sustaining device. Interesting calculations. On Sun, Feb 7, 2016 at 10:20 AM, David Roberson <[email protected]> wrote: > Bob, if each drop requires 100 joules of energy to vaporize then Mills > will need 100 kilowatts of power when 1000 shots per second is the cycle > rate. That amount of drive would seriously impact the COP figure that he > is achieving. Do you recall any mention of the number of joules required > to ionize the silver drops? My gut feeling is that 100 joules would be on > the low side of that requirement but I have not performed the calculation > so far. Perhaps someone has done the math and would like to offer their > data. > > If we assume 100 joules of energy per pulse then we can calculate the > amount of delta voltage required from the capacitor. I looked up the part > number and find it is 3400 Farads. If the initial voltage is assumed to be > 5 volts across the series/parallel combination then the drop becomes 6 > millivolts per shot. This is quite low and thus the current supplied from > the solar cell system would appear to be relatively constant during the > device operation. Since the capacitors are not actually discharged to a > significant degree I would be surprised if their life expectancy is > comparable to what is seen for complete discharge and recharge. > > Do we see any evidence for the amount of power required from the DC source > that is currently powering the experiment? I doubt that it can supply a > steady 100 thousand watts at 5 volts, which would be 20000 amps of DC. > This requirement suggests that my original estimate of 100 joules per shot > must be too high by a factor of 10 or instead, the actual firing rate is > 100 shots per second or less. Perhaps Mills intentionally failed to > discuss the DC input power requirement in order to avoid the contradictions > that exist. Or, the device demonstrated is not capable of the power rating > that they believe will eventually become possible with a significant amount > of engineering. > > Would you like to add to my speculations? > > Dave > > > > -----Original Message----- > From: Bob Higgins <[email protected]> > To: vortex-l <[email protected]> > Sent: Sun, Feb 7, 2016 11:11 am > Subject: Re: [Vo]: BLP demo video > > Sorry, I mis-counted my divisions ... the supercaps would expire after > 16.7 MINUTES - they are only rated for 1M discharges, so at 1000/sec, you > get 1000sec or 16.7 minutes. > > On Sun, Feb 7, 2016 at 8:45 AM, Bob Higgins <[email protected]> > wrote: > >> One of the things I noticed in Mills' apparatus is his use of supercaps - >> in this case Maxwell P285 supercaps. Supercaps sound great until you dig >> into the details. Supercaps are somewhere between a battery and a >> capacitor in specifications. One of the core specifications that is a >> problem for repeated discharges is the rated number of charge-discharge >> cycles. These supercaps are rated for 1M charge-discharge cycles >> (lifetime), which sounds like a lot compared to a battery. However, if you >> were doing as Mills describes and operating at 1000 pulses/second, these >> supercaps are going to expire after 16.7 hours of operation. This is a >> fundamental characteristic of supercaps, and makes them an unfit component >> for use in a power system. Such critical components should be rated for a >> minimum of 10k hours of typical operation to create a reliable system. >> This will be a painful engineering problem to overcome, but not >> insurmountable. They may just have to use a huge bank of conventional >> capacitors. >> > >
