On Wed, May 11, 2011 at 12:19 AM, Joshua Cude <joshua.c...@gmail.com> wrote:
> > > On Tue, May 10, 2011 at 9:12 PM, <mix...@bigpond.com> wrote: > >> >> >> This is based on the assumption that the actual operating temperature is >> indeed >> 400C @ 15 kW. If it's in fact much less, then 130 kW for a short period >> may not >> be a problem. Perhaps it only gets up to 400C when the output is really >> high? >> >> > That's true, so we can try to work in the other direction. If it's 400C @ > 130 kW, then at 15 kW it would be 370/9 + 30 = 70C. That seems rather low to > be able to heat water flowing through at 1 L/s by 5C. > > > Taking the temperature at 1500C (mp of steel) for the 130 kW spike, would > give 1470/9 + 30 = 190C at 15 kW. If the heat is transferred through copper, > then the limit would be its melting point at about 1100C, giving about 150C > @ 15 kW. Those values still seem pretty low, but maybe it's possible. > > > One can also try to calculate the necessary area required to transfer the > claimed power. The range of heat transfer coefficients for liquid water is > huge, but even at the highest value I found (10,000 W/m^2K), this would > require an area of 1.5 m^2 to transfer 15 kW at 40C temperature difference > (70C), or about .38 m^2 at 160C temp difference (190C). For a one inch id > pipe, this would require a 5-m length or 1.2-m length for the two cases. > Both seem hard to believe. > > > On the other hand, for a temperature of 1000C, you could get 15 kW with a > 20 cm 1" pipe. That begins to be believable, but rules out 130 kW. > > Let me once again correct my post-midnight errors, and at the same time improve the estimates, and withdraw this particular objection to the 18-hour experiment. I found a calculation of heat transfer coefficient for a situation reasonably close to Rossi's 18 hour experiment: See example 2 at http://www.arca53.dsl.pipex.com/index_files/conv4.htm. There, the flow rate is about 1.5 L/s through a 25 mm pipe and the mean water temperature is 40C. For this situation they calculate an overall heat transfer coefficient of nearly 12 kW/m^2K. For a deltaT = 160C (pipe temp 190C) and power transfer of 15 kW, the necessary area is 15 / (12*160) = 0.0078 m^2, corresponding to a 25 mm (id) pipe length of .0078 / (pi*.025) = 0.1 m. This seems entirely plausible. So if these calculations are right (and I admit a strong possibility of errors; I'm no expert), then 130 kW could be achieved with a 10 cm pipe at about 1500C, and still get 15 kW at about 200C.