John Fisher looked closely at the Rossi claims, including the patent
and blogs. He learned a great deal more than I did from these
sources. He has come to the conclusion that what Rossi is seeing can
be explained as an ordinary physical effect; the storage and release
of heat. The machine starts up roughly 4 kW of heat and then releases
3.2 kW, and then stops. The other 0.8 kWh is not captured by the
calorimeter. According to the blogs, the machine stops after about an
hour of operation. Rossi thinks this is a technical glitch which they
will soon fix, but Fisher thinks the machine has simply run out of
stored heat, and at this rate of heat release it cannot go for more than ~1 h.
Fisher is working on a detailed paper about this. The paper is not
ready yet, but he sent me the following preliminary note.
- Jed
- - - - - - - - - - - - - - - - - - - - -
In the startup phase of a run they turn on a 1000W electric heater
for four hours to heat up the nickel powder and its surroundings to
high temperature. Then they turn off the startup heater. 4 kWh of
energy has been stored as thermal energy. Because energy is
conserved, all of it will come out as the machine cools down. If the
machine is well insulated except for heat flow to an external
calorimeter, most of the energy will flow out through the calorimeter
where it can be quantified. Perfect insulation and a perfect
calorimeter would retrieve 100% of the energy. A less than perfect
system would retrieve less, for example 80%.
It was important to me that I should understand how it could come
about that Focardi and Rossi could have felt justified in omitting
startup energy in their energy balance. After studying their work I
deduced that they had misinterpreted the influence of hydrogen
pressure on the output power of the machine, attributing it to an
influence on a nuclear reaction rate where it actually was an
influence on thermal conductivity of the nickel powder. This is an
understandable human error. . . .
Yesterday I summarized my analysis and conclusion in the following
note to the CMNS group. However I do not intend to restrict it to
this group, and you are free to circulate it or otherwise use it as you wish.
Colleagues
Focardi and Rossi claim that the Rossi energy-amplification machine
generates a nuclear energy output more than 100 times larger than the
energy input. I have grave doubts that a nuclear process can be
responsible. The output energy may simply be withdrawal of startup
energy that was introduced at the beginning of each experimental run.
Relevant information can be found on the web site
<journal-of-nuclear-physics.com>. Startup energies ranged up to 4.0
kWh but were not included as input. The most recent of the tabulated
experiments (method C) ran for one hour and produced output energy of
3.2 kWh, only 80% of the possible 4.0 kWh startup energy (the actual
startup energy was not provided). No need for nuclear reaction was
demonstrated.
Other experiments were performed with different calorimeter designs
(methods A and B). Method-A experiments ran for from 1 to 1.5 hours.
Each of the three tabulated energies for method A appears to
correspond to combined results for many experimental runs, or perhaps
to a single run normalized to 24 hours operation at the experimental
power level. In method B, daily experimental runs appear to be
normalized to 24-hour operation. (Quantitatively the method-C output
energy scaled to 24 hours is comparable with the tabulated method-A
energies and to the tabulated method-B daily energies.) In each of
these interpretations the appropriate correction for missing startup
energies leads to output energy less than startup energy. Again no
nuclear reaction was demonstrated.
The question can be settled by a single experimental run in which
total input energy, including startup energy and machine operation
energy, is measured from a cold start.
John Fisher
