At 06:10 PM 2/26/2011, Joshua Cude wrote:
It is characteristic of artifacts that as experiments improve, they
get smaller. On the other hand, real effects -- desirable effects --
almost invariably become more pronounced as more people work on
them, and the experiments get better. This is true even if the
theory is not understood, just from systematic, or even
non-systematic, search of parameter space.
This is correct.
What is incorrect is Cude's claim that smaller effects -- in smaller
experiments -- are necessarily demonstrating this. The
Pons-Fleischmann effect is not small, necessarily. If experiments are
down-sized, which is wise from the point of view of experimental
cost, effects can be smaller, but the issue would be signal to noise
ratio, the statistical significance of the results.
Eventually, CF results became overwhelmingly positive, as to
correlation of the effect with experimental conditions and other results.
This does not mean that there is no artifact. Some artifacts do not
disappear when "experiments improve," until the artifact is
identified and controlled.
Better conditions and recipes are found, reported, repeated, and
extended. It's the way science works.
Technically, this is engineering. Engineering is difficult when there
is no guidance from theory. With cold fusion, better conditions and
recipes were found; however, the effect is apparently fragile,
dependent upon still-unidentified conditions, and thus, many times,
even with no apparent change in conditions, results that were clear
and striking have been, the next time, entirely missing.
And that class of experiment, P-F cells, is expensive and
time-consuming. One experimental run might be a thousand hours.
Especially science that can be performed on a benchtop.
Sounds easy, eh? Elsewhere today I see a comment from Jed Rothwell
about obtaining material that might match the original P-F material,
which was apparently high-performing. The company was willing to
fabricate it. However, there was no guarantee that this material
would work, because non-recorded details of fabrication, like perhaps
they waited a day to do an operation the first time, then did it
immediately the next time. However, the company wanted a 1 kg.
minimum order. Palladium prices have fluctuated greatly. At the
moment, the Kitco palladium price is over $800 per ounce. That's
relatively high, it's been lower (and it's been higher). That would
be $29,000 per kg. Plus some probably quite expensive fabrication.
Spare change?
And a thousand hours of benchtop time?
Some CF experiments are cheap, relatively. I was told by Jed, when I
first started checking out possibilities, that it would cost,
minimum, about $8,000 to do something minimal. He was thinking about
calorimetry, etc. Miles, at ACS 2010, announced a cheap calorimeter,
about $50 in materials. This was widely ridiculed as being of no
significance. However, that calorimeter might make it possible for
more people to do various kinds of investigations. Cost is not
irrelevant, when there is no big funding available, and, in spite of
the suggestions of both U.S. DoE reviews, the pseudoskeptics made
sure that no DoE funding appeared, and attacked not only other
sources of funding, but the real elephant in the living room of the
extended work of science, grad student labor. As soon as it got
around that if you worked on, say, cold fusion research showing
tritium production, your work might be tossed in the trash by your
thesis committee, there went that supply of labor. No matter what you
thought about cold fusion. People have to survive.
I have designed an experiment that someone could perform for about
$100, plus their own time (little) and a controlled-current power
supply. However, to do this, I spent about $6,000, buying materials
in bulk so that I could sell them for $100 and make a modest profit.
I also bought equipment that people might have lying about, not
needed for the specific $100 experiment, such as a digital storage
oscilloscope, measurement equipment and scales, etc.
Nobody else, as far as I know, has been attempting to do this. It's
taking me time, more time than I expected, simply because of my own
personal difficulties, business responsibilities, children, etc. But
I'll get there.
It takes time. While there have been hundreds of research groups
involved with cold fusion research, there is no overall coordination
of efforts, and a lot of work is wasted and/or duplicated.
I see a strange phenomenon with the work of Kim and Takahashi. Both
are working on theory, that Bose-Einstein condensates are involved in
cold fusion. Both are publishing in mainstream journals as well as
elsewhere. And they don't cite or mention each other, as far as I've
seen. I do not know why, but the omission is glaring.
Much cold fusion research has been an attempt to do just what Cude is
suggesting would happen. Many experiments consisted of varying a
parameter, adding some suspected "secret ingredient," and seeing what
would happen. And the common effect: nothing. There goes a thousand
hours, and a fair chunk of cash, with nothing to show for it but a
pile of apparently useless data.
And, unfortunately, that data has often not been published. So,
later, someone else wastes their time on the same dead end. Or was it
a dead end? Since there are uncontrolled conditions, that's obvious,
what if a magic ingredient was added and it *would have increased the
excess energy*, but a necessary uncontrolled condition was missing?
It's obvious to me that the field could benefit from coordination and
cooperation. Two knowledgeable theorists working together, even if
they are arguing and debating and disagreeing, should be able to
progress more rapidly than two working alone. All cold fusion
research results should be published, not just "successful" results.
And there should be more usage of tightly-specified protocols, ones
that show a high percentage of success at demonstrating the anomaly.
When that's done, it becomes possible to compare results, analyze
them for statistical signifigance, etc.
The standard protocol becomes a control, then, for exploration of the
parameter space. The protocol design can then be improved, to be come
a new benchmark.
I think that Technova was running hundreds of cells at a time,
perhaps. Without theory, this was, and remains, painstaking and
expensive work. However, many hands make short work. Inexpensive CF
cell design could make it possible for many to cooperate, and
amateurs can join in.
The absence of patent protection -- courtesy, again, of the
pseudoskeptics -- means that researchers sometimes have a financial
motive to keep their research secret.
It's a mess.
There are many ways to run a CF experiment and see nothing, and only
a few that are likely to expose the beast, sight the chimera. Hence
my approach, to make it possible for amateurs, students, down to the
high school level, to run experiments, involves premixed electrolyte,
hopefully shown to work, to produce a published effect. It involves a
cell that is, even if on a small scale, manufactured to close
specifications, from the same materials. No guesswork. My efforts may
be far from perfect, and I could fail. But you never succeed if you don't try.
(Imagine the Galileo protocol if the cell had been provided, with a
vial of electrolyte, premixed. Pour it in and run the current
protocol. Then send in the CR-39 for development. You get it back,
and you get a computer scan of the surface, which is also collected
by the coordinating group. Or "company that sold the cell and offers
the development service." And an adventuresome experimenter could do
any or all of this themselves, and could vary the protocol, add
stuff, pee in the cell if they want, to play on an old anecdotal
story from 1989-1990. It would all come out in the wash. You want to
trash cold fusion work, unfairly? Buy a cell and pee in it.... I
rather doubt this would happen often!)
A good example is high temp superconductivity, where higher
critical temperature and critical current densities seem to be
reported every year (although many claims do not bear up under
scrutiny), in the absence of a widely accepted theory. (The
relatively slow progress in plasma fusion is not difficult to
understand when you realize that experimental iterations are
measured in decades, considering progress should be expected to be
exponential in the number of iterations.)
Given that you understand this, Joshua, you should then not be so
influenced by slow progress with cold fusion, which also requires a
long time for each "interation." With plasma fusion, the theory is
quite well-understood, as to the basic physics, and the question
reduces to what might be called engineering theory, how to confine a
plasma that is massively energetic, long enough to get decent fusion
rates. Indeed, it was believed in the early days that this wouldn't
be so difficult.... I do recommend reading Seife, Fusion in a Bottle.
There are, in fact, desktop hot fusion devices. There is even a
handheld device, I think (piezonuclear fusion, used as a neutron source). So?
