On 12/28/2009 11:59 AM, Terry Blanton wrote: > On Mon, Dec 28, 2009 at 10:43 AM, Jones Beene <jone...@pacbell.net> wrote: > > >> Essentially that is what happens in a piston or Stirling engine, no? >> > With the size of those error bars, it's difficult to say what is going > on in some instances. But the phase one energy output of the H sample > in the PZ is a real puzzler. And why the heck does it match the D? > Is it something Casimir? Or is it an error? >
Speaking of error bars, I see what may be a nit and I have a question about it... In table 1, they give the average loading of D into PdZr as 1.1 +/- 0.0. That appears to mean 1.1 D per Pd with a zero sized error bar -- the result is exact. Is that a correct reading? I find that puzzling because the process they describe for measuring loading doesn't seem likely to lead to an exact value. They say (p. 4, first paragraph): " After the gas is introduced, pressure does not begin to rise for a while. During this phase (the first phase) the Pd powder absorbs almost all of the D2 (H2) gas atoms as they flow in, and heat is released as a result of adsorption and formation of deuterides (hydrides). After about 30 minutes, the powder almost stops absorbing gas; the gas pressure begins to rise, and the heat release from deuteride (hydride) formation subsides. This is the beginning of the 2nd phase, and the gas flow rate in the 1st phase is evaluated from the rate of the pressure increase. From the flow rate multiplied by the duration of the 1st phase, loading is estimated ... " This description was for loading determination during the runs using Pd powder; they don't repeat the description for the other runs but one would tend to assume it would be about the same. So what we seem to have is this: They time phase during which pressure doesn't rise, then they measure the rate of pressure rise once the Pd gas absorption slows down, and they use that measured pressure rise, along with the duration of the constant-pressure phase, to *estimate* the amount of gas injected into the container. Using that, plus the weight of the Pd, they arrive at an estimated value for the loading. Is that how other folks understood this? How can this approach lead to a zero sized error bar? Surely there is a good bit of wiggle room in a number of the steps in forming the estimate, or so it appears to me. > Listen to me, I sound like a skeptic. :-) > > Terry > >
