James Bowery <[email protected]> wrote: > > "Is the diseconomy of scale primarily driven by the large number of >> thermocouples implied by the squared law of the surface area?" > > > The answer was "Yes." >
Right. Yes. Note that this is not a problem with flow calorimeters, which is why I recommend one for this application. Abd ul-Rahman Lomax <[email protected]> wrote: > I am not familiar with Celani's specific methods. There is nothing to it. He measures the outer surface temperature of the cell. It comes to 120 deg C with 48 W input when the ambient temperature is 30 deg C. It comes to that temperature whether you use H or Ar, which I think is a pretty good indication the response is predictable. McKubre disagrees with me. He thinks that changes in gas conduction from the heater at the center of the cell to the outer surface might change the temperature. He also thinks the temperature may be inhomogeneous. I doubt such problems can be as large as 20 deg C. Has someone posted a photo of the device? You can see the TC mounted on the outer surface. There is also a TC at the core but it is not used for calorimetry. > I understand that there can be problems with the kind of calorimetry that > many use, informally. > There are definitely problems, but I do not think they are big enough to cause a 20 deg C false reading. Let me define what I mean by a "false reading." There is no chance the instruments are registering incorrectly. When NI installs $25,000 in equipment, and it measures a 20 deg C temperature increase, you can be sure that increase is real. The only question is: Does it come from internal heating, or from change the physical conditions? As far as I know, only two kinds of changes can happen: 1. Faster transport of heat from the core to the surface. In other words, decreased insulation. A coffee cup surface is a lot hotter than a thermos bottle surface because the heat escapes faster. For example, if you were to let some of the gas out of the cell, the core temperature would rise and the outer surface would cool, because the gas would not transport the heat out as quickly (mainly by convection, not conduction or radiation). In fact, this cell is leaking slightly. If anything, that should cause the surface to cool, and the metal at the center to heat up. Both heat up in this experiment. 2. Surface temperature inhomogeneity. In other words, uneven heat distribution, such that the surface is hotter but that does not actually indicate a real power increase. That would mean some other part of the surfaced is cooler. I am sure there are inhomogeneities but I am also sure they are far smaller than this. I say that based on the numbers from Mizuno's gas calorimeters. This reminds me a little of Taubes' claim that thermal gradients can produce a cell temperature "say fifty degrees hotter on one side than the other." No, they can't. 1. A thermal gradient is vertical not horizontal. Heat rises. It does not go South. 2. A gradient in liquid is much smaller than that. See: http://www.lenr-canr.org/acrobat/RothwellJcoldfusion.pdf By the way, the fact that this cell leaks so much precludes the possibility of doing mass spectroscopy. - Jed
