Xian Zhang Hang : nickel-hydrogen cold fusion experiment now quite reproducible

Quote from Thomas Clarke

(1) I never mentioned any theory. Nor is one needed. This is standard experimental methodology for flakey stuff.
(2) The issue is that until you understand your experimental methodology, including all its error artifacts, very well, you cannot know whether some change you make is optimising COP or optimising error. And no-one else can know whether your claims of COP are real or simply experimental error.

"Flakey stuff." Well, the saying is "extraordinary claims require extraordinary evidence." Commonly said but not commonly understood.

Cold fusion was very, very unexpected. Hence strong evidence was ultimately required, and that evidence mostly did not exist until 1991 -- and even the 1991 evidence wasn't confirmed immediately, but gradually over more than a decade.

Before then there was circumstantial evidence of a nuclear reaction, such as many people reporting anomalous heat, sometimes at significant levels. And then all kinds of other odd reports, mostly uncorrelated with the heat and therefore all suspicious. The direct evidence that was discovered was helium correlated with heat (and roughly commensurate from a fusion expectation, though not the classic fusion reaction).

NiH, we do not know what the ash is. That evidence, so far, doesn't exist, quantitatively. We are not even sure about the heat. So at this point, the default is "experimental error." Realize that this has commonly happened, there are many, many things that can go awry. So with each experiment, the sane approach is to stick with the experiment until it is understood, not as to theory, but as to the actual measurements.

(Before Miles, 2011, there were anecdotal reports of helium. What Miles did in 1991 was to correlate the helium with heat being produced, across many measurements. This is very different from merely "finding something unexpected.")

At this point, it is possible that the NIH reaction product, if NiH reactions are taking place, will be very difficult to detect, i.e., if it is deuterium. Looking for the reaction product at this point is not where I'd place priority. It would be in nailing down heat, though careful controlled experiment.

Yes, controls, and there are many ways to confirm calorimetry. One measurement isn't nearly enough. How do the results vary with the fuel? This experimental device gets hot on the outside. How does that compare with fuel/no fuel?
Songshen Jiang turned on a cooling fan in the middle of his original experiment, thus changing conditions. All uncontrolled changes confuse results.

The goal of a scientist here will not be to "improve COP," it will actually be to show that there is no unexpected heat, by demonstrating artifact through controlled experiment. Then ... if that effort, diligently pursued, fails .... then we have something to publish! And especially all the efforts to disconfirm! By this time, the experiment is understood well enough that one can describe all the details so that someone else can confirm or disconfirm.

Quote from BobHiggins

drive.google.com/open?id=0B5Pc25a4cOM2Uzk3NzdfMl9BeE0

This is quote nice. Yes, lots of missing details. Hydrogen generator? How used?

Why not a continuous calibration curve? I dislike the heater averaging, and using thermostatic control for the calibration, rather than constant input power, could introduce errors. With a continuous calibration, it would not be necessarily to aim for a particular temperature, one would just look at temperature vs input power.

The experiment assumes constant conditions. What if cooling conditions in the room changed between the calibration and the experimental run. Suppose, for example, there was a fan running in the room during the calibration and not during the experimental run? This would create something like these results. There are other possible changes.

Against temperature, I plotted input power, inferred excess power, and the ratio of excess power to input power.

What I could readily sees, with that plot: the input power of 569 W for a temperature of 1051 C is anomalously high, thus the inferred XP is anomalously low.

The ratio of excess power to input power (which is COP minus 1), is for the first three measurements, 15.4%, 16.0%, 15.3%. As mentioned, the third of these may have a measurement error or something, it would be higher than 15.3%, perhaps. (I'd expect it to be higher than 16%).

Then for the next two measurements, it is 23.2% and 27.1%. Somewhere around 1050 C there is an increase in XP. Below that temperature, the XP is a constant ratio of input power (from the measurements we have).

I would want to see the behavior at startup. From other work I would expect no XP at low temperatures, but it looks flat from 868 C to 1051 C. If there is real XP here, it would be starting at roughly 1050 C.

So what explains input power lower than calibration at those lower temperatures? This is a phenomenon that should immediately arouse suspicion of artifact.

To clear this up, I recommend exploring the lower temperatures, all the way down. I would monitor other temperatures: ambient, surface of the apparatus, internal temperatures. I would test this with various fuel mixtures.

I would run a long-term test at high power, long enough that if there is a nuclear reaction, there would be measurable products; in particular, one would want to run for long enough that if the product is deuterium from hydrogen, there would be deuterium well above the natural level in hydrogen.

At the same time, the long-term behavior at various temperatures would be studied. And calibration would be confirmed after the test as well as before, with as little change as possible.

Great test, if they nail it.

Quote from Hank Mills

The power of the Rossi Effect means that to PROVE the reality of the effect (once we figure out the critical parameters we need to allow for self sustain) we don't need to perform the style of testing you propose.

"Proof" is for non-scientists.

There is an effect observed here, an apparent increase in the necessary power to maintain a temperature, from which XP is inferred. The "Rossi effect" is not established. Period. This work from China is open and clear and apparently verifiable (though there are some details missing.)

What we look for in science is measurable results, predictable and independently confirmable, and then confirmed.

In this field, we must be aware of many possible artifacts, and claims of high XP are suspect, until confirmed. Low XP is enough of a problem! Self-sustain is known under some conditions. However, arranging it experimentally is often inconvenient.

For example, if there is XP in this experiment, then running it with good insulation could create self-sustain. That is, assuming that a temperature is needed for the reaction to take place, and if, at that temperature, there is heating, the heating can be enough to maintain the temperature with insulation. However, the insulation creates other problems with control and measurement. The input power is not a problem unless we are looking for some simple "convincing demonstration," which is the kind of desire that kept cold fusion from advancing much for over two decades. What is needed is well-studied and reliable results, a foundation on which to build.

We needed to study the FP Heat Effect as it was, not as it "should be." Miles did amazing work with relatively low XP. If not for the highly suppressive background of the rejection cascade, Miles would have turned the world upside down on cold fusion. Miles was independently replicable and the heat/helium *ratio* was reliable, within experimental error, if not the heat itself.

It was asked about Pons and Fleischmann and "bulk reaction." This is well-known among LENR scientists. The original reports expressed an opinion (or observation with few data points) that the reaction varied with the reaction volume, not with surface area, which was evidence for a bulk reaction. There are still some scientists holding to the idea of a bulk reaction, but the helium evidence is overwhelming at this point. That work is being done with increased precision, and if there is some problem there, I expect it to come out in the wash.

Once Pons and Fleischmann realized that helium would be trapped in the bulk and could not escape from it, once they found that bulk analysis of their cathodes was not finding helium that they expected, I infer, they refused to release their early helium analysis done by Johnson-Matthey. This refusal is what Park reported as having caused him to think this was bunkum. There is a quote from Fleischmann somewhere that he said he didn't want to be fighting on two fronts. Which is a sign that he'd lost his scientific objectivity and was sucked into a battle, where one wanted to avoid inconvenient facts.

Then the Morrey collaboration behavior is well covered in both Huizenga and Taubes. The actual Morrey report was published, after delay. In France, Pons and Fleischmann were producing enough heat that helium analysis should have been easy. But Fleischmann wrote that it was too expensive (and drastically exaggerated the cost).

It is easy to understand how Pons and Fleischmann thought the reaction would be in the bulk. The idea of the behavior of PdD in the bulk was what led them to do the work in the first place. That high loading was necessary also made this appear to be the case.

Quote from Tarun

Jack Cole wrote:
They appear to have measured their temperature through it single spot on the reactor core.

Good point. A hot spot is likely. Although since its on the end of the tube and the tube is made of steel, the chances are less.
So a re-calibration is necessary after the experiment. Probably with the vessel filled with only Ni and Ar.
A calorimetric setup will clear this issue (and perhaps bring up new issues).

Right. A variation in where the temperature is measured, or in the heat behavior of the fuel tube, could explain what I see, the apparent XP at lower temperature. It's possible this could also explain the behavior at a higher temperature. These are the kinds of possible artifacts to be explored.

For a dummy fuel, as one example, I would use the same fuel and heat it to drive off the hydrogen, leaving the nickel and lithium and aluminum, then fill it with argon. The idea is to eliminate as many variables as possible.

Many times, I suggested that the Lugano test would have been much more interesting with a calibration at the full range of input powers. Under those conditions, the use of IR emission would not have been nearly so problematic.

It's pretty clear to me that the Lugano team was aware of the problem, but ... my speculation is that they simply were not allowed to make that test, they gave a BS reason. Who told them that the same input power that was going to be used would cause the heating element to burn out in a calibration?

In any case, this is a very nice report, and studying it cannot fail to inform. Either XP will be confirmed by more careful study, or an artifact will be identified as a caution to future researchers.

If the goal is increased COP, it's only of value if that goal is reached, otherwise it's *failure.*

Ultimately, if a reliable reaction is found, it can be engineered not only for higher COP, but for self-sustain, at least as reasonably defined. One would still want power for control systems. Again, ultimately, if COP is high enough with an experiment that is large enough in scale, the heat can generate power to run the control systems. But first things first.

Until control is shown to be reliable, large-scale is dangerous.