Cold fusion/Nickel-hydrogen system/Rossi/Tests/Review 1

< Cold fusion < Nickel-hydrogen system < Rossi < Tests

This review was written by User:Abd who is generally responsible for the content. This page is open for editing, but material changes should be set off and signed. Outside of this, discussion of this page is welcome on the attached Talk page.

A scientist involved in cold fusion research commented privately to me that there were certain results from this first test that he found remarkable. So I am reviewing this test, looking for what impressed him. He is also aware of shortcomings in this report, but the opinions expressed here are mine.

The arXiv paper: .

I will start with the abstract.

An experimental investigation of possible anomalous heat production in a special type of reactor tube named E-Cat HT is carried out.

This test had been announced by Andrea Rossi, who claimed it was an independent test, implying this was new. However, the lead author was Levi, who had long been associated with Rossi, and the tests (or at least the second and third) were done on the premises of EFA Srl, which is a company controlled by Rossi's wife, apparently.

There were three reactors tested; the way this is described implies a commercial product, "E-Cat HT." The high-temperature versions of the w:Energy Catalyzer are called HT. The three reactors described in this report were not identical, and for the next report, Cold_fusion/Nickel-hydrogen_system/Rossi/Tests/#Lugano, the design is again different. The consequence of the difference is that data from the reports cannot be cleanly compared. It is not a single product being tested.

As far as I know, there is no "product." These were investigational devices. Rossi allegedly has a megawatt power plant in operation for some time now, reportedly being a reactor assembly, with individual reactors being 10 kW heat generators. The technology there is not HT, according to Rossi.

The three tests were

The reactor tube is charged with a small amount of hydrogen loaded nickel powder plus some additives.

This would be based on representations from Rossi, and is not properly attributed. If the design is as used later for Lugano, the fuel is likely nickel powder, plus a material that releases hydrogen when heated. That is, the reactor does not start with "hydrogen loaded nickel," but just nickel plus additives. At the time of this test, there was less information available on the fuel. There is speculation that the fuel is nickel plus LiAlH4. A single additive. However, the reality may be more complex.

The reaction is primarily initiated by heat from resistor coils inside the reactor tube.

This implies a reaction. This is a conclusion disguised as a description of the experimental conditions. The reactor is apparently heated from resistor coils inside the tube. The word "primarily" is interesting. Are there other initiators? These authors actually do not know what is inside the tube.

Measurement of the produced heat was performed with high-resolution thermal imaging cameras, recording data every second from the hot reactor tube.

As later analysis of the Lugano report has shown, this method is fraught with hazards, starting with a critical dependence of the results on the value of emissivity used. In this report, the device had been painted unevenly with

The measurements of electrical power input were performed with a large bandwidth three-phase power analyzer.

While skeptics have made much about possible errors in measuring input power, short of deliberate fraud (which can be clever enough to escape detection in all but fully independent tests), experts do not generally consider this a likely source of error.

Data were collected in two experimental runs lasting 96 and 116 hours, respectively. An anomalous heat production was indicated in both experiments. The 116-hour experiment also included a calibration of the experimental set-up without the active charge present in the E-Cat HT. In this case, no extra heat was generated beyond the expected heat from the electric input.

Notice: no calibration for the November and December tests. If the reactors were identical, this would not be so much of a problem. However, they were not.

Computed volumetric and gravimetric energy densities were found to be far above those of any known chemical source. Even by the most conservative assumptions as to the errors in the measurements, the result is still one order of magnitude greater than conventional energy sources.

This depends entirely on the accuracy of the calorimetry, which could fail in unexpected ways. If the claim is true that the March device was calibrated, the claim becomes stronger.

There are many problems with the report. For example, in images of the device in operation, in November (which burned out), show striations, the image caption states: "The darker lines in the photograph are actually the shadows of the resistor coils, which yield only a minimal part of the total thermal power."

How do they know what those shadows are, "actually"? The text has:

The thermal image shows a series of stripe-like, darker horizontal lines, which are confirmed by the five temperature dips in the Y Profile. This means that, in the device image, a darker line appears every 6.4 pixels approximately, corresponding to 2.2 cm on the device itself. As mentioned previously, the E-Cat HT needs resistor coils in order to work; these are set horizontally, parallel to and equidistant from the cylinder axis, and extend throughout the whole length of the device. By dividing the circumference of the base of the cylinder by the number of coils, one may infer that the 16 resistor coils in this device were laid out at a distance of 2.17 cm. one from the other. And, by comparing the distance between darker stripes and the distance between coils, one may reach the conclusion that the lower temperatures picked up by the thermal camera nicely match the areas overlying the resistor coils. In other words, the temperature dips visible in the diagram are actually shadows of the resistor coils, projected on the camera lens by a source of energy located further inside the device, and of higher intensity as compared to the energy emitted by the coils themselves.

They do not say how they know that there are 16 coils. The Lugano reactor apparently had three, though that is also unclear. The reactor shows three wires going into one end, so connections to the 16 heater wires would be internal.

They interpret the dark lines as shadows. There are two problems with that. It is just as likely that the reverse is true, that the bright lines correspond to wire positions, not the dark lines. Then, if there is energy being generated in the interior, with cooling being by radiation and convection from the surface, the interior would tend, from internally generated power, to reach a relatively uniform temperature. Then, if wires at that temperature are electrically heated, they will show elevated temperature, not depressed temperature. I intend to consult experts on this; I believe I have made this argument in the past, and it was not contradicted, but sometimes nobody says anything.

Why do the authors even talk about this? It is obvious: they are presenting evidence in one direction, toward the conclusion that the E-Cat HT is generating amazing amounts of power. This is not an independent test, properly skeptical.

So, the crucial issue: calibration. They are not using first-principle calorimetry, where heat is measured directly, they are using a complex method. Such methods can work, but require calibration. The lack of a calibration at operating temperature -- and emissivity of materials can vary greatly with temperature -- was a major flaw in the later Lugano report.

The March test is on a new version of the device, dubbed "HT2." HT2 was covered with black paint, of varying emissivity.

Unique to the March test, there was a dummy test, where an empty fuel tube was inserted and the device run with presumably only electrical heating. However, it was not the same. The experimental run was pulsed power, the dummy run was with continuous power.

The E-Cat HT2 was started approximately at 3:00 p.m. on March 18. The initial power input was about 120 W, gradually stepping up during the following two hours, until a value suitable for triggering the self-sustaining mode was reached. From then onwards, and for the following 114 hours, input power was no longer manually adjusted, and the ON/OFF cycles of the resistor coils followed one another at almost constant time intervals. During the coil ON states, the instantaneous power absorbed by the E-Cat HT2 and the control box together was visible on the PCE-830 LCD display. This value, with some fluctuations in time, remained in any case within a range of 910-930 W. The PCE-830 LCD display showed the length of the ON/OFF intervals: with reference to the entire duration of the test, the resistor coils were on for about 35% of the time, and off for the remaining 65%.

The input power to the calibration was 910-920 W. Because the calibration was done after the test, I at first incorrectly assumed that they set this input power to match the average power input of the experimental run. Apparently not. They apparently adjusted the calibration input power to match the experimental temperature result.

(I'd have greatly preferred to see a plot of input power vs temperature at equilibrium; as well, they confuse the issues by including control box power instead of reporting coil power and only comparing this overall to total input power in a separate discussion, that would confirm the coil input measurements. Control box dissipation may vary with power output.)

(This was not a clean calibration, which would ideally be *identical* to the experimental run except for the presence of fuel -- and an even better calibration would have a dummy fuel with similar heat behavior, but that would not have been possible here. The unnecessary difference was the duty cycle. This is a problem with tests that are not fully independent. They did not have access to controls, and they did not have access to the internal thermocouple, used as part of Rossi's control system.)

Power to the dummy’s resistor coils was stepped up gradually, waiting for the device to reach thermal equilibrium at each step. In the final part of the test, the combined power to the dummy + control box was around 910-920 W. Resistor coil power consumption was measured by placing the instrument in single-phase directly on the coil input cables, and was found to be, on average, about 810 W. From this one derives that the power consumption of the control box was approximately = 110-120 W. At this power, the heat produced from the resistor coils alone determined an average surface temperature (flange and breech excluded) of almost 300°C, very close to the average one found in the same areas of the E-Cat HT2 during the live test.

So the experimental run had total input power of 0.35 * 920 W, or 322 W, compared to the dummy run of 915 W, to maintain the same temperature. On the face of this, then, neglecting the issue of control box power, there was COP, very roughly 2.8.

Other critiques

Both these reviews place great weight on the possibility of fraud as to input power, not merely error. It must be agreed that this test was not adequately independent to rule out fraud. The Lugano test was later run in an attempt to address this by holding the test in a neutral facility, but Rossi remained heavily involved in the Lugano test, as to certain critical aspects, and the test was badly flawed for very different reasons.

In the first test, the power input was constant, while in the second test, the scientists experimented with turning the power on and off to test the self-sustaining mode. In the self-sustaining mode, they observed a periodic heating and cooling cycle that warrants further study.

First of all, three tests were reported (many articles claim two only).

Secondly, there is no clear "self-sustaining mode," that is, a mode where with no input power, the power output remains constant. The authors did not "experiment with turning the power on and off." Rather, this was part of the setup, presumably controlled by Rossi. I see nothing particularly anomalous about the heating and cooling cycle, it looks like normal heating and cooling. Peak temperature is seen shortly after power is turned off, which I believe would be expected due to delays in heat flow. The temperature then drops continuously until input power is restored.

This characteristic is, as far as we know, necessary for Rossi's control system, as has been explained many times. If he reaches true self-sustain temperature, the reaction is likely to run away, so he keeps the fuel temperature below that level. The concept of controlling heat with heat has often been ridiculed, but it is certainly possible.

As to the "dummy run," they say:

They ran the test in March on the E-Cat HT2 for about 6 hours, taking measurements exactly as they did when the cylinder was loaded. They found that no extra heat was generated beyond that expected from the electric input.

This was poorly stated in the report. We were not provided with dummy data "exactly as they did when the cylinder was loaded." The dummy run was different, with constant power, gradually increased to a level that generated the same average surface temperature as generated in the experimental run and measured by the optical camera. They did not measure the "heat expected from the electric input." Rather, they found the level of input power necessary to generate the observed temperature. Obviously, they do not know if there was "no heat generated beyond that expected from electric heat." What they have done is to calibrate presumably with electric heat only.

An arXiv critique

Ericsson and Pomp criticize the report as lacking in important details and data. I agree.

It is unclear why a group of scientists (including chemists, nuclear and theoretical physicists) should be assigned to perform this measurement. The task seems rather to require expertise in IR and electrical measurements.

Again, I've often made this point about the Lugano test. Many naive observers seem to assume that "scientists" are Jacks-of-all-trades. In fact, all through the E-cat history we have seen scientists make major errors that would not be made by experts, as in, for example, the use of a humidity meter to assess steam quality (that error was made in 2011 by Kullander and Essen, and Essen, one of the later authors, has never acknowledged it). As well, Kullander and Essen completely neglected the possibility of overflow water in that early report on the E-cat, and almost certainly, from the system design, there would be some. This was crucial because the heat output was estimated by the mass of water supposedly evaporated, and if the water was lost without evaporation, heat would be drastically overestimated.

Ericsson and Pomp raise some of the other objections that I also independently found, reported above.

Data on electrical input power were provided from the continuous video recordings of the PCE‐830 instrument. However, only an average value for the measurements during the resistive heating period is given. No data from the period with heating off are given.

That data would show the control box power in the "heating power OFF" mode. It is not a far stretch to assume that in that mode, heating power was zero. However, we should not be having to infer.

We would have expected a diagram or table showing the power drawn from the grid over (at least) one full 6 minute cycle, including both the 2 minute phase of resistive heating and the 4 minute phase without any external power to the device.

Well, what they have stated may be enough. Ideally all this data would have been automatically recorded and made available as spreadsheets, for those who want details. What is missing is the power during the OFF phase. I even assumed, above, that it was zero. It would not be zero. It would show the control box power.

It would seem to us that the power consumed by the control box could be obtained from the power measurements during the phase when no external heating was supplied to the reactor device. Instead, the power to the control box was obtained from the dummy measurement. Why?

When people ask "why," I tend to answer. Because the authors were naive and did not carefully look at their own report through critical eyes.

We would have expected to see a discussion and data on power consumption all the way from the start‐up phase of the device to the steady‐state phase, in order to show how and when the “anomalous” heating source kicks in. After all, that source allegedly provides about twice as much power as the heating coils and we would have expected the signature of its onset to be presented and discussed.

Again, this is cogent; however, this kind of data is exactly what Rossi would not want known. I have many times privately written that Rossi has carefully controlled what information is made available. The temperature at which the reaction begins is critical data. If the "independent testers" had knowledge of this, they were quite likely prohibited by Non-Disclosure Agreements from revealing it. These are not independent tests. For that reason, they cannot be considered conclusive.

A “dummy” test was conducted in conjunction with the March test, after the measurements of the loaded reactor. This involved a reactor without powder charge and end caps. The test was performed with a continuous electrical input power of about 810 W.

810 W is the power as measured at the device wires.

No data or figures associated with the “dummy” test are reported. This makes it very hard to assess the validity and usefulness of this measurement.

That is almost true, but not completely true. What is reported is that with that 810 W input, the surface temperature (as estimated by the optical camera) was about the same as during the experimental run, thus establishing a presumption that the total dissipated power in that March run was about 810 W. This would be more believable, more solid, if the full plot of input power vs temperature were given. I have seen this pattern in much cold fusion research. Data is not shown where the author considers it not relevant, even though it is closely related to what is claimed, and would provide a deeper understanding. If one wants to see more thorough reporting, read the SRI reports of McKubre et al, for EPRI, they are voluminous.

The dummy test was not performed in the same way as the test of the “loaded” reactor and can therefore hardly be seen as a conclusive no‐charge test. For example, in a proper dummy test, care should have been taken to supply input power in the same pulsed manner as in the loaded test. (It would then also have been clear to what extent the shape (time dependence) of the surface temperature curves as reported from the March test require any anomalous source, or can be explained as a simple consequence of heat diffusion.)

I find this correct as an abstract suggestion. It seems unlikely as adequate to explain the large increase in apparent power between the dummy and the experimental run. As an impeachment of the test, if the test is to be required to be rigorous, it stands. However, my investigation here was prompted by an experienced scientist who is thoroughly aware of all these issues, and who considered this result to be of high interest. I agree. Hence, from this test, I'd have been looking forward to the promised additional investigation. That actual ensuing report (Lugano), thoroughly flawed, was, then, a disappointment.

Here are the conclusions of Ericcson and Pomp:

The authors of the report published on arXiv [1] claim to have performed an independent test of what is called an E‐Cat HT reactor. We have shown that the tests lack many aspects of what would constitute an independent measurement, that the method chosen for testing seems unmotivated and, most of all, that the paper lacks several vital pieces of technical information that would be expected of such a report and, indeed, necessary to support the reported claims. We also find a few disturbing cases where the authors are jumping to the extra‐ordinary conclusion instead of critically assessing other, more established physics explanations. We note that the proposed claims would require new physics in several areas. Besides a cold‐fusion like process without production of any radiation also extreme new material properties would be needed to explain what rather seems to be a problem of correct measurement. We are surprised that the authors make such remarkable claims based on a report with so many shortcomings. We also find that much attention is drawn to trivialities while important pieces of information and investigation are lacking and seem not to have been conducted and considered. Wishful thinking seems to have replaced scientific rigor in some cases. Therefore, it seems clear to us that a truly independent and scientific investigation of the so called E‐Cat HT device is still pending.  Thus, we do not think that a convincing demonstration of “anomalous heat energy production”9 has been presented in the arXiv report, or anywhere else to date.

Except for analysis based on theory, and their final conclusion (which is unsupported by evidence and argument and is outside the scope of their analysis), I agree. So here is where my understanding differs from theirs:

General comment for students

I have found in this case and in many others that a careful reading of reports can develop an understanding that is often missed by casual commentators, and even by some where "casual" is surprising, where one might expect something more thorough. One will find that errors abound.

Many critics and supporters are only concerned with conclusions, not with attaining a firm grasp of what actually was done. We can see this even with me in my report above. I do not bother to understand the manner in which they analyzed the optical camera data, particularly because in this report, there is an independent measure of power dissipation: a calibration. My reason is simple: it could take me days to do this.

Where this could lead me astray: the calibration depends on temperature data from the optical camera. If the temperature data is captured and analyzed identically, that should not be a problem. But it may not be identical, and we already know one difference: pulsed power vs constant power. That difference may have an unanticipated effect, or there may be some other difference that is not identified. This is why it is desirable to have multiple independent methods of measuring power, in work like this.

I have not yet found a cogent criticism of the conclusion of the first report as to excess heat in the E-Cat HT2, beyond the general claim of "not convincing because of possible fraud."

This article is issued from Wikiversity - version of the Thursday, June 25, 2015. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.