Many people are frustrated when critics voice their opinions regarding eCat tests and that frustration is evident in the case of the May 2013 HotCat paper. However, when answering why the paper was published on Arxiv instead of a peer-reviewed journal, Andrea Rossi stated that he saw Arxiv as a first step where the paper would have its initial airing (and so peer reviewed of sorts). Preparing a proper scientific paper will take time and in time, such a step may follow. In other words we – along with Rossi – should welcome peer review. With that in mind, I am grateful to Peter Ekstrom of Lund University for his take on the report. A long-time eCat critic, it is no surprise that he is not convinced. The following thoughts and any errors are my own but it seems to me that Ekstrom’s justified observations are further coloured by his preconceived conclusions and embellished to amplify the result he expects.
He starts by pointing out that the test was not truly independent:
This report is advertised as an indipendent (sic) test of the functionality of Andrea Rossi’s E-Cat. Rossi has arranged a series of demonstrations of different versions of E-Cat since January 2011. The common feature is, however, that the demonstrations are completely under Rossi’s control, and that “testers” are very limited in what they are actually allowed to check.
While previous tests have been completely under Rossi’s control, this was not. True, by conducting it in his premises and setting the boundary conditions, he is limiting the tester’s options. However, this should not necessarily void a black box procedure.
It is very clear from the report that the authors had very limited control over the instruments and procedures used for the measurements. One even gets the impression that part of the report was written beforehand in Italian (by Rossi and Levi?):
”The authors would like to thank …. Prof. Alessandro Passi for his patient work in translating the text.”
According to Levi, they had full autonomy over the test and the equipment they used as long as they kept away from trying to examine the core itself and the control box. The rest appears to be nit-picking.
In my opinion, the main purpose of the report is to prove that the output power is greater than the input power. This purpose is not achieved.
Fair enough. Let’s see why:
The input power is measured with a PCE-830, which is probably a very sophisticated instrument, but the authors do not provide an account of how the instrument was checked so that they knew it gave correct values for the pulsed power input to the heating resistors. (In a test in September 2012 the input power was shown to be incorrect by a factor of 2-3 . A simple circuit diagram would have made it clearer how the watt meter was connected. Moreover, the testers were not permitted to inspect the control box regulating the input power.
Although unable to look inside the box, they were allowed to pick it up (it was light) and check for hidden wires or other signs of trickery. Indeed, we are told by Torbjörn Hartman that he did exactly this. I do think it is reasonable to remind us that a potential investment involving Hydrofusion was pulled because their testers claimed erroneous input measurement. The story is here and the following is Google’s translation of the announcement:
Investor Group had instructed the SP Technical Research Institute of Sweden, to monitor the measurement, and the researchers who attended measuring an input electrical power that was two to three times higher than Rossi himself measured.
Despite this, Rossi presented a measurement report September 9 based on earlier tests where a lower input power and an energy surplus reported.
I have no reason to doubt the current authors’ ability to check the input power and hope that the above incident made them mindful of the need for extra vigilance in that area. In my opinion, it is something worth keeping in mind without jumping to negative conclusions.
Continuing with the Ekstrom critique:
The output power is calculated from the temperature measured with an IR camera using Stefan- Boltzmann’s radiation law. This creates an uncertainty since the emissivity is not known, and the temperature varies considerably in different parts of the cylinder.
The emissivity for stainless steel could have any value from 0.8 to 0.075 . The lower value would obviously yield a much lower net power, in fact it could easily make COP=1. Furthermore, the paint used has an unknown emissivity
The emissivity value was used by the IR camera software to calculate the temperature and then again (in reverse) by the SB formula. While exact cancellation cannot be guaranteed, the difference is likely to be minimal. Indeed, in the March test, by using patches of known emissivity supplied by the IR camera manufacturer, this was checked and the difference was indeed small. Furthermore, spot checks were made with a thermocouple which compared well to IR temperature measurements within 2 Deg C.
Since a correct measurement of the output power is crucial to the functioning of E-Cat, a more unambiguous method should have been employed. The E-Cat could have been cooled with water (directly or indirectly) and the flow rate and input/output temperatures measured.
Perhaps, but I suspect we would simply have a whole different set of arguments to ponder. Few experiments are perfect but there is no reason this method should not work to an accuracy capable of measuring such dramatic power differentials.
In order to prove that the E-Cat works as claimed, it is very useful to perform a null test. This is done during the March test. Obviously, the dummy and the real device should be as similar as possible, and only the fuel (nickel, hydrogen, secret component) should be omitted.
“The electrical power to the dummy was handled by the same control box, but without the ON/OFFcycle of the resistor coils. Thus, the power applied to the dummy was continuous.”
The testers thus decided, for a reason known only to them, to change the power supply to the resistor coils. In my opinion, this completely invalidates the test. The dummy run should have been carried out with everything the same as in the test run, except the fuel. Presumably, it would then have run at a significantly lower temperature.
I believe that the testers made no such decision. They ran the null test in the same manner but, since there was no excess heat, the thermostat did not trigger the on/off cycle seen in the charged version of the run.
“The scope of the present work is to make an independent test of the E-Cat HT reactor under controlled conditions and with high precision instrumentation.”
It is disturbing that a local author is the main author in a so-called independent test. Furthermore, there is very little description concerning the control of the measurement procedures by the international authors. They also seem to have had very little say in the experimental procedures.
That a local author appears to have been the lead on the tests is not ideal and it is fair for a scientist already concerned about the validity of the results to question it. This should only be relevant in relation to perception, unintended bias (or over-trust) or direct complicity in fraud. The damage to perception will vary. I expect some business people will take it as a cautionary measure while those already sceptical will be unable to look beyond it. For my part, I am reassured somewhat by Hartman’s checks and by the involvement of the other authors. As for the charge of complicity in fraud (made elsewhere and not by Ekstrom) – that does not sit well. Giuseppe Levi is a respected member of Bologna University. I doubt he would involve himself in a deception bound to fall sometime in the near future.
“Later, an experiment  was carried out by S. Focardi and A. Rossi using an apparatus with a sealed container holding nickel powder plus unknown additives pressurized with hydrogen gas. When the container was heated, substantial heat was produced in excess of the input heat.”
This self-reference does not unambiguously show that excess heat was produced. The report is of poor quality, and the physical interpretation of the nuclear process is seriously flawed.
I’m not sure of the point being made here. There is no self-reference. The report was not authored by Rossi or Focardi and the context was purely historical. Further, as Ekstrom himself admits, the point of the paper was to determine if excess heat was evident, not to determine the nature of any assumed nuclear process.
”It was not possible to evaluate the weight of the internal steel cylinder or of the caps because the ECat-HT was already running when the test began.”
The weight of the cylinder is not that important, but it is crucial that the test team was present at startup to check the functioning of the device. In addition, it is stated that the non-local members of the test team were not present at the E-Cat HT test in December. Thus the December test is based on hearsay, and of very little value as an independent test.
The determination of excess heat can be made without being present at startup. In this context, the criticism is somewhat harsh. I do agree that the absence of non local members in December reduces the value of any claimed independence but once more, I think dismissing it as hearsay (the lead author was present) and of little value takes things too far.
“The E-Cat HT2′s power supply departs from that of the device used in December in that it is no longer three-phase, but single-phase: the TRIAC power supply has been replaced by a control circuit having three-phase power input and single-phase output, mounted within a box, the contents of which were not available for inspection, inasmuch as they are part of the industrial trade secret.”
The fact that the reactor itself was unavailable for inspection is acceptable, since it should be possible to measure the net power provided by a “black box” (especially using blackbody radiation), but the fact that the control circuit feeding electric power into the E-Cat was not open to inspection is very disturbing.
I was initially concerned by this but Hartman’s subsequent description of his own inspection reduces that concern. It is unlikely that anything significant could be squirrelled away in order to sneak extra power beyond that measured before the box.
“The authors of the paper noted that they weren’t in control of all of the aspects of the process, but they concluded that there was an energy production one order of magnitude higher than a conventional source.”
This statement is self-contradictory. If the experimenters were not in control of all aspects of the process, they can make no statement on the functionality of the device.
This is another overreach. In their opinion, the elements they had control of were enough to justify their conclusions. They take great pains to point out that, where uncertainty existed, they picked the most conservative assumptions. The apparent excess energy was so great that even if you assumed the whole volume of the reactor core was involved in the process and considering the fact that the tests were actively stopped while heat production was still in full flow, their conclusions do not contradict the statement of control.
“The weight that may be assigned to the powder charges? is therefore on the order of 0.3 g.”
A weight of the powder of 0.3 g would yield a power density of 533/0.0003 = 1.8*106 W/kg = 1.8 MW/kg. This is an enormous power density, and the nickel powder would presumably melt and be rendered useless.
They make no assumption about the mechanism driving this process. It is a physical test with measurable outcomes. If and only if the results are verified, theory will catch up and we can match fact to it. If they are not verified then the point is moot anyway.
“The most important element of the E-Cat HT was lodged inside the structure. It consisted of an AISI 310 steel cylinder, 3 mm thick and 33 mm in diameter, housing the powder charges. Two AISI 316 steel cone-shaped caps were hot-hammered in the cylinder, sealing it hermetically.”
It would have been interesting to have documented how the endcaps of the cylinder were hot- hammered in the cylinder with 0.3 g nickel powder floating around in a pressurised H2 atmosphere. But this is obviously a trade secret.
We know little about the core but I believe the Hydrogen is in solid form and not a gas.
Only Levi and Foschi were present at the December test, but all authors apparently claim that they have complete control of the test. How was that achieved? Since this is not documented, the tests in November and December should not have been included in the report with the Swedish authors.
The observation is well made but we know by subsequent comment that this is a preliminary paper created to illicit fast feedback such as yours. With that in mind and to allow us to view the tests in context, it seems reasonable to combine them in one place.
It is strange that the COP (coefficient of performance) of E-Cat decreases from the December test (COP=5.6) to the March test (COP=2.9). One would expect the device to become more efficient with time. Is this because the measurement procedure in the March test was more accurate? Maybe one could then expect COP=1 for an even more sophisticated measurement?
The authors state that the March device was of a different design. This seems reasonable in an ongoing R&D situation. If the November HotCat melted, perhaps control was more important than output at that stage of experimentation. It is purely my guess, but it seems that the inventor did not design something for the authors but gave them what was on the bench at the time.
Plot 9 shows COP and the ON/OFF status of the resistor coils. Is it a coincidence that zero feeding for two thirds of the time results in COP=3, but constant feeding would yield COP=1?
I noticed that too and it does make you pause to wonder. That said, if the input was measured correctly and proper measures taken to ensure no extra power was sneaked into the cylinder, then this is of no consequence.
If the results of the measurements are correct, it is obvious that some new physics is at play. The complete lack of any kind of interpretation of the results in terms of physical theory is thus surprising, to say the least. It is even more surprising that the authors have made no attempt to learn what is going on by using efficient, high-resolution detectors (for instance germanium detectors). According to the report, detectors were used, but the type was not specified, and no spectra were reported.
As Ekstrom himself stated, this was about measuring excess heat. I have no idea why he thinks they should have made it about something else, particularly as we know there were limitations put upon them by Rossi. In a black box test, these limitations are reasonable when we consider the commercial implications if this turns out to be true.
I would forgive anyone reading the above for thinking me an eCat believer. I am not. I am simply trying to strip away what I see as a natural (and often unintended) bias in so many of the critiques out there. This can diminish genuine concerns. I hope the authors continue to talk freely within the limits of any NDA they signed and correct anything we have got wrong. In my case, I would welcome such corrections.
To dismiss the HotCat report as of no consequence, I think you have to start from the assumption that the results are false and work backwards. I believe that there are unanswered questions and we have not reached the level of proof required to conclude that the world is about to change – but this report has made me sit up and take notice. Unless you are trying to make an investment call, there is no need to conclude one way or another. I fully understand why many people have jumped with both feet into the negative camp but suspicion and prior conclusions should not blind us to evidence. There is so much BS surrounding the eCat but to truly falsify this report without assuming fraud among its authors, is pretty hard to do.
We have been calling for an independent report for some time. While this falls short of the gold standard, you have to be determined to dismiss it completely. That on it’s own is a pretty impressive achievement.