Experimental Evidence on Rossi Devices

The TiO2 reference dots will provide reliable thermographic temperatures, since they have a known band emissivity.

The Profs used TiO2 dots (according to the report) as follows:

(1) they tested dots against thermocouples at low temperatures - and they matched as you'd expect.

(2) they tested dots on the (lower temperature) rods coming from the reactor. (Bottom of page 9). They used the results of this testing to measure band emissivity at lower temperatures. They say they "adjusted" the book emissivity data (which is wrong because it is total emissivity not band emissivity) to match the experimental temperatures from the dots.

The point is that this error mechanism is small at low temperatures. The total emissivity graph is high at 0.8, the band emissivity will also be high at 0.9 or so. Small adjustment will do.

They did not use this method of calibration at higher temperatures. They claim they could not put thermocouples on the reactor body (MFMP had no difficulty doing this). They claim that the TiO2 dots only work at low temperatures - possibly true - they are not designed for higher temp calibration.

What we see from this is in fact the reason why this error does not affect the dummy tests. They independently measured temperature and adjusted the low temp part of the emissivity curve to match - so no error. In any case the error here is quite small, so they could maybe not worry so much about the adjustment.

I had not realised this before, but we can now understand more clearly why the low temperature data matches pretty well. They changed the book emissivity graph for this!

The fact that then they base their results purely on uncalibrated high temperature book data is pretty amazingly bad practice!

This story goes on giving.

We now have a clear mechanism for:

Why the low temperature power estimate matched (they calibrated emissivity against reference dots and adjusted the data, but only at lower temperatures)

Why the "acceleration" occurs - as above. I guess that was why the Profs went on record with this even knowing the temperature measurement was flakey. They felt the acceleration alone showed an anomaly.

Why the COP = 3 results happen - because of the use of total emissivity instead of band emissivity.

And we have clear evidence based on their data correctly processed that the COP for the active tests is 1. The experiment has considerable uncertainty, so we know only that to the accuracy of this experiment the reactor behaves as a normal electric heater.

Alain,

Quote

You don't say it is a fraud, but this is the only possible explanation Double language is like Mary your style (her, it is not to admit she denies LENR).Claiming Rossi is a fraudster is possible, you just need evidence, or at least lack of countrary evidence.There are many countrary evidence for fraud, like tycoon investments, and test where he don't have full control.The test in Lugano may even be negative that the procedure itself show that Rossi was not predicting a negative result. This rule out fraud.

I am not using double language! I just don't feel comfortable dealing with all the human issues like psychology, fraud, etc. All I'm saying in this thread is that every independent test of Rossi's device has shown no evidence of anomaly.

You seem to argue:
(1) If I say this, then I must accuse Rossi of being a fraud.
(2) I do not have evidence to show Rossi is a fraud
(3) Therefore his device probably works.

If you will agree that there is no evidence from credible tests that the device works, and evidence from several independent tests that to the accuracy of those tests it does not work, then I cannot of course prove that Rossi is a fraud. Nor do I want to do that on the basis of these tests.

I will however insist that any evidence from Rossi on possible isotopic changes is viewed as null. You may feel that this is equivalent to accusing him of fraud. He himself however has nullified the previous isotopic tests, which at the time he allowed to stand. He now says that maybe the sample he provided was contaminated. Surely the same can be suspected here. If you call that fraud you should maybe call the previous "contamination" which Rossi now admits fraud? Ni-62 contamination is a bit unusual. But you must remember that Rossi is on record as saying that he uses Ni-62 as fuel. It seems that has changed, or did in this case, but maybe fuel batches got contaminated?

Mary would say this is most likely deliberate, and also most likely fraudulent. That may be true, but none of these conclusions about fraud have the strength of the scientific evidence which clearly shows an electric heater.

Quote

My way is to use ratio, and avoid assumptions and absolute value for emissivity or temperature.what I found is that the ratio of IR cam measurement between 450-800-900W, observing the calibration done at 450W was correct, but the emissivity was unknown, imply a COP>>1 unless emissivity grow hugely from 450 to 900W, which is absurd for alumina.my bet is that you don't take correctly the calibration with TC and known emissivity dots at 450W.You method is too complex for me.

Alain, I think you should be ashamed of yourself if you do not stick with this one and work it through.

In my post above I USED RATIOS. It is quite simple, and shows an error that explains the acceleration.

What you say you have found is just wrong. If you give me the numbers and how you use them I will show you exactly why and how. Please note:

(1) the Lugano Profs ignored their own values for emissivity and corrected it to match the correct temps at low temperatures. That is why the dummy test power measurement matches quite well (10%). I am indeed taking this calibration, but it only shows that the dummy test will not suffer from the systemic error the high temp (without calibration) tests do. Remember, they adjusted the low temperature part of the emissivity curve to match real band emissivty - not the high temp part.

(2) You are not understanding the way that emissivity affects COP, since when you calculate this it explains everything. I'm sorry if my way is "too complicated". How about you show me your way and without being complicated I will explain what you are getting wrong?

(3) If you'd rather critique "my way" you are welcome to do so - it is published on LENR CANR.

@magicsound

Quote

I call your attention to Bob Higgins' excellent analysis of the emissivity errors in the Lugano report:drive.google.com/file/d/0B5Pc25a4cOM2Zl9FWDFWSUpXc0U/viewand to MFMP's measurements of a physical replication of the Lugano reactor, using an identical factory-calibrated Optris camera and a Williamson dual-wavelength pyrometer:youtube.com/embed/0DY4TJmCJS8Based on this work, we concluded that the real COP at Lugano was around 1.2 - 1.4.

I'm glad you mention these two things. You will note that I referenced Bob's analysis. In fact I am indebted to him for a lot of the material I use. Bob suggests a COP of 1.6 or so in this report, but I think he has rowed back from it. He now says he prefers the MFMP experimental work.

In fact Bob's recalculation is wrong - hge uses the wrong method. When I asked him to check this he was not interested in doing so, saying that his estimate should not be taken seriously which is fair enough.

If however a significant number of people are misled by Bob's work (on which I depend quite a lot, but which has this error at the end) I will take the matter up with Bob again. He may now have more time to consider it.

Re the MFMP work. You have referenced a video. Could you tell me where the 1.2-1.4 estimate comes from and how it is calculated? It is not an unreasonable value except the given error tolerance (+/- 8%) is much too low. I agree with Alain that the experimental parameters are quite loose, and put the errors at more like 20%-30%. Any MFMP experimental work must add to these because of uncertainties in the thermal replication.

For the purposes of this thread I think Alain would agree that an estimate of COP=1.3 from Lugano (with no anomalous acceleration and no discrepancy with the dummy measurements) is a null result. There are just too many uncertainties relating to the alumina characteristics at high temperature, and other things. I've summarised these, and Bob has also eloquently pointed out one of the alumina issues (transparency).

I started this thread because Alain said the Ferrara and Lugano tests showed conclusively that Rossi's device works. I disagree with that - there is no evidence from these tests that his device works, and evidence from Lugano that it is an electric heater.

You need to follow Axil to square this circle.

@axil

Quote

Rossi gave the Lugano test team a prototype reactor that he has currently discontinued. This reactor has nothing to do with the 1 megawatt plant. All his activator reactors(mouse) have COPs of about 1.2. Rossi gets high COP by clustering up to 25 fueled drone reactors that are not electrically heated. The real secret to the Rossi technology is reactor clustering. It is useless to harp on a low COP for the Mouse because that is exactly how the Mouse is suppose to work. This just shows how little that the naysayers know what Rossi is doing.

What you seem to be saying is that Rossi gave for testing a device that he knows has only 20% excess power and therefore is within the errors of the flakey technique used to measure it?

So he expected a null result and was surprised when the Profs came up with their erroneous but highly positive results?

It seems far-fetched to me. But since it does not contradict my point on this thread - that no test has ever given credible evidence Rossi's device is other than an electric heater - I'll let it stand!

Quote

I computed the response (energy received by the bolometers) of the IR cam in the 7.5-13um (nearly affine law).From IR cam signal I get a temperature that depend on emissivity according to that affine law.This allow me to estimate the lowest possible temperature if emissivity is not what is assumed (like if emissivity at 900W is same as at 450W.this give me a minimum ratio of absolute temperature change between 450-800-900, which I can interpret as radiation power with power of 4The lower temperature, bounded by maximum emissivity change (I just assume to does not grow which is absurd), gives the number I show you,The "discount" on temperature that we agree on, finally create a discount on radiation and convection/conduction power.The convection par is much less discounted than the T^4, and the high temperature part, imply a bigger discount than the lower temperature part.Thus assuming all is radiation at maximum temperature, which is false, maximise the discount on power, following to the error (underestimation) on emissivity.Despite that discount on temperature, and T4 radiated power changing much more than the electric power, even with lower temperature, there is a COP>1.5.As I say, the 800-900W step is even more clear as emissivity cannot have changed so much, especially if the temperature is so low, and thus the temperature change so low.

Thank you Alain. This is enough for me to show you where you go wrong.

You assume that IR bolometer response is a linear function of temperature T: B ~ aT + b

In fact that is not true. The function is highly nonlinear over the range that matters. (It is linear between 1250C and 1400C but see below why this is not enough).

At 750C (as proved above using an online calculator) IR bolometer response scales as T^1.62 (T in K of course)
At 1350C (proved the same way) IR bolometer response scales as T^2.06 (T in K of course)
Example calculations using internet Planck function band response integrator

The COP acceleration comes from an overestimate of the temperature difference by the Optris using 2.06 instead of the correct 1.62 - obviously temperature differences scale according to the gradient of the bolometer response vs T curve and therefore proportional to the power of T. At higher temperatures we get a larger bolometer response difference for the same fractional temperature difference - so the fact that the Optris believes the temperature to be higher than it really is makes it underestimate the temperature difference. In this case by a factor of roughly 2.06/1.62 = 1.27.

The much lower temperature comes from the overall bolometer response curve. To get 0.4 of the response you must work out what T will give you bolometer response which is 0.4 of the 1350C value. You need to do this accurately using the real response vs temperature curve which incorporates the varying Planck spectral response with temperature. BUT

We get a first order approximation by supposing bolometer response ~ T^1.84 (average of the two limit values above)

Solving:

T ratio = (0.4)^1/1.84 = 0.608

Then

(x+273)/(1350+273) = 0.608 =>
x = 0.608(1623)-273 = 714C

This is not exact, we have made many approximations, but you can see it is in the right ballpark.

Overall the relationship between IR bolometer response and temperature scales between T^1 (at very low temperatures) and T^4 (at temperatures higher than those considered here.

You can get a more precise answer from numerical integration, as I do. But the approximate answer is enough to see what is going on.

Interestingly Bob Higgins also does not take the nonlinearity of the Planck Law into account which is why his value, like yours, is too high.

Quote

Now if you say it is not a fraud, which is proven by the protocol itself, whatever is the result, I don't care this test failed. Of course it work in their labs, or they would have stopped all business, or organized a pony show "à la Defkalion", à la "Orbo", "à la Kesche".

You argue here that the fact of Rossi allowing this test proves the device must work?

Really? I can think of two counterexamples:

(1) Suppose you end up agreeing with me that the test shows no excess power, Nevertheless for many months it appeared to do so, and it has served Rossi's PR well, got him very large amounts of money, so if he himself believed the device not to work, and wanted to deceive others, this test has been very successful. The measurement method comes from Rossi, it is not difficult to imagine also that the error in calculations - quite subtle - comes indirectly from Rossi too.

(2) Suppose Rossi has been doing the same test with the same error and observed high COP. This is very likely. When presented with my paper he calls me a snake and says it is wrong. You think he would be easily corrected by others? I don't, which is why I think it very likely he makes major errors of this sort. It is common for an inventor to go on believing his invention works on the basis of flawed evidence when everyone else can see it does not work.

Take your pick.

Quote

Rossi may be stubborn like some inventors, but neither Woodford, nor Darden, nor his engineers, nor Rossi's engineers.This turn to Apollo conspiracy theory.

If you think Rossi would hire an engineer who disagreed with him about his devices working you are surely joking. Rossi has shown himself good at detecting and avoiding anyone who might have an independent view of his devices - like Krivit. He calls such people snakes and they can come no-where near him.

I'd guess (if he has any sense) all engineers will be hired under NDA. They may suspect the device does not work but as with our arguments over the Lugano results it is not simple, and they will be in trouble if they break NDA, So even if one of these engineers proves independent-minded enough, and has access to enough data, to suspect the devices don't work we would not hear.

In fact it is easy for Rossi to keep data enough to himself so that no-one else would have obvious evidence.

As for Woodward and Darden. You think they sent in Krivit-like engineers? No way. They would do DD based on the independent reports of 6 mostly Swedish scientists, and background DD by talking to LENR theory types Rossi would suggest.

How could engineers coming in for a short time conduct a better test than the Swedes? And yet those tests were both highly flawed. Rossi has always avoided proper independent testing - e.g. from NASA. When it has happened anyway (Hydrofusion) he just says the thing they tested was broken.