Experimental Evidence on Rossi Devices

Yemsral wrote:

Quote

AlainCo wrote:
To Thomas about my computation, I follow your reasoning and mostly agree.
I used an approximation like Raley jeans, and I don't consider I know the temperature at al.
I just know the reading of the IR Cam, which is related to temperature via emissivity.
This is old computation, and given the current news, that ferrara and isotopic shift are good, I have more to do than dig old stories.

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.

Great, very valuable information.

Display More

Yemsral - you have perhaps not read the interchange between me and Alain after he posted that? It is not valuable information at all, it is 100% wrong. I'll summarise why below, it should be clearer than the multiple posts previously that said the same thing.

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.

The data we have is IR camera temperature reading skewed by the wrong emissivity entered. For the cal run, at a lower temperature, the temperature was measured by thermocouple and the (incorrect) method used led to the incorrect temperature (not that measured by TC). Rather than work out why their method was wrong the testers (they say this, bottom page (4? EDIT 9)) used the TC data to calibrate the book emissivity at this low temperature. Unfortunately the higher temperatures are on a completely different part of the "book" emissivity graph and they continue to use their wrong method here where because the book values are much lower the error is much larger. Since we don't know how much they fudged the book emissivity by to make their wrong method work, this can't be compared with the other two data points and is basically worthless.

So to address Alain's points:

• You can't assume comparison with the dummy test data because the testers adjusted their emissivity graph to make this fit. The discrepancy between active and dummy runs - corrected, is about 30% and this much comes from the adjustment the testers made to their book emissivity graph.
• You can't assume the COP difference between 800W->900W is too large to be accounted for by the data because of the math (given in detail above) which shows that the error introduced by the testers varies nonlinearly with power.

Alain implies I'm making assumptions. In fact he is the one doing this. There is some uncertainty in this method even doing it properly, but the results are exactly what would be expected if COP = 1 at all powers. No "Rossi effect".

Quote

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.

Betting is good, but like making assumptions it does not always work.

In this case my method corresponds to reality - which is complex. That is why the testers got it wrong. They were not (IR thermography) experts and for an amateur looking at this stuff it is easy to make mistakes. Although in my paper I worked out the answer numerically, when challenged on the other thread I worked out a much simpler way to show the same thing, algebraically, which proves that the COP "acceleration" is an artifact.

Best wishes, Tom

Shane - my arguments have not collapsed on scrutiny - a number of people initially disagreeing, and competent, have ended up understanding them and agreeing. Have a look. You would be able to join the enlightened crowd except I think you have said you do not want to do the required work: so your criticism here is hypocritical. Leave it to those who are prepared to get their hands dirty?

You are not looking in the right place. I missed it too on casual reading. Bottom of page 9:

Quote

We therefore took the same emissivity trend found in the literature as reference; but, by applying emissivity reference dots along the rods, we were able to adapt that curve to this specific type of alumina, by directly measuring local emissivity in places close to the reference dots (Figure 7).

I can't see that "adjust" is more dishonest than "adapt" but perhaps you would like to disagree? And you are too harsh, neither is dishonest - but it is poor experimental practice to do this adjustment for the dummy results but not for the active results (their excuse being that they could not affix suitable TCs to the reactor body securely - MFMP had no trouble doing this).

This also shows how closely you need to read experimental write-ups to work out what they really mean. Headlines and sound bites do not cut it. It will come to no surprise to anyone who has done a Doctoral LS of the research literature (one of the real things that doing a PhD properly teaches - but only learnt if it is you, and not your research supervisor - who makes sense of it).

Quote

Finally you admit your Lugano critique "is complex". Thank you for your honesty!

The report is complex. The paper I wrote is medium complexity - and the further amplification I was pushed to do on the thread here makes the key arguments simpler still. I should really rewrite the report to make the "approximate analytical" arguments clearer so people don't have to look at the precise numerical stuff.

Eric, I'm dissappointed. You are showing your credentials as a true follower of Kuhn, where scientific progress cannot exist because all is socially relative.

Quote

We cannot be sure that you're not exactly as wrong.

Yes we can. There are a number of uncertainties in the experiment which I explicitly state in my comment - I devote a whole section to them. I don't think there is anything I missed. See section VI (Caveats) and specifically subsection VI-C (Translucency errors).

Nevertheless I correctly identify an additional large issue which is an error in calculation not an uncertainty. When this is corrected, without fudging (it is not possible) the remarkable acceleration in COP goes away very precisely. That is added confirmation the correction is good, and also gives some assurance that even if the uncertainties exist, they are not so large as to invalidate my analysis.

Quote

As I've said before, all you've shown in your Lugano analysis is that such a byzantine approach cannot be trusted. Which was immediately obvious.

You rush to judgement. It is immediately obvious that there were severe methodological errors in the experiment, just as there were (see Pomp et al) in the previous one. I note that you seem less convinced of those.

It is not immediately obvious what is the actual import of the detailed data gathered - and it provides unique insight into the characteristics of one of Rossi's devices.

Quote

Your analysis is exactly no more reliable than theirs.

That statement is false and in the light of my argument above you might like to retract, or at least modify, it. Error bounds always exist, as do uncertainties. However adding a very large calculation error on top of this must reduce reliability, and I show in detail (via the "acceleration" which the testers highlight as significant) why this is.

Quote

A different approach was needed, including solid cross-checks. You mentioned this.

I agree, the world would have been better if...

Quote

That would have been sufficient.

That would, if the cross-checks were properly done and not fudged as in the "adapted" emissivity graph, have made these results more accurate, I agree. I reckon my bounds of +/- 30% would go down to say +/- 20%.

Why so little? My work related to an error in the temperature. That depends on emissivity at IR wavelengths where alumina is certainly opaque and there is much less uncertainty. TCs would nail the temperature down tighter, but it is not bad anyway. The larger uncertainty comes from total radiant power which includes higher frequency elements that are variable due to alumina translucence.

Quote

Even Joshua Cude/Popeye, whom you defer to, has said about your analysis (paraphrased): "I find such complex arguments often collapse upon themselves". To this day, he has yet to endorse your Lugano review by the way.

Shane - you are sort of making debating points here. Joshua quite properly reckons all these LENR experiments are known flawed because of the strong meta-evidence. I mostly agree with him, but I still do the LENR guys the courtesy of looking in detail at their stuff from time to time. I like decoding artifacts, and these experiments provide a fruitful source. Joshua has just said that he will not bother to look at Lugano, which is fair enough especially given Rossi's track record of bad demos.

I have never deferred to Joshua over the real experimental analysis - only over his mastery of the historical meta-evidence, which however I do not rate as highly as he does.

Quote

I concede that you may have fixed some of the second-order details -- this is not something that I've sought to verify.

This is numerically incorrect. The detail I fixed changed COP by factor of 3 from 3 to 1 and COP acceleration from 10% or so to 0.5%.

You claim some other uncertainty has a comparable effect to that? Substantiate it please! Or, if you simply claim that you cannot verify it your self and don't trust my (open, detailed, long published) results, say that.

Quote

Zeroth order methodological problems make such mathematical contortions moot.

As always details trump assumptions or theoretical statements such as this. What methodological consideration makes a possible error larger than, or even comparable with, the ones I corrected?

Quote

I like this advice. Perhaps you will apply it to yourself with regard to LENR experiments, which you cannot be bothered to read?

Retract that comment, or be specific quoting evidence. I have read a number of LENR experiments recently. Every time we discuss the matter somone brings up a new "killer" experiment, so it is difficult to keep track. None so far are what is claimed, or anything like. If people could agree on just one "as good as any" experiment I could read it in detail and report. But remember, a proper report takes 5X longer than reading and I do not have so much time at the moment.

Currently I'm awaiting Abd's claimed neutron data - I need a definitive reference to look at. Unless y'all think something else would be better...

Quote

Your reaction 58Ni(4p,?)62Ni in the "Isotopic Shifts" section does not even balance charge or spin.

As for the nuclear reaction, you will note that the calculation is the reaction change in total mass. I gave the nuclear transformation but use isotopic atomic weights for the numbers. These include with each nucleus the associated electrons (which indeed do not balance) and therefore the result is correct. To do it otherwise you would need to use nuclear masses rather than atomic weights and add in the electrons. Burdensome. (BTW if I had got the electrons wrong, by using nuclear masses, then the real answer would have been for even more excess energy than I show).

Take home, you need to understand papers, not just react to superficial reading:

Quote

But anyone reading your paper who knows nothing about nuclear reactions might conclude that you knew what you were talking about on the basis of the tone and scientific sound of the paper. That was a small problem in that section.

Well I did in this case know what I was talking about, so perhaps the (I agree generally unwise) conclusion would be correct.

Quote

A huge problem was the assumption about all of the nickel being converted, which would lead to 20x more energy than reported. Why on earth would we conclude that all of the nickel was converted via this reaction?

You would not. That is surely my point?

The Lugano authors advance isotopic change and excess heat as evidence for nuclear reactions. There is no measured excess heat to within experimental error. The isotopic results are striking, as ai state, but if whole they are impossible. If partial then they do not indicate nuclear reactions. That must be an assumption, because they could be due to fractionation. I say (which I think makes this argument fairly clear):

From the paper:

Quote

While speculation as to the meaning of this isotopic shift measurement is beyond the scope of this report there is one consequence that can properly be considered. We assume in what follows that the 99% measurement is considered accurate for the bulk sample. This is the simplest hypothesis, and is implicitly followed in the Report.

Note that assumption for the analysis is hypothetical, based on one specific case. This is however the case argued in the report, and is the simplest (though I agree erroneous) conclusion from the measurements.

Personally, I don't for a moment think this is the point: Rossi contaminating the sample is much more likely than any process, fractionation or nuclear transformation, that would lead to such complete conversion. Indeed contamination would be more likely even had the results been less complete (and therefore less obviously what could be obtained from bought 62Ni).

Quote

Even if we went with the assumption that all of the nickel was converted, which is shaky at best, there are lower-energy reactions than the 0.0379u (= 35.3 MeV) you come up with that would have led to a much lower energy balance. (This amount is high enough that it is almost a giveaway that the calculation is wrong).

Really? We have - take home - Ni58 converted to Ni62. Most of the LENR mechanisms to date that don't use D suppose proton or neutron capture. The bottom line from this is as I've stated whatever the intermediates. You could imagine some complex reaction in which Li is involved, but there is just not enough Li for total conversion given any simple reaction (and the energy excess while lower is still too high). The case where there is partial conversion I do not apply this energy calculation to, but instead note that fractionation is a possible non-nuclear mechanism for this result. If you disagree with this you need to argue the case for some other reaction, with figures which I will check for you. Don't be afraid!

Quote

I absolutely do not trust your results; there's no doubt about that.

I agree, but maybe you should examine your reasons for this lack of trust more closely. Maybe you do react to superficial and incorrect readings of papers? Or maybe you had some initial bias.

It rather makes my point that papers should be considered on their merits without factoring in trust or otherwise of the author except where it is engendered from the quality of their written output. It is too easy to be misled positively or negatively.