Celani's cell did NOT vary with pressure!
Dear world,
We have an admission and an apology to make, on the upside, some very important developments have occurred as a result.
When we first started seeing a correlation of sorts between Pxs and Pressure in the US cell, I asked for this to be investigated more deeply and Malachi, Ryan and the team did fantastic further work and analysis showing a potentially challenging finding. We are going to investigate this further and have made extra calibrations in US and EU cells to do exactly that moving forward.
What we did not say was that Celani had already been challenged at NI week and ICCF17 and subsequently on this exact potential issue and had carried out specific experiments to test for a measurable effect and reported back to his critics his findings which did not, in his cell, show the kind of relationship have seen. We had received this email a little after the 7th October 2012, but in the pressures of everything had simply forgotten about it, in hindsight, it was probably the reason we pushed for the investigation. Having said that, we were not given permission to share emails between MFMP and Celani openly, had we been able to then the community watching our journey would have surely reminded us.
Some have suggested that Celani had not considered this, but as you can see below he had and we could have made this clear without being specific. For this oversight, we apologise.
Here is a version of Celani's email, copied to us around the 7th October, in which I have adjusted the English for readability on Celani's request – the original is available below as a PDF so people can confirm the meaning is the same.
Subject: CMNS: Further tests on Pressure reduction and temperature variation in Celani reactor. Dear Colleagues, Some people, inside and outside of our community have made some STRONG CRITICISMS of the report I presented at ICCF17 and the 2 DEMOS held at Austin (National Instruments Week, 6-10 August 2012) and Daejeon (ICCF17, 13-17 August, 2012). * The key point of the critics is the possibility that the slow pressure reduction of Hydrogen and temperature fluctuations of the experimental room could greatly affect the high temperature (about 150°C) measured OUTSIDE the wall of our glass reactor to invalidate the claimed excess power. They think that the large, as expected, temperature increasing INSIDE the reactor (due to changing of convective motions) can have some notable effects on the OUTSIDE temperature of the reactor, supposed by me to behave like a Black-Body. In other words, they think that the thick-walled (3mm) Borosilicate Glass tube (Schott-Duran) didn't behave as a Black-Body in the temperature ranges of our interest (100-170°C outside, 160-240°C inside). * In particular: a) They had big reservations about the Fig. shown on pages 33 and 38 of my report (enclosed) at ICCF17. b) They dismissed as meaningless the positive results shown on pages 29 and 31. c) Moreover, they considered it just pure coincidence the effect of the apparent excess power increasing and the R/Ro ratio of submicrometric Constantan decreasing by a large amount (up to 30%!!!!). Other examples were shown on page 38. * In order to adress such possible doubts, in the last week [ first few days of October 2012 ] I decided (also under the "strong" suggestion/ recommendation/ control of one of our external Collaborators) to deload, as much as possible, the special Constantan wire from Hydrogen. It was exactly the same wire that had been working, continuously, from July 10, 2012 and, from that date, the reactor had NEVER been opened. The deloading procedure was quite difficult and took several days. By considering the R/Ro measurements it is thought that there was still some residual (i.e. at very low concentration) Hydrogen present in the Constantan lattice. * The measurements performed were as follows: a) The Reactor was kept without external thermal insulation, in a condition that was as similar as possible to the DEMOs held in USA and Korea (and previously at INFN-LNF, Italy). b) The power applied was about 48W, as typical in such demos. c) The power was applied to the "Active" special Constantan wire. d) The only difference was the gas: He (supposed inert) instead of Hydrogen (supposed active). We know that the thermal conductivities of such gases are quite similar. e) We simulated, as close as possible, the previous real experiments under H2, the behaviour of the "spontaneous" gas pressure decreasing (due to some gas leakages of our reactor with internal volume of, only, about 250cc). f) We kept the pressure constant at 6.6 bar: studies of room temperature spontaneous changing, 1 full day test, i.e. day-night cycle. Exp.#2 g) We studied the effect of slow pressure changing at the low region of pressures: 4.4-->2.8bar. Again overnight test, from 19:30 (Saturday) to 11AM (Sunday). Exp.#3 Results e1) Exp.#1
f1) Exp.#2
g1) Exp.#3
Comments and Conclusions Because the variations of external temperatures in the previous experiments with Hydrogen were of the order of 10°C, the effect due to the large pressure variations, as indipendently measured by Helium gas, are almost not significant when compared to the results obtained by our reactor and specific sub-micrometric Constantan starting from June 06, 2012.
Francesco CELANI |
When we published the pressure drop findings of our Celani inspired (but quite different) QUARTZ based cell. Even though we drew attention many times to the key differences in our cell, this still caused problems, quite understandably, for Celani at a time when he was not in the best position to respond to our findings.
There was several subsequent events, first, Celani copied an email to us that he had sent to the CMNS community rightly defending his position in relation to our set-up and findings.
From: Francesco Celani Dear Colleagues, In the very recent days, I got information that AGAIN there was some discussions going on about possible mis-interpretation of our previous experimental results due to the effect of pressure reduction inside the reactor. They made an experiment with the mixture Ar@75%-H2@25%, as I used in the previous experiments. They studied the effect of pressure reduction and found that such pressure variation affected in a large way even the output temperatures. I would like to note the following:
* Considering such experimental facts I am convinced that the effect dectected from our Colleagues is obtained in conditions very different from ours. Thanks for the kind attention and, again, I would like to tank the Colleagues that made in deep attention to our experimental set-up and data analysis. BTW. I am resending also the old mail of October were such results were discussed [as above] My best, Francesco CELANI |
It was at this point that we realised we had forgotten the previous email, never the less we did another run in pure He.
The second thing that happened was Celani said we could start publishing some of our correspondance with him, such as the previous one Ryan posted and these two. This is fantastic news as it feeds right into our Open Science model and will prevent these kind of disclosure difficulties in the future.
The third thing was the release of the other independent researchers findings, that of ST Microelectronics. Whilst this may well have been unrelated, the research was originally due for release later. In any case we welcome this great step forward in verifying the New Fire being made open as soon as possible.
We intend to load the EU cell with special Constantan wire of the same type as the US cell used and be loading it with H2 by wednesday and go for a run shortly afterwards. Let's see if the live data can give a present to Celani for the Rome conference.
Ryan adds,
We are pleased to have these comparisons pointed out. it is reassuring that his cell did NOT demonstrate this effect in Hydrogen, either. We did not pay this much heed at the time because, well, it seemed to say that there was no reason to care - the gas and pressure didn't affect it much.The next step is to try our cells with the borosilicate glass which is happening on both sides of the pond, EU first then US. The test cell in Eurpope should be available to give us some data soon.
Another important difference is that his cell has Macor ceramic disks at the end of his mica wire supports that serve to block the radiant heat, convection, and some conduction of heat to the cooler metal flanges at the end of the tube. Those may have made a big difference in susceptibility to the effect we observed. That is a function I did not fully appreciate till recently. Now we will work to add that function in the new test cells.
The US quartz based cell does change temperatures with all 3 gas combinations as the pressure changes. The purple line is the most recent test with Helium. It is odd shaped. That data was pretty noisy. Malachi took the 15 minute averages for these data points and it still gave quite an interesting shape. The green line is Hydrogen. The blue line is the H/Ar blend.
Comments
quantumheat.org/.../...
You have here the current looking of the cell. I am varying the size of the sheet to adjust the temperature cell and get 175°C for Tmica.
I had 165°C with 56V (max) and 0.9A. So I did try to find the sweet spot to reach maximum loading temperature.
In this calorimeter design, we eliminate the need to know the exact mass of air flow or the exact specific heat by measuring the number of watts it takes to heat the air 1 degree. Then we measure the temperature rise in the air after flowing through the chamber. Then it is simple math.
In a perfect world the math is linear. In the real world it will have a curve due to any number of losses and imperfections. They are compensated by calibrating the unit with a test heater stepped through many different wattages. The resulting calibration curve compensates for most errors.
The unit is insulated, but for high resolutions it will still need to be in a pretty constant temperature environment.
It is also noticed that the T_GlassOut is slowly rising as the pressure is slowly dropping during this test.
It would be interesting to correlate the time constant of the cell with the pressure. Perhaps the extra gas adds thermal mass to the system.
I was wondering if you'd planned for some insulation on the exterior to shield it from air currents. Otherwise, you may have the same problems you had before of instability. And are you measuring the air flow? If so, you can have an independent "first principles" method of measuring the enthalpy from knowing the mass flow of air, the specific heat, and the temperature rise.
This is the difficult way to go. A liquid or even a Seebeck "perimeter" or "envelope" style of calorimeter would have been way easier in my opinion.
No, because the glass will have same IR transparency during calibration period also. It's the excess that matters, not the absolute value of radiated power. Celani's setup is not suitable for accurately measuring the energy the glass radiates, his setup only shows that it radiates more than expected, hence the anomaly.
It's pointless to calculate power and all, if all variables are kept constant, the rising temperature will be proof enough.
If its not possible to keep the variables absolutely constant, then the temperature rise must be shown to be much more than that due to variables alone.
From my use of the black/gray body equation, any time you assume emissivity is less than 1, you decrease the watts you calculate based on the heat you detect. The equation for a gray body (anything with emissivity less than 1) is P/A = es(Tglass^4 - Tambient^4) , where P is power in Watts, A is surface area, s is the Boltzmann constant and e is emissivity.
If not all of the IR is being absorbed by the glass, then we'll have a lower Tglass, which'll lower the calculated power. There's no way losing IR detection could increase the calculation.
Yes. Less radiation absorbed will result in lower temperature of the outer surface and hence will show less excess energy.
As Dr.Celani is assuming that at those wavelenghts the glass tube is behaving like an ideal black body, shouldn't this have some implications on the expected results?
I did a wrong manipulation on the 12/8. I shut down the power supply believing the scrip crashed.
Then I realized I was reading wrong values. So I quickly changed the script to start it at the time I did this wrong move.
That is why you have this variation for a couple of minutes.
Sorry for that.
Sorry for the inconvenience.
RSS feed for comments to this post