Discussion and findings on success since Saturday 10/11/12
We had what appeared to be a very successful loading, with strong and compelling evidence provided by the change to a lower resistivity of the wire over the course of the loading cycle. There is the possibility that we may be able to load even more hydrogen into the wire in further experiments over a long period of time.
Additionally it has been found that during the course of the active experiment there was also a period of apparent loading as can be seen in this graph.
As you can see - initially the resistance increased, but at a certain point, there was some resistivity reduction even as more temperature was recorded in the T-Mica. This is extremely interesting. It may point to the loading/burning cycles suggested for commercial LENR configurations and why they need control and multiple reactors in a viable solution. Celani says that the bulk of the loading takes place in 10 - 20 minutes, this is what we found in our loading over the 10/11th November 2012 weekend. This would therefore require a reliable commercial design to have 3 active cells - 1 for cooling, one for loading and a third for “burning” and this may explain the configuration of 3 reactors in Rossi’s 1 MW reactor sub units or 9 in Defkalions configuration. When burning - a reactor working as part of a team could run at much higher temperatures producing dry steam even.
A single reactor would require times when it was at around 125-160 for loading and therefore would not be able to consistently output dry steam and therefore would be limited to producing low grade heat and may account for the specification of the home E-Cat and its need for kWs of heating 1 hour in six.
We also learned a lot about using the calibration data to determine the most useful calibration curves for excess power calculation.
Current test status
Our active run was done in a series of about 7 power steps to bring it up close to what we did during the calibration runs and allows us to compare how different wires are acting under different gas mixtures.
Based on our current assumptions, which we may need to review, we are measuring 2 to 3W of excess heat in the cell which has settled at about T_Mica= 265 C, however we do not consider this to be above the uncertainty/noise level, yet.
We will run the current experiment for a few days or until failure whichever is sooner - it will be interesting to follow:
- any potential events
- change in wire resistivity
- change in excess heat generation
Really interestingly, it might show us the time it takes for most of the H2 to be “burned” / lost based on the R/R0. This might give us more clues as to the duty cycle of these reactions. Maybe because the cell produces a modest excess output, it can run for days, if it produced 100s Watts then maybe the celani cell would have to go through a H2 loading cycle relatively quickly. Loading is separated from excess heat - but we believe we have shown that it can occur in the same cell.
It will also give us some time to explore the robustness of our data broadcasting that was severely stressed during the early part of the test with the number of connections and data downloads. This forced a change in server that whilst faster failed.
Temp in cell may not be high enough to trigger LENR.
Specifically, the local temperature of the wire may not be high enough to create the effect.
Our cell takes more power to heat it, because less of the IR is thermalized in the quartz glass than the borosilicate glass. T_mica shows the closest value of what the temperature of the wire is, and Pr. Celani explained that optimal temperature for the effect was between 250 C and 350 C.
Furthermore, increased current may create more near IR which is partially lost thought the quartz tube, see our previous analysis.
We don't want to risk any more current in the wire at this stage. We are already have twice the power that he had in his wire. It may damage it.
Reduce the pressure inside the cell to reduce thermal conductivity and decrease H2 recombination is our best option. At lower pressure we are able to increase the temperature inside the cell (T_mica) which might help triggering LENR effect without increasing the current inside the active wire. Because we did calibrations at 3 different temperatures, we are able to set what should be the pressure inside the cell for a certain amount of power injected.
If a set pressure of 2 bar at cold is not good enough to trigger the effect, we reduce the power input drastically and decrease the pressure at 0.5 bar. In this condition, the wire will tend to retain heat instead of dissipating it.
At some point we may decide to turn on both wires in order to achieve similar current and temperature in the Celani Wire to what he had in the demonstration.
Loading process may also need to cycle for best effect
When we started the loading, we thought we had to do it over 2 - 3 days, which we kind of did. We heated the wire up to the requisite temp in H2 and let it sit there. Celani says however, that the majority of the loading occurs in the first 10-20 mins as noted above - so why have it sit there for 3 days? Maybe we should be cycling it, ie, let it cool, bring it up to loading temp, repeat... n times.
In summary, after 11 hours since beginning to heat the Celani wire with current running through it, the experiment is just starting.
Data file to be added here ASAP:
Comments
It may not be problematic to have the wire at a high temperature, but high power to achieve this may be a problem. Celani advised to use a gas mixture and lower pressure precisely to reduce thermal conductivity so that the wire can have a higher temperature without compromising the integrity of the rest of the apparatus.
When we have been able to trigger the effect a few times - it might be interesting to have a power input curve that rather than stepping slowly, slides rapidly up to activation temperature range.
Celani does not reccommend using pure H2 after loading. We have shown the ability to load to some degree with a gas mix also.
We are planning to see if hard wiring the fan on will fix the cyclical voltage issues - then the voltage on the PSU will be more dependent on ambient but I believe much less an effect than the dodgy sawtooth we saw and also slower and easy to correlate and account for.
Celani says his wires activate at over 250 - 350 degrees - we were just over the bare minimum on a wire that was on its first load cycle. David is right in saying that the gas choice is to prevent the glass softening.
@ Al Potenza
We are just looking at the evidence from the Loading/Burning phases and seeing what is needed to do this and what temperature things occur at and making a conjecture.
If they have what is claimed, then it makes no difference to us, great I say. Others still made cars even when Henry Ford was churning them out in the 10,000s and people still research how to make better engines. Many of the competitors have made some of the best cars ever made.
We believe that a simple, repeatable, easy/cheap to modify, maintain and quantify experiment that does not run at super high temperatures and threaten neutron production has the potential to allow thousands of budding scientists and professional bodies to explore this phenomena and push the science and its understanding forward.
If Rossi and Defkalion are virtually ready, as they claim, to sell commercial, stable and certified kilowatt level LENR reactors, what are you doing with single wires and arguable, low power levels? What's the point if the problem is already solved?
Your research makes the most sense if Rossi and Defkalion can not be believed. Thus, how they configure their reactors is sort of moot, don't you think?
As it happens, I don't believe Rossi and Defkalion and I think what you are doing is an excellent idea! Sorry to interrupt but I think it needed to be said.
think long - post short...
If you are using 75H2/25Ar. Then changing to 10H2/90Ar might enable you to get to higher temperatures due to the much lower thermal conductivity of Ar compared with H2 (1:10)
In absence of such detailed plans probably it's better to simply use hydrogen directly at 100% concentration, I think.
@Robert Greenyer: if keeping the PSU fan enabled all the time solves the problem and if the fan itself does not suffer from continuous use, then it is definitely a good idea to perform this modification as soon as testing conditions permit.
I also noticed in Celani's papers that he obtained the best results at wire temperatures as high as 900 Celsius.
I would suspect (from some experience) that heating these kind of wires to very high temperatures with high current is not too problematic - provided you take care not to induce too great a thermal shock - i.e. increase/decrea se current/temp slowly.
On opaque design has already been made, but requires extra calibration and is not faithful to Celani's working design - so it is one for later.
@ Ecco
We do not feel there is anything to shout about apart form the very successful loading, the resistivity change during the run and the confirmation of IR/glass effects predicted by theory. We are also learning about the unloading curve as the experiment runs.
We are aware that these wires often do not work reliably until after a few runs and also we were unable to achieve clear activation temperature range - so there is more to do.
I TOTALLY agree on getting a better PSU. Enough said. However if it is just a matter of the fan - we could just keep the fan on at full blas permanently.
@ 123 star
It was written by an American, and because of us Brits, they still use Fahrenheit - what can you do?
@ Igor
We have considered using foil, that is our second choice for moving forward but we have no calibration curves for that.
As Ecco says, we will likely reduce the pressure for the next run after a new loading cycle - because we dont have to change the apparatus, wire etc and we already, on Celani's/Mathie u's advice, have a calibration run for this to happen.
The thermal conductivity of Argon is 10 times less than H2. The addition of 10/90 H2/Ar (for example) might make sense if reducing the thermal conductivity of the gas admixture enables the necessary temperature to be reached to initiate the LENR.
@ David Jones
When we saw the bulk effect of Nicolas' theoretical prediction of IR loss because of quartz, we thought along the lines of what you have suggested - but I had not considered how to mount it and your suggestion is a great one and is a good example of posters here adding real value to the way in which we have chosen to conduct this experimentation.
It is something we can try, but we would need to run calibration curves for set up and so it will be a potential solution down the road.
For everyone
Following Nicolas' analysis and collaboration on the science of glass choice with Celani, Mathieu ordered a custom made Borosilicate glass tube from a specialist blower and we understand that it is on the way to him now. The theory pointed to this problem, Celani told us of its significance and the experiment has demonstrated the theoretical prediction.
We are acting on the research. This is the beauty of parallel and open collaboration.
It is likely that no excess heat is being generated at all, currently. It's true that this is still the beginning, though.
It's true that in other blog posts here I did propose data decimation in order to save server bandwidth, but decimation, if not extreme, retains most of the trend characteristics for each data series (especially if we're just cutting out what cannot be displayed due to lack of chart display resolution), does not smooth out the existing data.
Note: to clarify, since I write here often, I'm not a sponsor or a member of the MFMP team.
Of course, improving heat insulation would be the best option (and for the occasion, replacing the glass tube - since the inside would not be visible anymore anyway - with a steel tube might be a good idea for several reasons). That probably goes outside the scope of the current experimental phase (more or less exact replication of Celani's results with the same set-up).
If this is true, then during the USA morning with people around the reactor, apparent excess heat should decrease again to yesterday's initial values.
A bit of a disappointment, but even Celani as far as I know doesn't have a 100% success rate with new treated wires. The ones that work, though, should clearly (and reliably) do so.
RSS feed for comments to this post