Published using Google Docs
LENR_Stick_1 Loading Report
Updated automatically every 5 minutes

Quantitative Measurement of Initial Hydrogen Loading of 400 Layer Celani Wire

Ryan Hunt, R&D Manager, Hunt Utilities Group

Malachi Header, LENR Engineer, Hunt Utilities Group

Wes Baish, LENR Technician, Hunt Utilities Group

Mathieu Valat - Chief Scientist, Martin Fleischmann Memorial Project

Starting on Friday, 19 April 2013, we started a hydrogen loading test on a 400 layer Celani wire in a new test cell and calorimeter.   The treated isotan wire was provided to us by Celani a few months ago and kept in a polyethylene bag till it was installed in the test cell.  We installed about 40 cm of the wire into the test cell.

The unique calorimeter is described in this document. The calibration data is also in that document.  The test cell containing the wire is described in detail here.

The experiment log for this test apparatus and test is here: Experiment Log

The Raw Data File (30 second averages of approx 2 second reads in a csv file format) is available to analyze, too.

The goal of this test was to try out the calorimeter on a real test to verify that it works as intended.  Due to limitations of the web site organization, we opted to consider this test just part of the commissioning of the calorimeter and not publish it ahead of time.  We soon regretted that as interesting data started coming out.  While we still could really use a web guru to help us organize the site for multiple experiments, we decided to share this data and get many more minds working on it ASAP.   Unfortunately, the live data stream will be a little slower coming since we haven’t figured out if we want to share the internal test setup or move that calorimeter to a different computer.  This document is intended to publish the data and invite others to examine the data, the analysis, and the conclusions, and add interpretation.  

At the time of this writing, we have done six experimental phases with this wire all aimed at watching the loading process in more detail.  (The Experiment log is here: Concentric Tube Calorimeter Experiment Log)    

Figure 1 - Power input levels of all the phases of this experiment to date

The test phases are:

Phase A

In order to explore the range of loading we did the Initial loading step by varying the power input level and watching the performance of the cell.  It is worth noting that our calibration parameters were not accurate on power out, but they were fairly close.

Figure 2 - Phase A Input Power and Output Power show slight difference - primarily from calibration parameters being out of date.  

Figure 3 - Internal Temperatures over time within the middle of of the test cell.

Figure 4 - Resistance of Celani Wire in the test cell over time.  The resistance drops over time at some temperature/power levels, but rises quite a bit at the middle power levels.  

Figure 5 - Pressure within the test cell.

Perhaps the most intriguing piece of data so far is the pressure drop of the hydrogen, providing our first direct evidence that the hydrogen does actually absorb into the wire.  The drop of pressure happens most rapidly at the 6 watt and 7 watt power steps, corresponding to cell temperature around 174 C.  The active wire is being directly heated with current, and is sheathed in fiberglass insulation, so its temperature may be somewhat higher than what the thermocouple measured.

An interesting thing about the shape of the pressure curve is that the slope decreases in the middle of the 7 watt step and then never substantially increases again, hinting that perhaps it “filled up” on hydrogen.

Another particularly interesting correlation is that the resistance of the wire rose considerably not exactly during the drop in Hydrogen pressure, but just after, instead.  That implies that the absorption of hydrogen and the reduction might not the same process, but may still be related.  

The small temperature changes during the loading are also worth noting.  As the pressure drops, the thermal conductivity of the gas decreases and that may change the relationship between the hot wire, the thermocouple adjacent to it, and the shell of the tube it is in.  

Phase B

 In the next phase of the experiment, we set the calibration parameters to the most accurate version we had, refreshed the hydrogen with a pressure and purge and then set the power to 6 Watts, which was the most active power level in phase A.

Figure 6 - Pressure over time at 6 Watts power input.  Note the sudden peak as the temperature rise drove a pressure increase, followed by a rapid absorption.

Figure 7 - Resistance over time corresponding to the same timeframes as in Figure 6.  

Note that on the far left of Figure 7, when the power was set to zero, the resistance measurement was noisy and offset due to very low current readings.  When the power came up to a measurable value, the resistance peaked at 8.2 ohms and then rapidly fell off.  It is curious to note that the resistance bottomed out after the majority of the pressure decrease had happened.  Then, as the pressure continued to drop the resistance rose, again.  

Figure 8 - Temperature in the cell over time throughout Phase B.

The temperature in the cell varied gradually a couple of degrees over the roughly 65 hours of this phase.  Perhaps the most interesting feature to this graph, though, is the peak at the very beginning.  I have no explanation for that feature.  Anybody else?

Phase C

After seeing the cell come to a very steady state, we decided to sweep the loading range, again, and see if we see any more notable changes.  We started at 6 W, again, for a brief period, then started to step from 3 W back to 12 W in 1 Watt, 2 hour steps.

Figure 9 - Phase C internal temperatures. 

Figure 10 - Power Steps in Phase C

Figure 11 - Pressure during phase C.  All small pressure changes with a little noise.  Definitely some pressure drops during the higher temperatures.  Furthermore, it appears that the rate of absorption is higher at higher temperatures.

Figure 12 - Resistance over time for phase C.  Very similar shape to the pressure.  The little steps in it during the initial 6 W step and the following 3 W step are interesting.  We zoomed into that, next..

Figure 13 - Resistance during first 2 steps of Phase C.

Figure 14 - Power input and output for first 2 steps of Phase C.  

The difference during the 6W step was very close to 0.25W, which was higher than any other time before or since that we had it at that power level.  The calorimeter’s uncertainty is approximately 100 mW at that input power.  After we turned it down to 3W, we see that the output power suddenly goes back to what it was before.  This small level of excess power could be from some adsoprtion/absorption phase, some quirk in the relatively untested calorimeter, or any of a number of other reactions going on.  It did end at the same time as a sudden resistance change, though.  It does not appear to coincide with anything notable in the pressures, though.

Figure 15 - Pressure during first 2 steps of Phase C

Phase D

After seeing some pressure and resistance changes at the higher temperatures in phase C, we decided to try some higher temperatures and see what happened.  Phase D went between 12 and 24 Watts input power in 3 Watt, 2 hour steps.  

Figure 16 - Power in and out during Phase D

Figure 17 - Pressure during Phase D

Figure 18 - Temperature of thermocouple during Phase D

Figure 19 - Resistance during Phase D

Both the pressures and the resistance showed interesting behavior during the power steps.  I don’t know exactly what to make of them.  There was, however, no difference between the input power and output power above the uncertainty of the calorimeter calibration.

The internal temperature as measured by the thermocouple show an interesting round topped shape to the higher steps.  Perhaps this indicates some strange time constants or heat transport modes?

Phase E

After seeing the slope of absorption at 24 W, we decided to bring the power back to 24 W, again and watch for additional loading.  The additional pressure drop fit the curve of the the last step of phase D beautifully and continued till it leveled off.  

There were no particularly interesting effects on resistance of the wire, or on the  output power.  The temperature on the thermocouple does seem to drop as the pressure drops, though.

Figure 20 - Pressures in Phases D and E.

Figure 21 - Temperature on thermocouple in phases D and E showing a trend for the temperature to drop as the pressure drops.

Phase F

After seeing the system come to a boring steady state at 24 W, we dropped the power to 6 W, again, to see if there would be any more loading or anything else notable to watch.

We ran across a few, minor, interesting things to note.

The first thing is that there appears to be a minor effect from ambient temperature swing over the day.  The next thing is that the pressure is modulated approximately 20 mbar from the temperature swing and, curiously, is overlayed with a sawtooth wave form about 1 to 2 mbar in amplitude that reverses depending on weather the pressure is trending up or down.  

Figure 22 - Pressure during phase F with interesting sawtooth wave function on it.

Another interesting observation was a set of  distinct changes on the thermocouple temperature that correlated with resistance changes and distinct noise in the Power Out measurement.

Figure 23 - Thermocouple temperature in Phase F with an interesting drop and rise.

Figure 24 - Resistance in Phase F with steep resistance changes corresponding to temperature changes.

Figure 25 - Power Out from calorimeter showing a 100 mW offset and +/- 30 mw swing out of phase with the temperature change and a few other little features.

The take home conclusion from this phase of the test is that there are dynamic little things going on in this system.  Understanding what they are and how any of them *might* relate to excess power generation will be critical going forward.

Hydrogen Loading Quantification Calculations

The calculations are all shown in this spreadsheet: LENR Stick Molar Loading.

By looking at the pressure drops for each phase of the test and knowing the volume of the test cells, we were able to calculate quantitatively how much hydrogen was absorbed.  Loading calculations were done with the ideal gas law, PV=nRT, estimating the volume of the cell to be ~13cc (P=Pressure, V=Volume, n=moles of gas, R=Universal gas constant in bar*m3*mol-1*K-1, and T=absolute cell temperature). Solving the ideal gas law for n yields moles of H2 in gas phase in the cell. Subtracting final moles present (using the same method) gave half the dissociated moles of monoatomic H molecules loaded in the material because H2 must split into 2 separate H atoms to dissociate into the metal.  Cell volume estimation and wire mass are the greatest contributors of error to this calculation because they are estimated instead of directly measured.

The most pertinent result is that the Hydrogen  absorption appeared to be rather complete as it shows a final molar ratio H/Ni of 1.52.  This seems particularly high.  Some of this may be Hydrogen that reduced a metal oxide, turned to water, and then condensed in the cool inlet side of the test cell that is at room temperature.  Leaking is not considered a significant contributor to pressure loss because of the extremely stable pressures during stable conditions in Phase F of this experiment.

We are assuming that the copper contributes minimally to the absorption of hydrogen because it’s solubility is orders of magnitude lower than Nickel’s (according to this graph from  http://www.rebresearch.com/H2sol2.htm)

Mark Snoswell from Chava Science in Australia was kind enough to share some insights on what may be going on.  He pointed out that the oxides on the surface do get reduced to metal and water.  If the whole cell was hot, the steam pressure would become an issue and we may even see the net pressure rise as the hydrogen gets turned to water in this way.  Since our cell is partly at room temperature, though, any moisture content above the vapor pressure of water at room temperature would condense out.  We may decide to adapt the apparatus to have a small glass tube at the low point so we can see how many microliters of water condense out.

Other conclusions?  

The delay between the pressure drop and the resistivity change observed at multiple times during these tests is something that I would appreciate some help understanding.  It makes me think that the resistivity only changes as the lattice is trying to squeeze in the last little bit of hydrogen, but that is pure speculation.

We have not pulled out the R/Ro data because of the problem with the power going to zero and giving us too little signal to test the resistance.  We encourage anybody out there to pull that info from our data, though.  Future experiments will include a pre-start-up and post-turn-off step where we will turn on the power to a few milliwatts and get an initial and an ending resistance at a baseline power.  

This has been a cursory report on the very first observations with this new calorimeter.  Did we miss something or misunderstand something?  What should we do next?  Let us know.