The first test run of the Steel and Glass cell suggested that it took approximately 6% more power to maintain the same volume of water at the same temperature in the control cell when compared to the active cell.

New Experiment Improvements

So, spurred on by the seemingly positive preliminary run of the Steel and Glass cells and taking a plethora of advice from the crowd, Nicolas sought to make some improvements to the experiment. The biggest revelation came from Celani, who informed us that the teens class wires (14 & 16L) that we used in the first run were actually the worst capability wires, this was explained in the previous blog post. So for this run, a new 14L wire was joined by a 360L wire.

You can see in the following images there were several improvements made to the experiment, many suggested by our followers. Chief of which was that the water containers were partially insulated with styrofoam boards.

Figure 1. Overview of the apparatus

Also, water stirrers were added and mounted in the same position and angle, pushing hot surface water down and creating a vortex.
Figure 2. View of stirrer blades set at angle in water and the driving motor mounted outside.

Water thermocouples were mounted on floaters to measure the water temperature at the same height from the water top level. This was ultimately found to be unnecessary, since the water temperature was very homogenous due to the action of the stirrers. In fact, the variation between bottom and top was only ­±0.1°C.

In the image below, the laptop on the left is running HugNetLab for data capture. The PSU with the blue display underneath is powering the water stirrers (2x DC motors in series), whilst the Dell laptop on the right is managing the duty cycle for them.

The 2 TTi dual channels PSUs are powering the 4 wires.

• The top PSU is for Cell#2 (active)
• The bottom PSU is for Cell#1 (passive)

Figure 3. Arrangement of power, control and data acquisition hardware.

A complete gallery of the setup can be found here.

The Second Run

The second experiment was set up and run between 9 and 25 May 2013, the logbook of which is here

https://docs.google.com/spreadsheet/pub?key=0Au5C1oywPfSKdGhxM1F6ZGNwSmxOZUh3bEZyMmRMMHc&output=html

Unfortunately, whilst the data was recorded on a laptop, it was not able to be uploaded to the MFMP’s data repository due to what turned out to be a failed hard drive and a problematic software upgrade process. Until the data is available on-line however (which should be very soon), it is contained below in some spreadsheets for you to work through.

Here is the data for the wire loading in Cell#2 and the runs for both cells (11 days).

https://docs.google.com/file/d/0Bz7lTfqkED9WSkpVYjdrc3ZWNzQ/edit?usp=sharing

Experiment Configuration

For each cell

• 1 TC inside the glass tube for T_mica
• 3 TC in the middle chamber for T_glassout1, T_glassout2, T_glassout3

Cell#1:

• 2 Nichrome wires
• Hydrogen pressure: 2.054 bars @38.5°C inside the glass tube
• Helium pressure: 127 mbars @ 35.0°C inside middle chamber
• Input power: varies from 60W to 70W (30W to 35W in each wire)

Cell#2:

• Blue channel: 360L constant wire
• Red channel: 14L constant wire
• Hydrogen pressure: 2.054 bars @38.5°C inside the glass tube
• Helium pressure: 127 mbars @ 35.0°C inside middle chamber
• Input power: 60W (30W in each wire)

Preliminary Results for tentative discussion

Finding 1: Approximately 19% loading (0.81 R/R0)

Graph 1. Hydrogen loading related resistance changes over time.

Finding 2: There seems to be good evidence that more power is required to get the passive (control) cell to achieve the same water temperature as the active cell.

The graph below shows total input power for each of the cells (LHS) and the temperature of their corresponding water bath (RHS) over time. Cell 1 is the passive, cell 2 is the active. Annotations have been added to note events recorded in the log book. Note that the rate at which ambient affected the cells is relative to the water temp in the cells to begin with. Unfortunately, no ambient temp was recorded, but we feel there is some important information in the data anyhow.

Graph 2. Over view of power and temperatures in cells with logbook event annotations.

For detail purposes, the graphs below show how resistance changes in the Celani wires required power to be adjusted to maintain an approximate 60W total in the active cell.

Graph 3. Resistance change in active cell over time.

Graph 4. Fine power adjustments in active cell over time.

Preliminary Conclusion

We have an experiment design that is extremely easy to understand and that if the results hold up to scrutiny and are repeatable, we might have a suitable lab rat.

Further Work

In the next run, we will be using 2 X 400L class wires. We also intend to run for a longer period of time in order to get a more settled result. Ambient forcing and water level top up are problems seen in run 2, these are to be dealt with in the next test. The former by sitting the cells in chest freezers with the power off and the latter by sealing the tops of the containers.

We may later want to consider a mass flow variant of the outer shell which would allow accurate measurement as well as faster response times. Also, a larger inner cavity/more passthroughs would allow experiments on potential trigger and stimulation approaches.

UPDATE#1 - Last block of data available

As the data upload is taking so long, Nicolas has pulled the last block of data for this run from the laptop. To avoid you having to wait to see it, you can grab it from the link below, we'll try to find time to produce an annotated graph similar to graph 2 above.

Last block of data for run two

Please dig into the data and yes, we know that ambient would have been hugely useful here.

UPDATE#2 - Should we use a different heater?

So a few weeks back, whilst this run was underway, Celani phones us and says that he is stopping using NiChrome wires as "passive" wires  in his experiments because of their potential to exhibit excess heat.