Differential Celani Cell Tests - Update 2

The US team has reconfigured their V1.3 Celani type cells to test in a pure differential mode between cell A and cell B.  Mathieu had done the same with the EU cells just before ICCF.  This configuration will allow us to set the cells to any temperature and pressure and power configuration and simply look for a large temperature difference between the active cell and the passive cell.  The US cells are now operating.  The EU cells will start up soon after some troubleshooting in the pass-throughs.

The reconfiguration in the US was made by taking the "used" Celani wire from cell A, taking a couple centimeter sample off of it, and placing the rest into Cell B in place of the heater wire.  Then we replaced that wire in cell A with a fresh nichrome wire.  That means cell B has 2 Celani wires and Cell A has 2 nichrome wires.  The cells are plumbed together  and the connection between them will be open so that they always see the same gas and pressure conditions.  

 US Cell Configuration

Loading:  Start at 5 bar H2 and 70C, raise the power in 2.5 Watt steps and wait for the resistance to level off at each step.  

First 24 Hours of loading test:

We started the low power first step of this test of this yesterday. As the power was turned on and then turned up a couple hours later, the CuNi2 wire showed a negative temperature coefficient by dropping in resistance while the CuNi1 wire, having already been loaded, increases , instead.

During the previous tests we did not see anything more than a watt or two, and because of ambient changes, limited power, vacuum variation, and other unknowns that may have been occurring, we could not call that test a definite excess energy.  With the flexibility that running in differential mode gives us, we should be able to test to a higher wire temperature and see if we just weren't getting it hot enough, or if our vacuum was too strong.  The tradeoff is that our calibrations will not be accurate at all since they were prepared in 1 mBar vacuum and only up to 25 or 30 watts.  The really nice thing is that abandoning calibrations mean we should be able to swap out these wires in cell B with new ones every week till we find one that shows enough signal to be compelling.

Update 8/16/13

Every once in a while I wish I was faking the whole thing and only going live for a few hours at a time so I could show some exciting results and be pretty sure things would look perfect.  Yesterday we made a mistake and it was, thankfully, caught by Ecco, our most consistent comment contributor.  When we reconfigured the wires in the cells, we accidentally moved the NiCr wire from cell A instead of the Celani Wire.  Malachi came in late last night to correct the problem.  (Note:  this does explain why the "loaded wire" in the graph above was showing us a positive temperature coefficient :-/ )

Ecco noticed what happened as he was scouring the data and noticed that one of the wires in Cell A was decreasing in resistance significantly.  The graph below is for Cell A.

Looking at the graph above, it is clear that the blue resistance line is in the range of the Celani wire instead of the NiCr.  What amazed me when I saw it, though, is that the "loading" started after a few hours at 70C mica temperature.  I did not expect to see resistance dropping at that temperature.  The knee on the slope about 09:00 when nothing else seems to be changing except the pressure is interesting.  This is the same wire that we spent a lot of time trying to load after the first few runs and thought we were seeing it get less active because it just wouldn't seem to load much, anymore.  What is going on inside this wire?  

You can keep watching this wire with us, but you'll have to look in cell B, now.  If we make more mistakes, let us know.  We want to make many mistakes and fix them quickly so we make the most rapid progress.

UPDATE 2 - 8/19 

Over the weekend, we continued the loading of the active wires in Cell B.  The plan was to up the temperature and wait for the resistance to level off.  Since were weren't going to be around to watch, though, we gave the cells 12 hours per step.  As you can see from the graph below, the rate of resistance drop didn't seem to change much over the weekend.  Now we will leave the power steady till it does.  The rate of resistance drop at such a low temperature is something pretty interesting to me.  

Since Hydrogen seems to be percolating into the new wire and the old wire was already mostly loaded, I thought I would see if there was any temperature difference, yet.

There is, but it was the control cell that was consistently slightly hotter.  Interesting, huh?  It should make any positive looking result a little more believable, anyway.