Burned out a wire

This is the type of thing that is not as much fun to share, but it is just part of the process.  On Saturday night at 10:45 we were nearing the end of a calibration cycle in Hydrogen gas when, all of a sudden, the temperature rose and then rose again to almost 500C on the face of the mica in the middle of the cell.  

Then it dropped off, again.  We studied and studies the data for hours trying to figure out what happened. Upon inspection we saw the following.  

The key things to notice are the broken nichrome wire with heat discoloring near the ends, the ceramic adhesive that has let go of the mica, and the the gap between the lower mica piece and the stainless center tube.

What happened is that, at near full power, the copper wires connecting the end terminals back to the the connection end expand.  Without the ceramic holding it in place, it sagged and shorted against the channel 1 wire.  This short made it draw more power and get the cell hotter.  There is evidence in the data that this happened in two steps till the channel 1 wire (being thinner than the copper) burned open and the cell cooled down again.

While frustrated, we are glad to discover this before we put the valuable wire in the cell.  It is also nice to know that the glass held at that temperature.  The higher power levels also caused one of the power sensors to give bad readings, which is also a good bug to get worked out now.

The solution we came up with is to lace the copper wire through holes we drilled in the mica.  (For you R&D newbies out there, this is all par for the course as we figure out the way things work in harsh environments.)

Since Malachi is out for a couple days, this is going slower than it has to, but we'll get it back in service tomorrow.  

In other news, the Euro Cell got delivered today!  I'm sure there will be a nice blog entry coming from the unpacking and setup of that.