Here's the inside view of the fan box mentioned in my previous post.  This box draws air through filters (removed here) from tubes buried under the building and motivates it to make the journey up through the panels and back down under the building again.  This simple process stores the sun's energy to be enjoyed by the building's occupants (plants and animals) in the cold Minnesota winter.

So, what was our process in developing and building this assembly? 

Paul estimated the flow rate required to move the energy we capture from the sun into storage in the thermal mass below the building.  The panel manufacturer's  (www.rreal.org) engineer, Graham, estimated the resistance expected from pushing the air through the panels and ducting; which was added to resistance Paul and I measured from pushing air through the tubes in the ground.  So, with total flow and pressure estimated we were ready to select the fans needed for the 12 fan boxes like this one.  We used Lau Fan's free fan selection program ( http://www.laufan.com/whirlwind.aspx) to come up with this 9" fan common to home furnaces. 

We are actually using the estimates and calculations to target a range of performance.  The real-time output temperature from the panels will be monitored and Ryan and Scott will use that indicator with their electronic control system to run the fans at whatever speed is optimum for the sun's energy at that moment.  So, any error in our estimates will simply mean the fan will run slightly faster or slower than we predicted.  Flexibility is nice when you're covering new ground. 

These fans often come supplied with a motor mounted to the top of the fan, with a belt drive; or with a motor stuck into one side of the fan directly driving the wheel.  We didn't have room to mount the motor on top, and we couldn't afford to sacrifice performance by obstructing the intake to the fan with a motor.  So, we elected to mount the motor to the side ourselves to provide for relatively free air flow.

 On the left of this engineering "solid model" you can see the representation of the two stacked solar panel frames and the open "trench" door at the bottom.  We used this model to understand how the geometry of the building would interact with the solar panel system.  Since RREAL's new solar panel design was just going into initial production as we needed to install them, the model was invaluable in working out many details even before their first panel existed.

Here is a detail drawing we used to take our first guess at what the fan box would look like.   

Here is a model of the final design. 

After building a wooden prototype, then a metal one (as our shop capabilities increased) several refinements were made.  When used properly, drawings and 3D models can really help you get close with your first proto-type.  But there is nothing like having the object in hand, trying to install it, service it, etc.  That's often where real problems are found and breakthroughs are made.  After Ryan and Ben went through that process to their satisfaction, and we all had a chance to kick the tires, it was time to go back to the computer and refine the model to our final design.

From the final design model this 2-D sheet metal pattern was generated.  Soon we will be able to send a pattern like this to our in-house CNC plasma cutter and cut it out of any sheet metal in minutes.  Since that wasn't ready, we simply emailed the file to a local shop and had them make up a batch of "blanks" for us to bend up.

This is our cool new "Magnabend" brake that we used to form boxes from the sheet metal blanks.  You can see the heavy steel bar sitting inside the small sheet metal box Ben is making.  It's held down by an electro-magnet while the flange to the left is bent up with a hinged steel bar as shown.  With typical brakes the hold-down bar is mechanically attached to the unit and thus much less flexible in what you can do with them.

This is our nice water-cooled spot welder.  Electricity shorting between the copper tips of the arms makes a strong little round weld. Cooling water circulates through them.  As you do many welds, heat would otherwise accumulate and melt the copper.  It even has an air cylinder inside that uses compressed shop air to pinch the arms together on your part.  It's rather fancy.  With the struggling manufacturing sector in this country, great deals on equipment can be found if your timing is right.

Here Ben is adjusting a motor in to align properly with the fan wheel's shaft so wear on the bearings and energy consumption is minimized.

 We designed the motor mounting plates to work with the slots on the motors to give us adjustment up, down and to either side.

All of the holes, including the slotted ones, were punched out in seconds using our ironworker.  It's like a combination paper punch and scissors but only for metal up to 3/4" thick!  Metal is a great material to build with if you have the right equipment to tame it with.