This page covers building a simple mechanical earth/sun simulator
that will allow you to build and test small scale models of reflector
designs and see how they do at various times of the day and year for any
point on the earth. It can be built in a couple hours.
|
Back to the main page on reflector design and use here...
The picture shows the basic parts and operation.
The earth's equatorial plane is represented by the "equator board" (see picture). The equator board is hinged so that it can tilt through plus/minus 23.4 degrees to model the tilt of the earth's axis toward or away from the sun as the earth goes through the seasons (23 degrees toward the sun in summer, 23 away from sun in winter, and flat at the spring and fall equinox).
The "turntable board" is pinned to the center of the equatorial board and can be rotated to model the hour of the day. The picture shows the turntable board at about 2 hours before solar noon.
The "ground board" is mounted to the equatorial board at a tilt of (90 -
latitude) degrees. This board represents the surface of the earth at the
latitude the collector will be located. In the picture, the ground board
is set for a latitude of 45 degrees.
If you are located at latitude 30, then the ground board would be mounted at a
tilt of (90 - 30) = 60 degrees to the equator board.
The sun is represented by a spotlight that remains in a fixed position as the equator and turntable boards are rotated to model different seasons of the year and times of the day.
Showing the "sun" spotlight and the simulator.
Normally sun is mounted further away to reduce light angle errors.
I use a 75 watt halogen "spot" light.
The height of the sun should be the same as the height of the middle of the collector/reflector model. The sun should be aligned with the south or solar noon axis of the turntable board with the turntable set in the 12 noon position.
To use the simulator, you mount a scale model of your collector and reflector on ground board. You then darken the room, and turn on the spot light that acts as the sun. You can easily see how well your reflector design actually reflects light onto the collector by varying the tilt of the equator board to represent the different seasons, and by rotating the turntable board to model the time of day.
Some more pictures of the simulator in use:
In the winter solstice position. |
In the equinox position. |
From the back toward the sun. |
Measuring light levels on reflector, collector, and ground using a photographic light meter. |
The simulator is very fast and easy to use -- you can go from 2 pm on the summer solstice to 2pm on the equinox or winter solstice with just one quick hand movement -- you can easy flip back and forth between different times of day and times of year in seconds and see how the pattern of light reflected onto the collector changes. By building the reflector out of aluminized Mylar cemented onto cardboard, you can try different reflector shapes, orientations and curves very easily.
Note that this earth/sun simulator could also be used for modeling shadow patterns, modeling light on collectors (especially ones of odd shape), modeling light on home models, shadows of mountains on valleys, ...
Light meter readings can be taken on the various surfaces of the model collector and reflector, and I see no reason why the relative values of these light light levels would not be in the same ratios of the actual sun light levels on the corresponding real objects. If these light meter readings are taken on different test runs, then the exact distance of sun to model must be duplicated, and things like aging or the sun spotlight and line voltage variations should be considered.
There are no doubt many good ways to build this kind of simulator, but here is what I did. If you have any improvement suggestions, please send them in.
See picture just below.
The equatorial board support should be high enough to allow the equatorial board to tilt 23 degrees down toward the front and also toward the back. When the equatorial board is parallel to the base board it is in the equinox position, when it is tilted 23 degrees down toward the spotlight it is in the summer solstice position, and when tilted 23 degrees down toward the back, it is in the winter solstice position.
See the picture just below.
See the picture just below.
Make the ground board so that it tilts up at an angle of (90 - latitude) degrees to the turntable. This board will then accurately represent the ground surface at your latitude. For example, if you live at 40 degrees latitude, the ground board should be tilted up at 90-40 = 50 degrees from the ground board. Think about the angle your front yard makes with the equatorial plane of the earth -- that's the angle you want (assuming your front yard is not a hill).
Add the angle stops so that the equatorial board is at the summer solstice position when against the front angle stop and at the winter solstice when against the back angle stop. For the equinoxes, I just us wedge piece of plywood of the right height between the equatorial board and the base. Assuming you live on earth, the stops will be cut at an angle of 23.4 degrees.
Board wedged in between equator board and base holds precisely in equinox position. |
Setup the spotlight so that it is a good ways away from the model.
You can set up the size of the simulator and the model to about anything you want or have a large enough shop for. On mine, the diameter of the turntable is 24 inches. This allows me to have models of collectors and reflectors that are in the 10 inches tall or wide area. This model size allows me to work in a scale of about 1 inch per foot compared to the actual collector and reflector. This seems like a good size to work with to me in that the models of this size are quick and easy to make out of plywood or cardboard, but are large enough that you don't have to be overly fussy to get good accuracy.
Since the spotlight is a lot closer that the 90 million miles our sun sits away from us, there will be some error in the angle that the light comes from on some parts of the model. The part of the model that the spot is centered on will have no error. But, if the spot light is 20 ft away, then a model with a radius of (say) 5 inches will have a maximum error in the angle of light of arctan(5/240) = 1 degree -- for most things this is not really a problem.
I will add more examples of using the simulator here as time goes on -- if you build one, please send pictures of it and how you use it.
The collector modeled in the picture just below is designed to reflect more light towards the center of the collector. In this way, the full reflected pattern stays on the collector for a long period of time -- in the case of the configuration shown below, the full reflected light pattern is on the collector from 10 am to 2pm solar time. A reflector without the side to side tilt up would have half of the reflected pattern off the collector by 2 pm.
blog comments powered by Disqus
Gary June 1, 2011