/

Solar-powered space flight

5b. Creating ultra-lightweight solar power concentrators: Rotationally symmetric aplanatic two mirror arrangements

[this page | pdf | references | back links]

Return to Abstract and Contents

Next page

 

5b.          Rotationally symmetric aplanatic two mirror arrangements

 

5.5          We concentrate on rotationally symmetric aplanatic two mirror arrangements as certain of these exhibit all of the above characteristics. For those readers not familiar with optical theory, ‘aplanatic’ in this context in effect means ‘creates a sharp focus’. Optical theory requires aplanatic arrangements involving strong magnification over a wide range of angles to have layouts that satisfy the so-called sine criterion, see Klein and Furtak (1986). The cross-sections can be iteratively determined as follows, see also Figure 7:

 

(a)    At the k ’th iteration, a light ray starting at exactly the object point, , should go through the image point, , after striking the first mirror at  and the second mirror at , say.

 

(b)   We arrange for  both to lie on the tangent to  and if a light ray starts at , strikes the first mirror at  and the second mirror at  for the ray then to pass through ,  small (positive or negative). We identify the position of  in a like fashion but so that if a light ray starts at , strikes the first mirror at  and the second mirror at  then it passes through . In the limit as , the layout is then image forming and if the correct choice of  is made (see later) then the mirror pair is aplanatic (of order 1).

 

Figure 7. Schematic diagram illustrating derivation of mirror pair cross-sections

 

5.6          This can be mathematically re-expressed as follows, see Kemp (2001) and Kemp (2003). We have for :

 

 

As the light rays are being deflected by reflection, we also have for :

 

 

We choose seed values of  and  (for ) so that the resulting mirror layout satisfies the sine criterion. For a far away source (e.g. f = –109) we can without loss of generality set . The sine criterion is then satisfied if:

 

 

We can iterate either from shallower angles to more oblique angles, or vice versa. We can switch between the two by using as seed values later iterated results and reversing the sign of . So without loss of generality (and if we wish to have the maximum possible angle span) we can iterate from highly oblique angles, choosing  equal to a small number close to zero, say 10-9, and choosing  and , where  and  are arbitrary real numbers (positive or negative). Then we need  for the initial parameters to satisfy the sine criterion. Without loss of generality we can choose . Given these initial seed conditions different layouts arise depending on whether  is positive or negative. The choice of the sign of  is made so that the sine criterion remains satisfied as  changes. Only one choice will work depending on the layout, and, given the conventions we have adopted above by choosing , the correct choice seems to be .

 

We iteratively update the values of  as follows for a small :

 

 

where as deflection occurs by reflection:

 

 

We end the iteration no later than when light rays cease to be able to pass freely through the mirror arrangement, once the cross-sections have been rotated around the x-axis to produce the complete three-dimensional mirror surfaces.

 

5.7          Each of the signs of ,  and  can be either positive or negative leading to 8 possible two mirror layouts. In 6 of the 8 cases there is a discontinuity in the feasible angle ranges of rays striking the image plane when we change the value of  from slightly below 1 to slightly above 1, see Kemp (2003). Thus there are 14 possible overall layout types, whose optical characteristics can be summarised for our purposes by the following metrics:

 

(a)    effective aperture area factor = area of first mirror perpendicular to sun's rays, expressed as a proportion of the maximum possible were the angle span of rays falling onto the image plane to be the complete range from wholly oblique to exactly perpendicular to the image plane. The higher this is, the closer to the thermodynamic upper temperature limit such a concentrator can approach.

 

(b)   mirror surface area factor = the total surface area of the two mirrors combined as a multiple of the effective aperture area. The closer this is to 1, the less is the mirror surface area required per unit of power delivered, and therefore the higher the power per unit mass that the concentrator can deliver.

 

(c)    aberration factor = average maximum second order aberration for sunlight in the vicinity of the earth (i.e. for a far away source subtending approximately a semi angle of 0.267°). The lower this is, the closer the mirror pair can get to the thermodynamic upper limit when concentrating sunlight without resorting to off-axial adjustments, see Kemp (2001).

 


NAVIGATION LINKS
Contents | Prev | Next


Desktop view | Switch to Mobile