Mirror Assemblies That Used Sealed Warped Glass Mirrors

Before 1984, our solar collectors either directly generated low pressure steam for heating, distilling alcohol, cleaning aircraft parts, and curing concrete blocks or remotely boiled water using heat-transfer oil. When converting thermal energy into power, higher temperature is better within the physical limits of materials. Also, tiny hot bodies lose less energy than bigger ones at the same temperature, so smaller is better (and often less expensive).

High performance solar collectors typically direct sunlight hitting a large area of mirrors into cavity receivers, insulated vessels open at one end. One optical goal for developing solar concentrators: direct as much sunlight as possible into the smallest hole. Since the sun has a finite diameter, sunlight reflecting off flat mirrors continues to spread and images get larger farther away. Flat mirrors that are much smaller than a target can reflect sunlight inside the target area but making and aligning each one takes time. Heat or other forming larger parabolic mirrors that would reflect more than 90% of the sunlight hitting it into an opening is difficult, and probably expensive. We found it straightforward and easy to warp flat glass mirror facets so each delivers, some distance away, an image of the sun much smaller than itself.

This Small Receiver Intercepts Sunlight Reflected Off a Large Mirror Area. Note: The Receiver Allows Only Brightness from Around the Sun in the Image. The Diameter of the Actual Receiver Opening Is Half the Diameter of the Insulated Cylinder.

These Mirror Assemblies Directed Concentrated Sunlight Into a Receiver That Generated Power with a Turbine

In 1983 we built our first solar collector where we bent flat mirrors into the shape that approximated a parabolic surface. Because we wanted our mirrors to last more than 30 years, we developed a technique to encapsulate the silver reflective surface inside glass (to isolate the silver from the environment). We modified approaches used to manufacture insulated glass windows that also have to perform well for a long time. These glass-mirror sandwiches (360 @16.5 x 36.5 inches in the first) used spacer strips along the two long sides between the mirror and glass. A thinner strip of adhesive was then applied midway between the spacer strips. This warped both the double-strength (0.12 inch thick) backing glass and the single-strength  (0.09 inch thick) mirror so each formed shallow troughs. We used insulated glass techniques to absorb moisture and seal the volume between the sheets of glass. The solar image on a target at the focal length of a mirror facet at this stage was a narrow line, brighter than the sun but longer than the facet. An aluminum strip to attach a “pull-back” was bonded on the back of the mirror facet down the long centerline. When mounted in a mirror assembly, the corners of each mirror facet were fixed so facet images were superimposed on a target at the concentrator focal length. A turnbuckle connected to the center of each facet was then adjusted until its long solar reflection became round with an area less than one tenth the size of a mirror facet. “Pull-backs” warped the trough-shaped facets into parabolic shapes.

Front View of the Above Solar Collector Showing Warped Glass Mirrors Mounted on 24 Mirror Assemblies, 12 with Extensions

In 1986 we began building a larger concentrator (3,200 square feet) that used larger glass-mirror facets (392 @ 24.5 x 48 inches). We used the same techniques as above but integrated an aluminum frame on each facet to make them much more rugged, able to be stacked for staging and shipping, and easier to mount on mirror assemblies.

View of Mirror Facets Mounted on Mirror Assemblies of the Solar Collector Shown at the Top of This Post

 

Close-up Showing a Reflection of the Receiver in One of the Mirror Facets. The Actual Receiver Opening Is Too Bright to See But the Aperture Skirt Tubing That Protects the Receiver Structure Is Visible. Curvature of Straight Elements in Reflections Illustrates Mirror Facet Are Slightly Curved

In 1990, in our final experiment with laminated glass facets, we adhesively bonded a large sheet (4 x 7 feet) of double-strength glass to a parabolic structure formed out of grid of aluminum extrusions. We bonded a single-strength mirror to this and used insulated glass techniques to seal the volume between the sheets of glass. This large assembly intensified sunlight over 30 times but proved cumbersome. Safely handling such large sheets of glass requires special fixtures and would be difficult to do without breaking some. Losing even a small corner of either a mirror or backing would probably require recycling it. This test also uncovered another limit: bending glass breaks if the tension at any surface exceeds the tensile strength of the material. Any flaw, such as a scratch or chip, concentrates local forces and greatly reduces how much stress can be applied without breaking. Thin glass bends easily and can safely establish a spherical (or parabolic) surface with a much tighter radius (focal length) than thicker glass. The large mirror facet above was aimed at concentrators larger than 6,000 square feet (able to direct more than 500 kilowatts of sunlight into a receiver).

Worldwide, serious research into high performance solar thermal technologies dwindled in the late 1980s. We had to find work in other arenas to put our kids through college. In my spare time I continued to make tabletop models over the next two decades to tease out solutions to many problems and awkward approaches we encountered in earlier work. Some of our mirror assemblies had been outdoors since the early 1970s and most of their facets were still in good shape more than thirty years later. With the right edge treatment and organic coatings, thin glass mirrors can weather outdoors in upstate New York for many decades. Since we have more than a thousand new and used glass mirrors from earlier work, I developed a technique for mounting them without a backing glass in an array to form a mirror assembly. The next piece will cover Mirror Assemblies That Warp Simple Glass Mirrors.

:

Dish Solar Concentrator Design

Over the next few months I will describe how I design point focus solar collector that will provide all the hot water and half the heat for my home. I will document how I manufacture the parts, assemble them, erect the structure, mount and align the mirror assemblies, connect the drives and plumb the receiver. The illustration below shows a computer model of an early version that has evolved over the past year.

Mirror Assemblies: Size, Quantity and Configuration

I will start by describing how I make mirror assemblies. They make up the concentrator and focus sunlight on a receiver designed to harness concentrated sunlight efficiently. Later I will cover the design of girders and trusses that support these components, the foundations required, and the gimbal that carries the concentrator and receiver. These all rotate together to follow the daily motion of the sun as well as allow all to follow the sun as it goes higher and lower in the sky through the seasons.

Computer Model of a Solar Dish with Mirror Assemblies and Receiver

A solar dish collector uses a concentrator to reflect sunlight into a receiver that transforms radiant energy into hot water or steam. The area of the concentrator may be 500 to 1,000 times larger than the opening of the receiver to minimize energy losses.  The receiver can be mounted close to the reflector, far away, or somewhere in between. Dividing the focal length, the distance from the mirrors to the receiver, by the concentrator diameter describes this (f/d) relationship. I work with an f/d ratio of 0.8, knowing that this works well with very high performance cavity receivers. Shorter focal lengths, say 0.6, require shallow cavities that lose more energy than deeper ones because hot surfaces inside lose more heat to wind and radiation. A longer focal length, say 1.0, where the receiver aperture is the same distance from the mirrors as the concentrator is wide, extends the receiver farther out from the main structure and requires more material and higher quality mirror assemblies.

Illustration Showing Various Focal Length to Diameter Ratios

Many point focus solar collectors have individual mirrors mounted on a structure to reflect images of the sun on a target. Attaching and aiming each one takes time and these individual mirror facets are difficult to clean. We made early solar collectors in the 1970s this way. We progressed to mechanically curved mirror assemblies that placed eight one-foot square mirrors end to end and were handled as a larger unit. Correct curves insure mirrors reflect sunlight to a spot. Mirrors are then continuous and easy to squeegee clean. Accurately curving aluminum extrusions is straightforward and simply attaching mirrors to them can establish a parabolic reflective surface that intensifies sunlight 30 to 60 times without having to adjust any mirrors. Attaching a few dozen of these assemblies to a structure and aligning them so they reflect sunlight to a single point creates the point focus concentrator. Permanently mounting mirrors in arrays minimizes labor when erecting a collector. An ideal: assemble and erect a solar collector in a single day. Accurately aligning mirror assemblies is easier and quicker than individually mounting nine to 16 times as many smaller mirror facets.

A specific design begins by choosing an appropriate size mirror assembly and then settling on a suitable array of these assemblies to make up the concentrator. Mirror facets come in many sizes and what I have on hand are 12”x12” and 12” x 36” mirrors that readily fit into square panels that are either three or four feet square or rectangles that are three by four foot. One person can readily handle these mirror assemblies. A 3’ x 3’ weighs 10 pounds and the larger ones weigh less than 20 pounds.

The table below shows a few ways that mirror assemblies can be arranged in a concentrator. A roughly circular arrangement gives each mirror a good view of the inside of the receiver. The farther away from the optical axis (a line through the centers of the concentrator and receiver) a mirror is, the poorer view it has of the inside of a cavity receiver, making it less effective than those closer to the center.

Mirror Assembly Configurations with Table of Mirror Area & Quantity

At this point I’m leaning toward configuration “C” that has 16 four-foot square mirror assemblies. In an hour in bright sun these would reflect 24 square meters of sunlight, 20 or more kilowatts, into a receiver that would displace burning five pounds of fuel oil and releasing over 11 pounds of carbon dioxide when I burned oil for heat and hot water. Over a year it should halve how much wood I have to gather and work into pieces for the stove: maybe by three cords, or six tons. I should be able to manufacture 16 mirror assemblies in a day or two and to fix and align them on the concentrator structure in one night so they hit a target.

It is easier to work in the dark using an LED flashlight shining at a mirror assembly, MA, from the center of a target at twice the focal length. Tighten the bolts at the four corners when it reflects the image to the target center. One could use sunlight during the day to do this but after aligning a few mirror assemblies the target becomes so bright it becomes impossible to see the reflection of a single MA. One could cover each fixed MA but this is cumbersome. Also, using the sun during the day works well around noon, when the concentrator faces up and is level, but because the structure must be tracking the sun for the process to work, climbing around the tilted moving structure in the morning or late afternoon becomes more difficult. And sunny days without clouds often don’t occur when you want them. Every night is dark, though, and working close to the ground with a stationary structure looks easier so we will be exploring this new MA mounting and alignment technique for this project.