Multifacet Parabolic Solar Concentrator




About: I like to build stuff and write code. I'm pretty tired of code though, and want to do more building.

I live in Phoenix, Arizona. Phoenix is among the areas in the US that gets the most insolation (incident sunlight) each year. This is partly due to the latitude and party due to the dry climate and therefore low humidity and low rainfall. These led me to think about building a solar water boiler. I've always been interested in old style steam piston engines and thought it would be fun to build a solar concentrator to boil water and run an engine. This project gets as far as building the solar concentrator. The engine and other aspects of steam piston engines operation are yet to come.

For now, let's content ourselves with just a parabolic solar concentrator.

Before we begin, it is important to know that concentrated solar energy is very dangerous. Reasonable precautions should be taken. I cannot take responsibility for individual choices you make in the design, construction, modification, or use of this project. Have fun and be safe.

Step 1: Obtain Parts

You will need several parts for this project. The cost ranges from $300-$500, so this is not a poor man's hobby. What you will need is:

1) About 17 aluminum angle stock rods 5/8 inches (or thereabouts). An aluminum siding store will probably have this much cheaper than your corner hardware store. They're friendly folks, usually.
3) Some chrome-plated aluminum sheeting. Again, an aluminum siding store will usually provide this. You may have to special-order it. I ordered it under the name "light sheet". I am not sure if it was a brand name or a description. You will need approximately 40 square feet of it and ask them to cut it into 6 inch square mirrors. They will try to talk you out of it and think you're crazy for asking them to. They will also charge you for the cutting service. Let them. They deserve to make an honest living, but ask them to be careful in cutting it into _exactly_ 6 inch squares.
4) About 300 machine screws, 500 machine nuts and about 500 large washers. I used 3 inch screws in a 10-32 pitch. 10-24 is sometimes cheaper. It doesn't really matter, but be consistent. 10-32 and 10-24 look just alike and with so many of them, you won't want to waste time sorting them.
5) A screwdriver, drill, wrench, and ruler that will measure inches and millimeters.
6) Various other common household tools.
7) Some creativity and basic high-school level geometry.
For the heat-exchanger (if you choose to build it)
8) A length of 1/2 inch copper pipe and about 20 "T" junctions, some end caps, solder, propane torch, gloves, a little soldering know-how.

Step 2: Measure, Cut, Measure Cut

Cut all but 4 of the aluminum angle stock to a uniform 6 feet 6 inches in length. Measure fairliy carefully, but an exact length is not too important.

Next, measure accurately an interval of 6 and 3/8 inches. This is VERY important. I took a small piece of medium density fiber board and drilled two holes VERY CAREFULLY to be exactly that distance apart. Then, I took a hole punch and a hammer and pounded small dimples into the soft aluminum at this interval all across the length of the angle stock. Do this for each of 13 of the angle stock pieces. This means more than 160 dimples exactly the same distance apart.

The reason they must be so carefully measured is that the mirrors are all 6 inches square. The machine screws are 3/8 inches wide. This means you will have space for one mirror, one bolt, one mirror, one bolt, etc.

Step 3: Drill, Drill, Drill

Now that all of the measurements are made, measure some more and then drill some holes.

Measure halfway in between the 6 3/8 inch intervals on the first, last, middle, and 1/4, 3/4 up each bar.
These will be the holes for the cross-beams.

Now, each of the dimples that was made will be drilled out with a 3/8 inch drill bit. This is a lot of drilling and I highly recommend an electric drill and some understanding neighbors. It will be loud because the aluminum will ring as you drill. That's ok. Also, because you're using power tools, don't forget to keep your fingers clear and your eyes protected. I got a piece of aluminum in my eye once. I do not consider eye injury to be an important step for completing the project, so wear eye protection.

Once drilled, you will want to file the holes flush because aluminum is soft and as such creates lots of sharp edges when you drill it out. An ordinary file will help you remove those edges so you don't slice your fingers.

Step 4: Assemble the Frame

Now, you will want to assemble the frame. Do this by taking some short 10-32 machine screws and nuts and simply attaching the 13 angle stock pieces to the 5 cross-beams. You should wind up with 13 parallel angle-stock pieces in the vertical and 5 in the horizontal. You may notice that it's difficult to keep them at a 90 degree angle from one another. There is a simple solution.

Take another couple of scraps of aluminum and do a pythagorean theorem problem (you know, x2+y2=d2) to figure out how long the triangle brace should be. A 3-4-5 right triangle is a pretty good one to apply. In the picture, you will see 2 of these triangle pieces.

Now that the assembly is fairly stable, start inserting the 3 inch machine screws. You may want to use 4 inch screws around the outer edges. You'll see why later. Hold them in place using a machine nut. This is perhaps the most time consuming part of the project.

Step 5: Mirror Alignment.

You will be holding the mirrors in place by washers sandwiched in between two nuts. This is tedious. You will begin by putting together a spreadsheet or other calculation to tell you how many mm each mirror corner should be offset from its neighbors. The basic formula for this is the formula for a parabola. y=x2. This is a parabola in 2 dimensions, however, so it's a little more complicated than that. Really, the formula is z=1/(4*f)d2 where d=sqrt(x2+y2) and f is the focal distance of the device. I chose a 72 inch focus. Don't do anything different unless you understand why you're doing it. Let me recommend that you study the math and don't try to align the mirrors by hand. It's even more difficult than measuring it and the focus isn't as good.

What happens is that when you get past the 3rd row out from the center is that your distance up each screw starts to get bigger than the screw is long. This is not really a problem. The trick is to offset all sides of a given mirror so that the lowest edge is as low as you can make it. What you get is rows of mirrors at the same angle, but not in the same plane. This is similar to a fresnel lens for those who understand optics.

If all of this is lost on you, I have attached a screenshot of the spreadsheet. This is a list of the mirror offsets for each of the mirrors as measured in millimeters. This will give you a focal distance of 72 inches. It works fairly well.

Step 6: Focal Assembly

The focal assembly is simply four angle stock pieces set up so that they are screwed to the corners of the concentrator and meet at the center. If you use 8 foot pieces, they will meet at a point about 7 feet from the center of the mirror assembly. This gives you about 1 foot between where they meet and the focal point so you can mount things easily at just about the exact focal point.

NOTE! When you're not around the concentrator make VERY SURE that it is pointed in a direction that the sun will NEVER be. It WILL catch things on fire. Think very carefully about how the sun moves and remember that it traces out an arc in the southern hemisphere (northern if you're an Aussie).

The broom below took only 30 seconds to burst into flame. I cannot stress enough that concentrated sunlight is VERY dangerous. Also much fun if you're careful.

Step 7: Completed Concentrator

In order to come closer to the goal of solar steam generation, it makes sense to build a heat-exchanger module. Next, we'll tackle that problem. It turns out to be a little simpler than the concentrator. The completed concentrator is shown here atop a makeshift box to hold it in place. It was originally intended to be a tracking frame, but it proved impractical.

Step 8: Heat Exahgner.

For the heat-exchanger, take a length of 1/2 inch copper pipe (about 8 or 10 feet). Cut it into lengths of 12 inches and use copper "T" junctions. Sweat (solder) them together using high temperature solder and a propane torch. If you've never done it before, read about how to sweat copper. Of utmost importance is to dry fit everything before you sweat your first joint. Also, keep everything very clean. Very Very clean. Use lots of flux. Otherwise, you'll have lots of leaks. Leaks are no fun.

The structure will look much like this. Note that there are gaps between the rows. You can take yet more copper and just solder them in between as shown on the right-most end of this example.

The gap filling ones don't need to be watertight. They're only there so that the light hits them and conducts the heat instead of passing through the gaps.

Step 9: Measure Performance

The whole project is now complete. You can now use the device for anything you please. You can measure the performance of the device by putting water into the heat-exchanger, mounting it in place, and measuring the temperature rise. How fast the temperature rises is an indication of the amount of energy being transferred per unit time. You can alternatively measure the amount of time it takes to boil a given amount of water or measure how much water boils per hour.

I finally abandoned this design and disassembled it mainly because I was not able to track the sun accurately enough for it to be practical. It sure was a lot of fun though.



    • Pets Challenge

      Pets Challenge
    • 1 Hour Challenge

      1 Hour Challenge
    • Sensors Contest

      Sensors Contest

    83 Discussions


    6 years ago on Introduction

    I love the mechanism for holding the corners of the mirrors. Sheer genius.

    I was thinking of making something similar, but using the long full length mirrors that you can get for mounting on doors. I have seen them for around $8 for a 5' long one. The advantage of using the long mirrors would be that fewer adjustment points would be required, and that the curvature would let you do a tighter focus, at least in one dimension.

    2 replies

    Reply 4 years ago

    Using ordinary mirrors has the problem of excess weight. Aluminium sheets are light and cheap (you can even use the inner surface of coke cans) which is best for the tracking system. Using long reflectors is suitable for cylinder-parabolic concentrator.


    1 year ago

    I sign up to this website just to comment on this. I have been investigating csp for over 1 1/2 years and the use of bolts and washers is by far the smartest thing I have seen in this field cannot wait to use this on my mylar plywood project. once again thank you sooooo much!!!!!


    2 years ago

    thank you so much, I love this project


    4 years ago

    I think it is too big for a sun tracking system. There's no need for such a complex structure and focus doesn't need to be that far. Just adjust the reflectors to converge closer.


    10 years ago on Step 9

    I really like this solar boiler project. I especially like your method of adjusting the mirrors to focus them. I've wanted to build one of these for a long time now. Unfortunately I live in an apartment and don't have anywhere to build it. I was wondering what your cost was for the aluminum angle pieces. Although it might make it slightly more difficult in drilling the holes, I was wondering if steel conduit might work better for providing strength and be cheaper as well (as little as $1.96 per 10ft). Also painting the heat exchanger flat black would help it's efficiency.

    4 replies

    Reply 10 years ago on Step 9

    I'm most gratified that you liked the project. Thank you. It was a most interesting and rewarding project. I don't have much room myself. That's why I built it so that it can be disassembled and stored when not in use. My wife is particularly happy that it's not taking up permanent room in the back yard. I don't recall exactly how much the aluminum angle braces cost. The most significant cost was the chrome plated mirror pieces. All in all, I believe I recall that I spent some $300 on the project. More accurately than that, I simply don't recall and didn't keep very good records. As far as painting the heat exchanger, I recall doing some experiments with that. It is a common misconception that simply painting something like this black will make it more efficient. It does indeed help for low temperature devices but works not at all for higher temperatures. It turns out that ordinary paint from the hardware store will burn readily under the power of a solar concentrator. Basically, it just burns up in the first 10 minutes or so leaving the copper bare again anyway. It seems that a better way to improve efficiency is to add an insulator to the back side. Something like fiberglass works well. Of course, the paper backing of the fiberglass should be removed to avoid fire. What you're aiming for is not allowing the excess heat to radiate from the back side.


    Reply 9 years ago on Introduction

    One of the best selective absorbers is oxidized nickel.  Since stainless steel contains a good bit of nickel, it turns out that it does a fine job itself.  You simply make the absorber out of stainless steel, heat it until it turns a dark purple, and you are done: a selective absorber that has an absorption/emission ratio of about 11 - excellent!  And, of course, it does not blister or peel.  Bare copper has an absorption/emission ratio of 3.  There are oxidizing chemical treatments that will turn copper black - and they are not paints.  They do not have good abrasion resistance, but that is not an issue here, yes?

    If the absorber surface temperature stays low - like at or below the boiling point of water - it does not need to be a selective absorber as boiling hot objects do not radiate much infrared energy.  On the other hand, selective absorbers do not absorb infrared energy as well as non-selective absorbers do.  Since sunlight at the surface of the earth has about 40%-45% of the thermal energy in the infrared range, you typically want a normal absorber like carbon black.

    If you are dealing with temperatures close to the boiling point of water you have many choices for paint, but if you need really high temperature paint, try Pyromark High Temperature Paint. It is available in 800, 1200 and 2000 degree F ratings.

    One of the inefficiencies of heat transfer is the temperature drop across the thickness of the metal barrier; another is mixing the fluid enough to bring all of it in contact with the hot media.  I feel that the best, most efficient way to solar heat water, especially with concentrator mirrors, is to place the absorber in the water.  That is, pump the water in transparent tubes into the absorber surface at the focal point.  One excellent media is an open-cell black ceramic sponge (looks like an aquarium aerator stone); another is to put a very concentrated (stable) black dye in the water.  (maybe use both techniques?)


    Reply 7 years ago on Introduction

    Great point about placing the absorber in the water. That is definitely the way to go.

    I would add to the absorber discussion that the emissivity values are for thermal radiation and emissivity varies with both frequency and temperature. For solar power from most reflective concentrators, the majority of radiation is visible, not thermal. So glass is a great transmitter of visible radiation but very poor for thermal radiation. A glass mirror that looks shiny bright in the visible range looks almost completely black when viewed (with a thermal imager) in the thermal range. For concentrators that use glass mirror reflectors, nearly 99% of the reflected power is in the visible frequency range because the mirrors absorb most of the thermal radiation. However, this design uses metal reflectors and these are efficient in both visible and thermal bands. Ideally, the translucent material should transmit both visible and thermal frequencies. In addition, the receiving face of the translucent material should be flat and normal to the incoming beam. One possible solution would be to use a multichannel polycarbonate panel used for greenhouse walls. The polycarbonate has excellent visible and thermal transmission. Its important to get the window-grade polycarbonate.


    Reply 10 years ago on Introduction

    The material they use in big sized solar concentrators is carbon. although will probably not be able to get your hands on big carbon tubes etc, you could try coating the copper in carbon by putting the copper tubes under and oxygen deficient flame. I have seen that acetylene produces a lot of black deposit on things when no oxygen is mixed with it, maybe you could just use a candle to coat it? The only thing i suspect may happen with this is that once the carbon gets hot enough, it will combine with the oxygen in the surrounding air and simply float of in the form of co2. still, worth a try?


    11 years ago on Step 6

    Another focus method :Cover all facets with pieces of paper. Suspend a fixed target at the intended focus point. Uncover one facet at a time and adjust until reflected light is cast onto target. when finished, turn assembly away from sun and remove all the paper sheets. No math, no spreadsheet and very accurate. Not my idea (found it on the web) , but simple and effective.

    8 replies

    Reply 10 years ago on Step 6

    I think that you would have to be very quick to set each panel and maybe reference to the first panels focus (by moving your hole frame) every five or so minutes to allow for the suns movement in the sky @ 15 degrees per hour. If you were taking 3 hours to set up, the last panel would have a very different focal point.


    Reply 7 years ago on Introduction

    1 Select one panel as a reference panel (ideally one centered on the desired focus point with the selected light source.) Use a large flat target to see the beam cast by the reference panel.
    2. Cover up all the reflectors EXCEPT the reference panel and the adjustment panel.
    3. Adjust the screws on the adjustment panel so that the light of its beam center coincides with the beam center cast by the reference panel.
    4. Once both beams are aligned, cover the panel adjusted and move on to the next panel, removing its cover.
    5. Repeat steps 2 to 4 until all panels are aligned.
    Reasoning: since the light of the reference panel moves as the sun tracks across the sky, adjusting all the other panels to it create a focus that is referenced to both reference panel focus and sun angle. Since the reference panel points to target, the entire frame can be moved to track sun. All other panels will always point to reference panel beam focus. Use of reference panel allows alignment procedure to no longer require an adjustment for sun movement.


    Reply 10 years ago on Step 6

    The focal point is set by the parabola formula. It does not have to be pointed at the sun for an accurate focus to be achieved. The system was designed to be mounted on a tracking device that moves the entire assembly to track the sun. I agree that if I was aligning the focus by moving the mirrors and aligning them to the sun, that would be highly impractical. This is why I decided to calculate the height of each corner based on a parabola and aligned it in my workshop in the shade. Only when it was complete, did I take it into the sunlight to test the quality of the focus and make minor adjustments.

    eston machariajarney1

    Reply 9 years ago on Introduction

    hey guys am eston,currently a final year engineering student in jomo kenyatta university in kenya.I wish to develop a project on a solar powered steam soil sterilizer to assist in control of micro-organisms in seedbeds.However am not sure whether to use parabolic dish concentrators or trough concentrators.i wanted to try the latter but am not sure how to go about it and what materials 4 the troughs would be best suitable.any ideas pls? your assistance will be greatly appreciated.u can reach me on

    Mechanic2011eston macharia

    Reply 7 years ago on Step 6

    I did something by accident one time that you could probably use .
    I laid a sheet of black rubber roofing membrane that was about 1.15mm or .045inches thick on a bright sunny day .
    It got soooo hot beneath the rubber sheet that it literally killed everything that was growing in the soil at that time even the weeds and thistles.
    I had to reseed my grass lawn to get it green again .
    Depending on how long it is left on the soil and how hot it got ,it might make a very worthwhile weed control system.
    I should do another experiment like this again and this time install some temperature probes at different depths in the soil beneath the black rubber sheet like right on the surface and then every 1mm or so and record the results.
    I believe to be effective ,the heat needs to get down to about 3 inches or 7-8mm in depth to kill all spores ,roots ,nodes etc.
    Again you would have to rely on the sun for the heat but it would be very clean as there would not be any chemicals involved .
    The rubber sheet could be moved in the mornings or evenings to a new location once it cools enough to hold onto .


    Reply 7 years ago on Step 6

    That is a pretty good idea. I have weeds etc. but I'm just not into poisoning my property. Just seems wrong. This sounds like a good alternative.

    dwrenneeston macharia

    Reply 8 years ago on Introduction

    Nice idea to use solar concentrators for soil sterilization. Did you proceed with the project. That said, with the solar potential in Kenya, Just using plastic for a month or two on a seed bed would probably be better. Building a solar system to create steam to use on a seed bed only every so often would be expensive.


    Reply 10 years ago on Step 6

    Sorry Jarney1 I was directing my thoughts to Burnerjack01 and his method, I forgot to address it properly. I fully agree with the 'measure first' then if thats not quite good enough correct it practically way.