Reducing the Cost of Concentrating Heat

One of the beauties of concentrated solar power is that nearly perfect reflective material (reflective mylar) itself is very inexpensive compared to photovoltaics. A 2'X4' photovoltaic costs hundreds of dollars, whereas the same area of reflective mylar costs $1-2. Unfortunately it is also only one part in a complicated system to convert solar heat into electricity. A piece of cheap reflective mylar by itself doesn't actually concentrate solar energy into anything useful. To make solar heat useful, we much take a large surface area of sunlight and focus is down to a small area then do something with the heat produced. We decided to stick with the traditional parabolic trough design. This design has been proven for decades and in numerous large scale power plants.

Without changing the physics of the design, we began asking what could be changed to reduce the cost of concentrating this solar heat? We started with the materials used to construct the reflector frames. Could we choose a cheaper material instead of galvanized steel or aluminum? Aluminum is much lighter than galvanized steel, which has benefits in the foundation of the system. But we wondered if we could go even cheaper? We're currently attempting to use PVC conduit pipe. It's 10% the cost of aluminum, very strong and easy to manipulate with off the shelf fittings. Our biggest concerns are dealing with weather, but we have some solutions in the works.

The second attempt at reducing the cost of the reflectors is the cost of the substrate. The substrate is the material the reflective mylar is attached to. The substrate maintains the parabolic shape and focus the heat at the focal point of the curve. The cheapest substrate being no substrate at all. We attempted roughly 10 different designs without any substrate at all. Each attempt proved unsuccessful as we could not avoid wrinkles in the reflective mylar. These wrinkles bounced the light in such a way that the sunlight was no longer focused correctly and did not produce adequate heat. We were successful with creating extremely low cost substrates. Three solutions have been effective, using monofilamant line every 1 inch along the length of the curved pipes, using shrink wrap tubing as a substrate, and using a very inexpensive, flexible fiber board. The monofilament line was an extremely inexpensive solution, but we feel it may be too complicated a process to mass produce. We're not only looking at direct costs, we're also searching for a long term solution that can be easily mass produced. The shrink wrap tubing solution is ideal, although to date we do not have a supplier than can produce "shrink tubes" wider than 24". We may reinvestigate this method if we can find a supplier that can provide wider tubes.

Our current solution we are experimenting with is a double-side plastic coated melamine board. These cost roughly $10 per sheet retail, can bend to take the parabolic shape and are stiff enough to hold the shape.

Pivoting the Reflectors, not the Steam Pipe

This reflector design pivots at the bottom of the trough, the pipe
moves with the reflector. It requires a flexible steam pipe at each end.

There seem to be two common designs to pivot parabolic troughs. The design that appears in traditional large scale plants (picture to the right >>) is to fix the steam pipe to the reflectors at the focal point and pivot the reflectors AND steam pipes together at the base of the reflector. The second design is to fix the steam pipe to the ground and pivot the reflector around the steam pipe.

The second design has a couple advantages. The first advantage is the steam pipes can be fixed to the ground, they do not need to move with the reflectors meaning one less moving part. This means there is no need for high temperature/pressure joints that must flex or pivot. Compared to a 90 degree elbow joint that cost about $1, these flexible joints are complex and expensive. The secondary benefit is how the support frames are constructed. When the reflectors pivot around the steam pipes, the horizontal support structure is also used to maintain the parabolic shape. These support structures keep the parabolic trough from spreading apart. The one disadvantage is the horizontal support structure casts a 2" shadow every 48". We feel the dramatic cost reduction with this design is worth the 4% reduction in energy collection.

Notice how the steam pipes and glass insulators are fixed in place, the reflectors pivot around a custom machined coupling.

An Ideal Parabolic Depth?

We have not found any research identifying the advantages or disadvantages of various parabolic depths for optical reasons. So we After considerable trial and error we feel we have found an ideal parabolic depth. A parabolic curve for a reflector has numerous variables (see picture below), but we have simplified these calculations down to three primary values. The "arc length" of the reflector. That's the length of the material if you took a cross section of the reflector, if we use a 4'X8' Melamine board, our linear length is 8'. The "Depth" of the reflector, that's the distance from the lowest point (the vertex) of reflector to the highest point when the reflector is pointing upright. Finally the "Focal Radius". This is the distance between the vertex and the focal point.

These numbers can be changed around which will change the dimensions of the parabola. Since we have not found any research that indicates that one type of parabola is optically superior to another, we decided to make the "focal radius" equal to the "depth". This allows our optical structure to match up with our mechanical structure. Since the size of the material we use may change over time, we created a calculator that tell us how to cut the parabola so the focal radius is the same as the depth. The reason for doing this is we can easily pivot the reflector around the focal point.

Tracking the Sun

For the first prototype we plan to track the sun using simple photoresistors which are separated by a sun shade. As the sun moves from morning to evening, one photoresistor will have more light than the other. We will send this signal to a single Netduino microcontroller. Based on the resistance values from the photoresistors, the microcontroller will make a decision about which direction and how much to turn motors to pivot the reflectors. We will need to have a secondary algorithm written into the logic of the microcontroller that will reduce the motors from jumping back and forth when there is cloud coverage and other false data inputs.

Pivoting the Reflectors

For the first version of the Zenman design, we will use a simple parabolic trough. Future versions may use a new compound parabola we're working on that won't require any tracking (or reduced tracking) of the sun, but for now we need move the reflectors to follow the sun. We're currently investigating two methods of moving the reflectors.

Perpendicular Motors with leadscrew - this method would have numerous motors mounted perpendicular to the reflectors. When the motors turn they would very slowly turn a long leadscrew which would attach to an arm on the bottom of the reflectors. As the screws turn, the reflectors would pivot along the focal point. The number and size of the motors necessary to pivot a single row of reflectors is not known at this time. Once we can identify the weight of the reflectors and how much friction is caused at the pivot points we'll have a better understanding of that

Parallel Motors with pulleys - this method may only require a single motor to pivot a row of reflectors. Each section of the reflectors has two wires attached on each side of the reflectors. The wires are attached vertically and are connected to a pulley changing the direction of force horizontally. A single motor is mount at one end of each reflector. As the motor spins it simultaneously tightens one set of wires and loosed the opposite side the same amount pulling the reflector one direction.

R-134 hydraulic pistons - If funding permits, we plan to experiment with a hydraulic piston that will use R-134 (freon) on each side of the troughs. As the sun heats up the freon, it will boil and cause a pressure differential on a piston which will pivot the reflectors until each side of the reflectors have equal pressure. The advantage of this design is there are no electronic failures.

Low Pressure Steam

Most large scale steam power plants use turbines. We plan to use a steam engine and not a steam turbine, because the price quotes we have received for turbines are astronomical compared to the price of the steam engine we hope to use. The steam engine we hope to use can operate between 20PSI and 200PSI. At this point we are unsure how much pressure we will operate the engine at as our prototype is not yet producing steam.

Type of Metal

Initially we planned on using copper pipe for the steam pipe as copper has the best thermal conductivity of any metal for the cost. After further research we discovered a solar scientist did a comparison looking at the the advantage of using copper pipe vs black iron pipe vs CPVC pipe (plastic). What was most interesting is that when the pipes are insulated from the wind the material the pipe was made from did not change how much heat was absorbed. We cannot use CPVC pipe (plastic) as it will not withstand the temperatures and pressure of steam. But we can use black iron pipe without a significant loss of solar energy. This is a common pipe used in gas lines and can be obtained at any hardware store. It is approximately 30% cheaper than copper pipes.

Black iron pipes generally have threaded ends, to connect multiple pipes together requires a female/female coupling between each pipe. These threaded connections may pose a problem for future maintenance as they may become difficult to disconnect. Whereas a copper pipe can easily be cut and a new sleeve soldered over the cut joint. When glass insulators break and have to be replaced, copper may be worth the added cost. Of course, if the black iron pipe are easy to disconnect then this may not be an issue at all. We'll know after we complete our first prototype.

The secondary advantage of black iron pipes is that we can obtain larger diameter pipes for a reasonable cost. The larger the diameter the less accurate the reflectors need to be. The last advantage of using black iron pipes is they are already coated in black. Copper pipes would require either a flat black paint or some other black coating to ensure the sunlight doesn't reflect off of it. This is an additional cost saving.

Insulating the Steam Pipe

We will construct our own custom insulating sleeves over the steam pipes. These sleeves consist of borosilicate glass, more commonly known as "pyrex" glass. This may or may not be a problem as our temperatures should not rise much above the boiling point of water. As the sunlight radiation (53% Infrared + 44% Visible light + 3% UV) passes through the glass pipe, it hits the metal steam pipe. The concentrated solar radiation heats up the water inside the metal steam pipe, without a glass sleeve over the metal pipe that hot water would just heat up the air around the pipe losing a considerable amount of the energy.

The advantage of using borosilicate glass is it has a low "coefficient of expansion". That simply means it doesn't expand much when it's heated. If the glass expands too much the o-ring seals will open allowing either our vacuum insulation to be lost or our replacement gas to escape. (We are still experimenting with how to insulate the pipe from the glass for improved performance.)

Evacuated Insulators vs low Thermally Conducting Gas

We want to keep that heat stored inside those steam pipes so the energy will cause the water to boil and pressurize. Any air between the metal steam pipe and glass sleeve will conduct heat. The steam pipe would eventually heat that small air gap, that air gap would then heat the glass and we're back to where we started without any insulator at all. This is still far better than no glass insulator since air is a fairly poor conductor of heat. An ideal insulator is to simply vacuum pump all the air from between the sleeve and the pipe. We will experiment with an automated system to continuously monitor and vacuum air if leaks do occur. But we're concerned this may become a maintenance nightmare over time if the seals prove ineffective.

Our secondary idea is to replace the natural air between the glass insulator and the metal pipe with a low conductive gas such as Nitrogen, Carbon Dioxide or even Argon. While natural air is mostly nitrogen and oxygen, it also contains a considerable amount of water vapor. This water vapor is what increases the thermal conductivity of air of a dry gas like pure nitrogen. Doing this should be relatively simple, we let the air out of one end of the glass pipes and pump the thermal gas in from the other. It will eventually push all the natural 'air' out and replace the air with the thermal gas. We then close the valves on both ends of the glass insulators and the thermal gas will be at atmospheric pressure. The advantage to doing this is that there will be no pressure differential that would cause that gas to escape. It should be maintenance free, unless a glass tube breaks. In that situation, if glass breakage occurs, it should be simple to replicate the process.

Once we can construct the first prototype, we can perform scientific tests to see how much of a difference a full vacuum is versus a low thermally conductive gas and make the decision about whether it is worth it to solve the maintenance problems of using a full vacuum or to have a maintenance free replacement gas.

Valves, Gauges and Safety

Joining Glass and Metal pipes

We are designing a custom joint that will connect two sleeves of glass together as well as hold the metal steam pipe at the center of the glass pipes. This coupling contains o-ring races to hold two o-rings on each side of the glass sleeve. Whether we use an evacuated system or a low thermally conductive gas as the insulator, we still want to keep gasses from escaping or entering at the joints where the glass ends. The metal steam pipe will slide through the center hole in the coupling. The hole is larger than the metal pipe. We want to allow the gas to move from one section of glass pipe to the other so we can replace the natural air in the system with either a low conductive gas or create an ideal vacuum using a vacuum pump connected at the end of each run. If we end up using copper pipes, the hole may be slightly larger than the copper coupling fittings which would allow us slide the couplings and glass over a long run of pipe instead of having to connect them one at a time. This may or may not make any significant difference in mass production.

Green Steam Engine

We love the design of the Green Steam Engine and hope it will work well with our solar boiler. The interesting part of this engine's design is that the pistons are placed next to each other in a circle. Using a flexible steel rod, the engine converts the linear motion of each piston into a circular motion pushing the flexible rod from one piston to the next. What is most fascinating to us is that the arrangment of pistons could allow this engine to grow to any number of pistons which appears to us that it could quickly increase the available horsepower without dramatically increasing the pricetag of the engine.

The engines are small and light, when we ultimately create a big box store rooftop concentrated solar power system these engines are light enough to sit on a roof, although we'll likely have them down at ground level inside of a mechanical shed. In either case, their size will allow us to house them without requiring a large amount of real estate.

Closed Loop System

On one end of the Green Steam Engine is a connection to the steam pipe which contains 'working steam'. This is steam that has been pressurized enough push the pistons and create mechanical energy. The Green Steam Engine has operating pressures between 20 and 200 PSI. Once the pistons have moved, the working steam loses energy and becomes 'dead steam'. Dead steam is still a gas, but no longer has enough pressure to turn the steam engine. We plan on running an underground pipe to the other end of the solar collectors and use the earth's constant 13C temperature to condense the steam back to water. We have not yet calculated the amount of piping we need to perform this state change. But it simply a matter of how much piping is necessary before the earth will force the gas into a liquid. This steam engine has a built in pump which will pressurize the liquid water to force the liquid water back into the solar collectors.

Mass Production

These engines are relatively new and are not currently mass produced. When we can raise enough funds from donors we hope to buy a manufacturing license to mass produce the steam engines selling them at our wholesale cost to cover the parts and license to those that will use them in solar steam power plants. If we can do this it will allow us to directly reduce one of the largest costs in a concentrated solar steam system.

We want to put a million of these engines in operation around the world all powered from cheap solar heat. Help us acheive this goal!

Pipes underground

information coming soon...Help us fund this research

Closed Circuit Steam System

information coming soon...Help us fund this research

Maintaining high pressure

information coming soon...Help us fund this research

1st Prototype - 10 horsepower motor

We want to use a 10 HP motor for our first prototype. This is roughly 7.3 KW of energy. Since a solar plant only operates when the sun is shining, we have a limited timeframe for how many hours a day the plant will produce energy. If we can power the motor for 5 hours per day, a 10 HP motor would produce roughly 35 KWh of electricity. The average American home uses 29 KWh of electricity per day. So if our solar steam engine can operate a 10 HP motor at it's capacity we could produce enough energy to offset a single home.

Remember, we're not ultimately trying to create a residential system. We want to see fields of these collectors and the rooftops of giant big box stores filled with solar concentrators. But this size engine gives us an ideal of what it takes to power a single home with a solar steam engine.

2nd Prototype - 100 horsepower motor

We want to increase the size of the motor for our second steam engine. We're not sure what the right 'unit' sizes are for our kits. Ultimately we're trying to get to a single megewatt. A 100 horsepower motor run for 5 hours a day should theoretically produce the same amount of electricity that 10 homes consume. A 100 HP motor costs about 3-4 times what a 10 HP motor does while producing 10 times the electricity.

3rd Prototoype - 1/2 Megawatt Power Plant

It should take roughly seven 100 horsepower motors to produce a single megawatt power plant. The reason to aim for 1/2 Megawatt instead of 1 Megawatt is for common net metering rules. In many states 500KW is the maximum capacity for businesses. Residential usage is often 25KW, which is another reason we're not aiming for residential usage. It's uncommon for businesses to have 500KW solar systems, we want to make it common so this number will increase in the future.

3-phase Induction motor

information coming soon...Help us fund this research

Induction Slip

information coming soon...Help us fund this research

Using the Electric grid to excite the field

information coming soon...Help us fund this research

Connecting the motor to the Steam Engine

information coming soon...Help us fund this research

NO EXPENSIVE INVERTER!

One of the biggest expenses with photovoltaic systems is the inverter. Since we are creating mechanical energy using the steam engine we will use that mechanical energy to spin an electric motor to create electricity. We can choose either a DC or an AC motor. Although large AC motors are dramatically cheaper than DC motors. DC motors also have the distinct DISADVANTAGE of having to convert to AC to connect to the electric grid. By using an AC motor as the generator we can bypass the need for the expensive inverter.

Synchronizing with the Grid

information coming soon...Help us fund this research

Coldframe Greenhouse

One of the ways we'll use low-cost, light-weight materials is we'll place our reflectors inside of a clear plastic greenhouse. The greenhouse will absorb any rain, wind, snow and ice keeping the weather off of our solar equipment. We plan on using a very simple greenhouse known as a 'coldframe'. A coldframe consists of a half-circle curved frame with a clear plastic roof. The plastic will allow almost all solar energy to pass through it, hopefully, not significantly affecting the performance of the solar collectors.

UV resistant Plastic Roof

We may opt for UV resistant plastic film. UV radiation only accounts for 3% of the energy in sunlight. Sunlight is made up of 53% IR, 44% Visible and 3% UV. Some films are able to stop UV from passing through. This will become an interesting experiment as UV damages materials and will reduce the lifespan of the equipment. It's feasible that if we can filter out UV with the greenhouse film, we can use cheaper PVC pipes and fittings (for the reflector frames) that do not have UV resistant addatives in them. Once we have the ability and funding to perform these tests we can find out the answer to this.