The single biggest hurdle in creating a solar concentrator capable of powering a house is the steam engine. It has to run all day every day reliably for many years and it has to be powerful enough to drive a decent size generator. That means it has to be very expensive, and it won't end up being as reliable year after year as you would hope. The power plants use turbines, which are insanely more difficult and expensive than any steam engine. The best thing out there right now is the free piston sterling engine with a high power linear alternator, which you can't build and which aren't for sale on the individual level.

If you want to see how easy it is to put together a solar concentration array just head over to SolarFire.org for a peek at some of their designs. These guys can turn water into steam like nobodys business, which is really awesome if you need steam. I need electricity.

The idea I am going to explore this summer is to force high pressure steam into the top of a 55 gallon barrel filled with water and suspended from a seesaw like arm. The water will be forced out the hose in the bottom of the barrel and then into the empty barrel which is suspended in the air at the upper end of the seesaw. When the initially empty barrel fills, it will weigh 459 pounds (208 kilos) and it wants to come DOWN. Notice in the diagram below that the motion of the seesaw drives a rotary wheel. I will be trying to get high enough water flow to cycle of water from one barrel to the other and back again in one minute. That's 6 feet per second through a 2 inch pipe - it's just a question of how much pressure will cause that much flow. That gives our rotary wheel a rotational speed of 1 RPM but with absolutely massive torque. If the seesaw has a radius of 6 feet the torque will be 6 x 459 = 2754 foot pounds. We need 1800:1 gearing to get that up to generator speed, which results in 1.5 foot pounds at 1800 RPM. There are other ways to arrange this whole thing, such as hanging barrels by a chain and sprocket gear, which can result in an initial main bar rotation of 20 RPM and only required 90:1 gearing, but after gearing the power works out exactly the same as this less complex seesaw design. Getting power out of both halves of the cycle is more troublesome with the hanging version, but the variable height does remove a few of the design constraints.

The reversing valve is trivially similar to steam engine slide valves or a number of rotating ball valves ganged together. In one position steam flows into the left barrel and the right barrel vents back to the condenser. In the other position the steam flows to the right and the left barrel vents to the condenser. I anticipate the valve operating by being mounted to the cross beam and using the relative motion of the stand to actuate the levers.

Part of the remaining complexity is that the barrels actually hang free and I need flexible connectors for the pipe. If I mount the barrels fixed to the cross beam I can get away with solid steel 2" pipe for faster water flow and lower maintenance.

Note that steam will condense in the barrels, they will slowly fill with water and become inoperable if there is no mechanism to continually keep the amount of water in the barrels correct. That mechanism is to time the system such that the barrel being emptied is blown fully clear of liquid, sending some small amount of liquid back through the vent return line.

Since steam has 1600 times the volue of water, it will take just a half a cup of water turned to steam to fill a barrel at 1 atmosphere of pressure

I'll keep you updated. I will want to be able to run a generator off of this before I invest in making the solar concentrator, so this is the priority part of this project

Unfortunatly when we run the numbers we find the process doesn't deliver on power, the result is less than 1 HP. The equation is HP = Torque x RPM ÷ 5252, so we know the HP of the original system with 10 foot arms and 459 lbs weight = 4590 foot pounds torque x 1 RPM ÷ 5252 = 0.87 HP, no amount of gearing is going to change that fundamental starting point.

One can see the HP equation is a direct function of the RPM. I wonder how fast we could get this thing to cycle. I also wonder if instead of a seesaw we put the whole thing into a big spin if we might increase the rotational speed. If I want 10HP out of this I have to get the RPM up to 11.5 (4590 ft lb x 11.5 RPM ÷ 5252 = 10HP). To achieve this, we must cycle 110 gallons x 11.5 times in a minute = 1265 GPM. The estimates from FlexPVC.com suggest a pipe in the 8" range depending on PSI and number of elbows. Also note that this amount of steam at 100 PSI is going to require more than 2 gallons of water vaporised per minute.

The problem here is that I tried to design this with maximum torque, which is presenting serious limitations on water transfer speed and RPM. Instead lets look at a design with a ferris wheel configuration of 10 food radius (20 ft diameter) cycling 5 gallons of water back and forth at a speed that generates 10 HP. HP = Torque x RPM ÷ 5252 = 10HP. Torque (max) = 417 ft lbs. That gives us a required RPM of 126, or 2 rotations per second. That's 10 gallons cycled in 0.47 seconds, which by an interesting conservation of energy comes out to the same volume as using 55 gallon drums. It would appear if I want 10HP that the amount of water to move per second is realtivly fixed regardless of other factors.

Started out as an interesting idea, turned out to be a bust in it's current incarnation, but maybe this is the spark for the next idea in weight displacement steam power.