Although water won't run uphill, some exceedingly clever soul discovered a long time ago that H2O can be persuaded to pump itself in that general direction. The hydraulic ram pump makes it possible.
A hydraulic ram pump looks something like this.
ILLUSTRATION: MOTHER EARTH NEWS STAFF
Are you planning to add a farm pond to your homestead, but bothered by the fact that the site you've selected is not naturally furnished with water year round? Or do you already have a pool which suffers from one or more of the maladies connected with a lack of sufficient incoming fresh water? Well—if your property contains a spring, creek, small stream, or other source with a flow of at least three gallons per minute (gpm) you can probably solve your problem easily and inexpensively with a hydraulic ram pump.
As I write this, it's been three months since we installed our ram pump in a nearby creek ... at a total cost of under $200. All that time the device has been pumping clear, cool spring water up over a 25-foot hill—a distance of 150 feet—and into our farm pond, without the use of any fuel whatsoever. In short, we're getting about 500 gallons of water per day at an operating cost of zero ... and we expect this to continue for ten years or more.
Prior to this installation, the water level of our 15-foot-deep, half-acre fishing and swimming hole dropped at least two feet each July and August, the pond's temperature rose to the tepid bathtub stage and the algae blossomed. Not so these days! The level remains constant, the water is clearer, cooler and more invigorating, and the plant population has been drastically reduced. Even the largemouth bass and bluegills (now five years old) put up a stiffer fight when we hook them.
Although our own pump is used solely to replenish that pond, the ram is a versatile machine with many other possible applications. For instance, it can drive water to a storage tank in or near a house . . . with the overflow first diverted to a barn or watering trough for animals and then finally to a pool. Or the device can be used purely for irrigation.
Before this country's rural electrification program such pumps were in wide use, since they employ only water power for activation. In Japan, in fact, the ram still commonly serves to bring water from the mountains into villages a mile or more away.
These days, a revival of the hydraulic ram seems to be underway among back-to-the-landers ... and you may be thinking about putting the water current-driven pump to work on your own spread. Whether or not you can do this successfully depends on five important and interdependent conditions:
 The amount of water (gpm) available from the source.
 The length of the main pipe from the source to the pump.
 The drop in feet from the source to the pump.
 The height in feet that water must be lifted.
 The distance in feet that water must be delivered.
Here's how we measured the first of these variables on our own place: We temporarily dammed up a creek a few feet downstream from its source (a spring) and inserted a pipe near the top of the dam. The water from the conduit was allowed to run into a gallon bucket while we timed the flow with a sweep second hand on a watch. During the dry month of August, the figure we obtained was three-and-a-half gpm.
Our next step was to construct a more permanent dam in order to form a small pool of at least 50 gallons. This reservoir would constantly be filled from the brook, and out of it would run our main pipe to the pump downstream. We began this project by building plywood forms across the creek in the shape of an elongated letter "U", with a base width of five feet and with two-foot arms extending back upstream. The walls were to be four inches thick and only two feet high.
After the forms were laid—but before the concrete was mixed and poured we temporarily diverted the stream around one side of the plywood mold to permit the box's contents to dry without being diluted or washed away. We also inserted a one-foot length of one-inch pipe, threaded on both ends, through the walls of the form about six inches from its top and inclined downstream. The upstream opening of this tube would later be covered with a strainer, and the lower outlet would connect with the long steel pipe leading to the ram pump.
The position for the ram itself was selected with the knowledge that we wanted at least a five-foot "head" or "fall" of water ... that is, a five-foot vertical distance from the pool's surface to the base of the pump. With the aid of a transit on a five-foot stand (a simple carpenter's level does just as well) we found that a distance downstream of 42 feet gave us a level sight with the top of the newly poured dam. At that spot beside the creek we built a simple concrete slab (two feet square by eight inches thick) upon which we would later bolt the ram.
Next, two sections of one-inch black pipe, each a standard length of 21 feet, were joined together with nipples and then connected to the one-foot piece of pipe which pierced the dam. We used more nipples and a simple one-inch union to hook the water line to the pump, and then attached the strainer (supplied by the ram's manufacturer) to the intake in the pool area. We completed our pipe-laying by running 150 feet of three-quarter-inch, flexible plastic tubing (the high-pressure type for durability) from the pump outlet up over a 25-foot hill to our pond.
Finally, we allowed the pool in the creek to fill by simply blocking up the temporary, diversionary stream around the side of the forms. Within ten minutes, and after a minor adjustment to the ram's release valve, the water rushing down the main pipe activated the device and commenced its "perpetual motion". Sure enough, a steady stream of water emerged from the plastic tubing into our pond. At first the flow looked disappointingly small ... only after careful measurement did we suddenly realize that our swimming hole would be receiving about 500 gallons of fresh spring water each day. That amounts to about 15,000 gallons per month or 180,000 gallons per year!
As I've already suggested, the same system could just as easily be used to solve more sophisticated problems pertaining to the delivery of drinking water or to the supply of storage tanks, irrigation networks, dairy barn needs, or livestock troughs. The higher and further water must be pumped, of course, the larger the ram required. All the same, we feel that the hydraulic ram offers a homesteader the most pollution-free and least expensive method of getting water from Point X to Point Y next to hauling it by hand in buckets ... and it's a helluva lot easier. Happy swimming and fishing!
Don Marier (Reprinted from Alternative Sources of Energy, July 1971.)
Here's how the hydraulic ram pump works: Water rushes down the drive pipe and escapes out the waste valve until enough pressure is built up to close that outlet. (The amount of this pressure increases with the "fall" or vertical distance from the source to the ram.)
The shutting of the waste valve forces water through the check valve and into the air chamber. The rushing liquid compresses the air enclosed in the compartment so that it rushes back like a piston. This action closes the check valve And forces water up the delivery pipe to a storage tank.
When the check valve closes, the water in the drive pipe rebounds for a moment and creates a partial vacuum that allows the waste valve to drop open again. The excess fluid which was not pushed up the delivery pipe thus flows out of the opening. At the same time, the vacuum draws a small amount of air into the ram through the air valve or "snifter" just below the air chamber. This gas—which is needed to replace the enclosed air because some is mixed with the water during each cycle—will be forced into the compartment when the incoming stream starts flowing down the drive pipe again. A small amount of water is lost through the air valve during each stroke of the pump, but the leakage is minute and serves to keep the opening clean.
The cycle just described is repeated about 25 to 100 times per minute ... the exact rate depends on how much tension is put on the waste valve spring by adjustment of the screws. The slower the ram works, the more water it will pump. The ideal setting is for the minimum number of strokes per minute at which the pump will still operate. I had to rework the waste valve spring on my ram a couple of times until it lined up properly and had the correct tension; otherwise, the pressure failed to build properly and the machine wouldn't work.
How much water the ram will pump can be calculated from the following formula:
D = [(S X F)/L] X 2/3
D is the amount of water delivered in gallons per minute (gpm). S is the amount of water supplied to the machine in gallons per minute. F is the fall or vertical distance in height between the supply of water and the ram. L is the lift or vertical distance the water is lifted from the pump to the storage tank. The fraction of 2/3 represents the efficiency of the
ram. Older models had efficiencies of about 40%.
The minimum fall from which a ram will operate is 18 inches, and that's the vertical distance I had to work with. I measured the supply flow at 10 gallons per minute by catching the water in a pail and timing how long the container took to fill. The lift I used was 10 feet. Thus the amount of water I should have expected to be delivered was:
D = [(10 gpm X 1 1/2 feet)/10 feet] X 2/3 = 1 gpm
I actually measured about nine-tenths of a gallon per minute.
It sounds inefficient to use 10 gallons of water to pump one gallon, but remember that the ram works constantly (unlike a windmill) ... so that one gpm adds up to 1,440 gallons per day. Besides, the nine gallons which go out the waste valve aren't really wasted since they can be returned to the stream or used for any convenient purpose.
Note that you can't pump the water to an indefinite height since pipe friction slows the flow down. This effect is reduced by using a sufficiently large water line and by keeping connections and bends to a minimum. It is much better to shape a long piece of pipe into a gradual curve than to use sections of tubing connected at a sharp angle. Garden hose is out of the question because all the kinks would produce too much friction.
According to Don Marier (Alternative Sources of Energy, No. 1, July 1971), the only American commercial source for the ram pump is Rife Hydraulic Engine Manufacturing Co (In 1973 - MOTHER EARTH NEWS). When the Heberts were planning their installation, they corresponded both with that firm and with a Japanese supplier: Ce Co Co Chuo Boeki Goshi Kaisha.
Each company was most cordial, informative and helpful, Bill Hebert says, and his order finally went to Rife for the following reasons:  The pump was slightly cheaper;  shipping costs to Ohio were considerably less than from Japan;  instruction and installation manuals were in clearer English; and  the replacement parts which may be needed ten years from now might be easier to obtain domestically.