When we first looked at the land that was to become Carraig Dulra, Gary Crocker showed us a natural spring, located down the hill, and suggested we use a "Ram pump" to bring the water up to us. This was the first time I had ever heard of this type of pump, which doesn't require an external power source to push water uphill. When we bought the land, we also got rights to use this spring for our water, and my research began.
There are good explanations & instructions on how to make RAM Pumps on the internet. There is also a good amount of older scientific research available. For those not inclined to plow through all of it, the basic idea is something like this:
Imagine you have a garden hose with water running out. If you hold the hose pointing upwards, the water will of course flow up into the air a short distance and then fall down to the ground. If you cover the end of the hose with your thumb for a moment and then let go, the sudden spurt of water from the build-up of pressure will cause the water to flow up somewhat higher for a moment before settling back into the previous flow.
A RAM pump creates a similar "water spurt" about once a second, and uses each spurt to drive a small amount of water into a hose or pipe that goes uphill. Over time, this water can eventually go a long distance uphill. An an example, our pump pushes about 1/2 litre of water a minute through 1 km of hose, to a point that is about 25 meters higher (measuring vertically) than where the pump is.
How we did it
I was all for building this myself, but having tried to create a special valve for our friend Phillip's own RAM pump, I decided to go with a kit the first time through. I chose a PVC kit from The RAM Company in the USA. The other alternatives included a cast iron version from the original patent holders, but the cost of this was many times more than the kit. I bought the 1 inch RAM with the additional valves and guages, total cost including shipping was about US $350 plus I had to pay additional customs duty of about 40 euro.
A RAM pump requires a flow of water, it can't pump water from a standing pool or well, for example. This is normally achieved by collecting water from a stream or spring and letting it flow into the pump located downhill from the collection point. Our spring has a large concrete collection box just below it, which was used to feed nearby houses and to water cattle. It isn't used any more although it is still plumbed as a "backup supply" for (I believe) a local country house. The slope below this box isn't very steep but it does go far enough down the hill to get sufficient "fall" into the pump.
In fact, Paul (an Irish volunteer) and I spent the better part of a day measuring this fall in the swampy overgrown pump area, and then from there up the hill to the intended destination for the water.
Going against existing research and the recommendations from The RAM Company, I drilled through the 10 inch concrete collection box towards the bottom, and ran a hydro-dare (plastic) pipe down the hill to the pump. I did this beacuse I already had the pipe and thought it worth trying before investing in metal pipes and because the local plumbers suplly people were sure it would be fine. All research says you need a metal or very rigid drive pipe, and I can now confirm this is true. The pump wouldn't run for more than a short while with this arrangement. The problem is that the pressure pulses get absorbed by the flexing of the pipe/
So the next investment was in copper pipe (copper because the water was for drinking). Paying attention to the research now, the distance and slope dictated a standpipe between the pump and supply, so the plastic pipe runs from the supply to a T junction with the standpipe, and from there downhill the pipe is copper. The slight slope required a large amount of copper pipe, which ended up being as expensive as the pump! But it worked!
This arrangement worked very well, although the standpipe, which is only supported by ropes and tree branches, fell over during high winds a few times, causing water to pump out of it and the pump to stop. Also the pump shifted around and flopped over sometimes, again stopping the flow. So I anchored the pump onto some paving slabs with wires, and re-tied the standpipe.
This worked very well for a long time when the spring's flow was light, but as the rains increased, the height of the water in the collection tank rose, created a larger fall from there to the pump. I assume this was the cause of the crack in the pump's gate valve cover. Water under high pressure was spurting out the side of the cover. Removing the cover it looked like I had screwed the cap on too tightly, but after a replacement was secured from the USA, it also cracked within a short time, in exactly the same way. Looking back at it, I realised that we had measured the vertical fall to the pump assuming a low water level, and had only allowed a foot or two up to the pump's stated limit, for when the water levels might rise. So my assumption (now) is that the increased pressure caused these cracks.
Some possible solutions include raising the pump during heavier rains, changing the pump for a more robust one, or repairing the crack and seeing if it holds together. I'm going for the last option first: I've just fiberglassed the crack so will try out the repaired cap next week. My next preferred option would be to replace some parts of the pump with metal equivalents.