Jess Stryker describes it well. This works well, but leaves a minor engineering problem. It is very hard to get these valves to close slowly! Engineers have made some great progress but they still haven't fully defeated what I call the 80/20 problem. The 80/20 problem is that valves close slowly until they are about 80% closed, then they tend to snap fully closed in a millisecond! This causes the water in the pipes to suddenly stop moving. Now we all know the story of Jack and Jill, and the reason Jack fell down is that water is heavy (and perhaps Jack was paying too much attention to Jill , and not enough to his bucket, but I'm getting off-track here.) A column of water moving through the pipe at 7 feet per second carries with it a tremendous amount of weight and momentum. While it's not a perfect example, the one commonly used is to think of the water in the pipe as a big freight train going through a long tunnel. The valve closing is like blocking the end of the tunnel with thousands of tons of rock. When the train slams into the blocked end of the tunnel there is going to be one horrific crash! The faster that train is moving, the worse the crash will be. Thus the problem with water velocity in the pipe. The faster the water is moving, the worse the crash is going to be when the valve closes. That crash is the cause of a big thumping noise when the valve closes. Secondary thumps that follow are essentially "echoes" in the pipe.