In this world of entropy, where everything constantly gets more complicated rather than less, water hammer arrestors are no exception. For the new engineers and designers who might be reading this article, water hammer arrestors are devices that absorb the pressure wave, or water hammer, that occurs when a quick closing valve shuts and rapidly stops the flow of water. I’ve seen pressure transducers measure water hammer, and it can easily spike momentarily to 500 psi just from the rapid closure of a kitchen faucet with a lever handle. Ironically, a lever handle kitchen faucet is not a location where you would provide a water hammer arrestor because it is not considered a quick closing valve, but that of course depends on how you use it.

Years ago water hammer arrestors were comprised of air chambers that were a capped piece of pipe mounted vertically and filled with air. The air provides the cushion to absorb that water hammer. Since air is soluble in water, over time these air chambers would become flooded because all the air would be absorbed into the water. Once flooded, the air chamber is no longer an air chamber but a water chamber and will no longer absorb the pressure from water hammer. (Remember, water is incompressible.)

Due to this anomaly, old fashioned air chambers required a drain and access such that they could be drained of water and filled with air periodically in order to function as intended. This of course was maintenance intensive.

Sometime in the 70s (I believe), water hammer arrestors started manufacture as a stainless steel bellows inside a steel housing. This device, manufactured by Zurn, Smith, Mifab and others, acted as a shock arrester that was maintenance free, since it used the expansion and contraction of the bellows to absorb the shock and did not rely on an air/water interface. Some years later, another device was created by PPP, Sioux Chief and others that used a piston inside a copper pipe chamber that moves up and down in response to pressure fluctuations, and like the bellows, is maintenance free.

Since I have been here in SoCal (16 years), it has been widely accepted that these devices do not require access since they are maintenance free unlike the air chambers of old ones. However, lately I have heard that some of the local plumbing inspectors are requiring access. The UPC is a little gray on this subject since it references installation per the manufacturer’s recommendations, and the manufacturers generally say that they are maintenance free so access is not required. But I’m hearing rumors that this is not satisfying all the powers that be, so you should probably check with your local authority.

One way to avoid the controversy of access panels is to use supply stops that include the water hammer arrestors on them so they are exposed below the sink or lav. This of course is a solution for sensor faucets but doesn’t help you regarding flush valves. The same solution can be applied to washing machine supply boxes and refrigerator ice makers.

The sizing and placement of the devices are dictated quite specifically by the various manufacturers, so you have to refer to their literature to perform this task. The requirement (or lack thereof) for the devices is far less specific and often a subject of hot debate. 

The general rule is that water hammer arrestors are required on all quick closing valves. Solenoid valves, the electronic snap-action valves that control water flow to sensor faucets, dishwashers, washing machines, ice makers and coffee machines, are all quick closing valves requiring water hammer arrestors — or do they? Water hammer is not just a function of how fast the valve closes but also the water flow, or more specifically, water velocity. So are water hammer arrestors really required on a 0.35 gpm sensor faucet with ¾” hot and cold supply pipes? In my opinion, no, but in strict accordance with the code, yes. 

We don’t usually see water hammer arrestors on coffee machines and refrigerator ice makers because the flow rate is small, the velocity is low, and historically they are never a problem. However, a washing machine has a much higher flow velocity so the likelihood of their necessity is much greater — or is it? Washing machines are fed by rubber hose connections between the unit and the wall supply. If these hoses are reasonably flexible, chances are they will absorb the potential water hammer before it returns to the piping system. But since as the engineer you have no control over the type of hoses used as the washer supply, you would be foolish not to install water hammer arrestors. For the same reason, a garden hose doesn’t create water hammer even if it is closed rapidly with a spring release trigger.

Flush valves are another gray area when it comes to water hammer arrestors, but again, you would be foolish not to specify them. Flush valves do not use solenoid valves to control the water flow, it is controlled by either a diaphragm or a piston, depending on the type of flush valve. Do these valves close quickly? The answer is “somewhat.” They do not close as quickly as an electronic solenoid, but the relative flow velocity is much greater than for a lavatory faucet, and a significant water hammer can occur depending on the valve operation and the pressure. I have heard flush valves — especially older models — literally rattle the walls.

The other day I was getting water from the drinking fountain at my gym — the old push button type of fountain — and I pondered the fact that the push button is a quick closing valve when released, so one could argue that a water hammer arrestor is required for a drinking fountain. Here again, the flow rate is small enough that water hammer does not occur in practice. An inspector could certainly argue that one would be required for the electronic ADA style drinking fountain because it does have solenoid control, but I have never heard of such a request.

If your need is to provide water hammer arrestors to satisfy code, then provide them wherever you have a solenoid valve or flush valve. If your need is to determine if water hammer is a legitimate concern, you need to consider the speed of the closure mechanism, the flow rate, pipe size and associated velocity. I have read that water hammer is rarely a concern below 4.9 FPS, but again, other variables come into play here. The idea of providing water hammer arrestors (and potentially access panels) willy-nilly bothers me, because it is the reckless application of an engineering concept. But unfortunately, in this age of litigation, we always have to err on the side of caution.

Timothy Allinson is vice president of Engineering at Murray Co., Mechanical Contractors, in Long Beach, California. He holds a BSME from Tufts University and an MBA from New York University. He is a professional engineer licensed in both mechanical and fire protection engineering in various states, and is a LEED accredited professional. Allinson is a past-president of ASPE, both the New York and Orange County chapters. He can be reached at lagunatictim@gmail.com.