Back in 1985, when I first entered this industry, my then boss at JB&B, Richard Reilly, wrote a feature article for Plumbing Engineer titled, "High-Rise Plumbing Design." I was very familiar with the piece because I helped him by drafting several of the images – back when we drafted by hand. A lot has transpired in the ensuing 30 years, including the passing of Richard Reilly. In his memory (and on St. Patrick’s Day) I am attempting to do him justice by summarizing the essence of what Reilly wrote, and perhaps give it a 30-year update as well.
We start by asking the question why is our “engineered science” frequently broken down into “plumbing” and “high-rise plumbing?" Don’t the same scientific laws of hydraulics and pneumatics apply whether the structure under design is a low-rise or a high-rise? Why single out high-rise for special attention?
Setting aside the obvious distinction of static pressure of water developed by height, it was Reilly’s opinion that the main difference between low and high-rise plumbing is only superficially related to height, and more directly related to increased population. As Richard said many times, “Fixtures don’t use water, people do.”
Increased vertical population acts to stress the demand on the piping system. It has long been known, due to the work of Dawson and Kalinske in 1939, that water traveling down a vertical drainage pipe will reach its terminal velocity in about 14 feet. So, from a waste velocity standpoint, the difference between one story and 100 stories is essentially moot. However, the increased stress on the piping system created by the dense population of a high-rise is where the danger lurks. A pipe sizing error made in a low-rise building would likely go unnoticed, whereas the same error in a high-rise structure creates headaches. Let us address this to avoid headaches.
First, let us look at sanitary drainage. It has been proven, both theoretically and empirically, that the maximum capacity of a sanitary stack is when the flow that streams down the inner wall of a vertical pipe, due to the surface tension of water, reaches a cross sectional area of about 7/24 the cross sectional area of the pipe. The flow drags with it a proportional amount of air, which if not relieved, will produce negative pressure in the upper portion of the stack, versus positive pressure at changes in direction and at the base of the stack. Relief vents or yolk vents are required to interconnect the waste and vent stacks, and give these pressure differences the opportunity to balance out. The goal is that the pressure differential be no more than 1 inch of water column to prevent traps from being blown out or siphoned dry. As an example, it can be calculated that a 4-inch stack flowing 180 gpm will move a corresponding 600 gpm of air flow. This is why large, heavily loaded stacks often have even larger vent stacks in parallel.
One means of controlling the pipe sizes and potential pressure fluctuations in very tall stacks was employed in the design of the Sears Tower (now the Willis Tower) by using intermediate building drains connected to an express riser that had no fixture connections. The beauty of this design is not only in reduced pipe sizes, but in the fact that the express riser had no fixtures connected thereto. Any pressure surges occurring in the express stack would not impact the operation of any of the building fixtures. Refer to Figures 1 and 2 for the conventional design compared to the express riser configuration.
Reilly wrote at length about controlled flow roof drainage, which is a system that uses specially designed roof drains that restrict the flow of water into the drain and store most of the rainwater on the roof in an effort to reduce pipe sizes. Drain down time of the roof can be as long as 24 hours. However, the system never really gathered traction in my experience, as the potential for roof leaks and voiding of warrantees outweighed the savings in piping. In its stead, siphonic roof drainage has developed momentum over the ensuing 30 years. Siphonic drainage uses special drains that allow water to pond around the drain and when the rainfall reaches a critical point the drain creates a siphon. Piping flows 100 percent full and runs flat due to the siphonic effect. This reduces pipe sizes and eases coordination. A building like the Willis Tower would have been an ideal candidate for siphonic drainage, due to its large, flat roof setbacks.
Since storm drain piping is subject to pressure fluctuations similar to sanitary piping, without having the benefit of the relief vents in the sanitary system, care must be taken at the base of stacks not to connect any drains, such as plaza drains, to the turbulent section of drainage flow. Plaza drains should be collected independent of the building storm drain and connected at a point downstream of the base of stack where pressure fluctuations have diminished.
Domestic water systems in any building must limit delivery pressure to fixtures to 80 psi. This challenge is obviously of great importance in high-rise commercial buildings. It is most often done using pressure regulating valves, but it can also be done by creating separate pumped zones. In the case of the World Trade Center, the towers were divided into pressure zones using break tanks on each mechanical level. The break tanks were filled by the pumps from the zone below and fed the suction of a set of pumps to serve the zone above. It was, in effect, a series of shorter towers stacked one on top of the other – a very efficient design if you have the real estate for the tanks, and if you are not in a seismic zone that makes elevated tanks impractical.
In Reilly’s article, he spoke at length to the pros and cons of variable speed drive pumping systems versus constant speed pumps. Thirty years ago, variable speed drives came at quite a premium, and for high-rise plumbing applications there was little basis to justify the cost. Today, Variable Frequency Drives (VFD) are much less expensive and have become quite the norm for pump control. Even though there is little savings in energy for the plumbing system, the benefit of the smooth VFD control makes them a no-brainer for direct pumped systems. The only real exception that I can think of is if the pumps are filling a water tank and operate purely in an on/off mode based on level control. For this application, VFD control would serve no purpose.
Pressure zoning of the domestic cold water is closely coupled to the design of the hot water system. In commercial buildings that use limited hot water at the core toilet lavatories, hot water is often served locally, either with instantaneous electric heaters at each grouping of lavatories, or with small electric heaters that might serve a grouping of floors – anywhere from three to 10. When this is the case, water pressure can be easily controlled with relatively small, direct-acting PRV assemblies that control the pressure zones of the building to within 30 and 80 psi. When large, central hot water systems are required – such as for a residential high-rise – pressure zoning becomes more sensitive because of the complexity of the hot water distribution and circulation system. In a commercial building, it is quite simple to circulate hot water (when circulation is required) back to the local electric water heaters. Just remember Richard Reilly’s mantra, "Thou shalt not circulate through a Pressure Reducing Valve." n