Up-Feed or Down?
By Timothy Allinson, P.E.
Murray Company, Long Beach, Calif.
Blood, Sweat and Tears sang it in 1969 in their song "Spinning Wheel"
What goes up
must come down
got to go 'round
The song continues with 1960s hallucinogenic metaphor describing the futile sin of talking about your problems rather than just letting the "spinning wheel spin." Is it possible that David Clayton-Thomas was a frustrated plumber disguised as a successful rock-jazz-blues artist? Not likely. But his lyrics hold special meaning for our industry even today.
When designing a high-rise plumbing system, one of the first decisions to be made is the water distribution scheme. Should the system be an up-feed system? Or should it be down-feed?
I've heard the argument that it is a waste of energy to pump all of the water to the top of the building or zone when it may only have to travel up a few floors to point of use. This argument doesn't hold water (excuse the pun) since the pumps must work to maintain pressure at the top of the zone regardless of where the water is traveling. The only real loss of energy is the friction in the pipe that is slightly greater for a down-fed system, when the full flow volume must travel to the top of the zone through a single pipe.
Generally what drives the answer to the up-feed vs. down-feed dilemma is the architecture, as it stipulates the location of water heaters and/or the physical space for the distribution pipes. If you are designing a 10-story residence, and the only place for the water heaters is in the basement, while the only space for piping distribution is in the 1st floor ceiling, guess what? It will be an up-feed system.
If that same building requires that the water heaters go in the roof equipment room (which is good if there is a gas flue), and there is room in the top floor ceiling for distribution piping, then a down-feed system makes more sense.
But what do you do if the architect leaves it up to you? "Where do you want your heaters to go? And where do you want your distribution pipes?" What so you say? Do you know the best answer? From a plumbing standpoint the preferable system is the down-feed system. There are several reasons for this.
First, the down-feed plumbing system uses gravity to its advantage. This means that when sizing risers, the static gain exceeds the friction loss. How can I make this statement without consideration of pipe size and flow? Let's look at the numbers.
Static gain is 0.433 psi per foot, or 43.3 psi per 100 ft. to put the units into the context of the standard pipe flow chart. At 43.3 psi per 100 ft. the velocity in 1/2 in. pipe is in excess of 10 feet per second. As the pipe size increases, so does the velocity. Most codes limit velocity to 8 feet per second or less, so given the static gain friction coefficient, down-feed risers can be sized purely based on velocity and friction can be neglected entirely. The end result is that down-feed risers are almost always smaller in diameter than up-feed risers.
Second, when properly arranged, the friction coefficient for the down-feed system is always greater than the friction coefficient of the up-feed system. Last month I wrote about the optimal arrangement of a direct pumped system as having the pneumatic tank at the top of the building or zone, with a pressure-regulating rig downstream of the tank. Thus arranged, the distance from the PRV rig of the down-feed system to the furthest riser will always be less than the distance from the PRV rig to the top of the furthest riser in the up-feed system. Since the friction coefficient of the down-feed system is greater, this means that the distribution pipes will be smaller. This proves that both the distribution piping and the risers of a down-feed system will almost always be smaller than the up-feed system.
Let's look at a real world example. The project is called EVO, a 23-story condominium in downtown Los Angeles. (Look for the National Geographic Discovery Channel documentary called "The Building" airing sometime next year. It's about the construction of this building and may include an interview with yours truly.) With 21 residential floors, this building has a 10-story down-feed upper zone, and an 11-story up-feed lower zone.
Tables 1, 2 and 3 indicate how the sizing of the water piping varies depending on zone configuration.
Table 1 summarizes the fixture unit limits for each pipe size of both cold and hot water of high-zone horizontal distribution piping (the top of the system). Dictating criteria are a friction factor of 2.2 psi/100' which is calculated from a 40 psi static pressure, 30 psi residual, and 450' TDL of pipe from the PRV outlet to the top of the furthest down-feed riser. Velocity limits come into play as well. Note how the fixture units for the cold and hot water are the same for smaller pipe sizes, as the hydraulics dictate that these pipes are sized based on the common friction factor of 2.2 psi/100 in. As the pipes grow in size, velocity limits take over and the difference between cold and hot sizing limits widens.
Table 2 indicates fixture unit limits for the high-zone down-feed vertical risers. Note the differences between Tables 1 and 2, and the fact that the hot water limits in the larger pipe sizes are the same in both tables since the governing criteria is velocity. Keep in mind that if Table 2 dictates a riser to be 2 in. in size, the horizontal feed to that riser may have to increase to 2.5 in. based on Table 1.
Table 3 indicates fixture unit limits for all of the low-zone piping. The friction factor has been reduced to 1.8 psi/100 ft. due to 110 ft. of additional up-feed riser piping. Here again the larger hot water pips are still governed by velocity, but the sizing for the smaller risers varies greatly between Tables 2 and 3. It is in this difference that the economy of the down-feed system can truly be seen.
I hope you all have a wonderful Thanksgiving.
Timothy Allinson is a Senior Professional Engineer with Murray Company, Mechanical Contractors, in Long Beach, Calif. Prior to entering the design-build industry he worked for Popov Engineers, Inc. in Irvine, Calif, and JB&B in New York City. Tim 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. Tim is a past-president of ASPE, both the New York and Orange County Chapters, and sits on the board of the Society of American Military Engineers, Orange County Post.