Subscribe to our newsletters & stay updated
Gary and I have covered a lot of ground in our last seven articles in Plumbing Engineer about practically perfect plumbing design. We focused on how high resolution hot and cold water peak flow data show that the Hunter Curves that have long been the foundation for pipe sizing vastly over-predict actual peak water use and on how we right-sized and designed compact energy-efficient plumbing in multifamily buildings.
We explained how these transformations resulted in much lower installation costs, faster arrival of hot water at fixtures, less wasted water and energy, and happier tenants and building owners.
In this and a companion article to follow, we delve into the origins and underpinnings of our approach and how we have confirmed that our designs really work well.
Fools Rush In
My dad used to admonish me as I stumbled through my youth — “Use your head for more than a hat rack!” When the developer of Solara, a net-zero multifamily project near Schenectady, N.Y., asked me to design the plumbing system for his buildings, I doffed my hat and dove into what I thought would be a sophomoric task. How hard could it be to choose pipe sizes and layouts to get water to the fixtures quickly and cheaply without much energy and water waste?
Not being trained as a mechanical engineer (ME) didn’t worry me. I had worked with water in pipe since I was a little kid and studied water flow management in my civil engineering classes. Plus, pipe size choices for solar thermal systems (a 10-year second career for me) are a centerpiece of every job I build.
In hindsight, not knowing about the seminal works by Roy Hunter published prior to World War II that MEs have long relied upon was no problem since there are first principles and equations that define pressure loss and time to tap. Also, there were tables that defined pressure loss in fittings, fixtures and pipe. And I had high-resolution water consumption data from other multifamily buildings collected as part of my solar thermal system design and sizing efforts (see Figures 1 and 2).
I set to work, comfortable with the notion that I had the tools engineers need to create a design that would achieve these goals.
Start with First Principles
I had a lot of experience sizing pipe as part of my solar thermal designs. I understood that big pipe costs more, requires more antifreeze and wastes more heat. On the other hand, big pipe reduces erosion-corrosion, pressure losses and pumping power requirements.
Right-sizing the piping became an obligatory task for each system. I relied upon plumbing guidance and equations like the continuity, Hazen-Williams and Darcy-Weisbach equations and their appropriate coefficients, as well as industry pressure-loss value tables for copper and PEX pipe and appropriate fittings’ equivalent lengths.
As I tried to apply my solar project knowledge to the Solara water distribution design, I pondered a different but central question that Roy Hunter faced: water draw simultaneity. How many simultaneous fixture draws should I design for? Peak flows generated by multiple overlapping draws would be the critical stressor that could generate tenant complaints when the shower flow turned flaccid: the dreaded PD — pressure dysfunction.
Thankfully, unlike Mr. Hunter, I had high-resolution hot water consumption data from similar buildings that I could hang my hat on to estimate peak flows. I reasoned that if the peak flow was about 4 gallons/minute (gpm) for the whole building (see Figures 1 and 2), designing for that flow was justified — the simultaneity quandary was resolved.
For example, 4 gpm of 105 F water flowing in a 3/4-inch PEX pipe segment 50 feet long would exert a pressure loss of only 1.9 pounds/square inch gauge (psig); for a 1-inch PEX pipe segment, 0.6 psig. Both are relatively minor losses for standard distribution systems. Goodbye to those 2- and 3-inch feed trunks! (See this link for the calculations: https://bit.ly/3EJYutT).
Fast-forward a couple of years. I reached out to Gary, who told me about the development of IAPMO’s Water Demand Calculator (WDC). It was during this discussion that I finally learned about the Hunter Curves and the requirements for water distribution systems found in the plumbing codes. Who knew?!
Version 1 (which Gary and I used for three 35-unit plumbing designs that were built in 2020) and Version 2 of the WDC predict peak water flows much lower than Mr. Hunter’s curves. The WDC is finding its way into codes and design guidance as more plumbing engineers become familiar with its benefits.
That said, my actual measurements of water consumption in multifamily buildings still suggest much lower peak flows, suggesting that the WDC features comfortable safety factors. Also, as temperatures of mixed hot water increase, actual hot water flows in feed piping diminish, adding an even higher safety factor for hot water pipe sizing.
Confirm Pressure Losses with Testing and Mockups
Trust but verify. It’s a great motto for diplomatic exercises — and for plumbing design projects. It was relatively easy to choose pipe sizes based on the draw data I had and first principles.
And I discovered pressure-drop data for modern pipe and fittings, time to tap research, and optimal plumbing layouts by this fellow Gary Klein, a true plumbing iconoclast. Small and simple is beautiful, he preached. But what if I made a mistake or he was wrong or both?
My shop offered the answers — mocking up the critical distribution elements and checking pressure losses under stress-filled draw scenarios. So, with Gary’s guidance, I built full-sized pipe-and-fitting mockups and measured pressure losses at various nodes. Doing so required the development of appropriate procedures and stress condition scenarios, and the procurement of reliable water flow generation and pressure measurement equipment.
I mocked up corridor trunklines to apartment branches and, separately, the branch to manifold to twig to fixture piping (see Figures 3 and 4). After surmounting numerous challenges, our data confirmed our pressure loss and time to tap estimates. The experiences were both reassuring and instructive; the design worked.
I delivered my design to Solara’s plumber, who rolled his eyes but went about creating a more detailed sketch to guide his crew to do the build. I thought my job was done and returned to building my solar systems at the same project.
But, to be super sure, after the plumbers leak-tested the network (see Figure 5), I did some trunkline pressure-testing at the end of the corridor trunk lines, fittings included. I used flows higher than I ever measured and predicted for this building — once again, the data confirmed relatively low pressure drops. Whew! Let those tenants in the door!
There are many ways to approach a plumbing design task. I chose to rely on first principles, high-resolution peak flow data I collected and mockup testing to right-size pipe and create an energy-efficient, net-zero distribution system.
Stay tuned for what happened next in the second article. A little teaser: the showers suffered from PD.