Why Most Engineers Mess Up Water Heater Sizing
I've been going through these calculations for over a decade, and I still see the same mistakes on project plans. People pull numbers from a handbook, plug them in, and call it done. The problem is that handbooks don't account for what actually happens in buildings where people live and work. A hot water system sized on paper looks fine until the first morning rush and then everyone's taking cold showers. There's no single formula that works for every situation. The fundamental approach uses the Hunter Curve method combined with fixture unit calculations, but you also need to factor in recovery rates, tank volumes, and simultaneous demand patterns that textbooks often gloss over. Here's how it actually works in practice. Start with a fixture list. Every fixture has a flow rate and a usage probability. A standard residential bathroom sink draws about 0.5 gallons per minute. A shower head is typically 2.1 GPM, though some old units push 5 GPM. Commercial restroom faucets are usually 0.5 GPM as well. You need to know exactly which fixtures you're dealing with because the difference between a 1.5 GPM and a 2.1 GPM shower head compounds fast when ten showers run at once.
Next, calculate the peak demand using the Hunter method. You assign a probability value to each fixture type. The more fixtures of a given type, the lower the probability that all of them draw simultaneously. A single shower might have a 50 percent chance of use at peak hour. Three showers? More like 75 percent. But here's the thing that trips people up: the probabilities aren't linear, and they change significantly between residential and commercial applications. Residential buildings have predictable morning and evening peaks. Hotels, hospitals, and dormitories have completely different demand curves that peak at unpredictable times throughout the day. The recovery rate matters just as much as the storage capacity. This is where most sizing guides fall apart. They tell you to pick a tank based on first-hour delivery ratings. That rating combines stored hot water plus the heater's ability to reheat incoming cold water within the first hour of use. For a residential unit, the FHR (first-hour rating) typically ranges from 35 to 80 gallons depending on the heater size and BTU output. A standard 40-gallon tank might only deliver 50-55 gallons in the first hour because you can't draw from the bottom of the tank without mixing cold water in. A 50-gallon unit might get you 65-70 gallons of usable hot water. I worked on a project last year where we sized a water heater system for a 24-unit apartment building. The engineering firm before us used a simple fixture-count method that resulted in a 75-gallon commercial unit. It failed within two weeks. The issue wasn't the calculation method itself, it was that they didn't account for the fact that all twenty-four units had electric showers and high-flow rain shower heads. The morning peak between 6 AM and 8 AM created a demand spike that the unit couldn't recover from. We ended up switching to a indirect-fired unit with a 120-gallon storage tank paired with a 200,000 BTU boiler. The system now handles the demand without issues. The cost was roughly 40 percent higher upfront, but the alternative was customer complaints and callbacks.
Commercial buildings require a different approach entirely. You need to look at the occupancy load, the type of establishment, and the operating hours. A restaurant's water demand during dinner service looks nothing like a hotel's demand during checkout morning. Healthcare facilities have their own requirements governed by ASHRAE Standard 170 and local plumbing codes. You cannot skip those codes. They exist for a reason, and violating them will get your plans rejected or your facility shut down during inspection. One detail that beginners consistently miss is the temperature drop across the system. The water heater might output 140°F water, but by the time it travels through pipes, mixes with cold water in the tank, and reaches the farthest fixture, the temperature could be 10°F to 15°F lower. In a large commercial installation with extensive piping runs, that drop can exceed 20°F. If you size your heater to deliver exactly 120°F at the outlet, the fixtures at the end of the line might only receive 100°F water, which is below code minimum for many applications and certainly below what anyone considers comfortable. Another common mistake is ignoring the makeup water temperature. Winter cold water entering the system might be 40°F while summer water could be 75°F. Your heater has to work significantly harder in winter to reach the same output temperature. I've seen systems sized for summer conditions that couldn't keep up during January. The fix is simple: size for the coldest expected inlet temperature in your climate zone, not the average. The equipment will cost a bit more, but it'll actually work year-round.
Get the Full Details
For tankless water heaters, the sizing calculation changes completely. You're no longer dealing with stored energy. You're dealing with flow rate and temperature rise. The formula is straightforward: GPM multiplied by the temperature rise in degrees Fahrenheit multiplied by 8.33 (the weight of a gallon of water in pounds) divided by 60 gives you the BTU per hour required. A tankless unit that delivers 3.5 GPM with a 70°F temperature rise needs approximately 36,000 BTU per hour. That sounds small. But if you're serving multiple fixtures simultaneously, you need to add their demands together. Two showers at 2.1 GPM each plus a kitchen faucet at 0.5 GPM equals 4.7 GPM total. At 70°F rise, that's roughly 49,000 BTU. Most residential tankless units max out around 100,000 to 150,000 BTU, so one unit can handle it. But in a commercial application with five or six simultaneous fixtures, you might need two or three tankless units in parallel, and the piping layout becomes significantly more complex. Demand characteristics also shift throughout the day in commercial buildings. A school has a huge spike during lunch when every cafeteria tap runs at once. An office building might see its peak during cleaning hours in the evening rather than during business hours. Understanding these patterns requires reviewing actual usage data when available. Some buildings have submetering that shows you exact consumption patterns. If you don't have that data, you're working from estimates, and estimates carry risk. The energy efficiency side of sizing is another area where shortcuts cause problems. Oversizing a water heater doesn't make it more efficient. In fact, a significantly oversized tank wastes energy maintaining temperatures for water that may not get used. The standby losses from a 100-gallon tank sitting idle are real and measurable. Undersizing causes the system to run constantly, which also reduces efficiency and increases wear. The sweet spot is sizing close to your calculated peak demand with maybe 10 to 15 percent margin. Anything beyond that is just throwing money at the problem.
For large institutional buildings, I recommend running a hourly load simulation. Software like H2X Analytics, Elite Software's Phast, or even basic spreadsheet models can track demand across all 24 hours of a typical day. This reveals peak loads that a simple fixture-unit calculation might smooth over. The simulation takes about 30 to 45 minutes to set up once you understand the building's schedule. The result is a sizing decision based on actual data instead of guesswork. Check your local plumbing code requirements before finalizing any size. Some jurisdictions mandate minimum first-hour ratings for certain occupancies. Others require recirculation systems for buildings above a certain square footage or with piping runs longer than a specified distance. Los Angeles, for example, has strict requirements for commercial water heater efficiency and often requires point-of-use heaters for distant fixtures. New York City has its own energy code amendments. Ignoring these because "the national code allows it" will cost you time and money during plan review. Finally, leave room for future expansion. A building that gets renovated or reconfigured six months after occupancy will stress an exactly-sized system. Adding a few extra gallons of storage or upsizing the recovery capacity by one tier costs relatively little compared to the disruption of replacing a complete water heating system. The additional upfront cost is usually 15 to 25 percent of the equipment price. The cost of a retrofitted system once the building is occupied is three to five times that amount.
