Critical gas pipe sizing

In the past I have written about gas pipe sizing, but some recent events have caused me to want to revisit the subject. As many of you know, the Code, be it UPC or IPC, is based on NFPA 54, which has no diversity built into it when it comes to sizing gas piping. That is to say, it assumes that every gas fired appliance and gas fired piece of equipment is firing at 100 percent capacity all at the same time. There is no Hunter’s Curve when it comes to gas pipe sizing – every day is Thanksgiving in the eyes of NFPA.

Imagine that? In a residential high rise, NFPA assumes that every burner on every stove is set to high; every oven and every broiler is on high temp; every gas dryer is running at the same time; every log lighter is lighting logs; every water heater is firing at full capacity; and so on. This, of course, is absurd and it ends up with gas piping systems that are dramatically over sized.

However, NFPA is written for the theoretical worst-case “critical” scenario. How often does that happen in gas systems? Never for a residential high-rise. But, it can happen for more unique installations.

Case in point: a very large residence (think billionaire) with an emergency generator that is dual fired, gas and propane. The gas service to the residence feeds the generator, as well as hot water heaters, pool heaters, spa heaters, stoves, ovens, dryers, patio heaters, and so on.

The generator of course stands alone fairly far from the house. It has a dedicated branch off the gas service main that serves nothing other than the generator. That pipe falls under what I would call critical gas pipe sizing. When the generator runs at 100 percent load, it defines the limits of the Code gas pipe sizing tables and NFPA 54. There is no diversity. There is no fudge factor. NFPA 54 does not have fudge factors – it just appears so most of the time because of the inherent diversity of gas appliances.

So, if the pipe serving this generator is engineered down to the “gnat’s ass”, for lack of a more delicate term, you can end up in trouble. Why? Because the NFPA and Code gas pipe sizing tables leave no margin for error when you are dealing with 100 percent load and no diversity.

With no margin for error, that means that if the contractor is required to run the gas piping in a fashion that is longer than what you had figured in your calculation, you are in trouble. If the contractor is forced to install more elbows than you had factored in with your 10 or even 25 percent fitting allowance, you are in trouble. If the gas regulator isn’t the appropriate distance in terms of pipe diameters away for the generator connection, you are in trouble. There are variable fixes to some of these issues such as increased pressure drop allowances etc. But for this case, let’s assume a fixed pressure that cannot afford a larger pressure drop.

In the case of this large residence the engineer in his attempt at regimental efficiency designed the gas system down to the gnat’s ass. In reality, the gas piping could not be run in the field in the direct route shown on his drawings, but had to be offset and run around the generator enclosure in such a fashion that it nearly doubled the length of the gas piping to the generator. The 10 percent fitting allowance figured into the calculation turned out to be far too little compared to the relatively short pipe run and the number of fittings required in the field to connect to the generator.

There is also the complicating factor of gas pipe volume and the requirements of the particular gas fired piece of equipment. The Code may determine one pipe size, but in reality the appliance and associated regulator might require a buffer volume of gas, similar to a cushion tank is a domestic water system. Pipe sizing might meet Code requirements, but that might not meet the operational requirements of a big piece of gas-fired equipment sensitive to fluctuations in pressure upon start-up.

Some manufacturers require a 1.5X pipe diameter size ahead of the generator to satisfy the draw of the equipment. During generator testing, you can find results of the power consumption in relation to the engine operation and find that an undersized gas line will not allow the generator to function properly due to larger than allowed pressure drop at the equipment.

The bottom line is, if you size gas piping that falls into the “critical sizing” category, meaning equipment that has no diversity, and you size it without any room for field changes and equipment sensitivity, then you are going to wind up in trouble. Such systems or system branches must be given a little fat as a buffer. The gas branch should probably be made one or two pipe sizes larger than what Code dictates to allow for field variance. The point of connection should probably be made larger than necessary to provide some cushion during start-up. I am generally not a fan of throwing arbitrary factors of safety into engineered designs, but in the case of critical gas pipe sizing I think, in fact I know, that it is warranted.

A word on propane tank sizing: I learned, in conjunction with the above lesson (many of my articles are based on lessons learned), that propane gas tank sizing is tricky business. I had never had to size a propane tank in 25 years prior to this residential project. The tank selected for this generator was based on required run time and space constraints. What I didn’t know then but know now is that propane tank delivery is a function of tank volume and associated surface area. A run time calculation might determine that you need a 500-gallon tank, but the propane flow demand might not be able to be delivered by a 500-gallon tank.

Why is that? Propane delivery is a function of surface area. Liquid propane can only evaporate so fast, so the evaporation (delivery) rate is a function of surface area and tank volume. The details are too specific to get into here, but be sure to speak with the propane vendor if you have to specify a propane tank. If the demand exceeds the evaporative delivery capabilities of the tank size you want to specify, then an evaporator will be required to convert the liquid propane into gas at the desired demand flow rate.

The devil, of course, is always in the details. I hope you enjoyed your summer. 

Timothy Allinson is vice president of Engineering at Murray Co., Mechanical Contractors, in Long Beach, Calif. 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

Content Type: