Gas pipe sizing is one of those tasks that looks simple until you do it wrong. The code books are full of tables, but the tables are useless if you do not understand the methodology behind them. The IRC and IFGC provide sizing tables for different pipe materials, pressures, and pressure drops. The plumber's job is to identify the longest run from the meter to the farthest appliance, calculate the total BTU demand on each segment of pipe along that path, and look up the minimum diameter that carries that demand within the allowable pressure drop.
Most sizing errors come from three sources: not using the longest-run method correctly, not adding up segment loads properly, and not accounting for fittings. A fourth error, less common but more dangerous, is sizing LP gas piping with natural gas tables. Propane has a higher specific gravity and behaves differently in the pipe. This guide walks through the complete methodology, shows worked examples, and explains the common mistakes that cause failed inspections and pressure complaints.
The Longest-Run Method: Why It Matters
The IFGC and IRC both use the longest-run method for gas pipe sizing. The idea is simple: find the pipe segment that runs from the gas meter (or point of delivery) to the most distant appliance. This is the critical path. The total length of this path determines which column of the sizing table you use. Every segment along this path is sized for the cumulative BTU load it carries.
The common mistake is sizing each branch independently. If you have a furnace 30 feet from the meter and a water heater 50 feet from the meter (branching off the main at 20 feet), the critical path is 50 feet to the water heater. The main from the meter to the tee at 20 feet carries the combined BTU load of both appliances. The branch to the furnace carries only the furnace load but is still sized using the 50-foot column, not the 30-foot column, because the total system length governs the pressure drop budget.
Why does this work? Because the allowable pressure drop (typically 0.5 inWC) is a budget shared across the entire system. If you size the main using a 30-foot table and the branch using a separate 30-foot table, you have implicitly assumed 0.5 inWC drop for each, which means the total system could experience up to 1.0 inWC drop. That exceeds the allowable budget and can cause pressure problems at the farthest appliance.
The longest-run method ensures that the total pressure drop from the meter to the most distant fixture stays within the allowable limit, no matter which branch you follow.
Gas Pipe Sizing Calculator
Size natural gas piping per IFGC/IRC using the Spitzglass formula. Select appliances, pipe material, and run length to find minimum pipe diameter.
Reading the Code Tables Correctly
The IRC and IFGC provide separate sizing tables for each combination of pipe material, gas type, inlet pressure, and pressure drop. IRC Table G2413.4(1) covers Schedule 40 metallic pipe for natural gas at less than 2 PSI inlet pressure and 0.5 inWC pressure drop. Other tables cover CSST, copper, PE, and different pressure/drop combinations.
Each table has rows for pipe diameters and columns for developed length. The cell values are the maximum BTU capacity of that diameter at that length. To use the table: (1) determine your longest-run developed length, (2) find the appropriate column, (3) for each segment, find the row where the cell value equals or exceeds the BTU demand on that segment.
Developed length includes equivalent lengths for fittings. A 90-degree elbow on 3/4-inch pipe adds roughly 2 feet of equivalent length. A tee adds about 4 feet. If you have a 40-foot measured run with four elbows and two tees, your developed length is 40 + (4 times 2) + (2 times 4) = 56 feet. Use the 60-foot column in the table (round up to the next available column).
If your developed length falls between two columns, always use the longer column. Using the shorter column would overstate the pipe's capacity and could result in undersized pipe with inadequate pressure at the appliance.
L_developed = L_measured + sum of fitting equivalent lengths
Typical equivalent lengths (3/4" pipe):
90-degree elbow: ~2 ft
Tee (through): ~2 ft
Tee (branch): ~4 ft
Field shortcut: add 50% to measured length for fittings
Calculating BTU Load Per Segment
Every pipe segment carries the total BTU demand of all appliances downstream of it. The main line from the meter carries the total connected load of the entire building. A branch that serves only a water heater carries only the water heater load. This sounds obvious, but mistakes happen when the piping layout is complex.
Start at the meter and work outward. At the first tee, the main divides into two branches. The main upstream of the tee carries the sum of both branches. Each branch carries its own load. If a branch subdivides further, the same logic applies recursively.
Use input BTU ratings from the appliance nameplate, not output ratings. A furnace with 60,000 BTU output and 96% AFUE has an input rating of about 62,500 BTU. A water heater with 40,000 BTU input is 40,000 BTU for pipe sizing regardless of its efficiency. The gas pipe does not care about efficiency; it cares about fuel volume.
For LP gas, the BTU content per cubic foot is 2,516 (vs 1,000 for natural gas), so LP piping carries a smaller volume of gas for the same BTU load. However, LP has a higher specific gravity (1.52 vs 0.60 for NG), which increases friction loss. The net effect is that LP pipe sizes are usually similar to or one size larger than NG for the same BTU load, depending on the table used.
CSST: Corrugated Stainless Steel Tubing
CSST (brands like TracPipe, Gastite, and CounterStrike) is flexible corrugated stainless steel tubing that is faster to install than rigid black iron pipe. It uses fewer fittings because it can bend around obstacles. But it has its own sizing rules that differ from rigid pipe.
CSST is sized using equivalent hydraulic diameter (EHD), not nominal pipe size. A 3/4-inch EHD CSST tube does not have the same capacity as a 3/4-inch Schedule 40 black iron pipe. The corrugations increase friction loss, so CSST typically requires one size larger than rigid pipe for the same BTU capacity and length.
Each CSST manufacturer publishes their own sizing tables. You must use the manufacturer's table, not the generic IFGC rigid pipe table. TracPipe and Gastite tables produce different results for the same EHD because the internal geometry differs between brands.
CSST also requires bonding per NFPA 54 and the manufacturer's installation instructions. A bonding conductor connects the CSST system to the building grounding electrode system. This is a code requirement that inspectors check. Failure to bond CSST is a common reason for failed inspections, and some insurance companies restrict or surcharge properties with unbonded CSST.
Common Mistakes That Cause Failed Inspections
The most common gas pipe sizing mistake is not using the longest-run method. Plumbers who size each branch independently often end up with undersized mains that cannot deliver adequate pressure to all appliances running simultaneously. The inspector may not catch it on paper, but the homeowner will notice when the furnace and water heater fire at the same time and the range burners turn yellow.
The second mistake is ignoring fittings. A 30-foot run with six elbows and three tees could have an actual developed length of 50+ feet. If you sized using the 30-foot column, you are significantly undersized. Either count fittings or use the 50% add-on shortcut.
The third mistake is mixing up gas types. LP gas tables are not the same as natural gas tables. If you use an NG table to size LP piping, you will undersize because the higher specific gravity of LP increases friction loss. Always verify you are using the correct table for your gas type.
The fourth mistake is not checking the supply pressure. Standard residential NG is 7 inWC at the outlet of the meter regulator. Standard LP is 11 inWC. If your supply pressure is lower (old regulator, high demand on the distribution main), your pipe capacity is reduced. Tables are published for specific inlet pressures.
The fifth mistake is not doing the pressure test. IRC requires a pressure test of 3 PSI for 15 minutes with no drop. If you skip this step or the test fails, the inspector will not sign off. Period.