SizingKitHazen-Williams
Pipe friction loss calculator
This works water friction loss by Hazen-Williams and returns it the way the trade tables print it: feet of head and psi per 100 ft, charted down every size of the pipe family so the cost of going one size up is a row away. Free and no account needed. Because the formula is an 1905 empirical fit with no viscosity term, the page also checks your inputs against the four conditions it was fitted inside — temperature, velocity, bore and flow regime — and says which ones you are outside.
- 100% free
- No signup
- 13 materials, 3 C columns
- NFPA 13 values included
- Validity checked, not assumed
Flow, pipe and C
Straight pipe only. This method costs the wall, and fittings are added separately.
The chart below covers every size in the family at once.
Independent of the dimension family on purpose: an unlined cast-iron main and a lined ductile one share a bore and not a C.
Steel, black: 140 new, 120 for design, 120 under NFPA 13. NFPA 13 gives 120 for black steel in a wet or deluge system and 100 in a dry or preaction system, because a dry system corrodes at every cycle. On untreated domestic water expect C to fall into the 90s within a few decades.
Nothing in the result moves when you change this. It is here so the page can tell you when the method has stopped applying, and to convert head into psi at the right density.
FRICTION LOSS PER 100 FT
0.84 ft
0.37 psi · 0.025 bar · 0.26 m per 30.5 m
OVER THE WHOLE LENGTH
2.11 ft
0.91 psi · 0.64 m
| Bore used | 4.0260 in · 102.26 mm |
|---|---|
| Velocity | 2.52 ft/s · 0.77 m/s |
| C factor | 120 — Steel, black |
| Reynolds number (checked, not used) | 69,966 |
| Darcy-Weisbach on the same pipe | 0.62 ft per 100 ft — 36% above this answer |
| Water density at this temperature | 62.366 lb/ft³ · 999 kg/m³ |
Inside the envelope on all four conditions
- Water between 40 °F and 75 °F — measured 60 °F. Condition met.
- Velocity at or under 10 ft/s — measured 2.52 ft/s. Condition met.
- Bore 2 in or larger — measured 4.026 in. Condition met.
- Fully turbulent, Re above 4,000 — measured 69,966. Condition met.
Loss per 100 ft across Steel, Schedule 40
100 gpm at C 120. Every size in the family, so the cost of going one size up is a row apart rather than a second calculation.
| Nominal | Bore, in | ft/s | ft per 100 ft | psi per 100 ft |
|---|---|---|---|---|
| 1/8 in | 0.269 | 564.53 | 446733.93 | 193477.99 |
| 1/4 in | 0.364 | 308.31 | 102407.12 | 44351.96 |
| 3/8 in | 0.493 | 168.07 | 23371.18 | 10121.93 |
| 1/2 in | 0.622 | 105.59 | 7534.40 | 3263.11 |
| 3/4 in | 0.824 | 60.16 | 1915.12 | 829.43 |
| 1 in | 1.049 | 37.12 | 590.94 | 255.93 |
| 1-1/4 in | 1.380 | 21.45 | 155.41 | 67.31 |
| 1-1/2 in | 1.610 | 15.76 | 73.35 | 31.77 |
| 2 in | 2.067 | 9.56 | 21.72 | 9.41 |
| 2-1/2 in | 2.469 | 6.70 | 9.14 | 3.96 |
| 3 in | 3.068 | 4.34 | 3.17 | 1.37 |
| 3-1/2 in | 3.548 | 3.25 | 1.56 | 0.68 |
| 4 in | 4.026 | 2.52 | 0.84 | 0.37 |
| 5 in | 5.047 | 1.60 | 0.28 | 0.12 |
| 6 in | 6.065 | 1.11 | 0.11 | 0.05 |
| 8 in | 7.981 | 0.64 | 0.03 | 0.01 |
| 10 in | 10.020 | 0.41 | 0.01 | 0.00 |
| 12 in | 11.938 | 0.29 | 0.00 | 0.00 |
| 14 in | 13.126 | 0.24 | 0.00 | 0.00 |
| 16 in | 15.000 | 0.18 | 0.00 | 0.00 |
| 18 in | 16.876 | 0.14 | 0.00 | 0.00 |
| 20 in | 18.814 | 0.12 | 0.00 | 0.00 |
| 24 in | 22.626 | 0.08 | 0.00 | 0.00 |
C by material, and what age does to it
| Material | New | Design | NFPA 13 | What moves it |
|---|---|---|---|---|
| PVC | 150 | 150 | 150 | Does not age measurably. NFPA 13 gives 150 for all listed plastic pipe. |
| CPVC | 150 | 150 | 150 | As PVC. |
| PEX | 150 | 150 | 150 | As PVC. The small bore, not the C, is what costs PEX its head. |
| Polyethylene (HDPE, PE) | 150 | 145 | 150 | As PVC; the small design allowance is for fusion beads on a long run. |
| Copper | 150 | 140 | 150 | Very stable. Design allowance is for scale in hard water, which is a real effect on hot lines over decades. |
| Stainless steel | 150 | 145 | 150 | NFPA 13 groups stainless with copper tube at 150. |
| Steel, black | 140 | 120 | 120 | NFPA 13 gives 120 for black steel in a wet or deluge system and 100 in a dry or preaction system, because a dry system corrodes at every cycle. On untreated domestic water expect C to fall into the 90s within a few decades. |
| Steel, galvanized | 120 | 100 | 120 | The worst ager in common building service. Old galvanized supply loses bore to tuberculation as well as C, so both terms of the equation move the wrong way at once. |
| Cast iron, unlined | 130 | 100 | 100 | The classic aging case: roughly 130 new, 100 at twenty years, 60–80 at forty on aggressive water. NFPA 13 requires 100 for unlined cast or ductile iron. |
| Cast or ductile iron, cement-mortar lined | 140 | 140 | 140 | The lining is why modern ductile mains hold their capacity. NFPA 13 gives 140. |
| Concrete | 140 | 130 | 140 | Depends heavily on the finish; a rough-formed sewer is nearer 100. |
| Riveted steel | 110 | 100 | — | Historic penstocks only. Conventional waterworks value. |
| Corrugated metal | 60 | 60 | — | Far outside what Hazen-Williams was fitted to. Use Manning's n for a culvert instead. |
Sprinkler column: NFPA 13, Standard for the Installation of Sprinkler Systems, Hazen-Williams C values table (numbered 23.4.4.7.1 in the 2022 edition, 22.4.4.7.1 in 2019, 14.4.4.7.1 in 2016) — unlined cast or ductile iron 100, black steel dry and preaction 100, black steel wet and deluge 120, galvanized all 120, listed plastic 150, cement-lined cast or ductile iron 140, copper tube and stainless steel 150, concrete 140. New and design columns are conventional waterworks and plumbing design practice (AWWA M11, ASPE Data Book), where C is chosen for the condition the pipe is expected to be in over its life rather than the condition it is in on the day it is laid.
C is not a measured property and there is no single right value. It hides roughness, bore condition and the Reynolds number in one empirical number, which is why it ages and why two engineers can defensibly pick figures 20 apart for the same main. Use the new-pipe value only for an installation being commissioned; use the design value for anything that has to work in twenty years; use the sprinkler value and nothing else for a fire sprinkler calculation, because there the code names the number. And note the direction of the error: a high C is optimistic, and an optimistic C under-sizes the pipe.
Where this method stops. Hazen-Williams is an empirical formula fitted to water in ordinary pipe. It holds for water at roughly 40–75 °F, in turbulent flow, at velocities under about 10 ft/s, in pipe of about 2 in and larger. It has no viscosity term, so it cannot see temperature and cannot be used for any other fluid — not oil, not glycol, not air, not hot water far from room temperature. Outside its envelope it returns a plausible wrong number rather than an error, which is why Darcy-Weisbach is the default on this site and Hazen-Williams is offered where a code or a published table requires it.
How to read a friction loss chart for a water main
The formula takes four numbers. Choosing the fourth one honestly is the whole job.
Set the flow and the pipe family
Type the design flow and pick which dimensional family the pipe belongs to. The chart then covers every size that family is made in, at that one flow, so you can see the whole curve rather than one point on it — and the point where doubling the pipe area stops buying you anything is usually obvious from the column.
Pick C for the condition the system will be in, not the day it is laid
The material select loads three C values: clean and new, a design figure that allows for the life of the system, and the value NFPA 13 names where it names one. Unlined cast iron is 130 new and 100 for design, because it will get there — and it will keep going, into the 60s and 70s after forty years on aggressive water. Choose the column that matches the question, then edit the number if you have field data.
Read the validity panel before you use the number
Four conditions are checked against the inputs you gave: water between 40 °F and 75 °F, velocity at or under 10 ft/s, bore of 2 in or larger, and a Reynolds number above 4,000. A fail is not a warning about precision — it means the formula has left the data it was fitted to, and the answer it returned is a number rather than a prediction.
Technical specifications
| Equation | V = 1.318 C R^0.63 S^0.54 in US customary units, solved for S, with the hydraulic radius taken as D/4 for a full pipe |
|---|---|
| Agreement with NFPA 13 | Reproduces the code's p = 4.52 Q^1.85 / (C^1.85 d^4.87) to better than 0.5%; the residual is NFPA's rounded exponents, not a different formula |
| C factor table | 13 materials × 3 columns — new, design and NFPA 13 — from PVC at 150 down to corrugated metal at 60, each with the reason it moves |
| NFPA 13 values carried | Listed plastic 150, copper and stainless 150, cement-lined iron 140, black steel wet or deluge 120, galvanized 120, black steel dry or preaction 100, unlined cast or ductile iron 100 |
| Validity envelope | 40–75 °F water, velocity ≤ 10 ft/s, bore ≥ 2 in, Re > 4,000 — all four tested against your inputs and reported individually |
| Aging spread | Unlined cast iron falls from C 130 laid to about 100 at twenty years and 60–80 at forty, which roughly triples the loss at the same flow |
| Cross-check | Darcy-Weisbach is run on the same pipe and printed beside the answer; at a new-pipe C the two land within a few per cent, and the rest of the gap is the aging allowance |
| Privacy | Nothing is uploaded — the chart is generated in your browser and the copy button writes to your clipboard, not to a server |
Frequently asked questions
Why is a formula from 1905 still in the codes?
Because it needs one number where the physical method needs three, and that number can be inspected. C rolls the wall roughness, the bore condition and the Reynolds dependence into a single figure a person can argue about, assign to a material, and age — which is what waterworks and fire protection actually need, since both are sizing for a pipe's whole service life rather than for its first week. NFPA 13 goes further and fixes C by material and system type, so that two engineers calculating the same sprinkler system get the same answer whether or not they agree about roughness.
What C should I use for forty-year-old galvanized supply pipe?
There is no defensible single answer, and that is the honest one: galvanized is the worst ager in common building service, and old galvanized loses bore to tuberculation as well as C, so both terms of the equation move against you at once. The table here carries 120 new and 100 for design, and field measurements on old domestic galvanized fall well below that. If the number matters, measure it: put a gauge at each end of a known length, run a known flow, and back-calculate C from the drop.
Does this include losses through elbows, tees and valves?
No — the length field is straight pipe only. Hazen-Williams costs the wall, and fitting losses are added separately as equivalent lengths from a resistance table, which is a different table with a different basis. The fitting schedule lives on the pressure loss page, where every fitting's resistance is stored in pipe diameters so it converts to feet of the bore you are actually using.
Why does changing the water temperature not change the result?
Because there is no temperature anywhere in the equation, and that is the single most important thing to know about it. It has no viscosity term, so it physically cannot represent the fact that hot water is thinner and loses less head — a 180 °F heating loop and a 55 °F chilled loop return the same figure here, while Darcy-Weisbach puts the hot run about 14% lower on the same 4 in pipe at 100 gpm. The temperature field on this page changes only two things: the density used to turn feet of head into psi, and whether the validity panel tells you to stop.
Can I use it on a glycol loop?
No. Glycol changes the one property the formula has no term for: a 50% propylene glycol mix runs around 4.5 cSt at 40 °C where water is 0.66 cSt, and near 55 cSt by −20 °C, where water is ice. That is enough to move the flow out of the turbulent zone entirely in a small pipe, which is the condition C depends on. There is no C factor for glycol and there cannot be one, because C is not a property of the fluid. Work a glycol loop on the physical method with the mix's own density and viscosity from its data sheet.
The two methods disagree by 15% on my pipe. Which one is wrong?
Probably neither, and the gap is telling you where the aging allowance is. Hazen-Williams at a design C is deliberately pessimistic — it is describing the pipe in twenty years — while Darcy-Weisbach with a clean-and-new roughness is describing it today. Set C to the new column and the two normally close to within a few per cent. If they still disagree badly, look at the validity panel first: a fail on bore or on Reynolds number will open a gap that has nothing to do with aging.
Is the C for a sprinkler system a recommendation I can override?
Not for a submitted calculation. NFPA 13 names the value by material and by system type, and a hydraulic calculation performed with a different C is not a code calculation regardless of how well justified the substitution is. Note that the code's own values split black steel by service — 120 in a wet or deluge system, 100 in a dry or preaction one — because a dry system corrodes at every trip cycle, not because the steel is different.
About Hazen-Williams and the C factor
Hazen and Williams published their hydraulic tables in 1905, fitting a power law to measurements on water in ordinary pipe. The result — velocity proportional to C, to the hydraulic radius to the 0.63 and to the head loss gradient to the 0.54 — has no dimensional consistency and no theoretical standing, and it has outlived every attempt to retire it because of what C is rather than what it is not. C is a number an engineer can defend to a plan reviewer: it belongs to a material, it can be measured on an existing main by putting a gauge at each end, and above all it can be aged, so a system can be sized for the pipe it will become rather than the pipe it is.
The direction of the error is the part worth internalizing: a high C is optimistic, and an optimistic C undersizes the pipe. Two engineers can defensibly pick figures twenty apart for the same cast-iron main and the one who picks 130 will specify a smaller pipe that fails its duty in year fifteen. Plastics and lined pipe barely move, which is much of why cement-mortar lining changed water distribution practice — a lined ductile main holds its capacity for decades while an unlined one loses half of it. What C cannot do is respond to what the fluid is or how hot it is, and that boundary is where this page hands over to the pipe pressure loss calculator, which carries the viscosity term explicitly.
One condition in the validity panel deserves more than a checkbox. The requirement that flow be fully turbulent is not a nicety: C stands in for a Reynolds dependence that only exists above about Re 4,000, and below it the friction factor follows a different function altogether — what those bands are and where your pipe sits in them is the Reynolds number calculator. And none of this applies to a pipe that is not full: a sanitary branch runs partly full by design and is an open-channel problem under Manning’s equation, which the drain pipe slope calculator handles, and a tank or line volume in gallons comes from the pipe volume calculator.
What leaves your browser
Every number on this page is worked out by JavaScript running in the tab you are reading it in. Nothing you type — loads, lengths, nameplate ratings, the rates your utility charges you — is uploaded, logged or kept, which is also why the calculators carry on working in a mechanical room with no signal.