What volumetric efficiency actually measures
Volumetric efficiency is a filling ratio. Take the volume of air a cylinder actually inhales in one intake event, divide it by the volume the piston sweeps, and you have VE. A cylinder that swallows exactly its own swept volume is at 100%. A stock truck engine at peak torque might reach 85%; a well-developed racing engine can exceed 100% because a tuned intake tract and the momentum of the incoming column keep pushing air in after the piston has stopped descending.
The reason VE matters more than any other single airflow number is that torque follows it almost exactly. An engine makes torque by burning fuel, it can only burn as much fuel as it has oxygen for, and the oxygen arrives as air. Plot VE against rpm and you have plotted the shape of the torque curve. That is why every cam catalogue, every head comparison and every intake manifold test ultimately reduces to a VE argument.
The word volumetric is doing real work. VE is a volume ratio, not a mass ratio, and volume depends on where you measure it. This calculator uses the convention a speed-density engine controller uses: actual air volume evaluated at manifold pressure and manifold temperature. That is the definition Heywood gives when it is referenced to inlet conditions, and it is the one that keeps a boosted engine's VE sensibly near 100% instead of reporting 220%.
The formula, and where 3456 comes from
Start with the denominator. A four-stroke cylinder fills once every two crank revolutions, so in one minute at N rpm the engine sweeps CID × N ÷ 2 cubic inches. Divide by 1,728 cubic inches per cubic foot and you get cubic feet per minute:
V̇d = CID × rpm ÷ 3456
The 3456 is nothing more mysterious than 1728 × 2. A two-stroke fills every revolution, so its divisor is 1728 and it displaces twice the airflow of a four-stroke of the same size at the same rpm. That single factor of two is why a 250 cc two-stroke and a 500 cc four-stroke sit in the same airflow class.
The numerator is the measurement. If your number is already a volume flow in CFM, use it directly — but it has to have been measured at the same pressure and temperature you enter, not corrected to standard air. If your number is a mass flow, which is what a MAF sensor and most dyno air meters give you, convert it with the ideal gas law:
ρ = 144 · MAP ÷ (53.353 · (IAT + 459.67))
Here 53.353 is the specific gas constant for dry air in ft·lbf per lbm per °R, 144 converts psi to lbf per square foot, and 459.67 converts °F to absolute Rankine. Feed it 14.696 psia and 60 °F and it returns 0.07633 lb/ft³, the familiar standard-air density. Then V̇a = ṁ ÷ ρ, and VE is the ratio of the two flows.
Notice what is missing: nothing about the cam, the heads, the exhaust or the throttle appears anywhere. VE is a measurement, not a model. Those parts determine the answer; they do not enter the arithmetic.
Worked example: 350 in³ at 6,000 rpm reading 38 lb/min
You log a small-block on the dyno at wide-open throttle. The MAF-derived airflow peaks at 38 lb/min at 6,000 rpm, manifold absolute pressure is 14.0 psia, and intake air temperature in the plenum is 100 °F. Displacement is 350 in³.
- Absolute temperature. 100 + 459.67 = 559.67 °R.
- Charge density. ρ = 144 × 14.0 ÷ (53.353 × 559.67) = 2,016 ÷ 29,860 = 0.06751 lb/ft³. That is 88% of standard-air density — the throttle and the hot underhood air have each cost you something.
- Actual airflow. V̇a = 38 ÷ 0.06751 = 562.9 CFM.
- Displacement airflow. V̇d = 350 × 6,000 ÷ 3,456 = 2,100,000 ÷ 3,456 = 607.6 CFM.
- Volumetric efficiency. VE = 562.9 ÷ 607.6 = 0.9263 = 92.6%.
Now do the same sum against standard air, which is what a flow-bench habit tempts you into. Standard CFM = 38 ÷ 0.07633 = 497.9, and 497.9 ÷ 607.6 = 81.9%. The two answers are both correct and they differ by nearly eleven points. The gap is exactly the density ratio: 0.06751 ÷ 0.07633 = 0.8845, and 92.6 × 0.8845 = 81.9. Whenever two people argue about a VE number, this is usually what they are arguing about.
Finally, sanity-check the power. At 12.5:1 air-fuel ratio the engine burns 38 ÷ 12.5 = 3.04 lb of fuel per minute, or 182.4 lb/hr. At a brake specific fuel consumption of 0.50 lb per horsepower-hour that supports 182.4 ÷ 0.50 = 365 hp. Both assumptions are gasoline wide-open-throttle rules of thumb; check your own with the brake specific fuel consumption calculator.
How to read the number you get
Read VE at one rpm only as a point on a curve. The number you care about is peak VE and the rpm it occurs at, because that is where peak torque sits. Heywood gives typical maximum volumetric efficiency for naturally aspirated spark-ignition engines as roughly 80–90%, and that remains a fair yardstick for a production engine measured at its torque peak. Anything materially below that band at wide-open throttle points at a restriction: an undersized throttle body, a clogged filter, a cam too small for the rpm, or an exhaust with too much backpressure.
Above the band, be suspicious before you are pleased. A manifold-referenced VE over about 115% on a four-stroke is rare, and by far the most common cause is an arithmetic one — displacement entered in cc when the field wanted cubic inches, g/s entered as lb/min, or gauge pressure entered where absolute pressure was asked for. Rule those out before you credit the cam.
The shape matters as much as the peak. A VE curve that is flat and broad is what a street engine wants; a curve with a tall narrow spike is what a race engine trades for. Cam timing moves the peak: later intake valve closing pushes the VE peak up the rev range and hollows out the bottom end, which you can see directly in the intake valve closing figure from the camshaft duration and overlap calculator.
Also treat VE as a diagnostic for your own speed-density tune. If your controller's VE table has to be filled with values that jump around between adjacent cells, the table is absorbing an error somewhere else — a leaking injector, an unmetered air leak, or a MAP sensor reading a pulsating plenum. A real VE surface is smooth.
Displacement airflow at 100% VE, four-stroke (CFM)
| Displacement | 3,000 rpm | 4,500 rpm | 6,000 rpm | 7,500 rpm |
|---|---|---|---|---|
| 122 in³ (2.0 L) | 105.9 | 158.9 | 211.8 | 264.8 |
| 183 in³ (3.0 L) | 158.9 | 238.3 | 317.7 | 397.1 |
| 244 in³ (4.0 L) | 211.8 | 317.7 | 423.6 | 529.5 |
| 302 in³ (4.9 L) | 262.2 | 393.2 | 524.3 | 655.4 |
| 350 in³ (5.7 L) | 303.8 | 455.7 | 607.6 | 759.5 |
| 427 in³ (7.0 L) | 370.7 | 556.0 | 741.3 | 926.6 |
| 502 in³ (8.2 L) | 435.8 | 653.7 | 871.5 | 1089.4 |
Two-stroke engines double every figure in this table, because they fill once per revolution rather than once per two.
Mistakes that produce a wrong VE
- Using gauge pressure instead of absolute. A boost gauge reading 10 psi means 24.7 psia in the manifold. Enter 10 and you will overstate VE by roughly a factor of 2.5.
- Mixing a standard-air CFM with a manifold-condition denominator. Flow-bench and compressor-map numbers are usually referenced to standard or ambient air; MAP-and-IAT density is not. Pick one reference and say which.
- Measuring intake air temperature in the wrong place. A sensor in the cold-air tube reads ambient; the charge that reaches the valve has picked up heat from the manifold. On a hot engine the difference is large enough to move VE by several points.
- Quoting VE at part throttle. Below wide-open throttle, VE mostly measures how far the throttle is shut. It describes the engine's airflow ceiling only at full load.
- Forgetting the two-stroke divisor. Using 3456 for a two-stroke doubles the reported VE.
- Ignoring humidity. The ideal-gas density here is for dry air. Humid air is less dense, so on a wet day a dry-air density slightly overstates the mass present and understates VE. The error is small — under 2% in ordinary conditions — but it is systematic.
- Trusting a MAF outside its calibrated range. A mass-air meter that has been clamped, rescaled or fitted with a different intake tube reports a curve nobody has validated. Verify against a fuel-flow check before you build a VE table on it.
Two conventions, both called VE
Manifold-referenced VE divides actual volume at manifold density by swept volume. Ambient- or standard-referenced VE converts the same mass to volume at a fixed reference density first. The two agree exactly when the manifold sits at the reference condition and diverge in proportion to the density ratio everywhere else. This page reports both — the primary figure is manifold-referenced, and VE referenced to standard air uses 14.696 psia and 60 °F dry. Note that SAE J1349, the engine power test code, uses a different reference again (29.23 inHg of dry air at 77 °F), so a J1349-corrected figure will not match either column exactly.
Where VE fits among the other airflow tools
VE is the hinge between measurement and specification. Once you know it, you can size everything downstream. Multiply displacement airflow by VE and you have the CFM a carburettor or throttle body must pass — the basis of the carburettor CFM calculator. Divide the per-cylinder share of that flow by the port's minimum cross-section and you have port velocity, which is what the intake port velocity calculator does. Convert mass flow to fuel demand and you get injector size, which the fuel injector size calculator handles.
Working the other way, VE is what you predict when you choose parts. A flow bench gives you a head's CFM at a fixed depression, and the cylinder head airflow horsepower calculator turns that into a power ceiling. A compressor map gives you a turbo's mass flow at a pressure ratio, which only becomes meaningful once you know what VE the engine will run at that boost level — see the turbocharger compressor airflow calculator.
One limitation worth stating plainly: this calculator reports VE, it does not predict it. Predicting VE from geometry requires wave-action simulation of the intake and exhaust tracts, which is the subject of Blair's work and of every commercial engine-simulation package. What you get here is an honest measurement from numbers you can actually log, which is usually the more useful thing.
Key terms
- Displacement airflow
- The volume flow the swept volume alone would produce at a given rpm, ignoring how well the cylinder actually fills. The denominator of VE.
- MAP
- Manifold absolute pressure — plenum pressure measured against a vacuum, not against atmosphere. Roughly 14.0 psia at wide-open throttle without boost, and 14.7 psi higher than a boost gauge reads.
- Charge density
- Mass of air per cubic foot in the manifold. Rises with pressure, falls with temperature, and scales VE directly when you convert between mass and volume flow.
- Speed density
- A fuelling strategy that infers airflow from rpm, MAP, IAT and a stored VE table, rather than measuring it with a mass-air meter.
- Delivery ratio
- The two-stroke analogue of VE — delivered charge mass divided by a reference mass. Conventions differ over whether the reference is swept or trapped volume, so state which you mean.