Engineering Guide

Reynolds Number in Compressible Flow: When Re Alone Fails at M > 0.3

Published July 3, 2026 · by Industrial Unit Converter Editorial Team

Reynolds Number in Compressible Flow: When Re Alone Fails at M > 0.3

A 36-inch natural gas transmission main at 1,000 psig and 100 MMSCFD has Re near 50 million and a Mach number of 0.034. The flow is deeply subsonic. Apply the same Re framework to a gas turbine first-stage blade cooling passage at 200 psig and 1,000°F, and the Mach number reaches 0.52. The friction factor barely changes, but the actual pressure drop is 17% higher than the incompressible prediction. For M above 0.3, density variation along the duct becomes a real correction.

For M < 0.3, density variation stays below 5% and the standard Re calculation is correct. The threshold comes from White (2016) and Anderson (2003). Most industrial pipe flow sits in the M < 0.1 region. The cases that exceed M = 0.3 are high-pressure natural gas with large pressure drop, gas turbine cooling passages, rocket engine channels, and choked orifices.

Speed of sound and Mach number regimes

For an ideal gas, the local speed of sound is:

$$a = \sqrt{\frac{\gamma R T}{M}}$$

where γ is the specific heat ratio (cp/cv), R is the universal gas constant (8,314 J/kmol·K SI), T is absolute temperature, and M is molar mass.

Gas	γ	a at 20°C (m/s)
Air	1.40	343
Natural gas (mostly methane)	1.31	450
Nitrogen (N2)	1.40	349
Steam (H2O vapor)	1.33	470 at 150 psig
Helium	1.66	1,007
Hydrogen (H2)	1.41	1,295

Natural gas is 31% faster than air because methane has lower molecular weight. Helium is 3× faster. Source: White (2016) Appendix A. The speed of sound is independent of pressure for an ideal gas. Real-gas effects reduce a by 5-15% above 1,000 psig. Use GERG-2008 above that pressure. The Reynolds number for air calculator uses Sutherland viscosity for air from -50 to 1000°C, paired with the speed-of-sound calculation for hot gas.

The Mach number M = V / a classifies flow into five regimes:

Regime	Mach range	Engineering treatment
Incompressible	M < 0.3	Standard Re, Moody chart
Subsonic compressible	0.3 < M < 0.8	Compressibility correction factor
Transonic	0.8 < M < 1.2	Compressible flow correlations
Supersonic	1.2 < M < 5.0	Shock waves possible
Hypersonic	M > 5.0	High-temperature real gas

The M = 0.3 boundary is conventional. The actual density variation scales as M², so M = 0.3 gives roughly 9% in the first-order density correction. The AGA uses M = 0.3 for natural gas transmission.

	M < 0.3	M > 0.3
Re < 2,300 (laminar)	Viscous laminar (most oil flow)	Viscous compressible (rare)
Re > 4,000 (turbulent)	Turbulent incompressible (most industry)	Turbulent compressible (gas turbines, high-pressure gas)

Most industrial flow sits in the bottom-left cell. The bottom-right cell is the focus: gas turbine blade cooling, rocket engine regenerative cooling, high-pressure natural gas transmission in long lines, and choked orifices at high pressure ratios. The Reynolds number for pipe flow calculator returns the incompressible Re; for M > 0.3, apply the correction manually.

Fanno flow and the compressibility correction

For flow in a constant-area duct with friction, the Fanno flow line describes how Mach number, pressure, and density change along the duct. Per Anderson (2003):

Friction factor is approximately the same as incompressible at the same Re
Pressure drop is higher because density decreases along the duct
Exit Mach number is higher than inlet Mach number for subsonic inlet
Maximum duct length is finite and ends at M = 1 (choked flow)

The Fanno line matters at high Mach and long ducts (L/D > 100). A 36-inch pipe at 100 ft has L/D = 33, well within the incompressible regime. A 1,000 ft run has L/D = 333. Gas turbine cooling passages can sit near their Fanno limit at peak power.

For M < 1, the actual pressure drop is higher than the incompressible prediction by:

$$\frac{\Delta P_{actual}}{\Delta P_{incompressible}} = \frac{1}{\sqrt{1 - M^2}}$$

Inlet M	Correction factor	Increase vs incompressible
0.2	1.021	2.1%
0.3	1.048	4.8%
0.4	1.091	9.1%
0.5	1.155	15.5%
0.6	1.250	25.0%
0.7	1.400	40.0%
0.8	1.667	66.7%

The correction grows rapidly above M = 0.5. Source: Shapiro (1953) Chapter 4. The formula applies to average Mach in the duct. For long pipes with significant pressure drop, average M is higher than inlet M. AGA Report No. 9 uses an integrated form for natural gas transmission.

Three worked examples from industry

Example 1: Natural gas transmission pipeline (incompressible)

A 36-inch (0.914 m) natural gas transmission main at 1,000 psig (1,014.7 psia), 60°F (520°R), 100 MMSCFD. Methane-dominated composition (γ = 1.31, M = 16.04).

Calculation per White (2016) and AGA Report No. 9: a = 256 m/s, ρ = 5.79 lb/ft³, A = 0.656 m², ṁ = 1,157 lb/s, V = 8.6 m/s. M = 8.6 / 256 = 0.034. Re ≈ 5 × 10⁷ (fully turbulent).

M = 0.034 is well below the 0.3 threshold. Most natural gas transmission mains in the 500-1,500 psig range fall in the M = 0.05-0.2 range. Use the SCFM to ACFM converter for actual flow at line conditions.

Example 2: Saturated steam at 600 psig (counter-intuitive, incompressible)

A 6-inch Schedule 40 steam line (D = 0.154 m) at 600 psig (614.7 psia), 50,000 lb/hr saturated steam. Saturated temperature 486°F, specific volume 0.0193 ft³/lb per ASHRAE (2024).

Calculation per ASHRAE (2024) Chapter 22: ρ = 51.8 lb/ft³, A = 0.0186 m², ṁ = 13.89 lb/s, V = 0.41 m/s. a ≈ 470 m/s. M = 0.41 / 470 = 0.00087. Re ≈ 6.1 × 10⁵.

Despite 600 psig, M = 0.00087 is essentially zero. Steam at high pressure has high density, which keeps velocity low. The incompressible assumption is excellent. The Moody chart friction factor guide covers the downstream calculation.

Example 3: Gas turbine blade cooling passage (genuinely compressible)

A modern gas turbine first-stage blade cooling passage, 0.5-inch (12.7 mm) diameter, 200 psig (214.7 psia) cooling air at 1,000°F (1,460°R), mass flow 0.5 lb/s per passage, γ = 1.36 (hot air).

Calculation per Cohen et al. (1987) Gas Turbine Theory: a = 491 m/s, ρ = 0.437 lb/ft³, A = 1.27 × 10⁻⁴ m², V = 257 m/s. M = 257 / 491 = 0.52. Re ≈ 1.05 × 10⁵.

M = 0.52 exceeds 0.3. The compressibility correction at the average M is 1/√(1 - 0.27) = 1.17, a 17% increase in pressure drop. For a blade with 200 cooling passages and a tight pressure drop budget, a 17% error is the difference between meeting the temperature target and running 25°C hot. The Reynolds number engineer complete guide covers the incompressible baseline.

Common mistakes

Mistake 1: Assuming high pressure means high Mach number. Steam at 600 psig is deeply incompressible (M = 0.00087) because high pressure gives high density, which keeps velocity low. The Mach number, not the pressure, determines compressibility importance. The cost: applying compressible flow corrections where they are not needed, or sizing relief valves based on the wrong density, with valve area off by 10-30%.

Mistake 2: Using incompressible correlations for high-Mach cooling passages. Gas turbine cooling, rocket engine cooling, and high-pressure instrument air at high flow routinely exceed M = 0.3. The standard Moody chart gives the correct friction factor at the inlet Re, but the pressure drop is higher by 1/√(1-M²). For a gas turbine blade, this is 15-20% of the friction budget. Blades run 20-30°C hotter than designed, with creep life reduction of 30-50%.

Mistake 3: Forgetting the Fanno flow limit. A constant-area duct with friction has a maximum L/D for a given inlet M. For inlet M = 0.034, max L/D is about 660. For inlet M = 0.5, max L/D is only 1.4. The mistake: designing a long pipe run without checking the Fanno limit, leading to a system that delivers less mass flow than predicted and chokes at 60-80% of design flow.

Mistake 4: Using the wrong speed of sound. Speed of sound depends on temperature, not pressure. Air at 1,000°F has a = 640 m/s, not 343 m/s. Using room-temperature a for hot process gas gives M three times too low, and pressure drop is under-predicted by 30-50%.

Mistake 5: Mixing compressibility correction with the Fanno line. The 1/√(1-M²) correction is for average M in a constant-area duct. Using it at inlet M for a long duct over-predicts pressure drop by 5-15%, and the engineer over-specifies the compressor by that margin. For a $500,000 compressor, that is $25,000-75,000 of wasted capacity.

Standards and best practices

White, F.M. (2016) Fluid Mechanics, 8th ed., McGraw-Hill. The M = 0.3 threshold and standard Re formula.
Anderson, J.D. (2003) Modern Compressible Flow, 3rd ed., McGraw-Hill. Fanno flow, Rayleigh flow, and the M = 1 choking condition.
Shapiro, A.H. (1953) The Dynamics and Thermodynamics of Compressible Fluid Flow, Vol. 1, Ronald Press. The original derivation of the compressibility correction factor.
AGA Report No. 9 (2017) Measurement of Gas by Multipath Ultrasonic Meters. Natural gas pipeline hydraulics.
Cohen, H., Rogers, G.F.C., Saravanamuttoo, H.I.H. (1987) Gas Turbine Theory, 4th ed., Longman. Gas turbine cooling passage design with Fanno integration.

When compressibility actually matters in industry

Four cases put industrial flow above M = 0.3:

High-pressure natural gas transmission with significant pressure drop. A long line at 1,500 psig with outlet pressure below 700 psig has elevated Mach near the outlet. Apply the integrated Fanno correction per AGA Report No. 9.
Gas turbine blade cooling passages. Modern first-stage blades operate at M = 0.4-0.6 in the cooling channels. The 1/√(1-M²) correction is mandatory for blade life prediction.
Rocket engine regenerative cooling. Liquid rocket engines have M = 0.3-0.5 in the cooling channels at full thrust.
Compressed air distribution at high pressure and high flow. 150 psig plant air at 5,000 SCFM in 4-inch pipe can reach M = 0.3 at peak demand. The compressed air piping pressure drop guide covers this case.

For everything else (most steam, most water, most HVAC air, most low-pressure gas, most process liquids), the M < 0.3 condition is comfortably satisfied and the standard Re + Moody chart is correct. The rule we use: if you are not designing one of the four cases above, you probably do not need compressibility corrections.

FAQ

Q: At what Mach number do I need to apply a compressibility correction?

A: Below M = 0.3, the correction is below 5% and the standard Re + Moody chart is acceptable. Between M = 0.3 and M = 0.7, apply the 1/√(1-M²) correction at the average M in the duct. Above M = 0.7, use the full Fanno line integration from Anderson (2003) Chapter 4. Above M = 1, the flow is choked and you need a compressible flow solver, not a hand calculation.

Q: Why is steam at 600 psig still incompressible?

A: Steam at high pressure has high density, which keeps velocity low. At 600 psig saturated, density is 51.8 lb/ft³ and velocity in a 6-inch pipe at 50,000 lb/hr is 0.41 m/s, giving M = 0.00087. The Mach number, not the pressure, determines compressibility importance. Use the saturated steam tables from ASHRAE (2024) or NIST WebBook for density at the operating pressure.

Q: How do I calculate speed of sound for a gas mixture?

A: Use the molar-mass-weighted γ and the mixture average molecular weight. A 95% methane, 5% ethane mixture has M = 17.0 and γ ≈ 1.30. For combustion products, use the equilibrium composition at the local temperature. GERG-2008 gives a = 450 m/s for natural gas at 1,000 psig and 60°F, about 5% below the ideal-gas value. The NCFM to SCFM converter handles standard-condition flow for gas mixtures.

Q: Does the Reynolds number change with Mach number?

A: The Reynolds number is defined at the local density and viscosity. As M increases and density drops, Re drops proportionally. The friction factor at lower Re is higher, which compounds with the compressibility correction. For the gas turbine example, inlet Re = 1.05 × 10⁵ and outlet Re (after density drop) is about 8 × 10⁴, increasing f by 5%. The combined correction can reach 20-25% for high-Mach passages.

Q: Can I use the compressibility correction with any friction factor correlation?

A: Yes. The 1/√(1-M²) factor is a separate multiplier applied to the incompressible pressure drop. The Moody chart or Colebrook-White gives f at the inlet Re; the compressibility correction modifies ΔP, not f. The combined form: ΔP_actual = f × (L/D) × (ρV²/2) × 1/√(1-M²). This is the form used in AGA Report No. 9 and in the gas turbine correlations from Cohen et al. (1987).

Q: When is the Fanno line limit reached in real systems?

A: The Fanno line terminates at M = 1. For inlet M = 0.1, max L/D is about 66 (γ = 1.4). For inlet M = 0.5, max L/D is about 1.4. Most industrial pipes are well within their Fanno limit, but gas turbine cooling passages and rocket engine channels can be near the limit at peak power. For the typical 36-inch transmission main at M = 0.034, max L/D is 660, so even 50-mile segments are well below the limit.

Related tools and calculators

References

White, F.M. (2016). Fluid Mechanics, 8th ed., McGraw-Hill.
Anderson, J.D. (2003). Modern Compressible Flow, 3rd ed., McGraw-Hill.
Shapiro, A.H. (1953). The Dynamics and Thermodynamics of Compressible Fluid Flow, Vol. 1, Ronald Press.
Cohen, H., Rogers, G.F.C., Saravanamuttoo, H.I.H. (1987). Gas Turbine Theory, 4th ed., Longman.
AGA Report No. 9 (2017). Measurement of Gas by Multipath Ultrasonic Meters, American Gas Association.
ASHRAE Handbook (2024). Chapter 22, Steam Properties, American Society of Heating, Refrigerating and Air-Conditioning Engineers.

Share LinkedIn X Email