Wind Tunnel Testing Guide — When Required, ASCE 7 Chapter 31 Procedure & Data Interpretation

Wind tunnel testing provides the most accurate wind loads for complex buildings. While ASCE 7-22 Chapters 27–30 cover the vast majority of regular buildings, certain conditions demand a site-specific wind tunnel study per Chapter 31. This guide explains when testing is required, how boundary layer wind tunnel tests are conducted, what the engineer receives, and how to interpret the data for structural design.

Related pages: Wind Load Calculation Example | Wind Load Basics | ASCE 7-22 Wind Load Guide | Wind Drift Design Guide | High-Rise Steel

When Wind Tunnel Testing Is Required

ASCE 7-22 Chapter 31 Triggers

Per ASCE 7-22 §31.1.1, wind tunnel testing is required or recommended when any of the following conditions apply:

The H/B > 5 Rule of Thumb

For rectangular buildings with an aspect ratio H/B > 5, cross-wind excitation from vortex shedding often produces base overturning moments 30–100% higher than the along-wind Chapter 27 values. Even if the along-wind drift satisfies H/400, the cross-wind mode may govern strength and requires wind tunnel verification.

The Boundary Layer Wind Tunnel

How It Works

A boundary layer wind tunnel (BLWT) simulates the atmospheric boundary layer — the layer of wind from the ground up to the gradient height (~1,000–1,500 ft) where wind speed and turbulence profiles match the target site conditions. Key components:

1. Upwind fetch: Roughness elements (blocks, spires) on the tunnel floor generate the correct velocity profile (power-law exponent alpha) and turbulence intensity profile matching the target Exposure Category.

2. Turntable: The model sits on a rotating turntable, allowing 36 wind directions (every 10 degrees) to be tested. For each direction, pressures are measured on all building surfaces.

3. Instrumentation: Pressure taps (~300–500 for a typical high-rise model) connected to electronic pressure scanners via short tubing. Sampling rate: 500 Hz for 60–90 seconds per direction (equivalent to 1 hour full-scale).

4. Reference instrumentation: A pitot-static tube measures the reference wind speed and pressure above the boundary layer for normalization.

Model Scale

Typical model scales are 1:300 to 1:500. A 600 ft tall building at 1:400 scale becomes an 18-inch model. The model must include:

• The subject building with accurate external geometry
• All significant nearby buildings within a 1,500–2,000 ft radius (at full scale)
• Topographic features within 1 mile if site is not flat
• The building's surface texture (balconies, mullions) if they affect pressure — typically modeled with equivalent roughness strips

Types of Wind Tunnel Tests

1. High-Frequency Force Balance (HFFB)

The most common method for preliminary structural design. The entire model is mounted on a high-stiffness, high-sensitivity force balance that measures the base overturning moment, base shear, and base torque in real time at 500 Hz.

Advantages: Fast (1–2 days of tunnel time), cost-effective ($20,000–$40,000 for a typical test), produces modal forces directly usable by the structural engineer.

Limitations: Cannot distinguish between windward and leeward surface pressures, assumes linear mode shapes (reasonable for most buildings under 60 stories), does not provide local C&C pressures.

2. Simultaneous Pressure Integration (SPI)

Also called the High-Frequency Pressure Integration (HFPI) method. The model is instrumented with 300–500 pressure taps covering the entire building envelope. Pressures at all taps are measured simultaneously (500 Hz), then integrated over the building surface to produce floor-by-floor forces and torques.

Advantages: Produces story forces directly (no assumptions about mode shape), identifies local pressure hot spots for cladding design, provides pressure coefficients usable with any structural configuration.

Limitations: More expensive ($50,000–$100,000), 2–3 weeks of tunnel time plus analysis, requires a model with internal pressure tubing.

3. Aeroelastic Model

A flexible model with scaled stiffness and mass distribution that physically vibrates in the wind tunnel, directly measuring the building's dynamic response including aeroelastic effects (motion-induced forces, aerodynamic damping). Required for buildings where the structural response significantly alters the wind loading (H/B > 8, or fundamental period > 6 sec).

Advantages: Captures aeroelastic effects that HFFB/SPI cannot, including negative aerodynamic damping that can drive large-amplitude across-wind vibration.

Limitations: Most expensive ($150,000–$300,000+), requires detailed structural design to be substantially complete, each structural scheme requires a new model.

Wind Tunnel Test Procedure

Step 1 — Meteorological Wind Climate Analysis

Before the tunnel test, the wind climate at the site must be characterized. This involves:

• Obtaining historical wind data from the nearest airport station (typically 20–30 years of hourly data)
• Fitting a Weibull or Gumbel distribution to annual maximum wind speeds by direction
• Establishing the design wind speeds for each return period (10-year for drift, 50-year for strength, 700-year for ultimate per ASCE 7-22)
• Accounting for the site's anemometer height, terrain roughness, and averaging time (converting 1-hour airport data to 3-second gust speeds)

Step 2 — Model Construction

A rigid foam or 3D-printed model is constructed at 1:300–1:500 scale. For SPI models, the model interior contains 300–500 plastic pressure tubes (1–2 mm diameter) running from surface taps to a connector board beneath the turntable. Tube length is minimized (< 24 in) and corrected for frequency-dependent attenuation using digital transfer functions.

Step 3 — Tunnel Testing

For each of 36 wind directions (0–350 degrees at 10-degree increments):

• Reference wind speed is recorded for normalization
• All pressure taps or force balance channels are sampled at 500 Hz for 60–90 seconds
• Data is split into overlapping windows for statistical analysis

Step 4 — Data Processing

The raw pressure data is processed into engineering coefficients:

• Mean pressure coefficients (Cp_mean) for each tap at each wind direction
• RMS (fluctuating) pressure coefficients (Cp_rms) for dynamic analysis
• Peak pressure coefficients (Cp_peak) using the peak factor method or observed extreme values
• Integrated base moments (Mx, My, Mz) for HFFB tests, resolved into along-wind, cross-wind, and torsional components

Step 5 — Structural Load Derivation

The tunnel data is combined with the wind climate statistics to produce:

• 50-year design loads: For strength design of members and connections
• 10-year drift loads: For serviceability and occupant comfort
• 1-year acceleration: For occupant comfort assessment
• Directionality factors: Actual Kd values by wind direction — often different from the generic 0.85 in Chapter 27

Interpreting Wind Tunnel Results

Design Pressure Coefficients

The tunnel report provides Cp values as a function of:

• Wind direction (10-degree increments)
• Building surface (each face, roof zones)
• Tap location (floor-by-floor elevations)

A typical output for the windward face of a 600 ft building might show:

Peak pressures are typically 2.0–2.5× the mean pressure, depending on the peak factor (gp = 3.5–4.0) and turbulence intensity.

Cross-Wind Response

For a slender building (H/B = 8), the tunnel may show cross-wind base moments exceeding along-wind moments by 50–100% at certain wind speeds — a result of vortex shedding locking onto the building's natural frequency. The design engineer must check:

• Cross-wind base shear > along-wind base shear? → Include both directions simultaneously (100% in one direction + 30% in orthogonal per common practice).
• Torsional moment exceeds 10% of overturning? → Accidental torsion (minimum 5% eccentricity) may not capture the true torsional response; use the tunnel's measured Mz.

The Acceptance Criteria Problem

Wind tunnel test results are typically 15–40% lower than ASCE 7-22 Chapter 27 directional procedure results for regular buildings, which raises the question: How low can you go?

Per ASCE 7-22 §31.4, wind tunnel results shall not be less than 80% of the Chapter 27/28 results for MWFRS, nor less than 80% of Chapter 30 results for C&C. This 80% floor prevents unconservative surprises from tunnel measurement errors, scaling effects, or wind climate mischaracterization. Most peer review panels also require an 80% minimum regardless of code language.

What the Engineer Receives — A Typical Wind Tunnel Report

Report Contents

1. Executive Summary: Key base shears, overturning moments, and peak accelerations for the governing wind directions.
2. Wind Climate Analysis: Directional wind speed distributions, return period curves, turbulence intensity profiles.
3. Pressure Coefficient Appendix: Full Cp tables for all 36 wind directions and all tap locations — typically 50–100 pages.
4. Floor-by-Floor Force Tables: Story shears (Fx, Fy), story torques (Mz), and eccentricities for structural analysis input.
5. C&C Pressure Tables: Zone-specific peak pressures (positive and negative) for cladding, girt, and purlin design per effective wind area.
6. Acceleration Predictions: Peak accelerations at the top occupied floor for 1-year and 10-year return periods, compared with ISO 6897 comfort criteria.
7. Cladding Wind Load Maps: Visual maps of peak positive and negative pressures on each building face.

Questions the Structural Engineer Should Ask

When reviewing a wind tunnel report, verify:

• Is the model scale adequate (≥ 1:500) and are blockage corrections applied (< 5% tunnel blockage)?
• Are the nearby buildings modeled to a radius of at least 1,500 ft at full scale?
• Is the wind climate data from a station within 20 miles and with at least 20 years of record?
• Are pressure tap densities adequate (≥ 1 tap per 500 sq ft of building surface)?
• Are the peak pressures derived using the correct peak factor (gp) considering the building's natural frequency?
• Does the report explicitly state the 80% floor comparison per §31.4 for MWFRS and C&C?

Frequently Asked Questions

How much does a wind tunnel test cost and how long does it take?

A basic HFFB test costs $20,000–$40,000 with a 3–4 week turnaround. An SPI test costs $50,000–$100,000 with 6–8 weeks from model construction to final report. An aeroelastic model test costs $150,000–$300,000+ with 3–4 months. Cost is small relative to the structural steel savings: a tunnel test showing 30% lower loads than Chapter 27 on a 40-story tower can save $500,000–$1,500,000 in steel tonnage and foundation costs.

Can I use wind tunnel data from a similar building instead of testing mine?

Generally no. ASCE 7-22 §31.1.2 requires the test to be for the specific building and site. Even buildings with identical geometry in different cities experience different wind climates. If your building is nearly identical to one previously tested at the same site, discuss with the wind tunnel laboratory — they may accept the previous test with a site-specific wind climate overlay, saving cost.

What wind speed averaging time does the wind tunnel use — does it match ASCE 7's 3-second gust?

Wind tunnel measurements are typically converted to full-scale equivalent pressures using the mean hourly wind speed at the building height. The conversion to 3-second gust for comparison with ASCE 7 involves the gust factor approach: q3-sec = (mean hourly q) × (gust factor)² where the gust factor for a 3-second gust is approximately 1.52–1.67 for typical terrain. The wind tunnel laboratory handles this conversion; confirm the methodology in the report.

How do I handle the case where the wind tunnel shows loads HIGHER than Chapter 27?

If wind tunnel loads exceed Chapter 27 values (e.g., for a complex-shaped building in a channeled urban site), the tunnel results govern per §31.4 — there is no code-based upper cap. Design the structure for the higher tunnel loads. This is the scenario wind tunnel testing is designed to catch: complex aerodynamics that produce loads beyond the conservative but simplified Chapter 27 approach.