Wind Tunnel Testing for Tall Buildings

(Last Updated On: June 14, 2018)

A Bluffer’s Guide
Wind tunnel testing is a ‘black box’ to many practicing structural engineers. This article strives to shine
a small beacon inside the box and provide the reader with a foundation for asking sensible questions of
wind tunnel laboratories. It is written from the author’s experience as a practicing wind engineer, an
advocate for wind tunnel testing on major projects, and a client in receipt of wind tunnel test reports.

A common puzzle for design teams is knowing when a wind tunnel test will be of value. As a wind
engineer, the obvious answer would be “always,” but the more honest response is “it depends.” The
typical issues for which a wind tunnel test might be commissioned for a tall building (Figure 1) include
life safety issues of accurate determination of local pressures and wind-induced structural loads and
responses. Also, typical serviceability issues of pedestrian wind conditions and building exhaust
dispersion that contribute to the public and occupant perception of the quality of the built environment
are of concern.

Figure 1. Wind tunnel pressure model.

Codes and Standards

Wind-induced structural loads and cladding pressures can normally be estimated by building codes or
standards. One regularly asked question is, “Why would a wind tunnel test be more accurate than the
code?” The simple answer is that all the values in the code were developed from wind tunnel tests and
intended to envelop the majority of cases. However, almost all codes include numerous caveats to their
applicability for unusual buildings or circumstances, where wind tunnel testing is mandated as an
alternative route to compliance. As such, the codes should provide conservative values for the majority
of ‘typical’ tall buildings in urban environments while a properly conducted wind tunnel test will give
more accurate project-specific design values. These updated values will either result in lower-than-code
values with consequent savings in construction cost, or capture unusual effects not covered by the
code to ensure adequate design reliability.

In the design process, it is very often the structural engineer who takes responsibility for recommending
whether wind tunnel testing is conducted. For many, this decision is predicated on the expectation of
the significance of wind-induced strength-design lateral loads to the design, often based on how these
might compare with the seismic lateral loads. This, however, does not take into account the savings
that can be achieved in the façade design or ensuring that occupants will not be disturbed by overly
frequent perceptible building motion.

With particularly slender buildings or those affected by local topography or excessive aerodynamic
interference from neighboring buildings, the use of codes may not achieve adequate structural
reliability. In these cases, the use of wind tunnel testing ensures that the desired levels of design
reliability and robustness are met, consistent with the structural engineer’s professional responsibility.
As performance-based design (PBD) gains more traction, wind tunnel testing will become an
increasingly important part of achieving design objectives, something that is anticipated to be covered
in an upcoming monograph from an ASCE Task Committee and an SEI Wind PBD Pre-Standard.

Test Types
There are three commonly used wind tunnel test types for the determination of wind-induced structural
loads and responses for tall buildings. These are the high-frequency balance (HFB), high-frequency
pressure integration (HFPI), and aeroelastic techniques.

The HFB and HFPI are the most common approaches and use rigid aerodynamic models. Both
measure the wind forces exerted on the building model and, for high-rise buildings, the dynamic
properties of the building are introduced mathematically into the analysis to determine the total
response to the wind loading. The HFPI approach (Figure 2) is now applied to the majority of projects,
as it uses the same model as the cladding pressure testing. Pressures measured simultaneously over
the building surface are integrated to determine the overall wind loads applied to the building. However,
for particularly architecturally complex buildings, it may not be possible to have a large enough number
of pressure taps to map the pressure fields over the building with sufficient resolution. For very tall
slender towers, the limited cross-section of the tower often provides a physical limitation to the number
of pressure tubes that can be extracted from the model at once. In this case, HFB testing is the logical
alternative.

Figure 2. Close-up of a wind tunnel pressure model.

HFB testing (Figure 3) uses a lightweight model mounted on a very stiff balance to measure the applied
forces at the base of the model. In this way, the HFB model is working as a mechanical integrator
compared with the numerical integration of the HFPI approach. As the construction of an HFB model is
less involved and more economical than a pressure model, this is also the technique that is used most
commonly early in the design process where the final architecture may not yet be complete. This model
is also easier to modify if a range of building shapes are being investigated. Shaping studies are
sometimes used during concept design of particularly slender and wind sensitive towers to optimize
building shape and minimize building responses. An appropriately designed HFB model can
incorporate a number of adjustable features to investigate various architectural changes.
Figure 3. High-frequency balance model. Courtesy of b&w structural designs, llc.

The aeroelastic approach differs from the aerodynamic model approaches in that the model
incorporates the appropriately scaled dynamic properties of the prototype structure: natural frequencies
of vibration, mass characteristics, and damping ratios. The aeroelastic approach is generally more
expensive than the aerodynamic techniques. The parameter which the aeroelastic modeling captures,
which is not measured in either the HFB or HFPI approaches, is the aerodynamic damping. For most
buildings, the aerodynamic damping is positive. This is beneficial in reducing the resonant dynamic
response of the building. However, the degree of positive aerodynamic damping is invariably much
smaller than the inherent structural damping and within the degree of uncertainty associated with the
estimate of structural damping. The aeroelastic test is more important when initial aerodynamic test
results show that there is the potential for strong cross-wind (or vortex shedding) response. As the wind
speed approaches the peak for vortex-shedding, negative aerodynamic damping is generated, thus
reducing the effective total damping of the building and increasing the building responses.

Wind Engineering Consultant

The first thing a trusted wind engineering consultant should be able to provide to designers is advice on
what wind effects should be of interest to the design team. This includes advice about what testing and
consultancy would be of value to a project, and identification of any design features that may be
particularly wind sensitive or, conversely, beneficial in the performance of the development. Early
consultation can help projects develop in a much smoother manner rather than waiting for unexpected
results when the form and structure of a building are close to being fixed.
Once a wind engineering consultant is on board and a project has reached the stage of preparing for
wind tunnel testing, then there should be regular interaction between the wind engineering, structural
engineering, and architectural teams. The architectural team is responsible for the supply of the
building geometry from which the wind tunnel test model(s) will be built. Generally, the wind
engineering consultant will take responsibility for gathering information about the surroundings to build
the proximity model. When the test models have been designed, drawings and/or 3D models should be
provided to the design team for their checking and approval. This helps to ensure that the model
reflects the current design and includes any critical changes that may have occurred since the original
issue of architectural information. This will typically happen before the physical test model is
constructed to allow for the incorporation of any modifications.

The wind engineering consultant should at all times be able to describe, and justify, the approach to
testing being used. For the design team, key issues to check are that an adequate radius of
surroundings buildings has been modeled. This is a balance of model-scale (for tall buildings this is
typically between 1:200 and 1:500 depending on the building height) and the cross-section of the wind
tunnel being used. For tall buildings, it would be normal to include a radius of at least 1200 feet around
the building, although 1600 feet is more common, and any other significant buildings outside of this
radius that would be expected to impact the flow onto the test building.

Reviewing Test Results

Most engineers do not have much exposure to wind tunnel testing and how to interpret and check
results, but there are a few resources to aid in this. The first is to make sure that the testing has been
conducted to a reasonable standard. This can be done with reference to a number of guides ranging
from the descriptive ASCE Manual of Practice No. 67 on Wind Tunnel Studies of Buildings and
Structuresto the more prescriptive ASCE/SEI Standard 49-12, Wind Tunnel Testing for Buildings and
Other Structures. A more concise document, created for design professionals working with tall
buildings, is the Council on Tall Buildings and Urban Habitat (CTBUH) publication, Wind Tunnel Testing
of High-Rise Buildings, which summarizes what should be expected from wind tunnel tests conducted
for tall buildings.

The most obvious first check is to compare the loads and local pressures with code values. This is
something that should also have been conducted by the wind tunnel laboratory and, if there are
significant differences, this should have been highlighted and explained to the design team.
Local negative (or suction) pressures, so-called “hot spots,” that are larger than code values are not
unusual in limited areas of the building. These are typically a result of very localized flow features, such
as conical vortices that result most commonly from architectural discontinuities. However, peak positive
pressures that are significantly larger than code values are a flag to raise questions, unless they can be
shown to be a result of approach wind speed increases, as from channeling between upwind buildings.

The same type of channeling can lead to increased structural loads in the along-wind direction.
However, the most common reason for high wind loads and responses of tall, slender buildings is
cross-wind response, which will often govern for buildings with a height to width ratio of greater than 5
or 6. This is not something that is covered in U.S. loading codes, but simplified estimates can be
obtained from online estimators and overseas design standards. An example of base moment response
dominated by cross-wind response is shown in Figure 4, identified by a rapid increase in the dynamic
response at a wind direction orthogonal to the load while the mean load is close to zero.

Figure 4. Graph showing strong cross-wind response.

A more common query is when loads are significantly lower than code values. This can occur when the
building is very sheltered by its neighbors. ASCE-7 has a lower limit on loads from wind tunnel tests to
account for the removal of such adjacent buildings unless it can be shown that removing such
significant sheltering buildings still results in low loads, in which case lower limits can be applied. If,
however, a wind tunnel reports loads significantly lower than the 80% cut-off used by ASCE-7, this is a
good cue for the design team to start asking questions.

Occasionally, a structural engineer will get the opportunity to compare wind tunnel results from two
different laboratories for the same building. These almost never agree exactly but should be within 10
to 15% of each other. Where differences are larger, the discrepancies are predominantly due to the
interpretation of the site wind climate. While pressures vary with the wind speed squared, dynamic
responses can vary with the wind speed cubed or greater. If there is one thing for a structural engineer
to check and understand, it is the wind engineering consultant’s interpretation of design wind speeds to
ensure that they are both scaled appropriately to any local statutory requirements and that directionality
can be rationally explained with respect to the wind climate and surrounding terrain.

Conclusion
Wind engineering is a combination of art and science, and it is important for structural engineers to
understand just enough to check that the right studies are being conducted and, if not the reasons for
unusual results, the appropriate questions to ask to elicit explanations.