How a Custom Wind Tunnel Project Is Delivered

Buying a wind tunnel is not like buying equipment. It is closer to commissioning a small industrial plant that happens to produce air of a specified quality — a machine, a building and a measurement system that have to be designed as one thing and handed over as one thing.

Most of the pain in these projects comes from the same place: decisions that look administrative are actually technical, and they are taken too late. This article sets out the sequence we use, what the client needs to bring to each stage, and where schedules break.

1. The six phases

1. Needs analysis. Not "what tunnel do you want" but "what has to be measured or achieved". Test objects, speed range, required flow quality, instrumentation, throughput, growth plans. This is also where the tunnel type is settled — see our

comparison of wind tunnel types

2. Concept and aerodynamic design. Circuit layout, section geometry, contraction ratio (how much the flow narrows and speeds up before the test section), the corner vanes that turn the airflow, fan selection, and a pressure loss budget — an accounting of where energy is lost around the circuit, which sets the drive power needed. The output is a concept that fits your building and hits your specification — or an honest statement that the two are incompatible, which is far cheaper now than later.

3. Mechanical and climate engineering. Structure, ducting, drive train, cooling strategy, acoustics, safety systems and control. Here the tunnel stops being aerodynamics and becomes an engineered product with schedules and suppliers.

4. Manufacturing and integration. Fabrication of ducts, vanes, contraction, fan assembly and the test section, plus integration of instrumentation and control systems — with factory testing wherever it is possible to test before the parts are on site.

5. Installation and commissioning. Assembly, alignment, electrical connection, sequential start-up, flow-quality measurement and correction, then acceptance testing against the contract.

6. Training, service and upgrades. Operator and maintenance training, spare parts strategy, service intervals, and the upgrade path that keeps the facility relevant for decades.

2. What the client has to prepare

The single biggest determinant of schedule is how quickly this information becomes available and reliable:

Site data. Clear internal dimensions and heights, column grid, floor load capacity, foundation conditions, crane access, door sizes and delivery routes. Oversized sections have to physically reach the room.
Electrical power. Available capacity, transformer arrangement, distance to the substation, and permissible harmonic distortion. Grid connection is often the longest-lead item in the entire project and it is rarely on the buyer's critical path list at the start.
Cooling and water. Availability, climate data for the site, and whether passive ventilation is viable — the decision that drives operating cost more than any other.
Permits and constraints. Noise limits at the boundary, working hours, environmental approvals, and any structural approvals the installation triggers.
Test requirements. Speed range and stability, required flow quality, section size, instrumentation, data systems, and the standards your results have to satisfy.
Operations. Who will run it, how many hours per day, and what skills exist in-house.

A supplier who does not ask for all of this is quoting a price, not a project.

3. Where schedules actually break

Power connection. Utility lead times routinely exceed manufacturing lead times. Start this first, not after design freeze.
Civil works dependencies. Foundations, penetrations and structural reinforcement have to be finished before mechanical installation starts. A two-week civil delay becomes a two-month installation delay when a crane window is missed.
Logistics of oversize parts. Duct sections, fan housings and contraction segments are large. Route surveys, permits and lifting plans are project work, not shipping details.
Late instrumentation decisions. Balances, traverses, optical access and seeding ports have to exist in the aerodynamic design. Adding them during installation means cutting into finished ducts.
Undefined acceptance criteria. If the contract does not say how flow quality will be measured, handover becomes a negotiation instead of a test.

4. What belongs in the acceptance criteria

Every performance claim in the contract should carry a measurement method, a location and a tolerance. At minimum:

Criterion	Define
Speed range	Minimum and maximum, with stability tolerance at both ends
Flow uniformity	Deviation across the section, measurement grid, plane of measurement
Turbulence intensity	Value, frequency band, instrument and measurement position
Temperature control	Steady-state target, allowable drift over a run
Noise	Level, position, operating condition
Availability	Uptime target and the conditions under which it is assessed
Instrumentation	Accuracy and uncertainty, calibration responsibility
Training and documentation	Scope, language, number of personnel
Spare parts	Critical items list and delivery commitments

The purpose is not legal protection. It is to make sure the person specifying and the person building have the same picture of "finished".

5. Single-point responsibility versus split packages

Splitting a tunnel into separate contracts — building here, aerodynamics there, electrical elsewhere — usually looks cheaper in tender and becomes more expensive in reality, because the interfaces are exactly where the risk lives. When flow quality is out of specification, the aerodynamic contractor points at the building, the building contractor points at the installation, and the client owns the problem.

Single-point responsibility means one party carries the performance guarantee for the finished, commissioned facility, and that party has the authority to make the interfaces work. It is the model we build under: concept design, aerodynamic layout, manufacturing, installation, commissioning and lifecycle support, with clear interfaces to the client's civil and electrical scope.

6. Timeline and budget expectations

Budget structure and the share taken by civil works are covered in

our CAPEX breakdown

, and for existing facilities the alternative path is set out in

retrofit versus new build

. Two principles hold in every project:

Design time is the cheapest time. Every change is an order of magnitude cheaper in phase 2 than in phase 5.

Commissioning is not a formality. Flow quality is measured, corrected and re-measured. Facilities that skip this arrive at a number on paper and a different number in the section.

7. Starting properly

The most useful first conversation is short: what has to be tested, at what speed, in what building, with what power available, and by when. From that we can produce a concept layout, a drive power estimate and a realistic programme — before anyone writes a tender document that locks in the wrong architecture.

Send us those five inputs and we will come back with a facility concept. Related reading: our article on

accepting a wind tunnel and verifying it meets spec

Sources and notes

The six-phase delivery sequence — needs analysis, concept and aerodynamic design, mechanical and climate engineering, manufacturing and integration, installation and commissioning, training and service — is TunnelTech's project delivery model.
Schedule risks, client input requirements and acceptance criteria reflect our project experience across research, sport and defence installations; specific lead times vary by country, utility and site.
Photographs are from TunnelTech installation and service projects.