Building test rig for propeller testing
Most small test rigs place the propeller with horizontal axis of rotation, which provides sufficient room in front of and downstream of the propeller. The floor should be ideally several propeller diameters away to enable undisturbed axisymmetric flow entry.
Vertical or horizontal?
Most small test rigs place the propeller with horizontal axis of rotation, which provides sufficient room in front of and downstream of the propeller. The floor should be ideally several propeller diameters away to enable undisturbed axisymmetric flow entry. With 2 meter propeller sizes this configuration could mean a structure perhaps 4 meter tall. With vertical axis of rotation and propeller blowing upwards, similar floor clearance is sufficient, with the advantage of axisymmetric flow entry and no need for floor attachment. The support frame can be significantly smaller, which is an important consideration at such scales.
Getting the dimensions right
The most important parameter which determines the size of the frame is the maximum diameter of propeller to be tested. With great diameter comes great responsibility i.e. larger thrust, torque and vibrations. These parameters generally grow with 2nd to 3rd power of diameter, so the decision between maximum diameter of 1.5 versus 2 meters could mean threefold increase in mass and cost of the test stand.
The support frame is a hexagon with outer diameter 2.6m and side length 1.3m. The structure is robust, designed to weight twice the maximum expected thrust. With such weight goal, frame structural failure is not an issue. Instead, the effort is focused on increasing the stiffness and frequency response of the frame.
The frame consists of 6 welded sides attached together by bolts. Each side weights less than 100 kg to allow easy manipulation. In reality, we needed every helping hand available when test-fitting the frame at our headquarters. After necessary modifications and making sure that everything fits, 4 people were needed to assemble the test rig at the final location.
Weight is an important safety factor. Without the ability to attach the frame to the floor (roof of a building in this case), improperly attached propeller or wrong direction of rotation could lift the whole test rig and cause a disaster.
Welded parts connecting load cells are subject to vibrations and fatigue and cannot be as massive as the rest of the frame. The design of two-component aerodynamic force balances incorporates a safety feature which prevents catastrophic failure in case of local failure of the joints between load cells.
The propeller is guarded by an expanded metal mesh to prevent objects and people entering the propeller disc area during operation.
Large propeller testing requires lots of space. Two meter propellers require more than 10m of ceiling height for indoor operation in vertical configuration. Not to say that everything light will be flying around the moment the propeller starts turning. It’s like testing full throttle of your Cessna in a general aviation hangar. Outdoors is then the only practical place to be. The weight of approximately 600kg and position on the roof of a building dictates stationary placement of the measurement rig.
The support frame is made of construction grade steel with hot dip galvanized surface which provides the best protection against corrosion and also looks nice without additional surface finish.
The electronic measurement devices are most sensitive to moisture due to rain fall and condensation. The equipment is placed in a water tight distribution box with connectors at the bottom. A watterproof tarp with a tent shape covers the balances with load cells, which have an IP65 protection on their own.
What else besides thrust do we measure?
Internet is full of homemade RC propeller test benches with kitchen scales to measure thrust and a multimeter measuring voltage and current to get power. This is great for picking the right propeller for your fancy drone, as long as the same electric motor is used. For serious propeller testing and comparison, the second and third most important parameter besides thrust is the shaft torque and rpm. Only then the non-dimensional propeller parameters can be calculated and propellers compared across different scales and power units. But not quite yet. Without knowing the air density, we could introduce an error of several percent in the thrust and torque coefficients and even figure of merit (static thrust efficiency). To establish air density, the atmospheric pressure, temperature and humidity have to be measured as well.
Last but not least we also measure the electric current and voltage. Not for propeller evaluation but to obtain information about motor efficiency by knowing the shaft output power as well as the electrical power input.
The signals from 6 load cells, rpm meter, voltage and current meter, barometer, humidity and temperature sensors are all measured by data acquisition boards developed and produced in-house at 4jtech. The measuring box is connected to a laptop using single UTP LAN cable. An interactive python based application was programmed to collect, display and log data.
The design, manufacture and assembly took about two months as planned to reach a nearly finished state. As always, the last modifications, improvements and testing took longer than expected and brought the total order-to-delivery time to around 3 months.