A Force Balance to Measure Lift and Drag in Small-Scale Wind Tunnels

Commercial wind tunnels are important for aerodynamics research yet are expensive to use. A popular alternative, Computational Fluid Dynamics (CFD) simulations are not perfect; physical experimentation is still needed to validate CFD data.

A key solution to this issue are the small-scale wind tunnels often used by aerospace students and researchers, but they typically lack a low-cost yet accurate way to collect lift and drag measurements.

This project aims to solve this problem by creating a force balance that can measure lift-to-drag ratios in small-scale wind tunnels. Two load cells were used for the data collection system, one for lift and one for drag, and extensive calibration was done to ensure accuracy and results were compared to simulations. The device can also vary the angle of attack tested.

Designs, code, and instructions are provided if you would like to build your own!

Tools:

1. A laptop for data collection and analysis
2. M4 allen key
3. Pliers
4. Wire strippers
5. 1 Soldering iron
6. 3D printer
7. Drill + drill bits
8. Fan (the faster the better) + small scale wind tunnel to conduct testing in (designs can be modified to fit a variety of small-scale wind tunnels, the one I used was around 6.5 feet long with a 10 in x 10 in x 12 in test section)
9. Protractor
10. Small level
11. Epoxy

Bill of Materials:

1. PLA filament
2. 8 Nylon M4 bolts + nuts (25 mm)
3. 24 metal M4 bolts
4. 2 long 100 mm M4 bolts
5. 2 #6x5/8" wood screws
6. 30 M4 nuts
7. 6 M4 washers
8. 2 (1 KG) load cells + 2 HX711 amplifiers
9. Breadboard
10. 10 jumper wires
11. Solder
12. Arduino Uno or Elegoo equivalent Microcontroller w/ cable
13. 1.5' x 2' wooden board for mounting
14. small Anemometer
15. M4 headset insert soldering iron tip adapter
16. Painters Tape
17. Set of 10 calibration weights (10g each)
18. Twine

Wire the two load cells and the HX711s to the Arduino microcontroller according to the diagram above. I've attached pictures of the HX711's if you need to clarify the correct attachments. You will need to solder the load cell wires to the E+, E-, A-, and A+ terminals on the HX711s and solder pin headers to the GND, DT, SCK, and VCC terminals in order to attach jumper cables. Sorry I can't go into more detail here, but you can find HX711 load cell wiring and/or soldering tutorials online if necessary!

Most of this device is made of 3D printed parts (except for the electronics.) 3D print the following parts:

1. 1 "bracket" at 30% infill
2. 1 "Lift Mount" at 30% infill
3. 1 "bottom mount sting" at 70% infill
4. 1 "AoA pivot" at 100% infill (print horizontally)
5. 1 "load cell calibration rig" at 10% infill - depending on the size of your printer you may have to split this print into two pieces. I recommended cutting it in the middle and adding dowel connections in your slicer before printing.
6. 3 "calibration pulleys" at 10% infill
7. 4 "lift pulley mounts" at 10% infill
8. 1 "drag pulley mount" at 10% infill
9. 1 "NACA 2412" at 3% infill (wing for testing)
10. 1 "NACA 4412" 3% infill (wing for testing)
11. 1 "NACA 0012" 3% infill (wing for testing)

Assemble the lift and drag data collection device and the calibration rig as follows and according to the images above:

Data collection system:

1. The bracket, bottom mount sting, and lift mount attach to the lift (oriented vertically and attached to the 8 and 11 pins on the Arduino) and drag (oriented horizontally and attached to the 6 and 3 pins) load cells using the 8 25 mm nylon bolts. Grease can help if your screws are struggling to go in.
2. The AoA pivot is attached to the top of the bottom mount sting using a metal M4 bolt and nut. Loosen the nut to change the angle of attack and tighten to keep the wing at the desired angle.
3. Screw the force balance into the very front end of the wooden board (to leave room for the calibration rig). Drill two holes (slightly smaller than the wood screws) aligned with the holes in the lift mount base and use a screwdriver to tighten the wood screws so that the assembly is firmly attached to the wooden board and won't move during testing.

Calibration rig:

1. (If needed) Glue the two halves of the calibration rig together using epoxy.
2. The long slots running along the top and sides of the rig are where the pulleys to calibrate the lift and drag load cells will be mounted. The 4 lift pulley mounts will be used to attach the 2 lift pulleys to the rig (one in front and one in back). Use two metal bolts, two washers, and two nuts to secure each mount to the rig, and one long metal bolt to support the pulley in between. Reference the 2nd photo above!
3. The placement of the pulleys can be adjusted by sliding them along the slots, and the mounts may need to be rotated or reoriented based on the rest of the assembly. The 2 lift pulleys should be in line with each other. The drag pulley should be attached to the rig using the drag pulley mount and another long bolt. When calibrating, the top of the drag pulley itself should be the same height as the wing you are testing.
4. Use a countersink drill bit (or two differently sized drill bits) to make two clearance holes for M4 bolts in the bottom of the wooden board. They should align with two of the holes on opposite sides of the bottom of the calibration rig, and make it so the calibration rig is a far forward on the board as possible without touching the data collection system. Secure the rig in place with two bolts facing upwards, with nuts on top of the flanges on the calibration rig.

Angle of attack measurement device:

1. Tape the small level to the bottom right corner of the protractor, making sure it is level when the protractor is flat against the ground.

To start, use the smallest drill bit you have to drill two holes in the top of each wing. The holes should be about a 1/4 inch apart on the top of the airfoil, directly opposite from the rectangular hole where the wing is mounted. Drill all the way through the airfoil, so from the top you can see through the two holes to the other side. If you are using a different model, you may want to attack you twine in a different way, such as by tying it around a part of the model or drilling the hole in a different place.

Then, thread the twine through one hole at the top and back out the other. Tie a knot as close to the airfoil as possible, and cut the twine to be about a foot long. Make a small loop at the other end of the twine to hang the weights off of. I calibrated each airfoil but if you feel this is unneeded for your purposes you can just do this with one. Now mount the airfoil on top of the data collection system by pushing it down onto the angle of attack pivot, and make sure that the airfoil is perfectly flat, at 0° angle of attack.

Use the calibration sketch code below to calibrate the airfoil. Start by calibrating the lift load cell. Place the twine that is attached to the airfoil (or your model) on the two pulleys that are on top of the calibration rig. Run the code, and add a weight to the loop at the end of the twine each time it says to do so and wait while the measurements are taken. There should be ten weights total, with weight one being the hook itself. Start the calibration with NO weights on the string and only add weights when it says so! If you notice that the weights are swinging excessively or are touching a part of the assembly when the code prints that it is now reading the load, I would recommend restarting the calibration. At the end, the code will print out an array with the lift and drag measurements at each weight. Copy it into this template spreadsheet (make a copy), under "Calibration: Pure Lift Applied" and "lift" and "drag" readings. You should start to see numbers populate in the matrix to the right.

Repeat the calibration process with the drag load cell (shown in the picture above). Use the same string and code setup but this time place the twine onto the pulley on the right side of the calibration rig. This will ensure that the force will be applied horizontally in order to properly calibrate the drag load cell. Copy and paste the final measurements under "Calibration: Pure Drag Applied". You should see the final matrix and inverse matrix values populate, and can test the accuracy of the calibration using the tester.

For the most accuracy, repeat this process for each airfoil or object you are planning to test, creating a new copy of the template spreadsheet for each one.

Use the anemometer to measure the wind speed in your wind tunnel test section (in m/s). Plug this value into the cell directly to the left of "max wind speed". This will be used to find the lift and drag coefficients for your model.

Detach the calibration rig from the wooden mounting board to prepare the system for data collection. Place the calibration rig underneath your wind tunnel, so the angle of attack pivot sits about halfway up your test section. You can raise and lower the mounting board by placing books underneath, and you will likely have to cut a small hole out of the bottom of the test section for the mounting rod (sting) to fit through. You also might want to put a weight on the mounting board to keep it from shifting. Place the wing or model you are testing on top of the angle of attack pivot, and use the angle of attack measurement device to adjust your wing to the desired angle. Hold the protractor at eye level, using the level to make sure it is parallel to the ground. Mark a line through the center of your model, and use the angle marks on the protractor and that line to determine what angle you are at.

Seal up your wind tunnel, and run the "Load_cell_data_collection" sketch below. Follow the instructions printed to the serial monitor. To begin, it will ask you to start the wind tunnel fan. After 20 seconds (to allow fan to reach top speed) it will take 100 lift and drag measurements and output the average of each. Copy and paste the data into the same template spreadsheet as before, under "lift" and "drag". The "Sting lift" and "sting drag" are the lift and drag created by the mounting rod. You can adjust for this by repeating the data collection process with only the sting in the wind tunnel (no model) and then subtracting the lift and drag it experiences (under "Lift - Sting Lift" and "Drag - Sting Drag") from the raw data collected in any other trials. The values in these columns will be used for all other calculations. I conducted 3 trials for 5 different angles of attack for each of my airfoils, but adjust the spread sheet if you use a different method. The raw data (- the forces on the sting) will be corrected using the inverse matrix you created during calibration, and after 3 trials at the same angle of attack the average of each will be shown. The measured lift coefficient, drag coefficient, and lift-to-drag ratio for your model at the angle of attack tested will then be populated.

After you have finished all of your trials at the desired angles of attack, I recommend generating graphs of angle of attack versus lift coefficient, drag coefficient, or lift-to-drag ratio. This is a great way to visualize the results and see if there might be any bugs in the system.

A 3D airflow simulation tool like XFLR5 is a great way to see how accurate your measurements are. While the simulation won't be perfect, it will give you a good idea of how well the system worked. I used XFLR5 because it is relatively easy to use (vs. CFD) and is built for low speed aircraft like RC planes (perfect for the lower fan speeds of small scale wind tunnels). You model the exact shape and size wing you used and it outputs a wide variety of information (including lift and drag coefficients and lift-to-drag ratios) across a specified range of angles of attack. However, if you are conducting experiments with models that aren't airfoils or aircraft XFLR5 probably isn't the best choice. XFoil is another popular option, but I found that its 2D simulation produced values that were way further off from my measurements than XFLR5. If you are comfortable with CFD programs the results from a CFD simulation would be even more accurate. I won't be able to explain the whole simulation process, but there are plenty of great resources out there.

Congrats on building your own lift and drag measurement system for small scale wind tunnels! I hope this helps you in conducting your own experiments and hopefully you learned some cool stuff along the way.