Cross-Flow Turbines

At the University of Washington had the opportunity to manufacture hydrofoil blades for Particle Image Velocimetry (PID) research. The first step was to design and machine two blades and a pair of clear acrylic plates so flow over the spinning blades could be observed from below the blades. Next, aluminum struts were manufactured to replace the acrlic plates.

This page is divided into three sections.

Highlights

CNC Machining a foil.

Turbine assembly in operation.

Foils

First, the assembly was modeled in CAD using Solidworks.

Next, Solidworks CAM is used to generate a toolpath to be used by a CNC mill. Operations are designed to first remove the bulk of the material, then to machine the profile. Part of the operation is shown in the video below. A bull-nose mill was used to minimize ridges on the curve of the foil.

After preparing the stock, the first side of the foil is machined. Another piece is stock is placed underneath the foil stock to add additional stiffness.


The reverse side of the foil is more difficult to manufacture because the thin trailing edge of the foil cannot be machined to a precise edge without additional support. To solve this problem, a new piece of stock was machined with the inverse profile. 

The half-finished stock is flipped and glued into the support piece and the second operation is performed. Acetone was used to remove the foil from the support after milling is complete.

Now, milling is complete but some additional steps are needed. 

In this image you can also see the three pilot holes that were drilled in the stock BEFORE the milling operations.

The remaining stock on the ends allowed the foil to be held in place and carefuly sanded to the desired surface finish.

Next the end stock was removed

After the end stock was removed, spacers were 3D printed.

These spaces make it possbile to support the foil in a vice.

The three holes were drilled and reamed to the desired size and depth.

The center holes were tapped. Each foil end required two pin holes and one threaded hole for a screw.

Clear acrylic plates were laser cut and the foils were secured using pins and screws. The screw heads were countersunk into the acrylic.

At this point, I handed the assembly over to Abby who coated the foils black and used them for her research. In the image below you can see the assembly in a flume water channel. There is a camera mounted below that is used for PIV data collection.

Struts

Next, I was tasked with building aluminum struts to replace the acrylic plates. The struts look like this when finished

This time the milling was more complicated. Multiple operations were needed for center-drilling, 3 different sized holes, countersinking two holes, milling the large center hole, and milling the foil shape with a bull-nose mill. Aside for pausing for tool changes, the operation was completed with one seemless CAM operation.

Again, the stock was glued to an extra piece of stock for support when milling the first side. Then then a custom support was made to aid in milling the second side

Milling the first side, including drilling holes.

Removing with acetone after milling and drilling the first side.

Milling the second side.

No holes are drilled on the second side, but an additional operation is need to cut the outside profile of the strut. This is done las since at this point the strut will be completely detatched from the remaining stock and will only be secured with glue.

Finally, the strut is removed from the support.

The first strut is made in this manner, but the second strut is a mirror image so new CAD and CAM was made for the strut and for the custom support piece.

Unfortunately there were some issues with my CAM for the operation milling the second side of the second strut, causing my part to be completely removed.  This error was particularly annoying because it was not observed in the CAM simulaiton meaning that I had no way of knowing it would happen until, at my shock, the mill destroyed my part.

After troubleshooting, I fixed the error. In total I milled four struts.

Credits

This project was completed for Dr. Brian Polagye's marine renewable energy lab at the University of Washington. This process was an amazing learning process that taught me valuable CAM and machining skills.

I am incredibly grateful for the support I received from  Eamon and Reggie , the University of Washington machine shop supervisors. Abby, whose research I built these foils and struts for, was very helpful and a pleasure to work with. I'm also thankful for additional support from Hanna, Isabel, Noah (who was abroad during this project but build some for the CAD models I used) and others in the lab. Finally, I must thank Brian for giving me this opportunity and for is exceptional leadership in the lab.