Revisiting BioSand filters, a type of water filter used in developing countries. Developing a concrete mold to slash costs from a quarter of a million dollars to around 200 dollars each using commonly found resources (wood and steel) as well as 3D-printing. Later revisions of the filter can bring the price as low as 5 dollars each.

Team members

Ying Wang, Designer, Printer, Concrete pourer

Computer Science & ECE, 2016

Project Updates

Project Notebook - Biosand Filter Part 2: Completion of the Proof of Concept


This is the completed biosand filter after the second revision. This is a proof of concept demonstrating that a 3D-printed mold can be used for concrete.

There were numerous improvements over the past filter. The 3D printed parts were redesigned such that they were easily removable after use, instead of being glued into one piece that adhered to the concrete too well. There was also much less use of mechanical ties such as wood staples; instead painters tape more than sufficed to hold the mold together. Instead of plastic sheeting covering the concrete-side of the interior and exterior mold, a form release agent made up of a combination of mineral spirits and petroleum jelly was spread onto the concrete, making it easier to remove as well as much faster to apply.


Additionally, the concrete mix itself contains far more water (often a quart or more for each bag). This allows the concrete to run into spots and harden into a solid block, instead of acting as many pieces of gravel. The amount of water that is used in the concrete mix can be seen in the below image.


The water pools at the top of the concrete as the concrete is saturated with water. This is accurate with other concrete pours from other people. Since I left the filter to dry for three weeks, the amount of water remaining is trivial.

One problem that I did experience is vibrating the concrete before it dried. Since the base of the filter is very small compared to the exterior mold, I was unable to screw in secure supports. When I stuck the vibrator in the concrete, the mold broke and the concrete poured out the broken sides. This happened twice before I decided to screw in as many steel fasteners to the sides together as I could and not vibrate the concrete. This worked and while the concrete did shrink some and contain bubbles, it did dry together successfully.

This project is complete and is awaiting pickup from the nonprofit. Possible next steps include a fully-3D-printed biosand mold on a large-scale 3D printer, which would cost around $5 and would be reusable. This would enable modularity as well as prototyping of unique aspects (vibrator attachment on the side of the exterior mold, handles, etc.). Water tanks/raised granary stands can also be made as well.

I’m quite happy with how this project has turned out and look forward to seeing how 3D printing and international development will work together at Duke in the near future.

Project Notebook - BioSand Filter Update: This is what Innovation Looks Like


My name is Ying Wang and I am a rising junior majoring in Electrical and Computer Engineering and Computer Science. Together with my fellow co-founders, I helped found DukeMakers, Duke’s 3D printing club and makers organization. We help integrate 3D printing into the Duke community as well as help organizations develop their 3D printing expertise. Out of all the projects I thought I would work on, building a 3D-printed concrete mold was the last on my mind.

I got started on this project when the founder of a Duke-based nonprofit approached me and asked whether it was possible to build a 3D printed concrete mold. He told me that nonprofits traditionally use steel molds to pour concrete, which cost a quarter of a million dollars apiece. I told him I’d get it done for about $500.

A biosand filter essentially is a large concrete block with a cavity. That cavity is filled in with gravel and sand. The water is poured through the cavity and filters through the sand and gravel, coming out of a hole in the bottom as relatively cleaner water.

The first step to making this biosand filter is designing the mold. As an electrical engineer I had only the basest of ideas of how to design a concrete mold (it’s made of wood…right?). The most important part was the interior mold which needed to be tapered for easy pull-out.  I decided to make most of the filter out of wood and use 3D-printed anchor pieces in as supports. The villagers can cut the wood themselves and the nonprofit only needs to bring in some lightweight and completely replaceable 3D-printed pieces in their carry-on luggage.

The filter was built beautifully, the sides of the interior tapering to the edge of the 3D-printed piece. The inner mold fit cleanly within the exterior mold and the form was ready for pouring.

As it turns out pouring concrete is not that easy. I mixed the concrete with too little water resulting in a gravelly consistency, and the mold was not fit in perfectly within the exterior mold. I also only left the stuff to set for 3 days instead of a week. The concrete was weak and broke very easily; the first iteration of the water filter was a failure. However I learned sufficiently enough information to make sure that the second iteration, coming by the end of the summer, will be a success. If so 3D-printed will have proven itself in yet another domain.