From pipe to a filter

This week we have spent hours in laboratory building and testing the filter hydraulics following the example of the article “Performance evaluation of a locally developed domestic drinking water filter” (Bolaji et al, 2010). The filter prototype is ready for hardness removal tests and it’s time to take a look to our lab diary and see what we did and learned.

Session 1: The first fill

Dia1

We built the filter into a plastic pipe which had previously been used for microplastics removal research. We filled the pipe with gravel, sand and bottom ash (which is more correct term than charcoal in this case) from Fat Lizard restaurant’s owen and put a piece of fabric in between the filter material layers.

We didn’t have the raw water tank yet so we took the water from the tap and evidently the pressure was too high as our filter bed started to float. For the next day we needed to have the small water tank to control the inflow pressure. Also the filter body is too high for us and it needed to be shortened.

 

Session 2: Fun with hydraulics

We decided to decrease the height of the filter material layers so that the water would infiltrate faster through the soil and because a smaller filter is easier to transport to Mexico. The final thicknesses of the layers are (from bottom to top): gravel 4 cm, sand 15 cm, bottom ash 11 cm and sand 15 cm. The diameter of the pipe is 12,3 cm, so the total volume of the filter material is ~5,4 litres. In addition, we added an extra fabric on the top of the material bed with some rocks to hold the fabric in place.

Now we had also a tank of 30 litres available for raw water, but the tank took the replacement air from the water hose going to the filter, which lead to under pressure and interruption of water flow. We had to remove the tank from the system and replace it with a funnel as we wanted to figure out the flow rate which comes out of the filter. We tried to hold the funnel on the same height as the water level in tank, approximately 92 cm above the bottom of filter, to keep the pressure stable. However, this was too precise a task for us and the filter bed started to float and the sand jammed the outflow passage. Now we weren’t able to define the flow rate but Henna had lots of fun playing with the funnel heights and observing the chances in water flow directions within the sand layers.

Session 3: Ready for hardness removal tests

Dia2

We started the third day by shortening the pipe and refilling it similarly as earlier. Trying to avoid the floating, we decided to water the material from top of the filter, and the watering time was a bit longer compared to previous tests as the hydraulic pressure pushing water through the filter was lower. However, this watering direction wasn’t ideal either: when we started to run water into the filter from the bottom, there were some air bubbles coming first that moved through the filter, disrupting the filter materials a little. Perhaps it is better to water from bottom while being careful with the pressure. In spite of the bubbles, the filter is ready now as we got a suitable tank of 10 litres connected to the system. The flow rate is approximately 4,17 litres/hour, which means that filtering 10 litres of water lasts a bit less than 2,5 hours.

References

B.O. Bolaji, G.A. Bolaji & S.O. Ismaila (2010) Performance evaluation of a locally developed domestic drinking water filter, International Journal of Environmental Studies, 67:5, 763-771, DOI: 10.1080/00207233.2010.514107

(Photos: Henna Jylhä & Sara Saukkonen, Text: Henna Jylhä & Juho Kaljunen)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Powered by WordPress.com.

Up ↑

%d bloggers like this: