Sand filters, PVC pipes, siphons and weirs 2016-08-31
SewageSlowSandFilterWater
0 Comment

We have been using a slow sand filter to treat our gray water for some months now and it has been working quite well. There was a problem due to normal intermittent use that the water level would drop due to evaporation and expose the sand, which basically kills off the biofiltration layer, but we fixed this by raising the level of the output PVC pipe.

Basically, the output PVC pipe acts as a kind of weir. By requiring the water that comes out the bottom of the sand filter to climb up a PVC pipe of a certain height before falling back down, the water pressure of the water going through the sand filter urn is actually proportional to the difference in height of the water level inside the urn and the height of the output PVC pipe. Thus, when these levels equalize the sand filter stops passing water through. By raising or lowering this level, you can control the height at which the incoming water stops being filtered and keep the biological layer in the sand filter from drying out. (But as I’ve learned, you need a little extra because the incoming water will still evaporate.)

Other than that, the system worked well.

That is, until we finally got around to mounting the final stretch of PVC pipe from the sand filter to the cistern.

Here is a picture:

You can see the smaller inlet PVC pipe climbing up the side of the sand filter urn to the top where it drops the graywater onto the diffuser plate inside. Then the water filters down through the layers of:

  • Burnt rice husk
  • Fine sand
  • Course sand
  • Fine gravel
  • Course gravel
At which point the filtered water is at the bottom of the urn. If there is pressure from above (i.e. the inside water level is higher than the top of the output PVC), it enters holes in the output PVC pipe at the bottom of the urn, passes out of the urn, climbs up a ways and then drops back down (this is the “weir” like section), and then feeds out to the lower right of the picture where it drops into a larger cement cistern where the water is held until it is used for watering trees.
Here is a closer up view of the side of the sand filter:
The problem was that this didn’t work. No matter how much water we put in, it never came out. 
I tried disconnecting sections of PVC pipe, and it worked at each intermediate section. It just didn’t come out the end into the cistern when it was all assembled.
This stumped me for most of a day.
The water dropping from the top of the “weir” should have enough pressure (from the height) to push out the much lower height of the PVC rising over the edge of the cistern.
When I finally figured it out, I was quite surprised. Basically, since the PVC pipe is entirely airtight, once the water level in the final rise by the cistern gets higher than the horizontal stretch, then a large air bubble gets trapped between the top of the “weir” and the bottom horizontal pipe. This air bubble has no downward force (because it has no weight) and thus the weight of the water in the final rise pipe adds to the weight of the water in the weir section resulting in more back pressure. In other words, the compression of the air bubble in the “fall” section of the weir allows the second rise to add its backward force to the first rise, resulting in a complete stop to the filter.
Now, my first thought was to remove the air bubble and make sure that the pipe was fully filled with water. But this would have been a very bad idea, since this would turn the weir into a siphon. As a result, the weir would have no significant effect at all and the water pressure would no longer be the difference between the water level in the urn and the top of the “weir”. It would become a function of the difference in height between the water level inside the urn and the bottom of the “siphon”. 
So instead, I drilled a very small hole in the very top of the weir piece to allow the air to escape rather than propagate pressure back.
Immediately after drilling that hole, I started hearing the hiss of air escaping (which confirmed my theory about an air bubble acting as a means to propagate backpressure from the final rise of the pipe), and shortly after, the sand filter functioned as designed.
Lesson learned. A weir must not be airtight lest it become either a siphon or… I don’t know what to call it when it had that air bubble in it. But it wasn’t controlling the level of the water in the urn like it was supposed to. Until the air was allowed to escape.
All’s well that ends well. 🙂