Recent Posts

Thursday, January 31, 2013

Helping Hand: Dengkil AP Up and Running!

So it took us a while to complete the project in Dengkil, however it was a good experience gained. The project was completed last Sunday 27th January. It took 2 days to assemble the plumbing and pipe routes. Needless to say, it wasn't without problems. There were a bit of a problem with the return pipe from the gutters since the tailpipe wasn't big enough, but in the end we tackled it in style. My thanks to Haji Anwar for presenting the opportunity for me to work with him and his sons. It was indeed a pleasant experience and I gained knowledge to apply in the system I have planned for my house.

Since I don't have an AP system at the moment, I really treated this as my baby. There is more to come when we reconfigure the pond filter which I've mentioned in the previous post. As the water flowed from the pump which is seated after the filter, we could observe sediment piling into the gutters. This showed that the pond filter is inefficient when it comes to mechanical filtration, which is to remove the solids from the water. I have planned an improvement project, which will commence soon when time permits. After all, I can't afford to have those sediments piling up in the gutters!

Lets see some pictures shall we?

 This is the overall layout of the system, the pond is right at the pillar on the right, the gutter beds on the background on the left. All in all, about 70ft of 25mm polypipe was used for the inlet, while about 50ft of 32mm polypipe was used for the outlet back into the pond. One thing I learned; the polypipe isn't the exact size the state it is, 25mm polypipe is smaller than 25mm PVC pipe, same goes for the 32mm polypipe which is actually about the same as 25mm polypipe. How does this affects the system? The outlet ends up being under designed!

Here are the gutter beds, holes being drilled. You can see the inlet manifold attached to gutter rows at the back. The inlets and the connectors of the gutters are 25mm PVC pipes.

My loyal assistant Syafiq at work. He's quite a handyman himself and has no trouble picking up and executing the tasks I laid out. The gutter connectors can be seen on right.

Partial view of Haji Anwar's pond. The tilapias in this pond are breeding. The largest ones are around 3kg!

Here's how the gutters are laid. Initially I just wanted to use a single piece of 20ft gutter but the earlier plan was to place the gutters at where we would need to bend them. This layout was a surprise to myself when Haji Anwar showed me the final placement of the gutters. I knew at that time, making a return which works by gravity at the length of 50ft is gonna be troublesome... 

Here's another view at the beds. They're about 7ft a piece. 

This is where the inlet to the beds is connected. The pump which powers the fountain of the pond is bypassed at a very low flowrate. This makes the system a continuous flow since the pump runs 24/7 and configuring a siphon with a 50ft tailpipe is very difficult I didn't even want to attempt running a siphon at that length...

After 70ft of 25mm polypipe, the inlet goes in here. A valve is crucial to control the flow though there is already a valve at the bypass at the pump. It just makes it more convenient to have it closer to the beds so you can immediately see the changes you make. You can never have too many valves...

This U-pipes are the connectors of the beds. It has a tee at the bottom with a stopper nut. Just in case the sediments plug up the pipe and there is need to drain it...

This is the swiveling loop siphon but in this application, it acts as an adjustable standpipe. When you swivel the loop, you can make changes in the water height in the beds...As you can see there is a reducer there however not by choice. The 32mm PVC pipe is indeed larger than the 32mm polypipe and connecting it requires that the 32mm PVC reduced to 25mm size. This is where the problem starts, a gravity dependent overflow must never be undersized. Ideally, I think the pipe should have been 32mm all the way but polypipe has the advantage of being flexible. Since all the stuff were there, we decided to just go ahead and make changes if problems arise.

After about 50ft of 32mm polypipe, the water from the beds return to the pond as shown...

So we open the flow and let it run. The water starts to flow to the beds at a very small rate (I need to verify this flowrate later). However, at first I was really anxious about the return pipe since after 10 mins of the beds started to overflow I still didn't get any returns to the pond! However then we start seeing the return flowing...but soon, it started siphoning!!! As shown in the picture below, the worst scenario occurred. I was being optimistic that maybe the siphon would break. However, my experience was right, it never broke no matter how many time I 'restarted' the cycle and made adjustments, the siphon kept going with the smallest flowrate I could just wouldn't break...The problem is obvious, the tailpipe was too long that no air bubble which is needed to break the siphon could be generated given the length of the return pipe...This truly calls for a drastic solution...

 Haji Anwar came with the brightest of ideas when he suggested that we empty the beds into a tank, connect the return pipe to the tank, and the tank itself will drain to the pond by gravity. By doing that, we could eliminate the neverending suction generated by the long tail pipe. I quickly assembled the what would be called a holding tank from a dustbin and put it to the test. I also got rid of the reducer at the loop siphon. Sure enough, we got lucky!

While I fill in the dustbin with water from the pipe to check if it's functioning, I notice once it's reached a level, a siphon was established! This is due to the base of the dustbin, being lower than the highest point that exists along the return pipe to the pond due to some incline in the layout...I just had a hunch that this the tank setup will turn into an autosiphon itself when I saw that the siphon actually breaks!!! I told Haji Anwar, we'll never get luckier than this. I was worried that the smaller-than-I-wanted 32mm polypipe was insufficient to act as the return pipe. Now that the new holding tank setup is working as an autosiphon, a huge rock was lifted off my shoulders!

This is the holding tank setup. As the base of the tank is actually lower than the highest elevation of the return pipe. An autosiphon was created. Once the water in the bin reaches about 2/3 of the height, it starts siphoning. Problem solved, in style! 

As you can observe from the photo above, the flowrate to the gutter is very minimal. The styrofoam and the pots in place, all left for Haji Anwar to do is lay some shade nets and start germinating the seeds...

I'll post more on this as we progress! Thanks for reading!


  1. Wow! It's been a long time since you post your new entry! And I'm excited to read this! The way you explain it is like what an engineer of aquaponic explain it. Keep it bro!

    1. Thanks vm bro! Will keep updating this blog. At the moment Haji Anwar has started germinating his veges. I'm very sure they'll grow well with that many excess nutrient in the water. It should make the pond water a bit less yellowish due to nitrates being used.

  2. Another solution for the long outlet pipe problem is to add a T-connector right after the siphon outlet. Air bubbles can enter through the junction at the T-connector cutting the siphon.

    I wish I could illustrate it here... :P

    1. If i were to try your solution William, I'd place a tee where the left elbow on top of the siphon loop as in the last photo. This design with an open top is a SLO (solids lifting overflow) design. However, I wonder if the water lever could rise above the horizontal pipe of the loop which would then still make a siphon possible. As I've explained before, due to the nature of the undersized return pipe, very minimal flow was needed to make it start siphoning. Well, the 'holding tank' concept worked and I don't think I'd tinker with it unless it starts giving me problem. Thanks for reading guys! I'll keep you updated with the progress of this project. Cheers!

  3. No, if u replace the left elbow (first junction) with a tee, no siphon action will occur. What I mean was to place a tee at the "end" of the downward outlet. That way, a siphon can occur and stop properly even with a long outlet pipe.

    I wish i could illustrate it for you... :P

    1. I've sent you a mail with an attached drawing... :D

    2. Yup got it! Brilliant idea you have. I'll keep it in mind for future projects. I'd really like to test all the plumbing concepts especially concerning the siphon and the overflows...

  4. Wow, you have been busy! It would be interesting to see the efforts of all the hard work!

  5. Looks great. The plumbing seems challenging. Will you be putting transparent roofing to prevent rain from wetting the plants too much or will you just use those black netting to reduce the amount of afternoon sun wilting the plants.


Popular Posts