Rainwater catchment in tropics (New build)

We measured the drop from the house ground level to where we roughly planned the tank location, only a drop of about 0.8m. The builder would prefer to build an underground cistern, similar to a swimming pool with a cement roof. He's not so happy about doing a wet system and I suppose an underground cement tank is structurally going to be a lot stronger than something above ground. So we're going to head in that direction now. At least I don't have to worry so much about my friction loss calculations anymore!

I will still use the sediment traps you developed (from Jnk40's post), even though it is now a dry system. Instead of having calming inlet piping (the "U" at the inlet pipe inside the bottom of the tank) I am going to add a small wall inside around the inlet pipes (which will extend almost down to the tank floor) to hopefully prevent too much sediment stir up.

I won't be able to do the bellmouth overflow anymore if the tank is below grade? Is having a horizontal overflow pipe out the side my only option? I was so set on the idea of a wet system that I'm not sure where to start with a buried cistern!

If you use leaf diverters, you can use a low restriction inlet that will supplement the flow rate to an above ground tank.

A sub surface tank's overflow will need attention to overflow capacity and if you have only one overflow outlet, a larger pipe will automatically reduce the head. You can still use a bell mouth in an underground tank.

Water really should pass through a silt pit on the way to an underground tank but it would have to be very large. This use to be done by the ancient Egyptians and some of their ancient and very large cisterns are still in use today. The alternative is to ensure that sediment can be easily removed every few years. Some people use pool pumps.

The sediment traps are ideally for flooded pipes, the flow is different otherwise but any improvement is worthwhile. It would be interesting to see what the capture is like at different rain intensities.

Having the second tank is still a good idea and it is critical that sub surface tanks are well vented.

After I posted I realised that the sediment trap might not be the best idea, there wouldn't be the pressure of the sitting water to flush through the smaller sediment pipes. If I did it would definitely need the inspection orifice as I would need to flush water into it to have the sediment pass through the sediment trap pipes.

I'll do some research about smaller silt pits, etc. Will see what I can find.

I can't visualise the bellmouth overflow for an underground tank?? Does it have a u-bend in the pipe or something? Any diagrams you could point me to?

A tank has an overflow pipe regardless of whether the tank is above ground (unless it is supplied by Supadivertas) or sub surface. A bell mouth could sit lower than the one in the photo if the elbow was heated and molded as a bell mouth. So long as it was effective, how it looked wouldn't matter because it is inside the tank.

So for the bellmouth you've shown, I continue that horizontal pipe out of the tank below grade all the way (at slope of 0% or slightly downward slope)? What I don't understand is that if the tank is below grade, than the overflow is also below grade until I can bring it to a point that is at a lower elevation than the pipe (in my case it's gonna be quite a long run!).

That's right.

You can have a tank that is partially dug in though, nothing wrong with that. Easier and cheaper too.

I've looked around for a suitable way to minimise sediment for my system (dry piping with sunken cistern) but recommendations of first flush systems would be a bit impractical as I've got 24 downspouts in total.

Like you mentioned, if I was going to do a silt trap it would need to be quite big.

My options are:
1. I could either have all the pipes run directly into the below ground tank and just use a pool vacuum to clean it every year or so. There would only be DIY leaf guards on the downspouts. Bit risky as not sure how much sediment I will get, but at least super easy.

2. Would a tank sticking about 300mm above grade with the "wet system" design from Junk40 work still? If the low restriction hose into the tank could be turned off, I could flush out sediment through the sediment traps (wouldn't be much head, but hopefully enough?) and also drain the last of the water through the low restriction hose into the tank (when water level is low enough). With the tank sticking a little bit above grade, it also makes it easier to run the overflow through the bellmouth as per your suggestion:

http://drive.google.com/uc?export=view&id=0B52grw6hCinQUGVDazdURkF5RTg



http://drive.google.com/uc?export=view&id=0B52grw6hCinQVm9YNGZ6d25VX0k


Am I making things too complicated for myself?

Sorry for the delay, I am a long way behind with my work at the moment and working every available minute.

The plans are all good, you know what you are doing. A couple of notes...

Your design to separate the pump from the infeeds is very good and common sense but amazingly, most tanks have the infeed directly above the tank's outlet that supplies the pump or above the submersible pump.

A calming wall is also a good idea when there is a submersible pump and for more than one reason (see video below), particularly for non settling tanks but I wouldn't have it totally enclosed if it was a large area. A calming inlet is best having a U shaped tee extension at the bottom.

https://www.youtube.com/watch?v=ULj6Ar2ncLE

I would like to do testing with a similar set up but have the water falling at different velocities vertically through a tank's standard aperture meshed inlet and have a floating intake hose 200mm below the water line supplying a pump operating at different infeed velocities.

By way of example, a 20mm internal diameter (ID) hose has an internal volume of 0.314 litres per metre, requiring 19 metres of pipe to supply 6 litres per minute (lpm) to a pump. A cistern (here in Australia) is designed to fill at no more than 6 lpm but water would still flow through through a 20mm ID hose at a velocity in excess of that needed to transport large air bubbles (greater diameters than 2 mm) down a vertical pipe.

I am also interested to see what effect a meshed inlet at the top of the hose would have on drawing close air bubbles.



The video's testing station is great and a credit to KSB, I am envious!

A low restriction inlet improves standard wet systems by supplementing the infeed flow rate, reducing the amount and height of water retained in the wet system pipe and oxygenating the anaerobic zone but your tank will be down slope and retain little water, you will have plenty of head and if you have a calming inlet, the anaerobic zone will receive all of the oxygen rich water anyway BUT because the low restriction inlet reduces the wet system's retained water, you will not be able to flush the sediment trap. Good leaf diverters on all downpipes are also a prerequisite if you have a low restriction inlet.

Could the tank possibly be constructed with a small depression that would capture a lot of any sediment that entered the tank?

I'm thinking that because of the Phil's weather pattern, you could leave the sediment trap open for a lot of the time but put a strong flap valve at the end because rats are intelligent animals.

The system you have designed is not complicated, it is simple, easy to understand and superior to standard systems.

Thanks again for your comprehensive reply SaveH20.

I've had a bit of time to think about it and found a bit of inspiration from an Ohio Dept. of Health document regarding rainwater cisterns.

This is what I've come up with at the moment. The silt trap isn't very big but hopefully it'll prove useful in all but the most intense rainfall events. I've added a sand/gravel filter at the bottom of the pit so that it can always drain (at least until the tank is half full). As the tank fills up I'm hoping the sand/gravel filter won't move much and further dirty the water. We've got a river about 40m below us (literally down the side of a cliff) so have access to clean gravel and sand.

I've just used calming walls on the inlets.

https://drive.google.com/uc?export=view&id=0B52grw6hCinQTENjcDQ2N25saW8



https://drive.google.com/uc?export=view&id=0B52grw6hCinQSHY0b3RubmVHTDg

Not much by ways of contributing to help from me here, but just chipping in my 2 cents worth to thank you guys for starting this thread. It's actually quite interesting reading through and I'm realising how the smallest things can make such a big difference. Especially in terms of geographical location and climate - you wouldn't think about having to account for all of these things to such an extent, but it is actually so necessary to make sure that you have a good working system in the end. So thanks again guys!