Gravity Pipe-Flow between storage tanks with inverted siphon - Help!

[jakeinuganda]

Hello,

I'm working on rural water supply with an NGO in SW Uganda and I have none of my engineering books with me and very limited internet access, so resources are very limited. Also, I haven't done much hydraulics since about 2004, so while it's in my head somewhere, I'm a bit rusty.

Currently, I'm working on a system to convey rainwater from the roof of a large building into small elevated tanks at the building so that it can flow from them by gravity through small underground pipes to a large tank about 30m away (and roughly ~3m below the tanks). Two small elevates tanks will capture water from two areas of the building, they will be piped to the base of the storage tank, and then combined into a larger pipe to flow into the storage tank (see attached picture). Funds are limited, so I want design for minimal expense. Obviously, this will mean determining the most efficient combination of small elevated storage tank capacity and underground pipe size for the gravity flow connection. The pipeline will extend straight down from the small elevated tank, turn once underground toward the large storage tank, and rise straight up to the inlet at the top of the large storage tank.

I have an estimate for the peak inflows based on observed rainfall data, so I can size the smaller tanks and the connection pipes if I can model the system in regard to various small storage tank sizes and pipe diameters.

I've made a model with Pipe Expert (picture of model profile included) and I can spec the materials based on the results, but I'm unsure about one thing: these pipes will remain dry most of the time and will only fill when it rains - do I need to be concerned with the a possible air trap problem with the initial vertical drop of 4m??? How can that be modeled??

I would greatly appreciate any help or advice!
Cheers,
Jake
US Peace Corps Volunteer
Rukungiri, Uganda

 

[BigInch]

From what I understand of your diagram, you don't really have 1 Barg of pressure at your inlets, that is 1 Bar Absolute, as it is exposed to the atmosphere, so you have no gauge pressure driving flow through the pipes, you have a couple of meters of differential head ... maximum. Flow will be very slow.

Secondly, 1" and 3/4" pipe will soon clog with dirt, dust, twigs and leaves, or whatever. No point I can see in building this at anything less than 2" maybe 3" diameter, if you could install and maintain it with clean-outs. Don't count on maintanance happening in Uganda. If you build it like that, as soon as you leave, it will stop up and never be used again.

I would recommend that you do it right or not at all. That would mean something of a much larger diameter and with a steady slope to above ground running from the roof to the top of the tank. Underground elevations in this case will just turn into a sediment trap, as the small line diameters and slow velocities underground (you won't get high velocity flow with just a couple meters of head) will just let whatever dirt particle drop out and sit there. If you keep all the pipe above ground you could easily install a drain at the low point and flush when it gets a bit on the dirty side. Maybe that'll stand a chance of happening once in awhile after you're out of there.

BTW, It is not a siphon, unless the height of at least one point on the pipe is higher than the elevation of the fluid surface level at the intake point. You have pressure flow, but with VERY little pressure, 0.2 - 0.3 Barg, you will have almost no flow through a 1" pipe, if it stayed clean. Sorry to say this, but nothing except trouble will come from the current design. Any precious little money spent will be totally wasted. Better to buy food with it.

If you want to redesign, there are a bunch of us here that will help you do it right.

What would you be doing, if you knew that you could not fail?

 

[jakeinuganda]

@BigInch

Thank you so much the thoughtful and thorough reply! I didn't expect one so quick!

I'll provide some more information to address some of your concerns so that we can discuss them. I'm not at all married to this design, but I've made more considerations than I outlined in the original post.

I see your point on pressure, but the pressure is nullified anyway since I applied the 1 bar to both ends of the system, correct?

3/4" was just what I had put thus far, it would actually be with 1", so that makes it a tiny bit better. That aside, the building is high on a hill where it is nearly impossible for anything other than some bird droppings to accumulate. Of course, some dust will make it to the roof, but this is a wet (not dusty) region of Uganda and the roof is far away and above the typical dust sources (dirt roads with traffic). I understand this well because I've installed a rainwater system on my own home here (supplying all of my water), which is close and a bit downhill from the church and I've lived here through one dry season without almost no dust accumulation. For added protection, the small upper tanks will have inlet filters that will should keep anything bigger than dust out. Finally, I plan to install a washout at the end of the line before it rises to the large storage tank.

Finally, in regard to maintenance, I understand the challenges specific to this setting. The good thing is that this system will be at the site of the office of the organization I work with, who have much higher capacity than most project beneficiaries. Also, I will be here for another 16 months at least, so there will be time to build specific maintenance capacity for this system.

I knew it was not a conventional siphon, but I remembered the term "inverted siphon" from my days in municipal civil engineering; I thought I remember the term being applied to sewer lines that work in this way to go under something before returning to traditional flow (in which case high velocities are needed to avoid settling). I never saw or knew of one, but thought I heard someone talking about one once...I may be wrong about this of course.

Of course, an above ground gravity flow system would be simple, but aesthetics are a big issue here as the building is a cathedral.

For these reasons, I'd like to explore this option a bit more before falling back on trying to convince the cathdral to go with aerial pipes (which would be needed for gravity flow).

I went back and removed the pressures to ensure an accurate model and the flowrates were the same, which will be sufficient with a safety factor with 1" pipes. I'm still curious if flow effects from air in the system at the beginning of a rain could be a problem though since it is not accounted for in the model.

I sincerely appreciate your comments.
Cheers,
Jake

 

[BigInch]

I'm glad you've had good luck with your personal system. Yes, I see you have 1 Barg on all ends of the pipe. Actually it is 1 BarA. Your only driver is the 3 meters of head, about 0.3 bars of differential pressure.

With such a small diameter and turbulent flow at those flowrates, I think that air would be pushed out ahead of flow, provided you didn't have a great relative high point at the middle of the pipe and the initial fill is relatively quick. I'm mostly worried that you will get plugging, be it from dust, roof tile bits, bird droppings, feathers, algae or whatever else wants to grow in there. If you insist on a small diameter, at least put in a flush connection, maybe in the middle of the underground segments and flush both ways towards the tanks.

I presume that the storage tank is more or less in between the two intakes and that combining the underground pipe runs is either not feasible for one reason or another.

You've done a lot better than most first-time posters with the hydraulic analysis. It's only a matter of it being a practical solution, or getting stuffed up with dirt.

What would you be doing, if you knew that you could not fail?

 

[bimr]

Regarding "BTW, It is not a siphon"

What has been described is a form of siphon known as an "inverted siphon". The poster is correct.

The 1-Inch pipes are too small and will be subject to plugging with debris. You would be better off hauling the water in buckets.

How do you plan to get the rain water into the tanks on the rooftop?

 

[BigInch]

OK, I guess everything is either a siphon or inverted siphon.
Whatever it is, it doesn't make me feel any better about the tiny diameter.

What would you be doing, if you knew that you could not fail?

 

[jakeinuganda]

Thanks for the further input.

The model is saying that flows from a light-medium rain will see a velocity of 1.9 ft/sec while a heavy rain should see flows of 4+ ft/sec. Coupled with inlet filters to keep out large debris and a washout at the bottom of the pipeline, could this be enough to prevent clogging?

 

[hokie66]

Where do the pipes come from? What material? If you use anything smaller than about a 90mm PVC stormwater pipe, much of the water from the roof will be lost over the gutters.

 

[jakeinuganda]

We're actually using sheet metal gutters, which collect much better than PVC half-pipe gutters in addition to being cheaper. They're just a bit more challenging to install properly. The concern of this thread is the unique hydraulic aspects of this application, which are described in the above messages.

Thanks!

 

[cvg]

I think clogging will occur, regardless of the screening used. Water will carry sediment from the roof and since you have an inverted siphon, the dirty water will sit in the pipe for days. The sediment will settle out on the bottom and eventually clog the pipe. There is probably not high enough velocity to carry the sediment vertically upward and out of the pipe. Unless there is a cleanout installed with a low level drain and the pipe is flushed periodically, it will probably clog. You can't run a snake down such a small pipe either, that is why you need a larger one, at least two inch or preferably three inch minimum to allow maintenance.

 

[bimr]

How do you plan to get the rain water into the tanks on the rooftop?

 

[hokie66]

I wasn't suggesting the use of PVC gutters. Metal gutters are the right choice. I was talking about the downpipes which go from the gutters to the tanks. A sump in the gutter at the downpipe helps with efficiency of collection.

I fail to see why you need such an elaborate system of tanks and piping. Just run the downpipes into a collector which takes the rainwater directly into your storage tanks.

 

[jakeinuganda]

The storage tanks are being built away from the building because there are plans to expand the building in the future. This situation would require aerial pipes to flow by gravity, which is not desired for aesthetic reasons.

With the inverted siphon design, a washout will be installed at the bottom of the pipeline and will be opened periodically to flush the pipes. So we could use 1" GI on the initial vertical run from the elevated tanks and then larger pipes on the underground run (where can use PE, which is tremendously cheaper than GI), that is where the settling would occur anyway. I'll look into that.

In regard to collection from the roof, it will be guttered and in some cases piped in 4" pvc to the elevated tanks. We've got the collection aspect covered as we have lots of experience with traditional RWH designs.

@hokie66: my mistake. please tell me more about a sump at the downpipe and how it would help collection. I don't think I've seen such a thing.

 

[hokie66]

jake,
The attached shows the idea of sumps and rainwater heads in gutter systems.

http://www.roof-gutter-design.com.au/

 

[jakeinuganda]

Oh, I see what you're talking about now. I've just always called it a collector box, and it will be used in this design.

 

[bimr]

Here is a link to a well written document describing the management of rainwater harvesting systems.

I am still not following why you have elevated tanks on the roof. Seems like the rooftop drainage should proceed directly to tanks at grade.

Since you have elevated tanks, you will need to ensure that you have some type of flooded suction on the entrance to your downpipe. Otherwise, you will trap air in the downpipe which will lead to poor water flow from air pockets, especially with the small diameter pipes.

I noted that the minimum downpipe size in the document is 1 1/2-Inch. Besides the clogging problem that should be expected with the smaller 1-Inch pipe that you are planning to use, there will be installation issues as well. Small diameter piping tends to be very flexible and it will be difficult to support and the pipe may not be asthetically pleasing once installed, because of that. PVC piping is also not resistant to UV light and should be painted to protect the PVC from UV light degradation.

In addition, you have not mentioned first flush, disinfection, etc; I would hope that you are addressing these matters.

Anyway, maybe there are some tips in the document that may be of some assistance to you.

Good luck.

http://www.unep.org/ecosystemmanagement/Portals/7/Documents/em-rainwater-handbook-caribbean.pdf