Rainwater Storing Gojo Is A Stroke Of Genius

A traditional Ethiopian Gojo. Image courtesy of easterntravelandtour.com

The traditional Ethiopian Gojo is a circular domed dwelling constructed from a central vertical beam, and a surrounding structure of curved beams made from wood or bamboo. A covering of dried grass and mud completes the outer structure. These buildings are found everywhere in rural areas, due to their ease of construction, and availability of cheap materials. One major problem living in rural areas in developing countries is access to water. Ethiopian inventor [Anteneh Gashaw] knows a thing or two about the practicalities of living in a developing nation, and has come up with an ingenious take on the traditional Gojo. The idea is to replace the outer structure with pipes capable of storing rainwater. A collector plate on the top of the roof directs rain water into the pipes — with some small balancing tubes connecting them at the bottom — distributing the stored water evenly. A tap at the bottom of structure allows the pipes to be emptied on demand. Another interesting point about this design, is that the water adds some extra weight, for free, which gives the structure much improved stability in high winds, increasing safety.

{Anteneh] notes that proper water infrastructure is incredibly expensive, and just simply won’t happen. Well digging, installation of underground water tanks, and other such stop gap measures are great, but still need significant investment, and he believes that his modified Gojo idea will help reduce the problem of storing water during the rainy season, and reduce the pressure on centralised wells and other such community-orientated solutions. What’s more, it should be cheap. We shall watch with interest where this goes.

We’ve seen a few hacks from Africa nations, not many, just a few, but they are interesting ones. Like this DIY Helicopter that didn’t quite get to fly, and this e-waste 3D printer. We’ll keep our eye peeled for more!

57 thoughts on “Rainwater Storing Gojo Is A Stroke Of Genius

      1. Will the pipes be covered with mesh? You will want some type of screen to keep mosquitoes from flying in and laying their eggs. If you don’t mosquitoes will breed and then feed on and infect the gojo’s residents.

      2. As [AwDuck] mentioned below, not all water needs to be drinkable. I was thinking of a separate collection point at the angle and up to just below the top collection point or the water can just flow across the normal roof surface to the second collection point perhaps for second grade water. This would greatly effect the total catchment surface area but would need shorter tubes obviously. I think that the balance tubes at the bottom are two small to be rigid, especially those right beside the entrance. I single breakage that low could result in the loss of all water.

        Absolutely a terrific idea.

    1. Not all water needs to be potable, though. Having lived through a severe 8 week drought myself, I’d have been ok with dirty water on tap for cleaning clothes and the like – we were using dish water to clean clothes (which then became water to flush the toilet). Water with some plain old dirt would have been a step up for the clothes – especially if we didn’t have to haul it up from the utility tap at the street.

    2. I used to live on a Caribbean island with no central water system. My house (like most) had a cistern under the floor where rainwater from the roof collected. (Really old houses had tomb-like stone cisterns next to the house) Everything that was on the roof (volcanic ash, Saharan sand, bird crap, etc) would end up there. Mosquito larvae swam in it, some plants grew in it. It only had a coarse particulate matter filter. I weaned myself on the stuff and never got ill. Never underestimate the capabilities of your body!
      (When I later moved I had a system with UV light and activated charcoal filtration)

    3. This is such an interesting design, could you use the system to structurally support the Gojo.
      I see comments about the system size but i understand what it is like to maintain crops and gardens, any rainwater capture is better than none at all.

  1. But only that small round collector at the top ?

    Since he will need to filter the water later anyway, why not some collector at the base of the roof that then takes the water to a reservoir ? There are some publications from the UN / others about this, and the ways of building the tanks with the materials and expertise of the place .

    1. @Rog Fanther building tank is very very challenging for majority of the population, it will take resources they don’t have and know how to build which they lack. It’s just the reality.
      Also small round is more than enough to collect the rain water.

      1. The ideas, from that UN book, involved putting value in the people of each village, using their knowledge about pottery, metalworking, etc, to build reservoirs with whatever material and technology was common there. .

        You do not describe how they will get the pipes and fittings for construct it this way. Are they easier to get in the regions you surveyed ?

        1. @Rog Fanther yes, I didn’t described about the implementation because the first focus was on the design, I will get in to it in time.
          If you only know the challenge, capacity and limitations of the peoples from the books, I invite you to come, see and understand. It’s very challenging and if they could solve it, they would have by now.

        1. Welding requires a great deal of energy. Robust electrical systems may not exist in most parts of developing country’s rural areas.

          Transporting pipe and fittings does not require access to such infrastructure.

        2. Just one tiny, tiny question – when did you last build a tank out of “sheet metal and a welder” or even “wood planks and banding etc.”?

          It would be so enlightening to read how you dealt with seam welding and reinforcement on a tank or waterproofing the joints between planks held together with “…. banding etc.”

          And do please describe from your personal experience – not your personal opinion.


        3. An even easier, cheaper and transportable solution might be a plastic or rubber bladder. Bladders need support but that can be a hole in the ground or built with local materials.

  2. There have been some very good ideas out of Africa.
    One from N Africa about 45 years ago was for pit toilet ventilation and disease control. Flat concrete slabs for floor & roof (Sahara), and a continuous spiral concrete wall. Key item for disease control was a 4-6″ ABS vent stack that went about 6′ above the roof. The black pipe got hot,heated the air in it, which rose, drawing fresh air in the “door”, down the toilet hole and up the pipe. It had an open bottom (small bottom) conical screen (oriented tip down) in the top of the stack. Flies would follow the scent, find their way to the hole, and go down to the pit. However, when the fly tried to leave, it flew as high as possible – the top of the cone, and then walked in circles at the top, never finding the hole, until it died and fell back into the pit.

    1. Turning every toilet into a fly trap.
      That is an idea I can approve of.

      About these pipes I am not so sure.
      It probably is not raining much in Ethiopia which translates to catching any water you can catch.
      And the higher you store the water, the more difficult it is to get it there.

      We used to grow pot plants under about 6000 square meters of glass here in the Netherlands and we had storage for about 450 cubic meters of water.
      And even then we had to pump up ground water (which is of lesser quality) every now and then. Therefore we extended the storage to around 900 cubic metres to even out the differences between rainfall thoughout the seasons. And the Netherlands is quite a wet country.

      1. That thought occurred to me. Assuming that the materials to build your model really are cheap, one could double, or triple the amount of vertical pipes. Thinking about it, the walls could be made entirely of pipes, and they can even be added over time as more pipes are acquired.

        Say you start with a hexagon, where each corner is a vertical pipe. If all pipes are connected together with pipes too, you have the volume of the 6 vertical pipes + the volume of the 6 horizontal pipes connecting them. More pipes can be added later by cutting one of the bottom horizontal pipes, inserting a T and another vertical pipe. You can repeat this process over and over as more pipes are acquired.

        I don’t know how close together these are built, but you could also collect more water by building smaller A frame shaped roof that connects 2 of these buildings and collecting and storing the water in the same fashion.

  3. Teaching them how to build a tank with the tools they have, would be more useful than a hundred rain collectors. That way they can teach their friends and everybody gets them.
    One untested Idea:
    A pit with a wooden frame for strength.
    Wood ash cement to make a smooth surface.
    Boiled sap to make it water tight.

  4. Sorry, but this is far from “genius”. It’s interesting and novel, but beset by a myriad of problems that the inventor hasn’t considered. Frankly, it’s a pipe-dream. (Sorry, not sorry….)

    Others have already alluded to water quality issues. That is one of the easier problems to solve. Pre-collection screening, and post-collection filtering will take care of the physical contaminants. Unfortunately, small efficient filters are not free and would be an ongoing cost, but you have to deal with that for any water collection system. A decent filter could be made using a free natural resource that is plentiful in much of Africa, sand, but making that work in a small space at ground level is challenging. That leaves the biological contamination to deal with. The cheapest reliable method of sterilizing is likely to be by introducing a measured amount of chlorine. At that scale, a cap-full of bleach… Other methods would be UV, or again, filtering. Both would be more expensive. Microbiological filters can be slow too.

    Someone has already mentioned collection area. In an arid area that becomes critical. You want to be able to collect as much as possible. The only negative of collecting at a roof-line gutter (instead of a small pan shown here) is that you’d lose the storage capacity of the pipework above the roof-line. Compensate with more pipes below. That leads me to the next 2 problems.

    Looking at that photo of a gojo, I’d estimate the height to the roof-line to be about 3m and the length of the roof about 6m. Let’s round up to a total of 10m to make the math easier. A pipe with 100mm internal diameter would therefore have a capacity of ~314L. That’s ~83 US gallons for those west of Ireland. With 14 supports (as shown in the CAD), that’s ~4400L, or ~1160 US Gallons, a reasonable capacity. But, 2/3 of the capacity is above the roof-line. That’s one problem:

    I’m going to work the numbers based on the CAD model, which I estimate to have 1/2 the capacity above the roof-line, or about 157Kg in each pipe, for a total of 2200Kg in the roof. I’d estimate the roof surface area to be about 27m^2. (3m radius, based upon the roof-line still being 3m high). That results in 80Kg/m^2 (16lb/ft^2) roof-loading from the water alone. That’s considered a class 1 snow-load in the US, and I’m willing to bet that the traditional Ethiopian gojo roof structure wasn’t designed with snow in mind.

    That force is going to push down and out, placing the roof-line fittings under significant bending stress. There’s no collar ties here resisting that force. That raises the 2nd problem – material choice. That hasn’t been specified, but it obviously cannot be PVC. PVC is non-structural. Steel or cast iron will rust. Aluminum is too reactive and would preclude the use of chlorine. Copper is very expensive at that size, and not considered structural. Stainless steel is very expensive. Bamboo doesn’t generally grow in nicely bent sections, and good luck drilling a 3m long cane into a pipe. I honestly don’t see what choice there is that would work that isn’t ridiculously expensive.

    Assuming you can solve that, that balancing ring around the bottom is going to be expensive. Those 14 fittings to join it all up are not going to be cheap. Even if you could use PVC, fittings in that size (4″x2″x2″) are about $20 each in US today.

    Next, it’s going to be a installation and maintenance nightmare. I count 97 connections, all of which have to be leak-proof. All of which must remain leak-proof even though they are being subjected to random structural loads. Good luck with that.

    We live in a world where IBC Totes can be obtained relatively cheaply. Put a gutter around the roof-line and dump the water through a screen, through a cheap plastic pipe, into the tote. Paint the tote with 2-3 coats of latex paint to prevent UV and inhibit algae. 1000L storage, done. Easily expandable with more totes.

    Another solution. Collect into a concrete cistern underground. If you really wanted to make your storage solution part of the structure, this could be a way to do it – make the walls/foundation a cistern. The water would be a huge heat-sink, helping to regulate the temperature inside. Concrete isn’t cheap, but it is effective, reliable, and capable to achieving both of those tasks.

    Ultimately, the best solution is polyethylene water tank designed for the purpose. They scale better than IBC totes because you need fewer fittings for one big tank than for several small ones, and fittings cost money too. A motivated person trying to solve water crises in the developing world would find ways to make those tanks more affordably available. Polyethylene is an easy plastic to work with, with a relatively low melting point. Transporting large tanks is expensive simply because they are bulky, whereas polyethylene pellets are easy to ship. It seems to me that the real problem that needs to be solved is one of manufacturing. This is demonstrative: https://www.youtube.com/watch?v=5xNdPJO3Ql0

    1. @genixia interesting read, I hope you are not a polyethylene water tank supplier or manufacturer.

      Some of your comments on your fear of the design drawback are exaggerated. But I will agree that material selection and cost are important here.

      In terms of the cost I will provide the cost breakdown next, from there you will understand it better and understand the benefits VS the expense argument.

      Material selection will be provided with the specifications.

      All I wanted you to grasp at this time is the basic Working mechanism and structurally system is sounds.
      I liked your comments very much, I will use it to my advantage.

    2. I agree with Genixia and some other commenters further down, using the pipes as “structural” components seems to me both structurally unsound and is imho completely unnecessary. Having that much water and piping up high seems slightly dangerous. And with access to all that pipe, it would make more sense to me to have a rain gutter around the lower edge of the gojo roof and duct it to pipes buried under the floor or next to the house some depth (at least 3 feet down). Digging that depth is entirely doable by hand and also has the advantage of keeping the water cooler. For efficiency fewer, bigger diameter pipes (or a single tank) would be better. Bringing the water back up could be as simple as a large diameter pipe serving as a well (but not a great solution due to leaving an entry point for critters), as complicated as an electrical pump running on solar power, and anything in between.

      However, it’s very easy to fall into the Engineer-saviour trap here (https://hackaday.com/2017/08/03/avoiding-the-engineer-saviour-trap/). We should be very careful in considering what materials are actually cheaply available before actually considering a solution good or bad. Mr. Gashaw probably already has a better idea on that than most of us.

    3. Simple solution to the structural problems: Take the angled pipes out of the roof, and put them vertically in the wall with the other pipes that are already vertical in the wall.

      That also makes it easier to collect water from a gutter around the edge of the roof rather than just at the top, resulting in faster water collection with any given amount of rain.

      Next, to reduce cost of pipe and fittings, replace the wall of (imported?) pipes with a double wall of local materials—wood/bamboo sealed with clay (wattle and daub), maybe. To keep the water pressure from bursting it, connect the inner and outer walls with plenty of tensile members (more at the bottom), which could be wood/bamboo covered with clay, or clay reinforced with straw, or something like that. (I don’t know how reliable that would be, or how easily you’d be able to tell when it was about to fail.) If plastic sheeting is readily available, it could be used for sealing the walls, but then I don’t know how you’d connect the walls together.

  5. Excellent design, Anteneh. I like how using using small diameter pipes solves lots of problems without special skills or equipment.

    I may be missing an obvious issue here, but could extra water be stored by digging a large hole, lining it with clay, and perhaps covering it with plastic sheet?

  6. If the houses are already being made from accessible/cheap local materials, what’s to be gained by building one with a bunch of imported pipes, over just setting the same pipes up for water collection next to the house? I’d be curious to see a cost comparison between building a normal gojo with a simpler pipe setup (now that it doesn’t also have to be a house) for collection, vs one fancy pipe gojo. What about over time as repairs are needed? Surely it’s easier to access and repair dedicated water storage than the combo structure that you also live in?

  7. I am questioning if they can get all this pipe, can’t they just get a more proper and cleanable water tank? It is one thing if all the pieces are made out of locally sourced stuff like bamboo but if it is PVC pipe and fittings it just seems silly.

    I thought of an idea for a mixed media pitcher pump with some parts being 3D printed and some parts being off the shelf PVC. I wanted it for my farm but I thought the idea could have third world implications, but again if you can get PVC and 3D print, you can probably get a more proper pump.

  8. My concern is that you use a lot of raw material to store a relatively small amount of water. i.e. given the same amount of raw material, you will store the largest amount of water in a single large pipe/tank vs multiple smaller ones.And even if the amount of water stored in the smaller pipes is sufficient for the requirements, you will use less raw materials making a single tank rather than multiple. In addition, as mentioned by other posters, you have a very small catchment area, which is less ideal in areas of low rainfall. You obviously want to capture as much of the water as possible when it is available.

    I’m not sure what raw materials are readily available, such as clay, wood, etc, but I would be more inclined to come up with a design using local raw materials that individual villages/villagers can build themselves. An underground cistern, waterproofed with clay, fed by gutters from the gojo. Gutters could be made from bamboo, for example, split in half, and the segments removed. Maybe a PVC pipe to make a pump to get the water out as needed, but worst case a bucket will do.

  9. Is this for an individual family or geared towards a community of some sort? Beyond the creative think/build/invent aspect of this, which is a good thing, are there any programs/charitable services available that dig wells? I can think of one off the top of my head. I see the awesomeness of the idea, but I’m just wondering if all the other options have been exhausted. I’m not sure how deep one would have to go, with all involved in that, but wells have been dug well before technology has automated/powered those systems. If this were to be the very last option, then I would do the best I could with what I had. If there were other options, much harder to accomplish but the water produced cleaner and safer, I would absolutely without question take the harder road keeping my eyes on the end of the road (safety). In the end, whichever way you decide to go, I hope it is successful, wildly, and your health benefits from the project as well as those in your sphere of influence.

  10. With a radius of 70cm the upper collector will collect approx 1.5L per mm of rain assuming optimum rain conditions, which would mean you would need 2 meters of rain to fill the system, which at that point the system would need to act as a boat. A gutter system on a 3m hut would probably produce approx 7L per mm of rain assuming optimal capture conditions.

  11. replacing the easy parts of a gutter with the hard parts. this is not brilliant. a tank made out of a tank is going to be easier from every perspective than a tank made out of pipe. a gutter with one downspout is gonna be easier than a gutter-pan with 14 of them. half a dozen water tight joints is going to be easier than 56 of them. a perimeter gutter is gonna be easier and more resillient than a central pan. especially given that this gutter system will overflow when the tank is full. africa is poor from colonial exploitation, not stupid.

    it’s tempting sometimes to think of pipe projects as cheap, but that’s only ever true if you have a pile sitting around already. if you have to go out and buy it for your project that just called for 168 feet of it, you’ll find it’s quite expensive. if you’re making a one-off hack, it might still make sense because the convenience is actually functional for you. you’re only making one so design/engineering costs are relatively large compared to construction costs, so you are willing to pay extra for a nice uniform predictable building material. mass deployment in impoverished areas is the opposite constraint.

    the simple fact of the matter is that if africa consideered 168 linear ft of high quality pipe per household to be affordable, it would have a permanent buried water distribution network already.

  12. Have a look at income and material prices.

    Average income in Ethiopa is about 936 US dollars. That’s 46500 Ethiopian Birr.


    Four inch PVC pipe is 1700Br for a single standard length.


    Assuming you only need one pipe for each of the supports shown in the sketch, that’s 14 * 1700 = 23800Br.

    That’s the income for a half a year – and you’d still need PVC glue, all the joints, the smaller pipes (and joints) for the internal connections, as well as the collector pan.

    That’s a mighty expensive framework for a (small) house, that won’t standup all that well to anything at all.

  13. Lots of cool overthinking up there! Having lived on tank water I don’t think it’s as complex people are thinking, I’m guessing because of the ick factor.

    Q: Rain that falls on a dusty roof, runs through dusty pipes? How can you drink that dirty water?
    A: You let the dirt settle in the bottom of the tank, and take the water out a few inches higher!

    The other thing people miss is that it’s a water capture AND STORAGE. The capture is good, but there’s more upside. It enables households sharing water, and water trading in greater quantities as the buyer has somewhere to store the water that’s bigger than a jug. It means it makes sense to drive a truck of water to a dry village as there somewhere to put enough to make a difference, it makes it viable to set up a large, leaky capture area to fill many houses when it does rain. It gives every house a tradeable commodity (water if you got it, storage if you don’t have it).

    Come on folks, think it through! Let’s lots of clever here, but it’s often constrained by an assumption that a thing is just a thing, and it needs to be perfect.. The things that make impact are things that enable other things – and those definitely seems to be one.

    1. Volume of water stored is only constrained by the number, type and cost of piping used. If there is concern about structural integrity of roof or gutter collection (and loss of volume of water). Drop the verticals or whole structure into the earth. Down side is the need for a pump unless your Good is already built on an elevation (natural or manmade mound). ** Increasing horizontals connections barring inhibitive costs of connections will add to rigidity and increase volume stored. *** Joining multiple Gojo / water units will help multiple storage volume. **** Building a Gojo unit like the one above with cleanouts at base will help flush sediment collecting in the pipes. **** Adding a collapsible/removable canopy (inverted umbrella, be creative ;) will help increase volume of water collected.

  14. Fiberglass reenforced resin lined mudbrick walls forming a double cylinder makes a lot more sense in terms of strength and use of plastics. It would also be strong enough to support the weight of a collection guter that also acts as a sand based water filter.

  15. I’m concerned by the about of seams on this and the need for sealing them without skilled knowledge and ready acccess to silicon sealants – if they are covered by structure you could easily have leaks in your house without immediatly knowing, causing structural damage.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.