EWB NY Professional Chapter in collaboration with The Happold Trust and AIA have organized a Lecture and Fund Raiser event at The Center for Architecture to benefit EWB NY projects in Usalama, Kenya and Belen, Peru. Thom Mayne, F.A.I.A., a highly acclaimed architect with recent projects such as Cooper Union building and Federal Tower in San Francisco, CA, presented a lecture on performalism. Such an informative and relevant presentation provided a great platform for design discussions regarding our work on the Belen Health Clinic and the library in Usalama.
The event was a great success, drawing well over 150 architecture professionals. The funds raised from the guests and sponsors of the event will play important role in fulfilling our project goals for 2010.
1.29.2010
11.16.2009
Please see below for a progressive narrative of construction of the Usalama School library. Funds for labor and construction were made possible by Global Power of Literacy.
Upon arrival in Kenya, Matt (left, engineer) and Charles (right, architect) spent two days with Fred Afwai of the Global Education Fund. After spending some time learning about Fred's CandleLight School and orphanage in Nairobi, Fred brought us down to Usalama to meet the community and help us with our first round of material purchases. At this time we met Jackson, the school Chairman and our primary contact with the community.
After exchanging formalities and submitting our order for materials, we drove the few miles down the road to the site where we quickly got to work. We laid out the plan 90deg. to our original design. While this was a little disheartening to the EWB team, as our ventilation strategy’s effectiveness was contingent on the building’s orientation to the prevailing wind, Jackson insisted that we place it where we did. He had other plans for future classrooms in the area. I was reluctant, but agreed – it was their library after all.
After making adjustments to the plan for new material costs and quantities, and a new understanding of square footage needed for the book storage, we redrew and laid out the plan. The layout itself was a good tool for discussion, as we walked through the projected space. After many nods of approval, we etched out the footprint for excavation of the footing.
The following day we arrived at 930am to see 30 or so parents beginning the dig. It was quite a large group to monitor, so Matt and Charles continued to measure throughout to ensure a level bottom surface. We also added a smaller channel down the middle of the site to support coming brick wall down the center of the structure.
The next day was the toughest day that we encountered. The main problem was that we didn’t start pouring the footing until 1pm. This prevented us from finishing until 730pm. This was a lot to ask of the parents - especially of the women ho are typically tasked with preparing dinner for their families. By the time we finished, everyone was exhausted. This was the first and only time we opted to pay the parents for their hard work. A total of 5000 shillings for approx. 30 volunteers.
The 3 reasons for our late start to the day was 1, we had to clear the site of the piles of dirt around the excavated channels (Matt and I later agreed that immediately after excavation, ALL dirt should be removed from the site). 2, The hardcore which arrived at the end of the day the day before, had not been broken down. (Hardcore is the volcanic glass/rock which is prevalent around the area and very strong). We started the day by placing the hardcore into the excavated channels, then proceeded to crush down the hardcore with sledge hammers. I had thought - and had been told - that the hardcore would be broken down at the end of the day the day before. However, it became clear that it is much better to break down the hardcore once it is in place. Our 3rd delay was that the gravel we received t
urned out to be a broken down product of the hardcore. It is a very porous stone that absorbs lots of water very quickly. Also, it is a very dusty stone. We opted to “wash” the stones to get rid of the absorptive dust. When transporting each wheelbarrow of gravel to the mixing stations, we doused the stones in water before dumping into the mixture. All of these factors delayed our begin to the pour. By 1pm we were pouring. We allowed the community to explain their process for mixing concrete. They typically mix the concrete on the ground. We tried using some plastic “tarp” (see photos) but it turned out to be ineffective.
Our original drawings called for a 2” layer of “blinding” (a course concrete mixture used to level the top surface of the hardcore) followed by 2 pours totaling a 6 - 8” footing. Once we started pouring however, we knew that not only would we have to finish that day, but we would have to limit ourselves to two pours. We opted to begin with a 2-3” initial layer, lay our rebar, and add a 4” pour on top. Matt designed, organized and bent the Y-10 rebar details during the first concrete pour; and by 4 pm we began our last pour. At the end of the day, 7pm or so, we paid the parents which left them very happy.
On Thursday July 16th and Friday the 17th, we laid the 9x9 stones for the foundation walls on the concrete footing. We opted to make only a single course width of 9x9 stones for the foundation walls as we decided that a double width would be overkill and too costly. These two days went fairly smoothly. During these two days we monitored the mason’s progress, an took careful note about how they worked. After laying the corners, they would lay a string between them to achieve a straight line.
On Saturday we filled in the cavities surrounding the foundation walls with soil and began leveling the terrain underneath the coming slab. This was immediately followed by the parents teaming up to distribute the hardcore around and within the foundation walls.
This was a sensitive moment within our design as we were monitoring the elevation of the hardcore to make sure it would work with the thickness of the coming slab. We needed to make sure that the final elevation of the slab would be a minimum of 2” above the highest point in the terrain. After distributing and hammering the hardcore into place, we decided we had reached an operable level to move forward with the concrete slab. Before the end of the day, we set up the formwork for the coming pour and got ready for the concrete.
The following work day we poured the entire slab. We learned our lesson from the pouring of the footing and made sure to get started earlier in the day. We made sure to set up a series of teams and to delegate responsibilities. Most of the women were on water duty. We had a constant need for water. The women were set up with the gravel, using the civs to take out the absorbent dust. Most of the men volunteers were on concrete mixing duty - shoveling sand, bringing bags of cement from the storage. With varying concrete mixtures from 1-2-4 to 1-2-3, we delegated the responsibility of counting each concrete batch’s ingredients to the eldest male volunteer. After the first pour we laid the light gauge welded wire mesh.
We allowed the slab to cure for approximately 2 ½ days. During this time we covered the entire slab using the black plastic material and the left over cement bags. The slab was watered 4 times a day - not only because of the arid climate, but also out of our concern for the poor absorptive quality of gravel.
As we laid out the location of the doors and windows, I indicated that I wanted small gaps in the first course to be positioned underneath the windows. Before departing for Kenya, I ordered a glass bottle cutting kit. This small device would score a small ring around the bottle. After applying a flame, then a bit of cold (room temp water) the bottle will split. I used the bottle tops and bottle middles (along with some mosquito mesh) to create fresh air openings into the space. As we prepared the bottle portions, we began to run low on bottles. At this point someone came by with a 
bag full of leftover pipe segments. We quickly cut them up and inserted them into place. Jackson and the masons weren’t quite sure what to make of these additions, but I assured them that they had a purpose. After explaining the natural ventilation strategy they were very excited and knew now that their library was going to be quite unique - and enthusiasm around the construction site began to grow.
The next 10 courses went up extremely quickly. Jackson insisted that the masons complete 3 courses per day in order to get paid. In order to stay on this schedule, we hired on a 3rd mason, Mr. Wambua. Matungi was head mason, and Peter rounded out the team. Each were paid 350 Shillings a day, and each mason’s assistant was paid 200 shillings a day. We tried to pay them at the end of each day - though having only large bills at times caused a few headaches. We kept close records of each payment - often within 100 or 200 of what was owed.
Constructing the ring beam around the top of the masonry wall was by far the most complex task undertaken. Matungi and the masons were knowledgeable of their local construction techniques and set to task making the form work for the concrete pour. We used pine (the cheapest) for the formwork. While I worked with the masons to select and cut the formwork, Matt was putting together the rebar cages.
We managed to pour half the ring beam by the end of the day - and allowed it to cure the following day before removing the formwork to place it on the other side.
The second day of pouring the ring beam turned into a bit of a fiasco. In order to stay on schedule, we had the masons working on the ring beam, and a large number of parent volunteers prepare the gravel and mix the concrete, help complete the rebar cages, and begin excavation for the exterior wall. There were too many people and too many tasks for Matt and I to oversee. The construction site became too hectic to be considered safe. After lunch, we slowed down the pace and asked Jackson to ask most of the parents to leave. We were still on schedule, but realized that getting ahead of ourselves would cause more problems that we could not foresee.
As we were finishing up the ring beam, I finalized the design for the trusses and had the first phase of cypress delivered to the site. Realizing that I could not simply present a drawing to the masons and expect each truss to be the same, I redrew the truss at 1 to 1 in the dirt. The masons then went to work building the trusses on top of the dirt drawing. At this point, Matt and I made a last minute change to the truss design. This not only improved the structural integrity of the trusses, but also minimized the amount of lateral bending that occurred when nailing the members together. We completed the first phase of trusses within two full days of work.
As we finalized the preparation of the trusses, the masons built up the gabled walls on the ends of the structure. Here we used the bottle bottoms from the fresh air intakes that we used on the bottom of the structure. I laid these out on top of the full scale drawing of the truss. While showing the dimensions of the courses, we came to a consensus of how to layout the “glass bottle bricks” and got to work. Over the next two days we finalized the majority of the masonry structure and located glass bricks.
After one more week of work, I was happy to see that not only were the trusses erected, but the interior walls had been finished. The ornamental bottles looked excellent, and the interior of the fresh air intakes had obviously been given much care. The precise work was very uplifting. Also, the windows and doors had been installed. (These were done prior to the interior finishing.) Approximately two weeks prior, Matt, Jackson and I went to Kalimani Enterprises in Kibwezi to order the windows 
and doors. I had taken the time to prepare a nice page of inked drawings for which Kennedy (the owner and operator) was pleased to see. He gave us a very good price (with help from Jackson’s bartering). I gave him the drawing to keep to ensure a exact product. Kennedy was one of the more charming subcontractors I encountered. He is very pleasant to talk to, was very happy to work with us to find a good price, and produced quality work. I recommend working with him as much as possible in the future. Not only was his work done well and on schedule, but his good nature makes working with him a pleasure. The only drawback is that he does not have a truck to transport his work. For this we coordinated with the hardware store down the street. We would place a small order with them and ask them to pick up Kennedy’s finished products on the way. They were happy to do so as long as our order was large enough to merit a delivery. (in the event that we needed a small delivery (i.e. 3 or 4 bags of cement), they charged a 300kes delivery charge.
Moving on to the exterior, we to work and began laying the foundation for the exterior seating. We dug a small, approx. 6” deep trench, 17” wide and placed both our remaining hardcore and remains from the dressing of the masonry. This was followed by a 2” concrete layer. We kept this to a minimum not only because of the small dead and live loads that would be exerted on the foundation, but also because we were running short on cement and money.
After allowing the foundation to cure for a day, we built up three courses of brick structure while being careful to accommodate the fresh air intakes (FAI’s) that were already in place. Before filling in the in the core of the seating, I cut strips of the left over plastic material that we had used to protect the slab during curing. The intention was to create a recycled waterproofing “flashing detail”. While the plastic material itself and become quite tattered with many openings, the gesture brought forth the opportunity to talk about water proofing strategies within the team. We then placed the “mud mortar” and stone core within void. Before placing the final course, I scrounged up some of the leftover rebar that we had laying around to create a basic rebar “bridge” detail over the FAI’s. After this last course and infill, we played a last layer of concrete about 1” thick. This would be followed in the final stages of completion with a finishing layer - a mixture of only water and cement.
While we were working the seating, we sent the third mason, Peter (A.K.A. “The Professor”) to begin work on the exterior finishes. Because Peter was by far the most precise of the masons, we made sure to put him on the tasks that required the most detail. He was a bit slower than the others, but diligent and consistent. He created 6” frames around the windows, ½” frames around the openings in the front feature wall and interior partition, and both the exterior “baseboard” and “frieze”.
Installation of the 2nd phase of the truss construction was a process that had to be precise and very methodic in order to achieve the consistent colonnade that our design called for. First of all, Jackson and others called out our wood columns as simply incorrect. Due to termite and ant colonies in the soil, we were informed that wood columns would not last. The only option would be to use black metal columns about 2” in diameter. These were not in our budget but very much required. After reviewing other options involving metal connections from footing to wood beam that proved too complex, we opted to take to financial hit and purchase the iron columns. The columns came in 20’ segments. We bought three and had them cut and welded appropriately for a small additional fee.
Preparation for the 2nd phase involved cutting the timber and digging the holes for the column footings. I insisted that all preparations were made before installation so that we could methodically move down the line. In order to precisely complete the colonnade, we attached the timber to phase 1 truss, poured the footing for the column, and attached to the column to the timber all at the same time. All the while I made sure to measure continuously. By delegating the tasks evenly, we were able to complete phase two of the roof structure within one full day. At one point, one of the columns was put in place incorrectly and off line with the previous three. Upon recognition, Jackson stepped in and insisted on removing the poured footing and essentially redoing the entire portion of the truss and column system. I was very happy to see his dedication to getting it right the first time. This reinforced my confidence in his strength as a leader, as I doubt that I would have had the weight to enforce additional hour of hard labor upon the crew. After completing the columns, we all began to see the structure coming together - and the final product was now within sight.
Installing the roof panels took 1.5 days. After having ordered the roof panels, roofing nails, and their arrival, we got to work. Day 1 of the roof was a very bright and sunny day. I would suggest that if it is possible on future projects, wait to install the roof panels on cloudy days. The reflection of the sun off the panels was clearly extremely unpleasant for the laborers. It was very hard on their eyes and I am sure that it was very hot. The following day was much cloudier and work went much faster. Installation and overlapping of the roof panels yielded two major lessons. First, we ordered a series of 2m, 2.5m, and 3m panels to span the dimensions of the roof. Our dimensions provided a close call between ordering 3m panels or increasing our order of 2.5m panels. It was recommended to overlap the panels (from end to end) anywhere from 6” to 18”. Because using 2.5m panels would have yielded a 4” overlap, we opted to go for ordering 3m panels. This created an overlap between 18” and 24”. I figured that this would not be a problem. With the thought of “better safe than sorry”, and a guarantee to stay on schedule, the small budget increase seemed to be worth the scheduling advantages. However, as we got roughly halfway through the roof installation, it became clear that a large overlap prevents the minute corrections needed to ensure parallel panels across the roof. These design constraints should be considered in the initial stages of schematic design - as the roofing material is among the most expensive. Second, during initial material quantity estimates, we estimated that each roof panel would overlap 2” with its adjacent panel. This was correct, but incorrect - as the panels in the middle overlap twice. This small detail caused us to be short 7 roof panels. This was a costly mistake - though we were able to add in our order of panels to an upcoming delivery, and they arrived shortly after the mistake was discovered. I would like to reiterate that the roof panel layout and the quantities should be a primary consideration in the first stages of design.
The final stage to the roof was to install the roof ridges over the top of the structure - binding the two faces of the roof. I opted to elevate the two center ridges with some leftover 2”x2” wood to create a roof vent. Coupled with the FAI’s along the first course of masonry, the roof vent completed the strategy to promote the stack effect. In the event that electricity in installed in the space, mosquito wire could easily be retrofitted over this opening.
Finishing the floor was a one day project during which Jackson took control and helped teach me the pouring process. After spraying gladiator (ant repellant) around the perimeter of the space, we got to work laying the first two layers. The first layer was a binding mixture by which we sprinkled pure cement over the rough concrete surface. This was followed by small amounts of water and a broom to spread the mixture evenly over the surface.
As this process moved across the floor, the following mason began distributing a 1-3 mortar mix across the floor. Using shovels, levels, and wood T-shaped tools for compacting, we spread an even 1” thick layer across the floor. By mid afternoon we had completed these two steps and moved on to the final stage - the floor finish. This was a pure mixture of water and cement. The masons took great care in distributing this final layer. After 2 or 3 hours we completed the final layer.
We then allowed it to cure for 15 to 20 minutes, then sent the masons in for a final smoothing of the floor surface. At the end of the day we poured a healthy amount of water over the floor and quarantined it off from everyone for the next day and a half.
By this time we had already begun excavation for the coming exterior wall. At this point we did not know how we were going to build the wall (or how we were going to pay for it). In our initial design we had a kind of wood screen wrapping around the exterior with the intention of talking to the community. Around this time the parents who had help excavate began asking questions. They did not understand why we were digging for more construction when we had already made so much progress on the main structure. I took this as an opportunity and asked Jackson to set up a meeting at which I could explain not only our vision for the final product and how the library would be used, but also start up a dialogue with the parents about how exactly to
complete this mysterious exterior wall. After showing them the plans and renderings from our initial design process, I explained our dilemma for the exterior wall. I brought up examples of wood structures and fences made from leftover crops from around the area. The parents however indicated that such a construction would not last the rainy season - and the best option was to make it out of brick. I suddenly felt a bit cornered, knowing that such an undertaking would require not only large amounts of sand and cement ($$) but would take much more time. However, the community was right. The only way to build a wall that would last the life of the main structure would be to make it out of brick. I went home that night to design the wall structure with minimal materials and maximum strength.
Upon laying the foundation for the brick wall, we used a similar strategy as was used for the exterior seating along the main structure. Laying a hardcore base of only 8-10”, we then laid a 2” layer of concrete to serve as our base for brick work. I consistently measured and opted to step up the foundation twice at major elevation changes. The foundation steps matched the dimensions of a brick, so the brick work above would be able to transition smoothly over the foundation along the rising topographic changes. Again, using the same detail as the seating, we opted to use a mud mortar and leftover stone mixture to infill within the space between the two walls of brick.
Once we reached three full courses (counted from the highest portion of the existing topography) we began building up the wall and laid a 2” concrete layer for the seating. In order to ensure a 
strong construction, I located buttresses on both the exterior and interior of the wall. Further, I placed misomo (or hoop iron, reinforcing) at every other course. By doing this I felt comfortable that the wall, with evenly distributed 6” square openings, would be extremely strong. As the wall grew however, we began to run short on reinforcing and available funds. As a result, I began using any left over rebar that I could find to use in the upper most portions of the structure. This seemed to be an even stronger option and resulted in a more solid final structure.
The masons continued to move forward at this point and began to finish the interior surface of the wall with a mortar mixture. (Because of the irregularity of the bricks the masons can only justify one side of the brick surface, hence the plastering over of one of the sides.) Once this was completed, we placed a smoothing layer over the seating (a mixture of water and cement, similar to the finishing of the floor).
As the wall was being completed, I worked with Jackson to organize some volunteers to work on adjusting the topography around the site to control water precipitation control. We used much of the soil surrounding the site from the initial foundation dig to surround the exterior. Also, for the space defined within the brick wall, we adjusted the topography direct all water run off to center of the space to irrigate the coming tree.
With the interior walls and floors finished, we began to adjust our sites on the interior furnishings. Through Jackson’s recommendation, I was introduced to Mutuku, a local carpenter from the Usalama community, to construct the bookshelves. After exchanging formalities, I took Mutuku 
through the plan and through the space itself. A few days later, Mutuku came forward with a modest bid. After a bit of refining and agreements on cost and materials, we put in our order at the hardware store and he got to work. I instructed Mutuku to first complete the checkout counter so we would be able to construct the security wall on top as he moved on to the bookshelves. He did so within one day.
The wood needed for the framework however, was not within our budget. For this, Jackson suggested we use some of the wood from the existing bookshelves in the administration building. We relocated some of the items on the shelves and removed only the wood that we needed. We laid out the wood over the full scale dimensions of the check out counter and soon began installation.
Unfortunately we ran into a wood shortage at the hardware store and we were unable to finish the bookshelves before I left. After finishing roughly half of the bookshelves, I charged Mutuku with the task of continuously pestering the hardware store for the remaining wood (which we had already paid for). All of the cypress comes from Tanzania, and the venders there were quite unreliable in providing accurate delivery times. I debated the possibility of delaying my return the states, but opted to give Jackson the remaining funds for Mutuku’s labor. This turned out to be a good decision, as the wood arrived a full 4 weeks after my departure.
As the interior of the space came to completion, Jackson’s recommended subcontractor for the glass installation. After a couple hours of measuring each window, he came forward with a modest bid for glass, putty, and the handles. (Be sure to include these final costs when budgeting out the initial numbers for the cost of the windows. The window frames are made by a separate contractor.)
After installing the fascia boards around the perimeter of the roof, and painting the primer, we began the final stages of painting. This actually became one of the biggest headaches.
After spending SO much time on the finishes - making perfectly straight lines for the window frames and precise work around the bottles - the crew was doing such a terrible job painting. They were throwing paint all over the structure - and I quickly became extremely strict. Muolu, one of the assistants, was doing such a horrible job that I got on his case quite a bit. Eventually he ditched his brush and made one out of a stick to be more precise. After finishing up the exterior, we used the remaining paint on the interior security wall.
In addition to leaving Mutuku’s final costs with Jackson, I also left funds for about 3 days of work for one mason (Matungi) and one assistant. This, along with the remaining 2.5 bags cement, would be enough for the finishing touches of the brick wall. I made sure that when presenting this money to Jackson, to also have Matungi present. This ensure that he would be paid. I was not worried that Jackson would embezzle the remaining funds. He worked very hard and exhibited a strong dedication to finishing the project to the highest standard. Further, the parents and other members of the community were well aware of how the library should be finished. I felt comfortable leaving such a sum with him. For all of his hard work, and in agreement with Global Power of Literacy, I paid Jackson a total of 23,000kes. He was very happy with this sum - and left our (EWB and GPOL) relationship with him in very good standing. For any further work in the area it will be very important to include Jackson. He is a very hard worker, has indispensable contacts, and is a good friend.
Global Power of Literacy has hailed this project as a complete success. The project (and bookshelves) have been completed, and the books from Global Education Fund are set in early November. While we did end up going over budget by a few hundred dollars, GPOL are very happy to have worked with Engineers Without Borders. Mwikali’s gift, the donor that initiated the relationship between Usalama and EWB, will continue to send aid to the community. They are doing this in the form of uniforms for the students, education seminars for the community, and distributing mosquito nets. They are also working on completing the water pipeline though AMREF. This project is very slow moving, though they are hoping to fund further construction upon completion of the water project.
Upon arrival in Kenya, Matt (left, engineer) and Charles (right, architect) spent two days with Fred Afwai of the Global Education Fund. After spending some time learning about Fred's CandleLight School and orphanage in Nairobi, Fred brought us down to Usalama to meet the community and help us with our first round of material purchases. At this time we met Jackson, the school Chairman and our primary contact with the community.After exchanging formalities and submitting our order for materials, we drove the few miles down the road to the site where we quickly got to work. We laid out the plan 90deg. to our original design. While this was a little disheartening to the EWB team, as our ventilation strategy’s effectiveness was contingent on the building’s orientation to the prevailing wind, Jackson insisted that we place it where we did. He had other plans for future classrooms in the area. I was reluctant, but agreed – it was their library after all.
After making adjustments to the plan for new material costs and quantities, and a new understanding of square footage needed for the book storage, we redrew and laid out the plan. The layout itself was a good tool for discussion, as we walked through the projected space. After many nods of approval, we etched out the footprint for excavation of the footing.
The following day we arrived at 930am to see 30 or so parents beginning the dig. It was quite a large group to monitor, so Matt and Charles continued to measure throughout to ensure a level bottom surface. We also added a smaller channel down the middle of the site to support coming brick wall down the center of the structure.
The next day was the toughest day that we encountered. The main problem was that we didn’t start pouring the footing until 1pm. This prevented us from finishing until 730pm. This was a lot to ask of the parents - especially of the women ho are typically tasked with preparing dinner for their families. By the time we finished, everyone was exhausted. This was the first and only time we opted to pay the parents for their hard work. A total of 5000 shillings for approx. 30 volunteers.
The 3 reasons for our late start to the day was 1, we had to clear the site of the piles of dirt around the excavated channels (Matt and I later agreed that immediately after excavation, ALL dirt should be removed from the site). 2, The hardcore which arrived at the end of the day the day before, had not been broken down. (Hardcore is the volcanic glass/rock which is prevalent around the area and very strong). We started the day by placing the hardcore into the excavated channels, then proceeded to crush down the hardcore with sledge hammers. I had thought - and had been told - that the hardcore would be broken down at the end of the day the day before. However, it became clear that it is much better to break down the hardcore once it is in place. Our 3rd delay was that the gravel we received t
urned out to be a broken down product of the hardcore. It is a very porous stone that absorbs lots of water very quickly. Also, it is a very dusty stone. We opted to “wash” the stones to get rid of the absorptive dust. When transporting each wheelbarrow of gravel to the mixing stations, we doused the stones in water before dumping into the mixture. All of these factors delayed our begin to the pour. By 1pm we were pouring. We allowed the community to explain their process for mixing concrete. They typically mix the concrete on the ground. We tried using some plastic “tarp” (see photos) but it turned out to be ineffective.
Our original drawings called for a 2” layer of “blinding” (a course concrete mixture used to level the top surface of the hardcore) followed by 2 pours totaling a 6 - 8” footing. Once we started pouring however, we knew that not only would we have to finish that day, but we would have to limit ourselves to two pours. We opted to begin with a 2-3” initial layer, lay our rebar, and add a 4” pour on top. Matt designed, organized and bent the Y-10 rebar details during the first concrete pour; and by 4 pm we began our last pour. At the end of the day, 7pm or so, we paid the parents which left them very happy.
On Thursday July 16th and Friday the 17th, we laid the 9x9 stones for the foundation walls on the concrete footing. We opted to make only a single course width of 9x9 stones for the foundation walls as we decided that a double width would be overkill and too costly. These two days went fairly smoothly. During these two days we monitored the mason’s progress, an took careful note about how they worked. After laying the corners, they would lay a string between them to achieve a straight line.
On Saturday we filled in the cavities surrounding the foundation walls with soil and began leveling the terrain underneath the coming slab. This was immediately followed by the parents teaming up to distribute the hardcore around and within the foundation walls.
This was a sensitive moment within our design as we were monitoring the elevation of the hardcore to make sure it would work with the thickness of the coming slab. We needed to make sure that the final elevation of the slab would be a minimum of 2” above the highest point in the terrain. After distributing and hammering the hardcore into place, we decided we had reached an operable level to move forward with the concrete slab. Before the end of the day, we set up the formwork for the coming pour and got ready for the concrete.
The following work day we poured the entire slab. We learned our lesson from the pouring of the footing and made sure to get started earlier in the day. We made sure to set up a series of teams and to delegate responsibilities. Most of the women were on water duty. We had a constant need for water. The women were set up with the gravel, using the civs to take out the absorbent dust. Most of the men volunteers were on concrete mixing duty - shoveling sand, bringing bags of cement from the storage. With varying concrete mixtures from 1-2-4 to 1-2-3, we delegated the responsibility of counting each concrete batch’s ingredients to the eldest male volunteer. After the first pour we laid the light gauge welded wire mesh.
We allowed the slab to cure for approximately 2 ½ days. During this time we covered the entire slab using the black plastic material and the left over cement bags. The slab was watered 4 times a day - not only because of the arid climate, but also out of our concern for the poor absorptive quality of gravel.
As we laid out the location of the doors and windows, I indicated that I wanted small gaps in the first course to be positioned underneath the windows. Before departing for Kenya, I ordered a glass bottle cutting kit. This small device would score a small ring around the bottle. After applying a flame, then a bit of cold (room temp water) the bottle will split. I used the bottle tops and bottle middles (along with some mosquito mesh) to create fresh air openings into the space. As we prepared the bottle portions, we began to run low on bottles. At this point someone came by with a 
bag full of leftover pipe segments. We quickly cut them up and inserted them into place. Jackson and the masons weren’t quite sure what to make of these additions, but I assured them that they had a purpose. After explaining the natural ventilation strategy they were very excited and knew now that their library was going to be quite unique - and enthusiasm around the construction site began to grow.
The next 10 courses went up extremely quickly. Jackson insisted that the masons complete 3 courses per day in order to get paid. In order to stay on this schedule, we hired on a 3rd mason, Mr. Wambua. Matungi was head mason, and Peter rounded out the team. Each were paid 350 Shillings a day, and each mason’s assistant was paid 200 shillings a day. We tried to pay them at the end of each day - though having only large bills at times caused a few headaches. We kept close records of each payment - often within 100 or 200 of what was owed.
Constructing the ring beam around the top of the masonry wall was by far the most complex task undertaken. Matungi and the masons were knowledgeable of their local construction techniques and set to task making the form work for the concrete pour. We used pine (the cheapest) for the formwork. While I worked with the masons to select and cut the formwork, Matt was putting together the rebar cages.
We managed to pour half the ring beam by the end of the day - and allowed it to cure the following day before removing the formwork to place it on the other side.The second day of pouring the ring beam turned into a bit of a fiasco. In order to stay on schedule, we had the masons working on the ring beam, and a large number of parent volunteers prepare the gravel and mix the concrete, help complete the rebar cages, and begin excavation for the exterior wall. There were too many people and too many tasks for Matt and I to oversee. The construction site became too hectic to be considered safe. After lunch, we slowed down the pace and asked Jackson to ask most of the parents to leave. We were still on schedule, but realized that getting ahead of ourselves would cause more problems that we could not foresee.
As we were finishing up the ring beam, I finalized the design for the trusses and had the first phase of cypress delivered to the site. Realizing that I could not simply present a drawing to the masons and expect each truss to be the same, I redrew the truss at 1 to 1 in the dirt. The masons then went to work building the trusses on top of the dirt drawing. At this point, Matt and I made a last minute change to the truss design. This not only improved the structural integrity of the trusses, but also minimized the amount of lateral bending that occurred when nailing the members together. We completed the first phase of trusses within two full days of work.
As we finalized the preparation of the trusses, the masons built up the gabled walls on the ends of the structure. Here we used the bottle bottoms from the fresh air intakes that we used on the bottom of the structure. I laid these out on top of the full scale drawing of the truss. While showing the dimensions of the courses, we came to a consensus of how to layout the “glass bottle bricks” and got to work. Over the next two days we finalized the majority of the masonry structure and located glass bricks.
After one more week of work, I was happy to see that not only were the trusses erected, but the interior walls had been finished. The ornamental bottles looked excellent, and the interior of the fresh air intakes had obviously been given much care. The precise work was very uplifting. Also, the windows and doors had been installed. (These were done prior to the interior finishing.) Approximately two weeks prior, Matt, Jackson and I went to Kalimani Enterprises in Kibwezi to order the windows 
and doors. I had taken the time to prepare a nice page of inked drawings for which Kennedy (the owner and operator) was pleased to see. He gave us a very good price (with help from Jackson’s bartering). I gave him the drawing to keep to ensure a exact product. Kennedy was one of the more charming subcontractors I encountered. He is very pleasant to talk to, was very happy to work with us to find a good price, and produced quality work. I recommend working with him as much as possible in the future. Not only was his work done well and on schedule, but his good nature makes working with him a pleasure. The only drawback is that he does not have a truck to transport his work. For this we coordinated with the hardware store down the street. We would place a small order with them and ask them to pick up Kennedy’s finished products on the way. They were happy to do so as long as our order was large enough to merit a delivery. (in the event that we needed a small delivery (i.e. 3 or 4 bags of cement), they charged a 300kes delivery charge.Moving on to the exterior, we to work and began laying the foundation for the exterior seating. We dug a small, approx. 6” deep trench, 17” wide and placed both our remaining hardcore and remains from the dressing of the masonry. This was followed by a 2” concrete layer. We kept this to a minimum not only because of the small dead and live loads that would be exerted on the foundation, but also because we were running short on cement and money.
After allowing the foundation to cure for a day, we built up three courses of brick structure while being careful to accommodate the fresh air intakes (FAI’s) that were already in place. Before filling in the in the core of the seating, I cut strips of the left over plastic material that we had used to protect the slab during curing. The intention was to create a recycled waterproofing “flashing detail”. While the plastic material itself and become quite tattered with many openings, the gesture brought forth the opportunity to talk about water proofing strategies within the team. We then placed the “mud mortar” and stone core within void. Before placing the final course, I scrounged up some of the leftover rebar that we had laying around to create a basic rebar “bridge” detail over the FAI’s. After this last course and infill, we played a last layer of concrete about 1” thick. This would be followed in the final stages of completion with a finishing layer - a mixture of only water and cement.
While we were working the seating, we sent the third mason, Peter (A.K.A. “The Professor”) to begin work on the exterior finishes. Because Peter was by far the most precise of the masons, we made sure to put him on the tasks that required the most detail. He was a bit slower than the others, but diligent and consistent. He created 6” frames around the windows, ½” frames around the openings in the front feature wall and interior partition, and both the exterior “baseboard” and “frieze”.
Installation of the 2nd phase of the truss construction was a process that had to be precise and very methodic in order to achieve the consistent colonnade that our design called for. First of all, Jackson and others called out our wood columns as simply incorrect. Due to termite and ant colonies in the soil, we were informed that wood columns would not last. The only option would be to use black metal columns about 2” in diameter. These were not in our budget but very much required. After reviewing other options involving metal connections from footing to wood beam that proved too complex, we opted to take to financial hit and purchase the iron columns. The columns came in 20’ segments. We bought three and had them cut and welded appropriately for a small additional fee.
Preparation for the 2nd phase involved cutting the timber and digging the holes for the column footings. I insisted that all preparations were made before installation so that we could methodically move down the line. In order to precisely complete the colonnade, we attached the timber to phase 1 truss, poured the footing for the column, and attached to the column to the timber all at the same time. All the while I made sure to measure continuously. By delegating the tasks evenly, we were able to complete phase two of the roof structure within one full day. At one point, one of the columns was put in place incorrectly and off line with the previous three. Upon recognition, Jackson stepped in and insisted on removing the poured footing and essentially redoing the entire portion of the truss and column system. I was very happy to see his dedication to getting it right the first time. This reinforced my confidence in his strength as a leader, as I doubt that I would have had the weight to enforce additional hour of hard labor upon the crew. After completing the columns, we all began to see the structure coming together - and the final product was now within sight.
Installing the roof panels took 1.5 days. After having ordered the roof panels, roofing nails, and their arrival, we got to work. Day 1 of the roof was a very bright and sunny day. I would suggest that if it is possible on future projects, wait to install the roof panels on cloudy days. The reflection of the sun off the panels was clearly extremely unpleasant for the laborers. It was very hard on their eyes and I am sure that it was very hot. The following day was much cloudier and work went much faster. Installation and overlapping of the roof panels yielded two major lessons. First, we ordered a series of 2m, 2.5m, and 3m panels to span the dimensions of the roof. Our dimensions provided a close call between ordering 3m panels or increasing our order of 2.5m panels. It was recommended to overlap the panels (from end to end) anywhere from 6” to 18”. Because using 2.5m panels would have yielded a 4” overlap, we opted to go for ordering 3m panels. This created an overlap between 18” and 24”. I figured that this would not be a problem. With the thought of “better safe than sorry”, and a guarantee to stay on schedule, the small budget increase seemed to be worth the scheduling advantages. However, as we got roughly halfway through the roof installation, it became clear that a large overlap prevents the minute corrections needed to ensure parallel panels across the roof. These design constraints should be considered in the initial stages of schematic design - as the roofing material is among the most expensive. Second, during initial material quantity estimates, we estimated that each roof panel would overlap 2” with its adjacent panel. This was correct, but incorrect - as the panels in the middle overlap twice. This small detail caused us to be short 7 roof panels. This was a costly mistake - though we were able to add in our order of panels to an upcoming delivery, and they arrived shortly after the mistake was discovered. I would like to reiterate that the roof panel layout and the quantities should be a primary consideration in the first stages of design. 
The final stage to the roof was to install the roof ridges over the top of the structure - binding the two faces of the roof. I opted to elevate the two center ridges with some leftover 2”x2” wood to create a roof vent. Coupled with the FAI’s along the first course of masonry, the roof vent completed the strategy to promote the stack effect. In the event that electricity in installed in the space, mosquito wire could easily be retrofitted over this opening.Finishing the floor was a one day project during which Jackson took control and helped teach me the pouring process. After spraying gladiator (ant repellant) around the perimeter of the space, we got to work laying the first two layers. The first layer was a binding mixture by which we sprinkled pure cement over the rough concrete surface. This was followed by small amounts of water and a broom to spread the mixture evenly over the surface.
As this process moved across the floor, the following mason began distributing a 1-3 mortar mix across the floor. Using shovels, levels, and wood T-shaped tools for compacting, we spread an even 1” thick layer across the floor. By mid afternoon we had completed these two steps and moved on to the final stage - the floor finish. This was a pure mixture of water and cement. The masons took great care in distributing this final layer. After 2 or 3 hours we completed the final layer.
We then allowed it to cure for 15 to 20 minutes, then sent the masons in for a final smoothing of the floor surface. At the end of the day we poured a healthy amount of water over the floor and quarantined it off from everyone for the next day and a half.By this time we had already begun excavation for the coming exterior wall. At this point we did not know how we were going to build the wall (or how we were going to pay for it). In our initial design we had a kind of wood screen wrapping around the exterior with the intention of talking to the community. Around this time the parents who had help excavate began asking questions. They did not understand why we were digging for more construction when we had already made so much progress on the main structure. I took this as an opportunity and asked Jackson to set up a meeting at which I could explain not only our vision for the final product and how the library would be used, but also start up a dialogue with the parents about how exactly to
complete this mysterious exterior wall. After showing them the plans and renderings from our initial design process, I explained our dilemma for the exterior wall. I brought up examples of wood structures and fences made from leftover crops from around the area. The parents however indicated that such a construction would not last the rainy season - and the best option was to make it out of brick. I suddenly felt a bit cornered, knowing that such an undertaking would require not only large amounts of sand and cement ($$) but would take much more time. However, the community was right. The only way to build a wall that would last the life of the main structure would be to make it out of brick. I went home that night to design the wall structure with minimal materials and maximum strength.
Upon laying the foundation for the brick wall, we used a similar strategy as was used for the exterior seating along the main structure. Laying a hardcore base of only 8-10”, we then laid a 2” layer of concrete to serve as our base for brick work. I consistently measured and opted to step up the foundation twice at major elevation changes. The foundation steps matched the dimensions of a brick, so the brick work above would be able to transition smoothly over the foundation along the rising topographic changes. Again, using the same detail as the seating, we opted to use a mud mortar and leftover stone mixture to infill within the space between the two walls of brick.
Once we reached three full courses (counted from the highest portion of the existing topography) we began building up the wall and laid a 2” concrete layer for the seating. In order to ensure a 
strong construction, I located buttresses on both the exterior and interior of the wall. Further, I placed misomo (or hoop iron, reinforcing) at every other course. By doing this I felt comfortable that the wall, with evenly distributed 6” square openings, would be extremely strong. As the wall grew however, we began to run short on reinforcing and available funds. As a result, I began using any left over rebar that I could find to use in the upper most portions of the structure. This seemed to be an even stronger option and resulted in a more solid final structure.
The masons continued to move forward at this point and began to finish the interior surface of the wall with a mortar mixture. (Because of the irregularity of the bricks the masons can only justify one side of the brick surface, hence the plastering over of one of the sides.) Once this was completed, we placed a smoothing layer over the seating (a mixture of water and cement, similar to the finishing of the floor).As the wall was being completed, I worked with Jackson to organize some volunteers to work on adjusting the topography around the site to control water precipitation control. We used much of the soil surrounding the site from the initial foundation dig to surround the exterior. Also, for the space defined within the brick wall, we adjusted the topography direct all water run off to center of the space to irrigate the coming tree.
With the interior walls and floors finished, we began to adjust our sites on the interior furnishings. Through Jackson’s recommendation, I was introduced to Mutuku, a local carpenter from the Usalama community, to construct the bookshelves. After exchanging formalities, I took Mutuku 
through the plan and through the space itself. A few days later, Mutuku came forward with a modest bid. After a bit of refining and agreements on cost and materials, we put in our order at the hardware store and he got to work. I instructed Mutuku to first complete the checkout counter so we would be able to construct the security wall on top as he moved on to the bookshelves. He did so within one day.
The wood needed for the framework however, was not within our budget. For this, Jackson suggested we use some of the wood from the existing bookshelves in the administration building. We relocated some of the items on the shelves and removed only the wood that we needed. We laid out the wood over the full scale dimensions of the check out counter and soon began installation.
Unfortunately we ran into a wood shortage at the hardware store and we were unable to finish the bookshelves before I left. After finishing roughly half of the bookshelves, I charged Mutuku with the task of continuously pestering the hardware store for the remaining wood (which we had already paid for). All of the cypress comes from Tanzania, and the venders there were quite unreliable in providing accurate delivery times. I debated the possibility of delaying my return the states, but opted to give Jackson the remaining funds for Mutuku’s labor. This turned out to be a good decision, as the wood arrived a full 4 weeks after my departure.
As the interior of the space came to completion, Jackson’s recommended subcontractor for the glass installation. After a couple hours of measuring each window, he came forward with a modest bid for glass, putty, and the handles. (Be sure to include these final costs when budgeting out the initial numbers for the cost of the windows. The window frames are made by a separate contractor.)
After installing the fascia boards around the perimeter of the roof, and painting the primer, we began the final stages of painting. This actually became one of the biggest headaches.
After spending SO much time on the finishes - making perfectly straight lines for the window frames and precise work around the bottles - the crew was doing such a terrible job painting. They were throwing paint all over the structure - and I quickly became extremely strict. Muolu, one of the assistants, was doing such a horrible job that I got on his case quite a bit. Eventually he ditched his brush and made one out of a stick to be more precise. After finishing up the exterior, we used the remaining paint on the interior security wall.
In addition to leaving Mutuku’s final costs with Jackson, I also left funds for about 3 days of work for one mason (Matungi) and one assistant. This, along with the remaining 2.5 bags cement, would be enough for the finishing touches of the brick wall. I made sure that when presenting this money to Jackson, to also have Matungi present. This ensure that he would be paid. I was not worried that Jackson would embezzle the remaining funds. He worked very hard and exhibited a strong dedication to finishing the project to the highest standard. Further, the parents and other members of the community were well aware of how the library should be finished. I felt comfortable leaving such a sum with him. For all of his hard work, and in agreement with Global Power of Literacy, I paid Jackson a total of 23,000kes. He was very happy with this sum - and left our (EWB and GPOL) relationship with him in very good standing. For any further work in the area it will be very important to include Jackson. He is a very hard worker, has indispensable contacts, and is a good friend.
Global Power of Literacy has hailed this project as a complete success. The project (and bookshelves) have been completed, and the books from Global Education Fund are set in early November. While we did end up going over budget by a few hundred dollars, GPOL are very happy to have worked with Engineers Without Borders. Mwikali’s gift, the donor that initiated the relationship between Usalama and EWB, will continue to send aid to the community. They are doing this in the form of uniforms for the students, education seminars for the community, and distributing mosquito nets. They are also working on completing the water pipeline though AMREF. This project is very slow moving, though they are hoping to fund further construction upon completion of the water project.
Subscribe to:
Posts (Atom)