Bonaccorso / Low-cost DIY Upgrade Strategies for Improved Comfort in Poor Brazilian Houses in Hot Climates

Low-cost DIY Upgrade Strategies for Improved Comfort in Poor Brazilian Houses in Hot Climates Process and Results

Author: Nadir Bonaccorso, Universidade de Èvora

Supervisor: Guilherme Carrilho da Graça, Lisbon University – Faculty of Science; Pedro Matos Gameiro, Évora University – Faculty of Architecture

Research stage: Final doctoral stage

Category: Paper


The 2020 Global Poverty update report 1 estimated that 46.2 % of the world population lived on less than 5.50US$ per day, with a slight increase of 0.2 % since 2015, where 12 % lived in slums 2. Poverty is a multi-dimensional problem related to monthly income, access to a job, education, health care, and social life. These effects, in Brazil, can be detected by looking at the urban tissue where, from one side, we see the typology of the closed condominium, an illusion of living on a private island, excluded from the outside chaos 3, while on the other, the informal constructions, self-built, filling the urban voids 4. Today, poverty reduction suffers a reversal tendency due to the Covid-19 pandemic and its global economic effects, armed conflict, and climate change 5. In hot countries around the world, this type of poor construction shows differences marked by culture, natural resources, and industrial development, which may influence indoor behavior 6. Low thermal mass envelope in poor houses, in high outdoor temperatures, tends to warm up the indoors beside the levels of comfort 7. Thus, this unsolved and complex spreading reality may be urgently addressed.


For the pursuit of the goal have been used Research by Design (RbD) driven strategy 8, complemented by a scientific methodology, for experimentation and analysis. Both qualitative and quantitative approaches informed and defined the extent and the limits of the investigation hypothesis. A constant realignment of the construction of the artefacts, based on trial and error, was followed to validate the proposed solutions 9. The research is divided into two phases organized into three different stages 10, repeated two times [ 1 ].

Map of the RbD methodology travelled.

Figure 1: Map of the RbD methodology travelled.

In the first stage, the literature review was complemented by in-situ research. This approach informed the question formulation, allowing the development of the first Design stage, characterized by the design project of a solar chimney. The first design proposal for improving comfort sensation by evapotranspiration, showed uncertain results, suggesting the implementation of different strategies to be explored individually and combined, compelling to return to the pre-design stage to re-design and verify the effectiveness of the new proposals. To overcome the limits of the results of the campaigns has been applied a dynamic simulation, using EnergyPlus v.8.7 11, to validate the model and to evaluate effectiveness in multiple hot climates.

Brazilian experience phase

On a planet of slums, how is a poor house? How is built? How is lived? How is its thermal space? How much could be the budget for improvements? What strategies may be effective to improve comfort in such communities? The first phase objective was set up to answer the many questions raised.

The first phase was held in the cities of Fortaleza and Sobral (1+2 years), both in Brazil. Establishing contact with a poor community during the first months seemed an improbable goal to achieve due to the violent and suspicious environment in Fortaleza 12. Yet, I could find a job (paradoxically), working on the front line of a Governmental expropriation in 5 slums. Each had a formal representative who maintained the team safe, guiding us through the community during sunlight and mediating with the residents. I visited more than 200 houses and spoke with their owners, observing the traces of living.

The poor permanent house has a spontaneous genesis and is built outside the regulations. Starting as a unique piece, that will be extended by the householder's means and time, the residence is always evolving, even when achieving the last stage of the shelter (the brick house, focus of this investigation), unfinished and ephemeral, like life itself 13. The constructive technique is the most rudimentary, and the used materials are the cheapest 3. The most common type of envelope is made by a single layer of brick commonly shared with the neighbour and rarely plastered on both sides, which forms the muscles house without bones [ 2 ]. On the walls is settled a wood structure, on which a ceramic roof tile lies directly, covering the internal space [ Table 1 ].

House construction composition in Brazil. (Source: IBGE. Residential Survey by sample of 2019, in Portuguese.)

Figure 2: House construction composition in Brazil. (Source: IBGE. Residential Survey by sample of 2019, in Portuguese.)

The internal distribution is organized by the structure support. The entrance is used, at night, as a shelter for the bike or the motorcycle owned. The rooms, are often open on the roof level, allowing air circulation. The bathroom is located close to the outdoor sewage solution, with no relationship with any internal space. Houses have hooks on the walls to support the hammocks. Each space support multiple usages. Most owners stores used or found construction materials to be employed in the future.

If Occidental architecture works on the transformation of the environment to adapt it to human’s necessity, this survival architecture tries to contain transformations, implementing the necessary for sufficient favourable surviving conditions 14.

D. Expedito poor neighborhood: a) context b) envelope construction c) roof composition.

Figure 3: D. Expedito poor neighborhood: a) context b) envelope construction c) roof composition.

Soon I understood that I have to reformulate my RbD approach to adapt and operate in this context, using what Levi-Strauss called the “first science”: the bricolage 15. In this type of fragmented construction, I had also to re-evaluate the concepts of (dis)order and chronology. Hypothesis pre-requisite was sketched: the roof, responsible for 50 to 70% of the indoor thermal behaviour 16, was set as the envelope side to be improved; all strategies should be efficient, low-cost and suitable for self-construction in a DIY mode.

In Sobral, supported by the University of Architecture INTA, I started contacts with the resident association of the community of D. Expedito, located on the left side of the Acaraú river, upgraded in 2000 with basic sanitation and road infrastructure 17. I was allowed to survey five houses and interview the owners. Houses overheat during the day and use as a mitigator tool a simple fan.

The first strategy proposal was inspired by Fathy’s houses, where natural ventilation acts as a cooler, improving comfort through evapotranspiration. A portable plug-in rooftop solar chimney (CHS), to induce indoor ventilation using cheap or recycled materials locally available, was studied, designed and prototyped [ 3 ].

 Brazilian Phase Atlas

Figure 4: Brazilian Phase Atlas

The project was presented to the community. I had the idea of creating a workshop with the interested residents, free of charge, to build up to five CHS and install them. Only Lucineide’s family agreed, being unable to help in the construction of the artefact. During an architectural summer workshop, the CHS2 was built and installed. After four months of use, the qualitative results collected were positives, reporting improvements in indoor comfort. Nonetheless, six days measurement campaign showed non-conclusive results, showing a slight indoor temperature increase. Data analysis observation appointed to the introduction of complementary strategies to achieve the goal 18.

Note: the day before the installation, Lucineide’s husband called me to cancel the experiment. In the meeting that followed, he was ensured that no cost will be supported by his family. In the end, he asked me if I was an inventor and finally agreed.

Portugal Experimental phase

The research continued at the Faculty of Science of the University of Lisbon, where was built a test cell, using bricoleur technics, to install passive strategies and measure it in a controlled environment. The CHS3 was improved and resized. Have been selected and studied for their low cost and simple construction [ 5 ], as complementary strategies: a reflective roof coat strategy (Albedo); Radiant Barrier insulation (RB). All strategies have been implemented and evaluated.

Scheme of all strategies applied - Design Stage. Reference note: CHS19202122; Albedo23;RB24.

Figure 5: Scheme of all strategies applied - Design Stage. Reference note: CHS19202122; Albedo23;RB24.

The data collected for each strategy were non-comparable, suggesting a better performance of the radiant strategy over the others. A simplified dynamic simulation to validate the model was performed with no conclusive results 25.

A second test cell was built beside the first one, to be used as the default model cell and the strategies were reviewed. A design of a REcycled Tetrapak Insulation Board (RETIB) and a simple tube pipe chimney (PC) for natural ventilation was implemented for the experiment [ 5 ]. A validated simulation model allowed to apply the suitable strategies in multiple hot climates.


RB and RETIB, presented a maximum decrease of interior operative temperature of 2.5ºC, at 27.5ºC outdoor temperature, and a 2.3ºC at 26.6ºC, respectively, with an increase of thermal comfort period up to 37.8%.

Heat map of TC indoor simulation in Teresina (BR): a) indoor heat of TC; b) ∆T using RETIB; c) ∆T using RETIB+NC.

Figure 6: Heat map of TC indoor simulation in Teresina (BR): a) indoor heat of TC; b) ∆T using RETIB; c) ∆T using RETIB+NC.

The introduction of natural ventilation, through the PC and the CHS3, proved ineffective during the day. The validated simulation applied, presented a maximum decrease of 3.45 ºC with RB and 3.39 ºC with RETIB [ 6 ]. Have been noted overheating during the night and corrected with a single-window opening, (0,40m2), only for night cooling [ 6 ].

Table 1: Validated simulation results using RB and RETIB.

Table 1: Validated simulation results using RB and RETIB.


The results show that Radiant Barrier (RB) and REcycled Tetrapak Insulation Board (RETIB), at the cost of 5€/m2 and 1-2€/m2 respectively, act as a mitigator and must be accompanied by a night cooling strategy. Both materials can be used as roof insulation, allowing to re-shape internal spaces, both physical and thermal, at a very low-cost price. The solar chimney proved to be non-effective in hot climates, inducing the hottest outdoor temperature inside the house. Nonetheless, this artefact should be studied in different environments and climates, using controlled ventilation, with cooler airflow.

From my experience, such improvements should be introduced in the community through resident associations, by recognized organizations (Universities, ONGs, others) to access this context and aid in their organizational process. Involving actively the population concerned 26, giving them the tools to upgrade their environments, maybe a strategy, while global approaches to eradicate poverty are implemented.

On the methodological plan, Research by Design proved to be effective, allowing to swing between qualitative and quantitative methods, when needed, in the search for both questions and answers. In this research, the achievement of equilibrium between both approaches turned to be the key to the investigation and may show the importance of the RbD methodology used, to better comprehend and participate in our complex human world.

Note: All Figures rights belong to the Author, except graphics on Fig. 4, referenced on 23 and 24.

  1. Atamanov (2020): March 2020 PovcalNet Update : What's New. Global Poverty Monitoring. Technical Note; no. 11 Washington, D.C. : World Bank Group.
  2. World Bank (2018). Poverty and Shared Prosperity 2018: Piecing Together the Poverty Puzzle. The World Bank. doi:10.1596/978-1-4648-1330-6.
  3. Montaner, Josep Maria (2014): Arquitetura e crítica na America Latina. Ed. RG , São Paulo.
  4. Fiori Arantes, Pedro (2008): O lugar da arquitectura num planeta de favelas. Opúsculo, Ed. Dafne. Porto.
  5. World Bank (2020): Poverty and Shared Prosperity 2020: Reversals of Fortune. Washington, DC: World Bank. doi: 10.1596/978-1-4648-1602-4.
  6. Geetha, Velraj (2011): »Passive cooling methods for energy efficient buildings with and without thermal energy storage: A review«, in: Energy Education Science and Technology Part A: Energy Science and Research Volume (issues) 29(2): p. 913-946.
  7. Sakka, Santamouris, Livada, Nicol, Wilson (2012): »On the thermal performance of low income housing during heat waves«, in: Energy and Buildings,Vol. 49,
  8. Hauberg (2011): »Research by Design: A research strategy«, in: Revista Lusófona Arquitetura, Educ./Archit. Educ. J. p. 46–56.
  9. Toeters, Ten Bhömer, Bottenberg, Tomico, Brinks (2013): »Research through Design: A way to drive innovative solutions in the field of smart textiles«, in: Adv. Sci. Technol, 80, 112–117.
  10. Roggema (2016): »Research by Design: Proposition for a Methodological Approach«, in:Urban Science. 1. 2.
  11. DOE, US, “EnergyPlus Engineering Reference.” The Reference to EnergyPlus Calculations, DOE, US, United States, 2010.
  12. Xavier (2016): »Fortaleza da desigualdade e violência: geopolítica do medo e anomia social como fator de produção da violência concentrada e da sensação de insegurança«, in: Conpendi Law Review, Onãti, Spain. v. 2, n. 1, p. 122-130. DOI: 10.21902/clr.v2i1.268 (in portuguese).
  13. Berenstein Jacques, Paola (2003): Estética da ginga: a arquitetura das favelas através da obra de Hélio Oiticica. Rio de Janeiro: Casa da Palavra. (in portuguese).
  14. Friedman, Yona (2009): Arquitetura da sobrevivência, uma filosofia da pobreza. Torino: ed. Bollati Boringhieri (in italian).
  15. Levi-Strauss, Claude (1966): The Savage Mind. Chicago: Ed. The University of Chicago Press, p. 16.
  16. Nahar, Sharma, Purohit (2003): »Performance of different passive techniques for cooling of buildings in arid regions«, in:Build Environment, v. 38, p. 109-116.
  17. Filho, J.F.,(2017): »O bairro Dom Expedito em Sobral-CE e suas transformações socioespaciais«, in:Anais do III Seminário Regional Comércio, Consumo e Cultura nas Cidades, Sobral. 19-22 de Junho. (in portuguese).
  18. Bonaccorso N., (2017): »CHS.2. Plug-in solar chimney: an installed experiment in Brazil«, in: SB-LAB 2017 Proceedings of the International Conference on Advances on Sustainable Cities and Buildings Development. Porto. pp. 29-40.
  19. Bansal, Mathur, Bhandari (1993): »Solar Chimney for Enhanced Stack Ventilation«, in: Building and Environment, v. 28, n. 3, p. 373- 377
  20. Neves, Roriz, Marques/Silva (2011): »Modeling a solar chimney for maximum solar irradiation and maximum airflow, for low latitude locations«, in: Proceedings of Building Simulation 2011: 12th Conference of International Building Performance Simulation Association, Sydney, 14-16 November.
  21. Hussain, Yit/Man (2015): »Experimental investigation of rooftop solar chimney for natur ventilation«, in: ARPN Journal of Engineering and Applied Sciences. v.10, n.21.
  22. Shi L. et al. (2016): » Developing an empirical model for roof solar chimney based on experimental data from various test rigs«, in: Building and Environment, v.110, p.115-128.
  23. Haschemi (2017): »Effects of thermal insulation on thermal comfort in low-income tropical housing«, in: Energy Procedia, Volume 134, Pages 815-824,
  24. Medina, Young (2005): »A perspective on the effect of climate and local environmental variables on the performance of attic radiant barriers in the United States«, in: Building and Environment. v.41, p.1767-1778
  25. Bonaccorso, Martins, Graça (2019): »Validated Simulation of Low‑Cost Thermal Envelope Upgrades for Slum Housing«, in: Proceedings of Building Simulation 2019: 16th Conference of IBPSA. Rome. Sept. 2-4.
  26. Santamouris, Pavlou, Synnefa, Niachou, Kolokotsa (2007): » Recent progress on passive cooling techniques: Advanced technological developments to improve survivability levels in low-income households«, in:Energy and Buildings, Volume 39, Issue 7, Pages 859-866.