BA-1302: Retrofit of a Multi-family Mass Masonry Building in New England

Executive Summary

Merrimack Valley Habitat for Humanity (MVHfH) has partnered with Building Science Corporation to provide high performance affordable housing for 10 families in the retrofit of an existing brick building (a former convent) into condominiums.

The condominium conversion project will contribute to several areas of space conditioning, water heating, and enclosures research.  Enclosure items include insulation of mass masonry building on the interior, airtightness of these types of retrofits, multi-unit building compartmentalization, window selection and roof insulation strategies. Mechanical system items include combined hydronic and space heating systems with hydronic distribution in small (low load) units, and ventilation system retrofits for multifamily buildings.

One of the initial questions in this renovation project was whether the existing mass masonry walls could be retrofitted with insulation without durability problems. To understand these risks, the team performed a site assessment followed by material property testing and hygrothermal simulations. This section covering the analysis includes a summary of the material property testing results, a description of the hygrothermal computer simulations, and an interpretation of the results. Key findings included the following:

• In both wall assemblies simulated, at normal rain exposures, brick freeze-thaw risk is predicted to be minimal with or without the thermal insulation retrofit;
• The existing brickwork damage is likely due to the combination of rain water concentrations due to poor maintenance and details (e.g., gutter/downspout failures); and
• Addressing the rain water management issues associated with the existing areas of damage must be addressed as the thermal insulation retrofit will exacerbate such problems.

Applying the results of the masonry retrofit analysis to a wider array of projects is beyond the scope of this report; an overview of the topic is covered by Straube et al. (2012). An analysis of the applicability of material property testing to similar projects is covered in an upcoming BSC report, which examines test results with other variables such as project geographic location, vintage, and other readily available properties (density, water uptake).

BEopt parametric analysis was performed for insulation options for the wall retrofit in order to determine the energy performance for each wall system and the most cost effective measure.  The project team decided to explore the following options for the wall retrofit: three layers of 2” XPS rigid insulation, two layers of 2” XPS rigid insulation, 5” of ccSPF and 2” of ccSPF with batt insulation. Advantages and disadvantages of each wall system were explored.  

The proposed 6” XPS rigid insulation on the interior of the masonry was selected as one of the best performing options. The energy analysis determined this option is not the most cost effective measure; however, it was selected due to the financial model of Habitat for Humanity, which includes volunteer labor and donated materials.

The cost of a project typically increases with the improved performance, but local incentives as well as state and federal tax credits can offset some of the cost. By performing an energy analysis of the project, homeowners and builders were able to make the early decisions on which measures are the most suitable and feasible. The results of this research provide information on what level of efficiency can be achieved within budget constraints. REM/Rate energy analysis software was used to determine the predicted energy savings for each housing unit by implementing the advanced retrofit packages. This work was driven by Massachusetts incentives of over $1000 per unit (ENERGY STAR Qualified Homes, Version 3 Tier II).

Effective air barriers are an important component for good energy performance, good indoor air quality, and control of interstitial condensation. In addition, an effective air barrier between units of multifamily housing reduces transmission of sound, odors, and smoke, lowers fire spread risk, and helps control stack-driven airflows. Therefore, BSC established targets for air barrier and compartmentalization performance on this project, and developed multiple details for airtightness (interior-to-exterior) and compartmentalization (unit-to-unit) strategies. 

The planned research intended to analyze the effectiveness of air barrier strategies implemented in this multifamily retrofit; however, no units have been completed to date, so effectiveness has not yet been measured. However, some information is available on the secondary question, of the difficulties in implementing these air barrier details.

Use of individual unit space heating and domestic hot water systems greatly limits the distribution losses associated with these systems, at a comparable or lower net cost to centralized systems. The use of individual unit HRV (heat recovery ventilation) systems greatly improves compartmentalization, and based on preliminary analysis, provides substantial first cost savings relative to a large centralized HRV system. Use of individual tankless water heaters and HRV units in each condominium will also provide the homeowners with the ability to control the settings according to their particular lifestyle and desired comfort levels. Therefore, BSC developed a mechanical design with individual mechanical systems for each apartment unit, and for the common areas of the MVHfH multi-family building. However, no mechanical systems have been installed to date, nor have any quotes been received for the design. Therefore, the cost impacts cannot be determined at this point, but details on the current HVAC design are available.

BEopt software was used to evaluate the cost effectiveness of the retrofit measures proposed for this project as well as the predicted site energy use. One representative unit was chosen for energy modeling (Unit 8, on the second floor). Several options for wall insulation, window types, air leakage, and mechanical systems were modeled in order to determine the combinations of measures that are the most cost effective. The difference in source energy use between the “Existing” and “Minimum Cost Case” projected by BEopt was 95.0 MBtu/year, or a 52.6% reduction. However, the “Design Case” which generates a slightly higher reduction (96.2 MBtu/year or 53.3%) but at a higher cost (per BEopt) was chosen by the design team.  This case includes measures that are the best approaches for this particular project and may not be the most cost-effective in all cases because of the financial model of Habitat for Humanity.

The retrofit work on the apartment units is ongoing and the project is slated for a completion date in December of 2014. Therefore, the utility bills were not available for the comparison of the predicted and actual energy use.  No additional work is planned for this project in the subsequent years due to the termination of funding at the end of 2012.

1 Introduction

Merrimack Valley Habitat for Humanity (MVHfH) has partnered with Building Science Corporation to provide high performance affordable housing for 10 families in the retrofit of an existing brick building (a former convent) into condominiums.

The research performed for this project provides information regarding advanced retrofit packages for multi-family masonry buildings in Cold climates. In particular, this project demonstrates safe, durable, and cost-effective solutions that will potentially benefit millions of multi-family brick buildings throughout the East Coast and Midwest (Cold climates). The retrofit packages provide insight on the opportunities for and constraints on retrofitting multi- family buildings with ambitious energy performance goals but a limited budget.

The condominium conversion project will contribute to several areas of research on enclosures, space conditioning, and water heating.  Enclosure items include insulation of mass masonry building on the interior, airtightness of these types of retrofits, multi-unit building compartmentalization, window selection, and roof insulation strategies. Mechanical system items include combined hydronic and space heating systems with hydronic distribution in small (low load) units, and ventilation system retrofits for multifamily buildings.

This report is divided into the following sections:

• Evaluating the risks of freeze-thaw to the masonry walls and identifying the suitability of the building for retrofit of insulation
• Exploring several wall retrofit strategies, to determine the most cost effective solution that will achieve specific thermal performance
• Energy analysis to determine the incentive levels the project is able to qualify for
• Development of specific guidance on establishing an air barrier for each of the housing units in the building, as well as identifying difficulties in developing robust air sealing and compartmentalization details for the proposed wall retrofit design and their implementation
• Comparison of the cost and performance of the proposed measures to be implemented on the project.

The following research questions were covered in this project.

• Does the addition of high levels of interior insulation present a risk of freeze-thaw damage to the mass masonry walls in this building?
• What are the predicted energy performance and cost impacts of the proposed wall retrofit design (6" of XPS) vs. the other considered strategies (spray foam, flash and batt)?
• How do rebates and incentives impact the decision making process of the builder, especially in cases of budget constrained construction projects?  In addition, how do these impact the overall energy performance?
• What is the effectiveness of air barrier strategies implemented in this multifamily retrofit project, and what were the difficulties in implementation?
• What are the cost impacts of implementing a compartmentalized ventilation strategy with individual apartment HRV systems vs. a traditional large single central HRV system?
• What are the cost impacts of using individual unit combination space heating/hot water units, including distribution piping design and installation specifics?
• How does the actual energy performance (i.e. utility bills) compare to the BEopt predicted site energy use?

1.1 Background

The former St. Patrick Convent (Figure 1) was purchased by MVHfH from the city of Lawrence in early 2008 for $300,000. The building consists of two parts, the original building (circa 1906 construction) and a rear addition (circa 1930 construction); a small third addition was demolished as part of the renovation (Figure 2). The building was divided into ten 3-bedroom units with designated common meeting and storage spaces.  The sale price for the 3-bedroom units is predicted to be between $125,000 and $130,000 with 35-year mortgages.

Figure 1: Merrimack Valley Habitat for Humanity retrofit building in Lawrence, MA

Figure 2: Merrimack Valley Habitat for Humanity retrofit building aerial views

The total budget for the project has been set for $1,000,000. The project heavily relies on the donated materials as well as volunteer labor, but there are several components of the retrofit that require various industry professionals, such as roofers, electricians, plumbers, HVAC technicians, and brick repair contractors.

The original plan for the project was a “phased” approach, completing contiguous blocks of units in groups. The plan was to have three phases: three units and the common space in Phase I, four units in Phase II, and the last three units (located in the addition) in Phase III.

However, the project team is now seeking a loan from the Federal Home Loan Bank of Boston for the amount of $200,000, and the terms of the loan dictate the completion date in December 2014. Dividing the project into phases would slow the schedule, because it would require the team leaders to teach volunteers the same skills in all three phases, and coordinate the work with the different trades. The final decision on phasing the project has yet to be made.

Before the building was acquired by MVHfH, it had been left vacant and unheated for two or three years. After purchase, the interior of the building was completely gutted, including all HVAC, plumbing, electrical, partition walls, and interior finishes. There were several issues that needed to be addressed with this retrofit project before proceeding with the installation of the high performance and energy efficient measures. Repair and remediation of structural members took priority and have been completed.  Structural issues associated with the settling of brick walls were evaluated by a Boston-based structural firm, and are being addressed as well. The roof leak in the rear addition also has been addressed before proceeding with any interior work in that area of the building.

The original energy efficient retrofit goals for the project were to provide an R-60 roof, R-40 above grade walls, R-20 basement walls, R-10 basement slab and R-5 windows, as promulgated by National Grid (2009), and similar to targets proposed by Straube (2011). The building section shown in Figure 3 illustrates the retrofit approaches to be implemented on this project.

Figure 3: Merrimack Valley Habitat for Humanity building section

2 Brick Material Property Testing and Hygrothermal Simulations

One of the initial questions in this renovation project was whether the existing mass masonry walls could be retrofitted with insulation without durability problems (see Straube et al. 2012). To understand these risks, the team performed a site assessment (documented in Appendix A), followed by material property testing and hygrothermal simulations. The latter two tasks involved laboratory testing of sample bricks, and hygrothermal computer simulations to diagnose the cause of current issues and predict the effect of potential interior insulation retrofit. This section includes a summary of the material property testing results, a description of the hygrothermal computer simulations, and an interpretation of the results.

The brick samples were collected from the building by BSC staff during a site visit on August 3, 2011. Two brick samples each were collected from the interior and exterior of the original and addition sections of the building (for a total of eight). These bricks were taken directly from the wall assembly. Additional samples were also taken which were sitting on the roof and part of a small demolition of an added structure. These additional samples were not used since their origin could not be determined.

2.1 Material Property Testing

The brick samples were subjected to a series of material property tests to facilitate the hygrothermal simulations and inform the assessment of freeze-thaw risk. The material properties, test method and results are summarized in the sections that follow. Further information on the testing procedures can be found in Straube et al. (2012).

2.1.1 Dry Density

Dry density is a fundamental material property that is used as a basic input for all hygrothermal simulation programs. Dry density is used to predict how much heat and moisture are stored in a material over a given time period.

To determine dry density, the brick samples are dried in a gravity oven and periodically weighed using a precision scale. Drying continues until there is no longer any change in mass. This process can take many days to complete.  The volume of each brick sample is then determined using a liquid displacement method. The dry density is simply the quotient of the dry mass and the volume. The test results are given in : Brick material property testing.

2.1.2 Water Absorption or Uptake Coefficient (A-value)

The water absorption coefficient or uptake (Aw or A-value) characterizes the capillary uptake of the material. The value is used in hygrothermal simulation programs to predict the movement of liquid water under capillary suction and redistribution.

The liquid water uptake test follows the method set out in DIN 52617. Carefully cut and oven- dried brick ‘chunks’ (approximately half a brick in size) are placed so that their exposed face (i.e. the outside face of the brick) is just in contact with a pool of water. The samples are periodically removed from the water, weighed using a precision scale, and placed back in contact with the water surface. The measured mass is plotted against the square root of the time of the measurements and normalized for cross-sectional area. The A-value is determined from the slope of this graph and has the rather unusual units of lb/in²s^1/2. The test results are given in : Brick material property testing.

2.1.3 Free Water Saturation

The free water saturation is practical maximum moisture content for masonry units in the field. Higher moisture contents can only be reached in the presence of dissolved salts or conditions that would cause condensation to occur within the empty pores. The value is part of the information used in hygrothermal simulation to estimate the relationship between moisture storage and relative humidity for the masonry units.

The free water saturation values were approximated from cold soak moisture contents. This is the moisture content of the brick after being left in a lab temperature water bath for at 24 hours. The test results are given in : Brick material property testing.

2.1.4 Vacuum Saturation

The vacuum saturation moisture content is used to estimate the maximum amount of moisture that can be held in the brick when all of its open pores are filled with water (W_max). This characteristic value is used to determine the degree of saturation when assessing the resistance to freeze-thaw action.

Carefully cut brick ‘slices’ (approximately 10 mm or 3/8 in. in thickness) are oven-dried, then placed in a desiccator, and a vacuum pump is used to remove any remaining water vapor molecules and 99.9% of the air. The vacuum pump is shut off and water is supplied to the desiccator. Nearly 100% of the open pores in the material are filled with water in this process. The brick slices are said to be ‘vacuum saturated’. The vacuum saturation moisture content is the difference of the mass of the water when saturated and the mass of the dry sample. The test results are given in : Brick material property testing.

2.1.5 Critical Degree of Saturation

The critical degree of saturation is the moisture content at which the material becomes susceptible to freeze-thaw damage.  The degree of saturation is the fraction of saturation relative to complete vacuum saturation. For example, at 0.5 degrees of saturation, the brick contains 50% of the moisture that it would at vacuum saturation.Fagerlund (1977) showed that below some critical degree of saturation (S_crit), no freeze-thaw damage is possible, regardless of the number of temperature excursions below freezing. Similarly, very few freezing cycles are needed to cause damage if the moisture content is above S_crit.

Brick slices are carefully prepared to present clean measurement surfaces. The pre-test length of each slice is measured using a precision micrometer.  The slices are then brought to various degrees of saturation (e.g. 0.7, 0.8, 0.9, etc.) and sealed in packaging so that moisture does not enter or escape. The sealed slices are allowed to ‘rest’ for several hours so that internal moisture can distribute uniformly throughout each slice. The slices are then immersed in a controlled temperature bath and subjected to at least 6 freeze-thaw cycles. The slices are brought back to room temperature, and removed from the bath.  The post-test length is measured, and the change in dimension or dilation is used to identify freeze-thaw damage. Numerous tests are performed on slices from each brick sample to facilitate an estimate of S_crit, the critical degree of saturation. The test results are given in Table 1.

Table 1. Brick material property testing results

These results can be put in context with qualitative descriptions by Lstiburek (2010), who considered “poor” S_crit values to be in the 0.4 or lower range, and “good” S_crit values to be in the 0.8 or higher range. The results appear to indicate good freeze-thaw resistance for the tested brick.

Several of the S_crit values are listed as “NA.” In those cases, the sample dilation at high moisture content was below the repeatability range of the length measurement. In other words, the expansion of the sample after freeze-thaw cycling was too small to be measured effectively. Although these results are inconclusive, an interpretation could be that it suggests the brick is more resistant to freeze-thaw cycling damage. . .

Download the complete report here.