This case study uses photographs and data from an actual building envelope investigation to promote techniques and protocols for using humidity/temperature dataloggers to acquire new, vital information for the forensic team. (While the word ‘forensic’ may have different meanings across North America, we define it as: the puzzle-solving application of a broad spectrum of technical knowledge and expertise to answer questions of interest).
Introduction
Consider an apartment complex, constructed circa-2002 in a city near San Francisco, containing about thirty 2‑story wood-framed buildings clad with traditional three-coat exterior plaster cement (stucco) – Photo 1. Typical for many large-scale residential projects here in California, the pump-applied stucco cladding system was applied to oriented strandboard (OSB) sheathing only at structural shear walls (Photo 2) and simply over steel line wire at the non-sheathed walls (Photo 3).
As documented in Photos 2 thru 7, moisture damage, wood decay and organic growth were found behind the cladding and sheathing at certain ground-floor walls throughout the complex, particularly at bedrooms. (The general absence of similar damage at the same second-floor bedroom walls of course was of great interest to the investigators.) The results of our extensive investigation, evaluation and testing processes did not credibly explain the localized extent and severity of the observed condensation damage at these localized walls.
Photo 1 – Typical Construction at Stucco-Clad Apartment Complex
Photo 2 – OSB Sheathing Supports the Stucco Cladding Assembly
Photo 3 – Line Wire Provides Support for the Stucco Cladding Assembly
Photo 4 – Moisture Damage, Wood Decay and Organic Growth at Back Side of the OSB Sheathing
Photo 5 – Moisture Damage, Wood Decay and Organic Growth at Back Side of the OSB Sheathing
Photo 6 – Moisture Damage, Wood Decay and Organic Growth at Back Side of the OSB Sheathing
Supplemental Investigation with Temperature/Humidity Dataloggers
We therefore supplemented our investigation by positioning within 23 of the occupied ground-floor apartments (distributed throughout the complex) about 100 dataloggers that measure temperature and relative humidity. In general, no prior destructive testing had occurred at any of the selected apartments.
Typically, we positioned four loggers per apartment to record ambient conditions at: a) hallway ceilings, immediately adjacent to kitchens and bathrooms; b) bedroom closets; c) plenums above the gypsum board ceilings at the bedroom closets (i.e., below the subfloor of the second-floor units); and d) exterior wall cavities directly adjacent to these bedroom closets. These walls faced all quadrants of the compass.
As demonstrated in Photo 7, the temperature and relative humidity (RH) sensors for the loggers are located at the ends of 6-foot cables. This feature allowed us to record conditions within the plenums and exterior walls simply by inserting the sensors through small holes drilled through the gypsum wallboard.
The dataloggers were programmed to take simultaneous measurements every 10 minutes (i.e., 144 readings per sensor per day). The accumulated data readily could be downloaded at any time with the optic scanners provided with these loggers, which will continue recording information until they are turned off (or when their batteries die after a year or two).
Photo 7 – Sensors at the Ends of 60-Inch Cables Were Inserted into Closet Walls and Ceiling Plenums
The broad goal of deploying these loggers was to search the collected data for patterns that might explain why some of these ground-floor wall cavities were so much wetter than others. Beyond considering the typical construction defects often found at such large-scale residential projects, we believed that possible additional factors for these conditions might include: relative exposure to the elements (rain, wind, sun, shade, and perhaps even night-sky cooling ); the presence or absence of wall penetrations (e.g., piping, conduit and ducting); variable amounts of interior lifestyle moisture (perhaps due to differing occupant loads); varying moisture content of the on-grade concrete slabs; and multiple other localized construction, architectural and mechanical system features.
The purpose of this article is not to summarize our findings for this ongoing litigation. Instead, we simply will demonstrate our firm’s process for evaluating the voluminous data produced by the loggers. In addition, this article documents and describes the daily solar-driven waves of water vapor that can occur within exterior wall cavities.
The Humidity Ratio
Once information is collected from these loggers, the provided software enabled us to convert the combined temperature and RH readings into the humidity ratio, which represents the actual ambient moisture content – measured by the total grains of water vapor per pound of dry air (GPP) – within each of the tested spaces at a particular point in time.
Chart 1 below is a very simplified ‘psychrometric chart’ demonstrating that the humidity ratio is derived from a non-linear relationship between temperature and RH. (Psychrometrics is the science of air/water interaction.) For example, if we know that the temperature is 80° F (at the vertical Red line) and the RH is 20% (at the curved Blue line), then the point where these two lines cross (at the horizontal Green line) informs us that the approximate moisture content of the air is 30 GPP. Using the dataloggers, comparative analyses of such continually changing moisture loads can be highly informative.
For example, Graphs 1 and 2 show 68 days of ambient moisture load data recorded 144 times per day (i.e., every ten minutes) during winter months of 2013/2014 within two of the ground-floor apartments at this complex. We can see that the occupants of Apartment ‘A’ live a relatively dry lifestyle (long-term average = 54 GPP), while the tenants in Apartment ‘B’ generate 33% more ambient moisture (72 GPP) on average.
It has been our experience with mass-produced residential units here in the San Francisco Bay area that occupants who generate a long-term average moisture load greater than 60 GPP are increasingly prone to exhibiting mold and moisture problems at cold gypsum wallboard (i.e., at exterior walls and corners — see Photo 8) during winter months. Further, these risks appear to increase exponentially with tenants who live ‘wet’ (average GPP > 63) or ‘very wet’ (average GPP > 67) lifestyles.
We generally believe it unreasonable to criticize residents who choose to live wet lifestyles.
(It is very important to note that all such wet versus dry assessments are highly relative and qualitative – in virtually all cases, the key factor for potential interior mold proliferation caused by condensation of ambient moisture is how poorly the exterior walls were insulated.)
The broad goal of deploying these loggers was to search the collected data for patterns that might explain why some of these ground-floor wall cavities were so much wetter than others. Beyond considering the typical construction defects often found at such large-scale residential projects, we believed that possible additional factors for these conditions might include: relative exposure to the elements (rain, wind, sun, shade, and perhaps even night-sky cooling); the presence or absence of wall penetrations (e.g., piping, conduit and ducting); variable amounts of interior lifestyle moisture (perhaps due to differing occupant loads); varying moisture content of the on-grade concrete slabs; and multiple other localized construction, architectural and mechanical system features.
The purpose of this article is not to summarize our findings for this ongoing litigation. Instead, we simply will demonstrate our firm’s process for evaluating the voluminous data produced by the loggers. In addition, this article documents and describes the daily solar-driven waves of water vapor that can occur within exterior wall cavities.
The Humidity Ratio
Once information is collected from these loggers, the provided software enabled us to convert the combined temperature and RH readings into the humidity ratio, which represents the actual ambient moisture content – measured by the total grains of water vapor per pound of dry air (GPP) – within each of the tested spaces at a particular point in time.
Chart 1 below is a very simplified ‘psychrometric chart’ demonstrating that the humidity ratio is derived from a non-linear relationship between temperature and RH. (Psychrometrics is the science of air/water interaction.) For example, if we know that the temperature is 80° F (at the vertical Red line) and the RH is 20% (at the curved Blue line), then the point where these two lines cross (at the horizontal Green line) informs us that the approximate moisture content of the air is 30 GPP. Using the dataloggers, comparative analyses of such continually changing moisture loads can be highly informative.
For example, Graphs 1 and 2 show 68 days of ambient moisture load data recorded 144 times per day (i.e., every ten minutes) during winter months of 2013/2014 within two of the ground-floor apartments at this complex. We can see that the occupants of Apartment ‘A’ live a relatively dry lifestyle (long-term average = 54 GPP), while the tenants in Apartment ‘B’ generate 33% more ambient moisture (72 GPP) on average.
It has been our experience with mass-produced residential units here in the San Francisco Bay area that occupants who generate a long-term average moisture load greater than 60 GPP are increasingly prone to exhibiting mold and moisture problems at cold gypsum wallboard (i.e., at exterior walls and corners — see Photo 8) during winter months. Further, these risks appear to increase exponentially with tenants who live ‘wet’ (average GPP > 63) or ‘very wet’ (average GPP > 67) lifestyles.
We generally believe it unreasonable to criticize residents who choose to live wet lifestyles.
(It is very important to note that all such wet versus dry assessments are highly relative and qualitative – in virtually all cases, the key factor for potential interior mold proliferation caused by condensation of ambient moisture is how poorly the exterior walls were insulated.)
Chart 1 – Humidity Ratio Is Derived from a Non-Linear Relationship between RH and Temperature
Graph 1 –Apartment ‘A’ Tenants Have ‘Dry’ Lifestyle (Average GPP = 54) – 144 Readings per Day
Graph 2 – Tenants at Apartment ‘B’ Live a ‘Wet’ Lifestyle (Average GPP = 72) – 144 Readings per Day
Photo 8 – Mold Growth at Poorly-Insulated Exterior Walls (at Corner of Building)
The moisture spikes seen in Graphs 1 and 2 typically represent cooking and bathing activities. Similarly, it has been our experience that extended periods of low moisture load commonly correspond to occupant absences: for example, in Graph 1 we see that the tenants at Apartment ‘A’ were not at home during a multiday period (the Christmas holiday) beginning at reading #7489 (Day 53).
At Graph 3 we see that the moisture loads within the exterior walls at these various apartments also differ greatly. With the wall cavity at Apartment ‘C’ (average GPP = 63) there is 75% more ambient water vapor than with the Apartment ‘D’ wall (average GPP = 36). The forensic team obviously then focused extra attention to comparing the differences between these walls.
Graph 3 – Why Is the Exterior Wall Cavity at Apartment ‘C’ 75% Wetter (on Average) than Apartment ‘D’?
Further, close consideration of Graph 3 leads to two other highly interesting questions:
Solar Heating and Solar-Driven Diffusion at Exterior Walls
At every cladding system that is ‘hygroscopic’ (i.e., has an ability to absorb and desorb water), some of these water molecules will be driven inward into the wall assembly when this cladding material is heated by the sun.
While some degree of solar-driven diffusion will occur at all hygroscopic cladding materials, its effects are most noticeable at reservoir systems (e.g., stucco and concrete) that can safely hold larger amounts of free water.
Similarly, solar heating of exterior walls also will free some of the water molecules adhered to the surface of (or contained within) the hygroscopic wood framing and sheathing materials within the wall cavity: “It is clear that any wet material …that is heated by the sun will generate large inward vapour drives.”
Then, at night when these wall assemblies cool, this excess moisture will be adsorbed and absorbed into the solar-dried cladding, framing and sheathing as the water vapor condenses. Further, during cold weather, even in the mild climate zones near San Francisco, this nighttime adsorption/absorption process also will be affected by an outward vapor drive from the heated interior: “The common assumption is that drying occurs predominately to the outside in cool and cold climates… This assumption becomes less true as the climate becomes warmer and as the enclosure is exposed to more solar heating.”
In short, the daily moisture cycles seen in Graph 3 simply document the effects of solar heating and nighttime cooling on the ambient moisture loads within the exterior wall cavities. Proof of this concept is provided at Graph 4, which charts:
Graph 4 – Exterior Temperature (F°) at the Residential Complex and Ambient Moisture Load (GPP) within the Exterior Wall at Apartment ‘E’ – 144 Readings per Day
The remarkably close correspondence between these two lines (which represent very different values) is strong evidence that the daily spikes of ambient moisture within this exterior wall are a function of solar heating of the exterior wall – which creates an inward vapor drive during the heat of the day that is countered at night by condensation (due to cooling) of excess moisture and by an outward vapor drive from the warmer apartment toward the colder exterior.
The long-term average ambient moisture load (Red dashed line) within the exterior wall cavity at Apartment ‘E’ is 47 GPP. Now consider the solid Blue line at Graph 5, which records the humidity ratio within a bedroom closet at Apartment ‘E’. Due to the tenants’ unusually wet lifestyle, the average ambient moisture load within the bedroom is 72 GPP. (Note: the hallway logger next to the kitchen and bathroom recorded a long-term average of 73 GPP.)
From a forensic perspective, an interesting feature of Graph 5 is that the shape of the daily rise and fall (consistent with the effects of solar heating) of the ambient moisture load within the bedroom at Apartment ‘E’ roughly corresponds to the shape of the solar-driven moisture load within the exterior wall (Graph 4) — seemingly indicating a direct interaction (via vapor diffusion and/or unintended air convection) between the two bodies of water vapor separated by a layer of gypsum drywall.
Graph 5 – Ambient Moisture Load (GPP) within Bedroom (Blue Solid Line) and Exterior Wall (Red Dashed Line) at Apartment ‘E’ – 144 Readings per Day
One reason that the rough correspondence between the two moisture loads (within the bedroom closet and within the exterior wall cavity) strongly interests the investigative team is that it has been our experience at other stucco-clad projects that interior moisture loads have not exhibited the daily rise and fall of solar-driven vapor — perhaps due to the reasonably airtight barrier formed when the interior wallboard is installed in compliance with the California Energy Code.
What is the Effect of Lifestyle Moisture?
While it is reasonable to speculate that a primary source of the water vapor recorded within these various walls could be lifestyle moisture from tenant activities, Graph 6 demonstrates that vacating an apartment may not greatly impact the daily waves of solar-heated moisture within the exterior wall.
Graph 6 – Red Line Is Ambient Moisture Load (GPP) near Kitchen and Blue Line Is Ambient Moisture Load (GPP) Inside Stucco-Clad Exterior Wall Cavity at Apartment ‘F’ — 144 Readings per Day
Graph 6 summarizes 17 days of ambient moisture load data, recorded 144 times per day, at Apartment ‘F’. As seen, the tenants vacated the unit midway through this testing period. There was no rain; every day was sunny. The sensor for Logger 1 (Blue line) was inserted into the exterior wall cavity at the master bedroom closet. The sensor for Logger 2 (Red line) recorded the ambient moisture load in the hallway (near the kitchen and bathroom).
When the apartment was occupied, the average moisture load within the exterior wall was only 9% greater than when vacant, seemingly demonstrating that while lifestyle moisture may have contributed to the daily waves of solar-heated ambient moisture measured within this exterior wall, it was not the driving source.
Further, after the apartment was vacated, the average moisture load within the exterior wall was 16% greater than the average moisture load within the unit — leading us back to a key forensic question: what are the sources of water vapor within the exterior wall cavity?
Static ‘Dew Point Methods’ Are Useless for Most Stucco Condensation Investigations
Despite the impressive advances in hygrothermal analysis (e.g., WUFI) made during the past two decades, many consultants still carry out forensic analyses of condensation damage with the static ‘dew point methods’ detailed many years ago by the American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE).
Given specific interior and exterior temperatures, these ASHRAE procedures calculate a theoretical dew point within the wall assembly by analyzing each component’s vapor permeance and thermal resistance (R value). However, it is clear that even the relatively simple processes of solar heating and solar-driven diffusion into and through stucco-clad walls are more than enough to render these static methods virtually useless.
Consider that leading experts on such dew point analyses write:
It is our firm’s opinion that these static methods are outdated and almost always useless for most forensic analyses of moisture condensation damage found within the existing building envelope.
Principles for the Investigation and Evaluation
For building envelope investigations, determining the humidity ratio (ambient moisture load) is closely comparable to using a traditional meter to measure the moisture content ratio of wood framing and sheathing. In either case, a common goal for field-level professionals simply is to determine which components or areas are substantively wetter (or dryer) than others.
The great value of these data loggers is that we can automate this moisture sampling process. Our deployment of 100 loggers at this project enabled us to collect 28,800 simultaneous readings (temperature and RH) every day over a 5-month period. By the end, we had more than 2,000,000 humidity ration (i.e., ambient moisture load) measurements that we could evaluate and process with Microsoft’s Excel program. As seen in the graphs for this article, highly informative patterns will emerge from wide variations in hourly and daily data.
Our previously published principles for this process of investigation and analysis include:
Summary Discussion
As noted, a key purpose of this article is to introduce and discuss unseen effects of solar heating of hygroscopic cladding materials, including reservoir systems (such as stucco and concrete) that can safely hold large amounts of free water.
Exterior wall assemblies with alternate claddings will perform differently on an hourly and daily basis due to varying material properties, e.g., permeability, absorptivity and drainage/drying potential. From a general perspective, these differences do not mean that any particular system (e.g., stucco) is inherently better, or worse, than any another (e.g., fiber-cement lap siding). The varying degrees of solar-heated water vapor occurring within our traditional well-proven exterior wall systems are natural phenomena that, in and of themselves, are not problematic. Instead, the key to successful long-term performance of these walls is to prevent the accumulation of unintended moisture from atypical sources.
However, if problematic condensation is discovered – perhaps years after original construction – then a full evaluation of these conditions might require consideration of the potential effects of solar-diffused vapor being added to an exterior wall’s existing moisture load. To this end, modern temperature/RH loggers that can track minute changes to the humidity ratios within both the building and its envelope can be an invaluable addition to an investigator’s toolkit.
While it is well beyond the scope of this article to identify specific design or construction deficiencies that may have contributed to the localized condensation damage seen at this project, we can report that our data analyses implicate air flow (convection) of warm humid air from the heated interiors into and through certain segments of the exterior wall system via unintended voids in the interior and exterior envelopes.
