Give us a call on 08445 611 646
Promote your
business with us
 
Business Site 
OverviewBlogsArticlesSocial MediaProjectBook TVPress
 
Testing for risk of condensation when retrofitting insulation to Victorian solid brick walls – a hygrothermal study
Print         Bookmark 
 
 
By Nicole Solomons

(Based on MSc thesis from CAT Graduate School of the Environment)

Fig. 1 Heat loss in older buildings
Source: German Energy Agency 
Abstract:
Installing additional insulation is one of the most effective ways of reducing heat loss from a building. However it is essential that the materials chosen are appropriate, particularly with historic buildings that are considered ‘breathable’ constructions.

There is a danger that when adding insulation to an existing wall, moisture can be trapped and become harmful to the structure and human health in the long-term.

This paper reports on the modelling of three different types of insulating material using a computer-based hygrothermal simulation programme. The materials are Woodfibre (a bio-based insulation), Calcium Silicate (a new mineral based insulation) and Polyisocyanurate (petrochemical based, with foil facing). Together these three types represent a realistic and contemporary range of alternative approaches.

The effect of adding each type of insulation to the internal surface of a brick wall is assessed through detailed study of changes in relative humidity, moisture content and temperature at the interface of the brick and the newly installed insulation.

Results of the simulations suggest that there is an increase in relative humidity after the installation of all three insulating materials, but the hygroscopic materials perform better in terms of buffering the moisture and negate the need for vapour barriers.

The effect of wind driven rain on the moisture content of the wall was considered, and it was found that reducing this rain effect, perhaps by adding an external rain screen, was successful in lowering the relative humidity build-up resulting from installing the impermeable insulation, but was not necessary in the case of the hygroscopic materials.

1. Introduction
Following the downturn in new house building (the lowest level since 1923) (www.independent.co.uk) focus in the UK has turned to the refurbishment of our current housing stock. Victorian housing (1840 to 1919) makes up a considerable proportion of UK housing, approximately 4.5 million houses (Yates, 2006). Property from this period usually have solid walls as opposed to more modern cavity walls.

One way of saving energy is to insulate these solid walled properties. Figure 1 indicates the proportion of heat that is typically lost through the walls of an older house.

In historic buildings it is often more appropriate to insulate internally due to reasons of appearance (e.g. in Conservation Areas) and difficulties in modifying the other external elements such as window surrounds and eaves details. This study therefore focuses on internal insulation.

Many in the historic building sector believe that the walls of such buildings should be breathable – i.e. readily exchange moisture with outside – and that to use impermeable materials may lead to a build-up in harmful moisture within the structure (Oxley & Gilbert, 1994).

This view has now been reinforced by changes to the building regulations for existing housing - the new Part L1B Conservation of Fuel and Power in Existing Dwellings states that special considerations are applied to “buildings of traditional construction with permeable fabric that both adsorbs and readily allows the evaporation of moisture” (Section 3.8 Approved Doc L1B).

Section 3.12 of Part LB goes on to state that materials used should: “enable the fabric of historic buildings to ‘breathe’ to control moisture and potential long-term decay problems

This paper, through the use of computer modelling, compares the effects of retrofitting two different natural breathable insulations to a solid Victorian Brick wall, with a non-natural impermeable material to look for differences in levels of moisture at the intersection of the wall and insulation.

1.1 Objectives   
The objectives of the study were to: 
a)         Observe patterns of moisture movement through the structure after retrofitting the insulation - both internal moisture vapour moving outwards and also moisture from driving rain entering the structure.
b)         Determine whether the non-breathable material traps moisture in the danger zone between the insulation and external brick.
c)         Ascertain whether condensation forms at any point within the structure as a result of the installation of the insulation.
d)         Assess the performance of the three different materials.

The materials chosen for the study are a generic woodfibre (Wood), which is becoming popular as an internal insulating material; calcium silicate (CAL), recommended for historic buildings which is highly capillary active and vapour permeable; and a generic polyisocyanurate (PIR) impermeable material with foil facing acting as a built-in vapour barrier. A thickness of only 30mm was used as appropriate for rooms where internal mouldings may be in evidence, and also in order to conserve as much internal space as possible.

2. Theory & Literature Review
Theory suggests that an impermeable insulating material will lead to a build-up of moisture in a structure, as it will not have an escape path from the building when trying to move from a situation of high to low pressure.

Work by Padfield (1999) showed the ability of certain hygroscopic materials to absorb and then release moisture again which he termed “moisture buffering”.

Padfield’s work also underlined the need to keep the whole structure permeable for the buffering effect to work i.e. not use gloss paints or impermeable renders on top of the breathable materials.

The fact that hygroscopic materials may negate the need for vapour barriers was studied by Cerny in 2009 and Straube in 2001. Kunzel (1996), the main developer of the WUFI software that was used for this study, looked at whether external rain sources affected the performance of internal insulation on solid walls, due to colder walls and rising moisture content.

He concluded that external coatings should be used to stop excessive moisture travelling into the structure caused by driving rain.
These factors are all considered when interpreting the results of the modelling reported here.

3. Methodology
The tests were carried out using the WUFI Pro 5.1 hygrothermal computer modelling programme. WUFI was developed by the Fraunhofer Institute in Germany and models transient heat and moisture flows in detail, as opposed to previous more steady state calculations of condensation risk (e.g. the Glaser method).

Many validation tests (Kunzel et al, 2004) have been carried out on the software and monitoring and testing of materials to be added to the programme database takes place at the Fraunhofer Institute on a continual basis.

Unlike other models, WUFI considers the dynamic movements of heat and moisture, taking into account external climate, boundary conditions and six key properties of the materials to be tested, Table  1 below details these variables.
Table 1 Data inputs to the WUFI model  Fig. 2   Build up with Calcium Silicate as shown in
WUFI programme 


Figure 2 above shows a cross-section of the Calcium Silicate build-up with the historic (19th C) brick to the left-hand side, then the installed insulation and internal render to the right-hand side.  A monitor has been placed in the render or adhesive between the brick and the new insulation as highlighted. Within the WUFI environment monitors are also automatically placed externally and internally.
Table 2 Wall build-ups (external to internal) for each material tested  Table 3 Properties of each material tested
The details of the material properties for each insulation studied in the trials are summarised in Table 3.  WUFI also enables time periods from a few months up to many years to be modelled.

Wall build-ups were assembled as per manufacturers’ recommendations.  For example, with the PIR an air gap was left between the insulation and the wall and with the calcium silicate the adhesive as recommended by the manufacturer has been used. 

The tests were run for a two year period with readings at hourly intervals and set in a lounge situation with north facing external wall.
The build-ups for each material and corresponding u-value of the overall construction are summarised in Table 2.

3.1 Material properties
In order to interpret the results and better understand the materials being tested it is useful to compare the material characteristics.

WUFI requires as an input certain key variables that together describe how materials may behave when wet:
•           Moisture Storage Function
•           Water Vapour Diffusion resistance
•           Liquid Transport Coefficient Suction
•           Liquid Transport Coefficient Distribution
 
Table 3  Properties of each material tested

The moisture storage function of calcium silicate is much greater than that of woodfibre and greater still than the PIR.  For example at 99% RH the water content of wood fibre is 81 kg/m³ compared to 17 kg/m³ in the PIR, but for the calcium silicate this figure is 546kg/m³.

Water vapour diffusion resistance explains which materials will resist moisture, and from Table 3 it can be seen that the PIR is most resistant to moisture. How the materials react to being wet is dependent on the liquid transport coefficients of suction and diffusion.

 
Fig. 3 Initial results graph showing relative
humidity at monitor position
The CAL has much stronger suction and diffusion coefficients than the Wood, but with the PIR it is non-existent, showing its inability to absorb moisture.

A key indicator to observe from the modelling is the level of relative humidity (RH) at the monitor point.  Any material that stays above 80% RH is considered to be at risk of mould forming. And of course at 100% RH condensation will occur.

4.  Results
Figure 3 represents initial results from the data,  showing relative humidity (RH) at the monitor position for each construction and the un-insulated historic brick. Immediately noticeable is the rise in relative humidity at the monitoring point for each of the three insulating materials.  However, as shown by the solid red line, the installation of the PIR leads to a much greater increase in RH – by approximately 30%.

Other screen shots (Figs. 4 & 5) show more clearly the relationship between temperature and relative humidity between the wall and the insulation, with the left-hand graph showing the situation after installation of the PIR and the right hand graph showing the un-insulated brick.

Fig. 4 & 5 Temperature (red) and Relative Humidity (green) after
installation of PIR (left)  and with un-insulated brick (right). 

The other noticeable difference (Fig 3) is the RH patterns throughout the year: the PIR RH moves in the same direction as the brick in relation to winter and summer months, but in the opposite direction to the other two materials (rising in summer and dropping in winter).

In order to understand why this is, it is necessary to consider how vapour pressure (VP) affects the air movement during different times of the year.
Table 4 Seasonal VP and RH

Air will always move from a position of high vapour to low (McMullan, 2007) usually meaning from inside a building to outside.  Calculations show in this instance that internal VP in the winter is much greater than in summer due to temperature differences inside and out. This is outlined in Table 4 which shows VP and RH internally and externally, from selected dates in the summer and winter periods.

These differences in vapour pressure lead to  a greater movement of the vapour from inside to out in the winter. The hygroscopic materials allow some warm, moist air from inside to move through the structure towards the outside, increasing the temperature and amount of moisture at the monitor position.  They will also allow a certain amount of moisture to dissipate back into the room.

Combined with incoming wind driven rain in the winter these vapour movements lead to an increase in moisture and RH at the monitor. However, for the PIR system the air gap, foil facing and impermeability of the insulating material mean that no warm air is able to pass through to the brick interface and combined with a lower k-value   the brick at the monitor position remains colder and RH at a similar level to the natural materials (figures show the temperature at the monitor with the PIR in winter as similar to outdoor temperature).

A
Fig. 6  Graph showing external temperature and RH throughout monitoring period 
s the outside temperature increases in the summer, the wall warms and there is less air movement from inside to out due to smaller temperature differences.  With the natural materials relative humidity is lowered  at the monitor point.  However with the PIR, solar radiation from outside moves inwards and causes solar condensation with moisture being trapped at the monitor position, increasing the relative humidity at that point.

 Additional monitors placed inside the insulation material support this and show the water content of the natural materials decreasing as temperatures rise, and the moisture evaporates.  However, the PIR, having a low moisture storage function and high water vapour diffusion resistance factor cannot absorb any additional moisture and this leads to an increase in RH at the intersection. 

Figure 6 shows the initial findings in relation to external temperature and RH.

The un-insulated brick is following the same seasonal patterns as the PIR, although at a lower level of RH.  This is due to it being more directly able to absorb internal heating in the winter causing lower RH levels, and also the fact that it is a more porous material meaning moisture can more easily dissipate.

The hygroscopic materials appear to be reacting to internal and external factors but with the PIR only to the external climate. This was tested by cutting out the rain factor and running the model again.

Fig. 7 RH readings with rain factor reduced
Figure 7 shows quite dramatically the RH of the PIR construction decreasing over the time period.

Table 5 shows readings of moisture content at the monitor position in the plaster for each wall before and after rain reduction which again indicates that the PIR system is being mainly affected by external climate. The water content of the plaster in which the monitor is placed is reduced by almost a third when the rain water is cut out.

5. Summary
Condensation has not occurred within any of the tested set-ups.  However, it can be concluded that the PIR system does create high levels of relative humidity at the interface of the brick and the insulation, remaining over the 80% danger level for the whole monitoring period.  Although this is unlikely to have a direct effect on the foil facing (as a non-organic material which is repellent to moisture and mould) it is possible that any mould spores could travel to other parts of the structure e.g. wooden joists, and create problems there.

The RH with the woodfibre system is at a lower level but still over 80% for almost 50% of the time period, which could lead to longer term problems of decay.   It should be noted that more sophisticated wood fibre insulating systems are now coming onto the market such as Pavadentro, which ‘includes a capillary active mineral functional layer made of silicates which actively redistributes moisture’ (NBT, 2011).

The calcium silicate readings stay mainly below the 80% RH level for the time period and the interface has consistently low water content.

Table 5  Water content with each system before
and after rain factor reduction 
6. Conclusion
Insulation is a key tool in the fight to save energy from Britain’s buildings.  However, in the rush to insulate consideration must be given to the types of materials used and how suited they are to the original construction.   Inappropriate use of cheap, impermeable materials can cause an imbalance in the movement of air and moisture in a building and could lead to problems of decay and damage which in the long run will cost more to remediate. 

Using organic, hygroscopic materials is one way to keep a building breathing, and these materials have the additional advantages of being non-toxic to human health and are obtained from renewable sources, which add to their carbon-saving credentials.

For prediction of the hygrothermal effects of insulating buildings the WUFI programme has the potential to be a highly effective tool and aid to understanding and analysis, providing the detailed input parameters and operating conditions are known. 

References
Burridge, N. (2011) Press Association, Available at: www.independent.co.uk/news/uk/poliics/uk-housebuilding-at-lowest-level-since-1923-2217888.html (Accessed: May 2011)

Kunzel, H. M. (1996) Effect of Interior and Exterior insulation on the hygrothermal behaviour of exposed walls. Materials and Structures Journal, Vol 31, March 1998 pp 91-103, DOI: 10.1007/BF02486471

Kunzel, H., Holm, A., Zirkelbach, D. and Karagiozis, A. N. (2004) Simulation of indoor temperature and humidity conditions including hygrothermal interactions with the building envelop, Solar Energy 78 (2005) 554–561 Volume 78, Issue 4, April 2005, Pages 554-561

McMullan, R. (2007) Environmental Science in Building. 6th Edition, Palgrave Macmillan, UK

NBT Pavatex Pavadentro Manual, Natural Building Technologies Ltd, Bucks, UK Available at: http://www.natural-building.co.uk/pavadentro_renovation.htm (accessed 29 June 2011)

Padfield, T. (1999) Humidity buffering of interior spaces by porous, absorbent insulation part of hygrothermal properties of alternative insulation materials Department of structural engineering and materials, Series R, No 61, 1999, Technical University of Denmark

Pavlík, Z. and Černý R. (2009) Hygrothermal performance study of an innovative interior thermal insulation system. Applied Thermal Engineering, Volume 29, Issue 10, pp 1941–1946, Elsevier Ltd

Straube, J. F. (2001) The influence of low-permeance vapour barriers on Roof and Wall performance. Proceedings of Thermal Performance of Building Envelopes VII, Clearwater Beach  Florida, December 2-7  2001

Yates, T. (2006) Sustainable Refurbishment of Victorian Housing. London: BRE Press
....................
 
Nicole Solomons  MSc BA (Hons)
 
Nicole has an MSc in Advanced Environmental & Energy Studies from the University of East London and is working as a freelance energy consultant, focusing on energy efficiency and renewable energy technologies.

Previous positions include Consultancy Coordinator at the Centre for Alternative Technology in Mid Wales, and European Project Manager at Marches Energy Agency, where she was Coordinator for a European-funded research project looking at the sustainability of historic buildings. She is a qualified Domestic Energy Assessor and Green Deal Advisor, and proficient at building thermography.
 
Website:  www.sustainablebuildings.info       
 
 
Post A Comment
You must be a registered user to comment on this page and be logged in.

If you already a member of ProjectBook, please click here 

If you are not yet signed up, please click here 
Submit This Page
 
Delicious Delicious    Digg Digg    Reddit Reddit    Facebook Facebook    StumbleUpon StumbleUpon