Neil May, GHA and NBT

Report
Internal Wall Insulation on Solid Wall
Buildings
Some challenges
Neil May
INTERNAL WALL INSULATION – WHY?
Performance of breathable materials in UK dwellings
INTERNAL WALL INSULATION – WHY?
Performance of breathable materials in UK dwellings
INTERNAL WALL INSULATION – WHY?
Performance of breathable materials in UK dwellings
Any one for EWI?
Assessing the execution of retrofitted external wall insulation for
pre-1919 dwellings in Swansea (UK); Joanne Hopper et al 2011
Assessing the execution of retrofitted external wall insulation for pre1919 dwellings in Swansea (UK); Joanne Hopper et al 2011
Assessing the execution of retrofitted external wall insulation for
pre-1919 dwellings in Swansea (UK); Joanne Hopper et al 2011
Background
• Government/EU commitment to 80% reduction in GHG by
2050
• All buildings to be near to zero GHG/ Carbon emissions by
2050
• = One building every 50 seconds from now on
• Green Deal/ ECO programme starting this autumn (?) with
particular emphasis on solid wall buildings
• 6 million plus solid wall buildings in UK, most in England,
most are brick.
• Minimum 2 million expected to use Internal Wall Insulation
• Many cavity wall and other buildings to use IWI as well
Research Concerns
• Thermal performance
–
–
–
–
Background issue of U values of traditional walls
Effect of IWI on thermal resistance of masonry
Thermal bridging issues
Overheating issues
• Moisture performance
– Effect of internal moisture
– Effect of driven rain and other liquid moisture sources
• Health
– Effect of above on occupant health
– Interaction with other factors especially ventilation
Thermal issues: Traditional walls
• Do not conform to type of wall suited to BR 443 (using BS
9496) – ie discreet layers of known materials
• Consequently in –situ testing of traditional wall U values
show that most walls perform better than under BR443 (incl
RdSAP (2009) default values. Typically traditional walls have
U values of 0.9 to 1.6W/m2K for walls over 225mm wide. The
thicker the wall the better the U value.
• Performance is much affected by moisture. More moisture
leads to lower thermal resistance.
• U value calculations given for IWI on traditional walls need to
take these issues into account.
Thermal Limits (German house)
Energy loss through external wall in %
External Insulation versus Internal
Thickness of internal insulation in cm
External
insulation
Practical limits: Thermal Bridges
Refurbishment of a traditional stone wall with 60 mm insulation on the
inside
 Reveal not insulated
 Reveal now insulated with 40 mm insulation
Thermal Bridges: Party Wall Issues
Before
13,1 °C
After
13,1 °C
15 °C
Partial fixed internal wall insulation:
 Displacement of isotherms, surface temperature sinks on the noninsulated side of the wall
 Risk of mould / mildew
12,6 °C
Moisture – research background
• Experimental work of Tim Padfield, Brian Ridout and others
based on material qualities and site testing – no or little
modelling used
• German work of IBP based on laboratory testing and
modelling
• Masses of good conservation work and even more bad work
on old buildings (no modelling or material testing, just
observation)
• Everyone agrees that Glaser (ie EN 13788 as per BS5250) is
inappropriate for IWI unless walls are absolutely dry and
protected. EN 15026 is correct standard at present
Modelling Protocols
• BS EN 13788 (BS 5250) versus EN 15026
EN 13788
Steady state
Monthly (averaged)
Limited materials criteria
No driven rain
No orientation
EN 15026
Dynamic
Hourly
Full materials criteria
Driven rain
Orientation
16
water content in kg/m2
Driven rain and internal VCLs: Average water
content of an external (German) wall
Driven rain absorption 100%
Driven rain absorption 50%
Driven rain absorption 0%
Variant 1:
without VCL
Variant 2:
with VCL
Insulation thickness (k-value 0.040) in mm
Source: Dr. A. Worch: Innendämmung: Bauphysikalische Aspekte, Probleme und Grenzen und Lösungswege für die Praxis
(engl: Dr. A. Worch: Internal insulation: structural-physical aspects, problems and limits and solutions for the practice)
Conflicting understanding of risk?
• Driven rain is not so important in Germany as UK
• IBP sees presence of oxygen as critical
• RH limits in IBP
– Max RH with air = 85%
– Max RH without air = 95%
• Part F limits
– 1 day 85%
– 1 week 75%
– 1 month 65%
Some Knowledge Gaps
•
•
•
•
•
•
•
Material data (thermal and moisture) for traditional buildings
Modelling (thermal and moisture) of traditional buildings
Thermal performance of traditional buildings
Moisture performance of all buildings esp traditional
Weather data – particularly wind driven rain
Mould formation processes and limits
Construction fault modelling? New DIN (68800-2) says 250g/m2
into structure; UK?
• Durability of different materials under moisture (ie gypsum plaster)
• Consequential effects on whole building performance and occupant
health
KTP approach
Aim is to find a safe, effective, saleable solution for mainstream
application. So focus on 9” to 13” brick buildings in England.
Three legged strategy:
• Modelling
• Case studies, real life monitoring
• Laboratory testing
Comparative testing of breathable and non-breathable systems
Modelling
• Use of WUFI Pro 5 1D
• Also use of Build Desk
Modelling can tell you a lot, however…..
Problems with Modelling
•
•
•
•
•
•
•
Human error
Manipulation
Data errors/ unknowns (ie OSB µ = 30/175)
Simplification of complex structures
Problems at junctions/ bits you can’t model
Issue of how to model bad application
False certainty
Moisture content - location
35
Swansea, SW
Moisture content [kg/kg]
30
Liverpool, SW
25
20
Manchester, SW
15
London, SW
10
40
60
80
Insulation Thickness [mm]
100
Pavadentro on 9” solid brick, 1%DR
Moisture content - orientation
35
Swansea, SW
Moisture content [kg/kg]
30
25
20
Swansea, N
15
10
40
60
80
Insulation Thickness [mm]
100
Pavadentro on 9” solid brick, 1%DR
Moisture content - orientation
35
Moisture content [kg/kg]
30
25
20
15
London, N
London, SW
10
40
60
80
Insulation Thickness [mm]
100
Pavadentro on 9” solid brick, 1%DR
Moisture content – different membranes
35
33
sd-value [m]
Moisture content [kg/kg]
31
0
29
27
5
25
100
23
21
19
17
15
London-N
London-W
Swansea-N
Swansea-W
100mm Pavaflex on 9”solid brick, 0 DR
Impact of density
35
Moisture Content (M-%)
30
100mm Pavaflex
25
Ρflex = 53 kg/m3
20
100mm Pavadentro
Ρdentro = 175 kg/m3
15
20mm Pavaclay &
80mm Pavaflex
10
5
Ρclay = 380 kg/m3
Ρflex = 53 kg/m3
0
0-10mm
10-20mm
20-30mm
Depth in construction
30-100mm
On 9”solid brick Swansea 1% DR
Case Studies
• Very few available
• 2 year KTP, but problems may take 10 or 20 years to develop
• So many variables between each case study
Neil May, February 2012
28
Case Studies
• Solid brick and Pavadentro
– 1 with external render
– 1 without render
• Solid brick and Celotex, without render, but brick
impregnated
• LEAF funded project
– 2 solid stone terraces with Pavadentro system & one new
breathable system (not started)
• Trinity College Cambridge (not directly linked to KTP)
• ERDF Aim High 10 solid wall brick houses in Birmingham
29
Trinity College
• WUFI modelling with 3 different companies in 4 iterations,
giving very different results
• Material Property Testing (GCU)
• Site survey (blower door, in situ U-value, RH monitoring, core
samples for density and initial MC) 2 times with very
different results
• Extensive monitoring planned after application
Neil May, February 2012
30
Laboratory testing
• Test methodology
• Laboratory test update
• Proposals for future tests
– Investigate the dry-out potential
– Liquid moisture ingress – wind driven rain
Test methodology
• 8 different internal insulation systems
• 4 conventional systems – the most
common IWI systems in the UK market
• 4 breathable systems from NBT –
development of two new systems
MOISTURE TRANSFER
Moisture convection:
leaks
Liquid transport:
Wind driven rain
Vapour diffusion
Construction moisture
summer
winter
Performance of breathable materials in UK dwellings
MOISTURE TRANSFER – TEST 1
Moisture convection:
leaks
Liquid transport:
Wind driven rain
Vapour diffusion
Construction moisture
Performance of breathable materials in UK dwellings
INTERNAL WALL INSULATION – LIMITS
T
[ºC]
EXT
T
[ºC]
INT
EXT
INT
Low temperature at the wall-insulation interface
Risk of interstitial condensation and mould growth
Performance of breathable materials in UK dwellings
INTERNAL WALL INSULATION – LIMITS
Low temperature at the wall-insulation interface
Risk of interstitial condensation and mould growth
Performance of breathable materials in UK dwellings
To what extent breathable materials
can reduce the risk of interstitial
condensation?
Performance of breathable materials in UK dwellings
TEST METHODOLOGY
Performance of breathable materials in UK dwellings
TEST METHODOLOGY
• Monitoring interstitial condensation by measuring
the RH at the wall-insulation interface
• 6 RH capacitance sensors each section
• Additional test: comparison
between monitoring and
hygrothermal modelling
(WUFI Pro)
Performance of breathable materials in UK dwellings
TEST 1
Settings:
• Driving force: vapour pressure
differential
• External conditions:
Manchester TRY file from CIBSE,
diurnal temperature variation
into account
Δ VP
ΔVP
ΔVP
• Internal conditions: WarmFront
data (UCL), 80th percentile
bedroom RH
• Rain is not simulated
Performance of breathable materials in UK dwellings
TEST 1
The wall is exposed to:
• November, December – winter:
vapour adsorption due to diffusion
• May, June – spring: vapour
desorption due to diffusion
Performance of breathable materials in UK dwellings
TEST 1 – COMPARISON OF RELATIVE HUMIDITY
Breathable materials: 22% average RH reduction
Non-breathable materials: 8% average RH reduction
Higher speed of desorption in breathable
materials
Performance of breathable materials in UK dwellings
TEST 1
Higher speed of desorption in breathable
materials
(measured at the wall-insulation interface)
Possible reasons:
• Low vapour permeability (vapour movement on both sides)
• The capillary suction moves the moisture away from the critical
interface
• Breathable materials can store moisture
(hygroscopicity)
ΔVP
ΔVP
Performance of breathable materials in UK dwellings
COMPARISON OF MONITORING AND MODELLING
Settings:
• WUFI Pro 1D
• Climate file from chamber
• Only diffusion (rain is off)
• Initial conditions from chamber
(trends comparison)
Performance of breathable materials in UK dwellings
COMPARISON OF MONITORING AND MODELLING
RH - simulated
RH - monitored
Dry-fit Pavadentro
Wetting well simulated – drying underestimated
Performance of breathable materials in UK dwellings
COMPARISON OF MONITORING AND MODELLING
RH - simulated
RH - monitored
Pavaclay and Pavaflex
Performance of breathable materials in UK dwellings
COMPARISON OF MONITORING AND MODELLING
RH - simulated
RH - monitored
PIR
Performance of breathable materials in UK dwellings
COMPARISON OF MONITORING AND MODELLING
• WUFI calculations agree with the measured data
during vapour adsorption (“winter”)
• The simulation underestimates the dry-out potential
of the materials
• Possible reason:
– Underestimation of liquid transport coefficient in clay
blocks and insulation materials
– Incorrect algorithms in model
Do we know the properties of materials in
traditional buildings?
Performance of breathable materials in UK dwellings
Some Specific Problems in Practice
•
•
•
•
•
•
•
•
•
•
•
•
•
Rising damp.
Different moisture levels at different parts of walls (ie corners).
Joist ends
Window reveals
Partition/ party walls
Uneven walls
Gypsum plaster
Knowing what walls are made of
Quality of workmanship/ bad application
Services
Application in wet areas (bathroom, below DPC,…)
What are extreme conditions/ limits? Human behaviour issues
Long term maintenance of fabric and building services
49
Some interactions to be considered
• Internal Wall Insulation and thermal performance due to changing
moisture levels
• Overheating
• Indoor air quality
• Ventilation requirements and systems
• Heating systems
• Occupant behaviour
50
Key findings so far
• No one really understands moisture movement.
• BR443 and BS 5250 currently inappropriate for modelling solid
walls and possibly any wall with internal insulation
• Correct modelling and testing indicates that
– External wetting is much more important than leakage of
moisture into the structure
– Location and orientation are critical for capillary open walls
– Breathability of IWI systems is vital where walls are wet
– Density of insulation is also vital
– Too much vapour openness is sometimes a problem
– In some situations only minimal or no insulation is possible
51
Way forward for IWI on Solid Walls?
• Must take into account faults and failures short and long term
of both IWI application AND other building maintenance
(incl external fabric, rain water, drains, ventilation)
• Need useful safe and buildable solutions, not over-optimised
solutions to allow for unknowns, faults and human behaviour
• Pointless and dangerous going for U values better than
0.40W/m2k (?)
• Need much more evidence, as well as proper data sets for
materials and weather
• Move towards simplified guidance rules and structure
• No “one size fits all” solution. Accept uncertainty and move
forward with awareness. Its as much about process and
people as technologies.
52
Thank you
www.natural-building.co.uk

similar documents