This action might not be possible to undo. Are you sure you want to continue?
... embedding simulation within energy sector businesses
Analysis of Thermal Bridges
Dr Aizaz Samuel
Scottish affiliate of IBPSA
Hosted by Energy Systems Research Unit, University of Strathclyde
Scottish Energy Systems Group
• 0930 Thermal bridging • 1000 Theoretical background • 1100 Morning coffee • 1115 Regulatory requirements • 1230 Lunch • 1330 Introduction to THERM 2 • 1400 Workshop exercise 1 • 1500 Afternoon coffee • 1515 Workshop exercise 2 • 1630 Exercise feedback and wrap up • 1700 Close
Scottish Energy Systems Group
A or geometrical interruptions to homogenous insulated 0.75 0.7 construction elements a path is created that allows heat flow 0.65 addition to the one dimensional heat loss through in 0.6 the construction elements 89/25 0.55 89/50 0.5 Disadvantage: 0.45 89/89 Greater heat loss through fabric 119/119 0.4 140/140 0.35 Localised cold spots on fabric elements 0.3 0.25 Lower radiant temperature (lower thermal comfort) 0.2 Condensation risk (mould hazard, maintenance) 0.15 0 5 10 Importance: Proportion ofwall bridged 15 through 20 heat loss bridge Percentage of increases as U values decrease
Scottish Energy Systems Group
Effect of Thermal Bridging on U value for a timber frame wall (timber thickness/insulation thickness) thermal bridge is created when due to structural
• Heat flow need not be perpendicular to lay of bridge.ac.Examples Example 1 Tie in Wall Construction flow through the bridge material is not proportional to the cross section area of the material.uk . • Heat Example 2 Structural beam below window © RenSolutions UK Scottish Energy Systems Group email@example.com.
Energy Savings Trust.References • Applicable standard BS EN ISO 10211 Thermal bridges in building construction. Enhanced Construction Details Building Research Establishment – Approved certifier of design samuel@esru. Heat flows and surface temperatures.strath.uk • • • • • • • Scottish Energy Systems Group . EN ISO 14683 Thermal bridges in building construction. Calculation methods CIBSE Guide A – Environmental Design BRE IP 17/01 Assessing the effects of thermal bridging at junctions and around openings Conventions for calculating linear thermal transmittance and temperature factors.ac. Simplified methods and default values. BS EN ISO 6946 More complicated EN ISO 13370 Thermal performance of buildings. fvalues and psi-values. BRE 497 MCRMA Technical Paper # 18. Heat transfer via the ground. General calculation methods. Combined method. Linear thermal transmittance. Conventions for calculating U-values.
Structural Thermal Bridges Scottish Energy Systems Group firstname.lastname@example.org .ac.strath. Geometrical Thermal Bridges Corner Junction between wall and balcony slab Step Service opening 2.Types of Thermal Bridges 1.
strath. Convective Thermal Bridges Scottish Energy Systems Group samuel@esru.Types of Thermal Bridges 3. Systematic / repeated Thermal Bridges Wall ties Studs 4.ac.uk .
Avoiding Thermal Bridges
Thermal barriers can be used to avoid thermal bridges
Joints where thermal bridging can occur
Scottish Energy Systems Group
• One dimensional heat transfer • Fourier’s equation for 1D and example and limitations • Two dimensional heat transfer • Fourier’s equation for 2D • Application for 2D equation and limitations • Transient conduction equation • Overall transfer of heat • Numerical solution and solvers
Scottish Energy Systems Group
One dimensional heat transfer
Fourier’s Law of heat transfer:
∆T Q = kA ∆x
k = thermal conductivity (W/mK) T = Temperature (K or 0C) x = length (m) Q = rate of heat loss (W) A = area (m2) Q/A = heat flux (W/m2) ∆T/∆x = temperature gradient (K/m) ∆x/k = thermal resistance (m2K/W)
Scottish Energy Systems Group
2W / mK 0 0 T1 = 20 C T2 = 220 C ∆x = 0.5 Q = 480W ∆x Scottish Energy Systems Group 20 Q = 0.2 ×10 0.strath.ac.3W / mK A = 10m 2 T2 T1 A = 10m 2 20 Q = 1.5 Q = 120W email@example.com .One dimensional heat transfer Fourier’s Law of heat transfer – Example: ∆T Q = kA ∆x T1 = 2 C T2 = 22 C ∆x = 0.3 × 10 0.5m k = 0.5m k = 1.
Overall heat transfer coefficient or U Value Tint T1 T2 T3 x1 x2 x3 Q = UA∆T x 1 1 1 = AΣR = +Σ + U hconv _ i + hrad _ i k hconv _ e + hrad _ e Scottish Energy Systems Group firstname.lastname@example.org T4 Text .strath.ac.
uk .Standard Surface resistances Direction of heat flow Upwards Inside surface Rsi Outside surface Rso 0.04 Horizontal 0.04 Q = UA∆T 1 1 x 1 = AΣR = +Σ + U hconv _ i + hrad _ i k hconv _ o + hrad _ o Scottish Energy Systems Group email@example.com 0.ac.1 0.17 0.strath.04 Downwards 0.
firstname.lastname@example.org Inside Scottish Energy Systems Group . • One dimensional analysis tends to overestimates heat loss in the case of a convex corner and underestimates in the case of a concave corner (if external dimensions are used) • Two dimensional analysis has to be used in order to accurately predict heat transfer.strath.Two dimensional heat transfer Outside • Transmission / conduction is one dimensional far from a corner but progressively shows two dimensional characteristics close to the corner.ac.
ac.strath.When is three dimensional analysis required? • • • 3D Ground heat transfer Significant Point thermal bridges Detailed analysis of individual (building) components Academic level results • Scottish Energy Systems Group email@example.com .
Multi dimensional heat transfer Q = Q ( x.ac.uk . y .strath. t ) ∂T Q = kA ∂x ∂ 2T =0 2 ∂x ∂ 2T ∂ 2T + 2 =0 2 ∂x ∂y ∂ 2T ∂ 2T ∂ 2T + 2 + 2 =0 2 ∂x ∂y ∂z ∂T k = ∂t ρC p & ∂ 2T ∂ 2T ∂ 2T Q 2 + 2 + 2 + ∂x ∂y ∂z ρC p 2 dimensional case Scottish Energy Systems Group samuel@esru. z .
strath.uk Scottish Energy Systems Group .ac.Linear thermal transmittance (ψ or psi value) Transmission / conduction loss governed by psi value Transmission/conduction loss governed by U value Edge Transmission / conduction loss governed by U value P D = ΣUA∆T 1 P2 D − P D 1 ψ= l∆T samuel@esru.
Linear thermal transmittance example Te lB B UB lA A UA Ti Q2 D − U A × AA × ∆T − U B × AB × ∆T ψ= l × ∆T Scottish Energy Systems Group firstname.lastname@example.org .ac.
strath.uk .15 in absence of other information 2.08 if all detailing conforms with Accredited Construction Details (Not available in SAP2009) 3. Use y=0. Use y=0.SAP calculations H TB = ΣLψ H TB = yΣAexposed 1. If psi values are known for each junction use these. Psi values can be taken from table K1 or calculated from BR 497 Scottish Energy Systems Group email@example.com. If y has been calculated from individual psi values use this value 4.
SAP calculations Scottish Energy Systems Group firstname.lastname@example.org .strath.
SAP calculations example H TB = ΣLψ H TB = yΣAexposed 4m 3m 1.08 ×12 = 0. If y has been calculated from individual psi values use this value Scottish Energy Systems Group email@example.com 2.08 if all detailing conforms with Accredited Construction Details (Not available in SAP2009) HTB = 0. Use y=0.ac.15 in absence of other information: H TB = 0.strath.96 3. Use y=0.uk .15 ×12 = 1.
16) samuel@esru.SAP calculations example 4.16 × 4 + 0.14) H TB = 0. If psi values are known for each junction use these.ac.09) 3m (corner psi=0.14 × 4 + 0.strath.74 Scottish Energy Systems Group 3m (corner psi=0. H TB = ΣLψ 4m (floor between dwellings psi=0.09 × 3 = 1.09 × 3 + 0.uk .09) 4m (ground floor psi=0.
ac.uk .strath.SBEM calculations example Scottish Energy Systems Group samuel@esru.
6 258.6 Load kWh/m2 259.4 60 70 80 90 100 45 40 0 10 20 30 40 259.4 0 10 20 30 40 50 60 70 80 90 100 % of default psi values Scottish Energy Systems Group firstname.lastname@example.org 259.8 cooling 258.strath.250 % of default psi values 259 heating 258.SBEM calculations example Loads vs Thermal Bridges 90 85 80 75 Load kWh/m2 70 65 60 55 50 Total load vs thermal bridges 260 259.uk .
lbl.gov/software/therm/therm.html Version 6.uk .3 (September 2010) Scottish Energy Systems Group email@example.comD heat transfer software THERM Free software from LBNL USA http://windows.
2D heat transfer software THERM Temperature profile Scottish Energy Systems Group firstname.lastname@example.org .ac.
ac.strath.THERM 2 Solution Procedure A PC program for analysing 2D heat transfer through building products Input data Geometry Material properties Boundary conditions Heat transfer analysis Error estimation Not OK Mesh refinement OK Automatic mesh generation Scottish Energy Systems Group Converged solution email@example.com .
strath.Calculating the psi value Thermal bridge of interest Te lB B UB Flanking element Use the modelling U value U` instead of U value. The modelling U value U` includes any effects of repeated thermal bridges in the construction of flanking elements. lA A UA Ti Flanking element Scottish Energy Systems Group firstname.lastname@example.org .ac.
ac.Worked Example Eaves detail (BRE 497 Exercise) Scottish Energy Systems Group email@example.com .
0.837mm Mineral wool 250mm.13W/mK brick 103mm.uk .21W/mK Plywood 10mm.77W/mK Air cavity 55mm.5mm.037W/mK 37mm 311mm Wood. 0. 0. 0.045W/mK plasterboard 12.13W/mK Mineral wool 115mm.306W/mK 800mm 2D model section with materials and dimensions Scottish Energy Systems Group samuel@esru. 0.037W/mK 100mm Fire stop 55mm. 0.strath. 0.ac. 0.
uk . • Rse of 0.ac. TL is taken as 10C for such cases. ′ ′ ψ = L − lW × U W − lC × U C Scottish Energy Systems Group samuel@esru.BRE 497 Roof junction boundary condition conventions (Not CEPH conventions) U`R Te lc lw U`W Ti TL U`C Roof eaves (insulated at ceiling) • It is usual to take Ti=200C and Te=00C • A heat balance between U`R and U`C can give TL • For ventilated ceilings TL~Te but ventilation component of U`R is not generally known.10m2K/W is taken for the upper surface of the loft space.strath.
1m2K/W adiabatic 0degC @ 0.ac.13m2K/W 1degC @ 0.1m2K/W 20degC @ 0.0degC @ 0.uk .strath.04m2K/W 20degC @ 0.13m2K/W Boundary conditions adiabatic Scottish Energy Systems Group samuel@esru.
g.THM Some information held in the model (e. material detail) may not be present in THERM libraries (repository of information held at system level and not model level) For such cases this message is displayed with the option to update libraries from the model Choose No and Use properties in THERM file Scottish Energy Systems Group firstname.lastname@example.org .Working with THERM Open BRE Validation Example 2.ac.
strath.Click on properties 4.uk .Close THERM 2.Overcoming reduced mouse functionality By default some versions of MS Windows operating systems do not support some mouse actions within the THERM software so it is recommended to change the settings.In the compatibility tab check disable visual themes and then click OK Scottish Energy Systems Group samuel@esru. This is done as follows: 1.ac.Right click on the THERM icon (on desktop or from Start > programs > LBNL software > THERM) 3.
Scottish Energy Systems Group email@example.com. It is productive to do it right the first time.strath.Important notes A THERM model contains the following three items (preferably in the same order) Geometry (dimensions and shape) Material specification (thermal conductivity and emissivity) Boundary condition (surface temperature and heat transfer coefficient) A THERM model does not have overlapping polygons.uk .
strath.ac. ctrl+F7 and right mouse click Task 2: Double click on one of the polygons Task 3: Double click on one of the outside edges of the shape Task 4: Inspect the dimensional information given in the status bar (at the bottom of the window) Task 5: Use the tape measure to measure various lengths Task 6: Use sticky keys to permanently select the tape measure (click sticky keys followed by tape measure) Task 7: Answer questions 1 to 5 on the worksheet Scottish Energy Systems Group samuel@esru.Basic tasks (15~20min) Task 1: Play with F7.uk .
strath.uk . Number of times mesh will be refined when run error is on 10211 compliance values are Quad Tree = 8.ac. If less than value shown in U-factors results reduce this number.Technical Parameters Select Options > Preferences > Therm file options Measure of error in heat flux calculated for each element. Error = 2% and 10 iterations respectively Scottish Energy Systems Group samuel@esru.
ac.Simulation and results analysis Task 8 Task 9 Task 10 Task 11 Task 12 Simulate either by pressing F9 or the simulate icon Toggle results by clicking the show results icon Inspect grid by switching colour off (view > material colours) and (calculation > display options > finite element mesh) View the various types of results available Do Q6 on worksheet (view > temperature at cursor) Scottish Energy Systems Group firstname.lastname@example.org .
Isotherms Scottish Energy Systems Group email@example.com .
Isotherms Scottish Energy Systems Group firstname.lastname@example.org .ac.strath.
ac.Temperature Scottish Energy Systems Group email@example.com .strath.
ac.uk .strath.Temperature Scottish Energy Systems Group samuel@esru.
ac.strath.Calculation of f value f min = f min f min Minimum temperature = 17.90C Tint − surf − min − Text Tint − min − Text 17.uk .6OC [PHI] Scottish Energy Systems Group firstname.lastname@example.org − 0 = 20 − 0 = 0.75 [MCRMA_TP18] Minimum recommended temperature = 12.895 Typical safe value for dwellings is f > 0.
04 0.04 7.51 1.lB Simple example using THERM R_si Surface resistance m2K/W R_so Surface resistance m2K/W H_si Surface HTC (film coefficient) W/m2K B UB lA A UA H_so Surface HTC (film coefficient) W/m2K λ Conductivity T thickness W/mK mm Material A Material B 0.1 0.4 200 200 Scottish Energy Systems Group email@example.com 10 25 25 0.13 0.uk .strath.
uk .strath. Scottish Energy Systems Group samuel@esru. For air cavities select Frame Cavity and use default values. Detailed radiation calculations are out of the scope of the current exercise.ac.Materials library Select Libraries > material library The two parameters of importance are Material name and Conductivity Emissivity is used in radiation calculations and can be safely ignored for most psi value calculations.
Also give it a different colour by pressing the colour button R_si Surface resistance m2K/W Material A Material B 0.13 0.04 H_si Surface HTC (film coefficient) W/m2K 7.1 R_so Surface resistance m2K/W 0.ac.69 10 H_so Surface HTC (film coefficient) W/m2K 25 25 λ Conductivity T thickness Task W/mK 0.4W/mK.51 1.04 0.4 mm 200 200 Scottish Energy Systems Group samuel@esru. From the materials library select New > give material name and conductivity 0.Defining new materials Task Define new material called Material_A with conductivity of 0.strath.uk .51W/mK as follows.51 > Close (there is no need to save) Similarly define Material_B with conductivity of 1.
0 Note that surface orientation is normal to heat flow e.ac.0 0.04 25.04 25.88 0.17 5. horizontal heat flow occurs at walls which are vertical.g.uk .strath. Scottish Energy Systems Group firstname.lastname@example.org 0.10 10. Data inputs include temperature and film coefficient (reciprocal of film resistance) Direction of heat flow Inside Rsi Inside co-eff Outside Rso Outside co-eff m2K/W W/m2K m2K/W W/m2K 0.13 7.Boundary Condition library Select Libraries > boundary condition library For psi value calculations the simplified model is sufficient.0 0.04 25.69 0.
The external part of the floor slab has a convective coefficient of 0 (no air flow under it) but CEPH reduction factor can be input following the ground sheet in the PHPP Scottish Energy Systems Group email@example.com .ac.strath.69W/m2K and temperature of 200C as follows: Libraries > boundary condition library > new > internal_wall > 200C & 7.69W/m2K > close Define the following BC (You may wish to give these different colours) Task 15 Name Temperature Celsius Film coefficient W/m2K 25 10 Exterior Ceiling 0 20 Note that for a floor junction the heat transfer coefficient for downwards heat flow will be used (5.Defining new boundary conditions (BC) Task 14 Define new BC called Internal_wall with film coefficient of 7.88W/m2K).
Inputting geometry Set snap on: Options > preferences > snap settings > Snap to grid – grid setting of 20mm. otherwise overlapping or geometrically separated rectangles may result.uk .strath. Also set smart snap on and make sure snap to vertical and horizontal is on Draw section A (vertical or wall) Select the rectangle button and VERY CAREFULLY single click anywhere on the white background of the drawing area (the mouse must not move after you have clicked). Now press the following keys 1000 [vertical length of this element] down arrow [command to draw 1000mm downwards from mouse position] 200 [horizontal length of element / thickness of wall] left arrow return Similarly draw section B (roof) Make sure make the initial mouse click exactly at the vertex of the previous rectangle.ac. (Snap to grid ensures this) More complex polygons can be made with the polygon button Scottish Energy Systems Group samuel@esru.
strath. similarly define the wall to be material A. Scottish Energy Systems Group firstname.lastname@example.org .Model Attribution (geometry) Select roof and define it to be material B using the drop down menu.
strath. 2.uk .ac. this makes BC around the perimeter of the model and attributes every BC to be adiabatic. The model should simulate (if not then copy from delegate pack) Scottish Energy Systems Group samuel@esru.Model Attribution (BC) 1. Click on the draw BC icon. Now select each of the boundaries in turn and attribute as relevant Exterior Adiabatic (leave as is) ceiling Internal wall Adiabatic (leave as is) 3.
Select the appropriate U Factor name and press OK Scottish Energy Systems Group samuel@esru.Calculating the psi value I U factor is the heat flow through one meter depth of the model.strath. This is used to calculate the psi value as follows: Psi value = U factor * length of 2D model – sum of U value * length of flanking element Associate a U factor with all internal surfaces of the model as follows: 1.uk . Create a new U factor: Libraries > U factor names > Add > “unique name” e.g. Simple_internal_UFactor > close 2.ac. Select the two internal BC: press the shift key while carefully clicking on the two surfaces > press return and the U Factor window should appear 3.
Calculating the psi value II 4.ac. The product of U factor and length will be used in the formula Psi = U factor * length of 2D model – sum of U values * length of flanking element This value should be less than maximum allowed error norm (see slide on technical details) Note that changing the projected direction changes both U factor and length but the product does not change Scottish Energy Systems Group email@example.com . 5. Rerun simulation Press the U factor icon 6.
13 + + 0.04 W/m2K 7.uk .78W / m 2 K → UA 1 0.0 1 thickness = RSI + + RSO U conductivity 1.04 UA 0.4 T thickness mm 200 200 0.2 = 0.3814*0.2 1.5622 U A = 1.4 1 = 0.2 = 0.2829 U B = 3.2) = -0.923W/mK Scottish Energy Systems Group firstname.lastname@example.org) – (1.1 + + 0.51 1.04 0.04 UB 1.54*1.ac.78*1) – (3.strath.1 λ R_so H_si H_so Surface Surface HTC (film Surface HTC (film Conductivity resistance coefficient) coefficient) m2K/W 0.54W / m 2 K → UB Psi value = (6.13 0.51 1 = 0.Calculating the psi value III Psi value = U factor * length of 2D model – sum of U value * length of flanking element R_si Surface resistance m2K/W Material A Material B 0.0 1 0.69 10 W/m2K 25 25 W/mK 0.
strath.uk .PassivHaus construction example AWm02 DAm02 DAm02 AWm02 Scottish Energy Systems Group email@example.com.
insect screen Open diffusion wind sealing with windproof glued joints Wood shuttering w. 5. 6. 7.04 300 6 0. 1mm gaps b/w boards Mineral wool b/w C posts Brick chipping concrete wall (clay blocks) Lime cement plaster 4 λ (W/mK) 0.12W/m2K Scottish Energy Systems Group samuel@esru.Construction AWm02 AWm02: 1.ac.8 15 U=0. 4. 3. Fibre cement panels Rear ventilation b/w upright aluminium lathes.uk .13 h (mm) 24 5 0.strath. 2.27 200 7 0.
2.13 h (mm) 24 7 0.1 200 9 ignore Ignore U=0. 7.10W/m2K Scottish Energy Systems Group samuel@esru. 3. welded airtight Wood shuttering Mineral wool b/w C sections Reinforced concrete Filler 6 λ (W/mK) 0. 8. PE seal mechanically bonded PP fleece Wood shuttering Ventilated cavity Open diffusion sheet.uk . 5. 4.strath.ac.04 400 8 2.Construction DAm02 DAm02: 1. 6. 9.
ac.strath. Hc=7.69W/m2K PH Ceiling T=20OC.uk . Hc=10W/m2K 2000 PH Sheltered wall T=0OC. Hc=7.Boundary Conditions and Dimensions 2000 PH Sheltered Roof T=0OC.69W/m2K Scottish Energy Systems Group samuel@esru. Hc=10W/m2K PH Internal wall T=20OC.
285*1.ac.uk .3922 Use the model provided if you are not satisfied with the results Scottish Energy Systems Group firstname.lastname@example.org.U factors Define these two surfaces to have a U factor and call it PHExample Now simulate and calculate U factor * length =0.376=0.
0.Psi value Psi value = U factor * length of 2D model – sum of U values * length of flanking elements Psi value = 0.324 = -0.uk .0802W/mK Assumptions built into model: • • C sections have not been included Ceiling filler has not been included Scottish Energy Systems Group email@example.com .ac.10x2.0.12x2 .
uk .Miscellaneous notes It is possible to import images and *.dxf files as under lays (File > underlay) If a space is completely enclosed by polygons then the void can be filled by a polygon (Draw > fill void) Scottish Energy Systems Group firstname.lastname@example.org.
Worksheet • Q1.uk . What is the function of the F7 key.ac. Which surface has a temperature of 10C? Is the heat transfer coefficient of the external wall correct? Which two boundaries are adiabatic and why? _______________________________________________________________________ _______________________________________________________________________ Scottish Energy Systems Group samuel@esru. What is the thermal conductivity of the materials called “BRE 497 – fire stop” and “BRE 497 – plasterboard”? BRE 497 fire stop = ________________ ________________ BRE 497 plasterboard = • Q3. the right mouse button and control + right mouse button? F7 right click control + right click = = = ________________ ________________ ________________ • Q2.strath.
strath. What are the dimensions of the trapezium at the junction of the wall and roof? (mark on image) • Q6.Worksheet • Q4.uk . What are the dimensions of the rectangle representing roof insulation? length = ________________ breadth = ________________ • Q5.ac. What is the minimum internal surface temperature for this model? Minimum internal temperature = ________________ Scottish Energy Systems Group samuel@esru.
ac.Mathematical proof of psi values from U factors (Not required for generating psi values from THERM) ψL∆T = P2 D − P1D P2 D − P D 1 ψ= L∆T U f L∆T − ΣUA∆T = L∆T U f L − ΣUL ×1 = 1 = U f L − ΣUL Scottish Energy Systems Group email@example.com .
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.