Pardo Seal A

154
Features, project design, dimensioning, laying and testing LO2 - 242/0 Valsir underfloor heating system TECHNICAL MANUAL

description

incalzire in pardoseala valsir

Transcript of Pardo Seal A

Page 1: Pardo Seal A

Features, project design, dimensioning, laying and testing

LO2 -

242/0

Valsir underfloor heating system

TECH

NIC

AL M

AN

UA

L

Page 2: Pardo Seal A

IINNDDEEXX

1. Characteristics of floor heating systems pg. 6

1.1. Thermal well-being1.2. Energy saving 1.3. Hygienic conditions 1.4. Aesthetic advantages

2. Components catalogue pg. 8

2.1. PEXAL and MIXAL pipes2.2. V-ESSE insulating panel2.3. V-ELLE insulating panel2.4. V-BAND edging strip 2.5. V-JOINT band for expansion joints 2.6. V-CLIP anchor clips2.7. V-FOIL anti-humidity film2.8. V-FLUID concrete fluidiser 2.9. Distribution manifold for residential systems2.10. Distribution manifold for industrial systems 2.11. Distribution manifold for high temperature circuits 2.12. End kit for distribution manifold 2.13. Interception valve kit for distribution manifold 2.14. Nut-ring-insert for distribution manifold 2.15. Plug for distribution manifold2.16. Flow meter for distribution manifold 2.17. Mixing kit2.18. Thermostatic valve with liquid sensor 2.19. Thermo-electric head2.20. Circuit control unit2.21. Expansion module for two zones 2.22. Mixing kit pump control module 2.23. Anti-shrinkage grid2.24. Encased metal cabinet for distribution manifold 2.25. Fixer for clips2.26. Pipe un-winder

3. Technical characteristics of the components pg. 15

3.1. PEXAL and MIXAL pipes 3.1.1. General characteristics 3.1.2. Characteristics of crosslinked polyethylene PE-Xb3.1.3. Characteristics of aluminium3.1.4. Mechanical behaviour3.1.5. Heat expansion3.1.6. Resistance to abrasion, deposits and corrosion3.1.7. Oxygen and UV rays barrier 3.1.8. Light-weight 3.1.9. Sound absorption3.1.10. Long life 3.1.11. Heat conductance 3.1.12. Heat output in comparison with other pipes 3.1.13. Pressure loss3.1.14. Quality control3.1.15. Approvals

VALSIR UNDERFLOOR HEATING SYSTEMFeatures, project design, dimensioninglaying and testing

111

222

333

3

Page 3: Pardo Seal A

3.2. V-ESSE and V-ELLE insulating panels3.2.1. V-ESSE panel3.2.2. V-ELLE panel

3.3. Mixing kit 3.3.1. Components of mixing kit3.3.2. Theory: adjustment of mixing kit3.3.3. Practice: adjustment of mixing kit3.3.4. Practical method of adjusting the mixing kit3.3.5 Assembled mixing kit

3.4. Distribution manifold 3.4.1. The manifold components 3.4.2. Practical method of adjusting and balancing the manifold

3.5. Fluidiser for concrete3.6. Control systems

3.6.1. Control units of the heating circuits 3.6.2. Expansion module for two zones3.6.3. Thermo-electric heads3.6.4. Circulator pump control module

4. Valsir underfloor heating systems pg. 55

4.1. Valsir residential system with pocketed panel4.2. Valsir residential system with smooth panel4.3. Valsir residential system for renovation 4.4. Valsir industrial system with smooth panel4.5. Valsir industrial system with insulating gush cement

5. Dimensioning of floor heating systems in accordance with the standard UNI EN 1264 pg. 60

5.1. Introduction5.2. Dimensioning: theory

5.2.1. Floor stratification5.2.2. Required thermal flow5.2.3. Characteristic curves 5.2.4. Thermal flow limit and maximum floor temperature5.2.5. Limit curves 5.2.6. Supply temperature5.2.7. Average floor temperature5.2.8. Downward heat dispersion5.2.9. Length of heating loops 5.2.10. Flow and temperature of heating fluid5.2.11. Design limits in the choice of pipe spacing5.2.12. Balancing of heating circuits

5.3. Dimensioning: practice 5.3.1. Floor stratification 5.3.2. The required thermal flow5.3.3. Thermal flow limit and maximum floor temperature 5.3.4. The characteristic and limit curves 5.3.5. Supply temperature 5.3.6. Circuit dimensioning 5.3.7. Balancing of heating circuits

6. Snow melting with radiant panel systems pg. 96

6.1. Introduction6.2. System types

444

555

666

4

Page 4: Pardo Seal A

6.3. System design6.4. Dimensioning: theory

6.4.1. Required heat output6.4.2. Stratification of radiant panels 6.4.3. Temperature calculation6.4.4. Downward specific heat output6.4.5. Calculation of circuit loops 6.4.6. Calculation of flow rate and temperature of heating fluid

6.5. Dimensioning: practice 6.5.1. Required heat output 6.5.2. Stratification of radiant panels 6.5.3. Temperature calculation 6.5.4. Downward specific heat output 6.5.5. Calculation of circuit loops 6.5.6. Calculation of flow rate and temperature of heating fluid

7. Installation pg. 119

7.1 Preliminary operations and controls7.2. Installation of the manifold and mixing kit7.3. Installation of the edging strip7.4. Installation of the insulating panels7.5. Installation of pipino7.6. Expansion joints7.7. Settlement joints7.8. Filling7.9. Leak test

7.10. Laying of the screed7.11. Heating up

8. Appendix pg. 126

A. Heat transfer B. Heat conductance and resistance of materials C. Wood as a floor covering D. Dimensioning of metal grid reinforcement in the floor E. Anti-freeze liquid in heating circuitsF. Calculation of quantity of concrete for layingG. Insulating panels for floor heating H. Measurement units

777

888

5

Page 5: Pardo Seal A

CChhaarraacctteerriissttiiccss ooff uunnddeerrfflloooorr hheeaattiinngg ssyysstteemmss

The first evidence of underfloor heating (ufh) can be found even in Roman times. The wor-king principles were fairly simple but ingenious; an underground fire was made and thehot fumes were conveyed through ducts under the floor of the home. Only following the war do we start to see the first ufh systems which used hot water insi-de pipes embedded in the floor; unfortunately the poor insulation in the buildings, the hightemperatures and the lack of adequate control systems caused this type of system to losepopularity for quite some time. The energy crisis of the seventies, however, and the issuingof European laws on thermal insulation made it possible to return to this type of heating. Floor heating is, today, certainly the most technically valid solution offered by the heatingmarket in the residential, commercial and industrial sector. The various solutions availableallow maximum flexibility and adaptability to all types of building and construction requi-rements.Furthermore, the use of a heating fluid at low temperatures and the particular stratificationof the heat in the room means important energy saving.

1.1. Thermal well-being When the temperature has a particular distribution in relation to height, a certain comfortis created in the room. Such a temperature distribution is defined the ideal curve of ther-mal well-being. In order to create a condition of “thermal well-being”, there must beslightly warmer areas near the floor and slightly colder areas near the ceiling. For every heating system it is possible to trace a distribution curve of the temperatures. Inunderfloor heating systems, the particular arrangement of the radiant panels and the heattransfer by radiation generates a stratification of the temperatures, which is closest to theideal curve.

1.2. Energy saving Systems with radiant panels, as compared to traditional heating systems, allow an avera-ge energy saving of 10% to 15% at equal environmental temperatures. The reasons for thismarked saving are due to the fact that the large surface of the floor allows the water to cir-culate at low temperatures. For this reason, it is convenient to use heat sources whose per-formance increases when run at low temperatures (heat pumps, condensation boilers, solarpanels, heat recovery systems, zone heating systems).The thermal gradient that is generated with floor heating systems is such that heat loss is lessthan in a traditional heating system. Unlike traditional systems, floor heating offers the possibi-lity of recovering heat that is usually wasted due to the stratification effect of the air, which rea-ches higher temperatures near the ceiling; the higher the ceiling the greater the heat recovery.

111

6

Page 6: Pardo Seal A

With a floor heating system the condition of well-being achieved at an average room tem-perature, which is generally 1°C below the temperature achieved with traditional systemsand therefore, at equal comfort, energy saving is possible. Furthermore, the use of insulating panels to hold the pipe, significantly reduces heat lossand contributes to the increase of system output; traditional heating systems do not requi-re such panels, from a design point of view, and therefore they are never used.

1.3. Hygienic conditions Floor heating naturally rules out the formation of humid areas on the floor, conditionsfavouring dust mites and bacteria are therefore not generated and there will also be noformation of mildew. Unlike traditional systems, there is no combustion of motes, whichprovoke a dry and irritated throat and there are no convective currents, which favour thetransport of dust in the room.

1.4. Aesthetic advantages There are no limits of an architectural nature linked to the presence of heating units; the-refore, there is total freedom in interior decorating. By eliminating the problem of con-densation and mildew, there will be no deterioration of wooden floors or windows andframes.Traditional heating systems limit the space available in arranging furniture whereas floorheating systems allow total use of all available space; it is also of advantage in buildingsof architectural and artistic importance.

More details on heat transfer and temperature distribution for different types of heating

systems can be found in the appendix.

7

Page 7: Pardo Seal A

CCoommppoonneennttss ccaattaalloogguuee

2.3. V-ELLE insulating panel■ (m2)COD.

109002

109003

12,00

12,00

Name

1000

1000

12

12

L (mm)

V-ELLE20/150

V-ELLE30/250

Smooth panel in rolls in expanded polystyrene with grey EPS film, with aluminium finish and redsquares for facilitating installation.

■ (pcs)

1

1

50

50

p (mm)Density

(kg/m3)

30

40

20

30

s (mm)

2.4. V-BAND edging strip■ (m)COD.

109200 125

Name

25 200

L (m)

V-BAND

H (mm)

White insulating strip in expanded polyethylene with self-adhesive side on the whole surface andprotective film divided in two. The strip is coupled with a 40 ºm polyethylene transparent film to pre-vent cement infiltrations.

■ (pcs)

57

s (mm)

■ (m)COD.

100101

113005

113007

113009

113011

100117

100

120

240

120

240

50

Name

14

16

16

20

20

26

2,0

2,0

2,0

2,0

2,0

2,0

De (mm)

PEXAL 14x2

MIXAL 16x2

MIXAL 16x2

MIXAL 20x2

MIXAL 20x2

PEXAL 26x3

thickness (mm) Di (mm)

10

12

12

16

16

20

Multi-layer pipe in crosslinked polyethylene with internal layer in aluminium

2.2. V-ESSE insulating panel■ (m2)COD.

109000

109001

12,15

10,12

Name

750

750

1350

1350

V-ESSE20

V-ESSE30

H (mm)

Pocketed panel in expanded polystyrene with blue EPS film.

■ (pcs)

12

10

20

30

s1(mm)

30

30

Density (kg/m3)

75

75

p (mm)

50

60

s (mm)L (mm)

H (m)

222

8

2.1. PEXAL and MIXAL pipe

Page 8: Pardo Seal A

2.5. V-JOINT band for expansion joints ■ (pcs)COD.

109201 5

Name

25 200

L (m)

V-JOINT

H (mm) s (mm)

7

White insulating band in expanded polyethylene with adhesive at one end by 20 mm to be used withV-JOINT/T support or to be glued to the “mushrooms” of the V-ESSE panel

■ (m)

125

■ (m)COD.

109203 12

Name

1,20

L (m)

V-JOINT/T

T-profile, self-adhesive for fixing to V-JOINT band for expansion joints

■ (pcs)

10

2.6. V-CLIP anchor clips COD.

109400

Name

14, 16, 20

De pipe (mm)

V-CLIP01

Anchor clips for pipe diameters 14, 16, 20 mm to be used with V-ELLE panel.

■ (pcs)

100

COD.

109403

109405

Name

16, 20

26

De pipe (mm)

V-CLIP02

V-CLIP03

Anchor clips for fixing pipes to metal grid for application on insulating concrete.

■ (pcs)

25

25

3÷5

3÷5

Net thread Ø (mm)

2.7. V-FOIL anti-humidity film■ (m)

100

Name

100 1200

L (m)

V-FOIL

H (mm)

Polyethylene anti-vapour film thickness 0,2 mm with self-adhesive ends by 25 mm.

■ (m2)

120

9

COD.

109600

Page 9: Pardo Seal A

2.8. V-FLUID concrete fluidiser COD.

109800

Name

V-FLUID

This additive permits improved concrete flow with less water. Optimises the covering of loops duringinstallation.

■ (kg)

10

2.9. Distribution manifold for residential systems ■ (pcs)COD.

110004

110005

110006

110007

110008

110009

110010

110011

110012

1

1

1

1

1

1

1

1

1

Numberexits

G1”1/4

G1”1/4

G1”1/4

G1”1/4

G1”1/4

G1”1/4

G1”1/4

G1”1/4

G1”1/4

D (inch)

4

5

6

7

8

9

10

11

12

d (inch x mm) I (mm)

214

214

214

214

214

214

214

214

214

Distribution manifold with flow-check valves, valves with thermostatic capacity and brackets forencased cabinet (cod. 112010 and cod. 112011).

G3/4”x18

G3/4”x18

G3/4”x18

G3/4”x18

G3/4”x18

G3/4”x18

G3/4”x18

G3/4”x18

G3/4”x18

L (mm)

247

297

347

397

447

497

547

597

647

Kg

3,90

4,72

5,54

6,35

7,16

8,51

9,33

10,15

10,97

2.10. Distribution manifold for industrial systems ■ (pcs)COD.

110016

110017

110018

1

1

1

Numberexits

G1”1/4

G1”1/4

G1”1/4

D (inch)

6

7

8

d (inch x mm) I (mm)

214

214

214

Distribution manifold with brackets for encased cabinet (cod. 112010).

G3/4”x18

G3/4”x18

G3/4”x18

L (mm)

347

397

447

Kg

5,54

6,35

7,16

2.11. Distribution manifold for high temperature circuits■ (pcs)COD.

110020

110021

1

1

Numberexits

G3/4”

G3/4”

D (inch)

2

3

d (inch x mm) H (mm)

87

87

Distribution manifold for high temperature circuits. Used for bathroom radiator or extra radiatorsin a floor heating system. (To be used with mixing kit cod. 110300).

G3/4”x18

G3/4”x18

L (mm)

155

205

Kg

1,95

2,59

10

Page 10: Pardo Seal A

2.13. Interception valve kit for distribution manifoldCOD.

110032G1”1/4 32

D (inch) d (mm)

Two valves (red and blue) with thermometer 60°C to be used with distribution manifold.

■ (pcs)

1151

L (mm)

2.14. Nut-ring-insert for distribution manifoldCOD.

110035

110036

110037

G3/4”x18

G3/4”x18

G3/4”x18

Fitting for connection of PEXAL and MIXAL pipes to distribution manifolds.

■ (pcs)

10

10

10

14x2

16x2

20x2

Pipe (mm x mm)

2.12. End kit for distribution manifold ■ (pcs)COD.

110025 1

D (inch)

14,5

d (mm)

G1”1/4

D1 (mm)

End kit with valve for system drainage and automatic air vent.

35

A (mm)

G1”1/4 28

B (mm)

95

C (mm)

99

E (mm)

0,566

kg

75

R (mm)

D (inch x mm)

11

2.15. Plug for distribution manifoldCOD.

110040G3/4”x18

D (inch x mm)

Plug for distribution manifold outlets.

■ (pcs)

10

Page 11: Pardo Seal A

2.16. Flow meter for distribution manifoldCOD.

110045

D (inch x mm)

G3/4”x18

Flow meter to be connected to distribution manifold (return side).

■ (pcs)

6

2.17. Mixing kit■ (pcs)COD.

110300 1G1”1/4

D2 (inch)

G3/4”

L (mm) I (mm)

214

Mixing kit with three-speed pump (predominance 4 m, 5 m, 6 m) with bracket for fixing to enca-sed cabinet (cod. 112010 and cod. 112011), safety thermostat, interception valve with thermome-ter, valve with thermostatic capacity at inlet and flow-check valve at outlet, adjustable safety by-pass, three-way motorised valve.

381

H (mm)

400

Kg

7,16

2.18. Thermostatic valve with liquid sensor■ (pcs)COD.

110400 1

A (mm) B (mm)

81,5

Thermostatic valve with immersion sensor.

52

L (mm)

160

d (mm)

11

2.19. Thermo-electric head ■ (pcs)COD.

110430 124

Operating voltage (V)

Thermo-electric head on/off with adaptor for distribution manifold. Controls the flow in floor hea-ting circuits. Is applied to distribution manifolds (return side).

Flow (l/min)

0,6÷2,4

D1 (inch)

D (mmxmm)

M28x1,5

Measuring field (°C)

30÷50°C

12

Page 12: Pardo Seal A

2.22. Mixing kit pump control module COD.

11062024

Module turns off the circulator pump when all the circuits are closed, to avoid activating the safetyby-pass of the mixing kit.

■ (pcs)

1

Operating voltage (V)

2.21. Expansion module for two zonesCOD.

1106102

Number zones

Module for increasing the number of zones controlled per control unit cod. 110600 andcod. 110605.

■ (pcs)

124

2.20. Circuit control unitCOD.

110600

110605

1

6

Control unit of heating circuits. It is the command box of the thermo-electric heads, which operatesdepending on the temperature picked up by the thermostats.

■ (pcs)

1

1

24

24

Number zones

4

14

Max. no circuits Operating voltage (V)

Operating voltage (V)

2.23. Anti-shrinkage gridCOD.

1097002000

L (mm)

Galvanised steel anti-shrinkage grid for anti-shrinkage structure.

■ (m2)

401000

H (mm)

50 x 50

Mesh (mmxmm)

2

Wire Ø (mm)

13

■ (pz)

20

Page 13: Pardo Seal A

2.24. Encased metal cabinet for distribution manifold COD.

112010

112011

1000

1200

Encased cabinet for mixing kit and distribution manifold, adjustable in height and depth. Fire-pain-ted steel.

■ (pcs)

1

1

110 ÷150

110 ÷150

L (mm)

720 ÷ 810

720 ÷ 810

H (mm) s (mm)

2.25. Fixer for clipsCOD.

112000

Fixer for clips cod. 109400 for anchoring PEXAL and MIXAL pipe to V-ELLE panel.

■ (pcs)

1

2.26. Pipe un-winder COD.

112001

Pipe un-winder for PEXAL and MIXAL coils.

■ (pcs)

1

14

Page 14: Pardo Seal A

43

3

2

1

TTeecchhnniiccaall cchhaarraacctteerriissttiiccss ooff tthhee ccoommppoonneennttss

3.1. PEXAL and MIXAL pipes

3.1.1. General characteristics For underfloor heating systems (UFH) Valsir has chosen to use the PEXAL and MIXAL pipefor their excellent thermo-mechanical properties. The PEXAL and MIXAL pipes are characterised by a particular multi-layer structure whichdistinguishes itself from other pipes used in UFH systems in that it possesses an internallayer in aluminium which is wrapped completely around the pipe and makes it completelyoxygen proof. The multi-layer pipe offers all the typical advantages of a metal pipe as wellas those of a plastic pipe and at the same time, the qualities of one material compensatefor the inadequacies of the other. The negative aspects of metal, such as corrosion, toxi-city, incrustations, rigidity, weight and elevated pressure loss, are neutralised by the cross-linked polyethylene, which is in contact with the fluid transported inside the pipe. Thenegative aspects of plastic, such as elevated heat expansion and dimensional instability,the fact that it is permeable to gas and sensitive to UV rays, are all overcome thanks to thelayer in aluminium.

The MIXAL pipe is used in diameters 16x2 and 20x2 for UFH systems both in civil andindustrial sectors. Its structure is composed of:1) An outer layer in high-density polyethylene PE-HD, white in colour, RAL 9003.2) An intermediate layer of aluminium alloy, butt-welded in an axial direction. 3) Two binding layers of adhesive, which unite the intermediate metal layers to the outer

and inner layers of plastic.4) An inner layer of crosslinked polyethylene PE-Xb.

Figure 3.1.1Multi-layer structure of MIXAL pipe

333

15

Page 15: Pardo Seal A

PEXAL pipe is used in diameters 14x2 and 26x3 for use in UFH systems in civil renovation jobs andsnowmelt systems for entrance ramps, car parks, helicopter pads, viaducts, etc.Its structure is composed of:1) An outer layer of crosslinked polyethylene PE-Xb, white in colour, RAL 9003.2) An intermediate layer of aluminium alloy, butt-welded in an axial direction. 3) Two binding layers of adhesive unite the intermediate metal layer to the outer and inner layers of

plastic. 4) An inner layer of crosslinked polyethylene PE-Xb.

The dimensional characteristics are indicated in the following table:

Figure 3.1.2Multi-layer structure of PEXAL pipe

Pipe PEXAL MIXAL MIXAL PEXAL

External diameter mm 14 16 20 26

Total thickness mm 2,0 2,0 2,0 3,0

Thickness of aluminium layer mm 0,3 0,2 0,25 0,58

Weight g/m 100 105 140 287

Volume of water l/m 0,077 0,113 0,201 0,314

Operating temperature °C 0÷80 0÷80 0÷80 0÷80

Maximum operating temperature °C 95 95 95 95

Maximum operating pressure at 95 °C bar 10 10 10 10

Coefficient of thermal expansion mm/m•K 0,026 0,026 0,026 0,026

Internal heat conductance W/m•K 0,44 0,43 0,43 0,47

Inner roughness mm 0,007 0,007 0,007 0,007

Oxygen diffusion mg/l 0 0 0 0

Radius of curvature without pipe bender mm 70 80 100 140

Radius of curvature with pipe bender mm 35 50 80 100

Table 3.1.1. Characteristics of PEXAL and MIXAL pipe for underfloor heating

16

43

3

2

1

Page 16: Pardo Seal A

3.1.2. Characteristics of crosslinked polyethylene PE-XbCrosslinked polyethylene PE-Xb has excellent mechanical characteristics in comparison withnormal high-density polyethylene. The elevated stability of its mechanical properties, even athigh temperatures, makes it ideal for use in heating applications where the fluid conveyed rea-ches elevated temperatures. During the crosslinking process the material undergoes a structu-ral modification, which improves its mechanical and abrasion resistance and its resistance ofchemical agents.

3.1.3. Characteristics of aluminium The aluminium used is the production of the PEXAL and MIXAL multi-layer pipes is made upof sheets of aluminium alloy. The sheet is formed around the layer of PE-X and the two ends,which run along the length of the pipe, are butt welded with a TIG welding process(Tungsten Inert Gas). This technology enables the production of a multi-layer pipe with analuminium thickness of 0,2 mm to 2,5 mm and therefore, large diameter pipes with thicklayers of aluminium. The most important characteristics of the aluminium alloy utilised in the production of themulti-layer pipe are good welding, elevated yield point, storage in dry areas to guaranteethe perfect conservation of the aluminium.

Density kg/m3 950

Minimum degree of crosslinking % 65

Softening temperature °C 135

Tensile strength at 23°C MPa 23

Tensile strength at 100 °C MPa 9

Thermal conductivity coefficient W/m•K 0,38

Specific heat at 23°C kJ/kg•K 1,92

Coefficient of linear expansion mm/m•K 0,2

Aluminium layer in PEXAL and MIXAL

pipes

Figure 3.1.3Aluminium layer in PEXAL and MIXAL pipes

Table 3.1.2. Some characterisitcs of crosslinked polyethylene PE-Xb

17

Page 17: Pardo Seal A

3.1.4. Mechanical behaviour The mechanical characteristics of the pipe make it ideal for use in underfloor heatingsystems. The bending radius corresponds to 2,5 times the pipe diameter and the circularsection in proximity of the bend remains constant. There is no spring-back, that is, oncethe pipe has been bent it remains in the desired position like a metal pipe; in this way,the use of fixing clips which are normally used with all-plastic pipes, is considerablyreduced.

3.1.5. Heat expansionThe heat expansion of PEXAL/MIXAL pipes is 0,026 mm/m•K; this value is comparableto the heat expansion of metal pipes. The table below shows how all-plastic pipes havemuch higher expansion coefficients and, in particular, PE-X has an expansion coefficientof 0,20 mm/m•K.

3.1.6. Resistance to abrasion, deposits and corrosion

PE-X does not corrode and its smooth surface does not favour the formation of incrusta-

tion. As it is not subject to corrosion, there is no build-up of rust particles resulting from

galvanic corrosion. Furthermore, PE-X is particularly abrasion resistant; this is an extre-

mely important characteristic in the proximity of bends where the abrasive effect of fluids

– and particles contained in the fluid – tends to be greater.

3.1.7. Oxygen and UV rays barrier

The inner layer of aluminium makes for a perfect barrier to the passage of gaseous mole-

cules, thus avoiding every danger of corrosion caused by the infiltration of oxygen and

damages caused by exposure to UV rays. In the following table, a comparison is made

between the coefficients of oxygen transmission (Oxygen Transmission Rate) of alumi-

nium, of the material used for the oxygen barrier (EVOH) in PE-X pipes with EVOH, and

of crosslinked polyethylene pipes with no oxygen barrier.

Type of material Heat expansion mm/m•K

PEXAL/MIXAL 0,026

Galvanised steel 0,012

Stainless steel 0,016

Copper 0,016

Plastic materials (PE-X, PE-HD, PB, PPR, PE-RT) 0,120÷0,200

Table 3.1.3. Heat expansion comparison

18

Page 18: Pardo Seal A

In PE-X pipes with barriers, the oxygen transmission coefficient OTR increases as the temperatureand relative humidity rises (Figure 3.1.4 and Figure 3.1.5). Even at 45°C and with a relative humi-dity of 65%, the EVOH barrier has an oxygen transmission coefficient of almost 3,0cm3/20µm•m2•day•bar.Many PEX pipes sold today present an oxygen barrier that is generally on the outside on the pipe.Such a layer is, therefore, significantly exposed not only to the danger of being scraped away butalso to the danger of being cut and the negative effect of humidity which drastically reduces itsimpermeability.

Pipe OTR at 25°C and 0% UR [cm3/20µm•m2•day•bar]

Aluminium 0

EVOH barrier 0,21

PE-X 12000

Figure 3.1.4Coefficient of oxygen transmission of EVOH in relation to temperature

Figure 3.1.5Coefficient of oxygen transmission of EVOH in relation to relative humidity

Table 3.1.4. Coefficient of oxygen trasmissione OTR

19

The oxygen diffusion value in PEXAL/MIXAL pipes is zero, thanks to the presence of

the inner layer of aluminium in all the diameters and independently of temperature

and humidity.

Temperature [°C]

OTR

at 6

5% U

RO

TR a

t 20°

C

Relative humidity UR [%]

Page 19: Pardo Seal A

3.1.8. Lightweight The specific weights of the materials that make up the pipe are low. A coil of 100 metres of 16x2weighs approximately 10,5 kg.

3.1.9. Sound absorption The acoustic properties of the pipe are very good. The inner and outer layers in polyethylene mini-mise the noises, which are normally not absorbed by metal pipes.

3.1.10. Long life The PEXAL and MIXAL pipes are designed to resist a pressure of up to 10 bar with working tempe-ratures of 95°C. The crosslinked polyethylene possesses, in fact, a very high ageing resistance.Artificial ageing tests carried out in laboratories guarantee the pipe a life of over 50 years. At working temperatures below 95°C, the pipe can support pressures of over 10 bar without anydamage being caused; at 20°C it can be used at a pressure as high as 25 bar.The technical characteristics of the PEXAL and MIXAL pipes are therefore of an elevated level, espe-cially if they are compared with the real working conditions of UFH systems which, on average, ope-rate at temperatures of 45°C and pressures which do not go above 2÷2,5 bar.

3.1.11. Heat conductance The heat conductance of PEXAL/MIXAL pipes depends on the multi-layer structure of the pipe, andin particular, on the thickness and the position of the aluminium layer. Whereas the value for PE-Xpipe is 0,38 W/m•K the PEXAL and MIXAL pipes have a value of 0,43 W/m•K for diameter 16x2and 0,47 W/m•K for 26x3 (see Table 3.1.1). This difference clearly favours the use of PEXAL andMIXAL pipes for UFH in that it is possible to install systems with an optimum heat output.

3.1.12. Heat output in comparison with other pipesAs seen in the previous paragraph, the presence of the aluminium layer, its thickness and its parti-cular position allow the achievement of excellent heat conductance properties. With PEXAL and MIXAL, it is possible to install UFH systems with higher heat outputs, in fact, thehigher conductance generates higher temperatures on the surface of the pipe than PEX pipes (seeFigure 3.1.6) and this advantage is reflected, for example, in the possibility of obtaining relativelylower supply temperatures (see Figure 3.1.7).

The safety margin of PEXAL and MIXAL pipes in UFH systems in very high. Consider thatat 95°C and with a safety margin of 1.5, the pipe can be used at 10 bar.At the same tem-perature, therefore, if used at 2.5 bar, the safety coefficient increases to 6 and, clearly, goesup even further if the temperature is reduced to 45°C.

20

Page 20: Pardo Seal A

Figure 3.1.6External surface temperature of the pipe (example)

Figure 3.1.7Supply temperature (example)

The better performance of PEXAL and MIXAL in comparison with PEX pipes is evident inFigure 3.1.8 where, at equal system conditions, a greater heat output is obtained. In thecase examined, with a spacing of 15 cm or of 22,5 cm, the heat output is increased by over2,2%.

21

PEXAL and MIXAL PEX PIPE

Page 21: Pardo Seal A

Figure 3.1.8Comparison of output of MIXAL 16x2 and PEX 16x2

22

Page 22: Pardo Seal A

It is evident that the flow and speed of the circuits are more or less the same and the-

refore, that the 16x2 diameter MIXAL pipe can be used instead of the 17x2 diameter

PEX pipe.

Characteristics PEX 17x2 MIXAL 16x2

Pipe spacing [cm] 15 15

Supply temperature [°C] 45 45

Loop length [m] 66,7 66,7

Temperature difference ∆T [°C] 18,6 18,8

Flow [l/h] 46,6 46,1

Speed [m/s] 0,10 0,11

Table 3.1. Comparison between PEX and MIXAL with pipe spacing of 15 cm.

The considerations examined allow us to reach a very important conclusion. WithPEXAL/MIXAL it is possible to use smaller diameter pipes than those used with all-plasticpipes. To simplify the concept, let us imagine that we need to install an UFH circuit for a 10 m2

room that requires a specific heat output of 80 W/m2. The floor is composed of a Valsir V-ESSE20 insulating panel, the layer of concrete above the pipes is 40 mm and for simplicitysake, we will not take any type of floor covering into consideration. In the two tables below,a comparison is made between the values of two circuits installed with a 17x2 diameter PEXpipe and a 16x2 MIXAL pipe with two pipe spacing values and a supply temperature of45°C.

Characteristics PEX 17x2 MIXAL 16x2

Pipe spacing [cm] 22,5 22,5

Supply temperature [°C] 45 45

Loop length [m] 44,4 44,4

Temperature difference ∆T [°C] 14,5 14,1

Flow [l/h] 61,4 60,0

Speed [m/s] 0,13 0,15

Table 3.2. Comparison between PEX and MIXAL with pipe spacing of 22,5 cm.

23

Page 23: Pardo Seal A

Figure 3.1.9Heat output PEX 17x2 and MIXAL 16x2

24

Equal thermal output 12 W/mPipe spacing 15 cm

Equal thermal output 18 W/mPipe spacing 22.5 cm

Page 24: Pardo Seal A

In sizing a UFH circuit it is necessary to keep in mind localised pressure losses due to the

continuous changes in direction of the radiant loops.

Table 3.1.5. Percentage increase of pressure loss in relation to the type of pipe layout.

Type of layout pattern Percentageincrease

Typical use

Single serpentine 17% Industrial systems, snowmelt (Figure 3.1.10)

Double serpentine 17%Industrial systems, heating systems of rooms withlarge surface, fitness centres, warehouses, etc.(Figure 3.1.11)

Counter flow spiral 13% Residential systems (Figure 3.1.12)

3.1.13. Pressure loss

The inner layer of the pipe has an extremely smooth surface with a roughness of 0,007 mm. This

surface does not favour the formation of incrustations or rust, which means that pressure loss is very

low and does not alter with time.

By using the diagrams of Figure 3.1.13, Figure 3.1.14 and Figure 3.1.15 it is possible to deter-

mine the pressure loss and flow speed in the PEXAL and MIXAL pipes in relation to the flow rate and

temperature of the water at 10°C, 30°C and 50°C respectively.

The linear pressure losses (calculated in the diagrams) must be increased by a percentage point, indi-

cated in Table 3.1.5, which depends on the type of pipe layout adopted in the system.

25

Page 25: Pardo Seal A

Figure 3.1.11Double serpentine

Figure 3.1.12Counter flow spiral

Figure 3.1.10Single serpentine

26

Page 26: Pardo Seal A

Figure 3.1.13Pressure loos with water at 10°C

27

Page 27: Pardo Seal A

Figure 3.1.14Pressure loss with water at 30°C

28

Page 28: Pardo Seal A

Figure 3.1.15Pressure loss with water at 50°C

29

Page 29: Pardo Seal A

3.1.14. Quality control

1. The principle test performed on the multi-layer pipe by theValsir Quality Control Function is to measure the diametersand the thickness of the individual layers. This test is perfor-med in the laboratory with the help of the last generation opti-cal measurement system, complete with software capable ofautomatically carrying out the dimensional tests (the dimen-sions of the pipe are tested in the process using laser detec-tors). The sophisticated optical projector also enables the weldcross-section to be checked and therefore to verify that theweld has been correctly executed.

2. A very important test which Valsir performs on the multi-layerpipe manufactured by the Company is the separation test; thistest is performed using a computerised dynamometer capableof assessing the force required to separate the Aluminiumlayer from the internal pipe (glued together). As a result, thetest provides a graph describing the value of the force (atevery point on the pipe's circumference) to be applied to sepa-rate the layers. The adhesion between the PEX and theAluminium is fundamental for the seal of a multi-layer pipeunder pressure: the higher the glueing strength the higher thepressures the product can undergo.

3. The 90° bending test is one of the mechanical tests performedon the Pexal pipe. This test is performed using a dynamome-ter that records the force required for bending. The test is pas-sed if no squeezing or wrinkling of the external layer occurson the test specimen.

4. The finished Pexal pipe (which has already completed thecross-linking process) is subjected to a test that measures thedegree of cross-linking achieved by the polymeric materials.The testing procedures are defined by international standardsand are strictly followed by the operators assigned to performthe test; the degree of cross-linking of the polymers is used toassess the greater strength of material's molecules aggrega-tion and is therefore important to appreciate the increase inthe mechanical and chemical resistance of the polyethylene.

In Valsir operating centres, the whole manufacturing process of the multi-layer pipe undergoes strictquality tests. Besides the tests foreseen by the protocols of the major international standard institutes,Valsir carries out important and high-level quality tests. The following is a list of the main tests perfor-med on Pexal pipe.

30

Page 30: Pardo Seal A

5. Samples of the pipe are taken (at pre-established intervals)during the daily production of the Pexal product. The samplesare used to perform hydraulic tests at different pressures andat different temperatures. The tests are designed to determinewhether the product is suitable for sale and to assess the pro-duct's hydraulic and mechanical-structural characteristics; thetests are performed by using suitable tanks or ovens at an elec-tronically controlled temperature; the pressure values set at theentry of each individual test specimen and the test conditionsare controlled and recorded at every moment by a computeri-sed system.

6. The pressure tests at 80°C and at 20°C provide importantinformation about the mechanical characteristics of the system.These tests are performed in suitable tanks containing water,and are followed by an in-air-test performed at 110°C byusing specific equipment.

7. Samples of Pexal pipe taken during the production phase atregular intervals are subjected to the cone test; the test is per-formed in compliance with the international standards, and iscarried out on-line by the production operators and in thelaboratory by the Quality operators (in this case the test is per-formed using a computerised dynamometer); the test is desi-gned to assess the seal of the weld and the sealing strength ofthe glue applied between the various layers, after havingexpanded the pipe by more than 13% of the nominaldiameter.

8. The diameters of the pipe are constantly monitored by laserinstruments during the production of Pexal, supported by acomputerised system in the successive phases of the manufac-turing process; in this way the production operators are ableto observe the trend graphs of the individual diameters on theline monitors at every moment; appropriate alarms are activa-ted when the values lie outside the pre-established range.

31

Page 31: Pardo Seal A

9. Aluminium is the fundamental raw material in the productionof the Pexal pipe; Valsir monitors every incoming delivery bymeasuring the dimensions and mechanical characteristics toavoid defects in the supply of this material (although primarysuppliers have been selected); the mechanical properties areverified by performing tensile tests (established by internationalstandards) on samples of the material taken at random fromthe delivered batch; the tests are performed using sophisticatedcomputerised dynamometric instruments.

10. All the polymeric raw materials used in the production of themulti-layer pipe are tested upon arrival to verify their maincharacteristics; this procedure enables Valsir to manufacturethe products while being certain of using suitable materials forthe production; cutting edge instruments are utilised to per-form the tests: for example, the melt index is measured usingthe latest generation automatic equipment.

11. The tests performed by Valsir's quality laboratory on the poly-meric materials utilised in the production of the Pexal pipe arenot limited to the acceptance tests, but continue after the pro-duction phase; the shrinkage and separation test of thevarious polyethylene layers is performed on the finished pro-duct by subjecting the pieces of pipe to ageing and thermalstress tests inside a thermostatic chamber.

12. Each coil of multi-layer Pexal produced is tested at the side ofthe line by inserting a steel sphere in the pipe which is thenforced inside the pipe by compressed air and by passingthrough ensures that no yielding has occurred or that noobstructions are inside the pipe.

32

Page 32: Pardo Seal A

MC - GOST: Certificate of conformity of the PEXAL system to be used to trans-port hot and cold drinking water in water distribution and heating systems.

RINA: Type approval, which guarantees the use of the PEXAL multi-layer system onships for water distribution, heating and air-conditioning systems.

Polish certificate of approval for installation of the Pexal pipes for heating and dis-tribution of hot and cold water for domestic use.

Polish certificate of approval for the Pexal system to distribute drinking water.

Type approval that guarantees the suitability of the Pexal multi-layer pipefor use in delivering hot water to heating systems.

IIP-UNI: Certificate of conformity of the Pexal multi-layer pipe systems to be utili-sed for hot and cold water and heating systems.

AS/NZS 4020: Type approval guaranteeing the suitability of the PEXAL multi-layer pipe for coming into contact with drinking water with special referenceto the following tests: taste - appearance – microbiological growth - cytotoxicactivity - mutagenic activity caused by metals extraction.

W270: Type approval that guarantees the suitability of the Pexal multi-layer pipefor coming into contact with drinking water with special reference to the propa-gation of micro-organisms on the material surface. (Certificate n° KU18946/1)

Type approval that guarantees the suitability of the Pexal multi-layer pipe foruse in delivering hot water to heating systems.

IIS: Qualification certificate of the aluminium welding procedure adopted in the pro-duction of the Pexal multi-layer pipe, in accordance with specifications EN 288-8:1997 Annexes No. 12.

AS 4176 SPEC. 438 LN IP083: Certificate of conformity of the Pexal system tobe utilised to distribute hot and cold drinking water under pressure.

Product certification referring to the suitability of the Pexal multi-layersystem for use in heating systems.

Product certification referring to the suitability of the Pexal multi-layersystem for use in sanitary systems.

Suitability certificate for the Pexal multi-layer pipe to transport hot and colddrinking water in sanitary systems.

Type approval that guarantees the use of the Pexal system to distribute hotand cold water in sanitary and heating systems.

Type approval that guarantees the use of the Pexal system to distribute hotwater in heating systems.

TYPEAPPROVAL

RI N A

18 6 1

UNI 10954-1

-LT 606--LT 813-

Spec. 438LN IP083

224811 MX03

AT/2000-02-0873

DVGW W270TGM-KU 18946/1

K 22518/01

K 22504/01

BS 6920

POTABILITÀ

AS/NZS 4020

Önorm B5157GEPRÜFT

Type approval that guarantees the use of the PEXAL multi-layer system on boardships for sanitary, heating systems.

A-794/2002-I

3.1.15. Approvals

33

Page 33: Pardo Seal A

3.2. V-ESSE and V-ELLE insulating panels

3.2.1. V-ESSE panels The V-ESSE panel is a pocketed, expanded polystyrene panel with a blue EPS film, whichgives it a good surface resistance to stamping. It has been studied and designed for resi-dential systems, commercial areas or warehouses where floor load is not very high. In factthe panel has a density of 30 kg/m3 with a compressive strength of 150 kPa.Spacing is 75 mm and the panel is available in two thicknesses. V-ESSE20 has a basethickness of 20 mm with a total of 50 mm, V-ESSE30 has a base thickness of 30 mm anda total of 60 mm.The V-ESSE panel is characterised by an L-profile joint, which allows a stable connection.The laying of the pipe is facilitated by alternated incisions on the bosses; this allows longlengths of pipe to be laid by following the bosses with the same incision (see Figure 3.2.1and Figure 3.2.2).

Figure 3.2.1V-ESSE panel

Figure 3.2.2Dimensions of V-ESSE panel

34

Euroclasse E

Page 34: Pardo Seal A

3.2.2. V-ELLE panelV-ELLE is a smooth panel in coils of expanded polystyrene with grey EPS film and red squa-res with a spacing of 50 mm to facilitate installation. It is supplied in coils, which are laidwith extreme facility by simply unrolling the panel onto the floor. It is supplied in two ver-sions differing in thickness and compressive strength; these structural differences makethem suitable for different uses. The V-ELLE20/150 panel has a thickness of 20 mm, a density of 30 kg/m3 and a compres-sive strength of 150 kPa. These characteristics make it suitable for residential heatingsystems but especially for areas where the available height for the installation is limited(less than 100 mm).The V-ELLE30/250 panel has a thickness of 30 mm and a density of 40 kg/m3. Given itselevated compressive strength of 250 kPa, it can be utilised both in residential and indu-strial jobs, wherever the surface load is very high. It is also ideal for snowmelt and de-icing systems (entrance ramps, car parks, squares, etc.).

Panel characteristics Measurement Unit V-ESSE20 V-ESSE30 Reference standard

Use -

Panel material - Expanded polystyrene Expanded polystyrene

with blue EPS film with blue EPS film -

Surface type - Pregrooved Pregrooved -

Dimensions used panel H x L mm x mm 1350 x 750 1350 x 750 -

Dimensions total panel H1 x L1 mm x mm 1370 x 770 1370 x 770 -

Panel surface m2 1,012 1,012 -

Minimum spacing p mm 75 75 -

Insulation thickness s1 mm 20 30 -

Total height s mm 50 60 -

Density kg/m3 30 30 UNI 6349

Compressive strength kPa 150 150 UNI EN ISO 13163

Flexural strength kPa 250 250 UNI EN ISO 13163

Fire resistance - Euroclasse E Euroclasse E EN 13501-1

Dimensional stability at 70° for 48 h % 0,5 0,5 UNI EN ISO 13163

Heat conductance W/mK 0,034 0,034 UNI EN ISO 13163

Heat resistance m2K/W 0,55 0,85 -

Packaging - Cardboard box Cardboard box -

Number panels per package - 12 10 -

Surface per package m2 12,14 10,12 -

Table 3.2.1. Characteristics of V-ESSE panels

Residential systems or commercial areas suchas offices and shops or warehouses with floor

loads of medium intensity.

35

Page 35: Pardo Seal A

Figure 3.2.3V-ELLE panel

Figure 3.2.4Dimensions of V-ELLE panel

Panel characteristics Measurement unit V-ELLE20/150 V-ELLE30/250 Reference standard

Use

Panel material

Surface type - smooth smooth -

Panel dimensions L x H mm x m 1000 x 12 1000 x 12 -

Panel surface m2 12 12 -

Total height mm 20 30 -

Density kg/m3 30 40 UNI 6349

Compressive strength kPa 150 250 UNI EN ISO 13163

Flexural strength kPa 200 350 UNI EN ISO 13163

Fire resistance - Euroclasse E* Euroclasse E* EN 13501-1

Dimensional stability at 70° for 48 h % <1 <1 UNI EN ISO 13163

Heat conductance W/mK 0,034 0,033 UNI EN ISO 13163

Heat resistance m2K/W 0,55 0,90 -

Packaging - Polyethylene bag Polyethylene bag -

Number panels per package - 1 1 -

* This characteristic refers to the panel base in expanded polystyrene.

Residential systems or commer-cial areas such as offices andshops or warehouses with floorloads of medium intensity.Studied especially for systemswhere height is limited or forrenovation jobs.

Residential systems or commer-cial areas but above all it is sui-table for industrial systems dueto its elevated compressivestrength. It is suitable for snowmelt andde-icing systems.

Expanded polystyrene with greyEPS film with red squares, spa-cing 50 mm

Expanded polystyrene with greyEPS film with red squares, spa-cing 50

36

Euroclasse E

Page 36: Pardo Seal A

3.3. Valsir mixing kit

The mixing kit is used to mix hot supply water from the boiler and return loop water in order to havea constant supply temperature. The kit is supplied with a three-speed circulation pump for the “secon-dary” circuit, thus making the Valsir mixing kit very versatile and applicable to all types of systems,be they small or big. The kit is fixed point but it can be converted to a variable point by motorizingthe three-way valve.

Figure 3.3.2Scheme of the components of the Valsir mixing kit

Figure 3.3.1Valsir mixing kit

37

Page 37: Pardo Seal A

3.3.1. Components of mixing kit ➊ System inlet flow-check valve. It is a valve with a thermostatic option, which regulates the inlet

flow to the mixing kit.➋ Liquid sensor thermostatic head. The temperature measurement probe is inserted in chamber 5

and commands the head that controls the incoming fluid to the system according to the tempe-rature set at the graded knob.

➌ Three-way valve. In this valve the in-coming water is mixed with the system return water. It isregulated by means of the graded knob; this adjustment is made manually (from which “fixedpoint” is derived), it is, however, possible to motorize the three-way valve by installing a motorconnected to a climatic control station (“variable point”).

➍ Circulation pump. It is a three-speed pump corresponding to three different pressure heads, 4m, 5 m, 6 m. Due to this characteristic the mixing kit is suitable for all system sizes.

➎ Chamber for the probe of the thermostatic head. ➏ Safety thermostat. The safety thermostat controls the motor of the pump, stopping it when the tem-

perature picked up in the supply circuit is higher than the set limit value.➐ Interception valve of the supply circuit with thermometer.➑ Interception valve of the return circuit with thermometer. ➒ Safety by-pass on the pump. The by-pass permits circulation of the fluid even when all the hea-

ting circuits are closed (e.g. when the rooms have reached the set temperature). Without a by-pass there would be the risk of damaging the motor of the pump; by keeping the flow-check valveof the by-pass open by a few turns (not more than three) the pump is not damaged in any wayas a minimum circulation flow is always guaranteed. It is possible to keep the by-pass closed byinstalling an electronic control station, which stops the pump when it picks up that all the ther-moelectric heads are closed.

➓ Mixing kit outlet valve.

38

Page 38: Pardo Seal A

Figure 3.3.3Characteristic curve of the three-way valve

Figure 3.3.4Characteristic curve of the thermostatic head

39

Page 39: Pardo Seal A

Figure 3.3.5Characteristic curve of the complete kit

Figure 3.3.6Drawing of flow of the Valsir mixing kit

40

Page 40: Pardo Seal A

3.3.2. Theory: adjustment of the mixing kitIn the drawing of Figure 3.3.6 the following are indicated:

the flow in the secondary circuit,the flow in the primary circuit,the re-circulation flow in the secondary,the temperature of the supply fluid in the primary circuit,the temperature of the supply fluid in the secondary circuit,the return temperature, which is the same in the primary and secondary and is given by:

[3.3.1]

[3.3.2]

[3.3.4]

m.s

m.p

m.x

Tp,m

Ts,m

Tr

[3.3.5]

[3.3.6]

considered the average based on the flow of the return temperatures of the n heating circuits.Applying the mass balance to the three-way valve, we have:

Whereas the energy balance applied to the three-way valve is:

Simplifying the two relations, we obtain:

and thus

[3.3.3]

41

Page 41: Pardo Seal A

With this last formula it is possible to calculate the flow that the boiler must supply to the heatingsystem and based on the diagram shown in Figure 3.3.3 we can determine in what position thethree-way valve must be set.

3.3.3. Practice: adjustment of mixing kitIn order to further clarify the use of the formulas shown previously, reference is made to the exam-ple in the chapter “Dimensioning of floor heating systems in accordance with the standard UNI EN1264”, the results of which are shown in the following table.

In the example, the calculation of the average return temperature gives:

Consider a supply temperature from the boiler of Tp,m = 70°C.In the secondary circuit Ts,m = 46 °C and m s = 0,0689 l/s, the flow in the primary circuitis thus calculated:

The in-coming flow percentage to the mixing valve is given by the following ratio:

From the diagram in Figure 3.3.3 it can be seen that the three-way valve must be set tovalue 4. From Figure 3.3.4 it can be seen that the thermostatic head must also be set tovalue 4.

m.

Table 3.3.1. Details of an example of a floor heating system

N. Room Spacing p [cm] Tm Tr

[°C] [°C] [l/s]

1 Bedroom 15 46 29,5 0,0154

2 Living room 22,5 46 33,5 0,0277

3 Kitchen 15 46 28,0 0,0121

4 Bathroom 15 46 35,0 0,0137

0,0689

42

Page 42: Pardo Seal A

3.3.4. Practical method of adjusting the mixing kitWhat we have seen previously can be put into practice by observing the following instructions, whichrefer to the same calculation example. a) Adjust the thermostatic head to the supply temperature required in the secondary circuit, in the

case of the example Ts,m = 46°C which corresponds to the value on the head equal to 4 (seeFigure 3.3.4).

b) Initially set the three-way valve to position 2.c) Check the in-coming temperature to the manifold by means of the supply thermometer.d) If the temperature is below the value set on the head (Ts,m<46°C): adjust the three-way

valve by moving slowly from position 2 toward position 3. Wait until the supply tempera-ture has stabilised.

e) If the supply temperature on the secondary is still below the requirement, repeat the previous stepby progressively increasing the adjustment position of the valve until the required temperature hasbeen reached.

3.3.5. The assembled mixing kit The Valsir mixing kit is made by assembling the basic components in relation to the type ofsystem that has to be realised as indicated in figure 3.3.5. To aid in the choice of the components, in the following tables the codes of the products whichmake up the complete mixing kit are indicated both for the version with high temperaturemanifolds (for supplying radiators, towel rails) and for the version without such a manifold. Furthermore, the sizes of the assembled kit are indicated and the code of the metal case is sug-gested for installing the kit.

43

Table 3.3.2. Composition of the mixing kit complete with high temperature manifold

112010

112011

No. outlets

4

5

6

7

8

9

10

11

12

796

846

896

946

996

1046

1096

1146

1196

480

480

480

480

480

480

480

480

480

110004

110005

110006

110007

110008

110009

110010

110011

110012

Size (mm)

A1(a) B1

Product codes (ref. Chapter 2)

Manifold ➋ Metal case(b)

110300

Mixer➊

110020

or

110021

High tempe-rature kit ➎

110025

End group➌

110400

Thermostaticvalve ➍

110035

or

110036

or

110037

Fittings (c)

110430

Thermo-electrichead (d)

(a) The sizes take into consideration the high temperature kit with 3 outlets.(b) Code suggested based on the total size of the system.(c) The code depends on the dimensions of the pipe used.(d) The use of thermoelectric heads depends on the type of circuit control system to be carried out.

Page 43: Pardo Seal A

44

Table 3.3.3. Composition of mixing kit complete with high temperature manifold

112010

112011

No. outlets

4

5

6

7

8

9

10

11

12

716

766

816

866

916

966

1016

1066

1116

400

400

400

400

400

400

400

400

400

110004

110005

110006

110007

110008

110009

110010

110011

110012

Size (mm)

A2 B2

Product codes (ref. chapter 2)

Manifold ➋ Metal case (b) Mixer➊ End group ➌

110400

Thermostatic valve➍

110035

or

110036

or

110037

Fittings (c)

110430

Thermoelectrichead (d)

(b) Code suggested based on the total size of the system.(c) The code depends on the dimensions of the pipe used.(d) The use of thermoelectric heads depends on the type of circuit control system to be carried out.

110300 110025

Keep in mind that the fittings are chosen based on the type of pipe that is used and thethermoelectric head is chosen based on the type of circuit control system.

Figure 3.3.7Assembled mixing kit

Page 44: Pardo Seal A

3.4. Distribution manifold

The pre-assembled Valsir manifold is obtained from a drawn brass bar with flow-check valves on thesupply manifold and thermostatic valves on the return manifold and it is supplied with mountingbrackets for anchoring it to the cabinet. It is equipped with flow-check valves with a “memory”, thatis, once the system has been balanced, the maximum opening of the valves can be blocked (bymeans of a special tool) so that they can be used as stop flow valves in the circuit. This is a specialsystem, in that the flow-check valve can be re-opened and it automatically brings itself to the num-ber of turns corresponding to the balancing value. Another important aspect, is that the Valsir mani-folds are equipped with an automatic air vent on both the supply and return manifolds with a systemdrainage valve. Finally, irrespective of the number of outlets, it was chosen to make manifolds witha 1”1/4 diameter, compared with other manifolds currently available on the market, in order toimprove flow.

3.4.1. The manifold components

Figure 3.4.2Scheme of the components of the Valsir manifold

45

Figure 3.4.1Valsir distribution manifold

Page 45: Pardo Seal A

The total width of the manifold and its weight are indicated in the following table:

➊ Supply manifold.➋ Return manifold.➌ Flow-check valves. All of the flow-check valves on the supply manifold have the

“memory” function to block the maximum opening. This function is particularly usefulwhen the manifold has been tampered with, as it avoids having to balance the systemonce more.

➍ Valves with thermostatic capacity. On the valves with thermostatic capacity, it is possi-ble to apply a thermoelectric head connected to the thermostat in the room to be hea-ted. Once the temperature set has been reached, the thermoelectric head actuates thevalve, progressively closing the circuit and thereby reducing the flow of the heatingfluid.

➎ Automatic air vent and system drainage group. All of the Valsir manifolds are equippedwith this group both on the supply and return. The group is made up of an automatic airvent, which eliminates air from the system; the air vent valve is equipped with a non-return valve which permits the total evacuation of air from the system. The group is alsoequipped with a discharge valve with an insert for system drainage.

➏ Mounting brackets for anchoring the manifold to the cabinet.➐ Hexagonal 6 mm key for regulating the flow-check valves.

Table 3.4.1. Dimensions and weight of Valsir manifolds

Outlets L (mm) Weight (kg)

4 296 5,05

5 346 5,87

6 396 6,69

7 446 7,50

8 496 8,31

9 546 9,66

10 596 10,48

11 646 11,30

12 696 12,12

46

Page 46: Pardo Seal A

Figure 3.4.3 Characteristic curve of flow-check valves on the supply manifold

Figure 3.4.4 Characteristic curve of the thermostatic valves on the return manifold

47

Page 47: Pardo Seal A

3.4.2. Practical method of adjusting and balancing the manifoldThe theoretical method of determining the flow rate and balancing the circuits is dealt with inthe chapter pertinent to “Dimensioning of floor heating systems in accordance with the stan-dard UNI EN 1264”. The adjustment and balancing, as well as blockage of the valve opening(“memory”) are carried out by following the instructions shown below and outlined in Figure3.4.5.

Figure 3.4.5Valve adjustment for circuit balancing

a) Insert the key in the valve and close the actuator ➊ by acting exclusively on the 6 mm hexago-nal key ➋.

b) Turn the fitted key ➌ in an anti-clockwise direction thus making the sleeve ➍ fully descend withoutmoving the 6 mm hexagonal key.

c) Adjust the sleeve ➍ with the fitted key ➌ to the number of turns obtained from the characteristiccurve of the flow-check valve.

d) Fully re-open the actuator ➊ using only the 6 mm hexagonal key ➋.

48

(a) (b) (c) (d)

Page 48: Pardo Seal A

Table 3.5.1. Characteristics of V-Fluid fluidizer for concrete

Additive designed to give concretes more workability at reducedwater ratio and increased concrete strength.

Placing of poured concrete floors.

Added to the paste water in making concrete.

0,2%÷0,3%

Reduction of paste water in the range of 10 to 15%Improved workability in the realization of screed for floor systems. Liquid

Brown

Characteristic

7

Soluble

1,1 kg/l

10 kg container

12 months in unopened package and out of direct sunlight

Not applicable below 5°C.Too much product can compromise the mechanical resistance of the cement.

Description

Field of application

Use

Dosage

Fundamental performance

Appearance

Colour

Odour

pH

Solubility in water

Density

Packaging

Storage

N.B.

3.5. Fluidizer for concrete

V-Fluid fluidizing additive has been studied to improve the plasticity and fluidity of concrete and mor-tar. At the same time it allows a reduction of the water used for mixing by 10% to 15%, without lowe-ring the resistance of the concrete, and obtains high fluidity and workability. Dosage ranges from 0,2% to 0,3% of the cement weight.

49

Page 49: Pardo Seal A

3.6. Control systems

3.6.1. Control units of the heating circuits The Valsir control unit allows management of all the circuits in the floor heating system at the distri-bution manifold by means of simple installation and wiring operations. It can be easily installed inthe manifold cabinet by using a standardised mounting guide supplied with the package or else itcan be mounted directly on the wall. The Valsir control unit is provided with a transformer which brings the network voltage (220 Volt) to24 Volt; in this way, the installation can be made directly by the heating system installer.

Figure 3.6.2Mounting of central unit on the guide

Figure 3.6.3Wiring

The connections are rapidly and easily carried out thanks to the colouring of the terminals which isthe same as the wires and also the use of a screwdriver is not required. Any extra modules are easilyassembled to the principal unit.

Figure 3.6.1Positioning of control unit

50

Page 50: Pardo Seal A

It is possible to use any type of thermostat or thermostat with timer, even if they differ from each other.Furthermore, the control unit allows the signals to be easily divided amongst one or more zones. Themodularity of the control unit means that the configuration can be changed at any time and extramodules allow the number of controllable zones, or the number of thermo-electric heads connectedto the same zone, to be increased.There are two types of Valsir control units: the smallest one controls up to four thermo-electric headsand it is commanded by a single thermostat (1 zone); the biggest one allows to control up to 14 ther-mo-electric heads and can be interfaced with 6 thermostats (6 zones).

Figure 3.6.4Colouring of terminals

The presence of LEDs allows monitoring of the circuit (open/closed) at any time and a fuseprotects the unit in case of wiring errors. The card of the control unit is protected in the caseof a short circuit of one of the thermoelectric heads of the mixing kit circulator.

Figure 3.6.5LED

Figure 3.6.6Modularity of Valsir control unit

51

Page 51: Pardo Seal A

Table 3.2.1. Control unit for 1 zoneCod. 110600

Supply voltage

Operating voltage

Base colour

Protection colour

Dimensions

Predisposed for connector

LED signals for

230V

24V 50/60 Hz

grey RAL 7031

transparent

L 93 mmH 70 mmB 75 mm

1 thermostat

Max. 4 thermo-electric heads (cod. 110430)

state of actuators state of supply state of fuse

Control unit via wire for the regulation of 1 zone, can be expanded with another module(cod. 110610).

Table 3.2.2. Control unit for 6 zones Cod. 110605

Supply voltage

Operating voltage

Base colour

Protection colour

Dimensions

Predisposed for connector

LED signals for

230V

24V 50/60 Hz

grey RAL 7031

transparent

L 305 mmH 70 mmB 75 mm

6 thermostats

Max. 14 thermo-electric heads (cod. 110430)

state of actuators state of supply state of fuse

The control units are equipped with spring terminals which allow wiring without using screw ter-minals and wires with sections of 1,0 to 1,5 mm2 which have to be peeled back by 10 mm.

52

Control unit via wire for the regulation of 6 zones, can be expanded with other modules(cod. 110610).

Page 52: Pardo Seal A

Each wire to be connected to the terminal unitmust be correctly stripped.

Divide the control unit into three main parts:base, unit and cover.

Press the wires leading from the thermostatsinside the guides present on the base of thecontrol unit.

Press the supply wires inside the guide pre-sent on the base of the control unit.

Replace the unit on to the base. Insert the stripped part of the thermostatwires inside the control unit connectorsmaking sure that the colours correspond.

Insert the stripped part of the supply wireinside the control unit connector with the helpof a screwdriver.

Press the wires from the thermoelectric headsinside the guides on the control unit.

Insert the stripped part of the wire of the ther-moelectric heads inside the control unit con-nector with the help of a screwdriver.

Verify that all the connections have beenmade. For each zone it is possible to connectup to 4 thermoelectric heads.

If necessary use a screwdriver to open theterminals and remove the wires from the con-trol unit to correct any wrong connections.

Replace the lid on the control unit.

53

Page 53: Pardo Seal A

3.6.2. Expansion module for two zonesCod. 110610

Supply voltage

Operating voltage

Base colour

Protection colour

Dimensions

Predisposed for connector

LED signals for

24V 50/60 Hz (through the same control unit).

24V 50/60 Hz

grey RAL 7031

transparent

L 93 mmH 70 mmB 75 mm

2 thermostats

Max. 8 thermo-electric heads (cod. 110430)

state of the actuatorsstate of the supplystate of the fuse

Module which allows the number of thermo-electric heads connected to existing zones to beincreased (cod. 110600 and cod. 110605).

3.6.3. Thermo-electric heads Cod. 110430

Supply voltage

Operating voltage

Supply wire

Length of supply wire

Characteristics

24V 50/60 Hz (through the same control unit).

24V 50/60 Hz

2 x 0,5 mm2

1000 mm

The pin is self-set.

Test function when first turned on.

Visual indication of open/closed state.

“Click-on” mounting.

System of protection for loss from valves.

Thermo-electric heads NC type for manifolds.

3.6.4. Circulator pump control module Cod. 110620

Module which allows the circulators to be controlled.With the module it is possible to interrupt the supply of a circuit in relation to the signal of athermostat or when the whole system is closed. Allows regulation of a start-up delay of 5, 10or 15 min of the pump.

Supply voltage

Operating voltage

Base colour

Protection colour

Dimensions

24V 50/60 Hz (through the same control unit)

24V 50/60 Hz

grey RAL 7031

transparent

L 93 mmH 70 mmB 75 mm

54

Page 54: Pardo Seal A

VVaallssiirr uunnddeerrfflloooorr hheeaattiinngg ssyysstteemmss

4.1 Valsir residential system with pocketed panel• 16x2 MIXAL pipe with internal layer in crosslinked polyethylene PEX and intermediate layer in

butt-welded aluminium. Total barrier to oxygen and gases in general. Produced in compliance withthe strictest international standards. 120 m and 240 m rolls.

• V-ESSE pocketed insulating panel made of expanded polystyrene with blue EPS film. Pipe spacingminimum 75 mm, thickness of 20 and 30 mm and total heights of 50 to 60 mm respectively. Paneldensity of 30 kg/m3 with compressive strength of 150 kPa. The panel is produced in compliancewith standard UNI EN ISO 13163.

• Mixing kit and 1”1/4 brass manifold with flow-check valves, valves with thermostat capacity andflow meters (0,6÷2,4 l/min). Three-velocity pump and three-way mixing valve with motor capa-city. Interception valves with thermometer (0÷60°C) and automatic air vent valves both on thesupply and return. Number of outlets: 4÷12.

• V-BAND edging strip, 7 mm thick and 200 mm high. The strip is self-adhesive on the entire sur-face and has a polyethylene film to avoid cement penetrating below the panels.

• V-FLUID fluidizing additive. When used in correct proportions reduces the quantity of paste waterand improves workability of the gush.

• Anti-shrinking grid in fusion welded steel. It is used to limit the formation and increase of crackingdue to the natural dimensional variation of the cement.

• V-FOIL anti-humidity film made of polyethylene with 0,2 mm thickness. Prevents any humidity pre-sent in the ground from spreading to the inside of the room. It has an adhesive strip to aid theattachment of one film to another.

Figure 4.1 Valsir residential system with pocketed panel

444

55

wallskirting board

V-BAND edging strip

floorMIXAL pipe 16x2

concrete

concreteanti-shrinkage grid if used

V-FOIL anti-humidity film if used

V-ESSE pocketed insulating panel

Page 55: Pardo Seal A

4.2 Valsir residential system with smooth panel

• 16x2 MIXAL pipe made with internal layer in crosslinked polyethylene PEX with an interme-diate layer in butt-welded aluminium. Total barrier to oxygen and gas in general. Producedin compliance with the strictest international standards. 120 m and 240 m rolls.

• V-ELLE smooth insulating panel made of expanded polystyrene with grey film and red squa-ring to facilitate installation. The 20 mm thick panel has a density of 30 kg/m3 with a com-pressive strength of 150 kPa, the 30 mm panel has a density of 40 kg/m3 and a compres-sive strength of 250 kPa. The panel is produced in compliance with the standard UNI ENISO 13163.

• Mixing kit and 1”1/4 brass manifold with flow-check valves, valves with thermostat capa-city and flow meters (0,6÷2,4 l/min). Three-speed pump and three-way mixing valve withmotor capacity. Interception valves with thermometer (0÷60°C) and automatic air vent val-ves both on the supply and return. Number of outlets: 4÷12.

• V-BAND edging strip 7 mm thick and 200 mm high. The strip is self-adhesive on the entiresurface and has a polyethylene film to avoid cement penetrating below the panels.

• Clips V-CLIP01 for anchoring the pipe to the panel.• V-FLUID fluidizing additive. When used in correct proportions, reduces the quantity of paste

water and improves workability of the gush.• Anti-shrinking grid in fusion welded steel. It is used to limit the formation and increase of

cracking due to the natural dimensional variation of the cement. • V-FOIL anti-humidity film made of polyethylene with 0,2 mm thickness. Prevents any humi-

dity present in the ground from spreading to the inside of the room. It has an adhesive stripto aid the attachment of one film to another.

Figure 4.2 Valsir residential system with smooth panel

56

wallskirting board

V-BAND edging strip

floorMIXAL pipe 16x2

concrete

concreteV-CLIP01 anchor clip

anti-shrinkage grid if used

V-FOIL anti-humidity film if used

V-ELLE smooth insulating panel

Page 56: Pardo Seal A

4.3 Valsir residential system for renovation

• 14x2 PEXAL pipe with internal and external layer in crosslinked polyethylene PEX withan intermediate layer in butt-welded aluminium. Total barrier to oxygen and gas in gene-ral. Produced in compliance with the strictest international standards. 100 m rolls.

• V-ELLE20/150 smooth insulating panel made of expanded polystyrene with grey filmand squaring to facilitate installation. The panel has a density of 30 kg/m3 and a com-pressive strength of 150 kPa. It is produced in compliance with the standard UNI EN ISO13163.

• Mixing kit and 1”1/4 brass manifold with flow-check valves, valves with thermostatcapacity and flow meters (0,6÷2,4 l/min). Three-speed pump and three-way mixingvalve with motor capacity. Interception valves with thermometer (0÷60°C) and automaticair vent valves both on the supply and return. Number of outlets: 4÷12.

• V-BAND edging strip 7 mm thick and 200 mm high. The strip is self-adhesive on the enti-re surface and has a polyethylene film to avoid cement penetrating below the panels.

• Clips V-CLIP01 for anchoring the pipe to the panel.• V-FLUID fluidizing additive. When used in correct proportions, reduces the quantity of

paste water and improves workability of the gush.• Anti-shrinking grid in fusion welded steel. It is used to limit the formation and increase

of cracking due to the natural dimensional variation of the cement. • V-FOIL anti-humidity film made of polyethylene with 0,2 mm thickness. Prevents any

humidity present in the ground from spreading to the inside of the room. It has an adhe-sive strip to aid the attachment of one film to another.

Figure 4.3. Valsir residential system for renovation

57

wall

floorPEXAL pipe 14x2

concrete

concreteV-CLIP01 anchor clip

V-FOIL anti-humidity film if used

V-ELLE 20/150 smooth insulating panel

skirting board

V-BAND edging strip

anti-shrinkage grid if used

Page 57: Pardo Seal A

4.4 Valsir industrial system with smooth panel

• 20x2 MIXAL pipe with internal layer in crosslinked polyethylene PEX and intermediate layer inbutt-welded aluminium. Total barrier to oxygen and gas in general. Produced in compliance withthe strictest international standards. 120 m and 240 m rolls.

• V-ELLE30/250 smooth insulating panel made of expanded polystyrene with grey film and squa-ring to facilitate installation. The panel has a density of 40 kg/m3 and a compressive strength of250 kPa. It is produced in compliance with the standard UNI EN ISO 13163.

• 1”1/4 brass manifold with 6 to 8 outlets.• V-BAND edging strip 7 mm thick and 200 mm high. The strip is self-adhesive on the entire surfa-

ce and has a polyethylene film to avoid cement penetrating below the panels. • Clips V-CLIP02 for anchoring the pipe to the support fusion welded grid with threading 3÷5 mm

or else clips V-CLIP01 for anchoring the pipe directly to the panel.• V-FLUID fluidizing additive. When used in correct proportions, reduces the quantity of paste water

and improves workability of the gush.• Anti-shrinking grid in fusion welded steel. It is used to limit the formation and increase of cracking

due to the natural dimensional variation of the cement. • V-FOIL anti-humidity film made of polyethylene with 0,2 mm thickness. Prevents any humidity pre-

sent in the ground from spreading to the inside of the room. It has an adhesive strip to aid theattachment of one film to another.

Figure 4.4. Valsir industrial system with smooth panel

58

wallV-BAND edging strip

flloor

MIXAL pipe 16x2

concrete

concrete

V-CLIP01 anchor clip

V-FOIL anti-humidity film if used

V-ELLE 30/250 smooth insulating panel

anti-shrinkage grid

Page 58: Pardo Seal A

4.5 Valsir industrial system with insulating gush cement

• 20x2 MIXAL pipe with internal layer in crosslinked polyethylene PEX and intermediate layer inbutt-welded aluminium. Total barrier to oxygen and gas in general. Produced in compliance withthe strictest international standards. 120 m and 240 m rolls.

• Support made of cement with special insulating additives. • 1”1/4 brass manifold with 6 to 8 outlets. • V-BAND edging strip 7 mm thick and 200 mm high. The strip is self-adhesive on the entire surfa-

ce and has a polyethylene film to avoid cement penetrating below the panels. • Clips V-CLIP02 for anchoring the pipe to the support fusion welded grid with threading 3÷5 mm.• V-FLUID fluidizing additive. When used in correct proportions, reduces the quantity of paste water

and improves workability of the gush.• Anti-shrinking grid in fusion welded steel. It is used to limit the formation and increase of cracking

due to the natural dimensional variation of the cement. • V-FOIL anti-humidity film made of polyethylene with 0,2 mm thickness. Prevents any humidity pre-

sent in the ground from spreading to the inside of the room. It has an adhesive strip to aid theattachment of one film to another.

Figure 4.5. Valsir industrial system with insulating gush cement

For snowmelt systems the 26x3 PEXAL pipe is available, which guarantees an elevated flownecessary in tough climatic conditions. In this case, the pipe can be installed either on insu-lating gush cement or on a V-ELLE30/250 panel and anchored with V-CLIP02 clips on afusion-welded support grid with threading 3÷5 mm.

59

wall

MIXAL pipe 20x2

insulating concrete

concrete

V-CLIP 02 anchor clip

V-FOIL anti-humidity film if used support grid

anti-shrinkage grid

V-BAND edging strip

Page 59: Pardo Seal A

DDiimmeennssiioonniinngg ooff fflloooorr hheeaattiinngg ssyysstteemmssiinn aaccccoorrddaannccee wwiitthh tthhee ssttaannddaarrdd UUNNII EENN 11226644

5.1. IntroductionFor the calculation of the thermal output and for the dimensioning of underfloor heating

systems in residential, office and other buildings, reference is made to the European Standard

UNI EN 1264, parts 1, 2, 3 and 4.

In this chapter, we will look at the mathematical instruments for calculating the thermal output,

the surface temperature and hot water flow, necessary for heating a room by means of a floor

radiant panel with PEXAL/MIXAL pipe.

In as far as possible, the symbols referred to in the reference standard will be used.

5.2. Dimensioning: theory

5.2.1. Floor stratification

The heat output of an underfloor heating system (ufh) is strongly connected to the structu-

re and composition of the floor; the type of covering (marble, cotto, ceramic, carpet, etc)

to be used must be known, as well as the thickness of the various layers and their thermal

characteristics, and the type of insulating panel. The heating loops are placed on the insu-

lating panel and buried in the concrete screed upon which the covering is laid. The room

may be positioned directly over the ground or it may be part of a multi-floor building; this

must be considered when calculating the heat resistance of the floor.

555

60

Page 60: Pardo Seal A

Figure 5.1 Stratification of a heated floor over another room

61

Page 61: Pardo Seal A

Figure 5.2 Stratification of a heated floor on the ground

62

Page 62: Pardo Seal A

• Sa,i is the thickness of each layer i which makes up the floor covering [m], the layer of screed overthe pipes is identified by Su,

• Sb,i is the thickness of each layer i which makes up the floor below the pipe [m],• λa,i and λb,i are the coefficients of heat conductivity of each layer i which makes up the floor both

above and below the pipe [W/mK],• αa a is the coefficient of convection heat exchange of the air [W/m2K] in the upper part of the

radiant panel and takes on a value equal to 10,8 W/m2K,• αb is the coefficient of convection heat exchange of the air [W/m2K] in the room below, in the

case in which the room is not in direct contact with the ground, equal to 5,9 W/m2K,• R λ,B is the heat resistance of the floor covering [m2K/W],• R λ,isol is the heat resistance of the insulating panel [m2K/W],• Ro is the total heat resistance of the floor above the pipe [m2K/W],• Ru is the total heat resistance of the floor below the pipe [m2K/W].

The heat resistance of the floor covering is given by the following equation:

[5.1]

In the appendix, the heat resistance values are given for the most widely used coverings, as

well as the most common construction materials. Some comments are also made on the

use of wood (parquet) as a floor covering.

The total heat resistance above the pipe is:

where Su indicates the thickness of the screed above the pipes and λE indicates its heat conductance.When the room is directly on the ground, the total heat resistance of the floor below the pipe is:

[5.2]

[5.3]

63

Page 63: Pardo Seal A

In the following table, these limits are indicated as well as the minimum insulation thickness required(polystyrene with typical heat conductance of 0.034 W/mK).

5.2.2 Required thermal flow Before dimensioning a heating system the heat flow (or specific heat output) required in heating thesingle rooms must be determined. The calculation must be carried out in compliance with the natio-nal standards and laws (references in the appendix) and it must take into account the structure andbuilding elements as well as the climatic area in which the building is situated. The heat flow delivered by the floor surface [W/m2] is given by the following ratio:

where Q is the required output to heat the room and AF is the surface of the floor covered byloops.

if there is a room lying underneath, the coefficient of convection heat exchange of the air is alsoused:

[5.4]

The standard EN 1264-4 establishes the minimum thermal resistance that the insulating

layers shall have depending on the thermal conditions under the floor heating structure.

Table 5.1. Minimum thermal resistance of insulating layers

[5.5]

Thermal resistance Minimum Condition

of insulation thickness of

R λλ,isol [m2K/W] insulation [mm]

0,75 22 Constantly heated room below

1,25 37 Unheated or intermittent heated room below.

1,25 37 Room directly on the ground.

1,25 37 External air temperature below Tu≥0°C.

1,50 44 External air temperature below -5°C≤Tu<0°C.

2,00 59 External air temperature below -15°C≤Tu<-5°C

64

Page 64: Pardo Seal A

In the following pages, the logarithmically determined average differential temperature ∆THwill be used, given by:

[5.6]

[5.7]

by means of the following coefficient of proportionality:

[5.8]

where the coefficients aB, aT, m T, aU, m U, aD, m D and aC depend on the structure of the heatingfloor. In particular: • aB is a coefficient depending on the thermal conductivity of the screed λE [W/mK] and the heat

resistance of the floor covering R λ,B [m2K/W].• aT is a coefficient depending on the floor covering R λ,B,• m T is dependent on pipe spacing p [cm] of the heating loop,• aU is dependent on pipe spacing p and floor covering R λ,B,• m U is dependent on concrete thickness Su [cm] above the pipes,• aD is dependent on pipe spacing p and floor covering R λ,B,• m D is dependent on the outside diameter De [mm] of the pipe,• aC is dependent on pipe spacing p, the outside diameter De and wall thickness s of the pipe as

well as the conductance coefficient λT [W/mK]. Its value is indicated in the following table.

where Tm is the temperature of the supply water in the circuit [°C], Tr is the temperature of the returnwater [°C] and Ti is the temperature of the room [°C].As specified in UNI EN 1264-2, the thermal flow is proportional to the logarithmically determinedaverage differential temperature:

65

Page 65: Pardo Seal A

The values of the coefficients aB, aT , aU, aD, are expressed in the attached diagrams at the end ofthis chapter.

5.2.3. Characteristic curves A diagram, showing the heat flow on the vertical axis, and the temperature difference on the hori-zontal axis, can represent the relationship between the average differential surface temperature andthe heat flow. The resulting curve is defined the characteristic curve of the floor installation; it is astraight line and the gradient is given by the coefficient 1/Rq. Once the floor structure has been esta-blished, it is possible to draw several characteristic curves depending on the pipe spacing. When thepipe spacing is increased, the curve gradient diminishes as can be seen in the graph below.

Table 5.2. Coefficient aC for the calculation of the specific heat output

Spacing

P PEXAL 14x2 MIXAL 16x2 MIXAL 20x2

[cm]

5 -87,044 -127,829 -145,262

7,5 -58,030 -85,219 -96,841

10 -43,522 -63,914 -72,631

15 -29,015 -42,610 -48,421

20 -21,761 -31,957 -36,315

22,5 -19,343 -28,406 -32,280

25 -17,409 -25,566 -29,052

30 -14,507 -21,305 -24,210

35 -12,435 -18,261 -20,752

37,5 -11,606 -17,044 -19,368

Coefficient aC

Figure 5.3. Characteristic curves

66

m T m U m D

Pipe

Page 66: Pardo Seal A

5.2.4. Thermal flow limit and maximum floor temperatureThe standard UNI EN 1264-2 establishes a physiological limit to the maximum floor temperaturedepending on the type of room:

Table 5.3. Maximum floor temperature

Tf,max [°C] Type of room/area

29 Occupied area

33 Bathroom and similar

35 Peripheral area

[5.9]

[5.10]

Only pipe spacing which respect the following ratio can be accepted:

[5.11]

If no spacing corresponds to this ratio, then the floor heating system alone is not sufficient

to heat the room and it must be supplemented with another source of heat, for example,

one or more radiators.

where BG and nG are two coefficients depending on screed thickness Su [cm] above the pipes andconductivity λE [W/mK] but above all, pipe spacing p [cm].Whereas the coefficient iis dependent on the room temperature and the maximum floor tem-perature:

The heat flow limit can be obtained with the radiant floor once all of its characteristics have beenestablished (spacing, screed thickness, type of covering, stratification). The heat flow limit is calcula-ted from the following equation:

67

Page 67: Pardo Seal A

5.2.5. Limit curves The union of the limit values qG forms the so-called limit curve of heat flow. This curve, drawn on thecharacteristic curves, indicates the pipe spacings, which are not acceptable, or, in other words, thespacings, which would result in a floor surface temperature above the physiological limits imposedby the standard.

5.2.6. Supply temperature The supply temperature can be determined by following the indications in the reference standard EN1264-3. For the calculation of the supply temperature it is possible to proceed in the following wayand as represented in the figure.a) Consider the room with the highest heat flow requirement qa,max (bathrooms excluded).b) The characteristic curves depending on the pipe spacing p available, are determined, assuming

a floor covering thermal resistance equal to R λ,B = 0,1 m2K/W. In the case of higher values, theseshall be used.

c) The limit curve is determined.d) The pipe spacing corresponding to the characteristic curves with a heat flow limit qG less than

required qa,max must be abandoned in the calculation of the supply temperature. e) Consider the characteristic curve corresponding to the minimum spacing. Calculate the minimum

average differential temperature ∆TH,min corresponding to the value qa,max of this characteristiccurve.

f) Consider the characteristic curve corresponding to the maximum spacing allowed. Calculate themaximum average differential temperature ∆TH,max corresponding to the value qa,max of the cha-racteristic curve.

Figure 5.4. Characteristic and limit curves.

68

Page 68: Pardo Seal A

g) The minimum supply temperature Tm,min is given by ∆TH,min considering a zero temperature dropbetween supply and return Tm – Tr= 0°C. The equation [ 5.6] becomes:

h) The maximum supply temperature Tm,max is given by ∆TH,max considering a maximum tempera-ture drop between supply and return Tm – Tr= 5°C. In this case the equation [ 5.6] becomes:

i) The supply temperature must be chosen between the minimum value Tm,min and the maximumvalue Tm,max.

Ulterior criteria based on the economical aspects of system management or investment can beapplied in order to further reduce the field of acceptable values of the supply temperature.

Figure 5.5. Calculation of the supply temperature.

[5.12]

[5.13]

69

Page 69: Pardo Seal A

5.2.7. Average floor temperatureThe average floor temperature is calculated in accordance with the equation indicated in the stan-dard UNI EN 1264-2:

where Tiis the environment temperature and qa is the specific thermal output.

5.2.8. Downward heat dispersionThe heating loops deliver heat to the cement screed, which, in turn, delivers upward thermalflow thus heating the room, but also downward heat which contributes to heating the roombelow or else loses heat to the outside (in the case of a room directly over the ground or offa terrace). The downward heat flow is calculated according to the following equation:

where R o is the thermal resistance of the floor above the pipe, R u is the thermal resistan-ce of the floor below the pipe, Tu is the temperature of the room below or the ground underthe floor, Ti is the temperature of the room to be heated and qa is the required thermalflow.

5.2.9. Length of heating loops The length of each loop, the chosen pipe spacing, and the total area to be heated determine the num-ber of heating loops.The length of the loops is limited by the length of the available rolls and above all by the maximumpermissible pressure loss. In the following table, it is possible to determine the maximum covered sur-face for a given length of loop and a pre-set pipe spacing.

[5.14]

[5.15]

70

Page 70: Pardo Seal A

Table 5.4. Maximum area covered by a single circuit

Heating area Aj [m2]

Pipe spacing p [cm]

[5.16]

The heating area AF [m2] is given by the following equation:

and as a consequence the loop length [m] can be calculated as follows:

[5.17]

where p is the pipe spacing [cm].

Loop length

L 5 7,5 10 15 20 22,5 25 30 35 37,5

[m]

25 1,25 1,875 2,5 3,75 5 5,65 6,25 7,5 8,75 9,4

50 2,5 3,75 5 7,5 10 11,3 12,5 15 17,5 18,8

100 5 7,5 10 15 20 22,5 25 30 35 37,5

150 7,5 11,3 15 22,5 30 33,8 37,5 45 52,5 56,3

200 10 15 20 30 40 45 50 60 70 75

71

Page 71: Pardo Seal A

5.2.10. Flow and temperature of heating fluid The required flow for heating the room [kg/s] is given by:

[5.18]

whereAF is the surface occupied by the loops [m2],qa and qb are the upward and downward specific heat outputs [W/m2],∆T = Tm - Tr is the temperature drop of the heating fluid and therefore Tm and Tr are respectivelythe supply and return temperature of the heating fluid [°C].When the average fluid temperature is known Tmed the temperature drop has been established ∆Tthe supply temperature is calculated:

[5.19]

and the supply temperature of the fluid:

[5.20]

The velocity of the flow [m/s] is calculated in accordance with following formula:

[5.21]

72

Page 72: Pardo Seal A

where ρ is the density of the water [kg/m3] at the average temperature Tmed and Di is the internaldiameter of the pipe [mm]. The speed can be obtained directly from the diagrams of the pressureloss, from which the pressure drop ∆ρ is also obtained.

5.2.11. Design limits in the choice of pipe spacing

The speed limits are linked to noise in the case of excessively fast fluids and to air bubbles in the caseof excessively slow fluids. The values of minimum and maximum speed for each pipe diameter are indicated in the followingtable.

The pressure loss limit is obviously linked to the capacity of the pump to circulate the flow inside theloops. The limit imposed is:

and it must include the loss in the flow-check valve on the supply manifold and in the valve with ther-mostatic capacity on the return manifold.

The maximum temperature for each circuit is

to avoid flows with excessive thermal ranges which would negatively influence the performance ofthe installation.

In choosing the pipe spacing, the limits to be imposed on the speed, pressure loss and tem-perature drop of each circuit, must be taken into consideration.

Pipe vmin [m/s] vmax [m/s] [l/s] [l/s]

PEXAL 14x2 0,05 0,45 0,0040 0,0354

MIXAL 16x2 0,05 0,45 0,0057 0,0509

MIXAL 20x2 0,05 0,45 0,0101 0,0905

Table 5.5. Velocity, flow and pressure loss limits

m.min

m.max

73

Page 73: Pardo Seal A

5.2.12. Balancing of heating circuitsBalancing of the circuits consists of determining the regulation point of the flow-check valves of eachheating loop. The procedure is as follows: a) Locate the circuit with the greatest pressure loss, here it is indicated as the k circuit. b) Calculate the total pressure loss of the k circuit:

where ∆pk, flow-check valve is the loss of pressure in the flow-check valve in the k circuit, conside-red completely open, given by the characteristic curve of the flow-check valve (Figure 3.4.3)depending on the flow in the circuit, ∆pk,valve is the pressure loss in the valve, given by thevalve characteristic curve (Figure 3.4.4) depending on the flow in the circuit, ∆pk,circuit, isthe pressure loss in the pipe loop of circuit k. c) Calculate the pressure loss for the other circuits at the flow-check valve ∆pi, flow-check valve, keepingin mind that the total pressure loss for each circuit is the same as the total pressure loss of the circuitwith the greatest pressure loss:

d) By using the flow-check valve characteristic curve, determine the number of regulations correspon-ding to the pressure loss calculated.

m.k

m.k

74

[5.22]

[5.23]i,flow-check valve k,total i,valve i,circuit

k,flow-check valvek,total k,valve k,circuit

Page 74: Pardo Seal A

5.3. Dimensioning: practiceThe underfloor heating system of the building indicated in Figure 5.6 and having the following cha-racteristics, must be sized:

The project is on the first floor, over constantly heated rooms, at a temperature of Tu=20°C. Incompliance with the standard requirements, the V-ESSE 30 pocketed insulating panel is cho-sen, with a 30 mm base and a total height of 60 mm; this panel allows a spacing of 7,5 cm,15 cm, 22,5 cm, 30 cm and 37,5 cm. The screed thickness above the pipes is 45 mm. TheMIXAL 16x2 pipe is chosen for the heating loops. The other characteristics of the floor can beseen in Figure 5.7 for the bathroom and in Figure 5.8 for the other rooms. The floor covering is 15 mm cotto tiles, in all of the rooms, except for the bathroom, wherethere 10 mm ceramic tiles have been chosen.

Table 5.6. Design details

N. Room Room Total Heating Required thermaltemperature Ti area area AF output Q

1 Bedroom 20°C 11,4 m2 11,4 m2 696 W

2 Living room 20°C 16,1 m2 16,1 m2 1320 W

3 Kitchen 20°C 10,4 m2 10,4 m2 832 W

4 Bathroom* 22°C 7,2 m2 5,4 m2 576 W

5 Storeroom 18°C 2,4 m2 2,4 m2 120 W

* The heating area of the bathroom must be reduced due to the presence of the bath.

75

Page 75: Pardo Seal A

Figure 5.6. Example

76

Room no.1

Bedroom Area 11,4 m2

Room no.3

KitchenArea 10,4 m2

Room no.4Bathroom

Area 7,2 m2 (5,4 m2)

Room no.2

Living roomArea 16,1 m2

Room no.5

Store roomArea 2,4 m2

Page 76: Pardo Seal A

Figure 5.7. Floor layers of the bathroom

Screed

MIXAL 16 x 2

V-ESSE 30 insulation

concrete subfloor

predalles deck 250 mm

plaster

Figure 5.8. Floor layers of the other rooms

77

Ceramic covering 10 mm

Screed

MIXAL 16 x 2

V-ESSE 30 insulation

concrete subfloor

predalles deck 250 mm

plaster

cotto tile covering 15 mm

Page 77: Pardo Seal A

5.3.1. Floor stratification The figures illustrating the floor stratification enable us to calculate the thermal resistance characte-ristics of all the rooms in the building.The upward thermal resistance of the rooms, excluding the bathroom, is:

whereas, for the bathroom, it is:

The downward thermal resistance is for all the rooms (including the bathroom):

Table 5.7. Floor thermal resistance of the rooms in the project

In the following table, the thermal resistance values of the rooms are summarised:

1 Bedroom 0,017 0,1447 1,4939

2 Living room 0,017 0,1447 1,4939

3 Kitchen 0,017 0,1447 1,4939

4 Bathroom 0,010 0,1377 1,4939

5 Storeroom 0,017 0,1447 1,4939

N. RoomsThermal resistance

of covering R λλ,B [m2K/W]

Upward thermalresistance

RO [m2K/W]

Thermal resistance of covering RU [m2K/W]

78

Page 78: Pardo Seal A

5.3.2. The required thermal flowIn order to proceed with the choice of spacing for each room, it is first necessary to calculate notonly the required thermal flow qa, but also the coefficient Rq given by the equation [ 5.8] and theaverage temperature differential TH for each available spacing. The coefficients aB, aT , aU , aD ,used in the calculation of Rq are determined respectively by the diagrams in Figure 5.12, Figure5.13, Figure 5.14 and Figure 5.15. The resulting values as shown in the table will be used in thefollowing paragraphs in the calculation of the supply temperatures and the choice of spacing.

The storeroom is not included in the calculation as it already contains the supply/return manifold ofthe heating loops and the same pipes will supply the required thermal flow, which is, in this case,very low (qa=50 W/m2). It is possible to verify this statement by calculating the heat flow deliveredby the pipes, which pass through the storeroom.

Table 5.8. Characteristic coefficients and average differential temperature for each roomand for each available spacing

m T m U m D

N. Room qa Spacing p aB aT aU aD aC Rq ∆TH[W/m2] [cm] [m2K/W] [°C]

1 Bedroom 85 7,5 0,97 1,00 1,00 0,980 -85,21 0,155 13,20

15 0,97 0,82 1,00 0,964 -42,61 0,191 16,25

22,5 0,97 0,68 1,00 0,955 -28,40 0,232 19,69

30 0,97 0,56 1,00 0,950 -21,30 0,281 23,89

37,5 0,97 0,46 1,00 0,948 -17,04 0,346 29,41

2 Living 82 7,5 0,97 1,00 1,00 0,980 -85,21 0,155 12,73

room 15 0,97 0,82 1,00 0,964 -42,61 0,191 15,67

22,5 0,97 0,68 1,00 0,955 -28,40 0,232 19,00

30 0,97 0,56 1,00 0,950 -21,30 0,281 23,04

37,5 0,97 0,46 1,00 0,948 -17,04 0,346 28,37

3 Kitchen 80 7,5 0,97 1,00 1,00 0,980 -85,21 0,155 12,42

15 0,97 0,82 1,00 0,964 -42,61 0,191 15,29

22,5 0,97 0,68 1,00 0,955 -28,40 0,232 18,53

30 0,97 0,56 1,00 0,950 -21,30 0,281 22,48

37,5 0,97 0,46 1,00 0,948 -17,04 0,346 27,68

4 Bathroom 107 7,5 1,01 1,00 1,00 0,980 -85,21 0,149 15,95

15 1,01 0,82 1,00 0,963 -42,61 0,184 19,65

22,5 1,01 0,67 1,00 0,955 -28,40 0,226 24,15

30 1,01 0,55 1,00 0,950 -21,30 0,276 29,51

37,5 1,01 0,45 1,00 0,948 -17,04 0,338 36,13

m T m U m D

79

Page 79: Pardo Seal A

The selected pipe spacing must be excluded in that they generate surface temperatures above thephysiological limits imposed by the standard EN 1264.

Table 5.9. Thermal flow limit for every room in the project

80

N. Room qa Spacing p Rq ϕ BG nG qG qa≤qG

[W/m2] [cm] [m2K/W]

1 Bedroom 85 7,5 0,155 1,0 94,5 0,013 97,90 Yes

15 0,191 1,0 74,5 0,077 92,97 Yes

22,5 0,232 1,0 54,5 0,148 84,68 No

30 0,282 1,0 35,1 0,247 74,45 No

37,5 0,344 1,0 17,4 0,411 60,65 No

2 Living 82 7,5 0,155 1,0 94,5 0,013 97,90 Yes

room 15 0,191 1,0 74,5 0,077 92,97 Yes

22,5 0,232 1,0 54,5 0,148 84,68 Yes

30 0,282 1,0 35,1 0,247 74,45 No

37,5 0,344 1,0 17,4 0,411 60,65 No

3 Kitchen 80 7,5 0,155 1,0 94,5 0,013 97,90 Yes

15 0,191 1,0 74,5 0,077 92,97 Yes

22,5 0,232 1,0 54,5 0,148 84,68 Yes

30 0,282 1,0 35,1 0,247 74,45 No

37,5 0,344 1,0 17,4 0,411 60,65 No

4 Bathroom 107 7,5 0,149 1,247 94,5 0,013 122,02 Yes

15 0,184 1,247 74,5 0,077 115,58 Yes

22,5 0,226 1,247 54,5 0,148 105,12 No

30 0,276 1,247 35,1 0,247 92,19 No

37,5 0,338 1,247 17,4 0,411 74,70 No

5.3.3. Thermal flow limit and maximum floor temperatureOnly pipe spacing that generates a flow limit qG above the flow required qa will be accepted. Inthe following table the flow limits and available spacing are indicated for each room. Also, thecoefficients ϕ, BG and nG required for calculating the flow limits are indicated and which are deter-mined by using the equation [ 5.10] and the diagrams of Figure 5.16 and Figure 5.17. For sim-plicity sake, the coefficients Rq, are indicated once more, as they are used in calculating qG , as wellas the comparison with the required thermal flow which permits evaluation of the applicability ofeach pipe spacing.

Page 80: Pardo Seal A

5.3.4. The characteristic and limit curves When the coefficient Rq is known for each available pipe spacing, the characteristic curves for thefloor heating system can be traced. If, on the same graph, the points corresponding to the flow limitqG are intersected, the limit curve is obtained. All the rooms with the same floor stratification havethe same curves; therefore, the characteristic curves of the bedroom, living room and kitchen are tra-ced in Figure 5.9, whereas the bathroom is shown in Figure 5.10.

Figure 5.9. Characteristic curves of the bedroom, living room and kitchen

Figure 5.10. Characteristic curve of the bathroom

81

Page 81: Pardo Seal A

5.3.5. Supply temperature To calculate the supply temperature, the room with the highest thermal flow, excluding the bathroom,must be considered. In this case, the bedroom has the highest thermal flow which is equal toqa,max=85 W/m2.As required by the standard, all the coefficients are calculated for tracing the characteristic curvesand the corresponding limit curve of the room, considering a floor covering with a thermal resistan-ce equal to R λ,B=0,1 m2K/W (from the project information, there is no other floor covering with ahigher thermal resistance).

Table 5.10. Characteristic coefficients and average differential temperature for the calcula-tion of the supply temperature

Room qa,max Spacing p aB aT aU aD aC Rq ∆TH[W/m2] [cm] [m2 K/W] [°C]

7,5 0,61 1,0 1,0 0,984 -85,21 0,247 20,99

15 0,61 0,86 1,0 0,972 -42,61 0,289 24,58

22,5 0,61 0,75 1,0 0,963 -28,40 0,334 28,35

30 0,61 0,65 1,0 0,958 -21,30 0,386 32,80

37,5 0,61 0,56 1,0 0,956 -17,04 0,448 38,10

Fictitious with structure as in project with Rλ,B=0,1 m2K/W

85

Table 5.11. Thermal flow limit for calculation of the supply temperature

Room qa,max Spacing p Rq ϕ BG nG qG qa≤qG

[W/m2] [cm] [m2 K/W]

7,5 0,247 1,0 94,5 0,013 98,50 Yes

22,5 0,334 1,0 54,5 0,148 90,22 Yes

30 0,386 1,0 35,1 0,247 82,53 No

37,5 0,448 1,0 17,4 0,411 72,92 No

Fictitious with structure as in project with Rλ,B=0,1 m2K/W

85

82

m T m U m D

Page 82: Pardo Seal A

In Figure 5.11 the minimum and maximum values of the average differential temperatures ∆TH,minand ∆TH,max are shown. These values are used in equations [ 5.12] and [ 5.13], to calculate theminimum and maximum values of the supply temperature:

The supply temperature Tm will have to be chosen from the acceptable range of 41°C to 50°C. In theexample dealt with, consider a supply temperature of Tm=46°C, an average value in the acceptablerange.

Figure 5.11. Calculation of the supply temperature

83

Page 83: Pardo Seal A

5.3.6. Circuit dimensioningIt is now possible to proceed with the calculation of all the other parameters: temperature drop,return temperature, downward thermal flow, loop length, flow, velocity and pressure loss, some ofwhich are fundamental for the definitive dimensioning of the circuits and, therefore, for the choice ofthe pipe spacing for each room. The temperature drop ∆T is calculated by using the diagram in Figure 5.18 when the room tempe-rature Ti, is known, the supply temperature Tm and the logarithmically determined average differen-tial temperature ∆TH. By using the equation the calculation of the return temperature is immediate[5.20]. The floor temperature Tf is calculated with the diagram in Figure 5.19 once the room tem-perature Tiand the required flow qa is known.The downward thermal flow qb is determined by equation [ 5.15].The length L of the heating loops is calculated by the equation [ 5.17].The flow is calculated with the diagram in Figure 5.20 depending on the total heat output andthe temperature drop ∆T.The velocity v and the pressure loss ∆p can be evaluated by using the diagrams of pressure losses.In the example, the layout pattern is counter flow spiral and the pressure losses are increased by 13%(see chp. 3.1.13).

The results calculated for each room are shown in the following table.

The final choice of spacing must be made excluding the spacing values that do not respect the limitsindicated in paragraph 5.2.11. In the above example, all the values are within the limits. If theMIXAL pipe 16 x 2 is to be used in 100 m rolls, reduced pipe spacing values must be chosen whichallow circuits under 100 m. For the bathroom, a spacing of 15 cm is chosen, so that the flow velo-city is not too near the minimum permissible value.

m.

Table 5.12. Results

84

N. Room qa Spacing p ∆T Tr Tf qb L v ∆p

[W/m2] [cm] [°C] [°C] [°C] [W/m2] [m] [l/s] [m/s] [mbar]

1 Bedroom 85 7,5 21 25 27,7 8,23 152 0,0121 0,11 52

15 16,5 29,5 27,7 8,23 76 0,0154 0,14 38

2 Living room 82 7,5 21 24 27,5 7,94 215 0,0165 0,15 122

15 17,5 28,5 27,5 7,94 107 0,0198 0,18 82

22,5 12,5 33,5 27,5 7,94 72 0,0277 0,24 97

3 Kitchen 80 7,5 20,5 25,5 27,3 7,70 139 0,0106 0,093 38

15 18 28 27,3 7,70 70 0,0121 0,11 25

22,5 13,5 32,5 27,3 7,70 47 0,0162 0,14 26

4 Bathroom 107 7,5 17 29 29,6 9,86 72 0,0089 0,08 15

15 11 35 29,6 9,86 36 0,0137 0,12 15

m.

Page 84: Pardo Seal A

5.3.7. Balancing of heating circuitsTo determine the regulation points of the flow-check valves of each loop, it is necessary, first of all,to consider the least favourable circuit, therefore, the one with the highest pressure loss, in this case,the living room.For the flow value indicated in the table = 0,0277 l/s = 99,72 l/h is determined:∆p2,flow-check valve=4,3 mbar from the characteristic curves of the flow-check valve in correspondence withthe maximum opening (t.a.),∆p2,valve=4,3 mbar from the characteristic curve of the valve with thermostatic capacity,∆p2,circuit is the pressure loss indicated in Table 5.13.The total pressure loss in the heating circuit of room 2 is, therefore:

From the characteristic curve of the valve with thermostatic capacity for circuit 1, we have p1,valve=1,3 mbar and

From the characteristic curve of the valve with thermostatic capacity for circuit 3, we have p3,valve=1,3 mbar and

From the characteristic curve of the valve with thermostatic capacity for circuit 4, we have p4,valve=1,3 mbar and

Table 5.13. Choice of final spacing

N. Room Spacing p Tr Q a+Qb L v ∆p

[cm] [°C] [W] [cm] [l/s] [m/s] [mbar]

1 Bedroom 15 27,7 1063 76 0,0154 0,14 38

2 Living room 22,5 27,5 1448 72 0,0277 0,24 97

3 Kitchen 15 27,3 912 70 0,0121 0,11 25

4 Bathroom 15 29,6 631 36 0,0137 0,12 15

4054 254 0,0689

m.

m.2

85

2,total 2,valve 2,circuit2,flow-check valve

2,total 1,valve 1,circuit1,flow-check valve

2,total 3,valve 3,circuit3,flow-check valve

2,total 4,valve 4,circuit4,flow-check valve

Page 85: Pardo Seal A

In correspondence with these losses, the regulations of the flow-check valves are obtained, as shownin the table:

For the calculation of kv the typical equation can be used:

with ∆p expressed in mbar and in l/h.

Table 5.14. Regulation points of the circuit flow-check valves

N. Room ∆pi,flow-check valve Kv Number

[l/s] [mbar] of turns

1 Bedroom 0,0154 70,3 0,21 0,25÷0,5

2 Living room 0,0277 4,3 1,52 t.o.

3 Kitchen 0,0121 82,8 0,15 0,25÷0,5

4 Bathroom 0,0137 92,6 0,16 0,25÷0,5

m.

m.

86

Page 86: Pardo Seal A

Figure 5.12.

87

Page 87: Pardo Seal A

88

Figure 5.13.

Page 88: Pardo Seal A

Figure 5.14.

89

Page 89: Pardo Seal A

90

Figure 5.15.

Page 90: Pardo Seal A

Figure 5.16.

91

Page 91: Pardo Seal A

92

Figure 5.17.

Page 92: Pardo Seal A

Figure 5.18.

93

Page 93: Pardo Seal A

94

Figure 5.19.

Page 94: Pardo Seal A

Figure 5.20.

95

Page 95: Pardo Seal A

SSnnooww mmeellttiinngg wwiitthh rraaddiiaannttppaanneell ssyysstteemmss

6.1 Introduction

Radiant panel heating systems can be used for eliminating snow and ice in outdoorareas such as roads, bridges, viaducts, runways and helipads, sidewalks, car parks, ath-letic fields, garage entrance ramps, hospitals, hotels, warehouse loading docks, etc. The problem of snow and ice is usually dealt with by scattering salt or other anti freezesubstances. As well as being unreliable, these methods cause pollution to ground waterand, in fact, some regional laws forbid their use. Heating systems with radiant panels have proven to be the most efficient and reliable forthis type of application. Furthermore, the Valsir Pexal multi layer pipe is the best choice for this type of installa-tion in that the internal aluminium layer renders the pipe more resistant than other tra-ditional pipes in plastic and at the same time, its elasticity makes it easier to install thanmetal pipes. Accurate sizing of radiant panel heating systems in closed areas is fundamental in orderto optimise heat output and avoid energy waste. Certain factors relating to climatic conditions and the type of installation must be consi-dered.

Climatic factors• Rate of snowfall: this factor is extremely important in designing a system; it is the

speed with which snow falls on the surface to be heated and it is measured in cm/h. • Air temperature: important in determining the thermal energy necessary for the remo-

val of both snow and ice. • Wind velocity: wind removes heat from the heating surface, thus obstructing snow melt

and facilitating the formation of ice. • Relative humidity: the higher the humidity the more thermal energy will be required for

melting the snow.

System factors • Surface temperature: in order to avoid ice formation, the temperature of the heating

surface must not be lower than 1°C. Higher temperatures would lead to energy waste.The system must be capable of delivering sufficient thermal output even in varyingexternal climatic conditions.

• Heating fluid: generally an antifreeze solution is utilized in these systems which ismade up of a mixture of water and ethylene or propylene glycol, in adequate propor-tions, depending on the minimum external temperature (see Table in the appendix).

666

96

Page 96: Pardo Seal A

• Water drainage: a suitable drainage system for the melted snow must be studied; anystagnant water left on the surface will cause a reduction in heat output.

• Thermal insulation: under the heating coils it is advisable to lay a layer of insulation, which redu-ces direct heat dispersion (heat loss). The type of insulation is closely related to the expected sur-face load. Extruded polystyrene insulating panels may be used for light loads (max. load 250 kPa)or insulating screed (generally concrete with insulating spheres) for high surface loads caused byheavy vehicles.

• Heated screed: the pipe may be buried in the concrete or directly in the ground. Asphalt pave-ments, however, are not advisable for two reasons. Asphalt is usually placed at 150°C and thistemperature may damage the pipes. Also, the compaction process (by means of rollers) maydeform or break the pipes.

6.2. System types

The heating circuits are embedded in the concrete in different ways, depending on the type of areato be heated. There are essentially three types.

The pipes are embedded directly in the layer of concrete, which also makes up the pavement.This is generally the case with entrance/exit ramps. The pipe must be embedded in the con-crete at a depth of 12 to 15 cm. On the surface a layer of cement or rubber is laid. The fol-lowing figure shows an example.

Figure 6.1

97

concrete or rubber

Page 97: Pardo Seal A

The pipes are embedded in the concrete and the pavement is laid on top. In this case the thicknessof the concrete, which contains the pipes can be reduced. The thickness can vary from 8 to 12 cm.The surface layer can be made of different materials, such as asphalt, earth, etc. This type of appli-cation is often used for roads and parking lots.

The pipes are immersed in a layer of sand or gravel, over which the pavement is laid. The bed ofsand or gravel in which the pipes are laid must have a thickness of 12 to 20 cm. The surface layercan be made of concrete or earth, as with athletic fields.

Figure 6.2

Figure 6.3

98

Page 98: Pardo Seal A

6.3. System design

This type of system utilises a mix of water and antifreeze solution, which cannot be heateddirectly by the boiler. A heat exchanger with stainless steel plates must be used between thesystem and the boiler. The design supply temperature is reached by using a three-way valvethat mixes the flow and return water. The three-way valve can have an on/off or modular servomotor. In the first case, a humidstatand a thermostat command the servomotor, which actuates the three-way valve in the presen-ce of atmospheric conditions that would create ice on the surface (humidity and temperature). In the second case, a command box connected to a sensor which measures the temperature ofthe ground under which the circuits are installed, commands the servomotor which acts on thethree-way valve in such a way that the heating circuits maintain the surface temperature above0°C.

Figure 6.4 Humidstat and thermostat command the servomotor

Figure 6.5 The surface temperature sensor commands the servomotor

99

Page 99: Pardo Seal A

6.4. Dimensioning: theory

6.4.1 Required heat outputThe sizing of a snow/ice melting heating system calls for the preliminary calculation of the specificheat output required. The following method used has been suggested by ASHRAE (American Societyof Heating, Refrigerating and Air-Conditioning Engineers, Inc.).The specific heat output required by the heating floor is:

where:

qs is the sensible heat transferred to the snow [W/m2],

qm is the heat of fusion [W/m2],

qe is the heat of evaporation [W/m2],

qh is the heat transfer by convection and radiation [W/m2],

ar is the ratio of snow-free area to total area A to be heated.

The coefficient ar takes on different values depending on the type of area to be heated.

The equations for calculating the various contributing factors to required heat output are:

Table 6.1

ar Area to be heated.

0 Roads and private walkways, private ramps.

0,5 Commercial areas, office and shop walkways, high traffic roads.

1 Squares, parking lots, entrance ramps to parking lots, runways and helipads, athletic fields.

[6.1]

[6.4]

100

Page 100: Pardo Seal A

where

sn is the rate of snowfall (cm/h),

Tea is the air temperature (°C),

vw is the wind velocity (m/s),

pav is the vapour pressure of moist air (Pa) which can be calculated with the relative humidity U.R.

(%), by means of the following equation:

6.4.2. Stratification of radiant panelsThe heating loops are embedded in the screed, which can be made up of different layers.The following shows the generic structure of a radiant floor for snow/ice melting.

• sa,i is the thickness of each layer i which makes up the screed above the centre line of the pipe [m],• sb,i is the thickness of each layer i which makes up the screed below the centre line of the pipe [m],• λa,i and λb,i are the coefficients of heat conduction of each layer number i which makes up the

screed, both below and above the centre line of the pipe [W/mK],• αa is the coefficient of convective heat exchange of the air [W/m2K] in the upper part of the radiant

panel valued at the temperature of the external air Tea which is generally equal to 20 W/m2K,• αb is the coefficient of convective heat exchange of the air [W/m2K] in the lower part of the radiant

[6.6]

Figure 6.6. Generic structure of radiant panel

101

Page 101: Pardo Seal A

panel valued at the temperature of the external air Teb if the panel is elevated (viaducts, bridges,etc.) it is equal to 16 W/m2K,

• Ka and Kb are the total transmittance coefficients of the screed above and below the centre line ofthe pipe [W/m2K],

• p is the spacing of the loops [cm].

The coefficient of total transmittance in the upper part of the radiant panel is given by:

The coefficient of total transmittance in the lower part of the radiant panel, if laid directly on theground, is:

if elevated (viaducts, bridges, etc.), is:

[6.7]

[6.8]

[6.9]

102

Page 102: Pardo Seal A

6.4.3. Temperature calculation With the specific heat output qa necessary to melt the snow or de-ice the surface, the mean tempe-rature of the heating fluid can be determined:

and the temperature of the heated surface:

6.4.4. Downward specific heat output The downward specific heat output [W/m2] which does not contribute to heating the surface, is cal-culated by means of the following equation:

where Teb is the temperature of the air under the panel in the case of elevated systems (bridges, via-ducts, etc.), whereas in heating systems laid on the ground, it is the temperature of the ground.

[6.10]

[6.11]

[6.12]

103

Page 103: Pardo Seal A

6.4.5. Calculation of circuit loops The number of heating loops is determined by their length, by the spacing and the total area to beheated. The length of the loops is limited by the length of the rolls available and above all by the maximumpressure loss allowable. In the following table, it is possible to determine the maximum surface cove-red for a given loop length and a pre-established installation spacing.

The heating area AJ is given by the following equation:

whereas the total area to be heated is given by the sum of the single heating areas:

and, therefore, supposing that the loops are all of the same length, the number of circuits is:

[6.13]

Table 6.2. Maximum area covered by a single circuit

Heating area AJ [m2]

L [m] 5 7,5 10 15 20 22,5 25 30 35 37,5 40

25 1,25 1,875 2,5 3,75 5 5,65 6,25 7,5 8,75 9,4 10

50 2,5 3,75 5 7,5 10 11,3 12,5 15 17,5 18,8 20

100 5 7,5 10 15 20 22,5 25 30 35 37,5 40

150 7,5 11,3 15 22,5 30 33,8 37,5 45 52,5 56,3 60

200 10 15 20 30 40 45 50 60 70 75 80

Installation spacing p (cm)

[6.14]

[6.15]

104

.

.

Loop length

Page 104: Pardo Seal A

6.4.6. Calculation of flow rate and temperature of the heating fluidThe flow rate necessary for melting the snow/ice on the surface is given by the following equation:

whereAJ is the surface to be heated by the loop number j[m2],qa and qb are the downward and upward specific heat outputs [W/m2],∆T = Tm - Tr is the temperature drop and therefore Tm and Trare, respectively, the supply and returntemperatures of the heating fluid [°C].

In the coldest part of the radiant panel, where the temperature of the fluid is Tr, the surface tempe-rature must not be below 1°C to avoid the formation of ice. It is necessary to make sure that the return temperature Tr is not below a minimum value determinedby the following equation:

With the average fluid temperature Tmed and having established the return temperature Trabove theminimum value allowed, it is possible to determine the temperature drop between the supply andreturn temperature:

and the fluid supply temperature:

It is also necessary to verify the maximum speed and pressure loss. By using the pressure lossdiagrams, depending on the pipe diameter and the recently calculated flow, it is possible todetermine speed and pressure loss. In the table the speed limits are indicated for each pipediameter.

[6.16]

[6.17]

[6.18]

[6.19]

.

105

Page 105: Pardo Seal A

6.5. Dimensioning: practice The use of the calculation diagrams is explained by means of a simple example of radiant panelsizing for snow melting. An entrance ramp, situated in the Brescia area, must be made with a length of 25 m and a width of4.2 m. The pipes are embedded in a layer of concrete having a 12 cm thickness from the centre lineof the pipes. There is a 4 cm thick layer of concrete under the pipes and a 2 cm thick layer of expan-ded polystyrol. Relative humidity is U.R.=70% and snowfall intensity is 0.2 cm/h. A minimum temperature of -7°Cand a wind speed of 1,5 m/s is indicated for the area in question.

6.5.1. Required heat output The required specific heat output qa is given by the equation [ 6.1].The diagrams in Figure 6.8 and Figure 6.9 allow us to calculate the sum qs + qm in relation to thesnowfall intensity sn and the outdoor temperature Tea for a snowfall intensity between 0 and 1 cm/hand between 1 and 2.5 cm/h.The diagram in Figure 6.10 allows us to calculate the sum qe + qh in relation to the relative humi-dity U.R., the outdoor temperature Tea and wind speed vw.

For the type of area to be heated, ar= 1 the required specific heat output for melting the snow/iceon the entrance ramp is easily calculated:

6.5.2. Stratification of radiant panelsThe radiant panel in the example is completely made of concrete and lies on a layer of insu-lation. From the tables in the appendix, the coefficients of heat conductivity of the two mate-rials are calculated:By calculating the total thickness from the pipe centre

the total heat resistance

Pipe Vmin [m/s] Vmax [m/s] [l/s] [l/s]

MIXAL 20x2 0,05 1,00 0,0101 0,201

PEXAL 26x3 0,05 1,00 0,0157 0,314

Table 6.3. Speed and pressure limits

m.min

m.max

106

,

Page 106: Pardo Seal A

and supposing there is a spacing p = 30 cm, from the diagram in Figure 6.11 the coefficient of totalupward transmittance is calculated:

Ka = 5,5 W/m2K

Then, by calculating the total downward thickness from the pipe centre

the total heat resistance

and supposing a spacing p = 30 cm, from the diagram in Figure 6.12 the coefficient of total down-ward transmittance for a radiant panel laid on the ground is determined:

Kb = 0,9 W/m2K.

6.5.3. Temperature calculationKnowing Ka and the required upward heat output qa the diagram in Figure 6.13 is used to calcula-te the difference between the average temperature of the fluid and the atmospheric temperature andthe difference between the surface temperature of the panel and the atmospheric temperature:

Tmed – Tea = 54°C from which the fluid supply temperature is calculated Tmed = 54 – 7 = 47°C,Tsa – Tea = 14,5°C from which the panel surface temperature is calculated Tsa = 14,5 – 7 = 7,5°C.

6.5.4. Downward specific heat output In order to determine the heat loss underneath the panel, it is necessary to use a hypothetical tem-perature for the ground; it is legitimate to consider a ground temperature under the panel equal tothe atmospheric temperature. From the diagram in Figure 6.14, in relation to the value Kb andthe difference between the supply temperature and that of the ground Tmed – Teb = 54°C, theresult is qb = 48 W/m2.

6.5.5. Calculation of circuit loopsConsidering circuits of 25 m, a spacing of p = 30 cm, Table 6.2 shows us that the area covered byeach loop is Aj= 7,5 m2 and therefore the number of loops is n = 105/7,5 = 14.

6.5.6. Calculation of flow rate and temperature of heating fluid Before calculating the flow rate of the fluid, the minimum return temperature necessary to prevent theformation of ice on the surface must be determined. From the diagram in Figure 6.15 the return temperature limit Tr,min = 22°C is calculated.The diagrams in Figure 6.16 and Figure 6.17 indicate the flow rate of the fluid for values of totalheat output requested both below 5 kW and above 5 kW once the temperature difference of the fluidhas been calculated.The maximum temperature drop for preventing ice formation is:

,

107

Page 107: Pardo Seal A

therefore, a difference of ∆T=15°C must be considered to avoid any risk and to obtain a relativelylow supply temperature. The actual return temperature is therefore given by:

With this flow rate and a 20 mm diameter PEXAL pipe with a 2 mm wall thickness, there is a fluidspeed of about 0,21 m/s and a pressure loss in each circuit of 0,016 bar.Figure 6.7 indicates a general outline of the heated ramp in which the spacing and the distributionof the 14 circuits are indicated.

and the supply temperature is therefore:

The total heat output of each loop is:

which, in Figure 6.16 corresponds to a flow rate of:

Figure 6.7 General outline of heated entrance ramp.

108

Page 108: Pardo Seal A

Figure 6.8.

109

Page 109: Pardo Seal A

110

Figure 6.9.

Page 110: Pardo Seal A

Figure 6.10.

111

Page 111: Pardo Seal A

112

Figure 6.11.

Page 112: Pardo Seal A

Figure 6.12.

113

Page 113: Pardo Seal A

114

Figure 6.13.

Page 114: Pardo Seal A

Figure 6.14.

115

Page 115: Pardo Seal A

116

Figure 6.15.

Page 116: Pardo Seal A

Figure 6.16.

117

Page 117: Pardo Seal A

118

Figure 6.17.

Page 118: Pardo Seal A

IInnssttaallllaattiioonn

7.1 Preliminary operations and controls Before installing the heating system, it is necessary to proceed with some preliminary controls:• Verify that the floor size allows the installation of the scheduled floor heating system. • Verify whether it is necessary to install the anti-humidity barrier; to be used in rooms pla-

ced in direct contact with the ground.• Control that the surface of the supporting base is flat, does not sink and is free of masonry

debris.• At this stage, the cabinet for the distribution manifold and the mixing kit should already

have been installed and encased.

7.2. Installation of the manifold and mixing kit• The cabinet must be installed in such a way that the distribution manifold is higher than the

pipe level in order to guarantee the venting of the system. • The position of the manifold and mixing kit will respect the design indications and will the-

refore be in a central position relative to the rooms to be heated, to avoid that the supplypipes to the circuits are too long.

7.3. Installation of the edging strip• The edging strip must be placed vertically in correspondence with the subfloor and must run

along all of the walls, pillars or stairs.• For the laying of the V-BAND edging strip all of the adhesive surface or just the bottom part

can be used by removing both strips or half of the protective strip. If the screed has a rela-tively small thickness, it would be better not to use the entire adhesive surface in order toavoid leaving traces of glue on the plaster. On the contrary, if the screed is to be quite high(industrial system) the entire surface shall be used.

7.4. Installation of the insulating panels • The V-ESSE insulating panels must be installed in such a way that the successive panel sur-

mounts the joint of the previous panel (Figure 7.1).• Laying starts at the longest wall or at any rate, along the wall at the far side of the entran-

ce to the room.• During laying of the panels along the walls, the polyethylene film of the edging strip must

be raised and placed on top of the panels themselves. This will stop cement from infiltratingbetween the panels and the walls (Figure 7.2).

• The panels can be cut with a rigid blade and the remaining panel, if of suitable dimensions,can be re-used.

• To install the V-ELLE insulating panels, it is sufficient to unroll the panels and place them nextto each other. An adhesive strip is used to attach them together so that cement does not getin between them.

119

777

Page 119: Pardo Seal A

Figure 7.1. Laying order of the V-ESSE panel

Figure 7.2. Placement of the polyethylene film of the V-BAND strip over the insulating panels

120

Page 120: Pardo Seal A

7.5. Installation of piping • Before installing the pipe, it is necessary to verify on the project which room to start with. The cir-

cuits shall be installed consecutively to avoid that the supply routes from the manifold to the hea-

ting loops of the rooms, do not cross over each other.

• Always use the pipe unwinder; without this tool the roll may be unwinded incorrectly, causing the

pipe to become tangled and twisted and thus preventing a linear installation.

• Unwind the roll starting at the outside coil.

• The laying of the pipe must start at the supply manifold and, it the pipe pattern is counter flow,

work should start at the perimeter of the room working inward. In this case, the distance between

the pipes must be double as compared to the design spacing, to leave space for the return route

to the manifold. Therefore, with a pipe spacing of 22,5 cm, the pipe must be laid starting from the

perimeter of the room toward the centre, leaving a distance of 45 cm between the pipes (Figure

7.3).

• When the centre of the room has been reached, invert the installation direction by bending the

pipe but at the same time respecting the minimum radius allowed (Figure 7.4).

• For pipe bending the minimum radii indicated must be observed.

• Connect the supply and the return of the circuit to the manifold.

• Remember to record the initial and final length of pipe used for each circuit, using the marking on

the pipe, in order to verify the total length of pipe used.

Figure 7.3. Laying of the pipe from the perimeter to the centre of the room

121

Page 121: Pardo Seal A

7.6. Expansion joints • The expansion joints must be made in order to compensate for the dimensional

expansion/contraction of the heated cement screed, caused by variations in the tem-perature.

• The joints shall cross the entire layer of cement and must be made when the surface isgreater than 40 m2 or when the length of at least one of the walls is greater than 8 m.

• In correspondence with these expansion joints, the pipes must be protected with a corruga-ted tube of 30÷40 cm (Figure 7.5).

• The arrangement of the joint line must not cross the entire heating circuit but only the supplyroutes (Figure 7.6).

• The visible cut on the surface must be filled with a special sealant.

Figure 7.4. Laying of the pipe from the centre to the perimeter of the room

122

Page 122: Pardo Seal A

Figure 7.5. Floor section in correspondence with the expansion joint

Figure 7.6. Arrangement of the expansion line

123

Corrugated tube Expansion joint

Page 123: Pardo Seal A

7.7. Settlement joints • Settlement joints must be made between one room and another and above all, when two different

floor coverings are used. • These joints are generally placed in door reveals and are made by cutting the cement screed by

approximately 1/3 of its thickness starting from the covering (Figure 7.7).• The visible cut on the surface must be filled with a special sealant.

7.8. Filling • Close the interception valves both at the supply and return of the manifold.• Close all the flow-check valves at the supply manifold.• Connect the supply pipe of the test pump by removing the insert of the supply manifold drainage

valve.• Open the drainage valve on the return manifold, connecting a pipe to the insert for the transport

of the water to the drain.• Open the flow-check valve of the first circuit and circulate the water until all the air contained in

the circuit has been expelled.• Close the circuit flow-check valve and repeat the operation for the other circuits.• Once the filling operation of the circuits has finished close the drainage valve on the return mani-

fold.

Figure 7.7. Floor section in correspondence with a settlement joint

124

Settlement joints

Page 124: Pardo Seal A

7.9. Leak test • Open all the flow-check valves • The test pressure is to be twice the working pressure with a minimum of 6 bar. • Verify the absence of leaks by controlling the entire circuit pressure on the pressure gauge.• The pressure shall be applied to the pipes during laying of the screed.• When there is a danger of freezing, it is necessary to add anti-freeze liquid to the water. If no fur-

ther frost protection is necessary for the normal operation of the system, the liquid shall be remo-ved and the system shall be flushed using at least 3 changes of water.

7.10. Laying of the screed• The cement must be prepared without the use of insulating agents. • During preparation of the mixture, the fluidising additive V-FLUID shall be used; the percentages

indicated on the product shall be respected. The fluidiser must be kept out of direct sunlight andthe room temperature must not fall below 0°C. Verify also the expiry date on the product.

• The fluidiser reduces water content necessary and improves workability. • The cement must be poured with the system working at a pressure of 6 bar and the tempe-

rature of the room must not fall below 5°C.• If fusion welded grids are to be used, these shall be laid in observance with the expected quotas

(1/3 of the height of the screed in relation to the covering).

7.11. Heating up• This operation shall be carried out at least 21 days after the laying of the cement screed or in

accordance with the manufacturer’s instructions but at least 7 days in the case of anhydrite screedsor, at any rate, also in this case, in accordance with the manufacturer’s instructions.

• The initial heating up commences at a supply temperature between 20°C and 25°C, which shallbe maintained for at least 3 days. Subsequently the maximal design temperature shall be set andmaintained for at least another 4 days.

• At the end of the test, the floor covering can be laid.

125

Page 125: Pardo Seal A

AA.. HHeeaatt ttrraannssffeerr Heat transfer concerns all physical processes in which a certain quantity of heat energy is transfer-red from one system to another due to a difference in temperature. Such processes occur accordingto the principles of thermodynamics:

• The heat energy given off by a system has to be the same as the energy received by the other.• Heat passes from a warmer to a colder body.

The speed with which the heat transfer process occurs is also very important and therefore the quan-tity of heat exchanged within the unit of time or the thermal output Q which is measured in Watts.

A.1. Modes of heat transfer The transfer of heat is a spontaneous phenomenon, which continues to occur between a hot and acold object, until both objects have reached the same temperature, the temperature of thermal balan-ce. The hot object transfers part of its thermal energy to the cold object and the heat transfer can beby conduction, convection or radiation.

A.1.1. Conduction Conduction occurs between solid materials when placed in direct contact with each other. It is thetemperature difference between the two objects that causes the hot object to transfer energy to thecold object, thus increasing the temperature until a thermal balance has been reached (same tempe-rature). In Figure A.1 it is observed how two objects at different temperatures are placed togetherand by conduction the heat flows from the hot object to the cold object until the balance temperatu-re is reached.

Figure A.1. Heat transfer by conduction

126

.

Page 126: Pardo Seal A

A.1.2. ConvectionConvection occurs when one of the substances is a fluid (water, air, etc.) and the heat trans-fer may be associated with a transfer of matter. When the temperature of a fluid is not uni-form, that is, it contains hot and cold areas, the difference in temperature and the speed ofthe fluid itself, generate a continuous movement of the particles in the fluid, causing them tobe mixed, thus favouring the transfer of heat from the hot to the cold particles. This pheno-menon is called natural convection. When, on the other hand, the movement of the particlesis caused mechanically, by a pump in the case of water circulation, or simply by wind, thephenomenon is called forced convection. Convection occurs, for example, when an interme-diate fluid circulates between two bodies (thermal fluid), which heats up when in contact withthe hot matter, and then transfers its heat when it comes into contact with the cold matter(Figure A.2).

A.1.3. Radiation Radiative heat transfer involves the emission and consequent absorption of electromagnetic rays. Theheat exchanged in this case rapidly increases with a difference in temperature. Radiation differs fromthe other modes of heat transfer, in that it does not require the presence of a means in order to trans-fer energy. Electromagnetic radiation that generates the transmission of heat is linked to the energe-tic state of the atoms of which it is made up. In this case, the hot matter emits electromagnetic raysthat are absorbed by the cold matter (Figure A.3).

Figure A.2. Transfer of heat by convection

127

Page 127: Pardo Seal A

A.2. Combined heat transfer processes Depending on the nature of the material, with heat transfer, one mode will be more predomi-nant than the others, or the heat will be transferred thanks to the combined action of two orall three modes. This is due to the physical characteristics of the materials, such as density,transparency, etc. The heat which is lost through the walls of a room to the outside of a buil-ding, is transferred by conduction across the various layers which make up the wall and byconvection and radiation through the air spaces between the bricks. From the external surfa-ce of the room, the heat is released into the surrounding outside air by means of convectionand radiation. Figure A.4 shows a room where the floor and ceiling are at different temperatures (TA>TB). Inthis case, the heat is transferred from the floor to the ceiling by convection and radiation. Ifthere was no air to act as a medium, then there would only be the phenomenon of radiationand less heat would be exchanged between the two structures. After a certain time, the totalheat output is given by the sum of the heat transferred by convection and radiation:

Figure A.3. Heat transfer by radiation

[A.1]

128

TO TAL Convection Radiation

Page 128: Pardo Seal A

The two transmission mechanisms act in a parallel way. There is, in fact, an analogy with electricalcircuits, which allows an analysis of the phenomenon of combined heat transfer (Figure A.4).

The flow of current which circulates in the two resistances R1 and R2, which represent respectively the

resistances of convection and radiation, behaves in an analogous way to the quantity of heat which

flows for a certain period between the two bodies caused by convection and radiation. The total flow

of current, in fact, is given by the sum of the currents, which circulate in the two resistances, just as

the total thermal output is given by the sum of the exchanged thermal output of the single transfer

mechanisms [ A.1].

On the other hand, if we consider a wall made of two layers of different materials, with the two exter-

nal parallel surfaces at different temperatures, as shown in Figure A.5, the heat is transferred by con-

duction only and it flows from wall A to wall B.

Figure A.4. Heat transfer between floor and ceiling

129

Page 129: Pardo Seal A

During its passage it meets different materials and the heat transfer intensity is different accordingto the type of material. We can consider the three layers as three resistances placed in a series(Figure A.5) for which the following relation is valid:

where, are the thermal outputs transferred by conduc-tion respectively through the cement, the insulating layer and the bricks. As noted in theequation [A.2] the total thermal output transferred is less than the output that would flowsingly in each material. The walls of houses are made of several layers in order to minimisethe heat loss though them.

Figure A.5. Heat transfer inside a wall

[A.2]

130

TO TAL Conduction1 Conduction2 Conduction3

Conduction1 Conduction2 Conduction3

Page 130: Pardo Seal A

A.3. Heat transfer in heating systems

A.3.1. Radiator systems In this type of system, the transfer of heat from the radiator to the room occurs both by radia-tion and by convection. The convection component, however, is greater than the radiative one.Hot air movement is therefore created at the ceiling and cold air at the floor as shown in figu-re A.6. This air current that is created transports the dust present in the room, which can crea-te breathing problems; furthermore, the combustion of such dust particles generates darkstains on the walls behind the radiator.

A.3.2. Fan heater systemIn this type of system, the radiative component is even lower than in radiator systems. The convecti-ve component is further accentuated by the fact that an internal fan inside the heat convector itselfincreases hot air movement.

Figure A.6. Radiator system

131

Page 131: Pardo Seal A

A.3.3. Floor heating systemsThe transfer of heat from the screed heated by the pipe occurs due a combined effect of radiationand convection. In this case, however, the convective component is negligible compared to the radia-tive component. The effects of dust movements, as seen in the other “traditional” heating systems, arenot created. Furthermore, layers of heat are generated between the floor and ceiling, that is, the tem-perature is higher at the floor and diminishes in proximity to the ceiling. This stratification is close tothe ideal conditions of thermal well-being.

Figure A.7. Floor system

132

Page 132: Pardo Seal A

133

Table B.1. Heat conductance

Material Heat Conductance λ [W/mK]

Asphalt 0.70

Reinforced concrete 1.51

Ordinary concrete 1.28

Chalk 0.43

Dry gravel in layers 0.93

Plaster with lime mortar 0.70

Plaster with lime and chalk 0.93

Linoleum 0.18

Cement mortar 1.40

Expanded polystyrol 0.035

Expanded polyurethane 0.028

Dry sand 0.58

Sand and cement 0.93

Cork leaves 0.04

Expanded vermiculite 0.07

BB.. HHeeaatt ccoonndduuccttaannccee aanndd rreessiissttaannccee ooff mmaatteerriiaallss..

Table B.2. Heat resistance of flooring materials

Materials Heat resistance R λλ,,ΒΒ [m2K/W]

Ceramic tiles 6 mm 0,0060

Ceramic tiles 8 mm 0,0080

Ceramic tiles 10 mm 0,0100

Ceramic tiles 12 mm 0,0120

Cotto tiles 10 mm 0,0111

Cotto tiles 15 mm 0,0167

Cotto tiles 20 mm 0,0222

Cotto tiles 30 mm 0,0333

Rubber 2 mm 0,0071

Rubber 3 mm 0,0107

Rubber 4 mm 0,0143

Rubber 5 mm 0,0179

Page 133: Pardo Seal A

Table B.2. Heat resistance of flooring materials

Materials Heat resistance R λλ,,ΒΒ [m2K/W]

Marble 10 mm 0,0029

Marble 15 mm 0,0044

Marble 20 mm 0,0059

Marble 30 mm 0,0088

Parquet 6 mm 0,0300

Parquet 8 mm 0,0400

Parquet 10 mm 0,0500

Parquet 12 mm 0,0600

Parquet 14 mm 0,0700

Parquet 16 mm 0,0800

Parquet 18 mm 0,0900

Parquet 20 mm 0,1000

Moquette 10 mm 0,1100

Moquette 12 mm 0,1300

Moquette 14 mm 0,1600

Moquette 16 mm 0,1800

Moquette 18 mm 0,2000

Moquette 20 mm 0,2200

Table B.3. Heat resistance of decks

Type of deck Heat resistance Rb [m2K/W]

Tiled deck 200 mm 0,320

Tiled deck 240 mm 0,350

Tiled deck 280 mm 0,370

Predalles deck 150 mm 0,360

Predalles deck 200 mm 0,400

Predalles deck 250 mm 0,430

134

Page 134: Pardo Seal A

135

CC.. WWoooodd aass aa FFlloooorr CCoovveerriinngg

Particular attention must be given to wooden floor coverings.The factors influencing the characteristics of the coverings are temperature for the glue andhumidity for the wood. The temperature of the screed does not influence the characteristics of the wood, however,it can be an important factor for the glue used. In fact, special glues must be used that donot depend on the temperature.The factor that can damage a wood covering is humidity. Variations in the concentrationof humidity generate “movements” in the wood itself; all types of wood are more or lessprone to such “movements”. Clearly the more constant the level of humidity, the less riskthere is of damage being caused to the flooring.The formation of cracks is due to low levels of humidity, bending, on the other hand, is cau-sed by levels of humidity that are too high. When the temperature increases, the humidity level generally falls; therefore, an overlyheated room will present a greater risk of cracking to the wood. When, however, the tem-perature goes down, the cracks will close again. In climates where winters are drier than summers, this phenomenon will be present whe-ther underfloor heating has been installed or not, and above all, if there is no control overhumidity in the surroundings. Both cracking and bending of the wood will occur, if it is not allowed to “move” and toadapt itself to the environment prior to installation. Various types of wood may be used for floor coverings; the choice should be made bet-ween woods that offer a greater dimensional stability than others.

The ground rules to be followed for the correct design of a floor heating system are: 1) low temperatures of the floor surface,2) the heat flow must be as uniform as possible on the surface and therefore spacing must

be kept relatively low,3) the cement screed must be completely dry before the wood flooring is laid, 4) the wood should be allowed to acclimatize to the conditions which will be generated in

the room where it is to be installed.

Page 135: Pardo Seal A

DD.. DDiimmeennssiioonniinngg ooff mmeettaall ggrriidd rreeiinnffoorrcceemmeenntt iinn tthhee fflloooorr

The metal grids placed in industrial floors have the function of limiting the opening of cracks in theconcrete screed due to shrinkage, in the proximity of contraction joints and are not intended toincrease the load capacity of the screed. By preventing the crack from increasing, the metal gridallows maintenance of a good load distribution on the concrete screed.

Dimensioning of the grid can be done by means of a mathematical equation defined as “slipequation”:

where:L is the distance between two successive contraction joints [m],f is the coefficient of static friction between floor and subfloor, ρ is the specific weight of the concrete including any permanent loads [kg/m3],su is the thickness of the concrete screed [cm],σs is the allowable traction stress of the steel [MPa],Aa is the grid area by linear meter of floor [cm2/m] or rather the ratio between the section of grid

and the dimension of the mesh as indicated:

136

Figure D.1. Floor with contraction joint

[D.1]

[D.2]

Page 136: Pardo Seal A

where da is the diameter of the thread [cm] and pa is the dimension of the mesh [m] (see Figure D.2).

Figure D.2. Characteristic dimensions of net for grid

Table D.1. Area of framework

In the following table, the area of several types of frameworks is given.

In the following table, the f values in relation to the type of subfloor are given.

Table D.2. Coefficient of static friction between floor and subfloor

Type of Subfloor Static friction f

PE film 0,80

Fine sand 0,90

Ganular mix 1,65

Gravel and sand 1,80

Clay 2,00

137

Net da [cm] pa [m] Area Aa [cm2/m]

Ø2/5 0,2 0,05 0,63

Ø3/10 0,3 0,1 0,71

Ø5/20 0,5 0,2 0,98

Ø5/30 0,5 0,3 0,65

Ø5/35 0,5 0,35 0,56

Ø6/20 0,6 0,2 1,41

Ø6/30 0,6 0,3 0,94

Ø6/35 0,6 0,35 0,81

Ø8/30 0,8 0,3 1,68

Ø8/35 0,8 0,35 1,44

metal grid

Page 137: Pardo Seal A

138

The principle on which the slip equation is based, is that the power of friction which is created bet-ween the floor and the subfloor is in proportion to the weight of the floor itself and the friction coef-ficient of the subfloor. The forces at the interface between the subfloor and the floor generate trac-tion stress in the metal grids due to the friction between the two surfaces.

In order to limit the degree of cracking in proximity of the contraction joints, the metal grid will haveto be laid as near as possible to the upper surface of the floor. In doing so, however, there is a riskthat the grid is cut during installation of the contraction joints.

D.1. Example of dimensioning of a metal grid Imagine an industrial floor made of 15 cm of concrete (specific weight 2300 kg/m3) and with a dis-tance between the contraction joints of 4,5 m. The net is made of steel with an allowable load trac-tion of 210 MPa. Considering the recommended friction coefficient of f=2,5, the grid area is calcu-lated:

According to Table D.1 it is possible to choose between a grid of Ø5/20 (0,98 cm2/m) or else agrid of Ø 6/30 (0,94 cm2/m) to be positioned at 10 cm from the subfloor (5 cm from the floor).

It is recommended to use static friction coefficient values f equal to 2,5 considering in this

way the irregularities of the ground on which the concrete floor is to be laid and which

significantly influence the effects of friction.

Figure D.3. Force of friction between the subfloor and the concrete floor

The correct compromise for the installation height of the grid is 1/3 of the total height of

the screed starting from the surface of the floor as shown in Figure D.3.

metal grid

concrete

ground

Page 138: Pardo Seal A

EE.. AAnnttii--ffrreeeezzee lliiqquuiidd iinn hheeaattiinngg cciirrccuuiittss

In floor heating systems for snow melting or for heating industrial pavements, or whenever there isa risk that the water inside the circuit will freeze, it is convenient to add an anti-freeze liquid in sui-table proportions to the expected minimum temperature. In the case of systems left unused during the winter season, it is recommended to leave any circuitsection valves open so that the entire circuit can absorb the variations in volume of the water, in rela-tion to the temperature.

In the following tables, the typical concentrations (volume percentage) of the most commonly usedanti-freeze liquids are indicated: ethylene or propylene glycol.

The concentrations may vary from one product to the other.The aim of these values

is to give a general idea of the quantities of anti-freeze required in relation to outdoor

temperatures.

Temperature Volume concentration

-4°C 10%

-9°C 20%

-17°C 30%

-26°C 40%

-37°C 50%

Table E.1. Concentrations of anti-freeze with ethylene glycol base

Temperature Volume concentration

-10°C 25%

-15°C 33%

-32°C 50%

Table E.2. Concentrations of anti-freeze with propylene glycol base

139

Page 139: Pardo Seal A

140

FF.. CCaallccuullaattiioonn ooff qquuaannttiittyy ooff ccoonnccrreettee ffoorr llaayyiinngg

The quantity of concrete to be used in a floor heating system depends on a number of factors someof which have very little influence.The height of concrete above the pipe and the panel bosses is the element which obviously has thegreatest influence, as does the panel type (V-ELLE or V-ESSE), other factors such as spacing and pipediameter are not as important. The quantity of concrete can be determined thanks to the following tables which give the volumenecessary for carrying out a heating system over a surface area of 100 m2.

In the following tables, the typical concentrations (volume percentage) of the most commonly usedanti-freeze liquids are indicated: ethylene or propylene glycol.

Table F.1. Quantity of concrete for pipe De=16 mm and V-ESSE panel

30

35

40

45

50

55

60

65

70

[mm]

Quantity of concrete in m3/100 m2 with external pipe diameter 16 mm and V-ESSE panel

Height of screedabove the bosses

4,8

5,3

5,8

6,3

6,8

7,3

7,8

8,3

8,8

7,5

Pipe spacing [cm]

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

15

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

22,5

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

30

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

37,5

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

45

Table F.2. Quantity of concrete for pipe De=20 mm and V-ESSE panel

30

35

40

45

50

55

60

65

70

[mm]

Quantity of concrete in m3/100 m2 with external pipe diameter 20 mm and V-ESSE panel

Height of screedabove the bosses

4,6

5,1

5,6

6,1

6,6

7,1

7,6

8,1

8,6

7,5

Pipe spacing [cm]

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

15

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

22,5

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

30

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

37,5

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

45

Page 140: Pardo Seal A

Table F.3. Quantity of concrete for pipe De=14 mm and V-ELLE panel

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

[mm]

Quantity of concrete in m3/100 m2 with external pipe diameter 14 mm and V-ELLE panel

Height of screedabove the pipe

4,1

4,6

5,1

5,6

6,1

6,6

7,1

7,6

8,1

8,6

9,1

9,6

10,1

10,6

11,1

5

4,2

4,7

5,2

5,7

6,2

6,7

7,2

7,7

8,2

8,7

9,2

9,7

10,2

10,7

11,2

10

4,3

4,8

5,3

5,8

6,3

6,8

7,3

7,8

8,3

8,8

9,3

9,8

10,3

10,8

11,3

15

4,3

4,8

5,3

5,8

6,3

6,8

7,3

7,8

8,3

8,8

9,3

9,8

10,3

10,8

11,3

20

4,3

4,8

5,3

5,8

6,3

6,8

7,3

7,8

8,3

8,8

9,3

9,8

10,3

10,8

11,3

25

4,3

4,8

5,3

5,8

6,3

6,8

7,3

7,8

8,3

8,8

9,3

9,8

10,3

10,8

11,3

30

4,4

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

9,4

9,9

10,4

10,9

11,4

35

4,4

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

9,4

9,9

10,4

10,9

11,4

40

Pipe spacing [cm]

Table F.4. Quantity of concrete for pipe De=16 mm and V-ELLE panel

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

[mm]

Quantity of concrete in m3/100 m2 with external pipe diameter 16 mm and V-ELLE panel

Height of screedabove the pipe

4,2

4,7

5,2

5,7

6,2

6,7

7,2

7,7

8,2

8,7

9,2

9,7

10,2

10,7

11,2

5

4,4

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

9,4

9,9

10,4

10,9

11,4

10

4,5

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

9,5

10,0

10,5

11,0

11,5

15

4,5

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

9,5

10,0

10,5

11,0

11,5

20

4,5

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

9,5

10,0

10,5

11,0

11,5

25

4,5

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

9,5

10,0

10,5

11,0

11,5

30

4,5

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

9,5

10,0

10,5

11,0

11,5

35

4,5

5,0

5,5

6,0

6,5

7,0

7,5

8,0

8,5

9,0

9,5

10,0

10,5

11,0

11,5

40

Pipe spacing [cm]

141

Page 141: Pardo Seal A

142

Tabella F.5. Quantity of concrete for pipe De=20 mm and V-ELLE panel

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

[mm]

Quantity of concrete in m3/100 m2 with external pipe diameter 20 mm and V-ELLE panel

Height of screedabove the pipe

4,4

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

9,4

9,9

10,4

10,9

11,4

5

4,7

5,2

5,7

6,2

6,7

7,2

7,7

8,2

8,7

9,2

9,7

10,2

10,7

11,2

11,7

10

4,8

5,3

5,8

6,3

6,8

7,3

7,8

8,3

8,8

9,3

9,8

10,3

10,8

11,3

11,8

15

4,8

5,3

5,8

6,3

6,8

7,3

7,8

8,3

8,8

9,3

9,8

10,3

10,8

11,3

11,8

20

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

9,4

9,9

10,4

10,9

11,4

11,9

25

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

9,4

9,9

10,4

10,9

11,4

11,9

30

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

9,4

9,9

10,4

10,9

11,4

11,9

35

4,9

5,4

5,9

6,4

6,9

7,4

7,9

8,4

8,9

9,4

9,9

10,4

10,9

11,4

11,9

40

Pipe spacing [cm]

Page 142: Pardo Seal A

GG.. IInnssuullaattiinngg ppaanneellss ffoorr fflloooorr hheeaattiinngg

Why install floor insulating panels? What influence do they have on system response time andefficiency, and therefore on energy saving? These are the questions we will try to answer inthis section of the manual.

G.1. Influence of insulating panels on system performanceG.1.1. Mechanical functionOne of the functions of the insulating panel in floor heating systems is to provide a support ontowhich the circuits are laid. The panels can be pocketed and contain slots for holding the pipe, orthey can be smooth, in which case the pipe is fixed by means of appropriate anchoring clips. Dueto the manner in which floor heating is conceived, insulating panels are therefore, a part of thesystem itself, and although their use is often neglected for economical reasons, they actually have avery important effect on the system's performance. It is, of course, possible to create radiant panelswithout using insulation, by using metal reinforcement grids on top of which the pipe is laid by meansof suitable clips; in this case, however, there would be no barrier to downward heat dispersion andthe performance of the system would consequently be compromised.

G.1.2. Reduction of thermal mass Thermal mass refers to the capacity of materials to store thermal energy (heat). A rock, for example,has a greater thermal mass than a piece of wood; it in fact retains heat for longer than a piece ofwood. This is due to the fact that the rock is denser and therefore contains a greater mass at equalvolume. When a heated thermal mass is subject to a stream of old air, for example, it starts to dis-sipate thermal energy (heat) at a speed proportionate to the difference in temperature between theair and the thermal mass. The principle of floor heating is to maintain constant and uniform the tem-perature of the thermal mass in such a way as to dissipate sufficient thermal energy to heat the room.The presence of insulating panels inside the floor structure reduces the thermal mass. The insulatingpanels, by separating the layer of concrete in which the heating pipes are buried from the load-bea-ring sub-floor, reduce the thermal mass to be heated and therefore create systems which respondquicker to the variation in room or outdoor temperatures. Let's evaluate a floor heating system in aresidential building of approximately 100 m2, the structure of the floor containing insulating panel isillustrated in Figure G.1.

143

Page 143: Pardo Seal A

144

In this case, the volume of the concrete in direct contact with the pipe, and which therefore influen-ces the thermal mass effect, is approximately 0.061 m3/m2 (quantity of concrete per square meterof surface). If the same system had been created without insulating panels (see Figure G.2), the volu-me of concrete in direct contact with the pipe would be 0.213 m3/m2 (given by the volume of screedabove the pipe and the volume of concrete). It is easy to verify that the relationship between the volu-me of concrete is from 1 to 3.5 and therefore there is a significant increase in the thermal massshould insulating panels not be used. Based on what has been expressed in the introduction, it fol-lows that the greater the thermal mass the longer it will take the system to respond to the variationsin temperature.

Figure G.1. Stratification of an insulated radiant floor

Figure G.2. Stratification of a radiant floor with insulation

Page 144: Pardo Seal A

G.1.3. Thermal insulation The energy produced by the coils is directed both upwards, bringing the necessary contribution tothe heating of the room, and downwards, generating a loss of energy towards the underlyingground.The flow of downward energy depends on the temperature of the ground and the room below andon the total thermal resistance of the layers of material below the pipes. Obviously, the better the insulation the less thermal energy that is wasted. To reduce energy loss, the Standard UNI EN 1264, which establishes the sizing rules for floor hea-ting systems, imposes minimum thermal resistance values for the layer of insulating material in rela-tion to the climatic conditions. The minimum value required is 0,75 m2K/W for rooms which lieabove other continuously heated rooms. A panel of expanded polystyrene which insures such a ther-mal resistance value must have a minimum thickness of at least 22-25 mm.A critical analysis of the difference between a building with an insulated floor and one without willsuccessively be made.

G.2. Numeric analysis of insulation

G.2.1. Calculation basis To evaluate the positive effect of insulating panels on the output of the floor heating systems, the pro-ject results of two systems to UNI EN 1264 will be compared: one with V-ESSE30 panels and theother with no type of insulation. According to the Standard requirements, and with reference toFigure G.3, the thermal resistance to the vertical thermal flow both towards the inside of the roomRo and towards the underlying room Ru must be calculated as indicated in chapter 5. The percenta-ge energy loss is calculated as the proportion qb /(qa+qb) between the downward thermal flow andthe total thermal flow created by the system.

145

Figure G.3. Energy flow of a radiant floor

Page 145: Pardo Seal A

G.2.2. Results To evaluate the energy loss through the floor, systems were sized with and without insulating panels,maintaining the floor structure and varying several fundamental parameters: the temperature of theunderlying room Tu and the useful thermal flow qa.A building of a residential type of approximately 100 m2 was considered and the system was instal-led with the multilayer pipe Valsir Mixal 16x2.

Variation of the external temperature at equal thermal flowA thermal flow of 70 W/m2 was considered and the downward energy loss in a room with floorinsulation and one without insulation was calculated at varying temperatures in the underlying room. From Table G.1 and Figure G.4 the following can be observed: going from a temperature Tu of -6°C (rooms over a terrace in cold climates) to a temperature of +4°C (rooms on the ground)until reaching a temperature of +16°C (rooms over heated rooms) there is an energy loss whichvaries from 26% to 12% for systems with insulating panels and which goes from 44% to 24% forsystems without insulating panels.A building without insulation, therefore, has significantly higher energy losses and on average, theyare double the losses in buildings with floor insulation. In the coldest conditions, the energy loss in asystem with no floor insulation can reach values as high as 44%.

146

Table G.1. Downward energy loss according to the variation in temperature of the room below

-6

-4

-2

2

4

6

8

10

12

14

16

[°C]

Temperatureof the room below

24,1

22,8

21,5

18,8

17,5

16,1

14,8

13,5

12,2

10,8

9,5

[W/m2]

Downward thermal flow withinsulating panel

26%

25%

23%

21%

20%

19%

17%

16%

15%

13%

12%

[%]

Energy loss

56,1

53,0

49,9

43,7

40,6

37,5

34,5

31,4

28,3

25,2

22,1

[W/m2]

Downward thermalflow without

insulating panel

44%

43%

42%

38%

37%

35%

33%

31%

29%

26%

24%

[%]

Energy loss

+19%

+19%

+18%

+17%

+17%

+16%

+16%

+15%

+14%

+13%

+12%

[%]

Tu qa qb

Page 146: Pardo Seal A

Variation of thermal flow at equal external temperatureA building situated on the ground with a temperature of +5°C was taken into consideration and, asthe temperature necessary for heating the rooms varied, the downward energy loss was calculatedfor rooms with and without floor insulation. From Table G.2 and Figure G.5 the following can be observed: going from a thermal flow qa of 50W/m2 to 95 W/m2 there is an energy loss which drops from 23% to 17% for systems with insulatingpanels and 41% to 32% for systems without insulating panels.Again, it has been revealed that the energy loss for systems without floor insulation is double the lossin systems with floor insulation. In conditions of low thermal flow (50 W/m2) the energy loss reaches41%.

147

Figure G.4. Trend of downward loss in relation to the temperature variation in the underlying room

Table G.2. Downward energy loss in relation to the variation of useful thermal flow

50

55

60

65

70

75

80

85

90

95

[W/m2]

Upward thermal flow

14,9

15,3

15,8

16,3

16,8

17,3

17,8

18,3

18,8

19,3

[W/m2]

Downward thermalflow with

insulating panel

23%

22%

21%

20%

19%

19%

18%

18%

17%

17%

[%]

Energy loss

34,6

35,7

36,8

38,0

39,1

40,2

41,4

42,5

43,6

44,8

[W/m2]

Downward thermalflow without

insulating panel

41%

39%

38%

37%

36%

35%

34%

33%

33%

32%

[%]

Energyloss

+18%

+18%

+17%

+17%

+16%

+16%

+16%

+16%

+15%

+15%

[%]

qa qa qb

Page 147: Pardo Seal A

148

Economic analysisFrom a simplified economic analysis it is possible to evaluate the difference in terms of heatingexpense between the two buildings, the first with floor insulation (which we will call case A), thesecond without floor insulation (which we will call case B). Let's suppose that the building is situatedon the round (Tu=+4°C) and the heating power is 7000 W (70 W/m2 for a surface of 100 m2).

From the table the following is read:- the downward energy loss in case A is 19%- the downward energy loss in case B is 36%

therefore:- the necessary heating power in case A is 8640 W- the necessary heating power in case B is 10937 W

If a boiler efficiency of 90% and methane heating power of 8200 kcal/m3 (34330 kJ/m3) is consi-dered, it is possible to determine the flow of methane necessary in both cases and the annual con-sumption, considering 75 days of heating per year and the cost of methane at 0,55 €/m3. The result is therefore:

- a flow of methane in case A of 1.01 m3/h,- a flow of methane in case B of 1.27 m3/h

and annual consumption is:

- 1810 m3/year equal to approx. 996 €/year in case A, and - 2290 m3/year equal to approx. 1260 €/year with a difference of approx. 265 €/year in

case B.

Figure G.5. Trend of downward loss in relation to the variation of thermal flow

Page 148: Pardo Seal A

G.3. ConclusionFrom the results obtained it is deduced that the absence of insulating panels has a significant nega-tive effect on the heat output of the heating system which could entail an energy loss of as much as40%. The use of insulating panels is therefore always strongly recommended; it allows, in fact, toachieve important reductions in annual consumption, in the example taken into consideration, asaving in methane consumption of approximately 20% was calculated. The considerations that weremade for residential floor heating systems can be transferred to an industrial type floor heatingsystem with the same results.The use of insulating panels influences not only consumption but also the performance of the system,it has been seen how the presence of the panel reduces the thermal mass and therefore enables thesystem to respond quicker to changes in the temperature.

149

Page 149: Pardo Seal A

150

HH.. MMeeaassuurreemmeenntt uunniittss

Table H.1. Basic and supplementary measurement units

Quantity Unit Symbol

Length meters m

Mass kilogram kg

Time seconds s

Electricity ampere A

Temperature kelvin K

Light intensity candle cd

Quantity of a substance mole mole

Plane angle radian rad

Solid angle steradian sr

Page 150: Pardo Seal A

151

Length

1 inch = 25,40 mm

1 ft (foot) = 0,3048 m

1 yd (yard) = 0,9144 m

1 mi (US mile) = 1,609 km

1 mi (nautical mile) = 1,852 km

Area

1 inch2 = 645,2 mm2

1 ft2 = 0,09290 m2

Volume and capacity

1 l = 0,001 m3

1 inch3 = 16,39 cm3

1 ft3 = 0,02832 m3

1 US gal (gallon) = 0,003785 m3

1 US gal = 3,785 l

1 UK gal = 0,004546 m3

1 UK gal = 4,546 l

Mass

1 kg = 2,204 lb

1 t (tonne) = 1000 kg

1 oz (ounce) = 28,35 g

Density

1 lb/ft3 = 16,02 kg/m3

Force

1 N (Newton) = 0,102 kgf

1 kgf (kg force) = 9,81 N

1 lbf = 4,448 N

Pressure

1 Pa = 1 N/m2

1 bar = 100000 Pa

1 bar = 1,019 kg/cm2

1 bar = 14,48 psi

1 atm (standard atmosphere) = 101325 Pa

1 atm = 760 mm Hg

1 at (metric atmosphere) = 1 kg/cm2

Table H.2. Conversion factors

Page 151: Pardo Seal A

152

1 at = 736 mm Hg

1 at = 10 m H2O

1 atm = 1,033 at

1 lb/inch2 = 6,895 kPa

1 inch H2O = 249,1 Pa

1 inch Hg = 3,386 kPa

1 mm H2O = 9,807 Pa

1 mm Hg = 133,3 Pa

1 torr = 133,3 Pa

1 mbar = 100 Pa

1 psi = 1 lb/inch2

Energy – Heat - Work

1 Btu (British Thermal Unit) = 1,055 kJ

1 kWh = 3,6 MJ

1 kcal = 4,187 kJ

1 J (Joule) = 1 N•m

1 J = 0,102 kgf•m

Output

1 W (Watt) = 1,36 CV

1 W = 1,34 HP

1 W = 0,860 kcal/h

1 kcal/h = 1,162 W

1 CV = 0,986 HP

1 HP = 1,014 CV

1 Btu/h = 0,2931 W

1 J/s = 1 W

1 lb/h (vapour) = 0,30 kW

Flow

1 ft3/h = 7,866 ml/s

1 ft3/min = 471,9 ml/s

1 l/s = 60 l/min

1 l/s = 3600 l/h

1 l/s = 3,6 m3/h

Energy flow

1 Btu/ft3 = 37,26 kJ/m3

1 kcal/m3 = 4,187 kJ/m3

1 Btu/lb = 2,326 kJ/kg

1 kcal/kg = 4,187 kJ/kg

Page 152: Pardo Seal A

153

Heat flow

1 Btu/ft2.h = 3,155 W/m2

1 Btu/in2.h = 454,2 W/m2

1 kcal/m2.h = 1,162 W/m2

Specific heat

1 Btu/lb.°F = 4,187 kJ/kg.K

1 kcal/kg.°C = 4,187 kJ/kg.K

1 Btu/ft3.°F = 67,07 kJ/m3.K

1 kcal/m3.°C = 4,187 kJ/m3.K

Heat conductivity

1 Btu/ft.h.°F = 1,731 W/m.K

1 kcal/m.h.°C = 1,162 W/m.K

1 cal/cm.s.°C = 418,7 W/m.K

Heat conductance

1 Btu/ft2.h.°F = 5,678 W/m2.K

1 kcal/m2.h.°C = 1,162 W/m2.K

1 cal/cm2.s.°C = 41,87 W/m2.K

Temperature scale

1 K (Kelvin) = 5/9.°R

1°R (Rankine) = °F + 459,67

1°C (Celsius) = 5/9.(°F - 32)

1 K = 5/9.(°F + 459,67)

1 K = °C + 273,15

Page 153: Pardo Seal A

NNootteess::

Page 154: Pardo Seal A

LO2 -

242/0

www.valsir.it

Uff.

Pub

. V

alsi

r -

Set

tem

bre

2004

(2.

000

- 09

/200

4)

You can indicate in which areas you would like to see improvement, indicating us your degree ofsatisfaction and filling in an appropriate form. Visit our Web Site www.valsir.it, enter the“Services” section and click “Give your opinion on Valsir” .

Among the first companies in the world

to obtain the ISO 9001:2000Certificate from DNV Certification Institute

VALSIR S.p.A.Località Merlaro, 2 25078 Vestone - Brescia - ItaliaTel. +39.0365.877011 - Fax [email protected] - [email protected] - [email protected]