COOL ROOF COATINGS ON INDUSTRIAL BUILDINGS

(1)

INDUSTRIAL BUILDINGS

An Energy Study of Reflective Coatings

Isak Sjödin

EN1822

Master thesis, 30 credits

(2)

microspheres on industrial buildings, a warehouse was built-up in the computer simulation software IDA-ICE. The warehouse was modelled in line with ASHRAE 90.1. 2004 ”Energy Standard for Buildings Except Low-Rise Residential Buildings”.

Four different cases were set up where the coating of the roof was the only variable.

Two coatings containing Expancel^® microspheres - one white and one red coating were compared to the same white coating without Expancel^® microspheres and the ”base case” where there is no coating at all. The same circumstances were also implemented in a practical laboratory test using a model warehouse with a detachable roof. Four interchangeable roofs with different roof coatings constitute the various cases in the laboratory tests. A ”sun” consisting of statically mounted IR light bulbs were constructed, as well as a cooling system to measure the difference in cooling effect (maintaining a constant indoor temperature) between the different cases as a result of the change in insolation.

The results of the computational simulations show that for a warehouse placed in Houston, Texas about 50 MW h in combined heating and cooling energy can be saved yearly between the best and the worst case, a reduction of 6.2%. Changing the geographic placement of the warehouse to Tepic, Mexico the corresponding savings were determined to 83 MW h or 13.5%.

A way of determining the yearly savings in heating and cooling amount for the warehouse with a generic roof coating, only knowing the SRI value of the coating, was developed. It was determined that for every unit-increment of the SRI value the yearly savings for the warehouse placed in Houston, Texas were 718 kW h and 0.1%. The corresponding savings for the warehouse placed in Tepic, Mexico were determined to be 1252 kW h and 0.22%.

The laboratory tests showed that the indoor temperature of the model warehouse decreased by close to 16°C between the best and the worst case.

(3)

to thank Expancel^® for the opportunity to execute my project and for all the help I have received during the project. I am especially grateful for my supervisor Jan Nordin and Olof Sandin, but also Jonas Vestin, Per Ajdén and Fredrik Svensson for every contribution, big or small, made to the project.

I would also like to thank my family, friends and classmates for the support and cheering throughout the project.

Isak Sjödin, January 2019.

ii

(4)

1 Introduction

By coating a roof white, or a similar pale color, less heat is absorbed if compared to a dark-colored roof. The lighter-colored roof or “cool roof” reflects heat, consequently lowering cooling costs of the building. The former US Vice President Al Gore, alongside Mayor Bloomberg, launched in September of 2009 the “NYC Cool Roofs” initiative by saying: ”The threat we face from the climate crisis is unsurpassed and smart policies like installing cool roofs are one way that we are going to meet the challenge” [1]. The term ”cool roof” is a generic term for roofs with the purpose of reflecting and emitting a sufficiently large fraction of the sun’s energy that otherwise would have been absorbed by the building.

During the summer when the indoor temperature in a building can get uncomfort- ably high, one solution is to install an air conditioning unit (ACU). However, this may not be the best solution to the problem. Yes – the temperature will decrease when using the ACU, but as a result the energy usage will increase along with the peak electricity demand, challenging the power supply. If we want to save the environment and slow down the greenhouse effect everything needs to be energy efficient and renewable energy must be used. Instead of having the most efficient ACU on the market and using renewable energy, firstly the reason how/why the temperature is too high should be examined. If the amount of energy absorbed by the building can be minimized, the indoor temperature will remain cooler. As previously mentioned - a white roof will reflect more of the solar radiation than a roof of darker color. However, the solar reflection properties of a white coating can be further improved. Coatings formulated with T iO2 as the pigment and Expancel^® thermoplastic microspheres as an additive excels when compared to other types of fillers as additive, and shows beneficial optical properties for solar reflectance [2].

The cool roof coating containing T iO2 and Expancel^® thermoplastic microspheres has been claimed to lower the external temperature of the roof by as much as 15°C, depending on intensity of solar radiation and roof type [3].

1.1 Background

AkzoNobel Pulp and Performance AB, also trading as Nouryon, is a multina- tional company with a presence in major parts of the world. The company pro- duces expandable thermoplastic microspheres (Expancel^® microspheres) used as lightweight fillers and blowing agents. The production and RD&I are stationed in Stockvik, Sundsvall, Sweden. When the unexpanded microspheres are exposed to heat they expand up to 60 times their original volume without increasing in weight.

The result is a very low density microsphere with a large area of applications - shoe soles, wine corks, paint, sealants and adhesives are just some examples. Another

(7)

application of Expancel^® microspheres is in elastomeric cool roof coatings. The microspheres ability of reflecting visible and near-infrared light combined with its insulating properties makes it a key ingredients in a high quality cool roof coating [2].

Many studies of cool roofs are primarily alluding the concept of the Urban Heat Island (UHI) effect – which briefly is the phenomenon where urban cities tend to become warmer than its surrounding residential suburbs and rural areas. In cities as Chicago, Sacramento and Salt Lake City, roofs and pavements can constitute up to 60% of urban surfaces [4]. These surfaces are often dark - mainly absorbing the sunlight reaching them, converting it into heat. Along with the lack of vegetation, the high percentage of roofs and pavements are significant reasons behind the UHI effect.

This study is examining the effect of cool roofs in the industrial sector rather than the residential sector. The main difference is the design of the buildings. One example of an industrial building is a warehouse/repository which often is a large building of corrugated metal sheets with little to no insulation. These buildings can often be non-air-conditioned, however there is still a need of lowering the indoor temperature. Examples where the indoor temperature is an important fac- tor is when there is heat sensitive material in the warehouses, but also the work environment for the employees in the warehouses. With a cool roof, the indoor temperature can be lowered and the peak temperature, often at midday on sunny days, can be lowered.

Air sealing and insulation are other ways of improving the comfort in non-air- conditioned buildings but requires access to walls and attic spaces – something that is not always an option in existing warehouses. With cool roof coating there is no need to alter the buildings structure and thereby often being the simplest and most cost-efficient option of the two.

1.2 Real life case studies of cool roof coatings

SkyCool is an Australian company based in Sidney that have developed and patented a very effective, maintenance-free cool roof coating called just ”SkyCool”.

”A Cool Roof Coating that almost defies the conventional laws of physics. SkyCool provides a profound cooling effect that can halve the running costs of mechanical air-conditioning in buildings with metal roofs” [5, 6], according to CSIRO, Aus- tralia’s national science research agency.

Examples of buildings in Australia whose roof have been coated with SkyCool

(8)

are: Melbourne International Airport, Australia’s premier national retailer ”Wool- worths”, and DuPont’s seed storage warehouse. In Melbourne International Air- port the application of SkyCool on the 43 000 m² of terminal roofing resulted in that the airport could shut down 16 of their auxiliary air-conditioning units.

In addition to not needing all of the available air-conditioning units, it has been reported that the airport’s cooling system’s peak loads is now operating at around 40% lower than before application of SkyCool [7].

A two-year field trial by Woolworths compared two supermarkets, one with Sky- Cool’s cool roof coating, and the other one having standard roof. It was found that the monthly air-conditioning energy decreased between 38% and 61%, depending on weather conditions, with an overall average saving of 47%.

The non air-conditioned warehouse of DuPont’s seed storage had problems with the interior temperature regularly exceeding the ambient temperature. After coating the roof’s with SkyCool, the interior temperature has consistently been 3°C to 10°C below ambient on warmer days [8].

1.3 Aim

The aim of this Master’s thesis is to investigate the effect of three types of cool roof coatings on industrial buildings. The effect will be analyzed through computer simulations and verified in practice using a physical model building. A white reference coating filled with CaCO3, a white coating filled with Expancel^® microspheres, and a red coating containing Expancel^® microspheres will be compared. In addition, template values of the roof coating’s SRI (Solar Reflectance Index) will be computed with the purpose of analyzing how a roof’s SRI value affects the energy consumption of a certain building.

1.4 Limitations

The limitations that have been determined in joint consultation with the supervisor at AkzoNobel Pulp and Performance Chemicals AB are as follows.

• The foundation of the results are based on the theoretical part of the project.

The practical moment is to verify the results of the computational simulations.

• Two viable reference cases for comparison of the cool roof coatings containing microspheres - bare stainless steel and stainless steel coated with the same coating with CaCO3 instead of microspheres as filler.

(9)

• One type of building will be examined - a large warehouse/repository.

• No slope on the roofs of the building.

• Geographic placement of the simulations is the American south (Houston, Texas).

• No site shading or nearby buildings are considered.

(10)

2 Theory

This section will elucidate important concepts of the paper to give the reader the information needed to sufficiently understand and be able to discuss the results of the report.

2.1 IDA-ICE

IDA Indoor Climate and Energy (IDA-ICE) is a simulation tool developed by EQUA Simulation AB that is used to study the thermal indoor climate and energy consumption of a building. The program allows the user to thoroughly define the structure of the building with either user-defined materials, thermal resistance etc. or using existing materials from the program’s vast database. Geographic location, orientation, climate, wind profile, thermal bridges, ground properties, daily occupancy, HVAC-system (Heating, Ventilation and Air Conditioning) are some of the settings/options within the program.

2.1.1 Verification of IDA-ICE

In his Master’s thesis from 2015, Gulliksson [9] compared three simulation soft- wares used for simulation of energy and climate performance. The study examined the difference of IDA-ICE, VIP-Energy and IES Virtual Environment, mainly con- cerning the simulation result. It was concluded that IDA-ICE is well suited for calculation of power demand for heating and cooling, yearly energy demand and climate simulations - thereby being the most applicable program for engineering consultants working with energy and climate simulations of buildings.

2.2 Heat transfer

Heat is defined as the form of energy that can be transferred from one system to another as a result of temperature differences [10]. The rate of net heat transfer into or out of a control volume is determined by

Q = ˙˙ mc_p∆T [kJ/s] (1)

where ˙m is the mass flow in [kg/s], cp is the specific heat capacity [(kJ/kg)°C], and ∆T is the temperature difference in [°C].

According to the second law of thermodynamics, the total entropy (degree of disorder) of the universe always increases - meaning that heat transfer only occur from high temperature to low temperature. The three kinds of heat transfer are

(11)

2.2.1 Conduction

Energy transfer from highly energetic particles in a substance to the adjoining particles with lower energy is called conduction and can take place in solids, liquids, or gases. The rate of heat conduction through a medium is depending on the temperature difference across the medium, as well as the geometry, thickness and material of the medium according to

Q˙_Cond = kA(T₁− T₂

∆x ) = −kA(∆T

∆x) [W ] (2)

where k is the thermal conductivity of the material in [W/m · K], A is the area [m²] of the material normal to the heat transfer, ∆T is the temperature difference across the material, ∆x is the thickness of the material. In the limiting case of

∆x → 0, Eq.2 reduces to the differential form Q˙_Cond = −kA(dT

dx) [W ] (3)

which is called Fourier’s law of heat conduction, where dT/dx is the temperature gradient through the material [10].

2.2.2 Convection

Convection is the energy transfer between a solid surface and the adjacent liquid or gas that is in motion. It combines both conduction and fluid motion - the faster the fluid motion, the greater the convection heat transfer. If there is no fluid motion, the heat transfer between the solid surface and the adjacent liquid is by pure conduction.

Heat transfer between a solid surface and an adjacent fluid is called forced convection if the movement of the fluid is forced, i.e the motion of the fluid is caused by external means such as a fan, pump or the wind. Natural (or free) convection is convection caused by buoyancy forces induced by density differences due to varia- tion of temperature in the fluid.

The rate of convection heat transfer is expressed by Newton’s law of cooling as Q˙_Conv = hA_s(T_s− T∞) [W ] (4) where h is the convection heat transfer coefficient in [W/m²· K], As is the surface area in [m²] through which the convection heat transfer occurs, Ts is the surface temperature, and T^∞ is the temperature sufficiently far from the surface [10].

(12)

2.2.3 Radiation

Energy emitted by matter in the form of electromagnetic waves (or photons) is called radiation and is a result of changes in the electronic configurations of atoms or molecules. For this study and in heat transfer studies, thermal radiation is of greatest interest. All matter with temperature greater than absolute zero emits thermal radiation. The particle motion results in charge-acceleration or dipole- oscillation which produce electromagnetic radiation. Radiation incident on a surface is either absorbed (A), reflected (R) or transmitted (T ) and through energy conservation, the following relation is obtained (at each wavelength) [11]

A(λ) + R(λ) + T (λ) = 1 (5)

For an opaque surface, i.e. a surface without transmittance, the above relation is simplified to

A(λ) + R(λ) = 1 (6)

The maximum rate of radiation that can be emitted from a surface at a temperature Ts (temperature in [K]) is given by the Stefan-Boltzmann law [10] as

Q˙emit,max= σAsT_s⁴ [W ] (7)

where σ = 5.67 · 10⁻⁸ [W/m²· K⁴] and is called the Stefan-Boltzmann constant, and As is the surface area in [m²]. An idealized body that emits radiation at this maximum rate, having perfect absorptivity at all wavelengths as well as being a perfect emitter, is called a blackbody. Radiation of such perfect emitters is called blackbody radiation. No real body emits radiation at the maximum rate, in fact, all real surfaces emits less radiation than a blackbody at the same temperature and is expressed as

Q˙_emit= σA_sT_s⁴ [W ] (8)

where is the emissivity of the surface, ranging from 0 to 1. Emissivity is a measurement on how close a surface approximates a blackbody, whose emissivity is equal to one.

The absorptivity (α) of a surface is another important radiation property, and is defined as the fraction of the radiation energy incident on a surface that is absorbed by the surface. Its value is in the range 0 ≤ α ≤ 1, where a blackbody has α = 1 as previously mentioned. Kirchhoff’s law of radiation states that an object’s emissivity at a certain wavelength is equal to the absorptivity at the same wavelength, when in thermal equilibrium [10].

(λ) = α(λ) (9)

(13)

2.3 Electromagnetic spectrum

The electromagnetic spectrum contains electromagnetic radiation of different wavelengths - from gamma rays with a wavelength of less than 1 ∗ 10⁻¹¹meter, to radio waves with a wavelength of more than 1 meter [10]. In Fig.1 all components of the electromagnetic spectrum can be seen.

Figure 1 – A graphic representation of the electromagnetic spectrum [12].

Wavelengths of 0.40 − 0.76 µm corresponds to the portion visible for the human eye. As mentioned in section 2.2.3, thermal radiation is of greatest interest in this study and is defined as the portion of the electromagnetic spectrum that extends from about 0.1 to 100 µm. The thermal radiation therefore spans the entire visible and infrared (IR) radiation as well as a portion of the ultraviolet (UV) radiation.

2.4 Solar radiation

Solar irradiation, or insolation, is the amount of solar radiation that reaches the earth’s surface. It is thereby a measure of the solar energy incident on a specified area over time. The unit (kW h/m²) per day, represents the average amount of energy striking an area each day. Solar irradiance, or solar radiation, is a measure of the average amount of power hitting an area over a year [13]. The average annual solar radiation at the outermost part of the earth’s atmosphere is approximately 1366 W/m² (called the solar constant) [14], but when passing through the atmosphere the solar radiation is attenuated yielding a maximum of 1000 W/m² at sea level on a clear day. The actual value at sea level is much lower and varies with the solar angle and atmospheric circumstances, resulting in a daily average insolation for the earth of approximately 250 W/m² or 6 kW h/m² [15].

2.4.1 Solar spectrum

The spectrum of the sun’s electromagnetic radiation at sea level (after attenuation from the atmosphere) is shown in Fig.2 below.

(14)

Figure 2 – Solar spectrum at sea level (air mass 1.5) according to ISO 9845.

Interval of light visible to the human eye is marked in blue.

The solar spectrum above can be divided in the segments UV< 0.40 µm, VIS 0.40−0.76 µm, and NIR 0.76−2.5 µm. The peak in intensity of the solar spectrum occur in the visible segment at about 555 µm, however the visible segment does not contain more than about 40% of the total energy from solar radiation. For example, the total energy of direct solar radiation incident at sea level at a solar altitude of 41.8° consists of about 3% UV, 38% visible, and 59% infrared radiation [10].

2.4.2 Incandescent light bulbs

Incandescent light bulbs operate by heating its central filament to temperatures where the filament starts glowing, thereby producing visible light. The radiation from an incandescent bulb resembles the solar radiation at sea level, but since the sun is hotter than the filament in the bulb, radiation from the sun is stronger at shorter wavelengths. Meaning, the incandescent light bulb’s ”radiation peak” is located at higher wavelengths as compared to the sun’s, as can be seen in Fig.3.

(15)

Figure 3 – The sun’s irradiance compared to an incandescent lamp’s [16].

Notice that in the preceding figure the main part of the radiation from the incandescent light bulb is NIR-radiation - similar to the sun. The light, i.e visible radiation, is a smaller portion of the total irradiance.

2.5 Cool roofs

From Eq.6 it can be seen that the radiation incident on an opaque surface (a roof for example) is either absorbed or reflected. The purpose of a cool roof is to reflect as much of the solar radiation as possible. As seen in Fig.2, the solar spectrum at sea level ranges from about 300 − 2500 nm, meaning that the reflectance in that range should be 100% for an ideal cool roof coating. In Fig.4a, the temperature dependency of a blackbody’s radiation peak can be seen. As the temperature rises, the radiation peak is translocated to lower wavelengths.

(16)

Figure 4 – (a) Blackbody radiation from surfaces at different temperatures, (b) solar spectrum before attenuation of the atmosphere, (c) absorption in the atmosphere during clear weather, (d) relative sensitivity of the human eye [17].

The temperature dependency is the reason why a roof emits radiation at higher wavelengths compared to the sun. Fig.4a also shows that a roof approximately emits radiation at wavelengths higher than 2 µm (at temperature below 100°C), meaning that the main part of the solar spectra at sea level differ from the radiation emitted by a roof (except a small interval between 2 − 2.5 µm), which enables for favorable optical properties in a cool roof paint/coating. The ideal cool roof coating should thereby have 100% reflectance of radiation up to a wavelength of 2.5 µm, followed by a sudden drop to 0% reflectance. For wavelengths above 2.5 µm the ideal cool roof coating would imitate a perfect blackbody and according to Eq.9 absorb, and consequently emit, 100% of the radiation.

(17)

2.5.1 Solar Reflection Index

The Solar Reflection Index (SRI) is a measure of how cool/hot a roof is based on its reflectance and emittance properties. A standard black surface with solar reflectance of 0.05 and thermal emittance of 0.90 is defined to have a SRI of 0. A white surface with solar reflectance of 0.80 and thermal emittance of 0.90 is defined as 100 in terms of SRI. This definition enables the SRI-value to be less than zero but also greater than 100.

The calculation of SRI is based on total solar reflectance and thermal emittance of the surface. A calculator for SRI has been developed by the Heat Island Group at Lawrence Berkeley National Laboratory using the ASTM Standard E 1980

”Standard Practice for Calculating Solar Reflectance Index of Horizontal and Low- Sloped Opaque Surfaces” [18].

The equation for calculating total solar reflectance (TSR) for a material is

T SR =

λ2

R

λ1

R(λ)I(λ)dλ

λ2

R

λ1

I(λ)dλ

(10)

where I(λ) is the solar radiation intensity [W/m²·sr]. The equation for calculating the thermal emittance of a material is

E =

λ2

R

λ1

(λ)J (λ)dλ

λ2

R

λ1

J (λ)dλ

(11)

where J(λ) is the blackbody radiation [W/m² · µm] at a given temperature, and is determined using

J (λ) = 2πhc²

λ⁵· exp(_kλT^hc ) (12)

where h = 6.6261 · 10⁻³⁴ [J s] is the Planck constant, c = 299 792 458 [m/s] is the speed of light in vacuum, λ [m] is the wavelength, k = 1.38065 · 10⁻²³[J/K] is the Boltzmann constant and T [K] is the absolute temperature [10].

2.6 Coatings

A coating/paint is build up by three fundamental components: a pigment, a binder and a solvent. The pigment provides color and coverage, while the binder accounts

(18)

for durability, flexibility and gloss. The solvent or diluent disperses the polymer and alters the viscosity of the formulation. In addition to the three main components, additives can be used to strengthen desirable properties to the formulation such as surface tension, appearance and biocides (to prevent bacterial growth) for example.

2.7 Expancel

^®

microspheres

The microspheres (henceforth called MS) produced by AkzoNobel Pulp and Per- formance Chemicals AB are micron sized hollow thermoplastic spheres containing a hydrocarbon blowing agent. During the manufacturing of the microspheres a sus- pension polymerization is used. The organic phase of the polymerization consists mainly of monomers and blowing agent, and the water phase contains an inorganic stabilizer and initiator. When the two phases are mixed using high speed stirring an emulsion is formed with small droplets of organic phase in water. As the temperature is increased to where the initiator begin initiating polymerization, a polymer shell is formed from the monomers - and as the polymer shell forms and the hydrocarbon blowing agent is encapsulated within. When exposed to heat the MS expands up to 60 times their original volume without increasing in weight.

When Expancel^® MS replaces the inorganic fillers in coatings, the reflectance of the coating is enhanced [2] - an important property for high performing cool roof coatings (see section 2.5).

2.8 Building heat balance

The heat balance of a building is determined from how much heat flows into the building and how much heat flows out of the building [19]. Typically there are three different losses of heat/energy in a building - transmission losses Ptrans [W ], ventilation losses Pvent[W ], and infiltration losses Pinf il[W ]. The heat/energy gain in the building are: insolation Pinsol [W ], internally generated heat (occupants, computers/devices etc.) Pintern [W ], and through the building’s heat system Pheat

[W ]. Consequently, the heat balance for a building is expressed as

P_trans+ P_vent+ P_{inf il}= P_insol+ P_intern+ P_heat [W ] (13)

(19)

3 Methodology

This section of the report describes the approach and implementation of the project, which is divided into two main parts - computational simulations and laboratory tests.

For the simulations there was a need for specifications/data of a typical industrial building (warehouse) placed in USA. ASHRAE (American Society of Heating, Refrigerating and Air-Conditioning Engineers) 90.1 2004 ”Energy Standard for Buildings Except Low-Rise Residential Buildings” is, as the name implies, a standard for building properties. By using data sheets and reports of the 90.1 2004 standard from the U.S Department of Energy [20, 21], the specifications for the warehouse was determined.

The cases on which the computational simulations and laboratory tests are based upon are as follows

• Case 1 - Bare stainless steel roof, no coating.

• Case 2 - White roof coating.

• Case 3 - White roof coating containing Expancel^® MS.

• Case 4 - Red roof coating containing Expancel^® MS.

Case 1 is a reference case with the purpose to act as a baseline for both the computer simulations and practical tests. Case 2 is also a reference case, where the coating has the same volumetric proportions as the coatings used in case 3 and 4.

To avoid misunderstanding due to that the above mentioned cases will be used in both computational simulations and laboratory tests, the two types of simulations have been assigned a number. For example, case 1.3 represents case 3 in the computational simulation. Similarly, case 2.3 represents case 3 in the laboratory test.

3.1 Computational simulation

IDA-ICE edition 4.7.1 was used for the computational simulations. From the specifications in the previous subsection, a building of warehouse type was defined within the software, as shown in Fig.5 below.

(20)

(a) South-west corner. (b) North-east corner.

Figure 5– The constructed building in IDA-ICE from two different angles.

The warehouse is a 100 ∗ 45 meter building with a ceiling height of 9 meter. It consists of three zones - an office positioned in the north-west corner of the warehouse, a large bulk storage located in the east part of the warehouse, and a fine storage in between.

The cases were implemented to the simulation software by changing the roof’s external optical properties according to:

• Case 1.1 - Longwave emissivity of 0.90, shortwave reflectance of 0.378, and a specularity of 0.101.

• Case 1.2 - Longwave emissivity of 0.90, and shortwave reflectance of 0.82.

The values for optical properties of stainless steel are taken from IDA-ICE database for the same material.

3.2 Laboratory tests

The laboratory tests demanded a model warehouse, built in scale to the simulated building. The purpose of the practical moment of the project was as a verification of the simulated results, as described in section 1.3-1.4. Therefore the main focus was not that the building should be an exact replica of the simulated building with accurate heat transfer abilities etc, but rather a relative comparison of the different cases.

The cases in the laboratory tests were as follows:

(21)

• Case 2.1 - Bare stainless steel.

• Case 2.2 - White roof coating with CaCO3 as a filler. ”Thermotek IMPAC 5000 White”.

• Case 2.3 - White roof coating containing Expancel^® MS as a filler. ”Ther- motek Bioreflection White”.

• Case 2.4 - Red roof coating containing Expancel^® MS as a filler. ”Thermotek Bioreflection Red”.

3.2.1 Set up

A simple model warehouse in scale 1:150 was built by ”Wikströms Plåt”. To be able to adequately compare the results of the laboratory tests there should only be one variable that changes between the different cases - namely the coating of the roof. Thus the building was built with a removable roof so that the roof could be swapped without alternating the set up (incident angle and distance to the "sun"

for example). One roof for each case/coating was constructed. The set up is shown in Fig.6.

Figure 6 – The set up for the laboratory tests (showing case 2.3, white roof coating containing Expancel^® MS.)

Commercial coatings were used to ensure correct components and thus eliminat- ing one source of error. The coatings were applied on its respective roof in two

(22)

thin layers using a fine structured paint roller. It is important that the coatings achieve total coverage of the roof surfaces, as well as not having too thick layers which could cause cracking upon drying. The thickness of the roof coating should not substantially affect the roof’s ability to transfer heat. The pigments used in the coatings were T iO2 and F e2O3 (white and red pigments respectively). The fillers used were CaCO3 and Expancel^® MS. Note that when substituting CaCO3

for microspheres it is important to use vol% instead of wt% because of the great density difference in between the two fillers.

An artificial sun consisting of three light bulbs emitting near infrared and infrared radiation was constructed by mounting them statically on a rack, see Fig.6. The light bulbs used as the artificial sun are Philips R125 IR [22]. Section 2.4.1-2.4.2 presents the necessary specifications to imitate the radiation from the sun. The spectral properties of the chosen lamp is presented in Fig.7.

Figure 7 – Spectral properties of Philips R125 IR light bulb [23].

Since the temperature of 2450 K is lower than the outer region of the sun (which is about 5800 K [10]) the radiation is translocated to longer wavelengths.

In section 2.8, the general form for a building’s energy balance is stated through Eq.13. The practical model was constructed to be airtight so there was no energy losses from infiltration, thereby simplifying the heat balance-relationship. By making sure the only heat generation was through insolation (no internally generated heat or by the heating system), and the only heat removal was through transmission losses - the model was optimized for the comparison of different cases by reassuring equivalent circumstances.

(23)

in heat gain between the different cases. The cooling system consists of a small radiator (from a computer liquid CPU cooling system [24]), and three small 12 Volt fans [25]. Water of constant temperature flows through small tubes into the building, through the radiator, and out of the building. The water used is taken directly from the municipality water faucet but in order to achieve constant temperature, the tubes distributing the water are submerged in an ice/water mixture.

The water flow is determined by a rotameter [26]. The 12 Volt fans circulate air inside the building (ensuring that the air is well mixed) and through the cooler - allowing convection heat transfer to the radiator. By adjusting the water flow so that the temperature of the air inside the building is the same in all cases and measuring the temperature of the water into and out of the building, the net heat transfer, or cooling power, can be calculated through Eq.1.

The locations of temperature measurement were: inside the building at half of the interior roof height, ambient temperature (sufficiently far away from the IR- lamps), and the inlet and outlet temperature of the water in the cooling system.

The temperatures were logged continuously throughout the test process.

3.2.2 Logging process

With the model warehouse in place, the cooling system is switched on. The system cools the building until the indoor temperature reaches a stationary state (henceforth called ”steady state 1”). When steady state 1 is achieved, the ”sun” is switched on resulting in a indoor temperature increment. The temperature will reach a new stationary state (”steady state 2”) when the amount of incoming solar energy is equal to the amount of cooling power + the building’s transmission losses. Now the cooling system is shut off while the sun is still on. The indoor temperature will rise yet again until stabilizing at a stationary state (”steady state 3”), where the amount of incoming solar energy is equal to the transmission losses of the building.

By awaiting steady state 2 before measuring the cooling power, using Eq.1, the one thing that changes between the different cases is the amount of heat transferred to the building - which becomes apparent in the temperature of the outlet cooling water.

3.3 SRI template values

As mentioned in section 2.5.1, the SRI value of a surface is a measure of how hot/cool the surface is based on reflectance and emittance properties. The definition of the value makes it obvious that a higher SRI value corresponds to a lower surface temperature. However, it is interesting to see how a difference in SRI af-

(24)

fects the indoor temperature and energy consumption for a certain building.

The corresponding reflectance and emittance value for a certain SRI value is implemented in the built up warehouse model in IDA-ICE. By varying the SRI value in steps of 5%-units per step, while the surfaces’ roughness and specularity are held constant, clear results of the effect of SRI value in energy savings can be obtained.

Since the SRI is dependent of both solar reflectance and thermal emittance, one parameter has to be held constant. The solar reflectance is the parameter that affects the SRI the most, while thermal emittance is relatively constant. The corresponding solar reflectance for a specific SRI value is presented below , in Tab.1, at a thermal emittance of = 0.9, a common value for the used coatings [2].

Table 1 – The corresponding solar reflectance values for certain SRI values.

= 0.9 SRI R (0-1)

70 0.585 75 0.622 80 0.658 85 0.694 90 0.729 95 0.764 100 0.800 105 0.835 110 0.870 115 0.904 120 0.939 125 0.973

129 1

The SRI value of 129 occurs when there is total solar reflectance, R = 1, and is used as an upper limit for the calculations. An SRI value of 70, typical for a pale/light roof coating, acts as a lower limit for the calculations.

(25)

4 Results

This section will present the results for the computational simulations and the laboratory tests, as well as the result of the calculations for the SRI template values.

4.1 Computational simulation

The results of the computational simulations are mainly presented in graphs and charts of indoor temperature in the different cases. In section 1.2 it is mentioned that the warehouses could often lack air-conditioning, however the amount of power consumed by the building’s ACU within the different cases while keeping a set indoor temperature provides evident result of the effect of cool roof coatings - which is why air conditioning has been implemented. The heating set point is 20°C and the cooling set point is 24°C for the building heating and cooling system. The results for the corresponding case without a cooling system will also be presented to display the temperature distribution over time and with regards to the thermal comfort improvement as a result of the coatings.

The results of each case will first be presented separately, followed by a comparison of the cases to clarify possible differences.

4.1.1 Case 1.1 - No coating

The results for the reference case without any coating (optical properties of bare stainless steel - emissivity = 0.90, reflectance = 0.378 and specularity = 0.101) is presented in Tab.2.

Table 2 – Delivered energy to warehouse with cooling system.

Component Purchased energy Peak demand kW h kW h/m² kW

Amount of cooling 714 825 161.5 227.4 Amount of heating 56 079 12.7 155.4 Total, heating and cooling 770 904 174.2

As expected the amount of cooling is substantially higher than the amount of heating for a building with low amount of insulation positioned in Texas (high insolation).

(26)

As previously mentioned not all warehouses have got cooling systems, however the thermal comfort in the building is still important. The results for the building lacking a cooling system (although having a heating system) is shown in Fig.8.

Figure 8 – Case 1.1 - Blue: Hourly temperature values in ”Bulk Storage” zone during 2018. Red: Average monthly temperature values in the same zone.

Note the high temperature from april/may (∼3000 h) to october/november (∼7000 h). The large fluctuations in the graph is due to the temperature difference during a day (daytime/nighttime). The average temperature for each month demonstrate the results more clearly. The highest average monthly temperature is 33.5°C and occurs in July. The highest hourly temperature occurs after 5000 hours (middle of July) and is 36.9°C. Presented in Tab.3 is the thermal comfort of the warehouse lacking cooling system.

Table 3– Thermal comfort in warehouse without cooling system. Case 1.1.

Percentage of hours when operative temperature is above 27°C in worst zone 56%

Percentage of hours when operative temperature is above 27°C in average zone 55%

Percentage of total occupant hours with thermal dissatisfaction 44%

(27)

As expected, the thermal discomfort in the building without a cooling system is high. This result lay the basis for the reference building and is anticipated to be improved in subsequent cases when the roof of the building is coated.

4.1.2 Case 1.2 - White coating (CaCO₃)

The results of the second case, with a white roof coating filled with CaCO3 (emissivity = 0.90, reflectance = 0.82) are showed in Tab.4.

The temperature distribution for the warehouse in case 1.2 without a cooling system is presented in Fig.9 below.

(28)

From the figure above it can be seen that the highest hourly temperature is 35.8°C at around 5000 hours. The highest average monthly temperature occurs in July and is 32.64°C. The thermal comfort in the building without cooling system is shown in Tab.5.

The thermal comfort is improved as compared to the reference case without roof coating.

4.1.3 Case 1.3 - White coating (Expancel^® MS)

The results of the third case with a white roof coating containing MS (emissivity

= 0.90, reflectance = 0.86) are showed in Tab.6-7 and Fig.10.

(29)

The temperature distribution over 8760 hours for the warehouse in case 1.3 without a cooling system is shown below.

From Fig.10 it can be seen that the highest hourly temperature is 35.76°C at around 5000 hours. The highest average monthly temperature occurs in July and is 32.57°C.

(30)

Thermal dissatisfaction in the warehouse with roof coating containing MS has not been improved when compared to the roof coating without MS.

4.1.4 Case 1.4 - Red coating (Expancel^® MS)

The results for the red roof coating containing MS (emissivity = 0.90, reflectance

= 0.37) is shown in Tab.8-9 and Fig.11 below.

The temperature distribution over 8760 hours for the warehouse in case 1.4 without a cooling system is shown below.

(31)

From Fig.11 it can be seen that the highest hourly temperature is 36.89°C at around 5000 hours. The highest average monthly temperature occurs in July and is 33.48°C.

The thermal comfort in this case is at the same level as the reference case (bare stainless steel roof without coating).

4.1.5 Comparison of the results from the computational simulation The lowest yearly amount of heating and cooling was seen in case 1.3 with the amount of 722 797 kW h (the white roof coating containing MS). The highest yearly amount of heating and cooling was produced in case 1.1, resulting in 770 904 kW h

(32)

(no coating). This result however is very close to the yearly heating and cooling amount in case 1.4 (red roof coating with MS). The difference between the best and worst case is 48 107 kW h, a reduction of the total yearly heating and cooling amount of 6.24%. The comparison is shown in Fig.12.

Figure 12 – Comparison of the total heating and cooling amount for a year for all four cases.

It is hard to distinguish a difference between the curves in the above figure, which is why Fig.13 shows an enlarged view of the last segment of the same figure where the results are more comprehensive. Notice that curve for ”Bare stainless steel”

and ”Red MS” are overlapping and are still hard to distinguish from each other.

Figure 13– Comparison of the end-values of the total heating and cooling amount for a year for all four cases.

(33)

The ”cooling peak demand” is lowered by 9 kW , a reduction of 4%, when comparing the best and worst case. Case 1.1 and 1.4 had a cooling peak demand of 227.4 kW, and case 1.3 had a peak demand of 218.5 kW . The difference in ”heating peak demand” differs by 0.4 kW between case 1.3 and case 1.1, with values of 155.8 and 155.4 kW respectively.

The percentage of total occupant hours with thermal dissatisfaction in the cases without a cooling system is improved by 5%-units from the worst case to the best case.

(34)

4.2 Laboratory tests

The results for the laboratory tests will be presented as the corresponding steady state temperatures and cooling effects for each case, as described in section 3.2.2.

4.2.1 Case 2.1 - No coating

The results for the first case for the warehouse with a roof without any coating, just bare stainless steel, are shown in Fig.14 and Tab.10.

Figure 14 – Temperature graph for case 2.1.

The figure above clearly shows the three stationary states. Observing the temperature inside the building (grey curve) it can be seen that ”steady state 1” occurs at approximately 0 − 25 min, ”steady state 2” at around 55 − 80 min, and ”steady state 3” at about 250 min. The cooling power is determined through Eq.1 where the temperature difference between ”H2O in” and ”H2O out” (blue and red curve) is ∆T . The yellow curve shows the ambient temperature sufficiently far away from the IR lamps.

The temperature inside the building and cooling power for each stationary state is shown in the following table.

(35)

Table 10– Temperature and cooling power for case 2.1.

Temp [°C] Cooling power [W]

Steady state 1 13.03 19.66 Steady state 2 26.40 74.18

Steady state 3 59.73 -

4.2.2 Case 2.2 - White coating (CaCO3)

The warehouse having a roof with white roof coating without Expancel^® MS showed the results presented in Fig.15 and Tab.11.

Note that the temperature inside the building in steady state 3 is significantly lower than in the first case. The temperature of steady state 2 is also lowered, despite a decrease in cooling effect of the cooling system.

(36)

4.2.3 Case 2.3 - White coating (Expancel^® MS)

The temperature distribution for the laboratory test of the warehouse with a white roof coating containing Expancel^® MS is shown in Fig.16.

The indoor temperature in steady state 3 is yet again lowered compared to the first cases. Notice also that the temperature at steady state 2 is lower than in the previous case even though the cooling power is at a lower value, as can be seen in Tab.12.

4.2.4 Case 2.4 - Red coating (Expancel^® MS)

The results for the warehouse with the red roof coating containing Expancel^® MS are presented in Fig.17 and Tab.13.

(37)

As expected, the indoor temperatures in steady state 2 and 3 for the red coating containing MS has raised as compared to the white coating containing MS.

4.2.5 Comparison of the results from the laboratory tests

The maximum difference in indoor temperature during steady state 3 is about 16°C, from 59.73°C in case 2.1 to 44.01°C in case 2.3.

(38)

4.3 SRI template values

By implementing the corresponding solar reflectance and thermal emittance values for each SRI value into the built up model in IDA-ICE, the results shown in Fig.18 were acquired.

Figure 18 – Annual amount of cooling energy as a function of SRI.

From the figure it can be seen that the yearly amount of cooling energy for the specified building decreases by 850 kW h for an increment of one SRI-unit (the slope of the linear regression).

The increase in solar reflection affects the yearly heating demand as well. The heat demand will increase (during cold periods when heat is needed) due to less solar energy being transferred throughout the building. This increment is shown in Fig.19 below.

(39)

Figure 19– Annual amount of heating energy as a function of SRI.

The increment in amount of heating energy per year is 132 kW h per SRI-unit, meaning that the total yearly energy requirement (combination of the amount of heating and cooling) will decrease by 718 kW h for every increment of one SRI-unit.

(40)

5 Discussion

The optical values for the different coatings were supposed to be analyzed using spectrophotometry to acquire the exact optical data for the used coatings. How- ever, the results of the spectrophotometry were not obtained in time to be included in the graduate project. The values used in this project are the solar reflectance and emittance values given by the distributor of the coating. For example, the data used for the red coating containing MS were for a similar product. This is one reason for why the results differ between the computational simulations and the laboratory tests. The laboratory tests shows that the indoor temperature during ”steady state 3” in the building with a red roof coating containing MS is significantly decreased compared to just bare stainless steel, but in the computational simulation the difference is barely noticeable. The lack of validated optical properties for the red coating containing MS leads to the computational results for case 1.4 being unusable since IDA-ICE bases its results on these properties. The results of the laboratory tests of the red roof coating containing MS are however still applicable since the methodology does not rely on the optical properties being predetermined.

When comparing the results of the computational simulations presented in this paper to papers investigating cool roof coatings using full-size models, it can be seen that the effect is more evident in the full-scale investigations. This phenomenon is examined in the technical modelling paper ”Issues and Solutions To More Realistically Simulate Conventional and Cool Roofs” by G Carter of Lend Lease [27]. The paper addresses the issue that a computer modelling program does not take into account all underlying factors affecting the end-result. How accurately the results of the computer simulations renders the results in real-life is beyond the scope of this paper, but is worth mentioning. To adequately acquire the effect of a cool roof coating, full-scale tests must be concluded.

5.1 Computational simulation

As expected, the yearly energy demand for the cooling system is reduced when the reflection properties of the roof is increased. Lowering the peak power demand is of major interest worldwide as the power grids get more strained. During a peak power demand the prices for electricity increase - a classic case of supply and demand. When the demand is high the power companies need auxiliary plants to manage and generate extra supply. These peak load plants are more expensive to use than the base load plants, and are often natural gas or coal powered because they are dispatchable, meaning they can be turned on and off quickly. The need for reducing peak power demand in cities is easily comprehensible when thinking

(41)

about the number of ACUs being installed as a result of the increasing urban temperature, as mentioned in the introduction.

5.2 Laboratory tests

The rotameter used in the laboratory tests was somewhat hard to adjust. Whether the problem of finding an exact volume flow was because of the rotameter or pres- sure differences in the incoming water has not been determined. This resulted in a difficulty in adjusting the building’s inside temperature. The idea was to adjust the water flow so that the temperature inside the building during ”steady state 2”

was equivalent in all cases, see section 3.2.2. As a result of these difficulties, it was decided that the flow was fixated at 0.005 l/s, which was the maximum flow for the used rotameter. The results however are not futile. In cases 2.1-2.3 the indoor temperature during steady state 2 is reduced for every case, even though the cooling power of the cooling system is also reduced. Since the energy of the

”sun” striking the building and the ambient temperature is the same for all cases, the reason for the decrease/increase in indoor temperature between the different cases is the coating of the roofs. The coating in case 2.3 reflects more than the coating in case 2.2, similar to the results of the same coatings in the computational simulations.

The red coating absorbs more of the incoming sunlight compared to the white coatings, which is why a white coating without MS performs better than the red coating, despite the fact that the red coating contains MS as a way of improving the optical properties of the coating.

5.3 SRI template values

Energy savings per year of 718 kW h for every increment of one SRI-unit may not seem as much, in fact, the savings per year is about 0.1%. However, when applying cool roof coating the SRI value of the roof increases by more than 1 unit (depending on existing roof coating). An increment of the roof’s SRI value from 70 (typical for a pale/light roof coating) to 109 (the SRI value for case 1.3) will result in yearly energy savings of 28.0 MW h, or 3.7% in savings per year.

5.4 Geographic location

The geographic placement of the warehouse has great influence in the amount of savings obtained by applying a cool roof coating. It goes without saying that the savings when applying cool roof coating to a warehouse are larger for a geographic placement with higher insolation than one with lesser insolation. In Appendix A

(42)

the results for the computational simulations with geographic placement of Tepic, Mexico is shown. The only parameter changed from the acquired results in section 4.1 is the geographic placement of the simulations.

The yearly savings in total heating and cooling amount between the worst case and the best case for the warehouse placed in Tepic, Mexico is 13.5%, compared to 6.24% savings in Houston, Texas.

Comparing the SRI template values between the two geographic placements also show a clear difference. The savings per SRI unit increment is 1252 kW h in Tepic, Mexico compared to 718 kW h in Houston, Texas.

(43)

6 Conclusion

A warehouse placed in Houston, Texas built to the building standard presented in the study shows a decrease of the annual amount of energy consumed by heating and cooling of 718 kW h for every unit-increment of the roof’s SRI value.

Due to the lack of the optical properties being precisely determined using spectrophotometry, the effect of the used coatings could not be accurately determined.

However, the results from both the computational simulations and the laboratory tests showed that the white coating containing MS showed the greatest reduction of the warehouse’s energy consumption and indoor temperature.

The geographic location has great influence on the effect of cool roof coatings.

It was seen that changing the location of the computer simulations from Houston, Texas (the American south) to Tepic, Mexico (central Mexico) could increase the annual savings (heating and cooling amount) from 6.24% to 13.5% for the same building.

Another conclusion is that the built up model warehouse used in the laboratory tests showed reasonable results as compared to the computational simulations.

The advantage of the practical set up compared to computer simulation is that there is no need to specify the exact properties of the used coating.

A general conclusion of using cool roof coatings is that it is a fast and easy way of saving energy by decreasing the amount of solar energy being absorbed by the building, without altering the structure of the building.

(44)

7 References

[1] NYC. ”Mayor Bloomberg And Former Vice President Gore Launch NYC Service Volunteer Program To Reduce Energy Usage And Green- house Gas Emissions By Coating Rooftops White”, [Online]. Avail- able: https://www1.nyc.gov/office-of-the-mayor/news/417-09/

mayor-bloomberg-former-vice-president-gore-launch-em-nyc/

-service-em-volunteer-program-to. [Accessed: September 4th, 2018.]

[2] O, Sandin et al. ”Optical and Mechanical Properties of Cool Roof Paint Containing Hollow Thermoplastic Microspheres”, 2013. [Online]. Available:

http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-206056. [Accessed:

October 15th, 2018.]

[3] AkzoNobel. ”Expancel for cool roof coatings”, [Online]. Available:

https://expancel.akzonobel.com/applications/cool-roof-coatings/. [Accessed: September 7th, 2018.]

[4] L. S. Rose, H. Akbari, H. Taha. ”Characterizing the Fabric of the Urban Envi- ronment: A Case Study of Greater Houston, Texas”, 2003. [Online]. Available:

https://escholarship.org/uc/item/3ww7p48f#main. [Accessed: Septem- ber 12th, 2018.]

[5] S. Davidson, ”SkyCool – Extraordinary Paint on a Hot Tin Roof”, Na- tional Year of the Built Environment, Melbourne, Apr-Jun 2004. [On- line]. Available: http://www.ecosmagazine.com/?act=view_file&file_

id=EC119p12.pdf. [Accessed: December 6th, 2018.]

[6] SkyCool. ”SkyCool Cool Roofs”, [Online]. Available: https://www.skycool.

net.au/. [Accessed: December 6th, 2018.]

[7] Airport Technology. ”SkyCool - Airport Tecknology”, [Online]. Avail- able: https://www.airport-technology.com/contractors/terminal/

skycool/#company-details. [Accessed: December 8th, 2018.]

[8] SkyCool. ”SkyCool on Warehouses”, [Online]. Available: http://www.

skycool.com.au/warehouses.htm. [Accessed: December 6th, 2018.]

[9] T. Gulliksson. ”Jämförelse av energiberäkningsprogram för byggnader”, 2015, Umeå Universitet. [Online]. Available: http://www.diva-portal.org.

proxy.ub.umu.se/smash/record.jsf?pid=diva2%3A816557&dswid=7131. [Accessed: September 28th, 2018.]

(45)

[10] Y. A. Cengel, A. J. Ghajar. ”Heat and Mass Transfer - Fundementals and Applications”, Edition 4, 2011, New York. ISBN: 978-007-131112-0.

[11] G. B. Smith, C. G. Granqvist. ”Green Nanotechnology - Solutions for Sus- tainability and Energy in the Built Environment”, CRC PRESS, 2011. ISBN:

978-1-4200-8532-7.

[12] Infinity-Theory. ”The Electromagnetic Spectrum”, [Online]. Available:

http://www.infinity-theory.com/en/science/Main_pages/The_

Electromagnetic_Spectrum. [Accessed: October 29th, 2018.]

[13] Energy Education. ”Insolation”, [Online]. Available: https://

energyeducation.ca/encyclopedia/Insolation. [Accessed: October 29th, 2018.]

[14] Encyclopedia Britannica. ”Solar Constant”, [Online]. Available: https://

www.britannica.com/science/solar-constant. [Accessed: October 31st, 2018.]

[15] Wikipedia. ”Solar Irradiance”, [Online]. Available: https://en.wikipedia.

org/wiki/Solar_irradiance. [Accessed: October 31st, 2018.]

[16] The Economist. ”Light-bulb moment - Illumination”, [Online]. Avail- able: https://www.economist.com/science-and-technology/2016/01/

16/light-bulb-moment. [Accessed: November 6th, 2018.]

[17] C.G. Granqvist. ”Radiative heating and cooling with spectrally selective surfaces”, Applied Optics, 20, 2606-2615, 1981.

[18] The Heat Island Group. ”Lawrence Berkeley National Laboratory SRI Calculator”, [Online]. Available: https://www.usgbc.org/resources/

lawrence-berkeley-national-laboratory-sri-calculator. [Accessed:

November 27th, 2018.]

[19] M. Soleimani-Mohseni. ”Kurspärm - Värme- och ventilationsteknik, Energi i byggnader, VVS, fukt etc.” 2016, Umeå Universitet.

[20] U.S Department of Energy. ”Reference Buildings by Building Type: Ware- house”, [Online]. Available: https://www.energy.gov/eere/downloads/

reference-buildings-building-type-warehouse-0. [Accessed: October 2nd, 2018.]

[21] U.S Department of Energy. ”Achieving the 30% Goal: Energy and Cost Savings Analysis of ASHRAE Standard 90.1-2010”, [Online]. Avail- able: https://www.energycodes.gov/sites/default/files/documents/

(46)

BECP_Energy_Cost_Savings_STD2010_May2011_v00.pdf. [Accessed: Octo- ber 2nd, 2018.]

[22] Philips Lighting. ”R125 IR 375W E27 230-250V CL 1CT/10 In- fraRed Industrial Heat Incandescent”, [Online]. Available: http://www.

lighting.philips.com/main/prof/conventional-lamps-and-tubes/

special-lamps/infrared/infrared-industrial-heat-incandescent/

923223543807_EU/product. [Accessed: September 27th, 2018.]

[23] Infrared heatlamps/healthcare. ”Incandescent reflector”, [Online]. Available:

http://www.infraphil.info/Philips_Infraphil-PAR38E.pdf. [Accessed:

September 27th, 2018.]

[24] Corsair. ”Hydro Series H60 High Performance Liquid CPU Cooler”, [On- line]. Available: https://www.corsair.com/us/en/Categories/Products/

Liquid-Cooling/Single-Radiator-Liquid-Coolers/Hydro-Series%E2%

84%A2-H60-High-Performance-Liquid-CPU-Cooler/p/CW-9060007-WW. [Accessed: October 30th, 2018.]

[25] Kjell & Company. ”Chassifläkt 2-pin”, [Online]. Available: https:

//www.kjell.com/se/sortiment/dator-natverk/datorkomponenter/

kylning/chassiflaktar/chassiflakt-2-pin-p98034. [Accessed: October 30th, 2018.]

[26] Kytola Instruments. ”Flödesmätare Modell E”, [Online]. Available: https://

www.kytola.com/sv/produkter/flodesmatare/flodesmatare/modell-e. [Accessed: October 30th, 2018.]

[27] T. Graham Carter. ”Issues and Solutions To More Realistically Simulate Con- ventional and Cool Roofs”, Proceedings of Building Simulation: 12th Confer- ence of International Building Performance Simulation Association (IBPSA), Sydney. 2011.

(47)

8 Appendices

A Computational simulations, Mexico

The computational simulations for the same warehouse, now located in Tepic, Mexico is presented below to show the effect of different climate zones and locations.

A.1 Case 1.1, Mexico

The results for the reference case without any coating (optical properties of bare stainless steel - emissivity = 0.90, reflectance = 0.378 and specularity = 0.101) is presented in Tab.14.

Table 14– Delivered energy to warehouse with cooling system.

Amount of cooling 616 997 139.4 223.4

Amount of heating 1 423 0.3 35.55

Total, heating and cooling 618 420 139.7

As expected the amount of cooling is substantially higher than the amount of heating for a building with low amount of insulation positioned in Tepic, Mexico (high insolation).

A.2 Case 1.2, Mexico

The results of the second case, with a white roof coating not containing MS (emissivity = 0.90, reflectance = 0.82) are showed in Tab.15.

(48)

A.3 Case 1.3, Mexico

The results of the third case with a white roof coating containing MS (emissivity

= 0.90, reflectance = 0.86) are showed in Tab.16.

A.4 Case 1.4, Mexico

The results for the red roof coating containing MS (emissivity = 0.90, reflectance

= 0.37) is shown in Tab.17 below.

A.5 Comparison of the results from the computational sim- ulation, Mexico

The lowest yearly amount of heating and cooling was seen in case 1.3 with the amount of 534 647 kW h (the white roof coating containing MS). The highest yearly amount of heating and cooling was produced in case 1.1 and case 1.4, resulting in approximately 618 MW h (no coating). The difference between the best and worst case is 83 459 kW h, a reduction of the total yearly heating and cooling amount of 13.5%.

(49)

A.6 SRI template values, Mexico

The annual amount of cooling and heating for the warehouse placed in Tepic, Mexico is shown in Fig.20-21 for SRI values from 70 to 125.

Figure 20– Annual amount of cooling energy as a function of SRI for warehouse placed in Tepic, Mexico.

The decrease of annual cooling amount per one increment of SRI value is 1263 kW h.

Figure 21– Annual amount of heating energy as a function of SRI for warehouse placed in Tepic, Mexico.

The increment of annual heating amount per one increment of SRI value is about 11 kW h, meaning that for every increment of one SRI unit the total annual amount of heating and cooling is reduced by 1252 kW h.

COOL ROOF COATINGS ON INDUSTRIAL BUILDINGS

INDUSTRIAL BUILDINGS

An Energy Study of Reflective Coatings

Contents

1 Introduction

1.1 Background

1.2 Real life case studies of cool roof coatings

1.3 Aim

1.4 Limitations

2 Theory

2.1 IDA-ICE

2.2 Heat transfer

2.3 Electromagnetic spectrum

2.4 Solar radiation

2.5 Cool roofs

2.6 Coatings

2.7 Expancel

microspheres

2.8 Building heat balance

3 Methodology

3.1 Computational simulation

3.2 Laboratory tests

3.3 SRI template values

4 Results

4.1 Computational simulation

4.2 Laboratory tests

4.3 SRI template values

5 Discussion

5.1 Computational simulation

5.2 Laboratory tests

5.3 SRI template values

5.4 Geographic location

6 Conclusion

7 References

8 Appendices

A Computational simulations, Mexico

A.1 Case 1.1, Mexico

A.2 Case 1.2, Mexico

A.3 Case 1.3, Mexico

A.4 Case 1.4, Mexico

A.5 Comparison of the results from the computational sim- ulation, Mexico

A.6 SRI template values, Mexico