9285 Buro Patterns st20 AW - Designing Buildings · 2013-08-07 · energy, see figure 1(opposite)....

PatternsGroundSourceEnergy

Introduction

Welcome to this edition of Patterns,which focuses on ground energy.

For centuries many a fine wine hasbenefited from the stable environmentafforded by a suitable cellar or cave,taking advantage of the great thermalcapacity and inertia of the ground. With the advent of HVAC systems we have replicated this stable internalenvironment in our homes, work andleisure spaces, while interestingly thewine stays by and large underground!

Today, in a time where energyconservation is king, we find ourselveslooking more closely at the temperaturestability and thermal capacity of theground as a source of free cooling and heating for our buildings.

Over the years we have developed a variety of techniques for coupling our buildings with the ground andexchanging heat back and forthbetween the two. The technology can take a number of forms using either water or air based systems.

As the drive to include a proportion ofrenewable energy into many public andprivate buildings increases, many see‘ground energy’ as a viable means ofenergy conservation. At Buro Happoldwe have gathered a wealth ofexperience in these exciting techniquesand have used the ‘Geekfest’ (see definition opposite) model to share and record this experience.

This sharing of experience is importantas the process of design in this area isevolving fast. The thermal behaviour ofthe ground is a complex science whichmust be analysed in conjunction withthe varying cooling and heating loads of the building and the characteristics of the HVAC system performance.

In the world of building design, prototypesare few and far between and solutions in a field such as this must be refined usingsolutions that at first are conservative andthen are honed project by project as actualperformance data is gathered. Theresimply is no substitute for experience!

We have split our discussion within Patternsinto two areas, water and air basedsolutions – with the exception of ScottBaird’s essay on the Burns Museum projectwhich incorporates both earth tubes onthe air inlet and a water based groundloop heat exchanger for heating/cooling.

Air based explores the use of thermallabyrinths and earth tubes.

Water based includes open loop ground water schemes and closed loop shallow (ground mat) and mediumdepth (piled) systems.

With water based systems used inconjunction with heat pumps for heating,or direct for cooling systems, it is alsoimportant to consider how such sourcesof heating and cooling are integratedwith appropriate heating and coolingsystems. This area is covered within a number of the essays contained in this edition of Patterns.

The topic of ground source heating andcooling is a fascinating one, being all themore rewarding as it fuses together BuroHappold’s collective skills and experiencein the fields of building services systemdesign and analysis, ground engineering,sustainability and alternative technologies(SAT) and computational simulation andanalysis (CoSA).

Geekfest:

A gathering of those engineers andconsultants within Buro Happold whohave specific experience in an area ofgrowing importance to the firm and ourindustry. Held as a colloquium, all presentmust set out and discuss their relevantexperience in terms of analysis, design,procurement, construction and post-occupancy evaluation. There are nospectators at a fest! By so doing, thegathering serves to share knowledge,educate others and close the all-importantfeedback loop to refine our analysis anddesign skills within a particular field. The fest usually includes the opportunity to ‘workshop’ live projects at inception/feasibility stage to explore whether suchtechnology can be successfully deployed.

1

PatternsGroundSourceEnergy

Ground energy options James Dickinson 2

Closing the loop Alan Shepherd 5

Massive Milanese scheme uses Steve Williamson 11ground water cooling/heating

To bore or not to bore? Mark Owen 15

Paper merchant pushes for Edith Blennerhassett 19water/water heat pump

Complementary technologies Jason Gardner 21

Ground source heating James Dickinson 23and cooling study

Double slinky in County Kerry Brian Doran 26

Finding the way through Mike Entwisle 29labyrinths and earth tubes

Dynamic modelling benefits Daniel Knott 34of Jennie Lee labyrinths

A question of coupling David Warwick 37

Double couple at Scott Baird 39Robert Burns Museum

Our authors 45

2

Ground energy options

When choosing liquid-basedsystems for ground source energythere are two main options – openor closed loop, and some variations.James Dickinson looks at theadvantages of the approach and the key actions required.

Throughout the year the ground absorbssolar energy and below a depth ofapproximately 7-10m the temperatureremains fairly constant at the meanambient air temperature regardless of the time of year. Depending on thelocation and depth this temperature canvary, typically, from 7-13ºC in the UK.In general, the use of ground energy toprovide heating and cooling in buildingrequires equipment (heat pumps) to upgrade the temperature of thesource temperature to a more usefultemperature level using additionalenergy, see figure 1 (opposite).

The energy can be transferred to this equipment using a ground heatexchanger (closed loop systems). This new science usually comprises a number of pipe loops, vertical orhorizontal, with a primary processmedium of water, or more normally a glycol solution which eliminates the possibility of freezing within theapplication’s seasonal temperaturerange. The alternative is to abstract and discharge ground water (open loop systems) from an aquifer beneaththe building.

In the case of the closed loop systemthe energy in the ground is, if the groundloop is sized appropriately, replenishedby solar irradiation, rain and, sometimes,for deeper vertical collector systems,underground water flow. With open loopsystems it is necessary to consider thesustainable yield available from the wells.

Variations of ground energy

Horizontal – closed loop

With this variation the energy or heat istransferred to the building using a seriesof ground collectors, laid horizontally at a depth of 1.5-2m, see figure 2(above). Each pipe run should be limitedto 100m to avoid the need for morepowerful circulation pumps. Pipe runswould normally be the same length to guarantee similar flow conditions,pressure drops and to ensure an even heat extraction from the ground.

The useable amount of heat or energy is dependent on the following:

■ Solar irradiation for the specific area■ Moisture content■ Soil type■ Size of pores.

Extraction rates are generally in the orderof between 10 W/m2 for dry sandy soil,to over 30 W/m2 for wetter loamy soils.Relatively inexpensive earth movingequipment is required for installation,although costs increase with greaterdepths. This type of collector is generallyused for applications with lower power outputs where there is a largeundeveloped area that is easy to excavate.

Vertical (probe) – closed loop

A vertical closed-loop system utilisesvertical ground heat exchangers orprobes that are inserted into speciallydrilled boreholes up to depths of 150m,see figure 3 (on page 3).

Extraction rates generally vary between20 W/m for loose dry substrate to~80W/m for damper sandstones,granites and basalts.

Figure 1 Heat pump technology

Figure 2 Horizontal closed loopsystem (Viessmann)

Free energyfrom the ground

Header duct with brine distributor

Brine distributor (flow)Brine distributor (return)

Ground collector

Electricity

Heatingor

Cooling

3

The useable heat or energy is dependenton similar factors to the horizontalsystem although more specialistgeological analysis is generally needed.Deeper test-bores can ascertain the typeand depth of each soil/rock layer, theheat transfer potential for the differentlayers over the length of the borehole,the presence and height of water tableand underground water flow.

Due to the requirement for a test borethis type of system lends itself to largerapplications where the initial testingcosts can be justified. The datagathered help to reduce risk during the design stage as non-optimum sizing has serious cost implications.

Vertical – open loop

In this variation ground water isextracted direct from the undergroundwater aquifer, eliminating the need for a closed loop ground heat exchanger.The used cooled or heated water canthen be returned to the ground via a return well, see figure 4.

Prior to the consideration of such aconfiguration it is necessary to contactthe Environmental Agency, initially togain consent for a pumping test, andthen for a final abstraction licence for a pumping test, and then for dischargeconsent. There is an additionalrequirement to consider the waterquality of the water source as this canhave an adverse effect on the materialsused within the heat exchanger.

Figure 3 (top) Vertical closed loop system

Figure 4 Vertical open loop system

Heating

Cooling

Ground at 9-12°Cthroughout the year

Heat Pump/Heat Exchanger

Heating

Cooling

Groundwater at 9-12°Cthroughout the year

Heat Pump/Heat Exchanger

4

Generic guidelines for ground energy systems

1 Start considering the technology at an early stage in the project.

Complete a ground energy desktop survey to establish the suitability of the geology and hydrogeology underneath the site to different types of ground energy systems. Suitable sources include the British Geological Survey and sitespecific Geotechnical Investigation reports.

Establish the spatial limitations around the building.

What is the indicative foundation design and is it suitable to act as part of the ground energy heat exchanger?

2 Optimise the heating and cooling building circuits.

Use high temperature cooling where possible (eg chilled beams and air based systems with over sized heat exchangers).

Use low temperature heat emitters (large radiators, underfloor heating and air based systems with over sized heat exchangers).

Simultaneous heating and cooling can be provided from the same heat pump unit.

Closed loop dos

1 For larger commercial systems, ie greater than ~100kW, a thermal conductivity test is advised to confirm the insitu thermal properties.

2 Carry out a desktop simulation using recognised software to ensure the long term performance can be guaranteed.

3 Ensure boreholes are spaced adequately to reduce thermal interference.

4 Try to balance heat abstraction and rejection to the ground.

5 Consider using less expensive conventional plant for infrequent heating and cooling loads and/or higher relative seasonal heating and cooling loads.

Open loop dos

1 For almost all open loop systems Environment Agency (EA) approval is needed for both abstraction and discharge of ground or surface water.

2 A pumping test will be needed to confirm the yield and to get permission from the EA to abstract and discharge a specified volume of water per hour/day/year.

3 Start the process to obtain an abstraction licence and discharge consent as early as possible (this process can take from eight to nine months in the UK).

Feasibility and Evaluation

5

In this report Alan Shepherddescribes the need for dynamicthermal modelling of Closed LoopGeothermal Heat Pump (CLGHP)systems. He looks at the variousCLGHP system permutations andhow they are applied and outlinesthe relative merits of the analysistools available.

Why dynamic thermal modelling isso vital to CLGHP system design

A ‘traditional’ gas fired boiler and vapourcompression chiller based HVAC systemdesign can be sized with a reasonablelevel of confidence simply by determiningthe peak heating and cooling load of agiven building. This is not the case forCLGHP systems. The heat source andsink for a CLGHP system is the rock andearth that surrounds the ground loops.Over the course of a year the groundtemperature varies sinusoidally as heat is either rejected into the ground (coolingoperation) or abstracted from the ground(heating operation).

The operating efficiency of a heat pumpdepends largely upon the temperaturedifferential between the source-sideentering water temperature from theground loops and the system-side(CHW/LTHW) water temperature. The smaller the temperature differential,the more efficiently the heat pump willoperate. To understand the seasonalefficiency of a CLGHP system it istherefore necessary to be able tosimulate the seasonal variation in the ground temperature surrounding the geothermal loops.

Furthermore, it is important that thereis a reasonable balance between the total annual heat energy rejectedinto the ground and that abstracted. A significant imbalance will result in thegradual increase in ground temperatureover successive years in the case of a cooling dominated load profile, or a gradual decrease in temperature for a heating dominated load profile.

An increase in ground temperature oversuccessive years will eventually result ina drop in heat pump cooling efficiency(as the differential between geothermalwater and CHW temperature increases)as well as a reduction in heat pumpcooling capacity and vice versa forheating operation.

CLGHP simulation process

The following steps describe amethodology that was used by Buro Happold to develop an in-houseCLGHP analysis tool. The methodologylends insight into the factors that affectCLGHP performance and also exposessome of the internal workings ofalternative commercially availableCLGHP analysis tools on the market.

Step 1: Generating annual heating and cooling load profiles

The first step in simulating theperformance of a CLGHP installation is to establish the annual heating and cooling load profiles. Derivingaccurate annual heating and coolingload profiles requires the use ofsophisticated simulation tools used in conjunction with realistic estimates of dynamic occupancy, lighting andequipment loads.

For peak load analysis occupancy,lighting and equipment loads are oftenassumed to be at a constant peak – this is of course highly unrealistic and,if used for annual energy analysis, will result in a gross over-estimation of cooling energy consumption and an equal under-estimation in heating.

Engineers should exercise caution in the use of the more basic DynamicThermal Modeling (DTM) software that is available on the market. Figure 1(above) shows a 3-dimensional renderingof a building model generated using IESDTM software.

It is also important that the DTMaccurately models the HVAC systemand controls. The use of generic systemtemplates can result in significantinaccuracies and should be used with caution. Figure 2 is an excerpt from a system model generated inApacheHVAC that incorporates anair-side economiser, cooling coil withwrap-around heat pipe, evaporativehumidifier with face and bypassdampers, along with all associatedcontrols. Figure 3 graphically displaysthe full hour by hour heating and coolingload results calculated by IES for thebuilding and system model shown infigures 1 and 2.

Closing the loop

Figure 1 3D rendering of an IES building model

6

Step 2: Calculating the abstractionand rejection of heat between theheat pumps and the bore field

Having derived the annual heating and cooling load profiles of the HVACsystem, the next stage in the analysis is to calculate the abstraction andrejection of heat from and to thegeothermal bore field.

When in heating mode the heatabstracted from the ground (QAbstraction)is calculated as follows:

QAbstraction = QHeating – QCompressor

where QCompressor = QHeatingCOPHeating

When in cooling mode the heat rejectedto the ground (QRejection) is calculated asfollows:

QRejection = QCooling – QCompressor

where QCompressor = QCoolingCOPCooling

The calculation of the heat ofabstraction and rejection creates a‘chicken and egg’ situation as it requiresthat the operating COP of the heatpumps be known. However, the heatpump operating COP can only bedetermined from knowledge of thegeothermal bore field temperaturewhich, in turn, is calculated from ratesof heat abstraction and rejection. The problem is circular. In order tobreak this stalemate it is necessary to make an initial estimate of heat pumpoperating COP. Figure 4 above showsthe annual heating and cooling loadprofile (from figure 3) displayed inmonthly ‘bins’ for clarity. Figure 5 shows the heat of abstraction and rejection.

Noticeably apparent when comparingthe two graphs is that the relativelybalanced heating and cooling loadprofiles displayed in figure 4 actuallyresult in an imbalance in heat exchangewith the bore field, with the heat ofrejection dominating over the heat ofabstraction. The simple reason for this is the fact that the heat emitted by thecompressor assists the heat pumpwhen in heating mode, but hampersperformance in cooling mode.

Figure 2 An excerpt from an HVAC system simulation model using IES ApacheHVAC software

Figure 3 Annual heating and cooling load profiles – hourly data

120,000

100,000

80,000

60,000

40,000

0

20,000

Jan Feb Mar April May June July Aug Sept Oct Nov Dec

Ene

rgy

(kW

h)

Heating Load (kWh)Cooling Load (kWh)

Figure 4 Annual heating and cooling load profile – monthly data

140,000

120,000

100,000

80,000

60,000

20,000

0

40,000


Ene

rgy

(kW

h)

Heating Abstraction (kWh)Heat Rejection (kWh)

Figure 5 Annual heat of abstraction/rejection profile – monthly data

7

Step 3: Sizing the geothermal bore field

Having calculated the annual heat ofabstraction and rejection, the annualvariation in the temperature of thegeothermal bore field can be determined.The relative capabilities of availableCLGHP analysis tools will be discussedlater in the report; however, for thisparticular analysis GS2000 was used.The input data required by GS2000and other CLGHP sizing software is broadly similar; requiring the user to define the following:

■ Bore field configuration (vertical, horizontal etc)

■ Ground temperature properties■ Ground layer description

(depth, material properties etc)■ Ground heat exchanger

pipe properties■ Geothermal circulation fluid

properties (ethylene/propylene glycol etc)

■ Heat pump details (peak capacity, COP etc)

■ Abstraction/rejection heat loads.

There are no hard and fast rules thatgovern the sizing of a geothermal borefield, although heat pump manufacturersrecommend that bore field leaving watertemperature should not be allowed tostray outside a minimum of 5°C and a maximum of 32°C.

For the annual abstraction and rejectionheat loads displayed in figure 5, andwith minimum and maximum leavingwater sizing limits of 8°C and 30°Crespectively, GS2000 calculated anannual leaving water temperature profileas shown in figure 6 and a requiredborehole length of 9097m (72 bores,each 128m deep).

Step 4: Establishing heat pump COP and peak capacity

Having established the annualgeothermal bore field leaving watertemperature profile, it is possible todetermine the annual variation in heat pump operating COP as well as the variation in peak heat and cooling capacity. Figures 7 through to 10 (right) were derived usingmanufacturer’s data for a ClimateMaster WW360 water/water heat pump. Figures 7 and 8 show the relationship between enteringgeothermal water temperature (sourcetemperature) and COP. Figures 9and 10 show the relationship betweenentering geothermal water temperatureand peak heating/cooling capacity.

Using the expressions given in figures 7 and 8 in conjunction with the annualbore field leaving water temperatureprofile given in figure 6 it is now possibleto derive the annual variation in heatingand cooling COP as shown in figure 11on page 8.

Using the expressions given in figures 9 and 10 in conjunction with the annualbore field leaving water temperatureprofile also allows us to derive the annualvariation in peak heating and coolingcapacity of an individual heat pump as shown in figure 12 on page 8. Thisallows us to determine the maximumnumber of heat pumps required and also how the number of on-line heatpumps varies over time – a key factor in determining the parasitic pump powerassociated with the system.

25

20

15

10

5

0Jan Feb Mar April May June July Aug Sept Oct Nov Dec

Tem

per

atur

e (o

C)

Loop Temperature (oC)

Figure 6 Annual geothermal bore field leaving watertemperature profile – monthly data

6

5

4

3

2

0 5 10 15 20 25 30 35 400

1

HE

AT

ING

CO

P

y = 0.0002x2 + 0.063x + 3.0875

Source Temperature (oC)

Figure 7 Relationship between heating COP and entering source water temperature

89

10

76

54

0 5 10 15 20 25 30 35 40

210

3Co

olin

g C

OP

y = 0.0016x2 - 0.2163x + 8.9792


Figure 8 Relationship between cooling COP and entering source water temperature

140

130

120

110

100

0 5 10 15 20 25 30 35 40

80

70

60

90

Hea

ting

Out

put

(kW

)

y = 69.908e0.0193x


Figure 9 Relationship between heating capacity and entering source water temperature

95

90

85

80

75

0 5 10 15 20 25 30 35 4065

70Co

olin

g O

utp

ut (k

W)

y = 0.0013x2 - 0.582x + 92.523


Figure 10 Relationship between cooling capacityand entering source water temperature

8

In Step 2 of the analysis methodologythe necessity to estimate initial COP values was discussed. Havingcompleted the first iteration of theanalysis and derived a complete annualvariation in COP (figure 11), these values should now be plugged back intoStep 2 of the analysis in order to obtainmore accurate heat of abstraction andrejection figures. This iterative procedureshould be repeated until such time that the COP values entered in Step 2match those calculated in Step 4.

Step 5: Parasitic Loads

When conducting a comparativeanalysis of potential heating and cooling plant options it is essential that the parasitic loads (pump power,cooling tower fans etc) associated witheach option are accounted. The BuroHappold in-house CLGHP analysis toolincorporates a parasitic load calculationspreadsheet. Depending upon whethera constant or variable speed pumpingstrategy is implemented the contributionof parasitic loads to the overall CLGHPsystem energy consumption can besignificant. Figure 13 opposite shows the annual variation in Geothermal, CHWand LTHW pump energy consumption.

Step 6: Comparative Analysis

In order to gain some relativeperspective on the performance of aCLGHP it is, of course, necessary toobtain comparative data for alternateheating and cooling system options.Ideally that data should be derived from the same annual heating andcooling loads used in the analysis of the CLGHP. The Buro Happold in-houseanalysis tool includes two alternatesystem options; air cooled chiller and water cooled chiller. Either optioncan be coupled with a gas, oil or LPGfired boiler. The results of a typicalanalysis are shown in figures 14through to 16 on page 9.

Imbalanced annual heating and cooling loads

The simulation process described in the previous section uses as its examplea somewhat idealised scenario in whichthe annual heat of abstraction is almostexactly equal to the annual heat ofrejection. This results in an annualgeothermal leaving water temperatureprofile that starts on 1 January at 8°Cand ends on 31 December at the same8°C (see figure 6). It is infrequently the case that a building will exhibit such a fortuitously balanced heatingand cooling load profile.

Heating dominated loads

Figure 17 on page 10 shows an annualheating and cooling load profile that is heavily heating dominated. Thisimbalance will result in a far greaterquantity of heat being abstracted fromthe ground during the heating seasonthan is replenished during cooling. As is clearly shown in figure 18 thisresults, over successive years, in agradual drop in the temperature of theearth surrounding the geothermal bores.

The drop in leaving water temperaturefrom the bore field will be accompaniedby a gradual drop in heating COP (see figure 19) and a consequentialincrease in operating costs. The peakheating capacity of the heat pumps will also gradually fall.

The drop in earth temperature willactually improve the cooling COP of the heat pumps. However, sincethe load profile is so heavily heatingdominated, the reduction in annualcooling energy consumption is relativelyinsignificant compared to the increase in heating energy.

8

7

6

5

4

3

2

1


CO

P

Heating COPCooling COP

Figure 11 Annual variation inheat pump COP – monthly data

105

100

95

90

85

80

75

70

60

65


Out

put

(kW

)

Max. Heating Output (kW)

Max. Cooling Output (kW)

2500

2000

1500

1000

500


Ene

rgy

(kW

h)

Geothermal Primary Pump

CHW Primary Pump

LTHW Primary Pump

Figure 13 Annual circulation pump energyconsumption – monthly data

Figure 12 Annual variation in peak heating and cooling capacity – monthly data

9

Cooling Dominated Loads

The impacts of a heavily cooling-dominatedload profile are fundamentally similar but in reverse. There is a gradual rise in earthtemperature over successive years, andconsequential reduction in cooling COP.The increase in earth temperature andthe reduction in COP are non-linear. Asthe temperature of the ground increasesthe heat loss to its surroundings alsoincreases. Furthermore, at highergeothermal leaving water temperaturesthe quantity of heat rejected into the

ground will begin to level off as the heatpumps are no longer able to meet peakcooling load requirements. The result of these phenomena is that the meanannual earth temperature will eventuallyreach a balance point: in one recentsimulation this was reached afterapproximately 12 years of operation.

Despite the negative impacts describedabove, an imbalanced load profile need not preclude the use of a CLGHPsystem. Described under the followingheadings are various mitigating measuresthat can be taken when dealing withimbalanced loads.

Increase the size of the geothermalbore field

A ‘solution’ that is often proposed whenfaced with the problem of imbalancedheating and cooling load profiles is toincrease the size of the geothermal borefield. This is a costly option and not themost effective. An increase in the size of the bore field does nothing to addressthe imbalance in load, it merely slowsdown the inevitable increase/decrease in earth temperature.

Load shifting by modifying the MEP system design

A far more effective approach is topurposefully manipulate the heating and cooling load profiles by modifyingthe MEP system design.

For example, a cooling dominated loadprofile can be brought back into balance,at least in some part, by making adesign change from electric resistancehumidifiers to evaporative type.

Bivalent systems – load side

Imbalances in annual heating and coolingload can also be addressed by sizingthe CLGHP system to meet a base load,while top-up boilers and/or chillers areused to meet peak load requirements.This dual approach is commonly referred to as a ‘bivalent system’.

Aside from load balancing purposes abivalent system design approach oftenresults in the optimum payback periodfor a CLGHP installation. The problemwith a bivalent approach is that itrequires the reintroduction of equipmentsuch as chillers, cooling towers, fluesetc, the elimination of which may havebeen one of the drivers for selecting a CLGHP system in the first place.

Bivalent system – source side

An alternative bivalent system approachto the problem of imbalanced loads is to target the source side of heat pumpsrather than the load side. This meansthat the heat pumps provide 100% of the heating and cooling load but that the geothermal bore field issupplemented in dealing with the heat of abstraction and rejection.

An example of this approach (shown in figure 20) is for a cooling dominated load profile the heats of abstraction andrejection from and to the bore field canbe put into balance by rejecting a portionof the heat via the cooling tower.

$4,000

$3,500

$3,000

$2,500

$2,000

$1,500

$1,000

$500

$0Jan Feb Mar April May June July Aug Sept Oct Nov Dec

Heat PumpWater Cooled ChillerAir Cooled Chiller

Figure 15 Comparative system cooling energy costprofile – monthly data

$6,000

$5,000

$4,000

$3,000

$2,000

$1,000

$0Jan Feb Mar April May June July Aug Sept Oct Nov Dec

Heat PumpGas Boiler

Figure 14 Comparative system heating energy costprofile – monthly data

Figure 16 The three systems – annual energy cost comparison

10

In addition to summer time heat rejectionoperation, a cooling tower can help tobalance a cooling dominated load profileby operating during winter. Running the tower during the winter effectivelyimposes a ‘false’ heating load on the borefield, pre-cooling the earth surroundingthe geothermal bores and therebyreducing temperature rise during summer.

For heating-dominated load profiles asimilar balancing effect can be achievedusing a bivalent system approachwhereby solar thermal collectors imposea false cooling load during summer.

Overview of CLGHP analysis tools

Buro Happold currently uses thefollowing CLGHP modeling tools:

■ In-house spreadsheet in conjunctionwith GS2000

■ GLHEpro■ Trnsys 16.

Figure 19 (left) Geothermal heat pump COP heating – 10 year simulation

Figure 20 (above) Bivalent CLGHP system with supplementary heat rejection

Water-to-WaterHeat Pump

Plate HeatExchanger

Ground Heat Exchanger

Diverter Valve

Cooling Tower

Tower Circuit Pump

VentilationAir

Water Source Heat Pump Units

Building

GroundLoop Pump

Alan Shepherd Figure 20

Figure 17 Heating-dominated annual load profile – hourly data

Figure 18 Geothermal inlet and outlet water temperature – 10 year simulation

11

Massive Milanese scheme uses ground water cooling/heating

Buro Happold is providing M&E engineering design on thegroundbreaking Garibaldi project in Milan. Steve Williamson explainshow this scheme has become aprime example of a large scalecommercial project responding todemands for low carbon buildings.

With around 200,000m2 of mixed use,predominantly office space, the Garibaldiproject meets all the requirements of a world class commercial centre.Moreover, despite value-adding featuressuch as highly-glazed facades, full airconditioning and a sound commercialapproach, the project aims to achieve a gold LEED rating.

The buildings have a combined peakcooling demand of 18MW along with12MW of heating demand, all of whichwill be provided by an open loop groundwater heat pump system. Combinedwith Varasene, its sister project of asimilar scale across the road, and by the same developer, it is believed thatthis is one of the largest ground waterschemes in the world.

Alpine sourced ground water

Early design studies identified an idealopportunity to make use of Milan’s coolground water fed from Alpine melt wateren route to the Mediterranean. This waswell received by the local authorities,who are most concerned with theincreasing smog created by local gasemissions from inner city buildings.

The key to making the scheme viable is the proximity of the Martesana Riverthat passes underground through the site. The combination of heatpumps and open well ground waterdischarging into the river actually costs less than a conventional chiller,boiler and cooling tower combination.

The project has been designed with 12 boreholes, each capable of providing35l/s (litres/second) of ground water. The buildings are provided with reversibleheat pumps to generate both heatingand cooling.

Mechanical plant

The principle of ground water cooling is a simple, low energy alternative toconventional heating and cooling plantutilising boilers and cooling towers. The system takes advantage ofconstant temperature water (circa 12ºC)from deep boreholes, which is used topre-cool air into air handling units, andto provide heat rejection for the chillers.The use of ground water for pre-coolingof ventilation air will be a direct energysaving, and using the ground water forheat rejection will improve the seasonalcoefficient of performance (COP) of theheat pumps to around 6.5, thus givingfurther energy savings in both heatingand cooling. A schematic of this systemis indicated in figure 1.

The primary equipment deemed mostsuitable to take advantage of the groundwater is a refrigerator/heat pump suchas the ‘frigorifero polivalente’.

A simple schematic is shown in figure 2. These heat pumps willsimultaneously produce hot water(LTHW at 50°C) and chilled water (at 7°C), and are able to utilise theground water for heat rejection.

As a result of the heating and cooling of the building, in winter the groundwater will be cooled by the heat pumpsfrom 15°C to approximately 7°C. Insummer, it will be heated from 15°C to approximately 30°C. After passingthrough the heat pumps, this rejectionwater will then be pumped locally into the Martesena River. The quantity of water must be such that thetemperature of the river is not increasedby more than 3°C, measured from apoint 5m upstream and 5m downstreamfrom the area of discharge.

The system produces hot and chilledwater with high efficiency, minimumnoise and without local CO2 emissions.

The Garibaldi development, Milan

12

It has been agreed that the building willbe designed with spatial allowance andfacilities for conventional plant (boilersand cooling towers) to be installed in the future. Correctly sized pipes will beinstalled in the risers, allowing for easyfuture connection from roof coolingtowers and boilers.

Ground water extraction

Ground water will be extracted from the 12 wells or boreholes located withinthe basement. The preliminary proposedpositions of these are indicated in figure 3. Each well will have a nominalflow of 35l/s and will be served by twopumps (duty and standby) that supply a basement-wide distribution main,providing a maximum flow of 420l/s.

The ground water flow rate to the heatpump will vary between the minimumwater flow required by the heat pumpsand maximum demand in peak summer.Therefore, the ground water distributionmain will be a variable volume pumpingsystem, to guarantee the minimumpumping energy and cost.

Water pumped from the wells will bemechanically filtered prior to passingthrough heat exchangers. Two heatexchangers will be installed for eachbuilding (100% duty and 100% standby)and will be accessible for cleaning.When a heat exchanger is shut downfor cleaning, the second heat exchangerwill provide all the necessary duty for full load operation. In this system,ground water never goes directly into the heat pump condenser/evaporator,thus preventing potential problems with respect to dirt and deposits.

In figures 4 and 5 the typical extractionwell with all necessary components is shown. Each well will be accessible from the top to provide maintenance andcontrol operations. Note that the positionof each well has been estimated with a minimum separation distance of 70m.

Figure 1 General ground water scheme

Figure 2 Frigorifero polivalente scheme

13

Well extraction will also relieve groundwater levels under the site, thus avoidingpotential problems generated by the waterbed level increase recentlyregistered in Milan.The waterbed level will be monitored by piezometers.

Ground water discharge

Ground water used by heat pumps willbe discharged in the Martesana Riverlocated under Via Melchiorre Gioia. TheMartesana River runs into the RedefossiRiver and then flows out of Milan to the south. The ground water is ’clean’and so may improve the water qualityin these rivers which provide irrigationwater for agriculture in the south Milan fields, where rice is grown.

The structural work associated with the ground water system will include a run-off pit which will enable the authorities tomeasure flows, and to gain water samplesfor laboratory analysis. Within this run-offpit, all pumped discharge pipes from eachbuilding will terminate, and accumulateddischarge water can then flow by gravityinto the river. A check valve will be installedto prevent a backflow of water from theriver in flood conditions. This system will

indicating the potential scale of floodsand their frequency during the year.

To ensure a robust solution for theGaribaldi site, an emergency system is proposed that will allow the system to work even if ground water cannot be discharged into the Martesana River.

Potential solutions include:

■ Injection wells■ A large volume tank to attenuate flow■ Using water from the river as a

heat sink instead of ground water during floods.

The best option in terms of reliability and feasibility is to utilise the injectionwells system. In studying this system(assuming the nominal water flow of each well for injection in the groundis 35l/s like the extraction well), fiveinjection wells are needed. Therefore,from the total 12 wells, six will be usedfor extraction and six for rejection.Hence, this system cannot discharge the maximum design ground water flowof 420l/s, but only half of it. However, it is most unlikely that the building plant will be required to operate at peak output(normal peak is high summer) during a Martesana flood (normally in winter ormid-season). The design must carefullyconsider the risk of ground water ‘shortcircuit’, to avoid extract water beingdischarged directly into an adjacent well used for abstraction.

It is also possible to take furthermitigating steps if the required power is likely to be greater than the 50%available. This includes programmingbuilding management systems toreduce demand by switching offsystems such as humidification/dehumidification, reducing externalairflow and other measures.

Figure 3 Extraction wells position

be designed to discharge at the maximumflow of 420l/s.

In all circumstances, according to ItalianLaw 152/99, the limit for the maximumincrease in water temperature in the riveris 3°C. Temperatures before and afterthe discharge point will be measured at a midpoint of the river, 5m before and 5m after the discharge point. This limit must be respected regardlessof the rate of flow in the Martesana.

Emergency discharge

The Martesana River comes from thewest of Milan and, before the Garibaldiarea, there is a confluence with theSeveso River. Due to the natural flow of water from the Seveso River the flowrate cannot be controlled. Therefore, in the position near the Garibaldi area,the Martesana will never be dry (theestimated minimum flow is 1m3/s).Therefore, flooding cannot be ruled out. To help improve the situation, the municipality is considering thecreation of an artificial river (canal) to help attenuate flood water at suchcritical times. However, there is noofficial data available from the authorities

14

Figure 4 Typical extraction well

Figure 5 Typical extraction well detail

Other uses

Ground water may also be consideredfor other applications in the locality. The most important one is irrigationwater for the campus (about 48,000m2

of green park near the Garibaldi area).The water may be stored in a tankduring the day and discharged on the campus overnight.

The proposed well positions are indicatedin figure 3. The authority responsible forground water extraction is the Provinciadi Milano. Our local engineering partners,Ariatta, had initial meetings with theauthority in September 2005, regardingmanagement of the discharge in the riverfrom the pollutants’ point of view andmanaging the extraction of water fromthe ground.

Approval process

It is the approval process which is often cited as the most difficult hurdlefor open loop ground water schemes, and Garibaldi was no different. Theprocess is concerned with two aspects;approval for the ground water extractionand approval to discharge into the river.

The extraction approval required aninitial request for permission to drill the wells. Once accepted, this aspectwas given a time limit of one year.

To help to understand the effects of our proposed extraction on the localwaterbed, it was necessary to undertakea mathematical simulation and desktopstudy. A laboratory analysis of groundwater quality extracted from the first test wells (there were three across thesite) was submitted, to ensure that the concentration of particulates wasacceptable for discharge into a river (law 152/99). Finally, an impact study of the waterbed in the area (Garibaldi,Varesene, new building of RegioneLombardia) was also provided.

The approval process for the dischargein Martesana was more complex andcreated the biggest risk. There weremany parties involved, and each had to be consulted individually and thentogether in a joint meeting, in order to agree a way forward.

The owner of the water in the MartesanaRiver near the Garibaldi area is theConsorzio Villoresi. The party responsiblefor the structure under the road adjacentto the site, Via Melchiorre Gioia, thatcontains the river, is the Comune diMilano. They in turn let the management of the river to the Metropolitana Milanese(Servizio Idrico Integrato).

The initial agreement for discharge was granted by the Consorzio Villoresi.However, this also had to be ratified by another body, the ‘Consorzio Navigli Lombardi’, who would be taking over responsibility for the river from 1 January 2006.

The client is still not clear as to the feefor discharge into the river. However,they are protected by the local law,DGR 1/08/2003 7/13950, which shouldensure that it is a nominal amount. Theclient has taken a view that the financialrisk is low, but has asked us to design a building which can be easily retrofittedwith cooling towers and boilers, shouldthe future users be held to ransom.

The project is not yet cut and dried but all approvals are in place. The civilengineering has now begun, and thebuildings were tendered in August 2007with the ground water scheme intact.

Client: Hines

Architect: Pelli Clarke Pelli Architects,Adamson Associates, Tekne

Services: Building services, LEEDenvironmental consultancy.

15

The refurbishment of the RoyalShakespeare Theatre in Stratford-upon-Avon is a great opportunity to create a high profile, low energybuilding. One of the key techniqueswill be ground coupling via a groundsource heat pump. Mark Owenexplains the design decision andits execution.

As part of the transformation of theRoyal Shakespeare Company’s theatresin Stratford, the design team set itself astrict energy/carbon emissions strategy,with the intention of reducing overallcarbon emissions from the redevelopedsite by some 20%.

To achieve the target carbon emissions,a number of solutions for the energysaving, energy sourcing and generationwere investigated:

■ Combined heat and power (CHP) ■ Site-wide energy loop ■ Site-wide power network■ Improving the building fabric■ Heat recovery and improved

control of the building services and environmental systems

■ Ground source heat pumps (GSHPs).

During the Stage C design development,due to a combination of budget and siteconstraints, the possibility of utilising a CHP system and site-wide energy loop were discounted, along with thesite-wide power option.

A study into the performance of the building fabric was undertaken.However, as a significant amount of thebuilding’s existing fabric is listed and,therefore, cannot be thermally upgraded,there were limited opportunities toenhance the overall performance.

To achieve the target carbon emissionsreduction would require improving theperformance of the building servicessystems, both in terms of efficiency and operation. This led to the focus on a ground source heat pump (GSHP)system as a low-cost, low carbon-emitting heating and cooling option,

along with the introduction of improvedcontrol systems, monitoring and heatrecovery systems.

Ground response test

A previous desk study, completed inAugust 2005, outlined the potential for aGSHP at the site. The study concludedthat the local geology was thought to be well suited to the technology and theextra capital cost could be justified bythe anticipated reduced operating costsand carbon emissions. Buro Happoldadvised that a Ground Response Test be carried out to ascertain the exactinsitu thermal properties of the ground,prior to taking this approach further. The test was carried out in January 2006 and a single borehole was drilled to adepth of 125m in the corner of TheatreGardens adjacent to the Swan Theatre.A summary of the results can be foundin figure 1.

The most important parameter requiredfor a GSHP system is the soil thermalconductivity. This reflects the rate ofheat transfer to and from the ground,and forms the basis of calculating the system’s performance. The actualtest result of 1.69W/mK, while beingslightly lower than the 1.9W/mK levelindicated in the desk study, was stillacceptable for use with a GSHPsystem. The soil thermal capacity and far field temperature results werealso within the limits acceptable for the installation of a GSHP system.

Ground source heat pump system capacity

IES Thermal models of the RoyalShakespeare Theatre (RST) and Swan Theatre have established that the buildings will require the buildingservices systems to cater for thefollowing peak loads:

■ Cooling: 350kW■ Heating: 1200kW

The test results from the boreholeindicate that the size of well field tocater for the required peak heating load would exceed land that is currentlyin the ownership of the RSC.

Theatre Gardens to the south-west of the RST/Swan was identified as thepreferred location for the well field orground loop heat exchanger (GLHE).The area has the potential to cater for 65 to 70x125m-deep, closed loopvertical boreholes, each rated at around5kW, spaced at between five and six metres (see figure 2). It will deliver a base load of around 350kW, in eitherheating or cooling mode. Althoughpotentially catering for the entire coolingload (eliminating the need for chillers) it would clearly require additional plant for the peak heating load (figure 3).

The refurbishment of the Royal Shakespeare Company’s theatres provides an opportunity to investigate alternative energy sources such as closed loop ground sourced heat exchangers

To bore or not to bore?

16

250000

200000

150000

100000

50000

0

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep Oct

Nov

Dec

Conventional

GSHP

Months

GSHP+Conventional Split - Heating

Hea

ting

(kW

h/yr

)

Figure 1 A summary of the ground response test results

Figure 2 Theatre Gardens borehole plan

Figure 3 GSHP heating profile

120000

100000

80000

60000

40000

0

20000

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep Oct

Nov

Dec

Conventional

GSHP

Months

GSHP+Conventional Split - Cooling

Co

olin

g (k

Wh/

yr)

Figure 4 GSHP cooling profile

Parameter Value Range

Soil thermal conductivity (W/mK) 1.69 1.66 – 1.72

Soil thermal capacity (MJ/m3K) 2.19 1.95 – 2.58

Deep far field temperature (°C) 10.0 9.5 – 10.5

Groundwater effect no

IES dynamic analysis

Due to their operational profiles, theatrebuildings traditionally encounter largepeak loads for relatively small proportionsof the day and then fall back to a basecondition. Dynamic thermal models ofthe buildings established daily/monthlyload profiles and these have been utilisedto generate a more accurate assessmentof the operation and integration of theGSHP system.

The estimated heating and coolingprofiles for the 350kW GSHP system, in conjunction with the dynamic heatingand cooling loads, are detailed in figures 3 and 4 respectively.

The results concluded that, although a 350kW GSHP system may only becapable of providing approximately 30%of the peak heating load, it would becapable of delivering approximately 76% of the building’s total yearly heating/cooling requirement and thus reduce theoperation of the supplementary heatingsystems to ‘peak lopping’. The GSHPis capable of providing the entire coolingload of the RST/Swan and the proposedGSHP system design enables heating or cooling at the same time, but withcooling taking precedence.

In early spring and late autumn, whenthere may be a requirement to both heatand cool the building at the same time, if the cooling load was small it couldresult in inefficient running of the system.However, the main cooling loads areassociated with the air systems and itis, therefore, anticipated that during thisperiod the free cooling potential of theexternal air will be utilised, reducing thecooling requirement.

Operational savings

The dynamic thermal models makeit possible to assess the system’soperational costs and the pay-backperiod for savings achieved, by utilising the 350kW GSHP system inconjunction with the top-up systems.

The operational costs assume that theGSHP system will provide the entirecooling requirement and the baseheating load.

The analysis studied the dynamic loadsand maximised the operation of theGSHP to achieve the best coefficient of performance (COP) for the overallsystem. Generally COPs of 3-4 can be achieved using traditional GSHPsystems, in either heating or coolingmode. However, by providingsimultaneous heating and cooling,COPs of up to 6-7 can be realised.

The analysis also included heat recoverysystems serving the air handlingsystems. Figure 5 highlights theestimated operational savings per year.

17

Two scenarios are presented above fordiffering gas and electricity unit prices.Recently, the UK has experiencedincreasing utility prices so scenario 2seeks to consider higher gas andelectricity rates. Higher relative gasincreases are likely as the supply fromthe North Sea diminishes. As the gapbetween gas and electricity pricesreduces, the operational costs of theGSHP system will offer better value and improved pay-back.

Carbon dioxide emissions

The GSHP system is required to providethe significant portion of the targetcarbon emissions reduction. The currentheating and cooling load for the existingRST/Swan, with no heat recovery andpoor control systems, generates a totalcarbon dioxide emission of around470,000kg/yr. This equates to 43% of the building’s total emissions ofaround 1,090,000kg/yr.

The current target is to provide a 20% reduction in this figure of around218,000kg/yr, resulting in revised totalemissions of 872,000kg/yr.

Figure 6 shows the calculated reduction in carbon dioxide emissions with theinstallation of a GSHP system, comparedwith a new conventional installation.3

Both the results include the operation of heat recovery systems and are basedupon the new building’s thermal models.

The electrical load for the new theatres will rise due to the inclusion of increasedtechnical stage engineering requirements,such as power flying. Emissions of thisincreased load are somewhat unknownand will depend on factors such as showrequirements. However, by assuming the current electrical loads as a base, itwas possible to establish the expectedcarbon emissions of the new building and compare the GSHP system against

Figure 5 Estimated operational savings

Figure 6 Carbon dioxide emissions

Fuel costs per year Savings per year

Scenario 1: Gas – £0.03/kWhElectric – £0.07/kWh

Conventional (heating – gas, cooling – electric)1 £52,633

GSHP (with peak gas and electric chillers)2 £32,584 £20,049

Scenario 2: Gas – £0.045/kWhElectric – £0.09/kWh

Conventional (heating – gas, cooling – electric)1 £87,678

GSHP (with peak gas and electric chillers)2 £46,572 £39,106

CO2 Emissions (kg/yr) Reduction (kg/yr)

Conventional (heating – gas, cooling – electric) 380,000

GSHP (with peak gas heating) 220,000 160,000 or 42%

Conventional heating and coolingExisting electrical emissions 620,000kg/yrConventional gas heatingwith electric chillers 380,000kg/yrTotal 1,000,000kg/yr

GSHP/peak gas heating Existing electrical emissions 620,000kg/yrGSHP/peak gas heating 220,000kg/yr Total 840,000kg/yr

1 Based on an efficiency of 85% for the conventionalheating system, and a coefficient of performanceof 2.5 for the conventional cooling plant.

2 Based on the performance of a typical heat pump and GLHE system.

3 The calculations are based on a carbondisplacement factor of 0.19 kg CO2/kWh for gas, and 0.43 kg CO2/kWh for electricity(CIBSE Guide F).

a new conventional arrangement of gasboilers and electric chillers.

18

Cooling Load

Chiller Boiler Plant

Bore Holes

Cold season operation with GSHP providing heating with top up from boiler plant. All cooling requirements provided by the chiller plant

Heat Exchanger

Hot water and High Grade Low Temp Htg

Underfloor and Low Grade Htg

GSHPs

1 2 3

Figure 7 Cold Season operation Figure 8 Hot Season operation Figure 9 Mid Season operation

The comparison shows that althoughthe new building will reduce emissionsby some 90,000kg/yr (9%), comparedwith the existing building, the installationof GSHP results in an overall reductionin emissions of 250,000kg/yr (23%).

System design

A key element of the design is tomaximise the efficiency of the heatpumps by allowing them to operate withthe potential to simultaneously provideheating and cooling. They would useheat rejected during the cooling processas low-grade heating for the underfloorheating and the ventilation systems.

High-grade, low temperature heating will always be a requirement to cater for the generation of domestic hot waterand to provide heating in retained areasof the building. Heating systems requirehigher temperature water, which will be generated by the gas-fired boilers.

While the analysis had established thepotential for the GSHP system to caterfor the entire cooling requirement, thedesign introduced the provision of a topup chiller. This recognised that a back-up system could be required in the case of GSHP equipment failure; the need to allow the borehole field to recharge

itself on occasions, and future proofingagainst increased cooling requirements.The diagrams above (figures 7-9)indicate how the GSHP and top upsystems will operate during variousdemand profiles throughout the year.

Project summary

In July 2007, the project took asignificant step forward with the closure of the existing RoyalShakespeare Theatre. The SwanTheatre will continue to operate untilAugust 2007. Full demolition of the RST auditorium will then begin, followedby its reconstruction along with thesurrounding and support areas.

The re-opening of the RST and SwanTheatre spaces is scheduled for autumn2010. Fortunately, the installation site of the GSHP systems does not lie on the critical construction path so it will be programmed during theintervening period.

Client: Royal Shakespeare Company

Architect: Bennetts Associates

Services: Building servicesengineering, structural engineering,infrastructure and development,geotechnical engineering, projectmanagement, fire engineering designand risk assessment, computationand simulation analysis.

19

Paper merchant pushes for water/water heat pump

In this case study of the Daintree Building in Dublin, Edith Blennerhassett looks at aground source heat pump closedloop system used for heating only.

The Daintree Building is the concept of Paul Barnes who owns and operatesa paper shop in Dublin. Not any oldpaper – Paul makes his own paper and imports unusual paper from all over the world.

The idea for this project stemmed fromhis vision to build a sustainable buildingin Dublin. Paul Barnes appointedSolearth Architects to develop thebuilding design. Buro Happold wasappointed as structural and buildingservices consultant.

This four-storey building is structurally asingle-storey reinforced concrete frame,topped by a three-storey timber frame.The concrete frame encloses groundfloor commercial and retail space, whilethe timber structure encloses primarilyresidential space, with some retail andoffice space located on the first floor.

Underfloor heating was originallydesigned to be installed throughout the building, and the use of lower thannormal flow and return temperaturesfacilitated the introduction of the groundsource heat pump (GSHP). The heatpump installation is located in thebasement plant room and was designedto be operated on low-rate electricityduring night time hours.

Domestic hot water (DHW) to thebuilding is being provided primarily fromsix solar water heating panels located at high level on the building. The panelsare of the evacuated tube type thatcollect energy even in cloudy conditions,which are prevalent at all times of theyear in Ireland. The solar panels areexpected to provide all of the hot waterrequirements during the summer. Thehot water generated by the solar panelsis piped to the basement plantroom andstored in a cylinder, which is then used

to supply pre-heated water to the mainhot water storage vessel.

The primary energy source for heating is the GSHP and for DHW is the solarcollector system. A gas-fired condensingboiler was installed to provide a backup/boost for the LTHW supply for spaceheating and for the DHW generation.The condensing boiler also feeds a smallair handling unit (AHU) for the basementarea. The hot water annual load wascalculated as 2200kWh and the heatingload for the basement AHU was23,000kWh per annum.

Grant funding

An application for grant funding wasmade to Sustainable Energy Irelandunder its House of Tomorrow scheme to part pay for the heat pump and otherenergy saving initiatives based on theseven apartments. Using the calculationspreadsheet provided as part of theHouse of Tomorrow package the use ofthe GSHP showed an energy reduction,compared with gas-fired boiler plant of71%. However, it also showed anincrease in carbon emissions ofapproximately 13% due to the highcarbon dioxide factor in Ireland for powergeneration. (These calculations are basedon a coefficient of performance (COP) of3, as stipulated in the SEI calculations).

However, the client intended to procureelectricity from a renewable supplier,such as Airtricity, to remove/reduce the carbon content of the electricity. The proposal, therefore, resulted insignificant CO2 savings, in addition to kWh savings. A grant of €35,000 was provided towards the total GSHPcapital cost of €50,000.

The heat pump for the building is a water to water heat pump rated at 30kW (based on 0°C out of theground and 50°C running which wouldrepresent a COP of 4. The output could be as much as 45kW, with a 7°C out temperature from the groundand a running temperature of 40-50°C). The heat pump was designed to takethe full space heating load, which wascalculated at 31,000kWh per annum on an overall area of 1346m2. The heatpump is linked to three 150m deep and150mm diameter boreholes spaced aminimum of 15m apart. This is used asa rule of thumb by heat pump suppliersin the absence of ground information– one 150m deep borehole for every10kW of output.

The heat pump unit is a Fighter 1310model supplied and installed by acompany called Unipipe. The refrigerantis R407c. The heat pump is used in thisinstance for heating only.

The primary heating energy for the building comes from the ground and for the domestic hot water from the sun

20

Within each borehole is a closed loop collector of polyethylene (Upanor‘energy system’) pipework filled with a water and anti-freeze medium (glycol).These loops are pumped and collectedat a single manifold adjacent to the heatpump unit. The pipework was installedwithin the boreholes as the boreholeswere being formed. The borehole is notbackfilled with bentonite or any othermedium; it is only lined through thealluvial layer. The borehole is cappedwith a neoprene cap to prevent directentry of ground water into the watercourse. The average cost of theboreholes is currently €25 per metrewith a further €70 being required for the neoprene cap.

The collector flow temperature isbetween 0°C and -4°C during theheating season, with return temperatureof between 7°C and 3°C. The heatpump is designed to deliver water at 45°C to the underfloor heating and radiator circuits.

The controls system was designed to allow the gas-fired boiler to feed onto the header only when the heatpump could not deliver the minimumtemperature of 45°C. A buffer vesselwas installed in the line to limit cycling.The controls are based on floatingcondensing technology but areessentially compensated controls.

Altered designs

The heat pump was performancespecified and a number of changeswere made to the installation whilst on site without reference to us asdesigners. In particular, the outletconnection from the buffer vessel was limited to half an inch which had a throttling effect on the heat pump, and the boiler was fed onto the systembeyond the buffer vessel rather than atthe header position. These changes andthe set up of the associated controlscombined to result in higher waterreturn temperatures than desirable, and the heat pump cutting out on itshigh temperature return thermostat.

The result is more continuous running of the back up boiler than envisaged in the original design.

In addition, for cost and constructionreasons, underfloor heating wasprovided to the basement and groundfloor only: the higher levels, including the apartments, are heated by radiators.This change led to the flow temperatureto the system being increased abovethat required for underfloor heating in order to keep the size of the radiators within the small apartments to acceptable levels. This reduced the usage and COP achievable from the heat pump.

We are currently reviewing theinstallation, with a view to altering the buffer vessel and back up boilerconnections to set the system runningas designed.

Client: Daintree Paper

Architect: Solearth ecologicalarchitecture

Services: Structural engineering,building services engineering.

21

Complementary technologies

The most environmentallyresponsible means of applyingground source heat pumps is whenthe motive energy can be renewablysupplied on-site. Jason Gardnerexplains how his team haveachieved this ideal combination ona project in Sheffield – integratingground source heat pumps todeliver a carbon neutral building.

The Advanced Manufacturing ResearchCentre (AMRC) at Sheffield University is an environmentally innovative facilitythat will be one of the UK’s first carbonneutral building of its type; it is capable of generating its entire annual energyconsumption. At the heart of the AMRC’senergy strategy are ground source heatpumps and on-site renewable electricitygeneration, shown above. Moreover, this is designed to be a financially viable,repeatable solution. With Carbon Trustfunding, there is a five year paybackperiod for the energy efficiency measures and electricity generation.

Reclaimed mine land forms the bulk ofthe AMRC site. The 4,200m2 facility willprovide a mixture of flexible workshops,laboratories and offices to support theUniversity of Sheffield’s work in the fieldof innovative manufacturing techniquesfor the aviation industry.

All heating and cooling energy isprovided by wind generated electricityto ensure the carbon neutral target ismet. Ground source heat pumps, linkedto a closed loop network and boreholes,provide the building’s heating, cooling and hot water loads.

Occupants will have a high qualityinternal environment. Workshops andlaboratories are closely temperaturecontrolled, primarily to maintainequipment calibration. In contrast, the offices will offer a comfortable,naturally ventilated environment.

Design process

During the earliest stage of the projectBuro Happold applied a ‘carbonmitigation’ design strategy to the designprocess (refer to diagram below). Thisinvolved focusing on reducing energyconsumption, initially through goodbuilding form and fabric design. Onlyonce the energy saving contribution of the building’s form and fabric hadbeen fully exploited did the design teammove onto developing the use of energyefficient services in detail, of whichground source heat pumps formed a key element. Applying this ordereddesign process enabled Buro Happoldto minimise the scale of the heat pumpinstallation, thereby maximising itspositive contributions by reducing theenergy consumption and system costs.

A vital part of the carbon mitigation design process was to ensure that the technologies applied to the projectcomplemented one another. Hence whenconsidering the fourth stage of the carbonmitigation design process, the introductionof renewable technologies, a key elementwas to ensure that the renewable

technology was compatible with the heatpumps. The wind turbine, whichgenerates 1,000,000 kWh of electricity perannum, ideally complements the groundsource heat pump installation by supplyingall the system’s power needs inconjunction with entire building electricaldemand. It should be noted that duringperiods of low demand, excess electricityis exported to the national grid, soenabling the building’s carbon neutralstatus to be realised.

Ground sourced system

A ground source heat pump systemprovides the entire cooling load for theAMRC building. The heating load for the offices, laboratories and ancillaryspaces is also provided by the pumps.The heating and cooling requirementsare supplied by four reverse cycle heat pumps, each providing 45 kW ofheating and 38 kW of cooling. Care wastaken to make the heating and coolingloads of similar value so as to achievethe highest efficiencies from the heatpumps and also to ensure that theoverall installation was as economicallyfeasible as possible.

The AMRC’s on site generation complements the ground sourced heat pumps

Carbon mitigation design process

22

The entire ground loop system is locatedbeneath the site car parking areaadjacent to the building. The car park is provided with a permeable surface to prevent the soil surface drying out, as this would decrease the ground loopheat transfer capability in this region. To achieve the required performance a total of 20 boreholes have been sunkat a depth of 100m each (see figure 2).

Heat pump hydraulic circuits arearranged to achieve free cooling fromthe ground loop whenever possible. Thecooling mode of the heat pumps is onlyactivated when this free cooling capacityis exceeded. In cooling mode, any heatfrom the heat pumps, normally classedas waste, firstly ‘looks’ for either adomestic hot water (DHW) or heatingcircuit load, before being rejected backto the ground circuit. This significantlyincreases the efficiency of the systemduring any periods when simultaneousheating and cooling are required.

Cooling energy is transferred to thedistribution system via a plate heatexchanger. A buffer vessel maintained at the required distribution temperaturesensures that chilled water is alwaysavailable.

The heat pumps are arranged with a lead unit that provides highertemperature water to a domestic hotwater cylinder. A distinct and separate‘hot gas’ circuit through the heat pumpsprovides additional heat recovery fromeach unit when they are in operation.The heat pumps contain an extraintegral heat exchanger to recover allavailable heat from the refrigerant gasbefore it enters the expansion side ofthe system. The harder the heat pumpunits work, the higher the amount ofsecondary heat is available for recovery.Flow in this heating circuit is varied, to achieve the higher low temperaturehot water (LTHW) temperatures requiredto heat the DHW cylinder.

Cooling is coupled with efficientdisplacement ventilation systems thatraise the flow and return temperatures

Figure 2 The 20 boreholes 100m deep are beneath the car park

Figure 3 Building and building services design was tuned to maximise the usefulness of ground sourced heatwith lower heating flow and return temperatures and higher chilled watertemperatures than usual

and reduce the demand for chilledwater (see figure 3). Displacementventilation supply air temperatures arein the region of 19°C instead of the12-14°C that would be required from a traditional mixing ventilation system.Increasing the supply air temperaturesignificantly reduces the amount ofcooling that is required, especiallyconsidering the fact that latent cooling is not required. Increased water flow and return temperatures of 11-15°C,instead of the more conventional6-12°C, have been used to furtherreduce the energy requirements to generate chilled water.

Heating distribution is via wet underfloorheating circuits throughout. This allowslow flow and return temperatures of40°C and 30°C to be used for theLTHW circuit. These temperatures are well matched to the most efficientachievable output temperatures of the heat pumps.

On-going monitoring

The final stage of the carbon mitigationdesign process, ‘Operation’, embracesthe need to ensure that the buildingservices operate as the designerintended. Only by continually monitoringthe energy and usage characteristicscan the low carbon credentials of abuilding be fully proven, and potentiallyimproved upon.

The factory of the future has beendesigned as a prototype, but theconcept is applicable to many energy-hungry facilities. It will undergo extensivepost-occupancy analysis and has beenprovided with targets and a means of monitoring performance.

The heat pump installation has beenprovided with sufficient monitoringequipment to enable the seasonalco-efficient of performance (COP) to be measured. To achieve this, the heatpumps heating and cooling generationwill be metered separately along withthe power consumption of the heatpumps in both modes.

This project is due for completion inDecember 2007, and the subsequentmonitoring will reinforce the value of the AMRC as a learning facility that will help teach the construction industry the way to achieve carbon neutralindustrial buildings.

Client: University of Sheffield

Architect: Bond Bryan

Services: Building services engineering,structural engineering, groundengineering, civil engineering, BREEAMconsultation and assessment,acoustics, fire engineering design and risk assessment.

23

Ground source heating and cooling study

James Dickinson from theSustainability and AlternativeTechnologies (SAT) group looks atthe application of ground source heat pumps in this case study of the Stockport Academy.

The new Academy is to be built onthe same site as an existing high schoolin Stockport. The relevant proposals for the building were summarised as follows:

■ Provide a sealed building due toacoustic restrictions on the sitearising from its close proximity to Manchester airport.

■ Mechanical ventilation with heatrecovery to be provided to alloccupied areas.

■ Ground source heating and cooling toprovide the required cooling for all theICT areas, cooling to air handling unitsand low grade heat to air handlingunits and underfloor heating systems.

A 3D representation of the building is shown in figure 1.

Design method

To optimise the GSHP, in terms of capitaland operational costs along with carbondioxide reduction, a detailed simulationof the system was completed.

Building thermal model

The thermal performance of the buildingwas evaluated using the softwarepackage IES. This generated the hourly heating and cooling loads for an average test year. The resulting load and energy profiles are shown in figures 2 and 3.

It was evident from this review that theheating and cooling loads and annualrespective energy requirements werevery different. Closed loop GSHPs canbe sized to meet a building’s dissimilarheating and cooling loads. However, it is possible to make considerablesavings if steps can be taken to equateboth these system parameters.

Heat is abstracted from the ground inthe heating mode and rejected to theground in the cooling mode. Net heatabstraction over the year means thatthe potential to ‘recharge’ the boreholefield over the year is reduced. Thecumulative length of the ground loopmust be increased to ensure continuedlong term performance.

GSHP sizing

Due to this imbalance it was decided to consider a bivalent1 system. In thisinstance, relatively infrequent peak loadscould be covered by conventional lowercost technology. There were two maincost benefits to this; firstly the boreholefield size could be optimised andsecondly, less expensive plant could be used for infrequent loads. The capitalcosts could be minimised but significantoperational benefits would still be realised.

Following interpretation of the simulation it was decided that a 300kW GSHP offered the best solution. This size has theadded benefit of eliminating the need forconventional cooling plant. The heating and cooling requirement is essentially out ofphase, with cooling demand (aside from ITserver rooms) in the summer and heating in the winter. The GSHP system will includefour heat pumps, each being able tomodulate between heating and cooling, so in mid season the system will be able tocover both heating and cooling loads. Highefficiency gas fired boiler plant is sized tocover residual heating loads in the building.

Figure 1 3D Model of the Stockport Academy building

Figure 2 Heating and cooling load comparison

Figure 3 Heating and cooling energy comparison (kW)

180,000

200,000

160,000

140,000

120,000

100,000

80,000

60,000

40,000

0 20,000


Ene

rgy

(kW

h)

Months

Cooling Energy (kWh)

Heating Energy (kWh)

Figure 4 Ground energy exchange comparison (kWh)

160,000

140,000

120,000

100,000

80,000

60,000

40,000

40,000

0

20,000

20,000


Ene

rgy

Exc

hang

e w

ith

gro

und

(kW

h)

Months

Heat Rejection (Cooling Mode)

Revised Heat Abstraction (Heating Mode)

Peak Sized Heat Abstraction (Heating Mode)

Figure 4 shows the resulting groundenergy exchange comparison with a peak heating sized GSHP system. The revised total heat abstraction and heat rejection from the ground is now much closer.

1 Bivalent – Where two sources contribute to the overall heating and/or cooling demand, as opposed tomonovalent where one type of equipment is used for the entire heating load, and similarly for the cooling load.

24

The total heat abstraction is stillmarginally higher, but this could changewith the future effect of global warmingcausing higher cooling demands andhence heat rejection.

The final calculated energy mix for theheating and cooling in the building ispresented in figure 5. Note that thebuilding heating energy provided by theGSHP is slightly higher than the heatabstraction from the ground – this isbecause of the added electrical energyfrom the heat pump. In the coolingmode the building cooling energy islower than the heat rejection as theelectrical energy is transferred in theopposite direction. This can be furtherunderstood with reference to thethermodynamic Carnot cycle.

Using GSHP simulation software Buro Happold estimated the cumulativelength of both the peak sized GSHPsystem and the bivalent option. Thepeak sized system was estimated toneed 210x100m deep boreholes whilstthe alternative requires approximately 45 boreholes of a similar depth. As theborehole field is the most expensiveaspect of the GSHP installation it isclear that the bivalent option would be less capital intensive and would be more cost effective.

Operational savings and payback

The GSHP simulation predicted theelectricity used by the heat pump plantso comparisons could be made withmore conventional plant. For thisanalysis the following assumptions were made:

■ Seasonal Gas Boiler SystemEfficiency: 80%

■ Conventional Electric Chiller PlantSeasonal Co-efficient of Performance(SCOP): 2.5.

The payback was based on additionalcapital expenditure of approximately£160,000 therefore taking into accountthe extra cost for the GSHP but also the consequent elimination

of conventional chiller plant andreduction in gas boiler capacity.

To assess the economic feasibility of theGSHP the system was modelled usingsensitivity analysis of future utility prices.Figure 6 shows how energy prices and specifically gas prices relative to electricity prices have increased over the last few years.

Forecasting for future electricity and gas prices is extremely complex and itcan be difficult to predict the long termpayback for more efficient, but higher

Figure 5 StockportAcademy final energy mix

Figure 6 Retail priceindex utility price trends(DTI) 2

0

20,000

40,000

60,000

80,000

100,000

120,000


Hea

ting

/Co

olin

g (k

Wh)

Months

GSHP Cooling (kWh)

GSHP Heating (kWh)

Gas Heating (kWh)

180

200

160

140

120

100

80

60

40

0 20

2002 2003 2004 2005 2006

RP

I 100

= 1

990

Year

Gas

Electricity

Figure 7 Paybackscenario analysis£200,000

£180,000

£160,000

£140,000

£120,000

£100,000

£80,000

£60,000

£40,000

£20,000

£-

0 5 10 15 20 Cap

ital

Co

st a

nd C

umul

ativ

e S

avin

gs

(£)

Years

Capital Over Cost Scenario 1 Scenario 2 Scenario 3 Scenario 4

2 Department of Trade and Industry, QuarterlyEnergy Prices www.dti.gov.uk/energy/statistics/publications/prices/tables

cost, plant. However, four scenarioswere constructed to reflect differentchanges in the energy market that aresupported by the recent trends reportedby the DTI. The third was chosen as themost likely based on reasoned analysisof energy imports and increases in thecost of the different utilities.

25

Figure 7 (on previous page) shows theresultant payback predictions.

A summary of the estimated paybackand annual savings for each Scenario in year 10 are:

■ Scenario 1: 15.2yrs/£11,300■ Scenario 2: 12.93yrs/£14,100■ Scenario 3: 11.5yrs/£17,900

(Predicted)■ Scenario 4: 10.2yrs/£24,500

This analysis also highlighted the added resilience the GSHP systemgives to future fossil fuel prices.

Building services

To maximise the GSHP efficiencyheating, low temperature hot water willbe delivered via an underfloor heatingcircuit and chilled beams at 45ºC. A gas fired boiler will deliver heat to allthe AHUs, radiators and radiant panels.Cooling will also be delivered via the chilled beam circuit and cooling coils in the AHUs. The cooling system hasbeen sized to allow for higher than usual cooling flow temperatures of 14ºC.

The GSHP will serve the cooling circuitas a priority but, via a sliding headerarrangement, the heat pumps will beable to modulate between the heatingand cooling loads. That the GSHP will act as the primary heating providerwhen the cooling demand is low tomaximise the benefits of its operation.The gas fired boilers will provide backup during peak loads via an injectioncircuit. During periods where there is spare GSHP capacity, heat will be diverted to preheat the domestichot water for the building.

A simplified schematic of thearrangement is shown in figure 8.

Conclusions

The bivalent GSHP system designed for the new Stockport Academy willsignificantly reduce the operationalcosts of the heating and coolingprovision and reduce the total effective carbon dioxide emissions from operating the building.

The estimated annual carbon dioxidesavings are 67mtCO2 (11.5% of the building total). The GSHP provides30.6% of the building’s total energyrequirement (23% of which isrenewable). The carbon dioxidereduction is an essential part of thestrategy to ensure the building meets2006 Part L Building Regulations.

The analysis and the consequentinstallation of a GSHP at StockportAcademy shows the potential for the application of this technology in educational buildings. The initialconsideration should include a review of the respective heating and cooling

Gas fired boilers

Injection circuit

Temperature sensor

Sliding header

Low temperature heating circuit • Heating to chilled beams • Heating to underfloor • Preheat to DHWS calorifier

Cooling circuits • CHW to chilled beams • CHW to AHU coils • CHW to server room buffer vessel

Heat pump 1

Heat pump 2

Heat pump 3

Heat pump 4

Ground loop heat exchanger CHW

LTHW LTHW

Figure 8 Simplified GSHP Building Services Schematic

loads to optimise the operationalbenefits with respect to the capitalcosts. When occupation is low duringthe cooling season (summer) the overallheating energy required for the buildingis generally much higher than thecooling energy.

In new academies or other schoolswhere the cooling requirement issignificant due to restrictions in providingnatural ventilation, there is a particularbenefit in sizing the GSHP to meet thecooling load. This enables the eliminationof extra conventional cooling plant whilereducing the net heat abstraction fromthe ground. Particular attention shouldalso be given to future utility prices asthis can make a significant impact on the economic case.

Client: ULT Projects Ltd

Architect: Aedas

Services: Building services, buildingstructures, SAT, CoSA.

26

Transferring the technology ofcommercial scale developments to a domestic scale of project canpay dividends. There are, however, a number of pitfalls to avoid, asBrian Doran explains in this groundsource heating case study.

There is some satisfaction in being ableto practice what you preach – a chanceto utilise new technologies within one’sown dwelling. The renovation/rebuildingof my house in rural Ireland provided the opportunity to examine and install a ground source heat pump and solardomestic hot water heating for me andmy family.

This case study summarises the upsand downs of an installation, whicheventually delivered a successfullyoperating 8kW water/water heat pump (HP) serving space heating via a horizontal ground loop array. The design justification was not skewedby any artificial factors such as grantaid, which was not available at the timein Ireland. The choices were justified on the straightforward capital cost and simple payback periods.

However, the process did highlight theconsiderable resistance to out-of-the-ordinary techniques in a marketplaceunfamiliar with heat pump technologyand energy-conscious construction. The heartening news is that in theintervening months there has been a sea change in terms of the situation in Ireland and it would now be mucheasier to progress this project.

Because this project was part existingbuilding and part new build, the scenarioof a very low energy building at orapproaching Code for Sustainable HomesLevel 6 (zero carbon) was not viable, even ignoring our budget constraints.

Why use a heat pump?

For us, the key design choices regardingenergy strategy and heat source were:

a) How low we could go, in terms of energy conservation

b) Minimising carbon emissions fromthe energy consumed – and lookingat available alternative technologies

c) Aspiration for a hassle free operation– low maintenance, simple energypurchase and delivery

d) Enthusiasm to learn a little moreabout domestic scale installation of alternative technologies, with a hands-on approach and postoccupancy monitoring.

Double slinky in County Kerry

After energy saving measures wereincorporated, the above criteria gave us an obvious steer to an HP solutionserving underfloor heating. However,there was a choice to make regarding the use of air source or ground sourceheat pumps. Though generally notclassified as a renewable technology,the former does offer a simple,cost-effective solution, assisted by the mild Kerry climate.

Scenario System

Dwelling emissions

Assumed SystemEfficiency

kgCO2/year% saving compared

with scenario 1

1 Direct electric heating 2556 – 99%

2 Gas fired heating 1306 -49% 80%

3 Oil fired heating 1719 -33% 80%

4Air source heat pump

(space heating only)1661 -35% COP: 2.2

5Ground source

heat pump (space heating only)

1332 -48%COP: 4

6

Ground source heat pump

(space heating).Solar water heating

1053 -59%HP COP: 4

Solar contribution: 30%

7Ground source

(space and water heating)

1265 -51% COP: 2

8 Biomass 235 -91% 80%

Notes Emissions are based on the following data:

Typical dwelling requirementsSpace heating demand 3500 kWh/year (delivered)Water heating demand 2000 kWh/year (delivered)

Carbon emissionsGas 0.19 kgCO2/kWhOil 0.25 kgCO2/kWhElectricity (Average for UK grid) 0.46 kgCO2/kWhBiomass (harvesting/transportation) 0.03 kgCO2/kWh

Figure 1 Approximation of dwelling emissions

27

Hot water

The other issue considered was whetherto utilise the HP for delivering thedomestic hot water demands or limitingit to space heating, operating only inwinter. We looked at using the HP todeliver the high temperature primarywater (>60°C) and alternatively assimply a pre-heat to the domestic waterload, thereby maximising the system’sCoefficient of Performance (COP). There are a number of manufacturerswho claim their high temperature(>60°C) domestic heat pump packagesachieve a COP of >3. They publish test analysis to substantiate this, but proven examples with monitoringtend to be scarce. It is clear howeverthat the thermodynamics dictate that the temperature difference of a HPsystem should be minimised in order to optimise COP.

As can be seen in figure 1, the overallsystem COP (space and hot waterheating) needs to be in excess of 2 to make it comparable with theemissions from gas-fired heating.

The additional cost of a domestic waterand space heating HP package alsogives credence to the argument toutilise separate alternative technologiesto serve the differing system needs (ie solar thermal for the highertemperature needs and heat pump for the low temperature heating circuit).

Space heating is therefore deliveredfrom the ground source heat pump, with the domestic hot water heatedprimarily from a flat plate solar collector(operating as a thermo-syphon) to a thermal store with supplementaryheating by direct electric means.

The procurement of the heat pumpsystem was on a DIY package basisfrom Kensa Engineering, which included the 8kW HP with integralcirculating pumps (£3,500 +VAT), and pre-measured and formed ‘slinky’ground source tube loop network (£550).

Even with domestic installations, somedebate takes place regarding the sizingof the ground array. Although there are instances where insufficient groundloops have caused freezing and groundheave, the relatively small cost of theexternal works (in a non-confined site)mean that it should be easy to install a generously sized ground array with the possibility of an additional loop(s).The ground loops serving the 8kW HPconsisted of two 40m long trenches(300mm wide x 2m deep) for thepipework formed from 40mm diameterHDPE pipe. Both loops are installed in parallel around 6m apart and rely on being of identical pressure drop to balance flow and minimise pumpingcosts. There was little concern regardingthe sizing of the ground loops, eventhough they were not particularly deep,given the southerly aspect of the siteand mild climatic conditions.

Lessons learned

Hopefully the following hard-earnedadvice offers some pointers for futuredomestic scale schemes:

1 Never underestimate the limitationsthat can be imposed on a project due to the skill base within the constructionindustry. To be fair, it is understandablethat contractors do not wish to moveoutside the comfort zone of previous

experience and standard solutions. But this can affect even simple designconcepts such as insulation thicknesses.We had to compromise on the cavitywall design, with a 150mm cavity, whichthe local contractor found too onerousto construct.

2 The ‘lead-in’ period for the site’selectricity supply may be protracted due to the larger than ‘normal domestic’load of the heat pump. In our case, the 8kW HP required a 25amp singlephase power supply (typically 65ampstarting current).

3 Never believe a JCB driver when hetells you that he’ll be there to dig trenches!

4 Make sure every inch of the groundworks are supervised at all times. Atone stage on our project, a 20m sectionof trench was backfilled, unsupervised.

5 The cost of the ground worksinstallation isn’t great and pales intoinsignificance compared with the cost of ground remedial works when it’s not done right the first time.

6 Make sure all ground loops arepressure and flow tested, both beforebackfilling and as soon as the trenchesare backfilled. To our detriment, wefound out that a kink in the ground pipe is much worse than a fracture.

View of garden containing septic tank percolation area and slinky trenches (around the perimeter)

7 Make sure your plumber carries out a flow test and checks for air locksbefore filling the system with anti-freeze.This avoids the need to repeatedly drainthe system (or pollute the ground) if anobstruction in the pipe is discovered.

8 Due to the points above, considerinstalling at least one extra slinky. Wedid not have the foresight to do this and discovered an obstruction in one of the slinkies. This was attributed toeither an air lock, debris or a kink in one ground pipe.

9 When the usual available pumpswere not able to remove the offendingobstruction, we resorted to a drainagepressure jet company to assist. Thismay be a fairly high-risk strategy as the pressure delivered to the pipeshould be in excess of its design rating.In our case, however, we had little tolose – safely exhuming the offendingslinky would be next to impossiblewithout further risk of damage and we had resigned ourselves to simplyreplacing a complete section of the below ground installation. As ithappened, the pressure jet equipmenteasily tracked down the problem by fracturing the pipe at the point of weakness (the kink caused duringbackfilling) and forcing water to thesurface. This was clearly not a scientificapproach. The remedial works led to a major additional cost – approximatelythree times the cost of ground loopinstallation. In addition, there could be significant health and safety issues with re-excavation of the trenches.

10 Never assume there is only onekink in a faulty installation.

11 Never underestimate the forceneeded to remove an air lock in a pipe (not just in the slinkies).

12 When installing a heat pump, don’tforget to spend a little more and installan electricity meter in its power supply.

28

View of trench with slinky installed13 Don’t be put off by these avoidableconstruction errors and take heart inthe fact that the technology is robust(and now working perfectly). Sincecompletion at the end of 2006, we have been through heating a full seasonand the system performs extremely well and efficiently.

29

Finding the way through labyrinths and earth tubes

Using the earth as a method ofthermally preconditioning the supplyof air is a simple, cost effective way of employing huge thermal masses.Mike Entwisle outlines theadvantages of this technique anddiscusses some recent examples.

History

One renewable energy source that hasbeen exploited for many years is thestability of the ground temperature. This is demonstrated in many extremeclimates by the use of ground shelteredbuildings, and indeed in China cavedwellings have existed for centuries,which take advantage of this lack ofthermal variation.

Water based ground source systemsharness this stability by passing a fluidthrough pipes that are in contact withthe ground, or use ground water fromdeep sources. However, air basedsystems are generally relatively low techand to date have received little attention.They require shallow interventions in the ground and as such are suitable for most sites. They can often beimplemented relatively cheaply with little mechanical equipment required.Whilst the Greater London Authority’srenewables guide does not recognisesuch a system as counting towards the10% contribution, there can be no doubtthat the energy recovered from theground is a true renewable source andcan indeed be considerable. However,the behaviour of air based systems isnot as well understood as that of buriedloops, and in particular the effect of thesystem itself on the temperatures of the relatively shallow buried air paths.

Serendipitous cooling

I first became aware of the magnitude of the energy available from this methodwhen I studied an office building in Peterborough in the mid-1990s (see figure 1). This was mechanicallyventilated using a raised floor plenum,and provided good internal conditions

as the thermal mass was accessedthrough the floor plenum.

The floor void ventilation system of the office building in Peterborough wasnot zoned, and the client was finding it impossible to provide comfortableconditions on all floors simultaneouslyduring summer; when the upper floorswere comfortable, the ground floortemperature was too low, and if thecontrols were configured to avoidoverheating on the ground floor, theupper floors were too warm. Thisprovoked consideration of the degree of heat transfer into the uninsulatedground floor slab. I took a detailed setof measurements and found that onwarm days (temperatures of around25°C peak), the air was exiting the floorgrilles at up to 5°C below the externaltemperature. On extremely warm days,the conduction into the ground providedeven more cooling – none of which wasavailable to the upper floors and gave aconsiderable difference in performancebetween the ground and upper floors!

Air exchanges heat with the ground

Air supplied through a B placement or swirl grilles

250-300mm raised floor

Central air plenum

Suspended ceiling

Ceiling void acting as extract plenum

Light fitting

Figure 1 1980s floor void ventilation system

Earlier uses of this technology includeThe Royal Victoria Hospital in Belfast,which proved to be a landmark inventilation of hospital buildings (seefigures 2 and 3). Air was supplied from a series of subterranean plant rooms, inwhich the air was filtered and/or cooledusing wet sprays (in the days beforeLegionnaires Disease was an issue!), andheated before passing through buriedbrick corridors on the way to the wards.While the ground connection would havebeen wasteful of heat energy (unlikely tobe of concern at the time), it would haveprovided some useful additional coolingin summer, thus maintaining the comfortof patients.

Wolverhampton Civic Hall’s ventilationsystem, constructed in the 1930s, also uses myriad buried ducts as a way of getting from A to B which reapthe benefits of cooling in the summer.

30

Principles

As with all ground coupling, the basicprinciple is that the ground temperatureis much more stable than the ambientconditions. Passing cold or warm air overa surface at a certain temperature willbring the air temperature closer to theground temperature. Very little additionalequipment or ground intervention isneeded, particularly if the building is to be mechanically ventilated anyway.

The heat transfer achieved depends onthe surface area available, the velocity(which to a large extent determines theheat transfer coefficient), the time spentin contact with the ground and theexternal temperature. These are oftencompeting with each other, but thecritical issue in any system is to ensurethat the airflow in the ground contactzone is turbulent. For a typical floorplenum 300mm deep, this occurs at avelocity of as little as 0.2 ms-1, reducingproportionately in larger ducts andincreasing in smaller pipes. Once in theturbulent zone, heat transfer coefficientwill continue to increase as the airmoves faster. However, it can still obeythe law of diminishing returns, with highvelocities increasing pressure losses,resulting in noisier fans and moreelectrical energy usage.

Analysis of the airside heat transfer can be performed relatively simply, butassessment of how the temperature ofthe ground varies with depth and time is more difficult, and requires detailedthermal modelling in four dimensions.However, as a general rule, the deeperthe ducts the better, although it’s notworth going below 2-3m deep!

The heat and coolth recovered from theground can not only reduce heating usein the building, but can effectively enablethermally lightweight buildings to havesummertime performance similar tothose with large amounts of thermalmass. This can provide a high degree of resilience against changes in use,such as increased IT loads, occupancies, and longer hours of use. Most critically

for the future, it is able to deal with theeffects of climate change.

To illustrate the energy available, on a hotsummer’s day the external temperaturecould easily be 30°C in the south of the UK, with a ground temperature ofmaybe 15°C. Heat transfer coefficients of 8 Wm-2K-1 are easily achievable, giving a cooling potential of 120 Wm-2!Our experience is that this can usefullyserve buildings of two to three storeys.Indeed, Buro Happold has recentlyreached completion on a three storeyschool served entirely by a labyrinth.

There are many possible configurationsof air based systems, but the followingare the most common:

Earth tubes

With earth tubes, pipes are buried inthe ground. This is a cheap technologythat can be easily utilised beneath the footprint of the building or in itssurroundings. Burial depths of 2m are preferred, but 1m of cover canprovide a good degree of stability.

Royal Victoria Hospital, Belfast Cutaway section of engine house and head of main duct

Filtering ropes

Draught control door

Heating chamber

Air inlet grilles

Branch ducts

Engine room

Fan shaft

Heating chamber

Pipe runs

Main duct

Branch ducts

Fan house

Ward roof

Roof of main corridor

Roof of operating theatres, etc.

Foul air extract duct

Foul air exhaust

Foul air extract duct

Extracts from wards

Air inlets to wards

Royal Victoria Hospital, Belfast Cutaway of the complete ventilation system

Figures 2 and 3 Diagrams of the early groundcoupling ventilation system at the Royal VictoriaHospital, Belfast

31

Therefore, mechanical ventilation wasintroduced; the supply air plant islocated in external pods and connectedto the rooms through a series of claypipes, buried with around 1.5m of cover.Air is then supplied to the classroomsthrough a perimeter trench heater,ensuring that draughts are avoidedduring winter. The rooms still haveopenable windows but in many casesthey do not need to be used in summer,as the earth tube system maintainstemperatures at least 4°C cooler than outside on a hot summer’s day.

Since then, earth tube technology has been applied to other buildings from project inception, including the 3 Ways Special School in Bath, St MaryMagdalene Academy in Islington, theDyson Skills Academy in Bath and theyhave also been proposed for the HerefordSteiner Academy. Further explanation of the last two is beneficial here:

The Dyson Academy (see figure 4) is a hybrid scheme where the buried air ducts perform as a series of earthtubes, but are much larger, and similar to labyrinth sections. The site is veryconstrained, and is located in a visiblysensitive area of the World Heritage Cityof Bath. In addition, a busy and pollutedroad runs down one side of the building.

Therefore, a mechanical ventilationsolution is achieved with plant inside the building rather than at roof level. The north side of the building is flankedby the River Avon and allows the flow of clean air into the building. The intakeducts run to the plant room as largeculvert sections, with the pre-coolingand pre-heating that they provide beingaugmented by the use of river water for further cooling and heating when necessary.

Lastly, the air is supplied to the spacesthrough the ‘Concretcool’ system whichuses ducts cast into the structure toincrease thermal mass contact stillfurther. Sadly, this scheme is unlikely tobe built in this form as the site has sincebecome available, but we are keen toexploit these principles and techniqueson its new site.

The brief for the Hereford SteinerAcademy was for a low energy andsustainable building that would assistwith the Steiner education methods and philosophy. After lengthy debate, an exceptionally well-insulated buildingenvelope was adopted, with the use of a timber frame. This of course, gave little opportunity for the inclusion of thermal mass, which would haveprovided resilience against the everwarming climate.

Concrete and clay pipes are preferred,as they have a thermal conductivitysimilar to the materials in which they are buried and (in the case of clay) a lowembodied energy. These ducts can beused as the inlet to a system or even on the supply to a room. However, caremust be taken not to heat air in winterbefore passing it through the buriedducts as heat will be lost into the ground.

One of our earliest systems of this typewas at Bristol City Academy. Like manyschools, the building was fundamentallydesigned to be mostly naturally ventilated,with classrooms of the appropriate depthand construction. However, shortly beforegoing to tender, the DfES’s new Acousticregulations, Building Bulletin 93, cameinto force and after much discussion we were advised that they applied to our scheme. On closer examination, itemerged that much of the school waslocated in an area of the site with ambientnoise levels that would make compliancewith the strict regulations impossible for a naturally ventilated scheme. Havingalready been through the planningprocess, we were reluctant to introducechanges that would radically alter theexternal appearance of the building, so rooftop mounted plant and ventilationstacks were not practical.

External dry-bulb temperature

Temperature reduction 3-4oC

Earth tubes on, windows closed. DfES Internal Equipment Gain of 1600Watts

Earth tubes on, windows closed

Openable windows only

TimeTe

mp

erat

ure

Upper Floor Air Temperature Profile, Hottest Day (31.1°C)35

30

25

20

15

10

5

0

01:0

0

02:0

0

03:0

0

04:0

0

05:0

0

06:0

0

07:0

0

08:0

0

09:0

0

10:0

0

11:0

0

12:

00

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0

23:0

0

24:0

0:00

Steiner Academy – graph of performance

Supply Air

Supply Air

Exhaust Air

Air Intake

Extract Air(Summer)

Extract Air(Winter)

Exhaust Air atHigh Level

Fresh Air Inlet atLow Level

Supply Air

WORKSHOP

WORKSHOP

Plant RoomNo 1

Dyson – schematic

AHU No 1

Figure 4 Ventilation schematic for the proposed Dyson Academy with intake air draw – through culverts

Figure 5 Graph showing earth tube performance at Hereford Steiner Academy

32

surface area available for heat transfer.However, there are a number of otherschemes that have been less high profile.West London Academy, shown in figures6-9, and one of the first waves of DfESCity Academies, is located very close to the A40 Westway dual carriageway on the outskirts of London. In addition to the traffic noise, part of the site alsosits in an area where the air quality is deemed to be unacceptable forventilation use. Given that the location ofthe building on the site was constrainedto a ribbon close to this road, thechallenge was to provide a solution that met the acoustic and air qualityrequirements, whilst drawing air from the far side of the building and includinga degree of passive behaviour.

As a solution, we placed a shallowundercroft below the majority of thebuilding footprint, which acted as asupply air plenum to the rooms closestto the road. In winter, the air is temperedby passing through the undercroft and is further warmed by passing it overradiators on entering the room. Once the heating had been commissioned fully

North

Air supplied under positive pressure to northern rooms

Air intake away from road. Air cooled/warmed by ground

Ventilation risers built into external facade

South

Plan on Baffle

SCALE 1:10

Galvanised steel baffle with returns fixed to back wall. E.G. HCP unilock (type TG) top grille type W panel (but with no grille)

Back wall Finned tube

Damper

Finned tube

Trunking

700 high baffle 120 off wall

Revised sketch of M502 with alternative pipework mounting details as per rotary request for ladder rack SK/Rotary RFI 079M/SK001 24/06/04 Buro Happold

Arch background A2502 rev 06

Ladder rack with HTG + hot + cold pipework

340 40

0 70

0 10

0

120

drainage gully AHU supported by 50x50 unistruts laid on slab

space needed to allow unit to be installed in undercroft

undercroft drainage sump in between fan units. 2no. per entrance zone with leak detection

block wall constructed around attenuators with airtight seal. 1000x1000mm access panel within block wall adjacent to each sump

200mm space for installation access & tolerance

removable egg crate grill (assumed 80% free area)

intake louvre & motorised damper

secondary steelwork supports intake grille/access panel

removable access panel (solid)

filters & fans need regular access

external ground level

1000

1900 clear

2030 2030

1000

719 1431

1502

30

0 20

0

200

200

500

500 1250

3000

260

410

240

90

1250

1200

1530

AHU/01

From top, left to right:

Figure 6 West London Academy – section throughsubterranean air handling plant

Figure 7 West London Academy – sketch sectionthrough air supply from undercroft

Figure 8 West London Academy – schematic section

Figure 9 Teaching space – alternative ventilationstrategy for BB93

400ø Concrete pipe with waterproof wrap

Temperature sensor controlling variable speed air handling unit located outside

Supply air 0ºC Winter 30ºC Summer

Supply air • 81/s/person - winter • 251/s/person - summer

Air warmed by ground in winter Air cooled by ground in summer

Extract via existing AHU

Teaching Space

Air handling unit consealed in bench consisting of: •Acoustic inlet grille with screen (insects, rodents etc) •Damper to stop draughts and back flow •Electric frost coil •Filter •Variable speed fan •Attenuator

Bench

T

We therefore introduced earth tubeventilation into the classrooms, utilisingpassive techniques in winter, boosted bylow pressure fans in summer when lowerwind and thermally driven pressurescoincide with increased ventilationrequirements. The analysis in figure 5shows that whilst the winter heat gain inthis case is relatively small, the thermalstability that this afforded the building in summer was remarkable, ensuring that peak summertime temperatureswere again 3-4°C below the external.

Labyrinths and undercrofts

The schemes above have all used thesimple and cheap technology of earthtubes. However, a more powerfultechnique that can harness the entirefootprint of a building is the use oflabyrinths and undercrofts. These canbe located under the building, or evenextend beyond it, and can be deepenough to allow access for maintenance.The most well known labyrinth scheme isprobably that at the Earth Centre, wherethe walls even have an irregular shapeto increase the turbulence and also the

the building produced an exceptionallystable internal environment, with theundercroft plenum tempering summerand winter temperatures by up to 6-8°C and internal peak summertimetemperatures in hot weather being 5-6°C below the external peak – a remarkable performance for anybuilding without mechanical cooling.

Seasonal variations were accommodatedby varying the fans from a winter trickle tohigh summertime rates, (a ratio of around10:1), night ventilation in summer, andcareful sequencing of optimum start for heating and ventilation in winter.

33

Having succeeded in this large scheme,the principles were developed for anexemplar design for the DfES, workingalongside Feilden Clegg BradleyArchitects, and shown in figure 10. A relatively noisy site and compactbuilding plan enabled the introduction ofa labyrinth beneath the building footprint.Intake air is circulated in a 2m deeplabyrinth below the building, configuredto maximise heat transfer at minimumfan energy. This is then tempered by air handling units before passing throughinsulated ducts to the perimeter of each classroom. The air provides theventilation, heating and passive coolingto the classrooms, which themselveshave exposed thermal mass. The airthen passes through passive attenuatorsto a central space, where it is extractedfor partial recirculation (in winter) ordischarged to outside (in summer). Thisrecirculation also ensures reasonablyhigh supply temperatures in winter,avoiding draughts. This ingeniousscheme has now been constructed as the Paddington Academy, which has been completed this summer. It is expected to deliver exceptionalinternal conditions in a hostileenvironment with low energy use and simple management. A rigorousprogramme of post-occupancy

monitoring and evaluation will be carriedout to ensure that its performance is optimised, and learn more about the behaviour of these systems.

Health and safety

Concerns are raised from time to timeabout the cleaning and maintenance ofundercrofts and earth tubes. The lattercan be dealt with in a similar way todrains, with access necessary at bothends, and preferably a pit at one end.CCTV techniques can be used to checkfor defects, and spray methods can beused to clean the ducts. To ensure thatcondensation is managed efficiently, the ducts need to be laid to a slight fall. Labyrinths are best dealt with by making them tall enough so that they are accessible. If this is notpossible, regular access points intoshallower voids should be provided.In any of these situations, it is crucial to ensure that dirt and vermin ingress is avoided by sealing any inadvertententries to the duct.

One issue with the current undercroftschemes is that while they recover heatduring the winter, and coolth in thesummer, there is a period in betweenwhen the air is likely to be cooled down

by passing across the ground and thenneed to be reheated before entering thebuilding. In future schemes we will allowthe intake air to be taken either from theground source or directly from outside.This will enable the optimum balance ofconditions, and reduce energy use further.

The thermal storage effect of largeunderground ducts can be increasedfurther by placing gabions within them,which provide an increased surfacearea. Furthermore, the roughnesscreates a more turbulent flow.

These ideas are amongst those whichwill develop in the next generation in low energy labyrinth schemes.

So, in summary:

Why do it?

1 To recover heat or coolth fromthe ground, and to reduce energyconsumption.2 To provide future proofing againstclimate change without needingmechanical cooling.3 The energy recovered is renewable.4 If a building is mechanically ventilated,the additional cost of an undercroft or earth tubes can be relatively minor.5 The solutions do not generally involvesignificant technology, and are simple.

What to watch out for:

1 Controls take some time to settledown, and post-occupancy evaluation,control modification and maintenancemodification regimes are critical.2 Make sure your air flow is turbulent,but not so fast as to generate largepressure drops.3 Ensure that clients and occupants are aware of the nature of the system so that they can be ‘on board’ andsupportive as it settles down, particularlyin the first few months of use.4 Allow for adequate cleaning and access facilities.5 Beware the mid season conditionwhen the ground might cool the airdown when in fact, it needs to be warm!

Draft BREEAM score of 76% –highest of all exemplar schemes

Classroom Heating & VentilationAnnual Carbon Emissions

Natural Ventilationbenchmark

Mixed mode withheat recovery

Mech vent withevaporative cooling

Undercroft scheme– gas heating

Undercroft scheme– electric heating

Figure 10 DfES exemplar scheme – now completed at Paddington Academy

34

When considering using labyrinthsor undercrofts to condition supplyair, an accurate assessment of theeffect and feasibility of the strategywith a computational model is vital.Daniel Knott explains the processand the benefits the OpenUniversity’s Jennie Lee project.

The Jennie Lee project is a facultybuilding on the Open University campusat Milton Keynes with a project value of €19m. The inspiration to provide a building with low energy consumptionand excellent green credentials wasintegral to the client and the localplanning authority.

The original aim to provide on-siterenewables was severely limited due to a restricted site. Therefore a systemfor preconditioning the fresh air supplyvia a series of thermal labyrinths,positioned in the building undercroft,was investigated and adopted.

The labyrinths are designed toprecondition external air used for the fresh air supply in the atrium andsurrounding inner offices. The outsideair is drawn through an undergroundsystem by the ventilation plant. Partlyusing the earth as a heat source andsink and partly using its own thermalmass, the labyrinth preheats the air inthe heating season and cools the supplyin the summer months. The technique issuitable for new mechanically ventilatedbuildings with appropriate groundconditions. The main benefit is thereduced peak demand for cooling andheating plant, which helps to reduce the size and cost of the HVAC system.

Buro Happold’s London CoSA(Computational Simulation and Analysis)team was asked to model the coolingeffect provided by the labyrinths for theair supplied to the atrium and internaloffice spaces using a high externalambient temperature. From our initialwork the investigation developed into awider study of the modelling capabilities of the IES Dynamic Thermal Model in predicting the temperature drops

achieved by an earth-coupled systemvia a dynamic temperature boundarycondition (see figures 1 and 2).

Challenges

The site had strict boundaryrequirements, which restricted theposition, depth and orientation of thelabyrinth. The only sub-surface accesswas from the south, as the north sidewas close to a retaining wall, the eastside had underground drainage and a trench occupied the west orientation.

Favourable factors for the use of groundconditioning include average groundtemperatures of less than 12°C andsoft, moist earth. Sacrificing thesepreferential conditions would lead to

smaller temperature gradients betweenthe labyrinth walls and the incoming air, resulting in a reduced cooling orheating effect.

Earth tubes or pipes are typically placedat a depth of 5m and sufficient landarea should be available for the outputrequirements. The recommendeddistance between pipes is 1m.

However, the project finances dictatedthat the labyrinth could not be fullysubmersed at a suitable depth belowthe building. It was instead containedwithin channels in the foundations. Inaddition, due to the cost of excavation,the design team was forced to furtherreduce the amount of earth aroundeach of the ducts, by connecting

Dynamic modelling benefits of Jennie Lee labyrinths

Figure 1 Temperature difference is reduced in both summer and winter for 1m comparative to 5m depth

24

22

20

18

16

14

12

Jan Feb Mar Apr

Temperature at 1m depth

Temperature at 5m depth

Tem

pera

ture

(ºC

)

Earth/Labyrinth coupling

Recommend depth of ground source system

May Jun Jul Aug Sep Oct Nov Dec Jan

10

8

6

4

2

35

the three separate labyrinths. Bothmeasures would decrease the desiredcharacteristics of the labyrinth due tothe lower adjacency areas exposed to the earth.

The distance between the intake and the internal zones suggested that a labyrinthine system of walls andpassages would increase the effectivelength of the undercroft and encouragefluid mixing, leading to an increasedcooling or heating effect. Unfortunatelythe introduction of internal walls will increase the pressure drop andconsequently significantly increase the energy consumption of thededicated extract fan.

Benchmarking

One of the main concerns in this project was to address the lack ofground temperature benchmarks. The determination of a thermal conditionfor labyrinth walls was essential topredict the thermal environment of the labyrinth and consequently thecooling and heating effects used to justify Part L compliance.

Although there is an equation forcalculating ground temperatures(Mihalakakou 1992, 1997), in thisproject the labyrinth and the adjacentearth were also dependent on thethermal characteristics of the buildingabove. For this reason we believed that modelling the thermal mass of the earth, labyrinth and adjoiningbuilding was justified.

The operation of earth tubes andlabyrinths are still not easily predictable,and they vary in success from project to project. This is understandable as thepassive system relies on many variableswhich are in continuous flux and changefrom site to site. The density of the earth,water table levels and sources of heatabove and below ground, all effect theheat sink characteristics of the labyrinth.

If a basic fixed temperature assumption isused for the earth at a certain depth thenthe results will differ vastly to a variable

earth temperature which takes intoaccount the diurnal changes, seasonalvariations, and the soil characteristicpreviously mentioned. At depths of below10m the earth temperature is steadyenough for such basic assumptions, butwith a labyrinth or undercroft constructedat a depth considerably above this, weshould consider a variable adjacent earthtemperature to be more accurate. In thiscase, the location of the labyrinth withinthe building foundation indicated that a variable condition would provide themost accurate results.

Contradictions on the thermal‘benchmarking’ of earth are prevalent.For example, the websitewww.actionrenewables.org states that “In the UK, several metres belowthe surface, the ground maintains a constant temperature of 11-13°C”, while Kensa Engineering argue that “the ground temperature is around10°C, the same as the inside of a fridge but there are obvious exceptionssuch as Bath and Southampton.”

South facing facade contains labyrinth supply

Jennie Lee building (IES model)

Labyrinth undercroft (Flovent model) Labyrinth Outlets

Labyrinth Inlets

N S

E

W

Figure 2 Computerised fluid dynamics model of the Jennie Lee faculty building labyrinth

36

One of the conclusions of the analysisfor the Jennie Lee Building was thattemperature benchmarks for the usein computation models should bepublished by CIBSE or a similarorganisation. The popularity of groundsource heat pumps and labyrinths isnow highlighting the need for somecommon conditions to base analysis on. This is increasingly important, as thepassive cooling and heating achieved by any ground source energy system is considered in the Part L assessment.

Modelling and verification

A Dynamic Thermal Model (DTM) wascreated in the IES Virtual Environmentsoftware. It was used to calculate thethermal characteristics of the Jennie Lee Building, the undercroft and thesurrounding earth down to a 10m depth.The intention was to create a large earththermal mass which reacted dynamicallywith the weather creating varyingboundary conditions for the buildingfoundations and labyrinth system.

The IES DTM was assessed againstempirical data and was found to have a high correlation for all depths. For theJennie Lee Building, the temperatures inthe adjacent earth zones at a 1m depthwere used as the dynamic boundarycondition for the labyrinth walls. Due to the proximity of the ground surface and the ground floor to the building, thetemperature profile varied significantly and closely followed the external ambient.

Conclusions

One of the conclusions from both theempirical data and results attained fromthe IES model simulation is that theearth temperature stabilises around12°C at a depth of 10m, as shown in figure 3. The observed temperaturefluctuations and season shifts follow the external ambient temperature, with a significant time lag due to theground’s large thermal mass. Theanalysis also highlighted the need fora bypass system for periods in the

Design guidance

The original target was for the labyrinthto be fully passive for the majority of the year. Wind and stack effects areexpected to drive the air through thelabyrinth with the fan providing drawonly when the flow rate is not sufficient.In practice, the control of the fan andlarge pressure drops in the labyrinthwill force the fan to run for a higherpercentage of the year.

There are several issues that arise during construction that the constructorsshould be aware of. These include water collection in the labyrinth due to rain water, the subsequent problemsof cleaning ready for use, and the need for compacted earth around thelabyrinth. The execution of the labyrinthconstruction is often overlooked and theobservation and testing of undercroftsonce installed is needed to further ourunderstanding of the success of projects.

Operational risks of condensation, fan noise and earth temperature shouldalso be monitored.

Client: Open University

Architect: Feilden Clegg Bradley

Services: Structural engineering,building services engineering.

0

5

10

15

20

25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Tem

per

atur

e (º

C)

Depth (m)

Stabilisesat 10ºC

Temperature of Ground at various depths (UK) comparison of Eng Geology,

Waltham 1994 with IES result, Knott 2006

1st Feb - IES 1st May - IES 1st Aug - IES 1st Nov - IES 1st Feb 1st May 1st Aug 1st Nov

Figure 3 A ground temperature study showing stabilisation at 10°C

autumn season when the thermal massof the earth causes large lagging effectsin ground temperature. When thetemperature of the labyrinth supply is higher than the external ambient, low level vents in the atrium corridorsare opened to supply fresh air directlyinto the atrium and adjoining zones.

Due to the proximity of the labyrinth to the ground floor slab, the doubleeffect of warmer than ambient groundtemperatures and heat recovery provebeneficial in creating large energysavings in the heating season. Thisrelationship is reduced by introducing a layer of thermal insulation on theceiling of the labyrinth and serves tolimit the undercroft/building coupling.This is a balancing act but the insulationis needed so that the earth heat sinkeffects in the cooling season are notcompromised by heat recovery.

The amount of heat removed from the supply air was greater than 2°C for all external ambient temperaturesabove 24°C. This minimum temperaturedrop of 2°C was used in the Part Lassessment for criterion one and threeand therefore represented a conservativeestimate of the reduced carbon footprintof the Jennie Lee Building.

37

A question of coupling

On the air side there are two maintypes of system commonly used topre-condition supply air – earth-to-air heat exchangers (earth tubes)and thermal labyrinths. However,each option has markedly differingdegrees of earth coupling andoperational characteristics. David Warwick runs through the key points and differences.

Earth-to-air heat exchangers(ETAHE)

An earth-to-air heat exchanger drawsventilation supply air through buriedducts or tubes, as shown in figure 1. As the temperature of the ground below 3m is practically constant, it substantially reduces ambient airtemperature fluctuations. It thereforeprovides space conditioning throughoutthe year, with the incoming air beingheated in the winter and cooled in thesummer by means of earth coupling.

System options

Systems can be driven by natural stack ventilation, but usually requiremechanical means. In some cases air is circulated via air handling units,allowing filtering and supplementaryheating/cooling. A simple controller can be used to monitor inlet and outlettemperatures, as well as indoor airtemperatures. Ground coupling ducts or tubes can be of plastic, concrete or clay – the material choice is of littleconsequence thermally due to the high thermal resistance of the ground.

ETAHE are suited to mechanicallyventilated buildings with a moderatecooling demand, located in climateswith a large temperature differentialbetween summer and winter, andbetween day and night. Location of theducts in sand or gravel below the waterlevel, with moving ground water, givesthe best performance. However, thepresence of ground water involvesextensive sealing precautions.

Size and output

The optimum pipe length is a function of pipe diameter and air velocity. Smallpipe diameters of between 200 and300mm are thermally more efficient – they should be buried at a minimumdepth of 2m and separated by 1-2m to allow heat dissipation. Optimum airvelocity is typically 2m/s.

Under constant load, the coolingcapacity of the ground may becomeexhausted and, therefore, generally it isnot possible to meet high loads. Withhigh loads, two separate duct systemscould be considered – one for use in the morning and one for use in theafternoon. A bypass can be used toimprove the performance of the systemduring periods when the ambient airtemperature can meet the coolingrequirements. In unoccupied periodswhen the ambient air temperature falls below the surface temperature in the ducts, night cooling can be used to pre-cool the system.

The ground temperature is based on‘undisturbed’ conditions. When theducts are installed beneath the building,or even within a built up area, this will beaffected substantially. The effect that theduct has on the ground temperature alsoneeds to be considered. Optimisation ofthe design requires a complete thermalsimulation of the system.

In principle, these are low-cost systems– the excavation is the major part of the installation cost. Maintenance is minimal, but regular inspection andcleaning of the ducts is recommended.

Summary

ETAHEs can be used on new buildingsor refurbishments to provide free coolingin the summer and pre-heating of air inthe winter. They have high capital costs,but over the life of the system will yieldsubstantial savings.

Figure 1 An earth-to-air heat exchanger can be equally well applied to domestic or commercial premises. Diagram courtesy of INIVE

Thermal labyrinths

A thermal labyrinth (see figure 2)decouples thermal mass from theoccupied space, usually by creating ahigh thermal mass concrete undercroftwith a large surface area. Decouplingthe mass means it can be cooled lowerthan if it was in the occupied space.This stored ‘coolth’ can be used tocondition the space for a number of days in hot periods.

Options

The labyrinth layout needs to balanceoptimum thermal storage with the airresistance of the system. Creating airturbulence, by increasing the roughnessand incorporating bends, improves heattransfer. However, incorporating morebends may increase the air resistancebeyond the point where the system can be part of a passive or naturallyventilated scheme.

Thermal labyrinths are suited to new,mechanically-ventilated buildings withcooling demand, located in climateswith a large temperature differencebetween day and night.

Size and output

As labyrinths are often constructeddirectly beneath a building, only the sidesand floor of the labyrinth are in contactwith the earth and the top of the labyrinthis directly coupled with the building. The labyrinth needs to be well insulatedfrom the building to prevent heat transfer.

The earth contact of the labyrinth does give the benefit of steady groundtemperatures. However, the undisturbedground temperature cannot be useddue to the effect of the building and theoperation of the labyrinth. Optimisationof the design requires a completethermal simulation of the system.

A bypass can also be used to improvethe performance of the system. Whenthe ambient air temperature can meetthe cooling requirements of the building,the labyrinth can be bypassed to retainmaximum cooling for use during peakconditions. During the unoccupiedperiod when the ambient air temperatureis low, night cooling is used to ‘charge’the labyrinth.

38

Running costs

Regular inspection and cleaning of the labyrinth are recommended,although thermal labyrinths are generallymaintenance free. The major cost iswhen fan power is required to supply air through the labyrinth.

Summary

Thermal labyrinths can be integrated into the building structure to provide freecooling in the summer and pre-heating of air in the winter. They have high capitalcosts, but over the life of the building theywill yield substantial savings by reducingpeak demand for cooling and heating.

Figure 2 A thermal labyrinth can provide a substantial amount of free cooling from stored ‘coolth’, as well as pre-heat supply air in winter

39

Double couple at Robert Burns Museum

The redevelopment of the Robert Burns Museum offered anopportunity to use two earth couplingsystems – a ground source heatpump (closed loop) and earth tubes.Scott Baird explains the advantagesof earth coupling HVAC systems on the water circuits and air-side.

Sustainable building design should aimto provide a balanced solution, offeringoptimum working/living conditionsalongside reduced environmentalimpact, both now and in the future.When you take the complete buildinglifecycle into consideration, there aremany factors involved; from the locationof the building, its design, subsequentoperation and maintenance, to theconstruction materials and practicesused, and how any future changes of use are addressed.

But energy consumption is an overridingconcern for building services engineers.This mostly relates to the efficiency of the building in use, with the mainmeasure being the carbon emissions for the building.

Previously the only way to sell this toclients was to demonstrate the economicefficiency of the design (whether this included renewable/alternativetechnologies or purely energy efficiencymeasures) in the hope they would investthe capital for the future and long-termdelivery of the project. However, newregulations have changed this and thereis ever-increasing consumer and politicalpressure for the construction industry to become more sustainable.

At Buro Happold for many years we have been challenging ourselves to review and, where applicable,integrate energy efficient practice and renewables/alternative technologyinto as many projects as possible.

We still need to question how far we are taking our research as engineers.Are we fully understanding thelong-term local and global impacts,

or are we simply providing a solution to one issue without fully understandingthe other impacts this may have?

For instance at the Robert BurnsMuseum we have looked at the viabilityof various technologies and systemapproaches. After a considerable amountof analysis we developed two sustainablesystems for the project ground sourceheat pumps (GSHP) and earth tubes.

Earth coupling

Museums are generally energy intensivedue to the very onerous environmentalconditions required for artefacts andexhibits. With this in mind, we evaluatedthe existing Burns collection to identifyany opportunities to house these withinsmaller volumes.

23.5

23

22.5

22

21.5

21

20.5

20

19.5

19

18.5

18

17.5

17

Out

let A

ir Te

mpe

ratu

re,o C

Tube Length, m

10 20 30 4 0

Ground Coupled Ventilation Systems, Outside air 24.8 oC, @ 4.0 m3/s

20 Tubes @ 0.25m Diameter 30 Tubes @ 0.25m Diameter 40 Tubes @ 0.25m Diameter 20 Tubes @ 0.5m Diameter 30 Tubes @ 0.5m Diameter 40 Tubes @ 0.5m Diameter 20 Tubes @ 0.75m Diameter 30 Tubes @ 0.75m Diameter 40 Tubes @ 0.75m Diameter

Ground Coupled Ventilation System Supply Air Conditions (30 number, 30m, 0.25m diameter tubes)

30

25

20

15

10

5

0 3 1 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71

Hour (3 days)

Air

Tem

pera

ture

, o C

Outside Air Supply Air

Figures 1 and 2 Computer modelling helped optimise the size and length of the earth tubes

40

Figure 3 Monthly heating load characteristics

As the Robert Burns collection waslargely manuscripts of his poetry andsongs, there was potential to house themajority of it within museum displaycases. Display cases are designed with a silicon drawer mounted below the exhibit. The silicon gel will absorb or release any humidity build up withinthe display case to retain a relativelysteady state humidity level. As the casesare not fully sealed, the temperaturewithin the case is controlled from the air within the main exhibition space.

The conditions required within the mainvolume of the exhibition areas were thenable to be relaxed, which would providelong-term energy savings for the client,while allowing a more sustainableservicing approach to be adopted for the exhibition areas.

From a low energy sustainable designapproach, it was thought that the bestmethod would be to passively ventilate the building. This is not standard practicewithin a museum as they usually havecontrolled facades with little or no glazingand can be relatively deep plan inconfiguration. There was, however, anopportunity to investigate the possibility of labyrinth ventilation or earth tubes.Through a period of investigation a numberof stumbling blocks appeared for thelabyrinth ventilation, including possible gasissues, substructure depths and so on.

Earth tubes were considered a betteroption. The principle of earth tubes is to bury a pipe made from materials withgood thermal transfer properties at adepth of 2m or so where the groundtemperatures are constant all year. As the air is drawn through the earthtube it is either pre-heated or pre-cooled, depending on the season.

The strategy uses an earth tubenetwork to provide partially passive, low energy ventilation. For the exhibitionspaces the air supply systems will be the primary source of heating or coolingthrough heating/cooling coils within theair path of the earth tubes. This will also

minimise the need for any wet servicesin the exhibition areas where artefactsare located.

Ventilation within the exhibition areas is achieved through natural ventilationusing the buoyancy of the air rising tohigh level extracts within the exhibitionareas. This in turn pulls the air throughthe earth tubes to make up for the airthat has been extracted. In periods oflow external pressure, low-speed fansinduce the air through the earth tubesinto the exhibition areas. Other areas of the building, with the exception of the toilets and kitchen, are naturallyventilated. The earth tube ventilation

system supplies the vast majority ofventilation.

The comfort-controlled exhibitionspaces are serviced via ten ventilationearth tubes and low-speed supply fans. The earth tubes run in soft groundwhere air can easily absorb heat or cool due to surface contact with thesoil. The tubes run from the integratedarchitectural landscaping feature to thesunken external plantroom where air is treated and supplied to the building.

Supply fans in the sunken plantroom are mounted in line with filters, coolingcoils and LTHW heating coils.

Figure 4 Monthly cooling load characteristics

41

Figure 5 Museum heating duration curve

Fans only operate when there isinsufficient wind pressure or naturalbuoyancy to allow the air to be passivelypulled through the earth tubes.

To improve the efficiency of the earthtubes it was also recognised that airturbulence within the earth tube wouldallow maximum thermal transfer to takeplace between the solid surfaces of the earth tube and the air passingthrough. Turbulence was generated by introducing bends in the earth tube paths before they entered thesunken plantroom.

After the air is treated within the sunkenplantroom, it is transferred throughsupply branches to insulated floor voids.Supply duct branches terminate in thefloor void, creating a positive pressure.Supply diffusers will be mounted in thefloor of the comfort controlled exhibitionspaces, allowing the air to be displacedinto the room. A solar thermal wallmade from fire clay brick has beenprovided in the exhibition area toimprove the thermal mass and providefurther stability to the rate of changewithin the museum environment.

From the modelling carried out it wasconsidered that for a 25m length buriedat 2m, a 4°C temperature differencecould be achieved. This would allow thepre-heat or pre-cool to reduce the coilsizes and loads so that the extremes and plant sizes could be vastly reduced.

Smaller diameter tubes wereinvestigated, as these would provide amore efficient transfer of energy due tothe greater air to solid surface contact.However, having smaller diameter earthtubes also means that more of them areneeded to provide the same volume ofair, and the spatial requirements wouldincrease due to the spacing betweenearth tubes. In the end, based on theavailable areas, we were able to use ten500mm-diameter (internal) earth tubes,leaving a 2m spacing between tubes.

Geothermal energy

Initial feasibility calculations showed thepossibility of applying a vertical closedloop ground source heat pump (GSHP)to provide heating and cooling for thenew Burns Museum. We examined both the adoption of GSHPs to meetpart of the heat load (in conjunction with supplementary heating), and aGSHP sized to meet the entire load.

A short review of the geology andbuilding heating and cooling loads wasfollowed by a simulation of a number ofGSHP systems. A review of the possiblesize of the ground loop heat exchanger,external area required and theoperational savings, both in terms of running costs and carbon dioxide,was provided from the model.

To analyse the geology down to 100m,the depth typified by the installation of a vertical GSHP system, it is usuallypossible to use data from the BritishGeological Survey (BGS). Unfortunatelythere are no suitable deep borehole logsin the vicinity of the site.

Using a combination of local geologicalmaps and literature we assumed thatthe prevalent bedrock is sandstone withbands of Westphalian coal measures.

Whilst coal has a relatively poor thermalconductivity, sandstone has a highervalue more suited to higher performanceinstallations.

However, as there is some doubtregarding the exact geological sequenceand relative depths of the respectivestrata, it was considered that a thermalconductivity test by a specialist GSHPcontractor was needed to confirm thesuitability of the site for a closed loopGSHP. This reduces the cost risk in theprocurement of the system and, if theground conditions are deemed suitable,enables the system to be optimisedusing insitu data.

Heating and cooling review

A dynamic thermal model was completedfor the building using the buildingsimulation software, IES. A summary of the monthly heating loads are shownin figure 3, whilst the cooling loads are shown in figure 4.

It can be seen that the heating loads,both in terms of peak (kW) and energy(kWh), are far greater than the coolingloads. Using the example weather yearfor Glasgow, the peak heating load of319kW occurs in January, whilst thetotal annual heating energy simulated is 386,934kWh.

Hea

ting

Cap

acity

(kW

)

Hours

42

The peak cooling load of ~45kW occurs in August and the annual cooling energy required is 2,550kWh. As a consequence, the GSHP designwill ultimately be led by the heatingrequirements in the new building.

The building heating profile can be further analysed by plotting the heat duration curve, which is shown in figure 5. This highlights theinfrequency of some of the higherheating loads; particularly above 75kW.Although 75kW is less than 25% of the total peak load, over 83% of thetotal heating requirements over the year can be provided with this capacity.

GSHPs are generally more expensive perkW installed than conventional systems.From this basis, and because the higherheat loads can be infrequent, it issometimes more cost-effective to reducethe capacity of the GSHP and to usecheaper plant to provide supplementaryheat at colder times of the year. This is often called bivalent heating, as opposed to monovalent where the heat is provided from one sourceonly. This way, the capital costs can be minimised whilst the majority of theoperational benefits can still be realised.

Figure 6 shows the heat duration curves for a monovalent and an examplebivalent mode, where the GSHPprovides 75kW of the heating load andthe remainder of the load is providedwith gas-fired boilers. This gives anexample of how the proportion of theheating can be split in the two modes.

GSHP analysis

Generally larger GSHP systems aremore suited to balanced heating andcooling loads, so heat abstraction fromthe ground loop heat exchanger (GLHE)in the heating mode can be replenishedduring the summer months, ie throughheat rejection in the cooling mode.However, GLHEs can be optimised to allow for heating only or heatingdominated applications, by ensuringthat both the cumulative length isadequate and the spacing between the boreholes is adequate to minimisethermal interference and ensure thermalcapacity in the long-term. The sizing of the GLHE is very important, as this is the greatest proportion of the totalcost of the GSHP.

This study firstly simulates the effect on the cumulative length of the GLHE

in monovalent mode with two differentborehole spacings, and then twobivalent systems with two differentborehole spacings. The differentsimulations are summarised in table 1(see page 43).

To enable the simulation, the followingassumptions have been made:

■ Ground thermal conductivity of 2.3W/mK

■ GLHE return temperature never falls below -2ºC in the heating mode to ensure continued systemperformance

■ Flow rate of ~0.15m3/hour/kWextracted to be maintained at all times

■ Each borehole is 100m deep, which may change depending on the bedrock and the consequencedrilling conditions.

Each simulation will be run for 20 yearsto ensure long-term performance of the system.

Results

Ground loop heat exchanger length

This is the cumulative length of theboreholes required for each proposedGSHP system. The results of thesimulation for the different modes are shown in figure 7.

Figure 7 shows the benefit in this case of maximising the borehole spacing dueto the heat dominated load. The shortestcumulative borehole length is 4,350mfor the 75kW bivalent system with 8mspacings. Assuming a nominal length of100m for each borehole in the GLHE, thisequates to 44 boreholes. The monovalentGSHP system has been calculated to be only 1,700m longer or, assuming a nominal borehole length, requiring only 17 further boreholes. There is very littledifference between the 125kW bivalentand 325kW monovalent system.

Figure 6 GSHP heat duration curve – monovalent versus bivalent

Hours

Hea

ting

Cap

acity

(kW

)

43

External space requirements

The external space required for thethree suggested systems is summarisedin figure 8.

This highlights the extra area required for the monovalent systems, but alsobetween different borehole spacings.Although fewer boreholes are requiredfor the respective 8m spacing systems,extra external area is still required.

Operational savings

In each case, the simulations alsoenable the electricity needed to run theGSHPs to be calculated. For this set ofcalculations the following assumptionshave been made:

■ Electricity unit price: 8p/kWh■ Gas unit price: 2.5p/kWh■ System efficiency gas fired

heating: 85%■ System coefficient of performance

conventional electric chiller: 3.

The calculations can be compared to a conventional gas-fired heating andelectric chiller cooling plant, as shown in figure 9. In the case of each bivalentGSHP system, conventional plant,efficiencies and utility prices are used to provide supplementary heat. No supplementary cooling is required as even the smaller bivalent GSHP cancover the entire cooling load throughoutthe year, as this does not coincidesignificantly with heating loadselsewhere in the building.

As expected, the savings are verysimilar for both borehole spacings foreach system so an average reductionin sterling and in percentage terms is shown in each case. The savingsbetween the different systems areless marginal, with the monovalentsystem offering the greatest potential for operational savings at 27% and£3,105 per annum.

Configuration Borehole Heating Cooling Supplementaryspacing capacity capacity heating capacity

Monovalent

Compact GLHE 6m 325kW 45kW 0kW

Low density GLHE 8m 325kW 45kW 0kW

Bivalent





Table 1 Simulation parameter summary

Figure 8 External area: comparison

Figure 7 Ground loop heat exchanger lengthcomparison

44

Carbon dioxide savings

Figure 10 shows the simulated carbondioxide emissions that will be realisedwith each GSHP system and also the conventional plant total. There isagain very little difference between theestimated reduction in emissions fordifferent borehole spacings for eachsystem so an average is shown in each case. All the GSHP systems show a possible reduction of over 40%versus a conventional system, with themonovalent option offering the largestsaving at 48% or approximately 42 tonsCO2 per annum.

Conclusions

In conclusion, the adoption of a GSHP at the Burns Museum offers the potentialto reduce both the operational costs and carbon dioxide emissions of thespace heating and cooling element ofthe building. All the systems modelledoffered significant carbon dioxidereductions of over 40% versusconventional plant.

In addition, there are significantestimated operational savings of over20% in each case. The earth tubes will significantly reduce the ventilationload, both in terms of fan energy and heating/cooling input. It is likely that 60% of the cooling costs will beremoved, as the pre-cooling providedfrom the earth tubes at peak externaltemperatures will reduce the airtemperature to a suitable level for the exhibition area.

The building space conditioning isdominated by the heating requirement.A large borehole spacing has significantbenefits in terms of the number ofboreholes needed, even if this doesmean a larger external area for theborehole field.

The simulations highlighted that thereduced capacity of the bivalent GSHPapproaches did not match an equalrelative reduction in the size of theground loop heat exchanger and the resultant space needed.

The heating profile remains relativelyhigh, even during unoccupied periods,due to the sensitivity of some of theexhibits. Therefore, there is reducedbenefit in applying a bivalent GSHP at the site, due to the high frequency of heating loads at approximately 30% of the peak capacity. This wouldimprove if the heating requirementduring unoccupied periods were toreduce significantly. This limited potentialis also exaggerated by the imbalancebetween heating and cooling loads over the year.

The GLHE is the most expensiveelement of almost all GSHP installationsso this aspect will undoubtedly reducethe cost effectiveness of the bivalentapproach. In addition, the bivalentapproach will add complexity to thesystem and supplementary heatingplant will still be needed.

To confirm the optimal approach it isadvised that both the bivalent 75kWand monovalent 325kW GSHP systemsare costed, including estimations for any supplementary plant that will still be needed. The results of this mayconfirm that the monovalent approachis the most cost-effective way to adopta GSHP for the new building. After the insitu thermal conductivity test is carried out to confirm the thermalproperties at the site, an informedjudgement can be made.

Figure 9 Estimated annual operational savings per annum

Figure 10 Estimated carbon dioxide savings per annum

Client: The National Trust for Scotland

Architect: Simpson and Brown

Services: Building services engineering,structural engineering, groundengineering, asset management.

45

Our authors

Daniel KnottBA MSc

A graduate engineer,Daniel has applied hisanalysis skills towards air flow modelling,displacement and naturalventilation as well asbuilding services of high rise buildings and wind modelling.

David WarwickBEng

David is a buildingservices engineer with an interest in the efficientintegration of responsivedesign elements withinadaptive buildings.

Steven WilliamsonBEng CEng MCIBSE

Steven is a highlyexperienced buildingservices engineer andteam leader who hasbeen involved innumerous large highprofile projects.

Mark OwenOND HNC CED HVAC IEngMCIBSE MASHRAE

With extensive experienceas a building servicesAssociate, Mark has been involved in diverselarge projects in the UKand abroad, and has a special interest in arts and cultural buildings.

Scott BairdBEng BSc

A building servicesdesign engineer of wideranging experience,Scott’s main areas ofinterest lie in efficient and sustainable buildingdesign with particularregard to whole life-cycle design andenvironmental impact.

Mike EntwisleMA PhD CEng MCIDSE MEI MASHRAE

An innovative designengineer, Mike’s particularareas of interest includepassive design, naturalventilation, renewableenergy, building form andalternative cooling sources.

James DickinsonBEng AIMechE

An environmentalengineer who specialisesin ground sourceengineering and theintegration of renewableenergy services andsustainable developmentinto building design.

Alan ShepherdPE CEng MCIBSEMASHRAE BEng MSt

Alan is an experiencedthermal modelling andsystem analylist with aparticular interest in theintegration of low energytechnologies.

Jason GardnerMSt BEng CEng MCIBSE

A building servicesengineer with extensiveexperience in designinghigh quality and carbonneutral environments.

Edith BlennerhassettBEng CEng MCIBSE MIEI

A civil and structuralengineer by training,Edith has howeverworked as a buildingservices andenvironmental engineer since joiningBuro Happold where she has realised her passion for integratedbuilding design.

Brian DoranBEng MSc CEng MinstE

A building servicesdesign engineer of wideranging experience.Brian’s main areas ofinterest lie in efficient and sustainable buildingdesign with particularregard to whole life-cycle design andenvironmental impact.

Patterns

Ground

Source

Energy

Buro Happold

Camden Mill

Lower Bristol Road

Bath BA2 3DQ

Tel +44 (0)1225 320600

www.burohappold.com

Bath

London

Belfast

Birmingham

Edinburgh

Glasgow

Leeds

Manchester

Newcastle

Abu Dhabi

Berlin

Boston

Copenhagen

Dubai

Dublin

Kuwait

Los Angeles

Moscow

New York

Pune

Riyadh

Toronto

Warsaw

Patterns is Buro Happold’s in-house

technical journal. Each issue in this

occasional series contains in-depth essays

on an industry subject or topical theme.

For details of how to obtain additional

copies, please go to the News section

of our website at: www.burohappold.com

Date post:	13-Mar-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

9285 Buro Patterns st20 AW - Designing Buildings · 2013-08-07 · energy, see figure 1(opposite)....

Documents