PROCEEDINGS, Thirty-Eighth Workshop on Geothermal Reservoir Engineering Stanford University, Stanford, California, February 11-13, 2013 SGP-TR-198

DEVELOPMENTS IN THERMAL PILE AND THERMAL TUNNEL LININGS FOR CITY SCALE GSHP SYSTEMS Nicholson D. P., Chen Q., Pillai A., Chendorain M. Ove Arup and Partners Limited 13 Fitzroy Street London, W1T 4BQ, England [email protected]

ABSTRACT The city scale development of horizontal trench closed loops and vertical borehole loops for ground source heat pump (GSHP) systems is limited by the lack of ground surface space. This paper discusses the use of thermal piles below city buildings. Consideration is given to the development of standards that maintain compatibility between the M&E, geothermal, and structural designs. The monitoring requirements are discussed. Cities also use tunnels to provide transportation and utility networks. Some of these tunnels must be cooled to meet operational requirements. This paper discusses the use of system-wide thermal tunnels with closed loops embedded in the tunnel segments. Water is circulated through the loops to cool the tunnels. The geothermal design of the loops and the impact on the structural performance is considered. The heat extracted from the tunnel is then used with GSHP systems to heat adjacent buildings via a district heating system. When planning a thermal tunnel system the future use of the heat for buildings must be investigated.

1 INTRODUCTION The limited availability of open land in cities reduces the use of horizontal trench closed loops for ground source heat pump (GSHP) systems. Vertical borehole closed loops systems use space more efficiently, but still require open areas next to buildings. City developments are often high rise buildings with piled foundations and basement car parks. Thermal pile closed loop systems provide another type of ground heat exchanger for GSHP systems. The pile depths are generally controlled by building structural loads rather than the heat storage requirements and therefore for tall buildings thermal piles are best used in combination with other heating and cooling systems. The design, installation and

operation of thermal pile systems are more onerous than vertical borehole loops because of the temperature effect on the piles structural performance. This paper considers the recent develops in thermal pile design and installation standards to ensure compatible structural and geothermal designs. In cities, tunnels provide access for rail, road and utilities. They can also be used as ground heat exchangers for GSHP systems. These tunnels can be considered as cold and hot tunnels. The cold tunnels access heat from the surrounding ground and the air inside the tunnel is similar to the ground surface temperature. The hot tunnels access heat from the surrounding ground and also equipment inside the tunnels, such as trains and HV cables. These hot tunnels also require air ventilation for cooling to keep the tunnels within the operational temperature range. This paper considers the design of thermal energy segments (TES) for thermal tunnels. These segments are installed system wide and therefore studies are required for future connections to surface and district heating systems to supply heat to adjacent buildings. 2 THERMAL PILES This section discusses the background to the UK‟s development of a range of thermal pile installation methods. The contractual arrangements for designing and constructing thermal piles are discussed. Full scale pile tests have been undertaken and this has lead to the development of design methods. 2.1 Background The thermal (energy) pile development in the United Kingdom is summarised in Table 1. Early thermal pile designs and construction work relied on Brandl (2006) and Austrian contractors to assist UK piling contractors. However, from about 2005 the design and construction work was largely undertaken by Cementation Skanska Ltd. and Geothermal International Limited. Figure 1 shows the UK‟s

increased use of thermal piles between 2005 and 2010. Table 1: Milestones for thermal pile development in UK Date Milestone Reference 2002 Brandl - Rankine lecture Brandl (2006) (2002) 2002 - Early projects – Keble Suckling 2005 College (2004) 2002 - PII - Ground storage of Arup (2005) 2005 heat energy 2005 Cementation / GIL Amis et al onward projects (2009) 2007 Lambeth College pile Bourne-Webb test, (2007) et al (2009) 2010 NHBC Piling for Houses NHBC (2010) Guide 2012 GSHPA Thermal Pile GSHPA Standard (2012)

the pipes around the perimeter of the piles, see Figure 2. Consideration is also given to the impact of freefalling aggregate into the pile.

Figure 2: Short cage with borehole thermal loop: a) Bundled, b) Lantern spacers to place pipes round pile perimeter, c) Borehole connecter showing minor impact damage from falling concrete In continuous flight auger (CFA) piles, double U tubes have been pushed into the fluid concrete using a central 32mm bar to depths of up to 25m.

Figure 1: Thermal piles installed in UK (2005 – 2010 In the UK and USA, Cementation Skanska Ltd have the UK Trademark No. 0606293.9 and the US Trademark No 77704419 on the term “energy pile”. Therefore the UK industry has adopted the term “thermal pile” to avoid conflict with trademarks. 2.2 Installation The early bored, cast in-situ thermal piles used full length cages with plastic pipes attached inside the reinforcement cages. The small pipes were bent around tight radii at the top and bottom of the cages. Since about 2005, the borehole U-connectors have been used at the top and bottom of the cages to maintain pipe continuity. Since 2010 the pile cover has been increased and the pipes are being mounted on the outside of the cages. This simplifies the thermal loop installation on long cages which have to be spliced together. The use of U-connectors allowed bundled pipes to extend below short cages in dry bored piles. Since about 2010, lantern spacers have been used to place

2.3 Contractual Responsibilities In the UK the ICE (2007) Specification for Piling and Embedded Retaining Walls (SPERW) is often used for pile design. This considers both Engineer and Contractor designed piles. The GSHPA (2012) have produced a Thermal Pile Standard which extends the SPERW to incorporate the roles and interfaces for the M&E and Geothermal Designers, see Figure 3. The construction interfaces during construction for Piling, Geothermal, Ground-works and M&E Fit-out Contractors are also shown. There is a need for a Main Contractor coordination role. The roles and responsibilities of the different organisations during the design and construction process are shown on Figure 4 for Engineer designed piles.

Figure 3: Roles and responsibilities for engineer designed piles

2.5 Thermal Pile Design The tasks for the thermal pile design team are summarized in Figure 6.

Figure 4: Design and construction process for engineer design piles 2.4 Thermal Pile Testing As part of the site investigation, the conventional borehole thermal response test (TRT) has been extended to assess pile thermal effects. The borehole is about 300mm diameter and backfilled with concrete to form a mini pile. Extensometers, strain gauges and piezometers have been installed to assess the thermal expansion, concrete stresses, and pore pressures induced in and around this mini pile. These can then be back analysed and scaled up to the contract piles. The test also provides data on the thermal conductivity of the soils. As part of the contract work, full scale thermal pile testing has been undertaken on well instrumented bored piles. The main example in the UK is Lambeth College, described by Bourne Webb et al (2009). The static loading, cooling and heating cycles are reproduced in Figure 5. This information has been used to develop the thermal pile design processes.

Figure 5: Lambeth College pile test

Figure 6: Tasks undertaken by designers The M&E Designer is responsible for defining the annual heating and cooling profile to the heat pump. This must consider variations in summer heat wave and winter cold periods. The Geothermal Designer for the ground source system is responsible for assessing the number of thermal piles that are needed to keep the temperature above freezing. The GSHPA (2012) Standard requires that the pile soil interface does not drop below freezing point. It recommends a minimum circulation fluid temperature of 2°C. This temperature may be reduced where detailed modeling is carried out on transient effects and pipe cover / spacing. The maximum fluid temperature will depend on the stresses induced in the pile but may be up to 40°C. The temperatures should be agreed with the pile designer. Conventional borehole loop design packages such as Earth Energy Designer (2012) or Pilesim (Pahud, 2007) have been developed to include pile effects and assess the variation of fluid temperatures with time. The Pile Designer is responsible for assessing the thermal effects on the pile. Where the pile is fully restrained, additional compression stresses will be mobilized at the head of the pile. Where the pile is unrestrained by the surrounding soil there will be movement of the building, see Figure 7. In practice, the pile behavior will be between these extremes. Arup have developed a thermo-hydro-mechanical finite element soil model using the LS DYNA code to assess the heating of the pile and the surrounding ground. The changes in stresses and the load bearing capacity of the soil are assessed. The changes to the water pressure in clay soils with time are also considered. This model has been calibrated against the Lambeth College pile test. In addition, the OASYS PILE program has been modified to incorporate the expansion of the pile and restraint from the surrounding soil (Bailie, 2012). Some

results for the end of cooling and heating stages of the Lambeth College pile test are shown in Figure 8. This work has been undertaken with Cambridge University.

Figure 7: Restrained and unrestrained pile model

2.6 City Scale On a city scale, the existing London underground system has experienced a long term increase in the temperature within tunnels and in the surrounding ground, Botelle et al (2010). The Crossrail project has introduced thermal piles and walls systems into the station boxes. These will be available for heating the overlying site developments above the stations and will help to cool the ground and the adjacent underground. In the UK at present, only open GSHP systems, which use advection, are regulated to achieve a balanced annual heating and cooling load. The schemes must be approved by the Environmental Agency to ensure that existing abstractors are not affected. Closed systems rely on conduction and are not regulated. 3 THERMAL TUNNEL LININGS This section discusses the background to the development of thermal tunnels using thermal energy segments (TES) in the UK. Additional background information is given in the paper by Franzius and Pralle (2011).

Figure 8: Lambeth College back analysis – Measured and predicted pile axial loads The seasonal expansion and contraction of the thermal pile has also been considered, see Figure 9. This is compared with the vertical cyclic loading of piles and the effect on factor of safety, Poulos (1989). The accumulative pile head settlement has also been considered based on Matlock and Foo (1980). Details are given in GSHPA (2012).

3.1 Concept The concept for the thermal tunnels is shown in Figures 10 and 11. The TES use plastic pipes embedded within either the steel cage or fibre reinforced segments. The plastic pile is PEXa grade to enable tight bed radii to be formed and ensure 100 year durability at 15 bar pressures. The pipes also enable permanent mechanical connections to be formed in the segment box-outs, see Figure 11. The box-out also allows for the pipe extension / compression if joint rotation occurs.

Figure 10: The thermal tunnel concept Figure 9: Seasonal cyclic expansion of pile leads to increased movements

surrounding ground temperature. Short-run tunnel air temperatures are often controlled by the surface air temperature and are called “cold” tunnels, see Figure 13. Longer metro and cable tunnels develop high internal air temperatures and are called “hot” tunnels, see Figure 13. For example, the motors in a London Crossrail trains will emit about 1MW of heat and one train passes at 2.5 minute intervals during peak times. Air conditioning adds an extra 0.1MW. At peak times this is about 22W/m2 and on a weekly average this is 13W/m2. Braking adds to the heat generation. The ground permeability and hydraulic gradient can also affect the thermal tunnel efficiency.

Figure 11: Thermal energy segments and box-out connections The ring to ring connections have to allow for the segment rotation to allow the tunnel to be steered. This flexibility is provided by using plastic pipes mounted on the segment surface. Intermittent access needs to provide for the header pipe connection, see Figure 11. The header pipe connections occur at about every fifth segment ring. A reverse return system is used to balance the head loss around all the ring circuits, see Figure 12. The header pipes transfer the flow back to the surface connection points at between 250m and 400m centres. Larger spacing of connection points should be avoided to control hydraulic head losses. The surface connection points can be provided at shafts, ends of the stations, cross passages or possibly from adjacent boreholes, see Figure 10. Where possible, a plant room is provided at ground level to incorporate circulation pumps and de-airing systems. Heat exchangers can be provided if the tunnel system is to be hydraulically separate from the building circulation system.

Figure 13: Cold and hot tunnels The thermal tunnel design is based on modeling the heat transfer from the tunnel air and the surrounding ground to the water filled pipes in the segments. The LS DYNA numerical model used to assess this transfer is shown in Figure 14. The model has been calibrated against periods with TES heat extraction rates of zero, 10W/m2 and 30W/m2 and the effects of the circulation water are modeled, see Figure 15. The model includes the heat load from the trains and the temperatures in the surrounding ground.

Figure 12: Header piles with reverse return and surface connections 3.2 Design Considerations The efficiency of the thermal tunnel is influenced by the air temperature inside the tunnel as well as the

Figure 14: TES model showing heat flow from tunnel to pipes and soil

Figure 17: Comparison of maximum principal stress distribution in segments during summer Figure 15: Temperature variation in the pipe – for zero, 10 and 30w/m2 continuous extraction rates The stress in the concrete segments and the surrounding ground was also studied in LS DYNA using a Thermo-Hydro-Mechanical (THM) soil model, see Figure 16. Some results are shown in Figure 17. These results highlighted that the hoops stress increased by about 7% due the heating and expanding the lining and the ground. compared with the „no heating‟ case. The circulating water also induced thermal gradients and stresses across the segments. This lead to a further 2% increase in hoop stress. Local tensile stresses were induced next to the pipes due to the cooling effects.

Figure 16: FE model used for stress analysis of tunnel segments with pipes and surrounding soil

A tunnel heating and ventilation program was used to calculate the combined effect of trains running through the tunnels. It included allowances for heat given off at station stops, braking heat during entry to the stations and acceleration heat when exiting stations, see Figure 18. The heat extraction from the TES was also included and some results are shown in Figure 19. These show that in summer the tunnel air temperature is reduced by about 5°C with the flow and return water circulation water temperatures of 21 and 26°C in central London. This leads to a heat transfer of about 10W/m². These water circulation temperatures are high enough to be used with a heat pump for domestic hot water.

Figure 18: Tunnel ventilation model extended to include TES heat extraction

Figure

19: Predicted tunnel air and pipe temperatures along Crossrail eastbound tunnel during summer peak hours, with and without TES installed throughout the tunnel

The effect of train fire on the TES system was also assessed. The majority of the segment pipe work is embedded at sufficient depth to avoid fire effects, see Figure 20. The pipes near the surface and at box-outs were not possible to protect. Instead the short-run tunnel length where a fire occurs would be isolated from the system and not reinstated. The smoke and fumes given off was assessed and considered to have little effect. This is because ventilation systems should have sufficient capacity to extract the smoke.

Figure 20: Fire modelling – Segments and pipes and header pipes 3.3 Buildings connections and city scale development The thermal tunnel system requires an early investment when the tunnels are being built to provide pipe work for heating adjacent buildings. This requires a city scale investment in long term heat supply. For Crossrail, a preliminary study was carried out for the buildings in a corridor extending 100m beyond the tunnels. These buildings were assessed using the London heat map and aerial photos. The buildings were divided into 3 tiers:Tier 1: 34 hotels, large residential, hospitals Tier 2: 4 schools, colleges, libraries, museums Tier 3: 327 offices, leisure centres, retailers An example of part of the Crossrail route map with building heat mapping is shown on Figure 21.

Figure 21: Comparison of Crossrail header pipe access points and building heat map study Based on header pipe access points connecting to 500m of twin tunnels, the heat output would be about 200 to 600kW for 10 to 30 W/m2 of tunnel surface heat extraction. Examples of header pipe access points are also shown on Figure 21. On a city scale, an Energy Supply Company (ESCo) could be used to market the distribution of this heat through local district heating networks. 5 CONCLUSIONS The main conclusions associated with thermal piles are:  The development of thermal piles in the UK is reviewed. The GSHPA Thermal Pile Standard incorporates recent experience and provides basis for their future use.  Thermal pile installation and testing has progressed from long to short reinforcement cages and installation into CFA piles.  The contractual roles and responsibilities for the pile and geothermal designers are becoming clearer and are linked to the existing piling contracts. The interfaces during construction are linked to handover testing.  The structural design methods are being developed to assess concrete stresses and pile settlements due to temperature changes.  At a city scale, thermal piles are used where there is insufficient space to install vertical GSHP borehole loops. They can support metro systems where cooling the tunnel is becoming increasingly important. The conclusions associated with thermal tunnel linings are:  The thermal tunnel concept and key components are explained.  The design is linked to hot tunnels where heat is extracted from the tunnel air and transferred to the surface for domestic hot water production and heating.



 

The DYNA modeling of the tunnel heat transfer and segment concrete stresses is explained. In addition the conventional tunnel heating and ventilation system model is extended to include the heat extracted by the circulating fluid. The impact on cooling the tube is considered. Fire protection is discussed but found not to be appropriate for the surface pipe work. At a city scale, investment is needed in the tunnel pipe work. Early studies are required to assess the way the buildings use tunnel heat and to make the business case.

ACKNOWLEDGEMENTS The work by the GSHPA thermal pile subcommittee is acknowledged in the standard development. The research work carried out by Cambridge and Southampton universities has provided a framework for instrumentation, monitoring and design. The thermal tunnel work required a multi discipline approach from within Arup / Atkins Crossrail team. The continuing support to the project by Crossrail is recognized, together with Mott MacDonald‟s input on the H&V modeling and Rehau‟s work on the pipework design.

Brandl, H. (2006), “Energy foundations and other thermo-active ground structures”, Geotechnique, 56 (2), 81–122 Earth Energy Designer. (2012), http://www.buildingphysics.com/index.htm. Franzius, J. N. and Pralle, N. (2011), “Turning segmental tunnels into sources of renewable energy.” Proceeding of ICE, Civil Engineering, 164, 35-40. GSHPA, (2012), “Thermal Pile Design, Installation and Materials Standards”, Ground Source Heat Pump Association, http://www.gshp.org.uk/GSHPA_Thermal_Pile_ Standard.html Institution of Civil Engineers, (2007), “The Specification for Piling and Embedded Retaining Walls, 2nd edition”. Matlock, H., and Foo, S.H.C. (1908), “Axial analysis of piles using a hysteretic and degrading soil model”, Proceedings of Conference Numerical methods in offshore piling, ICE, London, 127133.

REFERENCES

NHBC Foundations. (2010), “Efficient design of piled foundations for low rise housing design guide”, NF21 ISBN 978-1-84806106-4.

Amis T, Bourne-Web P, Amatya B. (2009), “Geothermal Business Bouyant”, Geodrilling International, 24 March.

Ove Arup & Partners Ltd. (2005), “DTI Partners in Innovation. Ground Storage of Building Energy Overview Report”, May, PII Ref: O-02-ARUP3.

Bailie P. (2012), “Sustainable Demands”, Ground Engineer, 22 September.

Pahud D. (2007), “PILESIM2: Simulation Tool for Heating and Cooling Systems with Energy Piles or Multiple Borehole Heat Exchangers”. User Manual, ISAAC –DACD – SUPSI, Switzerland.

Botelle, M., Payne, K., Redhead, B. (2010), “Squeezing the heat out of London's Tube”, Proceedings of the ICE - Civil Engineering, 163, (3), 114 –122. Bourne-Webb P, Amatya, B, Soga, K, Amis, T, Davidson, C, and Payne, P. (2009), “Energy pile test at Lambeth College, London: geotechnical and thermodynamic aspects of pile response to heat cycles”, Geotechnique, 59 (3), 237–248.

Poulos, H G. (1989), “The Mechanics of Calcareous Sediments”, John Jaeger Memorial Lecture, Australian Geomechanics, Special Edition. 8-41. Suckling, T. (2004), “Geothermal Piles used at Keble College, Oxford”, The Structural Engineer, 19.