Silver EMS is a heat networks and MEP design consultancy. Through our facilities management department we also operate and maintain 16 community heating networks across London with just under 3000 connected homes. In the following text, Technical Director Dr Anthony Riddle explores urban energy networks with a focus on waste heat integration, including through heat pump applications.

His focus is on retrofit buildings in London, where a lot of innovation has taken place over recent years, driven largely by regional policy and other enabling factors. The following six case studies present a different application context and are s either in operation or under development currently.  There is currently significant growth in heat pump applications in the building sector as well as a move to modern smart energy systems.

Drivers and challenges

Key drivers and challenges include:

  • Climate Emergency
  • Air Quality
  • Affordability
  • Energy Markets
  • Energy security
  • Lifecycle value proposition
  • Local Exploitation Opportunities
  • Fuel Poverty
  • Planning and Policy
  • Futureproofing and Resilience.

Exploitation Opportunities

A key driver is the abundance of low grade energy and waste heat within London.  This waste heat is available in a variety of forms and temperatures. The city also has access to the River Thames and to canals, reservoirs and the London Aquifer. Making use of low-grade energy for heating buildings requires primary energy input. Heat pumps offer a logical choice, with access to a rapidly decarbonising grid, thanks to ever growing contributions from renewable electricity plant.

Waste incineration provides a similarly attractive proposition, this time by extracting heat from steam turbines. The penalty is lost electricity production, but the process is highly efficient in carbon and cost terms. The cost and carbon penalties of heat production when linked to electricity generation depend on when the electricity is required and the electricity generators dispatched at that time. Operating energy networks flexibly and introducing thermal storage to allow load shifting to times of economic production, can therefore enhance value and improve economic returns. As networks grow and multiple heat sources connect, it becomes possible to dispatch them flexibly according to a merit order of carbon and cost, as is done in Copenhagen for example.

Modern Energy networks are being designed to exploit local opportunities to best effect. So for example a network connecting to an energy recovery facility will exploit the temperature of the available heat in its design. And where low-grade waste heat is available, such as from a data centre, a lower temperature distribution network will be the preferred solution.

This may be an ambient loop where there is also opportunity to share and exchange (i.e. to prosume), energy between adjacent buildings. In all cases, reducing end user distribution temperatures is critical to bringing down lifecycle operating costs, so much so that customer tariffs reflecting this are likely to gain more and more traction in the market in coming years. The equivalent opposite is also true where cooling customers are involved.


Turning to affordability, there are several factors to consider. Firstly the challenge of delivering affordable schemes in the face of complex construction and multi stakeholder environments with legacy buildings which are also often situated in low energy density contexts. Secondly operational costs, which generally far outweigh investment costs over a project lifecycle and thirdly the costs to society of not acting. Risk is also a key issue to factor in, for example, where a waste heat source is not in the ownership or control of the supplier of last resort, resilience to offset this risk must be built in. Taking all of this into account, the design challenge is to minimise the lifecycle cost per unit of carbon saved.

Fuel is not the only consideration for a network operator, but over the life of a scheme it can be a dominating contributor. The strong carbon case for heat pumps is clear, as is the economic challenge of a widespread roll out of heat pumps to individual residential customers and when seeking to replace behind the meter gas engine CHP such as in universities and hospitals. Equally striking is the strong carbon and economic case for energy recovery facilities, thanks to plant z factors which can be 6 or above for larger facilities.

The following case studies include both retrofit and new build. They demonstrate some of the principles discussed and that no one size fits all when designing urban energy networks.

Gospel oak district heating scheme in Camden.

Constructed in 2010, this involves economiser heat recovery from the exhaust gases of an open cycle gas turbine at the Royal Free hospital.   The turbine’s flue gases are cooled to around 110 degrees Celsius. Heat distribution to residents is through a variable temperature network, just sufficient to meet building temperature distribution requirements. The network was designed to accommodate future reductions in these temperatures. An interesting challenge in this scheme was the need to avoid erosion of the turbine’s electrical efficiency. This required careful implementation to prevent an increase in back pressure in the exhaust gas air path from the newly introduced economiser maintaining flue gas buoyancy and preventing plume formation in winter both unwanted by products of heat extraction, were also key design objectives.

Bunhill II

The Bunhill II scheme in Islington recovers low grade heat through an air to water heat exchanger located in the ventilation path of a TfL vent shaft along City Road. A 1 MW ammonia heat pump upgrades this heat for distribution to the network. Gas engine CHPs supplement this heat and generate electricity to supply the heat pump, the ventilation shaft fan and the landlords electricity supply in the adjacent residential block. The system is designed for flexible dispatch for Heat pumps to run when electricity prices are low and gas engines to run when prices are high, including during STOR periods.  Distributed thermal storage is incorporated to maximise flexibility. This is achieved by back feeding between energy centres. A key driver for TfL was the opportunity to provide cooling to the tube in summer.  This is achieved through reversal of the ventilation air flow direction.  Existing distributed communal boilers are retained for topping up.

New Mill quarter

The New mill quarter development in Hackbridge is served by an energy recovery facility via the Sutton decentralised energy network. Currently the facility operates biogas engines with heat recovery but in future it will operate as an incinerator with electricity production through a steam turbine. Heat is pumped over a km to an energy centre local to the development, where large thermal stores and peaking boilers are located. The SDEN network is sized to transmit the extracted heat at temperatures of up to 110 Celcius, thereby exploiting the temperature of the source. Temperatures mixing down at part load and in the secondary distribution networks minimises heat losses.

West King Street Hammersmith

The West king street mixed use development has been designed for SAP10 compliance. It will be served via centralised heating and cooling systems from an energy centre in the basement of an office block. Energy recovery will take place between a water-cooled chiller and a water source heat pump, with the balance drawn from the London Aquifer. Boilers and air source chillers provide topping up.

Careful attention was paid in the design to generation and distribution efficiencies, with customer end units being specified to minimise required refrigerant cycle temperature lift between heating and cooling systems. An ambient loop concept was also considered for this scheme. However, a number of integration issues made this challenging to take forward.


The GreenSCIES scheme in Islington is an ambient loop network, currently at stage 2 design. The first phase will serve a university, three residential blocks, and selected municipal and commercial buildings. Distributed energy centres will connect to the ambient loop providing heating and/or cooling to the buildings they serve. Surplus energy will be recycled back to the ambient loop for adjacent buildings to use. In a similar way to Bunhill II, flexible dispatch will harness value through the electricity market and thermal stores will provide load shifting, this time at each customer location. Further value will be captured via behind the meter PV and electric vehicle charging, connection to the London aquifer(planned to operate as a seasonal store), data centre chiller plant load displacement and using active temperature conditioning of the loop.

Pimlico District Heating Undertaking

The network currently operates gas engine CHPs with boilers and thermal storage. The engines are sized to run continuously to meet the network thermal demand. Decarbonisation options for this network are being examined and to that end, options for integrating heat pumps are being developed. Heat recovery from the River Thames, a TfL ventilation shaft, a UKPN transformer substation and air source heat pumps are all in the mix. Key challenges in this context are delivering heat efficiency at the prevailing network temperatures and integrating heat pumps in a way that allows continued CHP operation.

Integration challenges for retrofit applications

Integration issues include fabric, legacy controls, user behaviour, water quality and hydraulic balancing and commissioning. These are commonly known value proposition issues when connecting energy networks to legacy buildings. Effective integration at individual building level requires careful design. The technical solutions are available and the benefits are understood. But the costs and implementation challenges can be significant. Scaling this problem up as networks grow will be a challenge and requires strategic commitment from policy makers, if stifling the growth of energy networks is to be avoided.

What matters is that legacy issues are addressed so that energy networks are not over designed or operated inefficiently over the next generation.