Logical Heat Network Upgrade Path

From Heatweb Wiki
Jump to: navigation, search

This is thinking on the best way to upgrade existing housing stock to zero carbon.

The current date is 20th March 2022, and we have just completed numerous BEIS fundedn HNES schemes.


  • Existing networks vary from modern HIU networks, to ancient hot water storage systems.
  • Centralised energy centres create a single point of failure and a monopoly on heat supply.
  • Many older networks with storage, and some older HIUs, there is no differential pressure on the network. There are significant number of bypasses (short-circuits) on the network, and until they are nearly all fixed it is impossible to generate reasonable network DP. These include motorised valves on hot water cylinder, valves in HIUs, flushing bypasses, and pressure related bypasses commonly specified until recently.
  • Without differential pressure, it is impossible to satisfactorily run DHW using a modern HIU without an in-line shunt pump.
  • We want to move to heat pumps
  • Existing radiators can meet peak loads using CO2 heat pump optimum temperatures of 70/30C. There is no need to replace radiators to use heat pumps, only the TRVs with RTLs.
  • Installers will never be able to balance radiators based on flow rate to modern loads and temperature drops.
  • Fixed orifice TRVs set to a low flow rate have problems with dirt and air clearance. The long delays make them very hard to balance and tempt users into 'fixing'.
  • Return temperature limiting valves on radiators together with room thermostats are recommended in CIBSE Heat Network Guidance, and solve the three points above by directly targetting return temperatures. They provide users with enough trimming control within a preset range.
  • Compensating heating flow temperatures based of room temperature will reduce average daily heating return temperatures from 60C to 30C on systems running at 70C and balanced to a 10C drop, without further balancing or use of RTL valves. Such measures are not a replacement for balamcing, but instead make the most of current balancing.
  • Heat sources are most efficient if only run at their hottest (and least efficient) to elevate radiator temperatures on the coldest days. This also benfits pipe sizing, with the highest loads met by higher temperature drops as opposed to higher flow rates.
  • Heat networks and heat pumps are more efficient when weather compensated, if HIUs track this on feed to central heating, with either a setpoint offset for indirect HIUs, or inherently on direct HIUs.
  • During Summer months 61C would be an acceptable network temperature, thereby allowing DHW (storage) systems the ability to reach 60C to sterylise. Otherwise 51C would be acceptable in order to meet overnight (lower) domestic hot water needs.
  • It is politically difficult to force a user to turn on heating when they havnt asked for it for billing reasons.
  • Standard room temperature controllers often come with an Optimum Start function that adjusts the user on times, pulling them forwards to get to set room temperature on time. The offset timings are learned through operation.
  • With optimum start room temperature controls in properties, a network control system can pull heating load forwards by altering the morning start-up temperature relative to the weather compensating setpoint. For example, if the morning surge in central heating is resulting in too high return temperatures, then reducing morning temperatures will force room controllers to start earlier and reduce heating peaks.
  • Overheating in buildings is a common flaw with current installations. Weather compensating networks and reducing return temperatures to near ambient removes overheating, significantly impacting unwanted heat losses and Summer building temperatures.

The Green Heat Network Fund

18 minutes in (but please watch it all) is information on the Green Heat Networks Fund.

This puts hundreds of millions of pounds behind the proper solutions to the problem of converting heat networks to heat pumps.

The best solution is outlined below, and it is our role as the industry experts on renewable heat networks to educate the industry and heat network operators how best to accomplish the goals of the funding.

The Solution

  • Radiators fitted with RTL valves where possible, and room temperature compensated (HIU function) where not. Optimum Start function on room thermostats.
  • Fit CO2 air source heat pumps at roof level, back-feeding into existing riser tops. These will run between 65/30 in Summer and 85/30 on the coldest day, giving us 250% modulation on heating capacity at peak design flow rates.
  • Pipes sized on 75/30 peak, with flow pipes to <2m/s and returns to >1m/s (nearest). It is expected properties will take 15mm pipes (providing 50kW peak) or 22mm, and 28mm will be large enough for most risers up to 35mm at most.
  • Working from top-down, upgrade properties to an HIU with integral shunt pump. No additional DP is required on network for these to work. If possible, upgrade radiators with RTL valves as you go.
  • Risers can likely be reduced in size to 28mm or 35mm pipework. If they are in need of replacement then this can be achieved as one works down the building.
  • Once a riser has enough heat input to meet its own load, it may (temporarily) be isolated and run as a stand alone system while other works progress. This may be especially beneficial in regards to water treatment and cleansing of the network in stages. It should also be noted that at this point the monopoly on energy supply is broken, with the primary heat now coming from local heat pumps.
  • All HIUs would have a remote data system to ensure hand-over performance, and to enable optimisation and changes of settings as upgrades progress.

The 250% modulation is important. At 60/30 we have an average radiator temperature of 45C, and a difference to room of 14C. in steady state this us usually more than enough down to 5C external temperatures.

BUT... if we crank this up to 85/30 we have an average radiator temperature of 57.5C, and 36.5C difference to room. The radiators put out 2.5 x the output at 60/30.

IN OTHER WORDS... changes in central heating load, resulting from changes in external temperatures, can be perfectly met by adjusting the network operating flow temperature and maintaining a steady 30C return temperature regardless of season. As long as the system stays that bit hotter than required, room thermostats will do the finer control, as well as user trimming on RTL valves.

It also means, in reverse, that systems setup to operate at peak load requirements (those that do not weather compensate), are pumping out more than 2.5 times more heat into the building than they need to in Summer. A current system may have both a flow and return pipework system all above 70C at all times. The solution put forward would have a flow pipe nearer 55C in Summer at night, with the return pipe near 25C (riser air temperature). Insulation would be shifted from the return pipe to the flow pipe, with a standard 50mm on flow pipes and 15mm on return pipes. This further reduces losses for a similar material investment.