Weather-compensated system boiler short cycling and a heat pump ambition
A four-bedroom detached house has a 26 kW system boiler, a 250 L unvented cylinder with a 1.0 m² coil, and twelve radiators sized for 75/65 °C with a 20 °C room condition. The owner has had weather compensation fitted, and on a design day the boiler now runs 65 °C flow and 45 °C return. Rooms are slow to heat and the boiler cycles on and off every few minutes even when the house is cold. Measured at the boiler: flow 64 °C, return 49 °C, and the manifold shows 14 of the 18 L/min going through a bypass. Three radiators have a 10 °C drop, the rest about 4 °C. The cylinder reheats from a recorded 15 °C to 60 °C in 26 minutes. The owner wants the system to run at 45 °C flow so it is ready for a future heat pump.
1. Explain the cycling and the uneven radiator temperature drops, and set out the order in which you would correct the system.
Read the worked answer
The cycling and the uneven drops share a cause: flow is bypassing the heating circuit.
The bypass passes 14 of 18 L/min, so only 4 L/min reaches the radiators. The three radiators with a 10 °C drop are the ones actually getting water; the rest with a 4 °C drop are starved, which is why some rooms never heat. The boiler sees a 15 °C differential and a 49 °C return, so it is condensing, but the heat delivered to the circuit is small compared with its minimum modulation, probably 8 to 9 kW against perhaps 4 kW of radiator load at these temperatures. The boiler reaches set point quickly, shuts down, waits for the anti-cycle timer and repeats.
Order of work:
- Identify the bypass. If it is an automatic bypass valve, adjust it to the minimum setting that keeps flow above the boiler minimum with all TRVs open. If it is a fixed gate valve, close it and confirm the boiler internal bypass is adequate.
- Set the pump to proportional pressure or the manufacturer's recommended mode, so head falls as TRVs close.
- Balance with a differential thermometer: open all lockshields, set the index radiator first, work back aiming for 10 to 12 °C across each radiator at design flow, and record the settings.
- Re-check modulation and cycling after balancing.
Only after that should anyone conclude the boiler is oversized for the house.
2. Quantify the effect of running at 65/45 and then at 45/40, and state what else must change before a heat pump can be fitted.
Read the worked answer
Radiator output does not fall linearly with flow temperature. Output is proportional to the mean water temperature above room temperature raised to about 1.3.
Design: 75/65 gives a mean water temperature of 70 °C; with a 20 °C room, the difference is 50 K. Current: 65/45 gives 55 °C, so the difference is 35 K. Factor = (35/50)^1.3 = 0.7^1.3 ≈ 0.63.
So radiators specified to deliver, say, 12 kW at design conditions are now giving about 7.5 kW. If the house heat loss on a design day is 9 to 10 kW, the rooms cannot reach temperature, which matches the complaint.
For a heat pump at 45/40: mean water temperature 42.5 °C, difference 22.5 K, factor = (22.5/50)^1.3 ≈ 0.36. The emitters would need roughly 2.8 times their current rated output, meaning much larger radiators, underfloor heating on the ground floor, or both. A 45 °C flow cannot be achieved by turning the boiler down; the heat emitter surface has to change.
The cylinder needs review too. A 1.0 m² coil will not reheat 250 L usefully at 45 °C flow; a low-temperature coil of 2.5 to 3 m² or a plate heat exchanger is normally required, plus buffer volume to stop the heat pump short cycling.
Give the customer the emitter upgrade cost before they commit to a heat pump.
3. Assess the recorded cylinder reheat performance and explain what you would do about it.
Read the worked answer
Do the arithmetic before accepting the figure.
Energy to raise 250 L from 15 to 60 °C: Q = 250 × 4.19 × 45 = 47,138 kJ = 13.1 kWh. In 26 minutes (0.433 h) the average input would be 13.1 / 0.433 ≈ 30 kW.
That is more than the 26 kW boiler can deliver, so the measurement cannot be taken at face value. Possible explanations:
- The start temperature was higher than 15 °C; the cylinder may have been at 25 to 30 °C after draw-off.
- The time was recorded from burner start after a delay, or was estimated.
- The boiler is not the model stated, or is running above its rated input.
- The cylinder thermostat is set lower than 60 °C at the pocket.
Re-measure properly: record the pocket temperature at the start, the flow and return at the coil, and the time to burner shutdown, all with the same thermometer. Then compare with the cylinder data; a 250 L cylinder with a 1.0 m² coil typically reheats in 20 to 30 minutes at 20 to 25 kW input, so the result may be plausible but the boiler output or start temperature is wrong.
The judgement is that 30 kW cannot come from a 26 kW boiler, so the record is unreliable; do not use it to size anything until repeated. Also confirm whether the boiler gives DHW priority, because if a diverter valve sends all output to the coil, the heating is paused during reheat and the apparent cycling in heating may partly be DHW demand.