Why is my heat pump COP so low?

What COP should I actually be getting?

Benchmarks first, so you know whether you have a problem. The UK Electrification of Heat demonstration measured a median seasonal performance factor of 2.78 across 428 monitored air source installations. Open monitoring communities show the best domestic systems running above 4.4 — and those systems share one characteristic: an average design flow temperature around 34 °C.

Is the number you are reading even a COP?

Check this before changing anything. Three measurement traps produce alarming figures from healthy systems.

Boundary. A COP quoted for the compressor alone (H1) ignores circulation pumps, controls and backup heaters; a whole-system figure (H4) includes them and is typically 0.3–0.6 lower. Two "COPs" from the same house can differ by 20% purely on where the boundary is drawn.

Backup heat. If the electricity meter sees the immersion element or in-line resistance heater but the heat meter records only the heat pump's output, the arithmetic collapses. A 3 kW immersion running at COP 1.0 for two hours a day will drag a genuine 3.4 down towards 2.5.

Unit-reported values. Many controllers estimate COP from compressor tables rather than measuring it, and almost all of them flatter the machine. A discrepancy of 0.5 between the controller's number and a heat meter is common and is usually the controller being optimistic, not the meter being wrong.

The five things that actually destroy COP

1. Flow temperature. The dominant lever. Each degree of flow temperature is worth roughly 2–2.5% of COP, so a system running 55 °C where 45 °C would do is giving away a fifth of its efficiency permanently. See what flow temperature should I run my heat pump at?
2. A mixing buffer tank. A four-port buffer that blends flow and return raises the temperature the heat pump must produce by 4–5 °C to deliver the same heat to the emitters — roughly a 10% penalty, and entirely avoidable with correct piping or a volumiser.
3. Cycling. Every compressor start costs energy and produces no useful heat for the first minutes. A unit doing six starts an hour at mild ambient is oversized, over-throttled by thermostatic valves, or short of water volume; see why does my heat pump keep switching on and off?
4. An undersized cylinder coil. Hot water is where most systems lose their efficiency, because a boiler-era coil forces the heat pump to 60 °C to transfer heat that a properly sized coil moves at 48 °C. See what coil surface area does a heat pump cylinder need?
5. Throttled flow. Closed thermostatic valves, a fouled strainer, a pump on a fixed low speed or air in the circuit all reduce flow rate. The heat pump responds by raising flow temperature to deliver the same kilowatts — the efficiency loss appears in the bill, not on the display.

Why is hot water COP so much lower than heating COP?

Because the lift is bigger and the load is short. Charging a cylinder to 50 °C in mild weather runs the machine at a temperature it would never use for space heating, and each charge includes a start-up period at poor efficiency. A DHW COP of 2.2–3.0 against a heating COP of 3.5–4.0 is normal, not a fault.

It becomes a fault when the charge temperature is set to 60 °C daily, when the cylinder thermostat re-triggers several times a day on small draw-offs, or when a weekly legionella cycle is being run every night. One long charge a day at the lowest usable temperature beats four short ones.

How to fix it, in order

Measure first with a heat meter and a dedicated electricity meter, over at least two weeks. Then work down the list: open every thermostatic valve in the reference room, check the actual flow rate against the design figure, verify the weather compensation curve is enabled and not sitting on the factory default, confirm whether a buffer is mixing, lower the DHW charge temperature to the lowest that gives usable hot water, and only then look at the machine itself. Nine times out of ten the heat pump is fine and the system around it is not.

One note on datasheets, from a manufacturer that publishes them: Solimpeks rates its Varm Up Series at a heating COP of up to 4.9 at A7/W35, and the qualifier matters more than the number. A7/W35 is a laboratory condition, not your January.