Home / Advice Centre / Can My Existing Radiators Work with a Heat Pump?
ADVICE CENTRE · EMITTERS
Can my existing radiators work with a heat pump? Often, yes — here's how we check
One installer says your radiators are fine. Another wants to replace every one. Both looked at the same walls. This guide shows you exactly how a designer settles the question — radiator by radiator, with arithmetic you can follow.
- 11 min read
- Updated July 2026
- By Phil · Heat Loss Hub
QUICK ANSWER
Many existing radiators can work with a heat pump — but not because of the radiator alone. It depends on whether each radiator’s output at heat-pump flow temperatures (typically 40–50 °C rather than a boiler’s 65–75 °C) still covers its room’s heat loss. In our surveys, most homes keep the majority of their radiators and change a targeted few. Anyone who says “all fine” or “all must go” without a room-by-room calculation is guessing in one direction or the other.
Why flow temperature changes everything
A radiator is a simple device: hot water in, warmth out. How much warmth depends overwhelmingly on the temperature difference between the water inside it and the room around it — what engineers call ΔT (delta-T).
Radiator catalogues quote outputs at ΔT 50 — roughly a boiler running at 75 °C flow into a 20 °C room. Run the same radiator with 45 °C water and the ΔT falls to about 22.5, and the output drops to roughly 40% of the catalogue figure. The radiator hasn’t changed; the physics of the water inside it has.
This is the entire radiator question in one paragraph. A heat pump runs most efficiently at low flow temperatures — every 10 °C lower is roughly 20–25% off your running costs — so good design pushes the temperature as low as your emitters allow. The job is to find out, for each room, whether the existing radiator still does its job with cooler water. Some will. Some won’t. Measurement tells you which.
The radiator maths, made simple
You don’t need to be an engineer to follow the logic. For each room there are only two numbers:
- What the room needs — its heat loss in watts on the design cold day, from the room-by-room calculation.
- What the radiator can give at the design flow temperature — catalogue output multiplied by a correction factor for the lower ΔT.
If number two is bigger than number one, the radiator stays. If it’s smaller, you either enlarge that radiator or raise the whole system’s flow temperature. Because the whole system runs at the temperature its weakest room requires, one undersized radiator in the hall quietly taxes every kilowatt-hour the house consumes. That’s why blanket answers fail: the decision is only ever as good as its worst room.
WORKED EXAMPLE
A lounge loses 1,400 W at design temperature. Its existing double-panel radiator (600 × 1,200 K2) is catalogued at 2,050 W at ΔT 50.
At 45 °C flow / 40 °C return into a 21 °C room, ΔT is about 21.5 → correction factor ≈ 0.37. Output: 2,050 × 0.37 ≈ 760 W. Falls short.
At 50 °C flow, factor ≈ 0.47 → ≈ 960 W. Still short. Swap to a K3 of the same length (3,000 W catalogue) and at 45 °C it gives 3,000 × 0.37 ≈ 1,110 W… close, so we’d go one size longer or accept 47 °C. This is the granularity a real design works at — and why “your radiators are fine” from the doorstep means nothing.
How we check each radiator
During the technical survey we record every radiator’s type (single, double, K1/K2/K3, column, towel rail), height and length, plus its pipework. Back at the design desk each one is checked against its room’s calculated loss across a range of flow temperatures — 40, 45, 50, 55 °C — using manufacturer output data. The result is an emitter schedule: a table showing, for every room, the margin or shortfall at each temperature.
Then comes the design decision. We look for the “sweet spot” flow temperature where only a handful of radiators fall short, and compare the cost of changing those few against the running-cost penalty of running hotter. It’s a genuine optimisation — a few hundred pounds of radiator now versus a percentage on every heating bill for twenty years — and it’s different for every house.
Correction factors at a glance
APPROXIMATE RADIATOR OUTPUT CORRECTION VS CATALOGUE (ΔT 50) FIGURES — 21 °C ROOM
| System flow/return | Approx. ΔT | Output factor | A “2,000 W” radiator gives | Typical use |
|---|---|---|---|---|
| 75 / 65 °C (old boiler) | ≈ 49 | ≈ 0.98 | ≈ 1,960 W | How your radiators were sized |
| 55 / 47 °C | ≈ 30 | ≈ 0.52 | ≈ 1,040 W | Heat pump upper range |
| 50 / 43 °C | ≈ 25.5 | ≈ 0.42 | ≈ 840 W | Common retrofit target |
| 45 / 40 °C | ≈ 21.5 | ≈ 0.35 | ≈ 700 W | Efficient retrofit sweet spot |
| 40 / 35 °C | ≈ 16.5 | ≈ 0.25 | ≈ 500 W | New build / UFH territory |
Factors are indicative (exponent ≈1.3 per EN 442); we use each manufacturer’s own data in real designs. The lesson survives the rounding: halve the ΔT and you lose well over half the output.
What we usually find
Here’s the reassuring part. British radiators were mostly sized with generous rules of thumb, in an era of 70 °C water and pessimism. Many are bigger than their room ever needed — accidental headroom that low-temperature design can spend. Across our design work the typical pattern looks like this:
- Bedrooms: usually fine as-is — modest losses, decent-sized radiators.
- Living rooms and bays: the most common shortfall — big losses, radiators squeezed by furniture and window sills.
- Halls and landings: frequently undersized for their (surprisingly high) losses.
- Kitchens: mixed — often a small radiator doing little in a room with high gains anyway.
- Bathrooms: towel rails are poor heat emitters at any temperature; often supplemented rather than replaced.
The net result, house after house: most radiators stay; a targeted few change. Our full guide Do all my radiators need replacing? tackles the blanket-replacement quote in detail.
DESIGNER'S INSIGHT
“When a quote replaces every radiator, one of two things is usually true: nobody did the room-by-room arithmetic, or the price includes a healthy margin on ten radiators you didn’t all need. When a quote replaces none, the system will simply run hotter than promised — and the efficiency shortfall lands on your electricity bill, not the installer’s. The emitter schedule is where independence pays for itself.”
Phil
Founder & Lead Designer, Heat Loss Hub
Your options when a radiator falls short
A shortfall is a menu, not a verdict. In rough order of preference:
- Swap panel type in the same footprint. K1 → K2, or K2 → K3: up to double the output with no redecorating and no pipework moves. The workhorse fix.
- Go longer or taller where walls allow — more metal, more output, still an ordinary radiator.
- Accept a slightly higher flow temperature where the shortfall is marginal and confined — sometimes 47 °C beats buying three radiators; the design shows the break-even.
- Address the room’s loss instead: occasionally the cheapest watt is the one you stop losing — a bay roof insulated, a chimney balloon fitted.
- Underfloor heating in single rooms being renovated anyway — never a requirement, occasionally an opportunity. See the UFH guide.
What about microbore pipework and old valves? Both are survey items rather than deal-breakers: 8–10 mm microbore constrains flow rates and sometimes dictates strategy, and tired valves get replaced as a matter of course. It’s exactly the sort of thing the technical survey exists to catch before it becomes installation-week drama.
Frequently asked questions
Look at the end: count the panels (the flat steel faces) and the fins (the corrugated metal between/behind them). One panel = K1 or “type 11” with one fin set; two panels with two fin sets = K2/type 22; three = K3/type 33. Height × length plus type is enough to look up catalogue output — though room-by-room heat loss is still the other half of the answer.
“Oversized” only relative to boiler habits. A K3 occupies the same wall length as the K2 it replaces, just 60 mm deeper. Modern low-temperature designs rarely need the wall-of-steel some people fear — and where they might, we’ll tell you before anything is ordered.
Often, yes — column radiators have respectable outputs and period homes frequently have generously sized ones. They’re checked exactly like any other emitter. Their thermal mass suits steady low-temperature running rather well.
Usually during — it’s one system balance and one drain-down. But if you’re redecorating a room this year anyway, fitting the design-specified radiator early is perfectly sensible. That’s another benefit of owning the design before choosing the installer.
Yes, with a caveat: heat pumps like open, steady circuits. Design practice is to leave some rooms (typically where the main sensor lives) fully open and use TRVs as limiters elsewhere. Your controls specification covers this — it’s a commissioning detail that meaningfully affects efficiency.
CONTENTS
RADIATOR DOUBT?
Our heat loss calculations include an emitter check at heat-pump temperatures for every radiator in your home.
RADIATOR BY RADIATOR
Find out which of your radiators already work
Every Heat Loss Hub design includes a full emitter schedule — the room-by-room proof of what stays and what changes. Tell us about your home and we’ll take it from there.