Who carries out a TM59 overheating assessment

Ask who does TM59 assessment and the honest answer is that no register exists. The methodology, published by the Chartered Institution of Building Services Engineers as its technical memorandum on assessing overheating risk in homes, does not reserve the work to anyone. In practice it is done by building services engineers and dynamic thermal modellers, and the quality of the output depends almost entirely on assumptions the method leaves to the modeller.

That is the part worth understanding before you appoint. Two competent modellers given the same drawings can return different results, and the difference is usually window opening.

What the methodology asks of the modeller

The method sets criteria rather than a single temperature limit. For predominantly naturally ventilated homes it applies the adaptive comfort criterion from the companion memorandum to all occupied rooms, limiting the hours in which the operative temperature exceeds the comfort threshold across the summer period, and adds a separate night-time criterion for bedrooms so that sleeping conditions are assessed on their own terms.

Mechanically ventilated dwellings are assessed against a fixed threshold instead, because the adaptive argument does not hold where occupants cannot open windows. Buildings with communal corridors served by a heat network get a further check, since riser and pipework losses make corridors a known failure point.

The method also prescribes occupancy. Bedroom and living room occupancy patterns, the number of people per room, equipment and lighting gains, and the assumption that curtains or blinds are used are all set out, which is what makes results comparable between schemes. A modeller who quietly reduces the prescribed gains is not following the method.

Worst-case unit selection is the other prescribed step. The assessment covers the units most at risk, typically single-aspect dwellings facing south or west, top floors and those with the highest glazing ratios, rather than an average flat.

Competence without a register

What to look for instead of a licence:

• chartered membership of the relevant building services institution, or equivalent demonstrated experience
• a portfolio of dynamic thermal modelling on residential schemes of comparable form, not just non-domestic work
• familiarity with the authority's local requirements, which in London includes the cooling hierarchy and the energy assessment guidance
• willingness to state assumptions explicitly in the report rather than in a spreadsheet nobody sees

Accreditation schemes for energy assessors are a different thing and do not cover this work. An on-construction domestic energy assessor is qualified to produce the compliance calculation for a dwelling, which is not a dynamic simulation and cannot answer an overheating question.

Councils expect a suitably qualified author for overheating work, and professional review of any draft is strongly advised. Where an AI-assisted draft is used as a starting point, a professional sign-off on that draft is what gives it standing, and the scope of that sign-off should be written down.

Software, weather files and model validation

The modelling has to be done in dynamic simulation software capable of hourly calculation across the summer period. The software matters less than what is fed into it.

Weather files are the first thing to check in a report. The method specifies which design summer year dataset and emissions scenario to use for the location, and the choice changes the answer materially, because the design summer years are deliberately chosen to represent different kinds of hot weather. A report that does not state the weather file, the location and the scenario cannot be reviewed.

Model validation is the second. Reasonable checks include whether the geometry matches the current drawing revision, whether the context overshadowing has been modelled, whether the openable free area reflects the actual window ironmongery rather than the full opening, and whether the ventilation strategy in the model is the one the mechanical drawings show.

Window restrictors are the classic error. Above ground floor level, safety restrictors limit opening, and modelling an unrestricted casement produces a pass that the completed building will not deliver.

Inputs the design team has to supply

The modeller cannot invent the following, and the programme slips while they wait for it:

1. Glazing specification with solar transmittance and light transmittance values, plus frame details and the glazing ratio per elevation.
2. Construction build-ups with enough detail to establish thermal mass, not just U-values.
3. The ventilation strategy, including openable free areas, trickle provision, any summer bypass on a heat recovery unit, and purge arrangements.
4. External noise levels at each facade from the acoustic consultant, because noise decides whether windows can be assumed open at night.
5. Heat network and riser routes with insulation standards, where there is a communal system.
6. A context model showing neighbouring buildings that overshadow the facades.

Item four is the one that catches schemes out. On a site next to a main road or a railway, the acoustic assessment may conclude that windows have to stay shut for internal noise levels to be acceptable, and the overheating model then has to assume closed windows with mechanical ventilation. Running the two assessments in isolation produces a scheme that passes both and satisfies neither.

Reading a room that fails

A failing room is a design output, not a verdict. The response follows a hierarchy that local policy in London makes explicit and that applies as common sense elsewhere.

Reduce the heat getting in first: lower the glazing ratio, improve the solar control of the glass, add external shading such as reveals, fins or balconies. Then manage the heat: expose thermal mass, add night purge capability, improve cross-ventilation by changing the internal layout of a single-aspect unit. Then ventilate mechanically. Only then consider active cooling, which authorities treat as a last resort and will question.

Interpreting the margin matters as much as the pass or fail. A room failing one criterion by a small margin in a single worst-case unit is a different conversation from a scheme where half the dwellings fail both criteria, and the report should say which situation it is describing.

The simplified route and where it diverges

Building regulations introduced a dedicated overheating requirement with two compliance routes: a simplified method based on glazing area and openable area limits by location and risk category, and a dynamic thermal modelling method that uses this same memorandum with some specified modifications.

Two divergences cause trouble. The regulation route includes its own limits on noise through open windows, which the planning-stage assessment may not have considered. And the modifications the regulations require to the modelling method mean a planning-stage assessment is not automatically a compliance demonstration.

Plan for both. A scheme that passes at planning stage using the full dynamic method, with the regulation modifications applied from the start, avoids a second modelling exercise and a late redesign.

Reporting for planning and for building control

At planning stage the overheating work usually sits with or alongside the energy statement, and the authority is interested in whether the design avoids reliance on active cooling. The deliverable is a report with the assumptions, the weather data, the units assessed, the results against each criterion and the mitigation incorporated.

At building control stage the question is compliance, and the deliverable is evidence against the chosen route plus the as-designed specification. Measures relied on at planning should be secured by condition, because glazing specification and external shading are the first things value engineering removes.

Independent checking is worth the fee where the result is marginal, and the argument for peer review of technical evidence is strongest when a single assumption carries the conclusion.

Appointing early enough to change the design

Overheating is decided by glazing ratio, orientation, aspect and shading, and all four are fixed at the point the massing and unit layouts are set. An assessment commissioned after planning drawings are complete can only report the problem.

Appoint at the end of concept design, ask for an early sensitivity run on the worst-case units rather than a full assessment, and use the result to set glazing ratios per elevation. That run costs a fraction of the final assessment and it is the last point at which the answer is still inexpensive to change.

Reviewers in PlanScope's professional directory can review and sign off AI-drafted technical documents, and PlanScope's own report types include the narrative documents that sit around the modelling. For specialist technical reports, qualified professional input may be required.

Enter your postcode on PlanScope to see which reports your scheme is likely to need.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. This is general guidance, not legal advice; check the position with your local planning authority or a qualified professional before relying on it.