TM59 modelling inputs a reviewer should check

When you review someone else's overheating assessment, the result is only as good as the inputs behind it, and TM59 modelling inputs are where almost every disagreement starts. Check the weather file, the occupancy and gain profiles, the window opening assumptions, the shading credited and the room selection before you read the conclusion. A model that passes on the strength of a mild weather year and open windows at midnight has not demonstrated anything.

Ask for the model file and the input report, not the summary table.

What TM59 tests and against which criteria

CIBSE TM59 sets out a design methodology for assessing overheating risk in homes, and it applies two different tests depending on how the dwelling is ventilated.

For predominantly naturally ventilated homes there are two criteria. The first, taken from the adaptive approach in TM52, applies to living rooms, kitchens and bedrooms: the number of occupied hours in which the operative temperature exceeds the comfort threshold by one degree or more, across the period from May to September, must not exceed three per cent of occupied hours. The second applies to bedrooms only: the operative temperature between 22:00 and 07:00 must not exceed 26 degrees for more than one per cent of the hours in that period.

For mechanically ventilated homes the method switches to fixed temperature thresholds, assessed against occupied hours rather than against an adaptive comfort band.

Both criteria must be met. A report that passes the daytime test and fails the bedroom night test has failed, and a summary that averages across rooms conceals exactly the room that matters.

Communal corridors in blocks of flats carry their own check, because riser and pipework heat gains make them a source rather than a buffer, and a corridor running warm undermines the ventilation strategy for every flat off it.

Weather file and location selection

Start here, because the weather file can move a result from pass to fail on its own.

TM59 is run against a CIBSE Design Summer Year for the relevant location. The first design summer year stands for a moderately warm summer and is the compliance run. The second and third cover a short intense heatwave and a prolonged period of sustained warmth, and they test how the design behaves when conditions are worse rather than determining compliance. A report that only presents the first and claims resilience has not tested resilience.

Three checks:

1. Is the weather station appropriate to the site, and has an urban weather file been used for a city centre location where the urban heat island matters?
2. Is the emissions scenario and period stated, and is it the one the authority or the policy asks for?
3. Where future weather years have been requested, as the Greater London Authority does for referable schemes, have they been run and reported separately rather than replacing the base case?

Ask for the file name and version. Weather datasets are revised, and a run against a superseded file is not comparable with the rest of the evidence base.

Occupancy and internal gain profiles

TM59 prescribes the profiles so that assessments are comparable, which means a modeller who departs from them has to justify it in writing.

The method assumes bedrooms are occupied overnight and partially during the day, with one bedroom treated as continuously occupied to stand for a home worker or someone unwell. Living rooms and kitchens are occupied across the day and evening. Equipment and lighting gains are specified, and kitchen cooking gains are applied for a defined period. Domestic hot water distribution losses, cylinder standing losses and heat from risers and lateral pipework have to be included, and those are the gains most often left out.

What to look for in the input report: the profiles reproduced in full rather than cited, the hot water system described with its insulation standard, the heat network or communal system losses modelled if there is one, and the occupancy density matched to the bedroom count rather than to the floor area.

Understated pipework gains are the single most common reason a built scheme performs worse than its model. In a flat with a riser in the hall cupboard, that heat has to go somewhere.

Window opening assumptions and their limits

Natural ventilation models pass or fail on how much the windows are assumed to open, and this is where review effort pays.

Check the opening area actually available rather than the window area. Restrictors limiting openings where there is a fall risk reduce the free area substantially, and an assessment modelling a full casement opening on a fourth floor bedroom is modelling something the occupant cannot do. Check whether openings are top hung, side hung or tilt only, since the free area differs for each.

Then check when the windows are assumed to be open. TM59 opens them when the space is occupied, the occupants are awake and the internal temperature is above a stated threshold. Security and safety reduce that further at ground floor level and on accessible elevations, and noise reduces it everywhere that a closed window operation needs to be tested.

Noise is the part most often skipped. Where external levels mean a bedroom window cannot be left open at night without exceeding internal noise guidance, the overheating model and the acoustic assessment must share the same assumption. If the noise report says windows closed and the thermal report says windows open, one of them is wrong and the submission is not credible. Where the scheme relies on a mechanical route instead, that interaction belongs in the Energy Statement as well.

Shading, blinds and user behaviour

Credit for fixed external shading is legitimate. Overhangs, recessed balconies, brise soleil, external louvres and the shading effect of neighbouring buildings all belong in the model, provided the geometry matches the drawings.

Credit for internal blinds and curtains is a different matter. The dynamic modelling method in Approved Document O does not allow credit for occupant controlled internal blinds and curtains, so a scheme that passes only with blinds deployed is not a compliant result under that route. Where blinds are modelled, the report should say whether they are automatically controlled and permanently installed, and should present the result with and without them.

Glazing specification is the other input to check. Confirm the g-value and light transmittance come from the specified product rather than from a default, and that the frame factor and the shading from reveals match the construction. A low g-value glass that the client later substitutes for cost is a real risk, so the submitted specification should be conditioned.

Bedroom night time criterion

Treat the bedroom test as the governing one, because it usually is.

Check that every bedroom type, orientation and floor level that could be worst case has been modelled, not a convenience sample. Check that the criterion has been assessed between 22:00 and 07:00 rather than across the whole night or the whole day. Check that the result is reported as a percentage of hours and as the number of hours, since a small percentage of a long period is still a lot of uncomfortable nights.

Single aspect flats facing south or west, on upper floors, with a glazed elevation and a restricted opening, are where failures concentrate. A scheme with many of those and no mechanical cooling strategy needs a convincing explanation.

Reporting failures honestly

An Overheating Risk Assessment that reports a failure and explains the mitigation hierarchy is more useful than one that reports a marginal pass nobody believes.

The hierarchy runs in a settled order: reduce heat gains through orientation, glazing ratio and insulation of pipework and plant, then manage the heat that enters with fixed shading, then manage the heat in the building through ventilation and exposed thermal mass, and only then provide mechanical ventilation and, as a last resort, active cooling. A report that opens with a cooling system has skipped the first three steps.

Where failures remain, the report should name the rooms, state the margin, set out what was tried and say what would be needed to pass. A reviewer can work with that. A reviewer cannot work with a table of green ticks and no input appendix. The same discipline applies to the carbon side of the submission, as our note on energy statements for warehouse and logistics schemes describes, and to roof design where solar and planting compete, covered in our piece on biosolar roofs.

You can find reviewers for technical drafts through the professional directory.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. For specialist technical reports, qualified professional input may be required.