Phased retrofit of an 1800s sandstone detached house near Preston

Since internal wall insulation was being proposed and the house was in a reasonably exposed location, we felt it would be sensible to get some WUFI hygrothermal modelling done to assess moisture risk and how much insulation we could safely apply and to decide on the airtightness strategy (this is so that solid walls can still dry out as fast as they get wet, which might not be the case if you add too much insulation internally or have an airtightness layer that is a bit less vapour-open). Some of the existing walls already had 50mm of either XPS or EPS insulation internally, sometimes together with an impermeable membrane – coring out some of the existing XPS insulation in the utility room from a west-facing wall revealed humid conditions behind.

In order to do the WUFI moisture risk analysis properly, core samples were taken from different walls.

The WUFI analysis was then run on 6 walls in 4 orientations, and concluded that 15mm of diathonite followed by 80mm of wood fibre with 10mm of lime plaster internally would be safe, as long as there was an MVHR system installed and that the wall exterior was maintained in a good condition. The graph below shows the relative humidity at the critical interface between the masonry and diathonite for one of the walls.

However, it was not only the overall moisture content in walls that had to be checked, since the most vulnerable part within such a wall will be where there is timber. The analysis therefore also checked relative humidity levels at joist ends where these entered a wall (maximum allowed was 87% relative humidity). In order to remain below this threshold, the wall around any joist ends would have to only be insulated with 30mm of diathonite. The question then was what distance above and below the joists would have to be insulated like this before reverting back to the main wood fibre insulation – we felt that keeping 200mm both above and below would be sufficient, but this was not exactly based on a moisture model but was more of a best guess. Given such a scenario, we ran a thermal bridge model to check surface temperatures within this joist zone, since we also needed there to be no condensation risk (i.e. surface below 15 degrees Celsius) – we found that this was all fine as long as the edges of the wood fibre boards were mitred.

Once the lounge floor had been dug up, it was clear that there was water ingress at a couple of points along the base of the wall. This was resolved by digging a gravel-filled drain on the outside of the wall to take water away from the wall, as well as lowering the ground level slightly. The strategy for the floor was a vapour-open one, where any residual moisture would be able to escape through the glass aggregate insulation and lime screed to be extracted by the ventilation system (rather than making it impermeable which might push moisture towards the walls instead).

When insulating the walls starting with diathonite (which was applied over the dubbed-out walls), the walls were allowed to dry out fully before any wood fibre was applied. During this drying process, what we found was that there were two patches of moisture that recurred on the walls. The source of one of these was traced to gaps in mortar between the external stone blocks at the level of the flowerbed. The other always seemed to recur after a rain event, and what we assumed was that driven rain was soaking into what was a more porous gable wall and percolating down through the rubble to the base, where it appeared as temporary rising damp. These were resolved by some repointing and checking for cracks in the outer wall. Once these were resolved, we still decided to install Calsitherm on the bottom metre of the wall (instead of wood fibre), just to be sure.