Historic Building Conference 4: Retrofitting Historic Windows
This is a collection of notes from the Historic Building Conference organised by CPD Essentials and held at the 16th century Shaw House in Newbury, Oxfordshire on 3rd June 2026. I am collecting them here to build a record of my learning, creating a treasure trove of notes that I can refer back to in the future, but also make them available to others.
Marianne Suhr is a Chartered Building Surveyor specialising in historic windows and retrofit, and her session was grounded in a real tension. We have a climate emergency and a need to cut heat loss from old buildings, and at the same time we are at risk of stripping out irreplaceable historic fabric in the name of efficiency. The whole point is to do both intelligently.
Why historic glass matters
Her starting position is that historic windows, alongside doors, are the expressive, joyful part of an old building, and that the glass in particular is precious and finite. Historic glass was mouth-blown, either crown glass spun into a disc and cut, or cylinder glass, and it has that characteristic moving, dimpled texture you can see when you walk past an old window. We no longer make crown glass in large discs, and there are only two firms left in the world making mouth-blown sheet glass for conservation, one in France and one in Germany, with only around twenty people holding the full skill set between them. Once that glass is gone it is gone.
Repair first
The first challenge is simply finding a joiner who can properly repair a sash, including setting the sill back beyond the line of the window so rain runs off rather than sitting in the joint. SPAB publishes a free downloadable window repair guidance document. Old windows were made of slow-grown, tight-grained pine, far better than modern softwood, so good replacement timber is worth sourcing, and skilled repairs such as steam-bending a new glazing bar are entirely possible.
Understanding the heat loss
Heat leaves a window by three pathways. Conduction through the glass, frame and reveals. Convection, meaning draughts through gaps. And radiation, the infrared part, which is the one people overlook. Short-wave near-infrared from the sun passes through glass and heats the room, which is helpful solar gain on a south elevation in winter but an overheating risk in summer. Objects and people inside re-radiate long-wave far-infrared, which cannot pass back out through glass and so is trapped, the greenhouse effect, which is also why low-emissivity coatings exist, to reflect that near-infrared. On a human level, if you sit next to a cold sheet of glass your body radiates heat to it and you feel cold, which is why earlier generations hung tapestries and panelling on cold walls.
Draught-proofing alone is not the answer
Historic England's 2024 guidance on adapting historic buildings notes that draught-proofing windows is almost always acceptable for listed building consent and is the simplest, cheapest improvement. But Marianne's own house was the cautionary tale. She draught-proofed her sashes, and while it stopped most of the draughts it did nothing for traffic noise and did not stop condensation, arguably making it worse by reducing ventilation, so she got black mould on the sills. Shutters help a little by trapping a layer of air but do not stop vapour condensing on the pane. On top of that, routing a slender historic frame to take seals can compromise the frame, is irreversible, and delivers no meaningful improvement in heat loss.
Secondary glazing is the workhorse
It gives considerably higher thermal performance than draught-proofing alone, is generally acceptable for consent, and is excellent for acoustics if the cavity is right and the seal is airtight. Where you use secondary glazing, leave the outer window without draught-proofing so there is an insulating air gap. Marianne sorts windows into three categories and matches a solution to each.
Windows that are never opened, for example facing a busy road. A magnetic lift-out toughened glass panel with a magnetic edge seal, professionally templated and fitted to a slim aluminium frame. Hard to remove, but you take it out once a year to clean. Very good thermally and acoustically.
Windows opened only in the warmer months, closed from November to spring. Something temporary you fit for winter and remove in spring. The cheap, effective option is the old shrink-film kit applied with a hairdryer, which lasts a season. The better option is a clear plastic panel, acrylic or polycarbonate, held on with a magnetic strip system where the frame strip is taped over so it disappears and can be painted. DIY-able if the frame is square, or professionally templated. There are also friction-fit panels with a rubber seal now on the market.
Windows opened all year, such as bathrooms and kitchens. More complicated. You really only need one openable window per room in winter, and it is important to open one after showering to release vapour. Options include sliding secondary units, drawing on the long Northern European tradition of double windows, an openable secondary system, or individual friction-fit panes dropped into each pane of a multipane sash.
A standing point on fire escape. Do not compromise an escape window for the sake of secondary glazing. Maintain it, keep it usable without a separate key, agree a family escape plan, and if necessary replace a whole window to keep a workable escape.
On plastic versus glass for secondary panels, plastics are lighter, warmer to the touch and therefore better for radiant comfort, cheaper and more discreet with thin frames, but they scratch more easily and some grades yellow in sunlight, so you pick the grade for the application.
Replacing the glazing where the glass is already lost
Single glazing has a U-value of roughly 5.2 to 5.8 and is usually the weakest point in the room. Under UK building regs, new or replacement windows in an existing building must hit a whole-window U-value of 1.4, frame and glass combined, or a minimum Window Energy Rating of B, and it is worth knowing the difference between a whole-window U-value and a centre-pane figure. The options, from least to most intervention.
Laminated or coated conservation glass, with Histoglass the market leader, available even as hand-drawn and cylinder glass, achieving U-values around 3.6 to 3.9. Better than single glazing but nowhere near 1.4.
Slimline double glazing, which has a narrow cavity and edge seal so it can fit some old glazing bars. Conservation officers were initially comfortable with it, arguably overlooking the loss of precious crown and cylinder glass. Marianne no longer recommends it because the narrow edge seal gives a very high failure rate, with guarantees of only one to five years, and reglazing a failed unit is expensive and disruptive. There is also a thermal bridge through the glazing bar, and sashes need re-weighting for the heavier units.
Whole new units done well, which is her preference. The debate with conservation officers is over applied or stuck-on glazing bars, which they dislike because cheap ones fall off and look wrong, so they often insist on integral bars and glazing each pane individually. Her approach is a fully jointed external lattice, a single thermally efficient unit behind, and a second lattice over the top, with warm-edge spacer bars now available in colours and barely visible. Her tested point is that even conservation officers cannot tell her applied-bar samples from individual-pane ones.
On performance and sustainability, good modern timber units around 18 millimetres achieve roughly 1.6 with argon and around 1.1 with krypton, but krypton is a rare, energy-intensive noble gas and not very sustainable, so timber is the genuinely sustainable element here, and she now uses a modified, dimensionally stable timber that holds paint well. The ultimate is vacuum glazing, two panes with a vacuum cavity of about 0.3 millimetres, so thin it looks like a single pane, with a centre-pane U-value around 0.34 and a 25-year guarantee. The downsides are cost, very long lead times through a multi-country supply chain, and a regular grid of tiny support pillars you can see against the sky like specks of dust.
On paint, linseed oil paint is excellent on bare wood for in-situ redecoration, while factory-finished units take a water-based paint that lasts around twelve years, brush-applied for texture.
The closing line worth keeping
Always start by asking whether the window is significant. Single glazing does not automatically mean significant. In a climate emergency, replace the genuinely insignificant, beyond-economic-repair windows with something better, long-lasting and efficient, and lavish care on the ones that matter. Marianne also flagged a recently published book on heritage windows, and her own conference talks, "Through the Looking Glass", which are free on YouTube.
