How to not strengthen a traditional cut timber roof

I am currently engaged on a construction dispute in which the cost of building an extension somehow escalated to around £60k over and above the original £70k quote from the builder. I was initially called in to inspect and assess building defects relating to the works and the client was quite staggered at the range and number of defects found and I simply could not see where the additional £60k had been spent.

The builder had claimed an additional sum of money for ‘roof strengthening works’ and anecdotally I am informed that the client received a call from the builder when discovering that the roof rafters were structurally unsound and indeed, they were. At some point in the buildings history, changes had been made to the roof pitch and/or height and new rafters of the correct length and cross section should really have been installed to stretch from ridge board to wall plate. However, the builder at that time simply decided to nail extra lengths of timber onto the rafters and the roof was completed in the manner you see in this image.

Historic alterations to the buildings roof left the roof structure in this condition.

So, our builder, bless his cotton socks, was absolutely right; the roof structure was unsound and required strengthening, something that he could deal with at additional ‘reasonable’ cost.

Sadly however, what the builder actually did was make the same mistake again and simply nailed on even more bits of wood so we ended up with a three leaf rafter sandwich. Rafters were not properly coach bolted together and neither were the repair sections long enough to reach the wall plate. In fact, even the sections were undersized and the builder dealt with this by packing between the purlin and the repair sections with bits of cut plywood.

Three ply rafter sandwich but not much meat.

These were fairly major roofing work to a portion of the building and what is worse is that the builder only saw fit to report on the sections of the roof affected by his work to the new extension.

The main roof of the building was also structurally  unsound and whilst this had nothing to do with the current builder, surely a responsible builder would have reported this to the client and given them an opportunity to deal with the whole roof structure whilst works were ongoing.

New ‘strengthening’ rafters cut short and packed with plywood

I do know that the client would have taken the opportunity to deal with these structural defects in the whole roof but as it happens they also have to deal with the part of the roof that was allegedly strengthened . Needless to say, the cost of ‘roof strengthening’ works forms part of a rather large defect and cost schedule that is currently in the hands of my clients construction lawyer.



Please follow and recommend our blog page:

An Interesting Case of Roof Spread

Roof SpreadI was carrying out a dilapidation survey in the West Midlands quite recently and whilst checking the roof space came across an unusual case of roof spread. I’m sure you’re familiar with the basic cause of roof spread but if you aren’t then I’ve included a basic diagram that shows how unrestrained rafters can apply horizontal thrust to the walls and make them unstable. Generally this is due to inadequate collar ties or ceiling joists or indeed failure of those timbers.  In this particular case there were no external signs of cracking or structural problems and neither was I particularly looking for, or asked to investigate structural cracking in the building. However, on inspecting the first floor store room I noticed substantial cracking to the dividing party wall that warranted more detailed investigation.

There was a long crack to the internal corner of the ceiling to wall junction on the front elevation of the property and a further diagonal crack cutting down through the dividing party wall.

SRCrackThe damage was indicative of roof spread but was not quite serious enough to manifest any obvious visual problems to the front elevation of the property.

The roof space to this particular property was not fully accessible or safe to access so I had to inspect as best as I could from the loft hatch and after a couple of minutes I looked directly above my head and immediately noticed the problem. The ridge board was completely split which had resulted in a section of the rafters dropping just enough to apply horizontal thrust to the wall plate on the front elevation. If looking for the obvious and common cause of roof spread then this particular issue could so easily have been missed, especially  when limited to a head and shoulders inspection from the loft hatch.


Split Ridge

Split ridge board causing rafter instability and subsequent applied horizontal thrust to the wall plates.



Please follow and recommend our blog page:

A Flashers Embarassment

I carried out another detailed inspection of a new build property in Northamptonshire today and was immediately alarmed at the poor finishing seen to the stepped lead and apron flashings on the front elevation. Coincidentally, the brickwork was the worst I’ve ever seen on a new build property but that will be the subject of another blog. I am consistently finding that developers do not seem to understand the technical requirements when it comes to installing leadwork and I always advocate the need to investigate the quality of lead installation beyond surface appearances; particularly when surface appearances are as bad as seen in our first image.


Poor apron and stepped lead flashing details: Note lead securing peg visible to mortar bed joint.

Lead should be turned into the bed joint a a minimum of 25mm and should be securely lead pegged at 450mm centres. Once pegged the bed joint can be pointed up or sealed with a proprietary lead sealant. Lead has an incredibly high coefficient of expansion, which is why it should not exceed lengths of 1.5m when installed, this is all about limiting or managing the level of differential expansion at junction details. Failure to allow for differential expansion is a primary cause of lead flashing failure. This particular apron flashing was installed to the garage entrance porch so was accessible for closer inspection from a ladder. Lifting the apron revealed an easy route into the building for wind driven rain and NHBC guidance had not been followed with regard to extending the sarking membrane and turning it up the wall underneath the apron flashings. The NHBC recently released a technical bulletin that included a reference to this common defect, it is one they are consistently finding and my experience of this problem mirrors theirs. So even if the apron flashings were correctly installed there is no secondary barrier to prevent issues with wind driven rain.
The keen eyed amongst you may have noticed that a lead securing wedge is visible to the left hand side of the bed joint in the first image. Whilst this inspires some hope that the lead has been securely wedged this hope is offset by virtue of the fact that the wedge is visible. A correctly installed lead peg or wedge should not be visible and if it is then the weather resistance of the bed joint is compromised in that area.
When suspicions are raised to the degree they were raised on this inspection then there is a strong argument for gently tugging the corner of the lead. Correctly installed lead will not move whereas poorly installed lead will tend to do this…


Lead not turned into bed joint by the required 25mm

There was 5mm of lead turned into the bed joint as opposed to the required 25mm and lead wedges were installed but not providing a tight interference fit to secure the lead. In fact the only thing holding the 5mm of lead in place was a bead of grey mastic and this lead apron was always destined for very early failure. On the upside, the installer had managed to overlap sections of lead by the minimum required 100mm but there was very little else to redeem this installation.

The lead aprons to the garage porch were further susceptible to wind driven rain because there had been a complete failure to clip the free edge of the lead at 450mm centres; this would prevent wind from lifting the lead and therefore increasing vulnerability to rain.


Leading edge of lead apron not clipped

Moreover, the lead apron flashings were further susceptible to wind driven rain since the roofer had failed to comply with NHBC requirements relating to the need to overlap sarking membrane and turn it up the wall by the minimum required 75mm.


Where is the sarking?

Incidentally, I also measured the lead thickness with a micrometer to ensure that code 4 lead was installed and it was the required 1.8mm; it is not unusual for me to find that cheaper code 3 lead has been installed.
It worries me that the plumbers art of installing lead to a high standard is becoming lost but in this particular case the lead installations were as simple as it gets. You can make up your own mind as to whether this is poor quality workmanship, complete ignorance of the guidelines relating to lead flashing installation or a combination of both problems.
Joe Malone BSc(Hons) MCIOB MCABE

Please follow and recommend our blog page:
1 2