With the roof construction about to commence, our extension is now approaching the halfway point of ‘phase one’ — construction of the outer shell by the main contractor. So far the project has made good progress and is ahead of schedule — something that didn’t go unnoticed by one of our delivery drivers: “They’re not effing about are they?”
However, getting the main walls up to wall plate level hasn’t been without its challenges. One of the complexities stems from the planners wanting the front of our extension set 20cm lower than the roof on the main house, whereas at the back it needs to be the same height as the existing roof to create the desired dual gabled ‘twin peaks’ effect.
To make matters more interesting, the side wall kicks out 10cm with a short dog leg towards the rear. And although these intricacies are depicted on the drawings, amid all the hustle and bustle on site, it’s easy for details to get overlooked.
Conscious of the design complexities, I arranged periodic site meetings to run through the latest drawings. I know that some builders are more receptive than others to this, but fortunately Colin, our appointed chippie for the extension, is an experienced old school grafter, and we made good headway discussing the technical details of this key stage.
Carpenters building roofs are used to accommodating less than perfect wall plates, but there’s less chance of errors where the two trades are able to liaise in advance. I made a point of insisting that Daniel, the head brickie, checked the dimensions of our existing rear gable that we needed to copy. Getting this perfectly replicated is fundamental to achieving the desired ‘Toblerone’-profiled rear elevation.
To get the levels exact, he ran a water level (or ‘U-tube manometer’) at roof height across to the new replica gable wall being constructed. Liquid-filled tubing is an ancient tool that works on the principle that water always seeks its own level and is the same at both ends. Having double-checked all the measurements we could be confident of achieving the correct span at the base of the triangle with matching ridge heights and roof slopes (42°).
The Roof Structure
The roof design on our extension basically comprises two separate structures joined together. On the main front roof the ridge runs side-to-side. The adjoining back one runs front to rear, abutting the front roof at 90° with mitred valleys at the junction.
Before work could start in earnest the scaffolders popped back briefly to erect the next lift, and we also needed to construct a hefty internal loadbearing stud wall down below to support the main rear roof slope via a pair of doubled up purlins.
The next day, chippie Colin started work creating the skeleton of the rear roof. This is one stage of the construction process that looks extraordinarily precarious — with a single horizontal timber ridge board marking the highest part of the roof suspended in thin air other than the slenderest support from some temporary timber struts. Having carefully checked the dimensions and made any necessary adjustments, the carpenter can set about filling in the missing rafters. Before the day was out, the rear roof structure was fully fleshed out.
The next job was to build the main front roof structure. This was a more complex proposition due to the rafters needing to be lined internally to form sloping ceilings, creating an airy open plan interior.
The engineer had specified a large steel ridge beam (a 178x102x19mm universal beam) to avoid the need for obtrusive purlins supporting the rafters. At the outer end this would rest in the new side gable wall, with the inner end slotted into the existing side wall of the house, and both end bearings resting on padstones.
Health and Safety
I returned from work one day to find the job of lifting the beam was done. It transpired that a team of five blokes had lifted the heavy, 465cm-long ridge beam three storeys solely with ‘grunt power’. To my surprise they had manhandled it up the scaffolding, with additional support en route from a scaffold tower erected internally.
The primary decision about what plant to employ for the quoted works rests with the main contractor. But designers now have joint liability for health and safety, and you don’t need much of an imagination to envisage the potential dangers of slipping from height while carrying incredibly heavy loads.
Although we’d complied with health and safety regulations by drawing up a ‘Construction Phase Safety Plan’ at the outset, and the contractors had previously employed heavy lifting gear for the steels in the main walls, with hindsight, additional measures could have been taken to pre-empt the risk.
The next day work continued as normal, with the timber ridge plate bolted to the top of the newly installed steel and all the rafters fitted. There was some tricky detailing of valley boards and the space between the roof windows, and the valley was a little tight to accommodate the flashings, but nothing that an experienced chippie couldn’t handle.
All that remained was to fit the ceiling joists and bolt the collar ties in place to help secure the rafters. Trimming the projecting rafter feet was left for the time being until I’d determined the extent of overhang required to get the right ‘look’. Meanwhile the brickies could get on with finishing the gable end walls, neatly set off with two courses of indented red brickwork along the verges.
Spotting Potential Problems
No one likes their work being criticised, but getting errors swiftly corrected is an essential part of the project management role. When a mistake is pointed out the automatic response with builders is often to blame the drawings/designer, or if all else fails, another trade. Similarly, when fingers are pointed at designers there’s a tendency to automatically blame the builders.
In truth, mistakes are often down to poor communication on both sides. Either way, the project management goal is to get errors swiftly rectified without antagonising the workforce. The best approach is to avoid the blame game and come up with solutions. In most cases, once you’ve jointly solved the problem no one’s bothered about whose fault it was and work continues on an amicable basis.
That evening I stood in the garden and looked back at the freshly extended house with a mix of satisfaction and relief that my 2D drawings were being successfully translated into tangible reality.
I was busy admiring the stepped red decorative brickwork forming the projecting corbels at the feet of the new side and rear gables when it occurred to me that I ought to take a closer look. So armed with my trusty tape I shimmied up the scaffolding for a spot of measuring up.
The new matching ‘twin peaks’ rear gable was pretty much perfect. However, when it came to measuring the big side gable the figures wouldn’t compute. But the tape didn’t lie; the left-hand corbel was 40mm lower than the right-hand one and out of kilter by 220mm in width measured from a vertical centre line drawn down from the ridge. At this point our neighbours were treated to an explosive broadcast of ripe language.
Once I’d recovered my composure I marked the correct corbel position on the blockwork. The next day, with the mortar still just about malleable, uncharacteristically subdued brickies rebuilt the wayward corbel. Not for the first time a potentially serious problem had been nipped in the bud.
The following Monday, the skeleton roofs were draped with massive rolls of breather membrane sheeting and secured with a few preliminary battens nailed to the rafters. For the first time, my wife Ewa and I could venture into the extension without the precaution of a brolly!
To mark the occasion of the building being weathertight (apart from the windows), we invited the family round to drink a toast. Now all that remains to complete ‘phase one’ is the roof tiling, fitting the windows and rendering the walls.