Today’s visit to St Paul’s Church marked the start of the third year of activities for our ad-hoc society. Simeon Hawkins from the church’s congregation led the way. Sean Kisby of Salmond Reed Architects and Peter Liu and Dr John Jing of EQ STRUC came along to tell us about their work on the nascent project of strengthening and refreshing St Paul’s Church.
The church, explained
Simeon, whose Master’s thesis focusses on the church, gave us a brief overview of its history. The current St Paul’s is the third building to bear the name. It’s a Gothic Revival building, begun in 1894. The church is cross-shaped, as convention dictates. The main body of the church is stone, with brick veneers inside from shoulder height to roof and brick structure below the floor. The roof members are timber. The transepts (the arms of the cross) are also timber, and the chancel, the head of the cross, where the altar is, was made from reinforced concrete and wasn’t completed until 1936.
The building is not quite as its architect William Skinner conceived it. He’d intended for a gallery to sit above the door, housing the choir. The rough masonry you can see around the base of the church was to be covered with timber panelling. The chancel was to be faced with stone. And, most noticeably, no spire was ever built atop the northwestern stairs. Still, the building is Category I in Heritage New Zealand’s list, all the more poignantly since the HNZ listing says “St Paul’s invites comparison with Sir Gilbert Scott’s only New Zealand work, Christchurch Cathedral”.
For my part—although I don’t have favourite children!—I have a soft spot in my heart for St Paul’s. Something about the cheerful pick-and-mix of its irregularly-sized stone arches, its glorious melange of materials, and its air of patient worshipfulness makes my heathen heart glad.
First, do no harm
Sean Kisby picked up the thread. Salmond Reed, he explained, have a dual role at the church. Firstly, there’s a fair hatful of material repairs that need to be done: there’s stone to be replaced; roof coverings to be refreshed; lead and copper flashings to replace, too. It takes expertise in heritage materials to understand what should be done, and how best to do it. Tracey Hartley (and others) from SRA are supplying this knowledge.
In parallel, there’s a real gem of a design project to be done. The building needs seismic strengthening, with the aim of reaching 67% of the New Building Standard. The internal circulation needs to be improved—at the moment, masses of people must file up and down a tiny rear staircase. The main stair, a lovely wooden spiral, has literally rotted away, the victim of a “temporary” roof over the void that was to contain the spire.
With the access reconfigured, and the rooms beneath the chancel refreshed, the long-term plan is to build out into the carpark, and (best-saved-for-last here), at long last to design and build a spire for the church. Surely that’s a commission to gladden the heart of any designer.
Strengthening the church
“Bring the building up to 67% NBS” I wrote above, glibly. So, how might that be done? Peter and John from EQ STRUC have been considering that question. Over the last few months, they’ve carried out an analysis of the building, including a LiDAR scan of the building which created a detailed 3D image of the church. The LiDAR generates a point cloud, essentially a myriad of measured dimensions, allowing a viewer to see details of colour and ornamentation as well as capturing idiosyncrasies in the plan. The engineers also need to know how the mass of the building is distributed, and the point cloud helps to identify this, making later finite element modelling in Etabs or SAP2000 more accurate.
The EQ STRUC blokes were too modest to say this for themselves, so I’m going to say it for them: they’re really good at assessing and then using the strength of existing materials. This is important, very important, for heritage buildings. The conservative approach would be to discount the strength of unreinforced masonry down to almost nothing, necessitating a larger and more intrusive engineering intervention. Think whacking great steel frames marching down the aisles.
For the nave and the aisles, the solution that EQ STRUC have devised uses the existing materials to their own advantage. I need to caveat this by saying that, as yet, this is just a proposed solution, but here it is: the plan is to use the mass of the walls to dissipate earthquake energy by allowing the piers to rock.
How does this work? Let’s work our way down from the roof.
Firstly, a diaphragm will be inserted above the sarking and below the roof, allowing forces to be transferred. (The nave floor’s getting a new diaphragm, too.) The existing timber trusses will be enhanced with steel, stiffening them considerably. At the junction between the trusses and the wall, a moment joint will be created, meaning that the roof and wall can’t rotate towards or away from each other.
Along the walls, between the openings, the spandrels will be strengthened with fibre-reinforced polymer wraps (FRPs). This will greatly increase their ability to resist tension, helping them to stay intact in a quake. [see footnote for a follow-up on this][update April ’18]
Lastly, as noted above, the piers will then be permitted to rock. In a design-level earthquake, the pier will form two hinges: one above the plinth (at the second moulding) and the other below the capital (the leafy bit). The pier will then wobble back and forth on the hinges, dissipating energy as it does so.
How do you stop the pier from toppling over, and bringing the roof down with it? By tuning the stiffness of the moment joints up above, the quantity of deflection at the piers can be controlled. It’s elegant—if not easy—and the great virtue of the solution is that you don’t need to ruin the spatial qualities of the interior with ugly external bracing.
The real doozy of a problem might be the western gable end. How best to brace the wall, given the aesthetic and historic constraints at play? The solution may come from the planned completion of the choir gallery, absent from the church for so long. The gallery would run along the wall just below the level of the two long windows in the picture above. As at the Town Hall and Hopetoun Alpha, a gallery at the midheight of a tall wall can be a structural godsend, concealing crucial structural bracing, and reducing the risk of a potentially deadly of out-of-plane collapse
To the Crypt!
Beneath the nave lies the crypt. Although you enter the church at ground level from Symonds St, the ground slopes away steeply to the east, and there’s a generous space beneath the floor of the nave. The space is used for a range of church activities. Here, in this unfussy rough-brick room, it might be forgivable to insert some supplementary steel structure. EQ STRUC are proposing to do so, thus completing the rocking system described above by giving the piers a solid base to wobble on. (The slender piers of the nave are supported on the chunkier brick piers of the crypt, which you can see in the photograph above.)
Down in the crypt, we could see old signs of damage caused to the church by the city growing around it. The church abuts the motorway trench of busy Wellesley Street, dug in the 1970s (? don’t quote me on this date!) As the ground settled post-trench, the foundations rotated outwards, creating tension cracking like that shown above. Don’t panic, though: a geotechnical report showed that the settlement has long subsided, and there’s no fear of further rotation and damage.
What are you made of?
As noted above, and as you can see in these images, the church is made from a variety of materials. There’s stone, brick, reinforced concrete (RC), and timber. How, I asked, would you assess the interaction between the RC and masonry elements of the building?
Here, it seems, the timber transept is invaluable. The timber’s elasticity should take up any differential movement between the RC and masonry sections. This means that the RC and the masonry were able to be assessed as two separate structures, removing a layer of complexity from the analysis.
Given the building’s Category I status, it’s not easy to make a case for destructive testing to establish the strength of materials. EQ STRUC have a database of material tests from similar structures to fall back on, for establishing likely material properties.
Talk to me
In the crypt, I asked Peter, John, and Sean about collaboration between architects and engineers, hoping for dark and ghastly tales to befit at least the name of the surroundings. Instead, they talked about the pleasure of collaboration, of working together and modifying each others’ solutions to arrive at the best result. In all seriousness, such collaboration is a theme of our Society, since we’re composed of would-be members of each profession. It’s great to hear that the working world embraces this philosophy of co-operation.
Dr Jing made a comment that intrigued me. “You need to draw,” he said. “I always take a pen and paper to meetings and to site.” Calculations can be confusing, explanations are often ambiguous, but sketching makes for clear communication. Architects will be unsurprised by this, I think, but it’s food for thought for student engineers.
Sincere thanks are due to Sean Kisby from Salmond Reed Architects, and to Peter Liu and Dr John Jing of EQ STRUC, for their generous sharing of time and expertise. We’re also exceedingly grateful to Simeon Hawkins and Esther Grant from St Paul’s, and to the marvellous people who came and offered coffee and hospitality. Thank you for sharing your beautiful church with us. We look forward to seeing the project develop.
Tracey Hartley touched base, following up on the proposal to use FRPs to reinforce the spandrel areas. Regarding the bricks in the spandrels, she notes “I believe they are a rough pale brick with large joints, originally coloured over with a red ochre finish and lined out (tuck pointed) to make the brickwork smarter looking.” The final design may involve restoring this finish, so covering with FRPs may be more difficult. Also, the breathability of FRPs needs to be determined so that moisture isn’t trapped in the brickwork.
I thought it was interesting to include this as a footnote, to illustrate how the process of design is a compromise or collaboration between the expertise and priorities of the professionals working on a project. Working together, the architects and engineers will devise an acceptable solution.
Update April ’18
I ran into Dr Jing today and we talked about the FRPs for the spandrels and piers in the nave. Dr Jing explained that the FRPs in the spandrels were never going to be added onto the surface of the bricks, as I had thought. Instead, they would be mounted into little slots cut into the bricks. The slots would be ~3-4mm wide and span across several brick units, being anchored at top and bottom into bolts. This near-surface mounting would make the FRP intervention much less visually intrusive than if it were covering the surface (as per Tracey Hartley’s note above). To reiterate, it was never the plan to mount the FRPs on the surface: the EQ STRUC team were always intending them to be near-surface mounted.
One final point. I hadn’t picked up that the piers were also to be FRP wrapped, not just the spandrels. The FRP wrapping would naturally add great confinement strength to the pier. However, by varying the length and position of the wrap, the EQ STRUC engineers can force the plastic hinge to form in the places they want it to: at the top and bottom, as per the pictures above.