Tower cranes and pumping concrete 1km up – the unprecedented engineering behind the $1.2bn (£900M) Jeddah Tower, is elegantly simple and brutally complex.
More from: Super tall, super smart | Overview
“Conceptually, it’s very simple,” says structural engineer for the tower, Thornton Tomasetti principal Bob Sinn. “It’s easy for people to understand and build with the exception of the extreme height aspect.
“The trick is to come up with a system that’s not exotic, that’s not strange and take examples like the Burj Khalifa and the CN Tower and use them to show what can be done.”
Aerial view of a wing of the Jeddah Tower
Source: Copyright of Jeddah economic Company
The soon-to-be-iconic tower began its design life in 2009, after a team from architect Adrian Smith & Gordon Gill together with structural engineer Thornton Tomasetti won a competition to build a 1km high tower in Jeddah in the Middle East. Construction on site began in 2013.
Simplicity of form was an important part of the design and has led to a number of benefits.
One of the most important is that the tower is extremely stable under lateral loads, which when building at this height, and despite being located in a seismic zone, come primarily from the wind.
As soon as the team started designing the tower, it carried out a series of wind tunnel tests. In total, three different wind tunnel tests have been conducted throughout the design period to ensure that it can withstand the most critical loading of all.
Initially, the team carried out a high frequency force-balance test at a scale of 1:800 (around 1.25m tall) at a Canadian lab at RWDI, a wind tunnel testing company. The test measured a rigid model with all measurements taken at the base of the tower.
Architectural rendering of the kilometer tall tower
Source: Courtesy of Adrian Smith + Gordon Gill Architecture
“The results were quite positive, the shape of the tower didn’t really change much from the competition, we just had to keep building on the first test,” says Sinn.
The second test was a high pressure integration model at a slightly larger 1:600 scale (around 1.7m tall). This model allowed the team to put pressure taps, or sensors, over the surface of the model to instantaneously measure the different pressures on the structure. These pressures were then integrated to give design forces and movements.
However, even at this larger scale, physically threading the pressure taps through the building at the top of the slender structure proved difficult. To cope with this, a separate 1:400 (around 2.4m) scale model of the top of the tower was tested and the results were combined.
Lastly, the team carried out aeroelastic modelling of the building. During this test, the tower was allowed to move and therefore feedback on the effects of wind on the tower could be analysed, in a different way to the previous tests.
Latest construction july 2016 of Jeddah Tower
“The last one we did with RWDI was the aeroelastic model, which is the most expensive and the most sophisticated model of all,” says Sinn. “That’s the only type of model where the stiffness, damping and the mass of the tower are simulated, so the tower moves during the testing. During all the other tests, the tower is rigid.
“The results of that showed no significant differences and the building is very well behaved.”
Finally, the team did a series of tests on the model in another wind tunnel to compare the results.
“We did something which doesn’t happen a lot, but we did a companion model in another wind tunnel,” says Sinn. “The two were quite similar, not exact, but enough.
“So we had confidence.”
Wind tunnel tests explained
High frequency force-balance test (HFFB): Probably the most common method for obtaining wind loads on buildings – a force-measuring device is fitted inside a rigid model. It provides base overall loads and an estimate of accelerations at the top of the tower, as well as floor-by-floor loads. These tests are quick, affordable and can be used to optimise the building structure or aerodynamics.
High pressure integration model (HFPI): For this, a rigid model of the building is equipped with several hundred pressure sensors on all exposed surfaces. The measured pressures are integrated to obtain the overall wind loads at the base, and to gain a good understanding of the load distribution with height. Data can also be used to estimate façade loads. This is slower and slightly more expensive than HFFB, but it has the advantage of providing data for structural and façade design.
Aeroelastic modelling: This is the most expensive and detailed wind tunnel test method for buildings. Unlike the previous methods, the aeroelastic models are designed and built to vibrate just like the real structure, so they are dynamically calibrated. This is the slowest and most expensive form of wind tunnel testing for buildings, typically only carried out for super-tall flexible structures.
When concrete cures, it shortens which can have an impact on the structure as a whole where there are unconnected elements. However, in the Jeddah tower, the structure is made entirely up of shear walls which are all seamlessly connected by beams.
“One of the biggest potential problems is differential shortening but we don’t have that on the tower as it is relatively well behaved,” says Sinn. “One of the nice things about this system is that everything is connected so that there are no columns outside the core which want to shorten by themselves.”
Although the design of the tower is simple in its concept, the construction is anything but.
“The structure is relatively simple – it’s very similar to the [829.8m tall] Burj Khalifa, but once we get up to about to 600m things change drastically,” says Mace construction director Peter Savoy.
At present, the tower has reached level 43 with a milestone of around 25% of the total concrete now poured on the job. As of yet, the challenges of building have been relatively routine, but as Savoy explains, the difficulty of construction will increase drastically as it passes from super tall – more than 300m – to mega tall – more than 600m.
“Everything above the observation deck becomes difficult just because of the height and the logistics of trying to maintain the programme with the technical challenges,” says Savoy. “With a building like this with all of the constraints, there’s only one thing you can do in some cases.
“If you find at 700m that you’ve made an incorrect decision with the cranage, then potentially to fix that can have a massive impact on the permanent works, cost and programme.”
Cranes and the wind
Initially, cranes are mounted on the central core and each of the wings. But as the wings reduce in size due to the sloping exterior walls, only the central crane will remain.
Savoy explains that during construction cranes will hydraulically jump themselves up the building as it gains height. This is relatively straightforward, but jumping them fast enough will be one of the major challenges.
“If you have three cranes, they need to jump regularly to stay above the structure,” says Savoy. “If there’s any problem with jumping the cranes then the structure stops. The cranes are the life blood of the structure.”
The sky terrace, 638m up the tower is another area which has challenged the contractors. Traditionally the structure might be broken down into smaller pieces and lifted into place using tower cranes. Savoy explains that using this technique is slow and exposes steel erectors to more risk when reassembling the structure at high level.
To avoid this, an alternative solution to strand jack the giant 750t steel structure 638m up the tower has been proposed. However, this is not without its own challenges.
“Strand jacking – that in itself gives major challenges,” says Savoy. “When you look at the weight of the strands over 600m, it’s a significant weight – around 15t per strand, so about 60t on strands alone.”
”This, coupled with the fact that the lift will take an estimated 90 to 100 hours to complete, will make finding a suitable weather window to carry out the works more difficult, he says.
“Anywhere between 350m and 750m [up] is where the wind will blow its strongest,” explains Savoy. “We’re modelling and extrapolating the wind from 10m and we’re coming up with the likely forecasts every 100m.
“We have to plan the methodology around the weather circumstances, that’s a tremendously technical piece of work.”
One area where the team is thinking of using more traditional techniques is when lifting the dampers 750m into the spire, where they will control its movement. However, at the stage where they will be lifted into place, the tower will be relying on the sole crane mounted on the central core. Breaking the dampers down into 6t to 8t sections, the team has predicted that this will equate to a solid lifting time of around 10 days, which will put pressure on the construction of the surrounding structure.
“At that point, we have only one crane, so the challenge is that we don’t hold up or delay any other parts of the structure,” says Savoy. “Especially as, when we get to those heights, we’ll just be passing the Burj Khalifa and then all pressure will be on to pass that and become the tallest building in the world.”
Pumping concrete to 1km in the air
The height of the tower will also bring challenges for the structural concreting work.
“The structure is concrete until 40m to 50m from the top so we have to deliver concrete that high,” says Savoy. “Right now, the record for delivering concrete is about 624m.
“Is it challenging? Yes it is. There’s no pump that will get the concrete to the top of the building.”
“The higher you pump, the more pressure on the pump and there’s a limit to that,” says Sinn. “The slickline has to be supported along its entire length and there’s waste, because there’s some that sticks to the sides of it.”
To overcome this problem, the team will use a two stage pumping system. Initially, concrete will be pumped up to 500m after which it will then be pumped up to its final height by a second pump.
However, finding a concrete mix which meets the strength criteria, the modulus of elasticity criteria – on which the flexibility of the tower has been calculated – and which can be pumped to the vast height of the upper reaches of the tower, is no easy task.
“Once you get all those things meeting in the middle you should be ok,” says Savoy. “But it takes an awful lot pre-testing and monitoring of certain mix designs to get to a point where all three of those requirements meet in the middle. That’s extremely challenging.”
The team is currently trialling 21 different mixes. However, before adopting a mix, it has to go through a rigorous testing programme. Mixes are monitored at seven, 21, 56 and 90 days at which point the team gets an idea of whether they are suitable. Out of this possible set, only six, which are spread over three different strengths, have been deemed to satisfy the performance specification.
“If a concrete doesn’t work, we’ve got to start a new testing programme which can be fairly serious if we’ve got to go through to a 90 day programme again,” says Savoy. “That’s why we started out with so many mix designs.”
Progressing this further, the team is now taking the successful mixes, and carrying out 500m long horizontal concrete pumping tests as the precursor to pumping vertically.
Lifts and moving 9,000 people around the site
In the finished building, there will be a total of 54 lifts carrying people at speeds varying between 8m/s and a very fast 12m/s. Two double decker lifts will become the highest travelling lifts in one complete run up to the observation deck 600m up.
This may be impressive enough, but the hoists required to move people, materials and tools during the construction phase are just as important.
“The vertical transportation plan will determine whether the contractor is going to be superefficient or not,” says Savoy. “It’s not only bringing the lifts together at the right time, but during construction, trying to get beneficial use of them.
“If you can’t get people and materials up the building efficiently, it’s going to take longer and cost more money.”
With this in mind, the team is planning to complete up to 14 of the permanent lifts early, so they can be used during the construction period.
However, even with the additional lifts, it is estimated that starting at the base of the tower and going to an area of work could take as long as 13 minutes.
“If you calculate going up at the beginning and down at the end, that’s 10% of the guy’s production time,” says Savoy. “That amounts to quite a lot when you multiply that by the 1,000 to 2,000 people who are working in that part of the building. Then add in the time for going to the welfare facilities.”
“When you add all of that you realise why you need regular facilities and a good temporary hoist and vertical transportation plan.”
“If you think about it, tall buildings are one of the few man-made products which aren’t 100% validated at full scale,” says Sinn. When you think of an automobile, or an aeroplane they’re tested to destruction.
“With a building like this, you start with a computer model and then you go through the routines and then you go to a wind tunnel which isn’t close to the full size.”
With this in mind, the team has put long term monitoring systems in place to record the structural health of the building. From the bottom of the piles and throughout the tower, strain gauges measure the building’s movements, while near the top of the spire, anemometers record the wind speeds will be installed. The pieces of equipment will give engineers around the world a better picture of how these types of extreme buildings actually perform.
“By surveying the building throughout its life, we’re hopeful that with this tower we can get some of the data and then hopefully make it available for all engineers to use as a benchmark,” says Sinn.
Structural Engineer : Thornton Tomasetti
Architect Adrian Smith & Gordon Gill.
Contractor : Saudi Binladin Group.
Client project manager : JV comprising Mace and EC Harris
As is common practice in the Middle East, the entire building will be constructed from reinforced concrete with the exception of a circular steel “sky terrace” which cantilevers out from the main structure at a height of around 600m. Its form builds on that of other tall towers such as the Burj Khalifa, but is most like the CN Tower in Toronto, Canada,
The building will be occupied up to floor 168 (although the highest non-mechanical floor is the upper observatory floor at level 159, a huge 638m up) after which an unoccupied 330m high spire makes up the rest of the building. It has a simple triangular core in the centre with a inclined vertical “wing” extending from each the three sides. Each of the wings will be seamlessly connected to the core by a series of continuous shear walls which range from 1.2m thick at the tower’s base to 600mm thick at the top. The strength of the concrete used in the design was limited to 85MPa to meet contractor concerns that it would not be possible to pump higher strengths high enough up the tower.
Jeddah Tower Structure
The wings terminate at differing heights which is achieved by sloping their end walls inwards at differing angles.
The structure of the tower, which accommodates a hotel, office and residential grid, is relatively simple says Sinn. A clever layout means that there are no transfer structures and the vertical supports are made up entirely of walls, meaning that there are no columns in the building.
The building is supported on a piled raft system, made up of a combination of deep bored piles and a 5m thick, 18,000m3 concrete raft. The 270 bored piles reach depths of up to 105m and vary from 1.5m to 1.8m in diameter. Although the strength of the ground is good, there is no bed rock for the tower to bear onto, so the tower’s load is shared between the skin friction of the piles, which are bedded into coral limestone, and direct bearing of the raft on the surrounding ground.
The floorplates are simply 250mm thick flat slabs, designed for speed and ease of construction.
At level 168, a 4m thick reinforced concrete sky “raft” slab divides the occupied space from the unoccupied space and acts as the “foundation” for the spire above.
The spire is a massive 330m tall structure taking the number of storeys in the building up to a massive 240. To put this into context, western Europe’s tallest building, the Shard in London, is 306m tall. The spire is essentially a closed triangular pyramid with no openings in its walls to the exterior.
Up to the spire, due to the structure’s stability, no additional damping was required to control movements of the tower for occupancy comfort. However, in the spire two simple pendulum tuned mass dampers – one 870t installed in the bottom third of the spire and one 260t damper stacked vertically on top of the other in the upper third of the spire – will be installed to help control the tower’s movement for the occasional maintenance worker.