Zaryadye Park, Glass Grid Shell Roof

Date: 26 November 2018
Copyright:
  • G. Vasilchenko-Malishev, Malishev Engineers
  • First presented at GPD 2017 – go to www.gpd.fi for more info and 2019 plans
Artist impression of the completed grid shell structure
Photo source
www.gpd.fi

Date: 26 November 2018

This paper will describe the design, detailing, testing and construction of structural glass beams as part of load bearing grid shell structure in the newly built Zaryadye Park in Moscow, Russia.

First presented at GPD 2017

 

Abstract

This paper will describe the design, detailing, testing and construction of structural glass beams as part of load bearing grid shell structure in the newly built Zaryadye Park in Moscow, Russia, situated short distance from Red Square and the Kremlin. Glass beams (72 in total) are fixed into the main steel grid shell structure measured around 120m long by 60m wide and measured around 3m long by 0.2m deep.

Fig. 1 Artist impression of the completed grid shell structure
Fig. 1 Artist impression of the completed grid shell structure

Glass beams were designed to accommodate extreme weather conditions with snow drift loads of up to 350 kg/m2, as well as differential movement of the main structure which required sophisticated analysis with more than 100 load combinations as well as full scale testing to gain approvals from the authorities.

Due to lack of legislation on use of structural glass in Russia, so called “special technical standard” was created with our help to cover technical aspects of glass and it’s performance, which formed part of approval documents.

Fig. 2 Situation plan of the Zaryadye Park
Fig. 2 Situation plan of the Zaryadye Park

 

Design

3.1 Concept

Fig. 3 general arrangement plan of the grid shell structure with highlighted area representing structural glass beams as part of the main structure
Fig. 3 general arrangement plan of the grid shell structure with highlighted area representing structural glass beams as part of the main structure

The grid shell structure was designed to follow the curve of artificially created hill as an open structure (not enclosed), hence subject to thermal movements as well as snow loading and wind. The main structure is comprised of structural steel elements of approximately 3m long and 300mm deep.

However in certain locations clear less obstructed views were required, hence more transparency in the structure was necessary. Initially glass clad cable structure was proposed by the architect, however due to significant snow drifts high prestress forces were required which proved to be difficult to achieve hence different solution needed to be found.

Fig. 4 General arrangement plan of the glass beams, each approximately 3m long. Roof panels are not shown for clarity
Fig. 4 General arrangement plan of the glass beams, each approximately 3m long. Roof panels are not shown for clarity

 

3.2 Design loadings

Scaled model of entire project was wind tunnel tested to ascertain wind loads and possible snow drifts associated with that. Resulting loads are shown below:

In total there were 19 load cases which included: 3 cases for dead load, 5 variations on snow, 9 load cases for wind and 2 cases for the thermal movement. All those load cases generated over hundred different load combinations.

Design loadings

Fig. 5 Artist impression of the desired view
Fig. 5 Artist impression of the desired view

 

3.3 Load combinations used

Load combinations taken in accordance with Eurocode 0&1:

ULS: 1.35Gk + 1.5 Qk1 + 1.5ψ0 Qk2

SLS: Gk + Qk1 + ψ0Qk2

Fig. 6 Dotted lines indicate proposed structural glass beams.
Fig. 6 Dotted lines indicate proposed structural glass beams.

 

3.4 Glass selection

Based on draft version of European standard for use of glass in structures, following design values were adopted for this project, see table 1&2 on the left: Glass beam thickness was adopted as 5x10mm toughened glass laminated with PVB interlayer each 1.5mm thick.

Table 1: glass properties adopted, based on prEN13474, replaced by prEN16612
Table 1: glass properties adopted, based on prEN13474, replaced by prEN16612

 

3.5 Other materials used in the construction

Table 2: other materials properties
Table 2: other materials properties

 

3.8 Computer modelling

Node coordinates were imported from the main engineer’s model with associated deformations under individual load cases. Node deformations were combined with individual loadings for each glass beam and then analysed. Over 100 different load combinations were generated and analysed.

Fig. 7 Bending moment diagram
Fig. 7 Bending moment diagram
Fig. 8 Shear force diagram
Fig. 8 Shear force diagram
Fig.9 Axial forces
Fig.9 Axial forces
Fig.10 Torsion
Fig.10 Torsion
Fig.11 Max stress
Fig.11 Max stress

 

3.9 Bracket design

Initially, bolted connection was assumed however from mode detailed analysis and assessment it was clear that relatively rigid bolt connection details causing glass overstress in various locations. Hence softer, more flexible joint details was required.

Detailed analysis of node deformations in glass fixing points was carried out and summarized in Fig 13 on the right side. Based on this, nodes are predicted to be rotating approximately 1 degree around each axis. One rotation requirement was established, joint stiffness could be calibrated to minimize stress concentrations in glass.

To model and predict stresses within flexible joint, non-linear solid modelling was carried out using Mooney-Rivlin behavior model and stress strain curve tabulated data based on information from silicone manufacturer. Adopted model has shown silicone behaving within allowable stress range, see fig. 14 on the right side.

Fig.12 Fixing detail options a) original bolted solutions b) flexible joint
Fig.12 Fixing detail options a) original bolted solutions b) flexible joint
Fig.13 Angular displacement of a node to be acomodated by the connection
Fig.13 Angular displacement of a node to be acomodated by the connection
Fig.14 Non-linear analysis of the silicone joint was carried out to ascertain stresses and deforemations
Fig.14 Non-linear analysis of the silicone joint was carried out to ascertain stresses and deforemations

 

Testing

4.1 Full scale mock up

Due to unprecedented nature of the project in this country, it has been decided that full scape testing mock up is necessary. The tests were carried out in the Glass Institute facilities in Moscow, on the 26th April 2017.

Fig.15 typical triangular bay assembly
Fig.15 typical triangular bay assembly

 

Production

5.1. Glass production

Glass production was carried out during the course of April 2017 by Modern Glass Ltd, Chelyabingks, Russia with installation due to start on site in May-June 2017.

Fig.16 fully loaded mock up, taking 3 tonnes of sand bags
Fig.16 fully loaded mock up, taking 3 tonnes of sand bags

 

Conclusion

In conclusion, this project is our first experiment with use of structural glass beam element in grid-shell type construction which in our view has a potential to be scaled up.

 

References

Institution of Structural Engineers: Structural use of glass in buildings (1st ed). London, IStructE, (1999)

Institution of Structural Engineers: Structural use of glass in buildings (2nd ed), London, IStructE, (2014)

 

G. Vasilchenko-Malishev, Malishev Engineers

www.malishevengineers.com

600450 Zaryadye Park, Glass Grid Shell Roof glassonweb.com

Others also read

The four point bending test is one of the most commonly used and standardised tests to determine the mechanical properties of materials.
Structural sealant glazing systems are increasingly used in transparent glass constructions, but designing bonded joints still relies on outdated safety concepts, despite recent research proving the reliability of hyperelastic silicone bonds.
A Composite Approach in the Design of Glass-Aluminium Facades to Minimise Embodied Carbon Emissions
The aim of this work is to investigate how the load level affects the crack propagation and to generate an experimental database for L-shaped soda-lime glass specimens with curved crack paths.
The purpose of this study is to investigate the stress distribution along the sealant joint of a cylindrically curved glass panel subjected to wind pressure and to establish if the panel curvature influences the stress distribution along the joint length.
The aim of this paper is to provide a better understanding of the effect of the permanent shear deformation on SSG joints.

FROM INDUSTRY

Pol. Ind. Penapurreira Parcela C4-B,
15320 As Pontes de García Rodríguez A Coruña
Spain

2200 W. Salzburg Rd.
Auburn , MI 48611
United States

Ottergemsesteenweg 707-Zuid
9000 Ghent
Belgium

ARTICLES RELATED PRODUCTS

Add new comment