The document summarizes a case study of a bearing capacity failure that occurred at the Transcona Grain Elevator in Canada in 1913. Plate loading tests prior to construction underestimated the risk of failure. The structure experienced rapid settlement and ground heave, resulting in the collapse of parts of the structure. Theories of bearing capacity including Prandtl's solution and limit equilibrium analyses were later used to model the soil conditions and better understand the failure. The analyses showed that considering the weaker soil layer could have correctly predicted the failure under the load, highlighting the importance of accurately characterizing subsurface conditions.
The document summarizes a case study of a bearing capacity failure that occurred at the Transcona Grain Elevator in Canada in 1913. Plate loading tests prior to construction underestimated the risk of failure. The structure experienced rapid settlement and ground heave, resulting in the collapse of parts of the structure. Theories of bearing capacity including Prandtl's solution and limit equilibrium analyses were later used to model the soil conditions and better understand the failure. The analyses showed that considering the weaker soil layer could have correctly predicted the failure under the load, highlighting the importance of accurately characterizing subsurface conditions.
The document summarizes a case study of a bearing capacity failure that occurred at the Transcona Grain Elevator in Canada in 1913. Plate loading tests prior to construction underestimated the risk of failure. The structure experienced rapid settlement and ground heave, resulting in the collapse of parts of the structure. Theories of bearing capacity including Prandtl's solution and limit equilibrium analyses were later used to model the soil conditions and better understand the failure. The analyses showed that considering the weaker soil layer could have correctly predicted the failure under the load, highlighting the importance of accurately characterizing subsurface conditions.
The document summarizes a case study of a bearing capacity failure that occurred at the Transcona Grain Elevator in Canada in 1913. Plate loading tests prior to construction underestimated the risk of failure. The structure experienced rapid settlement and ground heave, resulting in the collapse of parts of the structure. Theories of bearing capacity including Prandtl's solution and limit equilibrium analyses were later used to model the soil conditions and better understand the failure. The analyses showed that considering the weaker soil layer could have correctly predicted the failure under the load, highlighting the importance of accurately characterizing subsurface conditions.
Case 4 Bearing Capacity Failure: T Transcona Grain G i Elevator, El Canada C d 加拿大特朗斯康谷仓 Outline 4.1 Case description 4.2 The theory 4.3 The analyses 4.4 Mitigation measures 4.5 Lessons learnt Where is it ?
Transcona existed because of the
railroad that ran through it. it
Winnipeg Transcona got its name from a
combination of "transcontinental" and d "Strathcona". "St th " Construction History
B Because TTranscona was a transportation i hub, h b it i would ld make sense to locate a grain elevator there to relieve the loads during the months of peak grain-shipment Construction started in 1911 and completed in Sep 1913 Residents of the Transcona area flocked to the site of the collapsed grain elevator
Construction in progress Photo courtesy of the Transcona Historical Museum Structure
Work house Bin house Foundation The Failure Photo taken before failure
Photo taken after failure
Photo courtesy of the Transcona
Historical Museum The Failure 18 Oct, 1913 11 am Movement noticed on the bridges
Noon The bin house had settled about 30 cm The 7.5 7 5 -9 9.0 0 m wide strip of ground around the bin- bin house (except for the south side, where the work-house stood) heaved up 1.2-1.5 m. 1 pm The settlement rate became higher on the west side, producing a tilt to the west. 9 Oc 19 Oct,, 1913 9 3 The structure continued settling and tilting Noon The earth on that side bulged up, forming a cushion which slowed down the movement. the connecting bridges carrying the conveyor belts breaking Night down and crashing to the ground the cupola structure housing the conveyor over the bins suddenly collapsed and fell to the ground. The Failure
27º
Heaved soil
1.5 m 9 m The Problem Differential settlement? The Problem Bearing Capacity Failure?
Plate loading test
The Problem Bearing Capacity Failure?
Pcr Pu P 0 1 2
Plate loading test S
p-ss curve p The Problem Bearing Capacity Failure? the smaller-scale plate loading tests predicted a safety factor of more than 1.3 1 3 !!! Pcr Pu P 0 1 2
Plate loading test S
p-s curve The Problem Bearing Capacity Failure?
(After Peck and Bryant, 1953)
The Problem
U Unconfined fi d compression i ttestt
τ φu=0 cu τf Dial gage Loading cell
O σ3=0 qu=σ1 σ=σ′′
Loading Soil frame
) τf=cu =qu/2 The Problem Bearing Capacity Failure?
cu= 54 kPa
cu= 31 kPa
(After Peck and Bryant, 1953)
The Problem
Excessive settlement vs. Bearing capacity failure
Fast settlement Ground heave Outline 4.1 Case description 4.2 The theory 4.3 The analyses 4.4 Mitigation measures 4.5 Lessons learnt The theory
Prandtl solution (1920)
Reissner (1924) Taylor (1948) P 45o＋ϕ / 2 d 45o－ϕ / d′ Ⅰ Ⅲ 2 Ⅱ Ⅱ Ⅲ c cc′ b The theory
Prandtl solution (1920)
Undrained bearing capacity ϕu = 0
The theory
Prandtl solution (1920)
Undrained bearing capacity ϕu = 0
Nc=5.14 The theory
The ultimate bearing capacity
Undrained bearing capacity formula
ϕu = 0,, cu Nγ=0, Nq=1, Nc=5.14
Terzaghi formula
Shape p correction factor Depth p correction factor The theory
1. For homogeneous soils, Prandtl solution provides an exact solution.
2 F 2. For two-layer l strata, Prandtl P d l solution l i cannot b be applied li d di directly, l and d its approximation using a weighted average of shear strength with depth is not rigorous. 3. Upper Bound Limit Analysis can provide some useful results, using simpler kinematic failure mechanisms, such as the scoop mechanism 4 The values of the bearing capacity factor are rigorous upper bounds 4. and are larger than the true collapse load. Therefore, they are not conservative and an important question has to be answered: how far are we from the true collapse load? Outline 4.1 Case description 4.2 The theory 4.3 The analyses 4.4 Mitigation measures 4.5 Lessons learnt Model parameters Model parameters Model parameters Pcr Pu P 0 1 2
Plate loading test S
p-s curve Predictions
1 Th 1. The b bearing i capacity it assumed d iin th the original i i ld design i
Terzaghi formula
Shape correction factor Depth correction factor
Consider the stiff layer
Cl Close to plat-loading l l di test result l !! It will not fail in a load of 300 kPa !! Predictions
1 Th 1. The b bearing i capacity it assumed d iin th the original i i ld design i
Terzaghi formula
Shape correction factor Depth correction factor
Consider the soft layer layer
I will It ill fail f il in i a load l d of f 300 kP kPa !!! Only 20% conservative than the true failure pressure. Predictions
2 Consideration 2. C id ti off ttwo-layer l strata t t
Plandtl solution
10% larger g than the true failure pressure p !! It will not fail in a load of 300 kPa !! Predictions
2 Consideration 2. C id ti off ttwo-layer l strata t t
Scoop mechanism
Marginally larger than the true failure pressure !!
It will fail in a load of 300 kPa !! Discussion
1. The analysis confirms that insufficient bearing capacity was the most likely cause of the Transcona Grain Elevator failure. 2. The Prandtl mechanism would provide a reasonably good prediction of the bearing capacity if the soil was homogeneous, as confirmed by the p plate load tests,, where the failure mechanism was entirely y confined to the upper clay layer. 3. The real mechanism was much deeper due to the large foundation width and penetrated a weaker lower layer layer. 4. An approximate approach using the Prandtl formula with averaged shear strength appeared to be neither accurate nor conservative. 5. The upper bound limit analysis using a scoop mechanism provided a remarkably good prediction. Outline 4.1 Case description 4.2 The theory 4.3 The analyses 4.4 Mitigation measures 4.5 Lessons learnt Mitigation
1 Emptying of the elevator
1.
Belt conveyor Mitigation
2 Underpinning of the work-house
2. By June 1914
Reinforcing the foundation and the structure before
underpinning operation. operation Excavating access tunnels under the foundation mat. Sinking a 1.5 m diameter pier under each of the 24 columns of the building. Mitigation
3 Straightening of the bin-house
3. Feb - Oct 1914
Timber pushers
Under-excavation at the east (high) side
Mitigation
3 Straightening of the bin-house
3. Feb - Oct 1914
Jacking-up the west (low) side
Mitigation
4 Underpinning of the bin-house
4.
The elevator in righted position after mitigation (October 1914)
Outline 4.1 Case description 4.2 The theory 4.3 The analyses 4.4 Mitigation measures 4.5 Lessons learnt Lessons learnt
Site investigation 9 The importance of the proper geotechnical site investigation.
Field load tests
9 The field plate load tests also did not help to avoid the failure. This happened due to the scale effect. 9 in order to be able to make meaningful predictions based on the field load tests, these tests have to be performed at thereal one-to-one scale or at different elevations.
Conservative design 9 A conservative prediction may be achieved by assuming that the soil is homogeneous with the undrained shear strength of the weaker layer.
Upper bound limit analysis
9 An upper bound limit analysis using a simple kinematic mechanism can provide a useful tool.
