For example, evidence like knowing a steel girder stands upright on blocks on the girder-factory floor or outside before it goes to the bridge construction site.  Not braced sideways, not even a little. (Ref. 1)  Nor pulled sideways at the factory by a sling and cable connected to a crane oscillating in wind so strong workers left the site for safety reasons.

***

The bridge on 102 road over Groat Road, Edmonton failed during construction when girders buckled early on the morning of March 15, 2015. (Ref. 2, also Sources below)

The bridge consists of seven, 40-tonne girders.  Each girder consists of two 7.5 metre long end sections and a 43 metre long middle section.  The end sections are 4.5 metres deep arching up to 3.0 metres at the middle section.  The sizes are approximate.

Six of the seven girders were in place at the time of the failure.  Three of the six girders failed when the 43 metre long middle sections buckled sideways.

The girders are lifted into place by a crane with a cable and a sling attached to the top (flange) of the girder.  The sling was still attached to the outside girder, the last one placed that day, when the workers went home.

I analysed the reason the girders buckled to demonstrate how experts form and modify hypotheses on cause during a forensic engineering investigation. (Ref. 3)

News reports state that the preliminary cause of failure was inadequate cross-bracing.  I don’t agree.  As well, some time ago, I concluded after my second modified hypothesis that the failure was due to a construction crane tugging on the top of the outside girder together with inadequate cross-bracing. (Ref. 3)  I no longer agree with that combo conclusion either.

A third modified hypothesis is needed.

The tiny bit of evidence for this? If steel girders can rest on the ground in the girder-factory yard with no cross-bracing, and not bend sideways in the middle or topple over, why not overnight on the bridge?

Do you want more evidence? The ends of the middle girders that buckled were also bolted to the braced end girders.  This added to their inherent stability that kept them upright on the girder-factory floor or on the ground outside.

So what happened in Edmonton in March, 2015?

There may have been contractual requirements to put cross-bracing in place but the lack of adequate bracing was not the cause of the girders buckling.  The girders buckled because the crane boom oscillated in the wind overnight causing the crane cable and sling to tug on the top of the outside girder until it bent sideways.  There was no other reason. The two inner girders bent because they were attached to the outer girder by a bit of cross-bracing.

***

Like I said, the wind was so strong that night that the construction site was shut down for safety reasons.  Crane operators lower their crane booms in strong winds. (Ref. Sources)  Why wasn’t the boom lowered that night?

The 1.0 metre buckle sideways was well in excess of the 0.3 metre buckle that would occur under service or construction loads. (Ref. Sources) The magnitude of the buckling indicated that the force from the crane tugging on the top of the girder was greater than the construction load, perhaps much greater.  Why wasn’t this tugging load included in the construction load by the bridge designer?

***

If all bracing was in place I would still put money on all the girders bending at least a little, say, a few millimetres to centimetres, as the crane tugged on the top of the girder.  Some pictures may show that the adequately braced end girders where they were connected to the middle girders bent sideways a little too.

***

Life was good for the quite stable, free-standing girders at the girder-factory.  Then they were taken to the bridge construction site and hooked up to the oscillating crane boom and tugged sideways.  They didn’t stand a chance after that.  The Edmonton bridge was destined to fail for that reason alone – a failure waiting to happen.

References

1. Jamie Yates, P.Eng., Civil Engineer, J. B. Yates Engineering Ltd, Fall River, Nova Scotia, Canada. Personal communication, March 9, 2020
2. Google: Edmonton bridge failure, Groat Road, Buckling, etc. to see photographs of the buckled girders.
3. Bridge failure in litigation due to inadequate bracing – City of Edmonton.  But, inadequate for what?  Posted March 15, 2016

Sources

I studied various photographs on-line including construction photographs taken at the time of the failure.

I spoke with Barry Bellcourt, the Road Design and Construction Manager for the City of Edmonton in 2015, also Bryon Nicholson, Manager of Special Projects. Barry mentioned the litigation and the city’s position.

I also learned from him that the bridge consists of seven, 40-tonne girders.  Each girder consists of two 7.5 metre long end sections and a 43 metre middle section.  The end sections are 4.5 metres deep arching up to 3.0 metres at the middle section.  The sizes are approximate.

I saw and photographed the underside of the repaired bridge girders from Groat Road in early August, 2015 when I was in Edmonton.

I understand it was windy the night the girders buckled and that was the reason workers were not on the job.

I spoke with four companies in Nova Scotia that operate cranes.  I learned that crawler crane booms move in the wind; flex and sway.  There is greater movement sideways because there is less strength that way.  Telescopic booms move more than lattice booms because of the greater surface area.  Booms are lowered to the ground in strong winds.  One company doesn’t operate its cranes in winds of 50 km/hr or more.

I also talked with Amjad Memon, P.Eng. a structural engineer with the Nova Scotia Department of Transportation, about the Canadian Highway Bridge Design Code.