Looks like the counterweight support structure gave way allowing face-plant on the far side.

Agree, Dave. Must have made a helluva noise when that span came down on the abutment. Looks like some web buckling near the end, but mostly just bending deflection.

Aug. 16, 2024
https://northernontario.ctvnews.ca/cn-begins-repai...
All of the newer news postings seem to be the same information.
Eventually Transportation Safety Board of Canada may publish a report on this collapse.

Pure speculation, but I winder if the control room convesation went something like this;
"Charlie. How many times do I have to tell you?
You have to retract the dogs and wedges before you engage the lift."

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

I would expect that there was an interlock which would have not allowed the counterweight mechanism to engage if the 'dogs' had not been disengaged.

John R. Baker, P.E. (ret)
Irvine, CA
Siemens PLM:

The secret of life is not finding someone to live with
It's finding someone you can't live without

Finding some pre collapse photographs was hard. I eventually found these.

What surprises to me is how rapidly railroads respond to incidents. It's like they have drawn up plans for every possible situation and have equipment and material at the ready.

Of course, they also seem to have situation that they don't do as much preventive maintenance as one might expect, but I just saw a water main repair of a break from 4 months ago that left a major intersection with a rather large hole, maybe 40 foot by 40 foot by 10-15 feet deep; completely impassible.

Rail road repair seems like that would be 2 days and ready to use.

I'm sure the timing is just a coincidence.

https://www.theguardian.com/world/article/2024/aug...

The articles don't say, but I would have to think it failed while lifting the bridge because why else would the span be damaged.

Quote:

failed while lifting the bridge because why else would the span be damaged

Opening or closing, gravity works in the same direction.

Like many utility companies, there is a limited amount of equipment and supplies staged around to respond to emergency conditions. What company of this size would not do that?
Besides with so much older rail removed in this country, and so few spare rail routs, why not pay more attention to the few problems you have. And the hold up of shipments is money sitting that can't be made.

My comment was made partly in jest, David.
Looking at the latest pictures posted by Human909 it appears that the counterweight is passive.
That is, it does not move with the drawbridge as the span opens.
The counterweight position does not change as the drawbridge opens.
It appears that the span may be opened a crown and pinion gear set.
A more serious suggestion is that this may have been a corrosion issue.
I wonder if someone in the maintenance and inspection planning department has MBA training?


--------------------
Ohm's law
Not just a good idea;
It's the LAW!

Its actually here https://maps.app.goo.gl/qWZ1cYafWYLybgzf8

So 5 mles from international falls onthe Canadian side of the border, not rainy river.

Looks to me like it was left int he up position as a default to allow boat traffic under

We can only assume the connecting rods failed for some reason.

Not sure how it operated though - difficult to find any information it. You can / could see it fromt he highway accross the lake if you go into street view.

Looks like they are busy repairing it any way. Old structure already gone!

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Oops409. The pictures you have posted above are not the same bridge. It is a nearby bridge but not the same one.

Quote (waross)

The counterweight position does not change as the drawbridge opens.
It appears that the span may be opened a crown and pinion gear set.

Yes that was what I thought. I'm not familiar with too many types of raising bridges so I thought I may have been stating the obvious.

It does seem to be a method that is very demanding on the mechanics. It did well to survive the time is has done.

Quote (waross)

We can only assume the connecting rods failed for some reason.

There doesn't seem to be any connecting rods to fail.

Casey Jones video explaining some of the misinformation posted......... and more photos and info

https://www.youtube.com/watch?v=hdhNePSeqRo

I think it works like this:

I don't think anything moves in the counter weight.

It's pretty clear that the "normal" position of the bridge is open. Every shot you see has it open. So this is where everything is in a "neutral" position.

SO I think the thing is just very well balanced and rotates around its hinges with minimal hydraulic effort - so a bit like Tower Bridge in London and a lot of these bascule bridges.

If a train went over it apparently according to old Casey just before they heard a bang it sounds more like this was on the opening procedure and the counter weight mechanism just seems to have completely broken free of the bridge structure and fallen backwards with the bridge at maybe part height then falling and collapsing.

So corrosion hidden in the structure seems a more likely explanation. The thing looks to b similar to the 1908 bridge so over 100 year sold, next to a lake and where it gets pretty cold in winter and hot in summer.

If this is the same bridge "5 mile bridge being 5 miles from Fort Frances, then it had problems with the bearings 5 years ago.

Note a lot of sites call it the rainy river rail bridge. It isn't, its the 5 mile bridge. The rainy river one is in fort frances itself and is quite different.

https://fftimes.com/news/district-news/five-mile-b...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

In the London Bridge, the weight rotates with the bridge. But that requires room underneath for the weight to go. This is too close to the water for a pit large enough.

Note the tilt of the upper counterweight support matches the tilt of the bridge and that the lower counterweight support isn't one that would be stable; it's on a V with the point down.

It's a bit confusing because the house is behind it and I've found a GE image of the bridge down and fair enough, the counterweight thing doesn't move with the bridge, but does seem to go up and down.

If you go to GM and streetview on the bridge and then different dates and then move along the bridge you see it move and the weight goes up. So maybe some sort of hydraulic lift or just a mechanical connection resting on the end of the semi circular end of the bridge.

Actually int he side view you can see the struts connecting the weight to the bridge and the weight to the lattice structure. So it hods it vertically as it moves.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Here's my forensic Lego model and analysis.



Failure of the pin on each side here


results in exactly the failed condition shown in the photos.

I am in agreement with 3DDave and OOps409.
This is a parallelogram mechanism.
The vertical structure (gantry?) does not move, It is the one rigid side of the parallelogram.
It is possible to design it with no lateral loading other than wind loading.
While it may not be designed for completely zero lateral loading the loading will be very low.
Given a combination of serious corrosion and a powerful wind storm, it may have deflected.
Once it deflects, the lateral loading will increase and it may proceed to destruction.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

A report indicated it was operated to allow a train to pass just before the collapse. Either it was fully up or was on the way up.

Here is the complete paragraph to go with the drawing above, from a Wisconsin Structural Inspection Manual for Chapter 3.2 Bascule Bridges. Link to pdf uploaded

"3.2.4 Articulated Counterweight Bascules – Strauss Type
Medium and long span simple trunnion deck bascules with the roadway profile close to the high-water level generally require bascule piers with counterweight pits. The walls of the bascule piers keep the water from entering the pits and so provide dry spaces for the counterweights as the bascule leaves rotate open. This is also the case for rolling lift deck bascules, but to a lesser extent. Counterweight pits can be avoided if the counterweights are located remote from the bascule girder, usually overhead.
Trunnion bascules may be balanced by counterweights that are not fixed to the bascule truss or girders. This concept is old, but U.S. Patents for bascules with remote counterweights were granted to Joseph B. Strauss around the year 1900 and bridges with his specific counterweight arrangements are often called Strauss bascules. Three common versions are shown in Figure 3.2.4-1. The distinguishing feature of all Strauss bascules is the parallelogram connecting the counterweight to the movable leaf. In all Strauss bascules the motion of the leaf is by rotation about a fixed horizontal axis.
In some situations, it is not the depth of the required pit but the front-to-back dimension of the pit that needs to be minimized, as in Figure 3.2.4-1 (a). In such cases articulating the counterweight by suspending it from the leaf may be advantageous. In this section, three of the articulated counterweight trunnion bascules promoted by Strauss will be described."

• https://files.engineering.com/getfile.aspx?folder=e908db04-a45d-416c-bdd0-5

I can't see the details of the bottom connection of the towers, but the location of the failure seems to be at the bridge pivot point.

A theory:
The bridge was lowered to allow a train to pass.
That would have put the counterweight in the upper position.
The bearings on the gantry end of the links seized or jammed.
As the bridge was raised, the counterweight lowered and the jammed bearings imparted a torque to the gantry structure that caused it to topple towards and on top of the counterweight.
Root cause:
Corrosion damage to both the gantry and to the bearings at the top of the gantry.
(Is gantry the correct term?)

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

This picture shows the failure to be above the pivot point. (maybe)



--------------------
Ohm's law
Not just a good idea;
It's the LAW!

This is why I love this site - far better explanation than any of the news sites. Great find for the diagrams etc. Amazing really that this design is still functioning 115 years later.

Maybe you didn't click on my link a few posts above, but it seems 5 years ago the bridge was stuck in the down position for over 6 months. This is the interesting bit.

"When constructions workers returned to the Five-Mile bridge in the spring, Boivin decided to ask what the issue was and when it would be fixed.
“They said that during their maintenance and upgrades to the bridge, they discovered that some very important bearings down in the bottom had failed and were not passing the test,” stated Boivin.

“They said the bridge would not be open until July 5.”

I think if any of the main joints, be it the main bridge joint or the ones connecting the counter weight to the bridge or the vertical structure seized or was creating a lot more friction, the torsional effects would just bend this thing apart. That vertical metal structure looks relatively weak compared to the weight of the counterweight so any serious forces would cause it it bend and then it just gets worse until it falls over.

It might not help that it looks like it was designed for a double track lift, but they lifted the rails and some of the sleepers on one side. Might be small, but for something as delicately balanced as this might make a difference long term?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

I lean towards the thin vertical leg supporting counterweight to the lever arm in the forefront of waross's picture failed. Thus causing the loss of support and torsion loads on gantry. Notice forefront side is collapsed lower than the other side. Thus also loss of counterweight to hold bridge up. Or pin connection failure at lever arm connection.

LittleInch, no fault of yours; when I went to that link I got a nearly full page "please subscribe" and skipped it. Thanks for the text extract. Quite informative.

I think being single track/other changes would only affect the load on the gear teeth and the attachment to the bridge; also a horizontal load on the main bridge bearings reacting the extra load on the gear.

I wonder if it was remotely operated or if someone sits in the little house all day awaiting a train and what they are doing with their time now.

Prob next to impossible to determine cause from effect here.

They might not even bother to find out...

I've seen somewhere on my travels that this line only has a couple of trains a day and is now automatic. Back int he day that little cabin was probably manned alright.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

I think the contributors on this site have created lots of content for the youtubers to create more revenue!

Great Work based upon very limited information!

Any bets that bridge has been making noises it should have been for a while and that someone working there would have heard them? That they would have heard the motor make a little groan if a bearing got stuck that usually would only happen if the bridge was unbalanced by a heavy ice load? That it would have been stopped?

Slight Correction. Foreground pin connection from counter weight to lever arm (leaf) is still attached in picture. While the background vertical appears to be broken away and separated from the lever (leaf) arm.

Second pictures shows bent vertical support for counterweight, due to loss of connection on other side.

https://www.youtube.com/watch?v=ri82m-o_Fys





https://res.cloudinary.com/engineering-com/image/upload/v1725034393/tips/patent_dkdqyb.pdf

The picture shows no contradiction to my theory.
The gantry and the top links serve to keep the counterweight CoG vertical over the lower pivot.
There is little side force on the gantry save for wind loading and possible slight construction errors.
Anyone who has ever held a heavy appliance balanced upon one bottom edge knows how little force it takes to keep the appliance balanced and how quickly the force needed multiplies as the object tilts away from the balance point.
When a link arm bearing seizes up on top of the gantry, the weight of the descending counterweight put an unanticipated force on the gantry.
Once the gantry deflected, the CoG of the counterweight shifted away from the vertical with respect to the lower pivot support.
That shift increased the unanticipated horizontal forces on the gantry and complete collapse was almost immediate.
The bottom of the counterweight support rests and is pivoted to the back end of the bridge deck extension.
As the gantry collapsed and the counterweight fell over, the counterweight supports would simply pivot without experiencing any great destructive force.

A suggestion MintJulep
A modification to your forensic model:
Weaken the white vertical members at or above the lower pivots to simulate corrosion deterioration.
I suggest cutting most of the way through the members with a hacksaw or something similar.
The model should still be able to function to simulate raising and lowering.
Now fix or pin the upper pivots on the white vertical members to simulate seized bearings and prevent rotation.
Do this with the deck in the down position.

Now, with the upper pivots unable to pivot as intended and the lower part of the white verticals severely weakened attempt to raise the deck.
I anticipate that we will see a failure similar to the actual event.
It may be interesting to actually load the arms simulating the counterweight with some weights.
The CoG of the added weight must be directly vertically above the lower pivots.
I anticipate that the model will function well with considerable weight even with the vertical supports compromised but that the model will collapse when the upper pivots are restrained.
Thank you for your consideration of this suggestion.


--------------------
Ohm's law
Not just a good idea;
It's the LAW!

From my research, it appears all the pivot points are likely sleeve bushing type bearings, which tend to loosen as gap or tolerance grows with wear. That said, uneven wear would perhaps created inexpected lateral forces.

Oops, consider the effect of expansion due to oxidation of the steel or iron behind the sleeve bushings pinching and seizing the bushings.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

waross, now you got my attention!!!

Excellent point, as bushing wore out and thins out, rust layers behind bushing could deform it and cause binding or differential friction coefficients between sides.

That earns a star!

3DD, you need to look at pdf attachment in my post above. It is Joseph Strauss's patent on trunnion bascule. Note counter weight concept is similar to your orginal counter weight concept!

Quote (Oops409 (Mechanical) 30 Aug 24 02:33 https://files.engineering.com/getfile.aspx?folder=...)

Lines B-D and C-E need not be vertical ...

The vertical tower structure will always be loaded in a skewed system. Couple that with acceleration forces during operation (the counterweight is alternately drawn towards or pushed away from the tower structure) and countless storm wind forces over the years, the tower was just plumb tuckered out.





The vertical counterweight support somehow sheared or tore above the bearing connection while the stub flopped down to its final resting position.

Great image finds that show the vertical counter weight separation, or break from the pivoting connection

Symple, I agree that the two parallel verticals B-D and C-E will not always be vertical due to arc of leaf they are attached to, and I agree to that over 100 years of use might be beyond useful life without good maintenance and repairs over the years of everything.

Great image finds that show the vertical counter weight separation, or break from the pivoting connection.

I also agree the vertical legs from the leaf to the counter weight are always under constant load.

So that leaves a clear question as to cause and effect. Did the break of the vertical counter weight steel column break first at leaf connector, or did the vertical gantry fatigue first and start the lateral movement that caused the break in the counter weight vertical?

As I pundicated, the gantry buckled first. This creates a runaway system where the counter weight leans away from and looses its influence over the span which then offloads more energy into the system. It would have been quite spectacular to be in the control house.

Quote:

Lines B-D and C-E need not be vertical

True, but why would it be designed to be not vertical? Vertical will be the simplest design solution unless there are extraordinary factors cause the need for not vertical.

Quote (Oops409)

I agree that the two parallel verticals B-D and C-E will not always be vertical due to arc of leaf they are attached to,

This is not correct.

B-C and D-E are equal length and parallel. Thus, B-D and C-E will also always be parallel. If they were designed vertical (and certainly they were), then they will always remain vertical. (until of course the bridge falls down).

The break at pivot C is a good find. If that was the initial failure point then I can see how it would introduce some twisting around the vertical axis of the gantry structure because that side of the linkage is no longer constrained to being parallel.

@waross
Your suggestions are interesting, but I'll not be cutting my Legos - I have too few good Technic pieces as it is.
"Weaken the white vertical members at or above the lower pivots to simulate corrosion deterioration." My initial experiment effectively did this anyways. It introduced a hinge at that point, where by design a hinge is not desirable.
I think the ratio between friction forces and gravity forces of Lego vs the actual bridge are very different, so the remainder of your suggested experiments aren't likely to scale correctly.

Mint, Notice the center line of the trunnion axle that the bridge pivots on is not in same vertical plane as the upper hinge point at gantry.

Then the length of upper control arm does not appear to be same length as lower control arm (leaf of bridge) nor are those control arms parallel to each other.

Thus the counterweight is not moving up and down in a perfectly parallel plane with gantry

A wealth of images of Strauss Bridges.

https://industrialscenery.blogspot.com/2020/07/str...

Lego with failed pivot C. You can see what it potentiality allows the towers to do.

https://res.cloudinary.com/engineering-com/video/upload/v1725070748/tips/PXL_20240831_021623932_tn0pgt.mp4

Excellent illustration Mint. I agree that any mechanical issues could stress the towers and lead to the collapse. That includes abnormal accelerations/decelerations.

Mint, I think your illustraion with a failed pivot C shows how the vertical support for the counterweight on the other side got bent before/during collapse.

I still lean towards failed connection at pivot C causing load transfers to rest of structure. The gantry legs bent like malleable steel that was overloaded laterally.

Quote (Oops409)

Mint, Notice the center line of the trunnion axle that the bridge pivots on is not in same vertical plane as the upper hinge point at gantry.

I had to think about this for awhile.
MintJulep's link to "A wealth of images of Strauss Bridges." helped a lot.
What should be done to minimize or eliminate unwanted horizontal forces is;
The CoG of the counterweight should be directly above the supporting pivot.
The links should be of equal length so that the CoG of the counterweight remains be directly above the supporting pivot throughout its travel.
It is not required that the pivots be in a direct vertical line as long as the mechanism is a true parallelogram, there need be no designed horizontal forces throughout the travel.

As far as the broken counterweight support, That would be only in compression during normal operation.
(The vertical counterweight support somehow sheared or tore above the bearing connection while the stub flopped down to its final resting position.)
The break had to be subsequent, not causative.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

Yes vertical supports are all ways in compression, but the horizontal vector forces are changing with movement of contol arms. I think when counterweight leg fractured at C, it walked off end of leaf connection.

Can't tell clearly from picture, but it appears it may have been bolt failures that attached 'stub' to bearing connection assembly.

Edit: I am surprised no one has mentioned Strauss was Chief Engineer for the Golden Gate Bridge, later in his career.

The bridge apparently had bearing work conducted during Winter and Spring 2019.



Finally reopened at end of July 2019.



I'm wondering if the counter weight has been beefed up behind its vertical support. Would this not place additional strain on the support columns and pivot tower? To my eye, the pivot tower looks out of plumb in many images.

The main bearing looks like it blew a gasket. Is that cause or effect? Also of note, the south vertical support appears broken just above the bearing connection similar to the north vertical. I suspect there was a transition that was vulnerable to the massive counter weight moving horizontally as it slammed down on the tracks.

The vertical support brace was damaged from contact with the pit wall (where the inspector is standing). This indicates that the bridge was fully lifted when the counter weight/pivot tower fell over. Similar damage is visible on the north side as well.



We can also see upgraded structural steel, likely from the 2019 work. CN didn't get much bang for their bucks, ... or maybe more than they bargained for?



Original steel frame work. This image also shows the proximity of the damaged brace to the pit wall when the bridge is in the lifted position.

Symple,

I just gotta ask what search engine or search tools are you using to find these deeply buried images? Either you are using some great open source tools or created your own, or you have an insider source for these images..... The Censored and Monetized Commercial Search Engine's today, only find information that puts CASH in their pockets!

I would like acknowledge that littleinch found article on bearing work in 2019 and posted above, and 3DDave mentioned V-configuration in an earlier post......

I need more time to fully digest your new images, but on surface I find things my eyes indicate may be flaws in some of your conclusions?

I will have to mark up your images to show some different conclusions. That said, I fully agree the gantry columns are a very weak point in original design, right above the pivot point. Clearly if the massive counterweight changes in any way to put more horizontal stress right above the pivot point, then it is just a matter of time before that joint fails. Warcross's example of balancing an washer or dryer on edge is a great example of any slight changes to balance of counter weight really increases stress and fatigue on the gantry columns right above the pivot joint, and father time does a number on that joint too.

I agree that their replacement of the original taper truss horizontal brace with what looks like a beefier I-shaped beam, indicates that had either movement issues around pivot, or that tapered truss had decayed. And strengthening or arresting movement at that point, adds stress to gantry right above pivot point. Some of the pictures show the bridge in a very rusted and low maintained state, which may be the state it was in most of it's over 100 years of use.

As far as the grease below the apparent sleeve bearings, I think that is normal. They should be changing out grease over time by pumping in more fresh grease into zerk type fitting with the old grease been expelled out the ends of the sleeves, and some will be expelled during use. I would argue the sheet metal cover over the outside sides of pivot bearing are solely there to hide the expected grease on the outer ends, where people could spot the grease and complain...... Of course if they were doing more and better maintenance, they would clean up the extruded grease periodically rather than just let it work it's way into the soil and water way.

Edit: littleinch earlier wondered what someone did when it had a man in the shack before automation. I would argue it should have been that mans job to constantly maintain the grease joints, and inspect for damages or decay. Especially if there is only 1-2 trains per day passing.

Google (various search terms dancing around the subject matter), Facebook, reddit, twitter, and most important GIMP, which I've leveraged to draw out details in very dark shadows with screen layering and scaling.

Ok, by zooming in on Symple's images, I see lots of rivets in the sheet or plate steel cover that appears to have been added over the trunnion ends. Mostly around pivot point and upward. So this plate steel mend cover may have been added to strengthen a weakened and failing area at pivot joint, or was part of original design to add strength to this critical area. So unlikely to hide extruded grease.

The hole is probably there to allow inspection of new grease flowing out end of bushing, to ensure fresh grease inside sleeve bearing pockets. Only way to know if full of grease, it that fresh grease coming out ends.

This is common approach used on trailer axle hub bearings. Trailers have dust caps. If bridge is up, when it has s sudden failure, the fall rate and rotational speed is far greater on pivot point, thus significantly more heating of joint than normal slow controlled and balanced rotation. Thus more grease expelled.

I still believe that a top bearing seized and the descending counterweight pulled the gantry structure down. The rest is subsequent damage.
If the counterweight supports had not bent and broken they would have supported one end of the counterweight off of the ground.
While they could support the vertical counterweight in compression, when the counterweight fell to the ground the supports bent and then broke.
But, hey, what I think doesn't matter and I've been wrong before.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

Sym P.le;
Are you able to find any photos of the top links and bearings?

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

I am certainly NOT an Expert at anything, but I know enough to get into trouble.

A short tale to illustrate: I recently gave testimony during litigation, and pointed out a flaw in attorney's documentation he asked my opinion on. My response got that exhibit thrown out by judge..... So the attorney, after thinking for a while, asked me if I was an Expert in that area, which is him baiting the hook for me....... I did not take the bait, and agreed I was not an expert.... Ah the fun and games of court....... and our so called legal process.....

Ok in first image, if we can agree red arrow points to bottom of concrete counter weight, and blue arrows point to pit wall, then the only way I see the vertical contacted pit wall, was when the bridge was in the full upright orientation, such that the vertical could contact pit wall as counterweight fell or lost balance while bridge is in up position.

Problem is the bend in the counterweight support leg, is in the opposite direction from what you would expect. However, the damage to the diagonal is the bend damaged by pit wall. I will add new yellow arrow to the diagonal damaged by pit wall. Clearing the bending indicates the counterweight was moving towards the gantry to cause that bend.... If that is case, that indicates perhaps the other vertical has lost it's foot, thus the bent vertical bends under load re-distribution before falling backwards. And yes the bent vertical would contact the inside of the pit wall on the way up and down.

If that is the case, that confirms Symple's conclusion that bridge was in upright condition, perhaps in static condition before failure. The damage counter weight joint at leaf circled in 2nd image below, looks like partial failure also and/or signs of decay. Symple's second image also shows concrete damage on inside of pit wall.

I have another image I will offer later from other side, that agrees with argument above...

Camera image may be playing tricks in this image, as I added some lines. One is parallel to what I think is back of gantry, and the other back of counter weight....

I have hard time explaining damage to building and ground below the stub end, if stub was fully attached when counterweight fell, and the sheared bolts happened after counter weight hit ground. My take is leg stub disconnected from bearing hub assembly while in air and slid off leaf into building doing ground and lower damage, prior to ending up horizontal and damaging building wall from ground up to resting point.

Notice also that the disconnected stub moved laterally into building, which also caused twisting of of the counterweight pivot connection on the other side.

Edit: Waross, I don't discount that there could have been issues at top of counterweight, as that is a hard area to inspect, and one that could go unnoticed for long periods of time. But I clearly think stub separation caused the twisting and over loading of the other counterweight upright.

Consider:
The bridge had been lowered.
The bearing at the top of the gantry may have done some damage to the gantry due to a jammed bearing, but not enough to bring the structure down.
As the bridge was then raised, the descending counterweight working through the frozen bearing, pulled the gantry over.
As the gantry fell over, the counterweight fell backwards.
The upper edge of the counterweight struck the ground first as the support members pivoted on the lower bearings.
The geometry was such that had the lower supports remained straigt and unbroken, the lower edge of the counerweight would have been supported above the ground.
However the counterweight supports did not have the strength in that position to support one edge of the counterweight above the ground.
As a result, the supports bent and then broke.
This would explain the alignment of the counterweight.
Had one lower pivot jammed and the support broken as shown and caused the counterweight to fall, there is a very high probability that the wreckage would be skewed to one side.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

waross, I do follow your logic, and I see that an upper issue could have been the initial domino, and any binding or loss of connection at one side of upper CW attachment could cause the shearing of bolts from stub to bearing connector assy. It appears in pictures that the bottom edge of CW on the side with bent lower support leg is touching ground, but the other bottom side is not. It also appears concrete CW contacted train rails, as you can see impact zone, then slid away from bridge pivot points, on building side, after it was on the ground.

But, I keep coming back to the image above and can't explain building damage below CW lower support stub. I do believe counterweight is skewed laterally into building and it has fallen further away from from counterweight lower pivot point. I can't tell for sure, but it appears the lower CW support leg may be bent inward towards the other support leg? Also the side away from building has collapsed lower to ground than building side.

The stub does not appear to have contacted pit wall on building side, like the other side, which indicates perhaps a lot of torsional and lateral movement in the gantry and CW assembly before splash down or on the way down.

It appears CW had to maintain some sort of connection at the top to the gantry to be able to pull the gantry down. Perhaps one side failed and the other side did not? I do feel the damage is skewed towards building and away from pivot point on building side.

I give up, I have exhausted all the brain cells I have, and I still cannot see any conclusive answers from the available data set....

Perhaps the cause is similar to this shipping incident.

https://www.youtube.com/watch?v=3m5qxZm_JqM

Quote (Oops409 (Mechanical) 1 Sep 24 18:34)

Arrow points to upper gantry bearing which detached and fell down. The proximal green (greyish) parts are from the upper gantry arm.

Looking at 409s picture above this double track bridge could never have operated that way looking at the single track causeway. So 115 years of slightly off balanced load had an effect?

The balance forces are very tightly matched to achieve low opening and closing on the rack and pinion so did this slight difference in lift force one side to the other just finally catch up with it?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Thanks for those final missing pieces that I could not see in images! Binding/failure and detachment of upper gantry control arm and asymetric loading fills in the missing pieces. Also notice all the scrap rails that spilled out of counterweight on building side of CW that was added to counter bslance heavier track side.

In the end, a tired old structure that got away on them. A feint blip in the budget of CN Rail, they just shrug and barely consider their luck in not maiming anyone or worse. To the locals, an insult, interruption and inconvenience to their quiet and semi-remote way of life.