Ottawa LRT

[Bytor]

What do we know about track geometry and speeds where the derailments happened? Or the wheel shape conical section specs?

Unsure of the track geometry but I heard this was a low speed derailment.

Derailments at low speeds are usually penultimately caused by two things.

The obvious one is when the vehicle itself lurches to the side, so the wheel bogie is lifted up enough that the flange on the wheel no longer prevents it from moving laterally.

The less obvious one is when the tracks are misaligned in some fashion that causes the flange to touch the rail in such a way that the friction from this new contact patch basically levers the wheel up past the flange and laterally out across the rail. The lighter the load on the wheel, the less friction is required for this to happen. Track shift around a curve can cause this by making the effective curve radius tighter than it should be and this increases friction. Or it can be caused by track twist that unloads a wheel and requires less friction to make it jump. A third cause is worn wheels with a reduced flange angle which increases the size of the flange's contact patch going around a curve which increases the friction. 

There's a lot of math and physics involved in this, beyond my ability to explain it more in depth than this, but if you are interested you can look up "Nadal's Criterion" (sometimes "Formula"), which is what this is. It's one of the things used to determine what the minimum curve radius for locomotives and cars of a given weight going at a specific speed. The related Wagner's Formula takes into account hunting oscillation at higher speeds.

Since the picture shows the whiles outside the track, we can rule out a gauge-widening misalignment, which would cause the wheel to fall of the rail to the inside.

So we're left with two likely ultimate causes.

The first is that the wheel sets are badly maintained and not being properly monitored for wear to the flange angle. At low speeds there shouldn't be any hunting oscillation and the flanges should never contact the rails on straight segments, but worn wheel sets will start to oscillate a lower speeds which will cause greater flanging friction around curves, but also potentially (but hopefully rarely) on straightaways. 

The second is bad track construction and or maintenance that has let the tracks go out of alignment resulting in higher frictions that violates Nadal's Criterion to keep the wheel from jumping. Unexpectedly high flanging forces can cause tracks to go out of alignment, but even with worn wheels that's AFAIK only at higher speeds that can do that, though tamping the rails can apparently reduce their ability resist lateral displacement.