Rumbling Red Rockets

Today I received a note that mentioned the rumbling of Toronto’s streetcars and the damage they have done to the streets and, possibly, surrounding buildings.  I thought it worthwhile to talk a little about track engineering and the huge improvements made by the TTC in the past decade.

Although the TTC decided to keep the streetcar system back in 1972, the message really didn’t sink through to management.  Track construction techniques kept getting worse and worse.  Where once we had continuously welded rail sitting on properly treated wooden ties, we eventually came to unwelded rail dumped on plain wooden ties and poured in concrete.  The result was quite predictable.

The lack of welds made for lots of small vibration points at the joints, the concrete at these joints (and along the side of the rails) flaked and broke from constant vibration, while under everything the wooden ties rotted leaving no support for the concrete.

The original wheel sets on the CLRV streetcars couldn’t have been better designed to destroy the trackbed if someone had deliberately set out to accelerate the demise of the streetcar system.  [Note to the paranoid among us:  It wasn’t a plot.  Sheer stupidity wins out over Machiavellian intrigue every time.]

The original wheels on the CLRVs were designed with a solid hub surrounded by a hard rubber ring.  The actual steel tire mounted over top of that ring.  This arrangement puts the rubber in compression and is rather stiff.  It is designed for use on a trackbed that is somewhat resilient.  Track paved into solid concrete is anything but resilient.

As if that wasn’t enough, the natural frequency of vibration of the wheel/rail interface was exactly the ideal frequency (a) to set the whole street in motion as a gigantic amplifier and (b) to hit the nerves of people who seem particularly attuned to this type of vibration.  (The value is around 70 Hz.)  So we have the street acting like a big guitar box, the concrete dancing around and disintegrating, and all at a frequency that could be felt by people a long, long way from the source of the noise.

There was a nice technical study on all of this, and part of the fix was rather simple:  change the wheels.  The ones that are now on the CLRVs (and were on the ALRVs from the outset) have the rubber mounted on the face up the axle hub, rather than around it, and the tire mounted onto that.  The rubber is in shear, and the whole assembly vibrates at about 120 Hz, far too fast for the roadway to keep up.  The vibration energy is radiated into the air rather than breaking up the pavement.

Astute readers will recognize this as a variation on the PCC wheel, a technology that had been around since the 1930s and was developed for more or less the same reason.

The next step was a redesign of the trackbed.  This took a lot longer, and it wasn’t until the late 1990s when we started to see welded rails again and robust tie and suspension systems.  Rubber casings around the rails damp the vibration and make a good seal between the track and concrete.  Steel ties will never rot.

The track built with this new design now accounts for more than half of the system, and it will last 25 years easily except at locations like carstops and tight curves where there is a lot of wear.  The locations still to be rebuilt are:

  • St. Clair (to be done with the LRT right-of-way project)
  • Bathurst (and Vaughan) from St. Clair to Bathurst Station (not heavily used)
  • Fleet Street (to be done in 2006)
  • Gerrard from east of Coxwell to west of Main (soon to be underway)
  • Dundas (almost all of it, scheduled for 2007)
  • Lake Shore (the outer end of the line scheduled for this year)
  • Parliament from Gerrard to King (in really rough shape in spots)
  • Various other bits of track in the downtown area that don’t see much service

This means that by the end of 2007 (or 2008 if budget cuts force delays in the projects), we will have a completely rebuilt streetcar track infrastructure.  Oh yes, I almost forgot.  The original Harbourfront line including Spadina south of King was not built with resilient track.  That’s why it is quite noisy.  It will be up for replacement sometime between 2010 and 2015 depending on how it holds out.

Intersections are also getting the new treatment.  All of the castings are wrapped in a thick plastic which is forced tight against the rails by the poured concrete.  This reduces a lot of the vibration that breaks up these complex layouts, although it doesn’t totally eliminate problems.  King and Dufferin was the test site a few years ago, and this arrangement is now standard for TTC special work.  The best chance to see this in 2006 will be the reconstruction of the Hillcrest Yard entrance tracks.

Finally, there is the question of the weight of the CLRVs.  They are tanks and there’s not much we can do about that until we get new, lightweight cars.  (Remember the PCCs?)  In case you are wondering, the heavy CLRVs were brought to you by the same bright sparks who gave us the Scarborough RT.  You see, they thought that a high-speed suburban streetcar needed to go really fast.  Really, really fast.  70 mph (about 112 kmh).  For that sort of speed, you need heavy, stable trucks (the undercarriage).  This added about 25 percent to the weight of the cars for no good technical reason at all.

I rode a train of CLRVs on the Riverside Line in Boston at 50 mph (their track isn’t up to anything faster) when they were demonstrated for the MBTA many years ago.  The basic design problem is that unless your stations are very far apart, there is not much point in accelerating to more than about 50 mph because you have to stop again.  Even the subway rarely gets above 45 mph.  [Sorry about all these miles, but I’m sure you will cope.]

And so we have:

  • Cars that are heavier than they needed to be
  • Wheels that were engineered to be as noisy as possible
  • Track that was a case study in shoddy design

No wonder the streets fell apart, and people asked why the streetcars were destroying our city.  The TTC track maintenance budget went through the roof thanks to these problems, and streetcar lines are beset with slow orders while they await rebuilding.

The TTC has learned its lesson, but we’ve lost so much in the process.  In one more way the TTC sent out the message that streetcars, and with them LRT, were not the way we should expand the network.  The subway junkies couldn’t have done better if they tried.