News   Nov 22, 2024
National Housing Day: City of Toronto and Riverdale Co-op Open Affordable Homes
 617     1 
News   Nov 22, 2024
Bauhaus Condos Rising in Toronto's Old Town
 1.1K     5 
News   Nov 22, 2024
49-Storey Waterfront Tower Proposed at Spadina and Queens Quay
 2.9K     8 

VIA Rail

I seem to recall YDS saying at one point that the cost-of-energy analysis was a wash for HFR, it was the carbon reduction that tipped the balance towards electric. Even that informed opinion might need to be revisited and the key data refreshed - is the cost per kg of carbon reduction competitive with what one could accomplish for the same investment in some other sectorj?

We armchair observers don’t have the data to prove one way or the other.....but.....There is certainly some intuitive logic to @kEiThZ ’s observation that a once-hourly passenger train line is not enough energy consumption to extract gains by switching modes - especially given a billion dollar pricetag.

- Paul
 
Last edited:
reaperexpress said:
I absolutely agree that's a totally respectable achievement. My thinking is not that we need to aim higher than that, it's that we really need to actually achieve that travel time in day to day operation, even if it means increasing the (rather optimistic) project budget. I have a bad feeling that we'll end up with a line which could theoretically be traversed in 3h15 if there were no other trains on the line, but in practice takes closer to 4 hours given schedule padding required to account for meets. To routinely schedule 3h15 travel times on a single-tracked line, the best case scenario needs to be under 3 hours.
Click to expand...

This.

An hourly service plan equates to a meet every 30 minutes en route. One train takes a minimum 5-minute hit for each of those meets, as it must take siding and await the other train, clearing early enough to not slow the other train. On a 3:15 schedule, that would mean each train needs 15 minutes of wait time just for meets.

Now assume a regulatory-imposed speed restriction in each of Peterboro, Tweed, Sharbot Lake, and Perth. Assume two minutes lost time in each of those. That's 8 minutes.

Now assume that due to regulatory issues there is no improvement in time over today between Ottawa and Fallowfield. And assume that VIA can go no faster than GO between Agincourt and Union (assuming the Scarboro Jct route is used, and not the Don route). And, be pessimistic about the speed through the junctions between VIA and CP at Smiths Falls, and between VIA and CP/GO at Agincourt (the concern about a lowest-cost design). Those segments eat up minutes.

I'm confident that there will be substantial speed achieved in between, even east of Tweed, but there are a lot of "minute eaters" in the plan.

Hopefully the JPO reports will put this concern to rest.

- Paul
 
crs1026 said:
There is certainly some intuitive logic to @kEiThZ ’s observation that a once-hourly passenger train line is not enough energy consumption to extract gains by switching modes - especially given a billion dollar pricetag.
Click to expand...

I would look at it as opportunity cost. Rail is bloody efficient as is. Electrification isn't going to save substantially on fuel cost and the associated emissions. Especially, compared to what the $2B penciled in for electrification could do for upgrades that yield higher ridership.

I'm happy to wait for a study. But intuitively, I don't think there's a case for electrification until they get into Higher Speed Rail/HSR territory with higher frequencies as well.
 
crs1026 said:
An hourly service plan equates to a meet every 30 minutes en route. One train takes a minimum 5-minute hit for each of those meets, as it must take siding and await the other train, clearing early enough to not slow the other train. On a 3:15 schedule, that would mean each train needs 15 minutes of wait time just for meets.
Click to expand...

You are ignoring the option of long sidings. They become less practical with higher speeds, but not impossible. The idea is to make the siding long enough that the trains can pass each other without slowing down. After the train passes the switch, it is thrown in the opposite direction for the oncoming train. To do this, you need to make the siding long enough for:
  1. time to throw the switch after the oncoming train has passed it, plus
  2. time to stop the train if the switch fails to throw, plus
  3. a reasonable amount of time for the oncoming train to be late.
There is also the time for the train to pass the switch, but given that the trains are likely only about 200m long, that is insignificant.

Using some back of the napkin math, you need about 1 minute to throw the switch, which at 177 km/h is 3km, maybe another 3 km for an emergency stop at 177 km/h, and assuming a margin of 5 minutes late for the oncoming train, another 15 km. That is a total of 21km. Given that at 177km/h a train can travel 88.5 km in 30 minutes, that is less than 1/4 of the ROW being long sidings, which is still a lot cheaper than double track and most of the time neither train needs to slow down.
 
roger1818 said:
You are ignoring the option of long sidings. They become less practical with higher speeds, but not impossible. The idea is to make the siding long enough that the trains can pass each other without slowing down. After the train passes the switch, it is thrown in the opposite direction for the oncoming train. To do this, you need to make the siding long enough for:
  1. time to throw the switch after the oncoming train has passed it, plus
  2. time to stop the train if the switch fails to throw, plus
  3. a reasonable amount of time for the oncoming train to be late.
There is also the time for the train to pass the switch, but given that the trains are likely only about 200m long, that is insignificant.

Using some back of the napkin math, you need about 1 minute to throw the switch, which at 177 km/h is 3km, maybe another 3 km for an emergency stop at 177 km/h, and assuming a margin of 5 minutes late for the oncoming train, another 15 km. That is a total of 21km. Given that at 177km/h a train can travel 88.5 km in 30 minutes, that is less than 1/4 of the ROW being long sidings, which is still a lot cheaper than double track and most of the time neither train needs to slow down.
Click to expand...

I'm assuming the precedent of the Ottawa-Brockville line, ie sidings about a half mile in length. And sidings placed about where they were in steam days, to save new grading of the roadbed. At that siding length, turnouts will not be high speed because there isn't any point in them when the train has little room remaining to brake to a safe stop after entering the siding.

The limiting event is that the train taking siding must clear the main line 30 seconds before the train holding the main reaches the "approach signal" (in fixed block mentality). The 30 seconds is the time required to throw the switch and clear the signal. Only then will the train on the main be able to proceed maintaining full speed, ie without ever reaching a restrictive signal.

Before the limiting event, the train taking siding must slow at service application (not emergency application, some safety margin is required) from full speed to turnout speed, transit the near turnout (likely a 30 mph affair) and clear the interlocking.

The train in siding must then wait for the train holding main to transit the approach block, transit the far turnout, and then 30 seconds to reverse the turnout and clear the signal. Then the waiting train can resume its journey, accelerating from zero to full speed.

I agree, a far better approach would be longer sections of double track with high speed turnouts, so meets happen close to full speed for both trains. Those sidings would have to be several miles in length to give adequate contingency. I just don't anticipate that VIA's budget stretches that far.

- Paul
 
Last edited:
crs1026 said:
I'm assuming the precedent of the Ottawa-Brockville line, ie sidings about a half mile in length. And sidings placed about where they were in steam days, to save new grading of the roadbed. At that siding length, turnouts will not be high speed because there isn't any point in them when the train has little room remaining to brake to a safe stop after entering the siding.

The limiting event is that the train taking siding must clear the main line 30 seconds before the train holding the main reaches the "approach signal" (in fixed block mentality). The 30 seconds is the time required to throw the switch and clear the signal. Only then will the train on the main be able to proceed maintaining full speed, ie without ever reaching a restrictive signal.

Before the limiting event, the train taking siding must slow at service application (not emergency application, some safety margin is required) from full speed to turnout speed, transit the near turnout (likely a 30 mph affair) and clear the interlocking.

The train in siding must then wait for the train holding main to transit the approach block, transit the far turnout, and then 30 seconds to reverse the turnout and clear the signal. Then the waiting train can resume its journey, accelerating from zero to full speed.

I agree, a far better approach would be longer sections of double track with high speed turnouts, so meets happen close to full speed for both trains. Those sidings would have to be several miles in length to give adequate contingency. I just don't anticipate that VIA's budget stretches that far.

- Paul
Click to expand...

I think Ottawa-Brockville is an excessively pessimistic precedent, because it was designed based on quite a different circumstance. Those sidings would have been placed with lower frequencies in mind (and thus less consequence of cheaping out on sidings), and only the vaguest form of schedule adherence (so frequency of sidings becomes more important than length).

I think a more suitable reference case would be the hourly off-peak GO train services that were introduced over the last couple years. These generally use sidings at least a couple kilometres long, with 45mph turnouts.
- The Barrie line started its 75-min service using the 3.3km siding north of Maple. Trains often got stopped waiting for each other and it added 2 minutes to the schedule (in addition to all the schedule padding that schedule already had).
- The current passing place for the Barrie line's 60-minute service is the 6.8km siding between Downsview Park and Rutherford. Trains meet around 100 km/h within the segment and the schedule is not adjusted for the meet (the normal schedule already assumed <45 mph at the locations of the switches).
- The Stouffville line uses the 2.1km siding between Milliken and Unionville. I don't ride that part of the line, but I would assume that it performs similarly to the Barrie Line siding north of Maple.
- The Kitchener line is fully double/triple tracked as far as Georgetown, but the hourly service operates a single-track pattern with a meet at Maple Station. The station stop serves to considerably increase the effective length (in time) of the passing track. But I don't see us double-tracking Peterborough station so that's not particularly relevant to HFR.

Given the above examples I personally think the base case scenario would be sidings a couple kilometres long with 45 mph turnouts in the locations meets are supposed to happen, and a smattering of Ottawa-Brockville style sidings in between (in case a train is off-schedule). But ideally the planned sidings would be more than just a couple kilometres long with turnouts suitable for more than 45 mph.
 
Last edited:
jelbana said:
How expensive is double track vs single track ? Is it mainly the grade-separations and/or the tighter areas that need a wider ROW that become more expensive? Or is it just the cost of laying down the track itself?
Click to expand...

The ROW may be wide enough to place two tracks, so in theory there is no added land cost.....but ..... the grading is the limiting factor. If the line has cuts or fills that were built for only one track, the earthmoving to widen the roadbed may be expensive. This is a factor east of Havelock where the line enters the Canadian Shield. Widening cuts means blasting rock. For this line, in many spots the line is draped onto the side of a slope or around a hillside - so the volume of material to be removed is greater. The other cost is fills through wetlands, which may require a lot of fill to stabilise all the way to the bottom of a marsh. And any bridges or underpasses that were built without future proofing for doubletracking. There are plenty of stretches where doubletracking would have a prohibitive cost, it’s a matter of how much double track is really needed and where it is most feasible.

And of course the cost of track materials and ballast doubles when there are two tracks over one.

- Paul
 
reaperexpress said:
Given the above examples I personally think the base case scenario would be sidings a couple kilometres long with 45 mph turnouts in the locations meets are supposed to happen, and a smattering of Ottawa-Brockville style sidings in between (in case a train is off-schedule). But ideally the planned sidings would be more than just a couple kilometres long with turnouts suitable for more than 45 mph.
Click to expand...

If this happens, I will be delighted.

But....baking those preferred passing locations in to the design strikes me as very risky, considering that CN controls timings at the east end and ML may control timings at the other end. One would need an ironclad agreement with CN. What if, for instance, CN abruptly declared that VIA can only have one track at any time east of Coteau? Or changed where in that territory it wanted VIA to hold its meets? That could shift the scheduled meeting points all the way to Toronto. VIA will probably tweak its schedules also with experience. It may start with some specific timing (eg always leave on the hour) but find some other timing is preferable. (On the hour ex Toronto but on the half hour ex Ottawa, as a hypothetical example)

That’s why I would argue for more fairly closely-spaced short sidings throughout the route. With a static service plan and perfect timekeeping, some of those sidings might rarely see a train.... but that gives maximum flexibility to alter the schedule, and the time penalty involved in any shifted meet is small. I had assumed VIA might invest in some extra software in their traffic control to predict actual times so that meets could be optimised.

- Paul
 
jelbana said:
Thanks! Would it be reasonable to think that the economic viability of double tracking in expensive spots will also likely go hand-in-hand with the viability of curve straightening? (my assumption here is that where the geography is likely to be sloped, is also where it is likely to have more curves?)
Click to expand...
The same variables are at play, yes. New construction may actually allow steeper grades and a straighter line is also shorter - less track to build and maintain. But rock is rock and swamp is swamp.

The real economic issue for straightening curves on this line is just the sheer number of them, and how close they are together. One doesn’t save much time (or generate ROI) by cherrypicking a few cheap/easy ones....one has to connect enough straight track to achieve a sustained speed improvement over a significant distance. Otherwise one spends millions to save seconds, and the business case isn’t there. I’d predict that the incremental improvements down the road will be packaged as $100-million + chunks of a dozen or more kms in length rather than smaller projects. The ROI only comes when you save increments of 5-10 minutes.

- Paul
 
jelbana said:
How expensive is double track vs single track ? Is it mainly the grade-separations and/or the tighter areas that need a wider ROW that become more expensive? Or is it just the cost of laying down the track itself?
Click to expand...

I wouldn't dismiss the cost of rail, ties, ballast, labour, and equipment to lay it. There is also the ongoing maintenance costs. The track needs to be regularly inspected to ensure it is within tolerance set for the class of track (the higher the class the more frequent the inspections) and if it is deficient, fix the issue(s). There is a reason why CP downgraded the Winchester Sub from double track to single track. Also, there is a reason CP has let the Havelock sub degrade to Class 1 (with a top speed of 10 mph for freight) even though that means it takes 12 hours (each way) to travel the approximately 140 km to Havelock. If you think about it, the 24 hour round trip is 3, 8 hour shifts. If you multiply that by 3 trips a week, you get 9 shifts a week. That is likely hundreds of thousands of dollars a year in labour alone (not to mention equipment utilization), much of which could be saved by maintaining it at a higher class. CP obviously knows that and the cost of maintaining it to a higher standard would be even higher. Imagine the cost of maintaining significantly more track to Class 6 instead.

This is why I beleive CP will have no issues selling the Havelock sub to VIA (though I have no doubt they will negotiate a hard bargain) because VIA will pay to upgrade it, allowing CP to do the trip in an hour or two instead (they would likely choose not to run at top speed), saving CP hundreds of thousands of dollars a year, while providing better service to their customers.
 
2transpo said:
www.highspeedrailcanada.com

VIA Rail Canada Fights to Keep Public from Seeing their Flawed $4 Billion High Frequency Rail Business Plan

VIA Rail Canada Fights to Keep Public from Seeing their Flawed $4 Billion High Frequency Rail Business Plan
www.highspeedrailcanada.com www.highspeedrailcanada.com

lol
Click to expand...

Also on this website: "HIGH SPEED RAIL CANADA is the only national educational resource on high speed rail, trains, past studies and current information."

The allegations about secrecy are not true. I found this in about 10 seconds:
2002 VIAFast Study

Fares change, stations change, and all of those details can be set and changed on a whim.

I agree that it's a step down from the 1970s, when the companies used their own rail (so less freight interference). The problem is that there's no public appetite for spending billions on HSR, as the article advocates for. If that plan came out, you would probably be at the front of the line to rally against it. This is the best way to test out a plan that allows for future upgrading, with VIA owning the entire line.
 
You must log in or register to post here.

Back
Top