Kelly writes,
I am also interested in cable propelled transit and am a big fan of The Gondola Project. Anyway, I just worked out some numbers on the Funitel at the resort where I work (Squaw Valley) and I thought I would share them with you.
Squaw Valley Stats
- 28 legal limit (This would be a crazy crush load)
- 18 normal mixed load – 9 seated and 9 standing (this is pretty typical, if there is a small line this is about how many people naturally pack into a car)
- 9 seated comfortably (kids sometimes squeeze 15 but adults who don’t know each other usually won’t sit more than 9)
- 1,296 pphpd seated comfortably
- 2,592 pphpd mixed seated and standing
- 4,032 pphpd official capacity (crush load/legal limit)
- Cable speed: 6 m/s (or 13.4 mph)
- Headway: 25 seconds / 150 meters (in practice they space the cabs out a little more than this.)
I did this because I wanted to compare Seattle’s new light rail to Squaw’s funitel. As it’s configured now, this is how are our light rail compares:
Seattle LRT Stats
- 1,184 pphpd seated
- 3,200 pphpd crush load
- Headway: 7.5 min
- Cost per mile: $179 million
Wish we would have built a CPT system!
Now, unfortunately Kelly didn’t provide us with how much the Squaw Valley system cost to build, without which makes comparison difficult (if anyone out there has that number, please post it in the comments below). But at 1.67 miles in length, I can assure everyone that the Funitel didn’t come anywhere near the cost of the Seattle LRT. The Squaw Valley system, likely, was built for somewhere around $25 million per mile, not much more (again, if anyone out there has the actual number, please post it with link in the comments below).
Furthermore, it’s not an entirely fair comparison. Seattle’s LRT has significant portions of the line underground and the line is far longer (17.3 miles) than what a CPT system typically sees. Additionally, Seattle’s prone to earthquakes and seismic activity. Any cable system built in Seattle would face increased costs in order to earthquake-proof the system. At the same time, some of those costs would be defrayed by cable’s ability to deal with Seattle’s topographical challenges in ways that LRT never could.
With those caveats aside, however, two things jump out: The dramatic difference in headway and the capacity premium of the Funitel. The LRT system has wait times 15 times longer and a (current) maximum capacity that is 20% lower than the Squaw Valley Funitel.
One would expect that if Seattle were to pay such a high cost for their LRT (as per my knowledge, it’s the most expensive LRT ever built), headways should be far shorter and capacity far higher. Especially when compared to a ski lift.
Comparisons such as Kelly’s are useful for de-bugging. It’s the kind of comparison that jolts you into questioning what you think you already know. As more and more research is gathered on the Squaw Valley, such comparisons could also be useful in more rigorous ways.
So here’s a question: What if The Gondola Project set up a central database of interesting, appropriate systems for people to use as comparisons? As systems are identified, they could be “put out there” and readers could contribute research and statistics as they find them. Would that be useful to TGP readers? Would TGP readers be willing to help build that database? How would it work?
Think about it and send us your comments.
29 Comments
Sure, database sounds good!
The comparison of capacity seems a tad unfair to me, for two reasons.
1) Headway: Trains can run much more frequent than 7.5 minutes. That headway is probably a function of the expected demand on the system. With greater demand, the headway would shrink, and capacity would increase. You could easily double the headway, doubling the capacity of the line.
2) Longer trains: According to Wikipedia, Seattle’s Central Link LRV’s can be coupled together into a 4 car train. From what I can tell, they currently operate 2 car trains. So this is another opportunity to double capacity (without increasing labor costs).
So what does this mean? Assuming 4 car trains and 3 minute headways:
5920 pphpd seated
16000 pphpd crush load
Headway is a little over 7 times longer than the funitel
Capacity almost 4 times the funitel.
What the system can actually carry when there is sufficient demand is much higher than current demand. It seems unfair to compare how many passengers it can carry now without considering what it can do (without any more investment in infrastructure) in the future.
Why isn’t this operated now? Because the demand isn’t there…. yet.
This is a line that will probably be the backbone of the city’s transit network in the future. In the future, the demand will most likely be there. This isn’t to say CPT is bad. It’s just to say that the cost is justifiable in the long-term. It is much cheaper to prepare for future demand now and build it before the system is operational, than to go back later and try to manage a construction project while keeping the line running while trains approach their capacity limits.
Jeff,
I both agree and disagree with what you’re saying. I agree in the sense that it’s good to build for the future and that the comparison isn’t entirely fair. The tunnels had to be dug and those will be there forever. That’s a good investment.
What I don’t agree with is your assumption of increasing offered capacity four-fold as a way to justify this system. Yes, the system could do that, but it’s not. You’re also not factoring in the additional cost of adding trains and increasing capacity. Trains are expensive and those extra vehicles would need to be purchased. They also need to be operated, stored, driven, insured and maintained.
Furthermore, it’s highly unlikely that an increase in capacity such as you envision would occur within the next 30 years. By then, the vehicles and tracks will have to be replaced anyways.
Lets say we took the existing crush load of 3,200 pphpd and then increased ridership by 2% (which is very generous) per year over a 30 year period. By year 30, that system is only going to require 5,796 pphpd – less than double what is currently needed and well within the current 6,000 pphpd limit of aerial cable systems.
Again, I understand and agree that the comparison isn’t totally fair and I say so explicitly within the post. Especially because we’re unlikely to see a 24km long cable line in the near future. There is, however, a massive disparity between the level of service characteristics of the two technologies given the price differential.
Good to know you think the d-base is a good idea.
I just found this thread, and I know a bit about Seattle’s LRT.
“it’s highly unlikely that an increase in capacity such as you envision would occur within the next 30 years” Probably more likely than you give it credit for. We’ll have our first extension built out to the University of Washington by 2016, then we’ll be able to run 4-car trains. “By 2030, the University Link line alone is projected to add 70,000 boardings a day to the light rail system.” By 2021 we’ll have our second extension both up north and down south. In 2023 we head east across the bridge and pick up a significant sized city and a major employer.
Even before any of these expansions, we’re up 10% over last year (ok, it’s a bit early to start using year/year numbers).
By the time the system’s built out, we’ll be near 2-minute headways with 4-car trains.
Don’t get me wrong – gondolas are great. And theoretically cheap. But their place is not the core of a regional transit system for a significantly large city.
One thing would be very interesting:
How long does the LRT take to cover the whole route?
How long would the gondola take using existing technologies and stations as the LRT does.
If the LRT is faster which distance do I need to travel so the LRT is always faster using travel time and average waiting time.
From far away it seems something is not rigt with the Seattle LRT. The price is very high but they still have grade crossings etc. Other cities build automated metros for much less cost. Vancouver is not that far and they have already two different automated metros Bombardier and Rotem in their city.
I can see some of your points Steven. And CPT would have distinct advantages at crossing natural features that are normally barriers to transportation in the Seattle area. It’d be neat to see them implement something!
I do like matthias’s question. How long would is take by CPT versus LRT on the same route? (Remember to factor in a wait time for LRT!)
I wouldn’t mind doing the calculation, but I wouldn’t know how to account for station creep times.
For your 3s calculation:
35 P á cabin
time between cabins inside the station: min. 30 sec up to max. 90 sec.
a complete station passage: depending on level of comfort 90 sec up to 150 sec. (around 120 is common)
moves with around 7.5 m/s
up to 27 km/h possible
I would think that if you were doing a side-by-side comparison of possible CPT and LRT systems on a given route, and considering possible future demand increases requiring higher capacity than your initial system could provide, you would want to game out the possibility of adding more CPT lines on the same route to increase capacity. That will add costs, but as Dale points out, adding more capacity on an LRT route would cost something too. And so the question you might ask is at what future demand level, if any, the LRT route would have a lower average cost for an equivalent capacity.
Good point…. so including the time-value of money, is it cheaper to make a larger investment now, or make smaller investments to meet demand later?
That’s essentially the trade-off I guess…which makes CPT more flexible. If demand doesn’t rise as much as we thought, but a new corridor arises that has a high-demand for public transport, then we can build a line over-there.
Jeff,
I tend to really agree with what you just said.
As I see it, transit planners have such an abysmal record of accurate ridership predictions, that gambling large sums of money on the assumption that ridership will just appear in the future makes little sense to me.
Id rather see more moderately priced systems with the potential for expansion or upgrading in the future once demand has been proven to exist. I believe that should be the case with cable or any other form of transit.
We simply do not have the money to spend on some planners or politicians guess about what may or may not happen in the future.
I think that’s a great way of looking at it, Jeff. Especially the time-value component. Typically we compare more expensive options in the future to cheaper ones in the present. In the situation you’re describing, the exact opposite is the case.
The speed really should be part of the comparison. The initial segment runs 14 miles in 40 minutes, for an average speed of 20 mph, much higher than the Squaw Valley system and as high as any Gondola system. That’s important given this is inteded to be a regional transportation system; as it is if you got to the airport just as a train was leaving and had to wait the full 7.5 minutes, it would still be faster than the Squaw Valley Funitel. And the extensions currently being built have wider stop spacings with average speeds around 30 mph (48 km/h).
Additionally, it’s not the case that the system will need to be rebuilt every 30 years. The supporting infrastructure, tunnels, and stations last much longer and they’re most of the cost. Replacing tracks and vehicles is much less expensive, and the tracks should really last longer than 30 years too.
That said I’m pretty sure 3 minute headways are not feasible due to the street crossings. Through downtown and north, though, this will be possible and that’s where the highest capacity demands will be.
What I’d like to see happen is for Sound Transit to focus on making this a high speed high capacity regional system, and then consider augmenting it with gondolas to connect to places a mile or a few miles from the line, or for secondary lines. I emailed them to ask if they’ve considered this technology and they wrote back that it makes rescues in emergencies difficult. Doesn’t seem to have stopped Portland, though, maybe our new mayor is crazy enough to be convinced to build one of these.
Eric,
Yes, tunnels and stations will remain, but especially the at grade parts of the LRT will need replacement every generation or so. Speed should definitely factor into the equation, too. Great point about the impossibility of 3 minute headways. That should really make one question investment in high capacity technologies based on future demand. If that future demand is physically impossible, does it make sense to install.
I think it’s funny when people say rescues are difficult, especially in places where they’re building transit tunnels underground in earthquake prone regions.
I believe we don’t need 3 minute headways down south within our planning horizon, and that’s where most of the street crossing happens. However, we will be near 2 minute headways by 2030 at the core of the system.
Also, when figuring speed don’t forget to add in stations. This isn’t just an airport shuttle – we’re building transit oriented development through residential areas. Each gondola stop will add at least a minute, probably two, to your time.
I absolutely agree that a great place to use gondolas is to extend the reach of high capacity transit. Gondolas shine at high capacity high frequency service over short distances, especially with difficult terrain. That’s perfect for connecting high density areas (or areas you want to make high density) with a core rail system.
“That said I’m pretty sure 3 minute headways are not feasible due to the street crossings.”
Good point. I’m basing that off grade-separated rail lines, and although I’m sure the technical challenges to doing so would be absolute dwarfed by the social challenges of dealing ensuing uproar by the populous who can’t get across the rail line 😉
But yes, people want to get places, so I think most cities do need some sort of high-capacity/high-speed backbone for their regional system. I’m in Austin this summer and it takes me 40 minutes to get from the north end of downtown to the near-south suburbs. That’s really not acceptable.
Great point, Jeff. We often forget about the social viability of a technology and its alignment. Sadly, I suspect cable is most susceptible to that problem.
Rather than a separate database, perhaps their could be Wikipedia articles with the relevant details added so as to make the information more accessible.
Great idea, Sam! I’d never actually thought of that. Will have to a back burner thing, though. Not enough time and resources to do it. I also have no idea how Wikipedia works. I’m also not sure they’d approve of me editing cable-related articles :).
I realize that comparing the Squaw Valley funitel to Link Light rail is hardly an apples to apples comparison, it was merely food for thought. That said, I think a combination of CPT and bus rapid transit could provide a similar degree of mobility at a fraction the cost. Imagine a CPT system taking the place of Link from Westlake station to Mt Baker station (the first 3-4 miles, which are also the most urban and geographically difficult) and bus rapid transit from Mt Baker to the airport (the last 10 miles.) Then use the money saved to build several similar such lines.
Kelly,
I thing you really hit the nail on the head there. This is just food for thought and should make us question the money we’re currently spending on standard technologies rather than exploring the possibilities new (yet proven) technologies present.
As I recall, in a similar discussion involving BRT versus LRT and possible future demand, one point typically made in favor of LRT is that the higher sunk costs of LRT translate into greater public confidence in future service levels, and so LRT systems tend to induce more demand. That makes me wonder if this is another area in which CPT offers something of a middle ground: the capital costs are lower than LRT, but they are pretty well sunk (I’m assuming it generally wouldn’t be worth it for a transit authority to try to move around CPT systems, although I gather that is possible with some of the existing demonstration systems). As a counterpoint, there might be a bit of a discontinuity when you would hit the need to actually add another line, meaning the public might be less than confident that would be done in a timely fashion.
I’ve heard that, too. I’ve yet to see the research to prove it, however. I think the idea of looking at cable as a sort of middle ground is very apt.
Any idea why this post generated so much discussion? Anybody?
Jeff: “Assuming 4 car trains and 3 minute headways:
5920 pphpd seated
16000 pphpd crush load
Headway is a little over 7 times longer than funitel
Capacity almost 4 times funitel.”
Okay, Jeff, how about this….?
Our gondola has 35 pax cabins (50 crush load) with 10 second headway. What does this mean?
Seated: 35 x 6 x 60 = 12,600 pphpd
Crush: 50 x 6 x 60 = 18,000 pphpd
But wait, unlike Wham Bam Tram (aka LRV), which has a one-track mind, Magic Carpet CPT is expandable – easily and cheaply – horizontally (note 1963 NY World’s Fair – relocated and still operating at 6 Flags NJ – with two separate systems hanging off a single arm) and vertically (just drop a new section on top and add a deck to stations). Of course, this means pre-engineering the system with expansion in mind. Presumably, during the expansion phase, the system could continue normal operations.
Single tier:
Seated: 12,600 x 2 = 25,200
Crush: 18,000 x 2 = 36,000
4 tier (expand horizontally and vertically)
Seated: 12,600 x 4 = 50,400
Crush: 18,000 x 4 = 72,000
Capacity 50,400/5,920 = nearly 9 times greater than Wham Bam. And a helluva lot more fun! At 1/4 the cost. Priceless.
While I certainly appreciate David’s enthusiasm, I’m not entirely sure I can endorse those numbers . . . let’s not get too ahead of ourselves here. It’d be nice just to see one small, modest system built in an urban environment. Then maybe . . .
Steven,
The tunnels aren’t as bad as you think. For one thing they’re not particularly vulnerable in earthquakes. And at least in a tunnel you can get out of the vehicle onto an emergency walkway and head for the nearest exit. Even elevators have escape hatches and ladders. Still pretty dangerous in a fire. I think part of it is people accept more danger if they feel like it’s in their hands, so a train that they could try to escape from feels safer than a gondola which they could not. To me the nice safety feature of a gondola is that if something goes wrong on one, not that many lives are at risk, so you can tolerate somewhat less safety. But that thought isn’t comforting to most people. Maybe “safer than flying” is what you promote?
Silly Ideas for Safety:
Include a length of rope, harnesses, and rappelling devices. Then if you NEED to get out, you can through the rope out and rappel down. The only problem is that, I think you need at least a little bit of training to be able to do do this 😉
“We simply do not have the money to spend on some planners (or politicians) guess (<- also wrong – scenarios and prognosises are not guessed, but analytically developed information) about what may or may not happen in the future."
It is crucial to do that! With any investment you have to know what we are talking about, otherwise it is … just a waste of time – a little tea time chat.
"they wrote back that it makes rescues in emergencies difficult." that quote and what Eric said recently about danger and so on: people are just not ready. they have no idea how a different kind of system could be useful. or why e-cars might be "a good" solution. there will always remain people which are against something, but saying that a cableway makes rescues in emergencies difficult is almost about the same than saying we should not live far from a fire station or build any fences around our houses. because the fire engine would need too much time to be there in time and probably couldn't enter an area because of the fence.
statistics prove that a cpt is by far the safest way to move around – for passengers and for people around.
and "Any idea why this post generated so much discussion? Anybody?"
because we are starting to join the part where we can talk about facts. facts are proof and when we're talking about ideas about projects worth millions of dollars we need proof. it is the only solid basis. otherwise it will once again remain just a little nice tea time chat. and with all the information and interest i think it's time to take the next step. it's a good round.
last: another very important point is the comfort of reaching your point of destination without changing the vehicle. it changes everything if you have to switch. so i think in case of seattles lrt there was here the suggestion of using a cpt for the last couple of stations, but they chose the more comfort way. and i have to admit besides that reason from a distance it seems to be more "whole". we need things which are easy to understand and make us feel like we are in control in case of an emergency.
The point is that the Seattle LRT is not a typical LRT. Stations are spaced very far apart so that not all areas along the line are within walking distance of a station. Just check out Paris metro or London tube station spacing for a comparison. A typical LRT would have twice as much stations for teh same distance as the Seattle Central link has. So the Seattle LRT is indeed a regional rail built with LRT technology.
Emergency evacuation is really an issue, other technologies like monorail also suffer from the same problems as they have to incorporate a emergency walkway along their track, which makes suspended monorail quite pointless and takes a advantage from straddled monorail. BTW a tunnel is probably the safest place to be during an earthquake. So for long distance city systems i think the installation of rail bound CPT with emergency walkways along the line and the ability to run in tunnels will have far more success than aerial gondolas. One reason for Portland and New York to choose aerial trams instead of gondolas is that the can have cabin attendants with aerial trams.
Christchurch City wants to build light rail as part of its post-earthquake reconstruction plan. All I could think of was miles of buckled track when the next one hits. Would cable cars be more resilient during a quake event – just a bit of retensioning and off we go again?
It depends what type of cable car you’re looking at. Were it a San Francisco type, it would be as UN-resilient as any other form of rail-based transit. You have to recognize that the underground conduit of cables, pulleys, etc. would likely be irreparably harmed in an earth-quake.
Alternatively, an above-ground system would be more resilient, though likely not as resilient as a gondola system.