It is a nice article. Summary is they wanted lighter cars so accel/decel would require less power and the gas/brake system would be lighter/smaller. So then the wider gauge is helpful to increase lateral stability (to prevent tipping over) since the cars are so light.
In my opinion these kinds of historical mechanical/civil engineering stories are quite interesting. Would love to see more.
That's not what the article says: "On aerial trackways during high wind events – which the engineers knew were not uncommon in places such as Contra Costa County -- the light BART cars could potentially tip over, the engineers hypothesized."
Marin /did/ want it, but after the Golden Gate Bridge authority insisted that the bridge couldn't support the extra deck needed for BART there was a risk that if Marin were part of BART they'd be on the hook for the money without ever receiving service. That, unsurprisingly, reduced enthusiasm - and if a reduction in the Marin vote had dropped the overall bond vote below the required 60 percent, none of the system would have been built, so Marin was kicked out.
Not hurricane force or anything, but gusts of 50 mph (45 knots) or more are not uncommon during Diablo wind events or storms. Gusts up to 70 mph (60 knots) have been recorded at times. Just this winter we had 90+ mph winds recorded in Sonoma (wouldn't have affected the current BART system, but still).
Some parts of the east bay are really friggin windy. I used to live on top of a hill near the carquinez strait and during windy season they'd easily exceed 60mph during a storm.
Like once every other year a bunch of fences would get demolished lol.
There's the Toronto gauge, the Japanese high speed rail gauge (which is standard, 1435mm, for the world but wider than what was used before in Japan, 1000mm and less) and the Spanish high speed rail gauge (which is also standard, 1435mm, even though they usually use Iberian gauge which is wider - and on this, they have some interesting train designs that are capable of automatically switching between the two).
I recently read that although the Spanish high speed lines don't have many passenger trains (the recent Perpignan-Figueres tunnel only has two per day, although that's partly because of corona), they also carry through freight traffic from France because there's no break-of-gauge. A bit unusual for a high speed line, but why not.
That really interesting. I wonder how much that impacts maintenance if at all. They definitely couldn't do that if timing on passenger rail was increased, but while there's sufficient gaps then as you said, why not.
Yes I was shocked at this too. Even the flagship Madrid Barcelona line only has 1 train per hour.
This seems ridiculously low given HS2 in the UK is planned to have 18tph on the core line (to various destinations, but still).
To make matters worse the spainish high speed lines have mandatory reservations so once the train has sold out you are not allowed to board - unlike the UK where you can still board but may have to stand for some of the journey - which is not ideal but at least you don't have to stand around for another hour waiting for the next train anyway.
Is this local transit? It is England so practically everything feels like local transit, right? It is better than what we have, sure but it isn’t as impressive as long distance routes.
> The schedule offers trains every 3 to 5 minutes on the most used sections during rush hours. During other traffic periods, intervals range usually from 4 to 12 minutes or up to 20 minutes on outer sections. During late nights, only selected express services are operated and all late-night services usually run every 20 minutes.
17 says Wikipedia, but the information might be out of date. See [1].
The existing quad-track line carries 15 trains per hour [2], and is full, but this is because it carries trains of varying speed -- express passenger trains, commuter/regional trains and freight.
(The Victoria Line is a metro line, pretty irrelevant comparison.)
If you maintain similar stopping patterns you can get higher than that on a two-track railway - metro services can do 25tph,30tph or even slightly more.
From December the west coast mainline is bringing in a roughly pre-covid timetable. The railway is a 4 track railway, but looking at just the fast tracks leaving Euston, there will be 13 departures an hour on the fast lines
3tph Manchester
3tph Birmingham fast
2tph Liverpool
1tph Chester and North Wales
1tph Glasgow
1tph Crewe via Trent Valley stops
2tph Birmingham via Northampton (switch from the fast lines to the slow lines after about half an hour)
None of these are high speed - I'm currently on a train to Manchester and my phone says we're doing about 125mph (200km/h). HS2 is a more reasonable 200mph.
To be fair, Accela Washington-New York-Boston is more akin to the London-York-Edinburgh services which are only 2tph, or 31 trains per day (44 run as far as York, but only 31 continue to Edinburgh)
The frequency on the WCML can't be higher as some of those trains are non-stop as far as places like Stoke, Crewe, or even Warrington (about 90-110 minutes out of Euston), where others do stop at places like Watford, Milton Keynes and Rugby.
(In addition to the 5 trains per hour to Birmingham above, there is another one or two from a different London station - Marylebone, and there's a handful of additional trains from Euston to Edinburgh via Birmingham)
My normal London-Stoke train is 145.9 miles in 1h28, average of 99.4mph, with 125mph along great swathes of it.
The hourly Glasgow departure averages 105mph to Warrington
Once you get up north speeds drop -- the eng to end average for Glasgow-London is about 84mph (401 miles in 4h47), Edinburgh-London is a similar distance but a bit faster at about 4h26, so 89mph.
Washington DC to Boston is 440 miles and seems to take about 6h40 at best, just 66mph
America, where I’m free to drive anywhere at any time because public transit doesn’t effectively exist for most people :(
I miss the trains in other countries so much. They’re reminders of what we could have had here had we not let NIMBYism win. https://www.quora.com/Why-wasnt-BART-built-initially-to-loop... has some of the history, there’s probably a better article out there somewhere.
It had less to do with NIMBYism and more to do with car company profits. [1]
Here in the Denver metro area, we used to have trains running all over the place, and 100 years ago, there were daily trains between Denver, Boulder, Golden, and Longmont. Then they ended service and tore up most of the tracks.
Fast-forward to 2022, and now they're trying again to put rail between Boulder and Denver, but it's a $1+ billion boondoggle that hasn't gone anywhere in 15 years. [2]
I suspect that's counting both the intercity trains and the regional trains. If so, the Northeast Corridor in New Jersey and southern Connecticut will be hitting roughly those numbers already--NJT and Metro North add a lot of traffic to the NEC.
Oh for some reason I interpreted this as that many trains departing to the same place. Like as if there were a train from New York to Boston every few minutes.
These are all intercity trains to cities north of London.
There will be 6 trains per hour to Birmingham (120 miles from London), three of which continue beyond Birmingham. An additional 11 trains don't stop at Birmingham, but cities beyond it.
The thing that you're counting as "intercity" is roughly comparable to what is classified as regional service in the North East US.
Destination | Operator | Distance from NYC
Camden / Philly | NJT / Amtrak | 82 mi
Montauk | LIRR / Amtrak | 110 mi
New Haven | MNR / Amtrak | 67 mi
Except for Montauk, these places are cities in their own right, but we don't really think of them as intercity trains because the primary reason someone would board is as a commuter not as a traveler.
An intercity train stops primarily at large towns and cities. Acela looks like the only frequent intercity train from NYC, though I may have missed one as there are several daily Amtrak trains.
The planned schedule for the new line in England will have the shortest train make only these stops, 3 times per hour:
London Euston 0 miles
London Old Oak Common 5 miles
Birmingham 125 miles after 52 minutes
And a longer one (also 3 times per hour) will make only these stops:
London Euston 0 miles
London Old Oak Common 5 miles
Birmingham 125 miles (1 of the 3 per hour)
Manchester Airport 205 miles
That is nothing like a regional/commuter train. The existing regional train to Birmingham stops at Watford, Milton Keynes, Wolverton, Rugby, Coventry, Canley, Tile Hill, Berkswell, Hampton-in-Arden and Birmingham — that is like the Philly/LIRR/MNR trains.
No it's not. This is purely new high speed intercity services. If you include the WCML out of Euston ("legacy line" - can still do 125mph/200kmh - which will have increased regional services as space is freed up by moving the intercity ones to HS2), then you can at least double that number. Probably triple given it is a 4 track railway.
LGV Sud-Est, France's first and Europe's most used HS line, sees 240 trains/day/direction, with a capacity of 13 trains/hr, to be improved to 16 tr/hr in 2030. It's a trunk line with many different service patterns though, mostly linking Paris to southern destinations.
Alacatel/Thales promised Muni something like 48 trains per hour (across two tracks and five lines) with their new fangled train control system. For the most part that never happened.
For the time being BART and Muni run 2–3 minute headways during peak hours (so somewhere around 25–30 tph). One of the big things BART remedied with the new cars was the lack of doors (which slows down boarding and increases dwell time).
In the UK/Germany (and probably others) you can stand if you want to and it is sold out. This is obviously preferable (though not ideal) to having a wait a week for a seat, or get a very slow coach/bus instead.
We are talking about 2-4 hours of travel time standing up, almost nobody would like that.
In Spain there are several railway companies: RENFE, Ouigo and (TBD) Irio. There are enough trains for the current demand. If they detect a surge (holidays for example) new trains are added to the timetable.
Short-distance trains do indeed allow you to be standing up but the travel time is at most 1h.
Well the point is it's an option to stand even if it's far from ideal. I personally would prefer to stand for an hour or two (trains tend to empty out en route) than stand at the station to wait for a seat. I've done it multiple times in the UK and Germany.
There are not enough trains in Spain for the cost of the high speed infrastructure. Looking today both the 19.30 and 20.40 MAD BCN are sold out, which means you would have to wait until the next day to travel if you couldn't get to the station before 19.00 (which isn't uncommon).
> This seems ridiculously low given HS2 in the UK is planned to have 18tph on the core line (to various destinations, but still).
Before COVID-19, there were more than 40 trains a day from Madrid to different destinations (Sevilla, Cordoba, Barcelona), and over 20 to many others, so the difference is not that large.
But then also, London is around three times larger and has a bigger weight in its country economy, than Madrid. It just makes sense that more people commute or travel to London anyway.
I’m not sure it’s legal to have standing passengers in high-speed trains. That’s why you have to buy your ticket before getting into the train, and showing the ticket to railway company personnel if requested. You have an assigned seat there.
In the short-distance trains you can get into the train only by passing a card through a rfid reader. Maybe you should compare this trains to UK ones.
Thus, Spain has less population than the UK, is less centralized than UK (where almost everything is in London) and Spain is much more mountainous. We can’t really compare these two countries.
Security concerns. I’m case of an accident you can sit and maybe survive.
Edit: keep in mind that the Madrid-Barcelona high speed railway line reaches almost 250kph. That’s more than the double maximum speed of a car in Spain.
But you don't have assigned seats on German high-speed trains. It certainly happens (although rarely) that trains are so full that some people need to stand
fyi it's possible to buy Tokyo to Hakata (over 1000km, takes 5hr, max 300km/h) ticket without reserved sheet. (Though eventually free sheet will be available at the middle stations.)
That France-Spain connection is fairly special. Firstly, international connections tend to have much less traffic than intranational ones. Secondly, the line was built with a PPP project (long since failed), assuming it could be very profitable based on track access fees - and the track access fees for that stretch of railroad are basically the highest anywhere in Europe except for the channel tunnel. But unlike the channel tunnel (and thirdly), the connections across the line aren’t as useful as the channel tunnel, which connects a giant metropolis (London) with several other metropoli in 2-4 hours, a sweet spot for rail, whereas Barcelona and Paris/Marseille are more like 6h. Once the French side will update their Mediterranean rail lines to hsr standards, the time will go down and ridership will increase, despite the high costs.
Lastly, there isn’t much freight on the line, because (besides the high track fees) there isn’t much of a standard gauge freight rail network in Spain.
In some sense the Perpignan Figueras line was supposed to be like the channel tunnel, but it really isn’t. It’s not exactly comparable to the rest of the Spanish hsr network - which is arguably overbuilt for political reasons to get hsr into many even smaller cities, and perhaps under-utilised at this point, but it’s not as bad as that one line might suggest when picked as a representative.
The Perpignan-Lyon has been postponed by the French government. When they complete it, the connection with Barcelona will see much more railway traffic.
There’s also the 1600mm Irish gauge which historically was down to picking a compromise between two standards used in lines built in the 19th Century. Some Irish rail engineers brought it with them to places like Australia and Brazil.
There's a small gauge (I think 2 feet) used mainly for moving pineapples and sugar cane around on plantations. The rail is readily available so it's used for other purposes too. I saw it used for material handling in rail shops in Canada, and it was always called "pineapple gauge".
another reason they're nonstandard is to keep freight rail companies from demanding to use them. part of the reason rail on the us is so slow is freight companies have had a 150 year monopoly on the right of way.
another reason they're nonstandard is to keep freight rail companies from demanding to use them. part of the reason rail on the us is so slow is freight companies have had a 150 year monopoly on the right of way.
You're confusing two different topics.
Freight railroads cannot "demand" to use rail lines owned by passenger railroads. Passenger railroads can, if they choose, allow freight lines to use their tracks, and vice-versa.
You can see examples of both in Chicago. CTA, a passenger railroad, used to carry freight traffic, but chooses not to do so anymore. Meanwhile, the local freight railroad companies choose to allow Metra, another passenger railroad, to use their tracks and rights-of-way.
It's not anywhere close to the black-and-white situation you present.
Where passenger railroads own their tracks, they get to decide what runs on them. The problem is that passenger railroads in the United States own very little actual track, and rely too heavily on using freight-owned tracks.
There's lots of railroad magazines that write about this ongoing problem every month.
No, CTA cannot do that. If they did, CTA would have to take a bunch of expensive steps to make their stuff complaint with FRA regs.
And no, the local freight railroads aren't "choosing" to let Metra be. They pretty much have to. That was part of the agreement when the regulators allowed the railroads to drop passenger services (which they were previously legally obliged to keep). There's a process, which hasn't been put into practice for a while but is still on the books, for Amtrak to demand the right to use certain tracks (the exact scenario where this happens escapes me).
> freight companies have had a 150 year monopoly on the right of way.
In the US, railroads had to run both freight and passengers or else leave money on the table, 153 years ago, when the first transcontinental line was finished. And if they wanted to advertise speed to passengers, the freight trains had to yield.
When Amtrak started of course things changed a lot, immediately and also gradually, but that was 1971, 52 years ago.
There's a fun story about BART: the US decided to go with a blank slate approach to railway design by hiring not railway engineers but aerospace engineers, and consequently the project relied heavily on reinventing the wheel. This track gauge thing happens to be one of many examples.
As BART decided to forego any standard way of doing mundane railway tracks, they had to endure decade-long delays and huge budget overruns to cover up their loss. Otherwise there would be nothing to talk about a mundane rapid transit system which already covers pretty much the whole Europe.
Good to know it was for steadiness against the wind given its light construction and the original intent to have if go on (under?) the GGB deck.
I’d like to know why the tracks are flat and not angled and require regular grinding down to even out the washboard effect this causes… which results in that high pitched sound whenever it enters a curve due to slippage.
Correction: the rails are flat, it's the wheels that are also flat. So pretend the above are the upper half of the wheel profiles.
End result is more surface contact = more noise/squealing, also less self-centering of the wheels on the tracks, so the lips also brush against the rails. I think this also exacerbates the washboarding of the rails.
BART had regular rail profile and non-tapered wheels, so flange squeal (and wear) was apparently a real issue. I'm not sure why they decided on this, since it runs counter to normal railway practice. This article has a story but I wonder if it wasn't just more NIH.
Are you sure? I was just watching a video that explained the wheel edges are sloped, not the rails, to help fixed axel wheels to around curves. Wish I could find it again - quite ingenius!
Yes. The wheels were flat, apparently Bombardier saw room for improvement and suggested a tapered profile that was retrofitted to the older cars. Unfortunately the track pitch (I'm forgetting what the proper term is) is also wrong, the spacing in the curves is too tight.
Normally you'd increase the spacing a bit in curves to encourages the wheels to rotate at different speeds instead of destroying the track. Unfortunately even with the new profile BART is still incredibly loud, and it's not helped at all by low speed curves like the Oakland Wye.
I think the term you're looking for is track gauge.
And as for a need to widen it in curves, the answer is basically it depends.
On mainline railways gauge widening in curves basically stems from the days of steam locomotives, which mostly didn't have bogies (at least for the main driving wheels) and comparatively enormous wheelbases.
On straight(ish) track, some minor gauge widening as compared to the nominal standard gauge might nowadays also be beneficial, especially on high speed lines.
Within the context of streetcars/trams, which often need to negotiate very tight curves on the other hand, current state of the art seems to indicate that a slight gauge tightening might actually be more beneficial in order to avoid too large a skew angle between the bogies and the rails. (Compare http://interfacejournal.com/archives/472)
I'm no wheel/rail-interface guru, so quite what the conclusions from that would be for BART I can't say, either, but just wanted to point out that the topic is somewhat more complex, and nowadays gauge widening in curves might not actually always be the right answer.
> I think the term you're looking for is track gauge.
No. Track gauge is the nominal width, no? I'm talking about how the width changes in the curves. A while back a former BART engineer went on the record claiming the spacing in the curves was contributing to the noise.
Top to bottom BART fully embraced NIH. Fifty years ago there was good reason but now the rest of the rail industry has moved on and BART is stuck with a lot of tech and engineering debt. For e.g. BART can't grease or sand their tracks either, and you can feel the older trains slip and slide (and damage the track) as they exit the station.
Edit: Here's an interesting piece on noise. Note how nearly every other rail system lubricates the tracks (BART can't) and how no other rail system uses (noisy) solid aluminum wheels.
> No. Track gauge is the nominal width, no? I'm talking about how the width changes in the curves.
I'm not intimately familiar with the English terms, but I think it still all boils down to the track gauge. That Boston article I linked to certainly talks about track gauge and gauge-widening and the like.
Normally you do both: The running surface of the wheels is sloped, but the rails are also mounted onto the sleepers with a corresponding inclination. (Some grooved rail profiles for streetcars/trams have the inclination built directly into the profile of the railhead, so the rail itself is mounted upright, but you get the same effect at the railhead level.)
It's rail wheels which are beveled, with the radius decreasing to the outside of the wheel. Your top image has the wheels reversed. There's also a flange which extends below the rail depth on the inner edge of the wheel
As the car rounds turns, the inner wheel rides down and the outer wheel rides up, increasing the relative rotational radius of the outer wheel. This compensates for the fact that rail wheelsets rotate on a single fixed axis, not independently, and without a differential.
It makes it more complicated. The wheels need to be fixed to the axel for breaks to function, so you can’t just stick a bearing in the wheel and call it a day.
The fixed wheels, plus the bevel create a self-centering effect.
As the wheels move off centre, the bevel makes one wheel have a larger effective diameter and the other wheel a smaller effective diameter. Because the wheels have a solid axel, that forces one wheel to start travelling slightly faster than the other, causing the pair of wheels to turn.
With the bevel making the wheels larger on the inside, it means that as a pair of wheels drifts right relative to the tracks, the right hand wheel rides up, starts behaving as if it’s larger, and then starts to overtake the left wheel (which has ridden down, and behaves as if it’s smaller), creating a turning effect to the left, correcting the rightward drift.
When setup correctly, and travelling at a reasonable speed, the big metal lip on train wheels should never touch the track. The self-centering effect does all the work, the lip is just a backup to deal with overly tight turns, points and extremely slow running.
I get the principle of how it works - though it seems imperfect bc it would vary depending on the weight/load. But why favor such a mechanism to just independent wheels?
Or are you saying the whole setup actually tilts the carriage into the turn - helping to prevent it from tipping?
If the wheels were independent the lips on the wheels would constantly ride on the inner sides of the rails. That increases rolling resistance and wear.
There are some trains which use independent wheels, but they need complicated mechanical centering mechanisms to replace the self-centering effect, and these are more expensive to maintain.
> though it seems imperfect bc it would vary depending on the weight/load.
Weight and load have no impact of the effect. The only requirement is that the wheels remain adhered to the track, and don’t slip. Which is expected anyway, because it would be tricky to accelerate and decelerate a train with slipping wheels.
With independent wheels you have no self-centering effect, so only think keeping your wheels point in the right direction would be inner flange on the wheels colliding with the track. Such collisions obvious involve quite a bit of friction, noise and track/wheel wear.
Not to mention independent wheels are more complex to build. Two wheels directly attached to a single fixed axel is about a simple as you can get. You only need two couplings between your single axel and non-rotating part of your train. With independent wheels you need four, or more complex wheel designs to allow bearings at the wheel/axel coupling.
There no need to tip carriages into turns to prevent tipping, gravity, track banking, and speed limits already assure that tipping isn’t an issue. Tilting trains only exist to increase passenger comfort, not safety. Tilting trains are prefect capable of operating a full line speed with their tilting mechanism disabled or jammed, it’ll just be uncomfortable for the passengers.
One minor correction. Tilt trains exist to allow higher speed service on older lines where tipping would be a safety problem. They can run just fine with the tilt mechanism disabled, provided they go at the lower speed of the older trains.
> Tilt trains exist to allow higher speed service on older lines where tipping would be a safety problem.
That isn’t true. No train network is going to allow train to operate at speed, if the safety of those trains depends on non-failure of a complex tilting mechanism. Having a tilting mechanism fail or jam before a corner (which has happened) can not result in a high speed derailment. That risk is simply unacceptable on modern railway.
Tilting trains make it possible to run at higher speeds on older lines by allowing older lines to have the camber on corners increased to accommodate the higher speeds. But the camber can’t increased enough to hit optimal amount of camber without make the rail unusable to older trains.
Tilting trains provide a useful compromise. Increase the camber enough for stability (but not comfort), while keeping the camber acceptable for older trains, then make up the rest of the tilt with carriage tilting mechanism. The total amount of tilt is then comfortable for passengers. There is a minor safety element provided by tilting trains, in that lateral forces in cornering are lower, thus passengers are less likely to fall, and items won’t come of tables. But the tilting mechanism certainly is a safety critical element to running faster on old lines.
It’s also worth mentioning that carriages on passenger trains have a pretty low center of gravity, as all the heavy equipment (motors, bogies, breaks, suspension) sits below the passenger compartment. As a result tilting a passenger train won’t significantly move its center of gravity, so impact on total train stability is going to pretty minor.
I'd always understood tilt trains to be a comfort factor rather than safety.
Tilting a carriage isn't going to adjust the CoG by any appreciable amount. It will tend to make the net centripetal moment more virtical from the reference frame of passengers, however.
Wikipedia's article corroborates this:
As a train (or other vehicle) rounds a curve at speed, objects inside the train experience centrifugal force. This can cause packages to slide about or seated passengers to feel squashed by the outboard armrest, and standing passengers to lose their balance. Tilting trains are designed to counteract this by tilting the carriages towards the inside of the curve, thus compensating for the g-force.
With separted axles, you wouldn't get the climbing behaviour of the outside wheel. With a fixed axle, the idealised shape of the wheels is of a double-ended cone (see the video for an example) which can ride higher on the outside of a turn. Even with conical wheels, with independent axels there's no way to impart a turning force on the carriage. I'm not entirely sure what the end result would be, but I suspect it would be a far harsher ride, and much more track and wheel wear. Probably a much higher probability of derailments as well.
Cone angle varies by type of train and speeds. The video shows both freight and high-speed rail designs and discusses characteristics of both.
This is what causes a lot of noise and wear and tear. I actually researched this and it seems that it was not as stupid of a decision as a lot of people say it is. There is a problem with beveled wheels -- it is called hunting. This is when the car oscillates from side to side. Hunting is unpleasant, also causes wear and tear and noise and creates a danger of derailing.
Hunting becomes more likely if the car is lighter and if it is going high speed. The bart was designed to be lighter than most existing metro systems and to go much faster than existing metro systems in certain corridors. So they wanted to prevent hunting and thus made a portion of the wheel cylindrical.
So it seems that they just made another engineering trade off. And I am sure they had justification. But I do wish that when engineers make these trade offs they take human comfort into account. I does not seem that they gave sound much of a weight when making their trade offs.
I think the solution is to keep the wheels cylindrical but just decouple them. I.e., not to have the right and left wheel connected by an axle. If they are independent they can move at different speeds on the turns and will not cause grinding and noise. And the wheels that are driven may be driven by different electric motors that drive the wheels by the necessary different speeds on turns. This is called vectoring and is already being done for some electric cars.
Of anyone wants to see what hunting looks like, then watch a video of the London DLR.
The DLR has very tight corners, and was originally built on very tight budget. As a result it uses very steeply bevelled wheels to handle the corners.
The wheels do a great job on the corners, but on the straight they suffer from significant hunting, resulting in the trains wobbling side-to-side (and slightly up-and-down), at a rather alarming frequency and amplitude. Not quite enough to make you sea sick, but it’s pretty close.
The DLR is a low-speed network with tight turning radii. I haven't found any footage clearly showing hunting (in part because it's difficult to distinguish other lateral-movement-inducing elements such as turns and switches), but the basic operating parameters are fairly clear.
I suspect the wheels have high conicity. The rolling stock is also likely lighter than a heavy-rail system.
Well, there's a reason they named it the Docklands Light Railway!
(Large stretches of the original network built in the 1980s ran on elevated track, supported on pylons over ground that's a mixture of chalk and clay and tends to waterlog. Heavy rail can be built over and under that part of the Thames, but it has its own special challenges and the DLR was built as the cheapest way of getting commuters in and out of the Canary Wharf office towers, from Tower Gateway and the actual London Underground network to the west. Then from the 1990s onwards mission creep set in ...)
its like the lifeline of east london, and it connected large aread of land that used to be borderline slums. Now there are highriae apartments and the fact that they have access to piblic trandport has increased their valuation by billions.
Providing improved transport and transit is amongst the best way of increasing property values. A stellar example is Denver, Colorado, which grew in size tenfold in a decade after a branch line connected it to the Transcontinental Railroad in 1870, becoming the dominant Rocky Mountain region city, a title which might otherwise have gone to Chyenne, Wyoming.
It's also something of a two-edged sword: the property value boost makes further transit rights of way far more expensive to acquire. I've noticed in particular that much rail development happens either as greenfield projects within countries as they're first industrialising (UK and US in the 19th century, China in the late 20th), or following a devastating war or other economic setback --- Japan's shinkansen and to a somewhat lesser degree France's TGV are both legacies of WWII.
In the US, aside from political opposition from other parts of the transportation sector, it's wealthy landowners who are the most significant obstacle to new development.
Oh trust me, the hunting on the DLR is pretty bad. You’ll notice the second the trains get up any speed on a straight. The problem is bad enough for it appear in its own section of the DLR Rolling stock wiki page.
The video confirms what I'd surmised: that tight turns were the reason for more conical wheels. The sections I watched didn't actually show the wheel geometry, though, and I've not been able to find any photos online (DDG and Google Image search, various combinations of "DLR | Docklands Light Railroad | (conical|cone|conicity) wheel"). The WIkipeidia DLR Rolling Stock article mentions the conicity and hunting issues, but doesn't show the wheels either:
The hunting phenomenon is real and has some interesting consequences. For freight trains, empty boxcars have a lower speed limit than fully-laden ones, as the hunting behaviour is greater with less weight.
High-speed rail systems required understanding and managing hunting action. Getting a case of the wobblies at 300 kph is an exciting but short ride.
The Shinkansen also use cylindrical wheels on flat rails. The solution is to spend a large amount of effort and money into maintaining both the wheels and the rails. The amount of maintenance that BART train wheels and tracks get is much lower than what is done to keep Shinkansen running.
Hunting in modern rail vehicles is largely a solved problem using axle yaw dampers and other systems. I struggle to see why BART doesnt use a standard wheel profile and buy from a rolling stock supplier with experience of designing for good ride quality.
Huh, the MBTA Green Line in Boston has a couple of sharp turns where it makes unbearable high pitched shrieks too, I wonder if it's also using flat tracks...
The Green Line still is very much a streetcar at heart, and at those kinds of tight radii no amount of rail inclination (that is no reasonable amount of inclination, after all the whole shebang still needs to work on tangent track, too) will be able to compensate for the difference between outer and inner wheels.
The best you can try is add some sort of lubrication, but even that won't fully quieten it down, since you can only lubricate the flanges, but not the running surface for hopefully obvious reasons.
U can individually suspend the wheels, rather than using a fixed axle. Some low floor trams (streetcars) have those, because it allows passenger areas where otherwise the axle would be.
Yes, but then you're losing the self-steering, which isn't ideal, either. For low floor vehicles you take that compromise (except possibly at the vehicle ends if you don't need fully 100 % low floor), but otherwise it looks like fixed axles still are the preferred solution.
Everyone is trying to explain this but the reason is that BART was designed and built by a company that previously built dams, and in their hubris decided that the rail industry had nothing to teach them. It was the Space Age and Americans were absolutely overflowing with hubris.
Seems highly likely that keeping freight trains off the tracks also factored into the decision. Huge shame that they didn't consider the ongoing maintenance cost of using a non-standard gauge.
I have to be honest, I frequently find myself with NIH syndrome. I just don't understand how we can make progress if we defer to "everything is a solved problem!"
Truly, help me understand how to better navigate when to defer and when to have the courage to invent here.
There is a difference between inventing something different to understand and not using it and inventing something and trying to force it on the world. Standards are often useful.
In the case of trains they could be a lot cheaper if everyone agreed one one standard and only made changes when everyone agrees to upgrade at once (over 40 years counts as at once,so long as there is agreement). All the different standards mean manufactures have to have more expensive jigs, and also cannot optimize their lines as much.
Though wheel gauge within limits that Bart fits is not too big a deal to adjust on a line.
Often the cheapest way of building new rail lines is to repurpose existing ones. Having to regauge or parallel them instead significantly raises costs.
BART has actually built or planned to build standard gauge since then to reduce costs on future extensions. Some extensions, like BART absorbing Caltrain to go to San Jose, have failed due to the cost.
Toronto's streetcars and subway trains also have a unique gauge, but all new lines built and about to be built will use standard gauge. Extensions to the existing three subway lines and the streetcar network will continue to use the old gauge.
Every single bogie on every single LRT or subway car is "custom". Given all the tooling, training, and skills the TTC has with TTC standard gauge, introducing a non-standard gauge so that a politician can be seen to be making a difference will effectively just double the cost. Sure, a guy in a blue tie gets to benefit from "subways subways subways" including a "subway" of a bus on tracks in a tunnel in his own riding, but does it benefit the people who use and pay for the extra maintenance, training, tooling, and storage?
Note that the new lines are not TTC lines, they are Metrolinx lines. And Metrolinx is also building LRT systems in other cities that use standard gauge, so it definitely makes the most sense to stick with standard gauge instead of switching to the TTC's unique gauge.
Now imagine that Metrolinx's GO trains also used a non-standard gauge? Not only could they not share track with freight trains, or even Via Rail, but simply delivering the train cars from where they are manufactured in Thunder Bay would be a nightmare. As it is, they're just attached to a freight train and pulled along like any other car, and I've even seen them on a CP freight train.
To add onto this, the only modern streetcars that have been built to Toronto's gauge, the new Flexity streetcars, have issues, so there's not really a compelling reason to buy those.
The new new streetcars (for Line 6 Finch) are the Alstom Citadis Spirit, identical to those used in Ottawa and also destined for the Hazel McAllion. Given the wheels literally keep falling off, the gauge is the least of the worries.
Standards mean you can share track maintenance tools, track maintenance is obviously required, but some of the processes need very expensive tools that bart needs for only a few weeks per year.
Standards also mean you can buy standard parts instead of getting something custom made. This can greatly reduce costs as mass manufacturing is only cheap when there is a mass needed. (At least Bart choose a guage in use elsewhere so there is some demand for those parts).
Standard means CA HSR or Caltrain could share less used track which can save everyone money. (It makes scheduling hard, but that is a lot cheaper than building track) who knows, maybe freight rail would use that track for some low demand customers they can't serve at all now, and trains are more environmentally friendly than trucks.
Things being standardized is in itself a benefit. Anything unique has to outweigh that.
If you invent a competitor to microSD that somehow stores 50% more data in the same space(For example only, just ignore the fact that form factor is independent of the flash storage tech), I would be annoyed.
If you push to make it the global standard and succeed, I'll still be annoyed but appreciate the progress.
If you just make a line of devices with unique storage, I'll probably be like, yeah, but MicroSD was fine and this doesn't work with my readers, and now I have to buy a new card.
In software NIH often doesn't offer much benefit at all. It's usually just "We want to build it ourselves and really understand it". Sometimes features are even worse("We don't need to support importing those formats, build something smaller that only support one"). Sometimes performance is worse ("Who needs a million line library with crazy vector instructions? 5 lines of C will do").
> I frequently find myself with NIH syndrome. I just don't understand how we can make progress if we defer to "everything is a solved problem!"
Progress involves making the best changes we can. NIH syndrome means re-litigating small changes that aren't impactful rather than focusing on the new big ticket items that will really make a difference.
What's the point in optimizing something that will have a negligible return when there are better opportunities? And there's also the important downside of NIH syndrome. It might just be that the people who solved the problem initially knew what they were doing and you'll discover years down the road that your solution doesn't help anyone make progress, it was actually a terrible mistake.
Like the BART solution. The people who did this were aerospace engineers. They thought they were improving on train design. Well, they were not. They were ignoring important constraints you only learn about later: the fact that interoperability between train tracks is critical to long-term development and efficient operation. Extending BART is a nightmare, it cannot share track with other systems, you cannot move trains on other systems to get them to BART which makes it far more expensive to do anything, the parts are incredibly expensive now that the supply chain for them is tiny.
Basically every commuter suffers every day because of this idiotic decision which keeps the system fragmented.
And that's what NIH syndrome gets you. You think you're making progress. But you often just have a terrible understanding of the problem and you're building yourself long-term technical debt.
The secret lies in the difference between addressing the problem and solving the problem. Society is full of problems that are addressed with "solutions" that don't work, or don't work in the way we think they do.
Arguments like "the best code is no code" basically allude to the idea of defining the problem in a way that allows a solution without addressing it through some complex technical dependency.
In transportation there is a lot of popular discussion based around different modalities like car, train, bike. Each one has its enthusiasts arguing for a mode being a solution. This is a very engaging topic because it creates the grounds for an ideology: "everything will be great once we eliminate x and build more y". A great way of addressing transport issues while not discussing the overall problem in terms of land use, tripmaking needs, accessibility, safety, etc. The path to the actual solution is through studying all those other things to identify what key metrics are in play and then specifying your solution in terms of "must meet this metric". That's where the meat of any problem lies, because the person who lays down the finishing touches is building on all that preparatory work - supply chains, scientific principles, engineering tolerances and so on.
And, of course, there's another path into addressing without solving by selecting unrealistic metrics that don't suggest how to move forward. Before you even have the metrics you need good critical thinking skills, ability to collect and analyze data, and some resulting media that communicates the findings like a document. That's a whole job in itself, and then selling and marketing it is another. And then maybe we get to actually implementing something.
It takes a lot, in all. It's almost imperceptible how we progress on big issues. And the whole way through someone is trying to steal credit for some part of it.
From reading TFA: it was a cheap way to improve lateral stability of the lightweight cars in high wind conditions. And the gauge used was common in some other countries like India.
Don't know what NIH syndrome has to do with it, doesn't sound like interoperating with other rail systems was in the spec.
The people who developed the BART were Rohr Inc., which up until that point was an aerospace company. In that sense, they basically ignored most of the previous lessons learned by the industry.
They also entered busmaking at this time and produced the Flxible 870, a notorious lemon of a bus.
From TFA: “The pamphlet does not specify why only the wide gauge of five-feet-six inches was tested and not others.”
Standard gauge was ubiquitous in the US at the time, and this engineering firm didn’t even evaluate it in their report, despite the fact that the original BART plans called for standard gauge. Stability of standard gauge has since been proven to be a complete non-issue.
If you read more about the history of BART, everything about it was NIH. They thought they were going to design some new space-age standard and usher in a new era of public transportation.
meanwhile the peninsula joint powers board (caltrain) had the option to kick freight off their right of way and opted not to. and spends hundreds of millions of dollars building grade separations to freight standards (low max grade) while getting the rails destroyed nightly by heavy badly maintained freight cars
Just coming from Hong Kong or Tokyo I don't understand how someone can write an article like this without feeling completely embarrassed. Bart is a symbol of American incompetence.
It's like, ok, unique decision on the broad gauge. But the Bart is a piece of shit. So why should I care?
A real topical blog post would be one comparing Bart to say the train systems in China; but instead they ignore the obvious and focus on this irrelevant detail.
And just because the bay area average is 431 there are plenty of cities in the bay area with much higher densities and there are comparable cities else where that manage to have great public transportation.
I’m seeing a density of 2,419 per sq km for SF urban core at https://www.newgeography.com/content/007367-toronto-solidifi... and that beats New York at 2,054 people per sq km. Probably has something to do with how the urban core is defined and NYC sprawl, but still interesting…
Would've helped a lot if they'd upzoned near transit, so people would actually live next to it instead of parking lots and SFH. Eventually SB50 will come back and do it, hopefully.
There's also VTA, which guessed completely wrong about where people would move to and so goes from MTV to nowhere.
@deltaonefour - no you don't understand. The article isn't trying to sell you anything. I'm not trying to sell you anything. The article is literally just a nice short explanation of a relatively obscure (but apparently popular) topic - why the BART chose a specific gauge. It's not saying BART is faster, higher capacity, cheaper, cleaner, more modern or in any way better than anything else in the rest of the world. It's just about the gauge, nothing more. There's no need to get so upset about this.
Not just carpet, but padded wool seats -- those were replaced with padded water-repellant vinyl a decade ago, which sounds like a long time until you realize that it had cloth seats for 40 years before that.
Absorbent wool was, as you might imagine, unbelievably nasty. Among other things I saw on BART, I once saw a guy with severe flesh necrosis on his foot sit with said foot, bandages taken off, pressed against the seat. Of course, minor issues like diarrhea and vomit stains were more typical.
As why carpets and wool, the answer is actually just about as interesting as the linked article. Back in the 60s / 70s when BART started operating, it was one of the first metro subways. Its creators envisioned it as a futuristic luxury ride, whisking businesspeople and tourists in and out of far-flung suburbs to the glimmering high-rises of San Francisco. Everyone would be wearing a suit or a nice dress, and they would sigh in relief each time they daintily sat down in these airline-style seats.
The actual future turned out to be a bit more disappointing...
>No one imagines homeless and drug addicts riding their transportation.
That's a pretty strong absolute; I'm certain that, at the very least, folk involved with public transportation do regularly consider homeless and/or addicts when covering various aspects of their service. Maybe not so much in the past, but these days it's basically a given.
I'm not so sure. Bay area is the only transit systems I've been on that seems to commonly have homeless and drug addicts. I've used transit systems all over the world. The rest don't seem to have that issue. Maybe it's because of better social services or maybe it's because so many normies use those other systems that it doesn't become a place for homeless and drug addict to hang out.
It’s not being an actual circle line means it’s not as attractive to homeless people. You can’t just fall asleep and just ride in circles, you get up periodically and change trains now.
I mean, you don’t need the homeless and drug addicts to make wool and carpet nasty, you just need people to be outside before they come and use public transit.
Sweat, rain, dirt etc. are all going to come into the train.
I know, the way I said it was poorly worded. It was a first in the minds of the planners and builders for, among other reasons, the approval process, the engineering conditions they built it under (a compressed air tunnel was required to keep out the mud and water from the Bay, I'm not sure if that was a first for just the US or the entire world), the amount of funding required (largest US public project to date), the specifications of the system, and the supposed comforts of the trains themselves.
In short, it was supposed to be a revolutionary metro, not just "yet another metro system". As it turned out, they would have been much better served by copying what was tried and true.
Have you ever seen the drawings from the late 1800s of what the subway was supposed to be in NYC? The future has been disappointing people for a very long time.
Because BART is an urban subway marketed to the white flight suburbanites. It was supposed to be luxurious not usable. That's why you got two doors at the cost of boarding speed and tons of seats at the cost of standing room.
Dallas runs trains slower, and at certain temps, stops them completely above ~104 sustained for ~>4 hours because the rail can/will buckle due to expansion. I think they also do a stop travel when the temp gets below a certain temp as well. It's not unheard of, and not a bad idea when it comes to public safety and potentially derailing a commuter train in downtown or whatever your worst case scenario is.
Didn’t they modify an existing Union Pacific right of way to accommodate BART for that extension? Changing the gauge would have probably be an excessive cost on top of that (although maybe it would have been worth it in the long run).
I think they plan on doing something similar with the Valley Link project between Dublin/Pleasanton to Lathorp in the San Joaquin Valley.
Despite the District’s claim, I still wonder if the Southern Pacific had a hand in the choice of a non-standard gauge, to help make sure they’d never be ordered to take over BART and bear its operating losses. That risk might seem slight today, but it would have been more of a concern in that era, years before Amtrak, when railroads were being required by their regulators (ICC and CPUC) to operate other unprofitable passenger services.
As the posting concedes, “The pamphlet does not specify why only the wide gauge of five-feet-six inches was tested and not others.”
There was little if any cooperation like that. The BART engineers were there own little tribe and let people know from up on high how things would go. This wasn't just the choice of gauge, it was everything about the design. In particular the collision avoidance system was conceived with very little input and took many years to sort out. The idea that BART engineers took input from Southern Pacific or anyone else has no supporting evidence and is counter to what is known about how they operated.
I feel like people always approach this question from a cost perspective. Ive actually literally had people insinuate that if they used standard guage, they could somehow just repurpose existing trains or existing parts.
I do wonder how much that plays a role. I suspect not much. Bombardier or Kawasaki or whoever is going to mill everything from scratch. And deck heights whatnot would probably make repurposing very expensive.
To some degree, there is sharing of parts and carbodies. Most of the subway systems built around the era of the DC Metro have compatible specifications, most new systems today (and soon Caltrain) use a lot of slightly modified Stadler European rolling stock, etc. And even if not the entire car can just be taken, a lot of the pieces are also modular, but BART doesn't really benefit from any of that.
The bigger issue is that they can't use existing railroad rights of way without expensively converting the gauge or building a whole new set of tracks. A bunch of proposed/new extensions are just regular standard gauge rail to deal with that issue.
If your goal is to make cars that are 38% lighter than other rolling stock then you are not going to want to make use of the parts from the heavier cars.
The issue is that the fleet is so custom that it took over 40 years to replace them, and the replacements had issues since a new company had to do it (the original exited the train business.)
Okay, but for example Stadler's Metro can be configured such that it's lighter than the original BART stock. (I don't believe there's currently any in operation in the US; most of what's in the US is FLIRT/KISS based, which are decidedly heavier stock designed for non-rapid-transit usage.)
Not necessarily. Trucks are one of the most expensive parts of a train car, but can largely be interchangeable. As I understand it, they essentially have to be custom made for the BART.
It plays a massive role for logistics. A compatible gage means you can transport a train by train. It’s quite common to even transport underground cars on tram networks if they exist to bring them closer to the destination. Sometimes tram networks are even completely connected to subway lines to allow trains to go on and off without lifting.
This was the railfan argument for years, but BART replacement cars were priced similarly to standard cars. Modern mass transit cars are built on customized production lines, so there really isn't any economy of scales they are missing out on.
I've used public transportation all over the world (Japan, Singapore, Kuala Lumpur, Shanghai, Hong Kong, Seoul, London, Paris, Rome, Istanbul, Barcelona, Brussels, Antwerp, Berlin, Copenhagen, Amsterdam). Of those, SF's is by far the worst. SF might be #3? in the USA (NYC->Chicago->SF) but it's still arguably crap compared to all those others. The Muni effectively has no schedule. Just show up and pray. There's no way to plan your trip except to go 30 minutes early. The Bart rarely comes since there's effectively one bottleneck track so you have to wait 20-30 minutes for the correct train. The muni actually has the same problem. The Bart is also so loud that I suspect it breaks pretty much every 1st world country's occupational safety hazard rules and is effectively ruining commuter's ears. It's a horrible experience.
In the United States, being the 3rd best city in public transportation is infinitely better than having literally zero reliable public transport because everything is car first. suburbia is quite a miserable experience and it's tragically justified because we don't know what we don't know.
i haven't traveled the world and experienced the worlds best public transit. rather, i've grown up in Los Angeles where we can't conceive that a public transit system could possibly be of any value. "what poor lowly miserable soul can't afford a car? SHAME SHAME on you poor soul"
Anything is better than the suburban denial that is the concrete jungle. I love Bart.
It’s hilarious that LA is building out a ton of track but they put all the stations in highway medians, guaranteeing that it will never be a useful mode of transportation ever.
There are, but if you have to walk 5 minutes through tunnels up and down flights of stairs just to get out of the station, and then another 10 minutes to walk to whatever is actually of interest (since nice things don’t tend to be positioned right next to a noisy highway) then nobody is going to ever use that transit as a preferred mode of transportation. You’re adding 20-30 mins of misery to any transit experience. Even if LA is betting that the areas around stations will get developed because of the presence of the station, they put them in the middle of highways, which guarantees they can’t even develop on that land without destroying the highway or with some engineering heroics. So all in all, it’s a huge waste.
A few minuten to exit the station and a 10 minute walk (who lives next to a station?) do not seem unusual or badly designed to me. Can you park your bike at the station?
You’re doing this wherever you arrive too, and then you’re doing it again when you return. It sucks, as someone who lives in a city that actually values transportation.
So, how would you like it if I said that the nearest parking to both your home and your favorite store is a 15 minute journey away? And the whole time it’d be loud and annoying? Would you still drive there?
You’ll bike to the station, then you’ll need to bring your bike on the train (not sure if they allow this) and then bike again at your destination (or they need to build out share bike infrastructure), and keep in mind that LA is bike theft central so you’ll need to seriously lock down your stuff (unlike the Netherlands or where I live, Japan), and keep in mind that the “highways” I’m referring to aren’t sensible European/Japanese sized highways, but instead 8 lane each way hellscapes. Add the fact that for much of the year it’s 35C+, people already think transit is bullshit there, you’re probably on the train for at least an hour since everything is so sprawled out and there’s no express trains, and like I mentioned, the stations are 100% of the time not even close to anything of interest - no thanks, I’ll stick with a sensibly designed system.
Sure, it isn't ideal, but there is something in between dense urban PT and no PT at all: sprawl goes pretty well with (light) rail combined with busses or bikes.
Crappy PT means they won’t build good PT pretty much ever since they already made the bad capital investment. It’s like sealing the fate of LA to never have good PT.
It's the nature of the problem. In suburbia, station must be further away on average, the difference in density with a major inner city imposes that. A 15 main ride on each end is something that is nothing unusual here, and millions travel that way.
Modern North American transit is designed exclusively for the poors. The time of those people has little or no value so we can save money on getting them to our kitchen, gardens, and factories by spending their time walking more. Besides, it's good exercise and keeps them fit for their lot in life.
All the money we save at the expense of the poors can be put into making the capital works impressive and highly visible so when the taxpayers are driven by (or drive by, some people like that kind of thing) they can behold how much their elected representatives have done for them and get reminded of that next time they vote.
In terms of urban rail transit SF can’t really be considered #3
SF is #5 in total ridership and significantly worse than that when you look at ridership per mile.
I think with a mix of subjective and objective measures I can declare with some confidence that most people would take NYC, Washington DC, Chicago, and Boston over SF’s transit.
But it also brings to mind buying a house with an unusual roof. It ended up giving us tons of problems and costing a ton of money to fix and replace because it required a ton of specialist work.
The roof guy left me with some good advice: "you want a roof that looks like all the other roofs around you".
In the 19th century, two competitors set out to build railways in England. Brunel in the south, Stephenson in the north. Stephenson used what's now called the standard gauge because the axon length was already in use for horse chariots and therefor easily available. Brunel however, set out to find the optimal gauge and settled on a wider one (around 2m). He also invented the black box as a side effect to prove his gauge was better and provided more stable rides. When it became time to connect the two railway systems, they settled on Stephenson's preference for political reasons.
From wikipedia:
Stephenson's gauge was chosen on the grounds that existing lines of this gauge were eight times longer than those of the rival 7 ft or 2,134 mm (later 7 ft 1⁄4 in or 2,140 mm) gauge adopted principally by the Great Western Railway
> I thought Stephenson gauge was chosen because there were a lot more rail miles and rolling stock using that gauge than Brunel gauge.
That was key reason. The other was that 7ft gauge limits how tight your corners can be versus standard gauge. There are other minor disadvantages, such as that because trains are bigger, tunnels need to be wider.
The 7ft gauge trains were probably a lot more smooth though. And that extra ~3ft would make them feel much more spacious. Luxury was a big prestige element on the Great Western.
> He also invented the black box as a side effect to prove his gauge was better and provided more stable rides.
There's an interesting chapter in the memoirs of Charles Babbage (he of the Difference and Analytical Engines) where he describes his experiences with the first railways and how he did some experiments on Brunel's Great Western route and built (one of, or possible the?) first dynamometer cars:
That chapter also contains this priceless bit where after Babbage had only by luck narrowly avoided a head-on collision with Brunel going the other way on the same track, Brunel replies that if Babbage's train had actually been approaching the other way on the same line, "he should have put on all the steam he could command, with a view of driving off the opposite engine by the superior velocity of his own."
I think its wheels are designed flat, where normally rail wheels are angled. This means they're always in full contact and have to slip when going around a bend because trains don't have differentials. Not sure why they designed them flat.
> A wider gauge, the pamphlet notes, would also increase “riding comfort” by enabling the use of wider trains with more room for passengers.
But from some web searches, it looks like BART actually uses standard-width cars, and their hard construction standards (tunnel clearances, etc.) do not allow for more.
So - why did BART go broad-gauge, but chicken out on the width? Wider tunnels obviously cost more...but so the longer sidings, longer station platforms, etc. that long-and-skinny trains need.
In a nutshell, for added stability back when it was thought that BART tracks would follow the lower deck of the Golden Gate Bridge:
> Michael Healy, a BART historian and the former BART Director of Public Affairs, believes the use of the broad gauge “goes back to when BART was planned to go across the lower deck of the Golden Gate Bridge,” he told BART recently.
> “High winds coming through the gate would have required more stability for the aluminum-shelled cars. Thus, the wide gauge, as I recall,” he said.
Why that particular gauge (five-feet-six-inches) and not the more widely-used standard ( four-feet-and-eight-and-a-half inches)?
> The pamphlet does not specify why only the wide gauge of five-feet-six inches was tested and not others. One possibility is that the broad gauge was among the more widely used gauges in other countries. Another theory is that complex calculations were required, thus limiting alternatives that could be practically evaluated at the time.
I mean, if your train car has a width W, why not also make the track gauge also W? That would maximize stability, and you already need the footprint of W cleared and leveled anyway so you might as well just set it at W and not risk anything.
Because there's a only slightly smaller standard track gauge that's well understood and has evolved over 150 years, has suppliers available for your every need, and can interoperate with standard rolling stock. Seeing how BART turned out they would have done well to stick to more standard approaches instead of badly reinventing the basics.
But that’s where the idea comes from that BART is somewhat of a consultant fee driven conspiracy. If u design a bunch of unnecessary engineering from scratch, it’s easier to create more costs and funnel the spending to your friends.
Standards exist for a reason. Almost all rolling stock in the US and much of the world is set to this gauge because it was good compromise. By using a nonstandard gauge, they have forever limited themselves to completely custom rolling stock and related components.
The decision to use wide gauge, however it was it was decided, contributed to BART becoming one of the most expensive rail systems in the US to maintain.
If you look, you’ll be unable to find a metro system in the US that does not have custom rolling stock. BART has unique requirements, but you’ll see that everywhere you look.
Most metro systems have more in common with each other than with BART. Nearly everything on BART is one off. For instance Muni has a bunch of small radius 90° turns so they need to have articulated cars with relatively narrow front/rear ends. So yeah that's custom, and a relatively minor inconvenience at service/repair time.
Compare that to BART where a power surge damaged thyristors on fifty(!) cars and BART struggled to find replacements. Other metro systems use thyristors, but most other systems run 600–800 VDC. On any other system you could probably source from a variety of vendors or operators. BART runs 1000 VDC and thus needs expensive, unique, hard to find, thyristors.
Anyways for a hot moment back in the 30s rail operators standardized on a train design. It's totally doable.
It doesn't seem like the world has much of a standard though, as much as they have a term "standard gauge". I was on a train from China to Mongolia where they lifted each train car, one by one, passengers inside and everything, to change the boogies for a different track gauge so the train could continue.
At the France/Spain border my train changed its gauge mechanically on-the-fly. The tech is there, it's probably some stupid patents that prevent the rest of the world from having this efficiency.
Even India has three gauges with local/intercity trains. Narrow gauge, which is mostly used for mountainous terrain (2% of the total Indian rail network), and meter gauge (3.5% of total network) which is being converted into broad gauge (5' 6").
It's interesting that the the new Marin train[1] is standard gauge, and the connector train in Pittsburg[2] too. Amtrak in the bay area is standard guage, as is Caltrain. Bart is coming to the south bay, but I wonder if more expansions wouldn't be better off going with standard gauge. Bart transfers aren't too bad.
263 comments
[ 2.6 ms ] story [ 207 ms ] threadIn my opinion these kinds of historical mechanical/civil engineering stories are quite interesting. Would love to see more.
the Bay area has intense winds?
Like once every other year a bunch of fences would get demolished lol.
There's the Toronto gauge, the Japanese high speed rail gauge (which is standard, 1435mm, for the world but wider than what was used before in Japan, 1000mm and less) and the Spanish high speed rail gauge (which is also standard, 1435mm, even though they usually use Iberian gauge which is wider - and on this, they have some interesting train designs that are capable of automatically switching between the two).
It was four per day in each direction pre-Covid.
This seems ridiculously low given HS2 in the UK is planned to have 18tph on the core line (to various destinations, but still).
To make matters worse the spainish high speed lines have mandatory reservations so once the train has sold out you are not allowed to board - unlike the UK where you can still board but may have to stand for some of the journey - which is not ideal but at least you don't have to stand around for another hour waiting for the next train anyway.
the next train comes into the station as the previous one is leaving
> The schedule offers trains every 3 to 5 minutes on the most used sections during rush hours. During other traffic periods, intervals range usually from 4 to 12 minutes or up to 20 minutes on outer sections. During late nights, only selected express services are operated and all late-night services usually run every 20 minutes.
https://en.wikipedia.org/wiki/List_of_New_York_City_Subway_s...
I believe that the 1, 2 and 3 lines in NY share the same tracks in the core and can hit about 30tph in the peak
if you're talking about total trains then somewhere like clapham junction is way, way higher (something like 180/hour)
back to stopping services: thameslink core manages 24, and that entire route is several hundred miles long
and crossrail core will do 24 on opening, designed upto 30 (the railway on either end is also several hundred miles long)
Wikipedia suggests the Wenhu line in Taipai was testing upto 50tph, and a post from 2006 on District Dave's forum said Lille Line 2 was at 47tph [0]
Both of those lines are short vehicles though - 52tph. The forum does post that many Moscow lines are 38tph (95 seconds) at long trains (140-160m)
[0] https://districtdavesforum.co.uk/thread/5210/worlds-most-fre...
The existing quad-track line carries 15 trains per hour [2], and is full, but this is because it carries trains of varying speed -- express passenger trains, commuter/regional trains and freight.
(The Victoria Line is a metro line, pretty irrelevant comparison.)
[1] https://en.wikipedia.org/wiki/High_Speed_2#Proposed_service_...
[2] https://www.railforums.co.uk/threads/how-many-trains-run-per...
From December the west coast mainline is bringing in a roughly pre-covid timetable. The railway is a 4 track railway, but looking at just the fast tracks leaving Euston, there will be 13 departures an hour on the fast lines
3tph Manchester 3tph Birmingham fast 2tph Liverpool 1tph Chester and North Wales 1tph Glasgow 1tph Crewe via Trent Valley stops 2tph Birmingham via Northampton (switch from the fast lines to the slow lines after about half an hour)
None of these are high speed - I'm currently on a train to Manchester and my phone says we're doing about 125mph (200km/h). HS2 is a more reasonable 200mph.
To be fair, Accela Washington-New York-Boston is more akin to the London-York-Edinburgh services which are only 2tph, or 31 trains per day (44 run as far as York, but only 31 continue to Edinburgh)
The frequency on the WCML can't be higher as some of those trains are non-stop as far as places like Stoke, Crewe, or even Warrington (about 90-110 minutes out of Euston), where others do stop at places like Watford, Milton Keynes and Rugby.
The full proposed timetable from Dec 2022 is here: https://railforums.co.uk/threads/december-2022-timetable-cha... which might be at https://files.smallpdf.com/files/e54b2d4e9c7a08be02a4be30522...
(In addition to the 5 trains per hour to Birmingham above, there is another one or two from a different London station - Marylebone, and there's a handful of additional trains from Euston to Edinburgh via Birmingham)
The hourly Glasgow departure averages 105mph to Warrington
Once you get up north speeds drop -- the eng to end average for Glasgow-London is about 84mph (401 miles in 4h47), Edinburgh-London is a similar distance but a bit faster at about 4h26, so 89mph.
Washington DC to Boston is 440 miles and seems to take about 6h40 at best, just 66mph
Average over 24 hours (yes including night when there are no trains) is one departure every 43 seconds.
I miss the trains in other countries so much. They’re reminders of what we could have had here had we not let NIMBYism win. https://www.quora.com/Why-wasnt-BART-built-initially-to-loop... has some of the history, there’s probably a better article out there somewhere.
Here in the Denver metro area, we used to have trains running all over the place, and 100 years ago, there were daily trains between Denver, Boulder, Golden, and Longmont. Then they ended service and tore up most of the tracks.
Fast-forward to 2022, and now they're trying again to put rail between Boulder and Denver, but it's a $1+ billion boondoggle that hasn't gone anywhere in 15 years. [2]
[1] https://www.youtube.com/watch?v=p-I8GDklsN4 [2] https://www.cpr.org/2022/03/01/rtd-fastracks-commuter-rail-b...
These are all intercity trains to cities north of London.
There will be 6 trains per hour to Birmingham (120 miles from London), three of which continue beyond Birmingham. An additional 11 trains don't stop at Birmingham, but cities beyond it.
[1] https://en.wikipedia.org/wiki/High_Speed_2#Proposed_service_... and https://en.wikipedia.org/wiki/High_Speed_2#/media/File:HS2_v...
Destination | Operator | Distance from NYC
Camden / Philly | NJT / Amtrak | 82 mi
Montauk | LIRR / Amtrak | 110 mi
New Haven | MNR / Amtrak | 67 mi
Except for Montauk, these places are cities in their own right, but we don't really think of them as intercity trains because the primary reason someone would board is as a commuter not as a traveler.
The planned schedule for the new line in England will have the shortest train make only these stops, 3 times per hour:
And a longer one (also 3 times per hour) will make only these stops: That is nothing like a regional/commuter train. The existing regional train to Birmingham stops at Watford, Milton Keynes, Wolverton, Rugby, Coventry, Canley, Tile Hill, Berkswell, Hampton-in-Arden and Birmingham — that is like the Philly/LIRR/MNR trains.For the time being BART and Muni run 2–3 minute headways during peak hours (so somewhere around 25–30 tph). One of the big things BART remedied with the new cars was the lack of doors (which slows down boarding and increases dwell time).
In Spain there are several railway companies: RENFE, Ouigo and (TBD) Irio. There are enough trains for the current demand. If they detect a surge (holidays for example) new trains are added to the timetable.
Short-distance trains do indeed allow you to be standing up but the travel time is at most 1h.
There are not enough trains in Spain for the cost of the high speed infrastructure. Looking today both the 19.30 and 20.40 MAD BCN are sold out, which means you would have to wait until the next day to travel if you couldn't get to the station before 19.00 (which isn't uncommon).
When high demand, extra trains are added.
Before COVID-19, there were more than 40 trains a day from Madrid to different destinations (Sevilla, Cordoba, Barcelona), and over 20 to many others, so the difference is not that large.
But then also, London is around three times larger and has a bigger weight in its country economy, than Madrid. It just makes sense that more people commute or travel to London anyway.
In the short-distance trains you can get into the train only by passing a card through a rfid reader. Maybe you should compare this trains to UK ones.
Thus, Spain has less population than the UK, is less centralized than UK (where almost everything is in London) and Spain is much more mountainous. We can’t really compare these two countries.
why wouldn't it be?
Edit: keep in mind that the Madrid-Barcelona high speed railway line reaches almost 250kph. That’s more than the double maximum speed of a car in Spain.
I’d say that having to buy a ticket with an assigned seat in high-speed trains in Spain has been always been that way.
However, short-route trains (Cercanías) don’t have assigned seat, but they are not high-speed trains.
It will be to keep the quality of the service high, reduce boarding times, or something like that.
Lastly, there isn’t much freight on the line, because (besides the high track fees) there isn’t much of a standard gauge freight rail network in Spain.
In some sense the Perpignan Figueras line was supposed to be like the channel tunnel, but it really isn’t. It’s not exactly comparable to the rest of the Spanish hsr network - which is arguably overbuilt for political reasons to get hsr into many even smaller cities, and perhaps under-utilised at this point, but it’s not as bad as that one line might suggest when picked as a representative.
https://en.m.wikipedia.org/wiki/Track_gauge_in_Ireland
7-ft 1/4 inch (2,140 mm) gauge railway - Isambard Kingdom Brunel : https://en.m.wikipedia.org/wiki/7-foot_gauge_railway
Great Western Railway : https://en.m.wikipedia.org/wiki/Great_Western_Railway
List of 7-foot gauge railway locomotive names : https://en.m.wikipedia.org/wiki/List_of_7-foot_gauge_railway...
British narrow-gauge railways : https://en.m.wikipedia.org/wiki/British_narrow-gauge_railway...
You're confusing two different topics.
Freight railroads cannot "demand" to use rail lines owned by passenger railroads. Passenger railroads can, if they choose, allow freight lines to use their tracks, and vice-versa.
You can see examples of both in Chicago. CTA, a passenger railroad, used to carry freight traffic, but chooses not to do so anymore. Meanwhile, the local freight railroad companies choose to allow Metra, another passenger railroad, to use their tracks and rights-of-way.
It's not anywhere close to the black-and-white situation you present.
Where passenger railroads own their tracks, they get to decide what runs on them. The problem is that passenger railroads in the United States own very little actual track, and rely too heavily on using freight-owned tracks.
There's lots of railroad magazines that write about this ongoing problem every month.
And no, the local freight railroads aren't "choosing" to let Metra be. They pretty much have to. That was part of the agreement when the regulators allowed the railroads to drop passenger services (which they were previously legally obliged to keep). There's a process, which hasn't been put into practice for a while but is still on the books, for Amtrak to demand the right to use certain tracks (the exact scenario where this happens escapes me).
You should tell the CTA. It will be news to them.
In the US, railroads had to run both freight and passengers or else leave money on the table, 153 years ago, when the first transcontinental line was finished. And if they wanted to advertise speed to passengers, the freight trains had to yield.
When Amtrak started of course things changed a lot, immediately and also gradually, but that was 1971, 52 years ago.
If they didn't wanted to reinvent the wheel, they could have adopted another standard railway track gauge: the Iberian gauge
https://en.wikipedia.org/wiki/Iberian-gauge_railways
There's a fun story about BART: the US decided to go with a blank slate approach to railway design by hiring not railway engineers but aerospace engineers, and consequently the project relied heavily on reinventing the wheel. This track gauge thing happens to be one of many examples.
As BART decided to forego any standard way of doing mundane railway tracks, they had to endure decade-long delays and huge budget overruns to cover up their loss. Otherwise there would be nothing to talk about a mundane rapid transit system which already covers pretty much the whole Europe.
I’d like to know why the tracks are flat and not angled and require regular grinding down to even out the washboard effect this causes… which results in that high pitched sound whenever it enters a curve due to slippage.
End result is more surface contact = more noise/squealing, also less self-centering of the wheels on the tracks, so the lips also brush against the rails. I think this also exacerbates the washboarding of the rails.
https://sf.streetsblog.org/2018/06/08/bart-makes-headway-aga...
Is this it? I found it in my YouTube history.
Or this one, which superimposes free body diagram (of forces) on the wheels to show how it is self stabilizing? https://youtu.be/XzgryPhtc1Y
Normally you'd increase the spacing a bit in curves to encourages the wheels to rotate at different speeds instead of destroying the track. Unfortunately even with the new profile BART is still incredibly loud, and it's not helped at all by low speed curves like the Oakland Wye.
I think the term you're looking for is track gauge.
And as for a need to widen it in curves, the answer is basically it depends.
On mainline railways gauge widening in curves basically stems from the days of steam locomotives, which mostly didn't have bogies (at least for the main driving wheels) and comparatively enormous wheelbases.
On straight(ish) track, some minor gauge widening as compared to the nominal standard gauge might nowadays also be beneficial, especially on high speed lines.
Within the context of streetcars/trams, which often need to negotiate very tight curves on the other hand, current state of the art seems to indicate that a slight gauge tightening might actually be more beneficial in order to avoid too large a skew angle between the bogies and the rails. (Compare http://interfacejournal.com/archives/472)
I'm no wheel/rail-interface guru, so quite what the conclusions from that would be for BART I can't say, either, but just wanted to point out that the topic is somewhat more complex, and nowadays gauge widening in curves might not actually always be the right answer.
No. Track gauge is the nominal width, no? I'm talking about how the width changes in the curves. A while back a former BART engineer went on the record claiming the spacing in the curves was contributing to the noise.
Top to bottom BART fully embraced NIH. Fifty years ago there was good reason but now the rest of the rail industry has moved on and BART is stuck with a lot of tech and engineering debt. For e.g. BART can't grease or sand their tracks either, and you can feel the older trains slip and slide (and damage the track) as they exit the station.
Edit: Here's an interesting piece on noise. Note how nearly every other rail system lubricates the tracks (BART can't) and how no other rail system uses (noisy) solid aluminum wheels.
https://www.bart.gov/sites/default/files/docs/noisereport.pd...
I'm not intimately familiar with the English terms, but I think it still all boils down to the track gauge. That Boston article I linked to certainly talks about track gauge and gauge-widening and the like.
Should be, showing from the axel to the rail:
As the car rounds turns, the inner wheel rides down and the outer wheel rides up, increasing the relative rotational radius of the outer wheel. This compensates for the fact that rail wheelsets rotate on a single fixed axis, not independently, and without a differential.For a non-ASCII image see: https://steemitimages.com/DQmUjS7PbhzAYpkTi5A3c2QQQa7jn1c52s...
Longer explanation: https://www.popularmechanics.com/science/a25581/science-behi...
Wikipedia also discusses geometry: https://en.wikipedia.org/wiki/Train_wheel
Does it make it more complicated? Or there is a disadvantage?
https://connectorsupplier.com/regenerative-braking-systems-r...
A solid axel allows placing an external bearings between the truck and the axel.
Hub bearings are internal and service would require dealing with the nontrivial weight of a loose wheel.
See https://en.m.wikipedia.org/wiki/List_of_railroad_truck_parts
As the wheels move off centre, the bevel makes one wheel have a larger effective diameter and the other wheel a smaller effective diameter. Because the wheels have a solid axel, that forces one wheel to start travelling slightly faster than the other, causing the pair of wheels to turn.
With the bevel making the wheels larger on the inside, it means that as a pair of wheels drifts right relative to the tracks, the right hand wheel rides up, starts behaving as if it’s larger, and then starts to overtake the left wheel (which has ridden down, and behaves as if it’s smaller), creating a turning effect to the left, correcting the rightward drift.
When setup correctly, and travelling at a reasonable speed, the big metal lip on train wheels should never touch the track. The self-centering effect does all the work, the lip is just a backup to deal with overly tight turns, points and extremely slow running.
Or are you saying the whole setup actually tilts the carriage into the turn - helping to prevent it from tipping?
Weight and load have no impact of the effect. The only requirement is that the wheels remain adhered to the track, and don’t slip. Which is expected anyway, because it would be tricky to accelerate and decelerate a train with slipping wheels.
With independent wheels you have no self-centering effect, so only think keeping your wheels point in the right direction would be inner flange on the wheels colliding with the track. Such collisions obvious involve quite a bit of friction, noise and track/wheel wear.
Not to mention independent wheels are more complex to build. Two wheels directly attached to a single fixed axel is about a simple as you can get. You only need two couplings between your single axel and non-rotating part of your train. With independent wheels you need four, or more complex wheel designs to allow bearings at the wheel/axel coupling.
There no need to tip carriages into turns to prevent tipping, gravity, track banking, and speed limits already assure that tipping isn’t an issue. Tilting trains only exist to increase passenger comfort, not safety. Tilting trains are prefect capable of operating a full line speed with their tilting mechanism disabled or jammed, it’ll just be uncomfortable for the passengers.
That isn’t true. No train network is going to allow train to operate at speed, if the safety of those trains depends on non-failure of a complex tilting mechanism. Having a tilting mechanism fail or jam before a corner (which has happened) can not result in a high speed derailment. That risk is simply unacceptable on modern railway.
Tilting trains make it possible to run at higher speeds on older lines by allowing older lines to have the camber on corners increased to accommodate the higher speeds. But the camber can’t increased enough to hit optimal amount of camber without make the rail unusable to older trains.
Tilting trains provide a useful compromise. Increase the camber enough for stability (but not comfort), while keeping the camber acceptable for older trains, then make up the rest of the tilt with carriage tilting mechanism. The total amount of tilt is then comfortable for passengers. There is a minor safety element provided by tilting trains, in that lateral forces in cornering are lower, thus passengers are less likely to fall, and items won’t come of tables. But the tilting mechanism certainly is a safety critical element to running faster on old lines.
It’s also worth mentioning that carriages on passenger trains have a pretty low center of gravity, as all the heavy equipment (motors, bogies, breaks, suspension) sits below the passenger compartment. As a result tilting a passenger train won’t significantly move its center of gravity, so impact on total train stability is going to pretty minor.
Tilting a carriage isn't going to adjust the CoG by any appreciable amount. It will tend to make the net centripetal moment more virtical from the reference frame of passengers, however.
Wikipedia's article corroborates this:
As a train (or other vehicle) rounds a curve at speed, objects inside the train experience centrifugal force. This can cause packages to slide about or seated passengers to feel squashed by the outboard armrest, and standing passengers to lose their balance. Tilting trains are designed to counteract this by tilting the carriages towards the inside of the curve, thus compensating for the g-force.
https://en.wikipedia.org/wiki/Tilting_train
For a good explainer see: https://yewtu.be/watch?v=SRsm7mv0Oh8
With separted axles, you wouldn't get the climbing behaviour of the outside wheel. With a fixed axle, the idealised shape of the wheels is of a double-ended cone (see the video for an example) which can ride higher on the outside of a turn. Even with conical wheels, with independent axels there's no way to impart a turning force on the carriage. I'm not entirely sure what the end result would be, but I suspect it would be a far harsher ride, and much more track and wheel wear. Probably a much higher probability of derailments as well.
Cone angle varies by type of train and speeds. The video shows both freight and high-speed rail designs and discusses characteristics of both.
This is what causes a lot of noise and wear and tear. I actually researched this and it seems that it was not as stupid of a decision as a lot of people say it is. There is a problem with beveled wheels -- it is called hunting. This is when the car oscillates from side to side. Hunting is unpleasant, also causes wear and tear and noise and creates a danger of derailing.
Hunting becomes more likely if the car is lighter and if it is going high speed. The bart was designed to be lighter than most existing metro systems and to go much faster than existing metro systems in certain corridors. So they wanted to prevent hunting and thus made a portion of the wheel cylindrical.
So it seems that they just made another engineering trade off. And I am sure they had justification. But I do wish that when engineers make these trade offs they take human comfort into account. I does not seem that they gave sound much of a weight when making their trade offs.
I think the solution is to keep the wheels cylindrical but just decouple them. I.e., not to have the right and left wheel connected by an axle. If they are independent they can move at different speeds on the turns and will not cause grinding and noise. And the wheels that are driven may be driven by different electric motors that drive the wheels by the necessary different speeds on turns. This is called vectoring and is already being done for some electric cars.
The DLR has very tight corners, and was originally built on very tight budget. As a result it uses very steeply bevelled wheels to handle the corners.
The wheels do a great job on the corners, but on the straight they suffer from significant hunting, resulting in the trains wobbling side-to-side (and slightly up-and-down), at a rather alarming frequency and amplitude. Not quite enough to make you sea sick, but it’s pretty close.
I suspect the wheels have high conicity. The rolling stock is also likely lighter than a heavy-rail system.
https://yewtu.be/watch?v=i1pAfKsL1bk
(Large stretches of the original network built in the 1980s ran on elevated track, supported on pylons over ground that's a mixture of chalk and clay and tends to waterlog. Heavy rail can be built over and under that part of the Thames, but it has its own special challenges and the DLR was built as the cheapest way of getting commuters in and out of the Canary Wharf office towers, from Tower Gateway and the actual London Underground network to the west. Then from the 1990s onwards mission creep set in ...)
It's also something of a two-edged sword: the property value boost makes further transit rights of way far more expensive to acquire. I've noticed in particular that much rail development happens either as greenfield projects within countries as they're first industrialising (UK and US in the 19th century, China in the late 20th), or following a devastating war or other economic setback --- Japan's shinkansen and to a somewhat lesser degree France's TGV are both legacies of WWII.
In the US, aside from political opposition from other parts of the transportation sector, it's wealthy landowners who are the most significant obstacle to new development.
Thanks for the, er, bedrock explanation as well.
https://en.m.wikipedia.org/wiki/Docklands_Light_Railway_roll...
Here’s good video about the DLR time stamped to point that demonstrates the hunting well.
https://youtu.be/_WosYRVYMEM?t=907
The video confirms what I'd surmised: that tight turns were the reason for more conical wheels. The sections I watched didn't actually show the wheel geometry, though, and I've not been able to find any photos online (DDG and Google Image search, various combinations of "DLR | Docklands Light Railroad | (conical|cone|conicity) wheel"). The WIkipeidia DLR Rolling Stock article mentions the conicity and hunting issues, but doesn't show the wheels either:
https://en.wikipedia.org/wiki/Docklands_Light_Railway_rollin...
BART's since adopted a very-slightly-beveled wheel design pictured here:
https://sf.streetsblog.org/wp-content/uploads/sites/3/2016/0...
Story:
https://sf.streetsblog.org/2016/04/07/new-bart-cars-show-age...
Note that that's as of 2016, six years ago.
The hunting phenomenon is real and has some interesting consequences. For freight trains, empty boxcars have a lower speed limit than fully-laden ones, as the hunting behaviour is greater with less weight.
High-speed rail systems required understanding and managing hunting action. Getting a case of the wobblies at 300 kph is an exciting but short ride.
Links a 2016 article.
The best you can try is add some sort of lubrication, but even that won't fully quieten it down, since you can only lubricate the flanges, but not the running surface for hopefully obvious reasons.
In short there is no reason. It was a mistake.
Some more historical research here:
https://www.kalw.org/show/crosscurrents/2016-07-05/hey-area-...
Seems highly likely that keeping freight trains off the tracks also factored into the decision. Huge shame that they didn't consider the ongoing maintenance cost of using a non-standard gauge.
Truly, help me understand how to better navigate when to defer and when to have the courage to invent here.
In the case of trains they could be a lot cheaper if everyone agreed one one standard and only made changes when everyone agrees to upgrade at once (over 40 years counts as at once,so long as there is agreement). All the different standards mean manufactures have to have more expensive jigs, and also cannot optimize their lines as much.
Though wheel gauge within limits that Bart fits is not too big a deal to adjust on a line.
Not really any ideal solution here, if the standard isn't suitable, you need something nonstandard.
BART has actually built or planned to build standard gauge since then to reduce costs on future extensions. Some extensions, like BART absorbing Caltrain to go to San Jose, have failed due to the cost.
Then again, the original choice to use the broader TTC gauge was also petty political.
The point is, it's not an economic or technical decision.
Now imagine that Metrolinx's GO trains also used a non-standard gauge? Not only could they not share track with freight trains, or even Via Rail, but simply delivering the train cars from where they are manufactured in Thunder Bay would be a nightmare. As it is, they're just attached to a freight train and pulled along like any other car, and I've even seen them on a CP freight train.
Standards also mean you can buy standard parts instead of getting something custom made. This can greatly reduce costs as mass manufacturing is only cheap when there is a mass needed. (At least Bart choose a guage in use elsewhere so there is some demand for those parts).
Standard means CA HSR or Caltrain could share less used track which can save everyone money. (It makes scheduling hard, but that is a lot cheaper than building track) who knows, maybe freight rail would use that track for some low demand customers they can't serve at all now, and trains are more environmentally friendly than trucks.
If you invent a competitor to microSD that somehow stores 50% more data in the same space(For example only, just ignore the fact that form factor is independent of the flash storage tech), I would be annoyed.
If you push to make it the global standard and succeed, I'll still be annoyed but appreciate the progress.
If you just make a line of devices with unique storage, I'll probably be like, yeah, but MicroSD was fine and this doesn't work with my readers, and now I have to buy a new card.
In software NIH often doesn't offer much benefit at all. It's usually just "We want to build it ourselves and really understand it". Sometimes features are even worse("We don't need to support importing those formats, build something smaller that only support one"). Sometimes performance is worse ("Who needs a million line library with crazy vector instructions? 5 lines of C will do").
https://www.youtube.com/watch?v=taJ4MFCxiuo
Progress involves making the best changes we can. NIH syndrome means re-litigating small changes that aren't impactful rather than focusing on the new big ticket items that will really make a difference.
What's the point in optimizing something that will have a negligible return when there are better opportunities? And there's also the important downside of NIH syndrome. It might just be that the people who solved the problem initially knew what they were doing and you'll discover years down the road that your solution doesn't help anyone make progress, it was actually a terrible mistake.
Like the BART solution. The people who did this were aerospace engineers. They thought they were improving on train design. Well, they were not. They were ignoring important constraints you only learn about later: the fact that interoperability between train tracks is critical to long-term development and efficient operation. Extending BART is a nightmare, it cannot share track with other systems, you cannot move trains on other systems to get them to BART which makes it far more expensive to do anything, the parts are incredibly expensive now that the supply chain for them is tiny.
Basically every commuter suffers every day because of this idiotic decision which keeps the system fragmented.
And that's what NIH syndrome gets you. You think you're making progress. But you often just have a terrible understanding of the problem and you're building yourself long-term technical debt.
Arguments like "the best code is no code" basically allude to the idea of defining the problem in a way that allows a solution without addressing it through some complex technical dependency.
In transportation there is a lot of popular discussion based around different modalities like car, train, bike. Each one has its enthusiasts arguing for a mode being a solution. This is a very engaging topic because it creates the grounds for an ideology: "everything will be great once we eliminate x and build more y". A great way of addressing transport issues while not discussing the overall problem in terms of land use, tripmaking needs, accessibility, safety, etc. The path to the actual solution is through studying all those other things to identify what key metrics are in play and then specifying your solution in terms of "must meet this metric". That's where the meat of any problem lies, because the person who lays down the finishing touches is building on all that preparatory work - supply chains, scientific principles, engineering tolerances and so on.
And, of course, there's another path into addressing without solving by selecting unrealistic metrics that don't suggest how to move forward. Before you even have the metrics you need good critical thinking skills, ability to collect and analyze data, and some resulting media that communicates the findings like a document. That's a whole job in itself, and then selling and marketing it is another. And then maybe we get to actually implementing something.
It takes a lot, in all. It's almost imperceptible how we progress on big issues. And the whole way through someone is trying to steal credit for some part of it.
Don't know what NIH syndrome has to do with it, doesn't sound like interoperating with other rail systems was in the spec.
They also entered busmaking at this time and produced the Flxible 870, a notorious lemon of a bus.
Standard gauge was ubiquitous in the US at the time, and this engineering firm didn’t even evaluate it in their report, despite the fact that the original BART plans called for standard gauge. Stability of standard gauge has since been proven to be a complete non-issue.
If you read more about the history of BART, everything about it was NIH. They thought they were going to design some new space-age standard and usher in a new era of public transportation.
It's like, ok, unique decision on the broad gauge. But the Bart is a piece of shit. So why should I care?
A real topical blog post would be one comparing Bart to say the train systems in China; but instead they ignore the obvious and focus on this irrelevant detail.
The SF Bay Area has a density of 431 people per square kilometer.
Obviously it's easier and much more efficient to build rapid transit when you are serving areas fifteen times more populated.
https://youtu.be/MnyeRlMsTgI?t=50
And just because the bay area average is 431 there are plenty of cities in the bay area with much higher densities and there are comparable cities else where that manage to have great public transportation.
There's also VTA, which guessed completely wrong about where people would move to and so goes from MTV to nowhere.
Absorbent wool was, as you might imagine, unbelievably nasty. Among other things I saw on BART, I once saw a guy with severe flesh necrosis on his foot sit with said foot, bandages taken off, pressed against the seat. Of course, minor issues like diarrhea and vomit stains were more typical.
As why carpets and wool, the answer is actually just about as interesting as the linked article. Back in the 60s / 70s when BART started operating, it was one of the first metro subways. Its creators envisioned it as a futuristic luxury ride, whisking businesspeople and tourists in and out of far-flung suburbs to the glimmering high-rises of San Francisco. Everyone would be wearing a suit or a nice dress, and they would sigh in relief each time they daintily sat down in these airline-style seats.
The actual future turned out to be a bit more disappointing...
That's a pretty strong absolute; I'm certain that, at the very least, folk involved with public transportation do regularly consider homeless and/or addicts when covering various aspects of their service. Maybe not so much in the past, but these days it's basically a given.
It’s not being an actual circle line means it’s not as attractive to homeless people. You can’t just fall asleep and just ride in circles, you get up periodically and change trains now.
Homeless people are more likely to be on buses (it's cheaper) but even that isn't common.
Sweat, rain, dirt etc. are all going to come into the train.
What does that mean? It wasn't the first in anything by any stretch of the imagination.
In short, it was supposed to be a revolutionary metro, not just "yet another metro system". As it turned out, they would have been much better served by copying what was tried and true.
https://images.squarespace-cdn.com/content/v1/56200fcee4b020...
Uhh. No. London had the Underground literally 100 years before this and Boston had a subway around 1900. Paris, Moscow.. many subways predate BART.
Even local transit, e.g., BVG frequently uses fabric as seat material.
You just have to design it to be able to be cleaned easily and quickly.
I think they plan on doing something similar with the Valley Link project between Dublin/Pleasanton to Lathorp in the San Joaquin Valley.
As the posting concedes, “The pamphlet does not specify why only the wide gauge of five-feet-six inches was tested and not others.”
I do wonder how much that plays a role. I suspect not much. Bombardier or Kawasaki or whoever is going to mill everything from scratch. And deck heights whatnot would probably make repurposing very expensive.
Build for what makes sense.
The bigger issue is that they can't use existing railroad rights of way without expensively converting the gauge or building a whole new set of tracks. A bunch of proposed/new extensions are just regular standard gauge rail to deal with that issue.
i haven't traveled the world and experienced the worlds best public transit. rather, i've grown up in Los Angeles where we can't conceive that a public transit system could possibly be of any value. "what poor lowly miserable soul can't afford a car? SHAME SHAME on you poor soul"
Anything is better than the suburban denial that is the concrete jungle. I love Bart.
So, how would you like it if I said that the nearest parking to both your home and your favorite store is a 15 minute journey away? And the whole time it’d be loud and annoying? Would you still drive there?
All the money we save at the expense of the poors can be put into making the capital works impressive and highly visible so when the taxpayers are driven by (or drive by, some people like that kind of thing) they can behold how much their elected representatives have done for them and get reminded of that next time they vote.
https://www.amazon.com/This-Pacific-Electric-Stephanie-Edwar...
SF is #5 in total ridership and significantly worse than that when you look at ridership per mile.
I think with a mix of subjective and objective measures I can declare with some confidence that most people would take NYC, Washington DC, Chicago, and Boston over SF’s transit.
https://en.wikipedia.org/wiki/List_of_United_States_rapid_tr...
But it also brings to mind buying a house with an unusual roof. It ended up giving us tons of problems and costing a ton of money to fix and replace because it required a ton of specialist work.
The roof guy left me with some good advice: "you want a roof that looks like all the other roofs around you".
In the end, it seems it's good enough. Both high speed trains as well as the heaviest freight trains ever work fine with standard gauge.
That was key reason. The other was that 7ft gauge limits how tight your corners can be versus standard gauge. There are other minor disadvantages, such as that because trains are bigger, tunnels need to be wider.
The 7ft gauge trains were probably a lot more smooth though. And that extra ~3ft would make them feel much more spacious. Luxury was a big prestige element on the Great Western.
There's an interesting chapter in the memoirs of Charles Babbage (he of the Difference and Analytical Engines) where he describes his experiences with the first railways and how he did some experiments on Brunel's Great Western route and built (one of, or possible the?) first dynamometer cars:
https://en.wikisource.org/wiki/Passages_from_the_Life_of_a_P...
That chapter also contains this priceless bit where after Babbage had only by luck narrowly avoided a head-on collision with Brunel going the other way on the same track, Brunel replies that if Babbage's train had actually been approaching the other way on the same line, "he should have put on all the steam he could command, with a view of driving off the opposite engine by the superior velocity of his own."
But from some web searches, it looks like BART actually uses standard-width cars, and their hard construction standards (tunnel clearances, etc.) do not allow for more.
So - why did BART go broad-gauge, but chicken out on the width? Wider tunnels obviously cost more...but so the longer sidings, longer station platforms, etc. that long-and-skinny trains need.
https://www.csx.com/index.cfm/customers/resources/equipment/...
https://www.bart.gov/about/history/facts
https://en.wikipedia.org/wiki/Superliner_%28railcar%29
The 9'6" width in the CSX data is inside widths on freight cars.
The Golden Gate Bridge has a lower deck?
edit: Looks like it was only planned but never came to together https://www.sfgate.com/bayarea/article/Golden-Gate-Bridge-lo...
> Michael Healy, a BART historian and the former BART Director of Public Affairs, believes the use of the broad gauge “goes back to when BART was planned to go across the lower deck of the Golden Gate Bridge,” he told BART recently.
> “High winds coming through the gate would have required more stability for the aluminum-shelled cars. Thus, the wide gauge, as I recall,” he said.
Why that particular gauge (five-feet-six-inches) and not the more widely-used standard ( four-feet-and-eight-and-a-half inches)?
> The pamphlet does not specify why only the wide gauge of five-feet-six inches was tested and not others. One possibility is that the broad gauge was among the more widely used gauges in other countries. Another theory is that complex calculations were required, thus limiting alternatives that could be practically evaluated at the time.
The decision to use wide gauge, however it was it was decided, contributed to BART becoming one of the most expensive rail systems in the US to maintain.
Compare that to BART where a power surge damaged thyristors on fifty(!) cars and BART struggled to find replacements. Other metro systems use thyristors, but most other systems run 600–800 VDC. On any other system you could probably source from a variety of vendors or operators. BART runs 1000 VDC and thus needs expensive, unique, hard to find, thyristors.
Anyways for a hot moment back in the 30s rail operators standardized on a train design. It's totally doable.
https://en.wikipedia.org/wiki/PCC_streetcar
At the France/Spain border my train changed its gauge mechanically on-the-fly. The tech is there, it's probably some stupid patents that prevent the rest of the world from having this efficiency.
It's interesting that the the new Marin train[1] is standard gauge, and the connector train in Pittsburg[2] too. Amtrak in the bay area is standard guage, as is Caltrain. Bart is coming to the south bay, but I wonder if more expansions wouldn't be better off going with standard gauge. Bart transfers aren't too bad.
[1] https://www.sonomamarintrain.org/
[2] https://www.bart.gov/about/projects/ecc
https://en.wikipedia.org/wiki/Pittsburgh_Light_Rail