Isn't the Transit way above 93 North? This is basically a two level tunnel where the floor of the transit way is the ceiling of 93 N below, correct? You can see from the linked street view that the ceiling of the 93 N tunnel appears to be the underside of the concrete structural slab. (This is different than the I-90 tunnel where the ceiling is suspended below the roof structure.)
If after the end of the existing transit way the roof of 93 N (floor of an extended transit way) changes construction to something that can not take the load of an extended transit way then yes the transit way tunnel can not be extended without great cost.
If that is not the case the transit way seems like it could ascend up the surface along with 93 N.
Thanks for pointing that out. I did get it that backwards. Doesn't change the infeasibility, however.
To clarify:
93 mainline is underneath the Transitway. End of the loop is where Transitway floor and 93 mainline's ceiling meet, corner of Summer and East St.
The onramp from the first divider peg out to the portal is on significantly steeper grade than the 93 mainline because its portal is much closer. So there's a 'hump' in the roof until that same East St. intersection spot where the divider wall ends. That accounts for the double-thick wall.
Now here's the math that serves up the same bad news as before.
Tunnel width:
- 10 ft. between track centerline and wall. (source: GLX design specs)
- 6 ft. thick walls (source: Mass Pike tunnel specs, http://www.concreteconstruction.net/Images/Concrete Boxes_tcm45-591056.pdf). Since this is same soil and depth, and the Pike walls are waterproofed where the leaky 93 walls are not, these walls will be same thickness to achieve same waterproofing.
- A center dividing wall between the tracks, since the tunnel will be supporting the commuter rail tracks/platforms above and the plaza in at the Atlantic Ave. entrance. Can be thinner than the side walls, but because of proximity of the nearest CR track to the Transitway hook-in I doubt it's any less than 3 ft. wide.
Tunnel width maths:
- (10 ft. centerline clearance x 2) = each track bore 20 ft. wide
- x (2 track bores) = 40 ft.
- + (6 ft. side walls x 2 = 12 ft.) = 52 ft.
- + (3 ft. center wall) = 55 ft.
Your tunnel is 55 ft. wide. Tangent track can get squeezed by 2 ft. on the centerline clearances per the design manual, but this is a curved tunnel so that's not in-play. 55 ft. or bust.
The loop is a double-loop--"inner" and "outer"--surrounded by a small center area arranged with support pegs and utility room. Inner loop is the minimum turn radius of a 60-footer, outer loop somewhat wider; loops are separated by pavement stripes. Only picture of it I was able to find, from Opening Day in 2004:
https://sery2831.smugmug.com/Transit/Silver-Line-Opening/i-3pCHt8D/A.
Insertion point for the trolley tunnel has to stay
clear of the loops if you have any intention of keeping SL1 (and any other present/future) bus service in the Transitway. Safe assumption that it's a dealbreaker if you're kicking the buses entirely out.
This creates a big problem. You can't ram the trolleys into the middle of the loop and follow the contours of the NE quadrant of the loop back onto the mainline (i.e. take a chunk out so it's a "two-thirds" loop / "one-third" mainline). The least-invasive means of construction ends up creating several incredibly dangerous safety issues for the transit vehicles:
- Bus drivers on the loop can't see to the right because of the curving wall. Outbound trolleys would be merging in at oblique angle mid-loop, with dividing wall ending too abruptly for a bus driver to see them. Extreme sideswipe risk.
- Bus drivers on the loop have obstructed view to the left because of the center supports + utilities and can't see the mainline ahead. Inbound trolleys would be crossing wrong-way across the NE quadrant of the very tight inner loop. Extreme head-on collision risk...and that's an instant bus driver fatality given the weight differential.
- You can't close off the inner loop and just leave the slightly less-dangerous outer loop because there has to be a place to stuff out-of-service buses and has to be a way around a bus disabled on the loop.
- Bus drivers slamming brakes on the loop to avoid a trolley they didn't see are at severe risk of jackknifing the 60-footers inside an enclosed/unventilated tunnel. Unacceptable risk for emergency personnel access and fire suppression.
- Can't install wayside traffic light signals mid-loop because of the obstructed sightlines and drivers paying attention to staying within their lane markers not the lights on the wall. A signal turning red at the last moment would be too easily missed, and the jackknife risk makes any sudden braking inherently dangerous.
- No potential for an auto-stop signal system in the Transitway like there is on Red/Orange/Blue and most likely will be installed in the future on Green. Impossible to co-mingle modes that way, so it's wayside signals + line-of-sight between the merging point and SL Way. Even if you could install collision avoidance on the trolleys alone, it makes no difference because the buses won't have it and they are the ones dealing with the most acute sightline dangers (see also: sudden involuntary brake application on bus rounding the loop = jackknife).
^There's no give with these kinds of safety considerations.^ It won't be approved for build if the lowest-profile insertion point makes the loop unsafe for buses.
Therefore, you have to insert
after the loop has almost completely straightened and there's adequate sight distance for bus drivers coming off the loop to see if they've got a green or red signal where the trolley tracks merge. The notch in the wall for Phase III up Essex is roughly at this spot (depicted in the Blue Book, p. 46 as the "Bellmouth" on the Transitway wire schematic map), if you want a very rough indication of what those sightlines need to be. You also have to insert after the tunnel has narrowed enough that you can end the dividing wall between the loop and the trolley tunnel and have one cavern for the rest of the merge, because--recall--you're supporting the plaza and commuter rail entrance upstairs and are a lot closer to the surface than 93N.
To get a sense of where this insertion point has to be, see the tunnel diagram a few posts up. Tilt your head at the loop so the mainline towards SS is pointing at the 12:00 position. Your insertion point for the trolley tunnel is probably no lower than the 2:00 position. That corresponds on the surface with the curb across the street from the Essex Hotel front door, along the taxi stand parking row.
At that spot you have a less-than 50 ft. radius between the SW cornerstone of the historic South Station building and that 2:00 spot on the loop, so you've run out of room before that 55 ft. wide trolley tunnel can thin out and shed its inbound-side and center walls for the merge into the Transitway cavern. It's close, but not close enough to build to-spec while avoiding catastrophic building impacts. Even if that radius estimate is on the low end and you do wind up with a very snug-fitting merge of your your 55 ft. wide trolley tunnel, you are not going to be able to dig up to the last inch of the SS cornerstone. It's a 120-year old granite foundation. Several feet of buffering is mandatory (no less than 5, but probably more like 10) if you don't want to crack the SS foundation and walls like the Copley station elevator shaft cracked Old South Church from a distance of 15 ft. away. Several feet of buffering gets you right back where you started: below the minimum allowed tunnel width for making the merge at the soonest possible safe spot for mixing trolleys and buses.
It's no-go. Unless you kick out the buses so you can cannibalize the loop. But kicking out SL1 lowers the value enough that it will never get approved for build, so it's still no-go.
So...to recap, I got my
vertical levels flip-flopped there. But the exact point on the map where you run out of room to insert the tunnel ends up same as before (now verified by calculator). It can't be done without grave impacts to the SS building, or safety impacts that won't be allowed for co-mingled buses.
If you want to call it "So you're saying there's a 1/1000th of a chance it'll work?"...yeah, sure, knock yourself out. It doesn't make it any less insane to keep quadrupling-down on the 0.001% for sake of proving...
something?...when this is a transit project the Seaport must ultimately have in order to sustain its growth. The existence of a Silver Line Phase III equivalent isn't a Crazy Transit Pitch; it's an officially proposed unfunded mandate. Nobody tries to fulfill a public works mandate on a scratch ticket odds of engineering feasibility at twice the price. They left us Essex St. Don't waste time endlessly debating functional impossibles; bloody use it and get the Seaport its transit mandate.
I do have a few questions regarding the Hudson St route. In Google Earth you can look at aerials back to 1995 for this area. In 2002 there is a view of the current Chinatown park which shows the former exit ramp from 93 south in the Dewey Sq tunnel. This ramp goes right up to the Danity Dot Building (now Radian). This is where the green line transit way tunnel would go. If this ramp was simply filled this seems like a very easy place to build the tunnel (however if the ramp was incorporated into the current 93 S tunnel, there is no place for a tunnel to Hudson Street). Also the Dewey Sq tunnel ventilation stack in the Chinatown Park is very close to the building with the Hot Pot Buffet at the Chinatown gate. Is there room here to slip a green line tunnel between 93 S, its ventilation stack and this building?
It was infilled. They deleted that ramp, cannibalized the northbound side of Surface to make it a SB-only one-way, and put a new onramp on the old NB side. The infilled offramp was on what's now the parking row and bike lane on that block of Surface. If you go back to that grainy shot on Historic Aerials you'll notice the Surface SB travel lanes are in the same exact position but it entirely lacks a shoulder.
At the narrowest point of the park directly in front of the Radian you have 60 ft. of width to work with for the tunnel. However...
unlike the Transitway approach that comes underneath the CR tracks, has to support the weight of commuter trains...and merges too soon into the Transitway to shed its center wall and narrow...this trolley tunnel has more room after underpinning 93S to shed its center wall and shrink in width as a single bore. That means (per GLX specs). . .
(10 ft. of inbound track centerline clearance to wall)
+ (10 ft. of outbound track centerline clearance to wall)
+ (13 ft. of between-tracks centerline clearance)
=
33 ft.
A 7 ft. reduction in width eaten up by the centerline clearances, minus another 3 for deleting the center wall. Therefore, if the walls stay 6 ft. thick, you have a
45 ft. wide tunnel instead of 55 ft. and
15 ft. of buffer space in front of the Radian. There's no risk to the Radian's new-construction concrete + rebar foundation from that distance like there would be to South Station's century-old granite foundation from a lesser distance.
Equal 15 ft. slack space for slipping between the Hot Pot Buffet building and the tunnel vents, so that's likewise a non-issue.
And as for Hudson St...because the tunnel is tangent and not curved, you can shrink those 10 ft. centerline clearances to the minimum allowable 8 ft. (source: GLX specs) and lop another 4 feet off the tunnel width to net a
41 ft. wide tunnel for a short stretch. That gets you through the extreme-narrowest portion of Hudson between the Golden Leaf building and the Wings Kitchen building (tight...careful mitigation required...but no worse than what SL III was going to have to mitigate on whole blocks worth of upper Essex). All points on Hudson south of Kneeland it's >50 ft. thanks to the plaza in front of One Greenway, and then you're home free in Pike urban renewal land.
Don't get me wrong; this is an expensive-ass project with building mitigations on any potential route. $1B-2B range when inevitable (but hopefully sub-GLX) bloat gets baked in. It's just not the $3B-4B that Silver Line Phase III was surging towards and assuredly going to top by the time design-bid and all that contractor bloat hit its final estimate. But figure that SL Phase III projected +20,000 new daily Silver Line riders and +4500 all-new transit riders in its headway-crippled incarnation shackled to the Dudley route through speed-killing Boylston Loop. As a much faster Green Line branch hitting all the downtown transfers, all heavy rail lines (SL III didn't go near Blue), and both North and South Stations the projected ridership is a whole lot higher and capital cost per rider craters to pretty reasonable level. Compared to the GLX sticker shock, $1-2B is about the going rate for an all-tunnel project with 1 all-new transfer subway station (Tufts), 1 all-new infill subway station (Ink Block) and 3x the projected ridership.