Re: Copley grade separation
Clearly if a deal could be worked out with Angell Animal Medical Center to build a yard on some of their potentially underused land, and to extend the E branch one stop south to a stop just south of Bynner St, that would be valuable.
Clearly the number of trains arriving at Heath needs to more or less match the number of trains departing Heath if we don't have a yard, but if it were to consistently take 12 minutes for a train to get from Boylston to Symphony and 15 minutes to do the reverse, I don't think that would be a problem. Can you elaborate on why you think this would be a problem? Indeed, does the Heath to Brigham Circle travel time currently match the Brigham Circle to Heath travel time during rush hour? I'm sure variable delays, or very slow traffic, are suboptimal in that case, but that's different from a consistently different travel time in each direction.
I just elaborated why this is a problem. You're not paying attention, you're relying on your own intensity of belief to square a discrepancy without providing any evidence for why that's so. Including assuming that Angell is going to respond with pure giving altruism at what in effect is a gun to their heads: "Fork over your parking lot so we can have a bigger-than-normal storage yard or we have no choice but to bustitute all your nearby rapid transit service out of inability to run this kooky asynchronous schedule."
It's a dispatching nightmare to run an asynchronous schedule. The E runs a 6-minute bi-directional headway at peak, same as all 3 other branches. A 3-minute discrepancy in inbound vs. outbound scheduling...plus uncertainty padding...is a gigantic difference. You need more cars than there is available land for yard space to square that discrepancy. Kludging it up with Brigham short-turns fouls slots when the cars changing ends can't get out of the way. Cutting frequencies in one direction beyond Brigham is transit loss, pure and simple; it won't fly with City opposition. And dispatching that whole shit sandwich of awkward kludges on one branch while minding all manner of Central Subway slotting on all other branches is going to make OTP across the whole GL wildly inconsistent (or, assuming you pull out all other sensible SGR and straightforward performance tweaks to roll back OTP attrition to something more consistent...end up making it wildly inconsistent all over again).
On what empirical evidence is there demanding such an incredibly convoluted solution when, as EGE notes, the main problem with Copley Jct. congestion is "garbage in, garbage out" at the portals from lack of signal priority on B/C/E? If this is just a brain teaser in how to invent the hardest possible way to address a straightforward problem...stop right there. This is the wrong thread for that.
And with competent transit signal priority, I wouldn't expect the running time on the surface to be terribly different from the running time in the tunnel. And I bet building a tunnel because you can't figure out how to collect fares at surface stops in a sane manner is going to cost more than just giving away free fares at those stops if fare collection time were going to be the deal killer for the surface stops.
More assumptions not rooted in evidence. Surface running times will never be equivalent to the tunnel. The tunnel is a hermetically-sealed, grade-separated, single-mode system with centrally controlled automatic signaling for that one mode. And the Huntington tunnel in particular has the highest speed limit on the entire Green Line outside of the hinterlands of Newton on the D. The surface has mixed traffic and mixed traffic rules: crossing traffic, crossing pedestrians, signal phases for crossing traffic, line-of-sight pauses for crossing traffic at unsignalized intersections, and a lower City of Boston speed limit. The traffic properties are too divergently different to equalize unless you whack speeds in the tunnel with long schedule adjustments dwells at Pru & Symphony every...single...time. What possible sense does that make?
This makes me think we aren't communicating clearly about what I was intending to propose. I was thinking that outbound E branch trains would use the existing underground stops at Park and Boylston, then make surface stops in the vicinity of Arlington and Copley, and then go back into the tunnel for Prudential and Symphony, and continue to make all of the existing E branch stops to Heath or wherever the Hyde Sq extension process takes us.
I understand completely what you're proposing. It's overcomplicated, impractical, and completely unnecessary by any real-world (or idealized real-world) measure of bread-and-butter Green Line operations, any measure of fact-based mechanics of surface vs. subway traffic management, and any measure of fact-based mechanics of what the root causes of Green Line congestion are vs. what they aren't. The only thing that isn't clear is why one's intensity of belief in doing the most unnecessary and overcomplicated thing overrules traffic modeling facts.
Sorry...there's not a whole lot to debate here if this isn't going to grounded in some acknowledgment of real-world Traffic Modeling 101 and applying some sort of practicality-based filter on how far is too far for a transit solution to have to twist itself into a pretzel to perform its basic function. The graveyard of ArchBoston thread sidebars is chock full of dead-end circular arguments like this where belief un-moored from factual evidence always has inexhaustible reserves of personal hunches and what-if's and yeah-but's. It doesn't bring any clarity to the discussion of a possible real-world transit solution when carried out as a steel-cage match evidence vs. belief.
