San Francisco’s Muni Metro system was originally designed to operate one and two-car trains on city streets, where longer trains weren’t needed, and then to couple the short trains into three and four car trains suitable for subway travel. This arrangement was used from the time the Muni Metro system opened in 1980 until the mid-1990’s when the General Manager of the SFPUC (which then operated Muni) made his fateful decision to discontinue the coupling. This meant that in order to keep up with demand it would be necessary to operate the one and two car trains in the subway. As demand increased it was necessary to continually increase the number of these short trains to the point of their completely overloading the subway and causing the entire Muni Metro system to be snarled up, virtually on a daily basis.

In early 2020 the drop in transit use caused by COVID temporarily put the problem on ice.

But now demand is starting to rise again, meaning that if nothing changes the SFMTA will likely soon be faced with the same problem. Absent coupling, the dreary choice will be to 1.) again overload the subway with short trains, or 2.) send unnecessarily long trains out into the avenues to disrupt cross traffic, or 3.) force a substantial number of Muni Metro riders to transfer between trains to get to the eastern parts of the city. None of the three is acceptable.

For this reason it is necessary to take another look at coupling, which the SFMTA has so far refused to do or even talk about, despite the fact that passenger trains are successfully coupled and uncoupled during revenue service in many parts of the world.

It is emphasized that in the case of the Muni Metro system, six branch lines all feed into one Market Street subway which make San Francisco’s system much more dependent on coupling than most other systems. Here are several ways of proceeding without gumming up the subway, disrupting the avenues or degrading the service for a sizable percentage of Muni Metro riders:

Read more here:

M Line Trains: Three-car M trains could operate to SF State University, thereby increasing the carrying capacity of the line. To avoid sending three-car trains all the way to the Balboa Park M line terminal, two cars could be uncoupled at State University and be ready to join the next inbound train. This would make it possible to eliminate the West Portal Shuttle trains, thereby reducing the number of peak period trains operating in the subway by 5 trains per hour. This in turn would create time slots for additional trains if needed.

M/K Trains: K trains could join M Line trains at the St. Francis Circle Station. For this to work K trains operating along Ocean Avenue would have to receive transit-preferential treatment and signal pre-emption at St. Francis Circle.

J/N Trains: The official SFMTA position seems to be that because of defects in the current Automatic Train Control System (ATCS), coupling cars together at the Duboce Portal would be disruptive and potentially dangerous. For this reason the idea of coupling at this location has apparently been shelved until the existing ATCS is replaced with a new $300 million train control system years from now. A closer look at coupling at the Duboce Portal is warranted. Bringing the benefits of coupling to this location is apt to be much less daunting in 2022 than it was in 1965 when the original system was being designed.

To improve N-Line regularity, the effective 3-inch high transit median between 9^th to 19^th Avenues should be extended in both directions.

A few weeks ago BATWG suggested that a modernized and carefully planned set of changes that included efficiently coupling some avenue trains into longer subway trains could both eliminate the need for riders to have to transfer between trains and the subway jams. But for reasons unknown, the SFMTA brass continues to refuse to tackle automated coupling, apparently preferring to downgrade service in its multi-billion dollar Muni Metro system.