The Washington, DC region is great >> and it can be greater.


Testing the "Invisible Tunnel"

Steve Offutt has joined the GGW contributing team. He lives in Arlington and has also been contributing to Arlington's county-run CommuterPageBlog. Welcome Steve!

The technology exists to allow Metro riders to transfer between the two Farragut stations and treat them as though they were transferring within the system. Metro should implement this idea immediately, since there is no downside, many riders will save time, and congestion at Metro Center will be reduced.

For background, see my previous posts (1st, 2nd, and 3rd) on CommuterPageBlog.

Recently I had a meeting at American University, which provided me the perfect opportunity to try out the transfer for myself. I was traveling from Arlington, so I got off at Farragut West, walked up 17th Street to Farragut North and then took the red line to Tenleytown. I made the same trip in reverse on the return.

I took a stopwatch with me to see how long it would take. For the initial trip I reached the top of the escalator at exactly the wrong time to cross I St. and had to wait the full light cycle. I waited about 25 seconds to cross K St.  I was standing on the platform at Farragut North 5' 13"; after the doors opened on my train at Farragut West. On the return trip I arrived on the street during the walk signal at K St. but had to wait about 20 seconds at I St.  I was on the platform 4' 10"; from the time the train doors opened at Farragut North.  I did not run. I walked at a normal able-bodied speed. Someone in a hurry could make this transfer faster; if one stands on the escalators, it will take longer.

So what does this mean? In both cases I then had to wait a little bit for the train, so I likely ended up on the same one as I would have had I made the usual transfer at Metro Center. However—particularly on the return trip—had I arrived on the platform just in time to catch a train, that train would almost certainly be one train earlier than what I would have caught at Metro Center.

Based on this one experience, I would guess that a person making this transfer during rush hour will catch the earlier train at least a couple of times a week and possibly as much as half the time if they hustle. During periods with longer periods between trains, one will catch the earlier train less frequently, but it will save a lot more time when it happens.

The bottom line is that the transfer works and many riders will learn how to take advantage of it if it's made available to them. If you work at Metro or know who to contact to help push this forward, please do so. If you're a rider who would avail themselves of the transfer, please contact Metro and request it. I know that Chris Zimmerman has made at least one inquiry about it; perhaps he can continue to pursue this along a faster timeframe than sometime in 2010, if ever. A woman named Cyndi Zieman was recently put in charge of SmarTrip cards. Perhaps she can take a leadership position and make this happen. It's a no brainer; let's build the invisible tunnel!

Steve Offutt has been working at the confluence of business and environment for almost 20 years, with experience in climate change solutions, green building, business-government partnerships, transportation demand management, and more. He lives in Arlington with his wife and two children and is a cyclist, pedestrian, transit rider and driver. 


Add a comment »

Ask Kwame Brown to focus on this if/when he is re-elected! He likes to get things done and brag about it later-- and there's nothing wrong with that for a politician.

by Tom A. on Oct 30, 2008 9:30 am • linkreport

The whole "invisible tunnel" episode reminds me a bit of the NextBus fiasco: I recall a conversation I had years ago with Jack Requa, then head of Metrobus, about the bus tracking GPS data they acquire. Apparently, ever since the GPS units were installed, they could get the bus position every two minutes. So presumably they could collect huge data files of GPS location and timestamp correlated with each bus serial number. But they had no (computerized or automated) way to combine the bus serial number with the route and run to which it was assigned. Even after-the-fact processing would be a gold-mine of data to figure out which buses are on time and where delays are happening. But without code from their vendor, they just threw their hands up in the air.

In part, I think that they knew how bad bus on-time performance was and were afraid to collect any data which might undermine the sort of best-case, rose-colored-glasses scenario they present to a (non-bus-riding) board.

But--as with the invisible tunnel, and with passes on SmartTrip--what this shows is that WMATA needs its own in-house programming staff, and it needs to write contracts for technology-heavy vendors (like SmartTrip or the bus tracking hardware) that gives it full and unfettered access to the raw data. There's no reason for these projects to get delayed by years waiting for some vendor to write the code.

by thm on Oct 30, 2008 9:58 am • linkreport

Thanks for the data, Steve. I have asked WMATA's public relations people for information on the standards in use for Smartrip cards and readers to see if we really are locked in with a single-source vendor. Newer card standards are more open and would tolerate a change in vendor more easily.

I don't expect a response from them because I have been very critical of WMATA recently and they probably feel no obligation to respond to me. My audience is small and to them I'm "Just a Blogger". My emails and phone calls go unreturned for weeks.

Hopefully, a resolution of the "Are blogs media?" question, as well as a change in federal FOIA requirements late this year may improve the response I get.

by Michael P on Oct 30, 2008 11:23 am • linkreport


There is basically one smart-card vendor in the industry (Cubic) that is the source of the fiascos that happen with most smartcard related projects - there just isn't another option for most transit agencies. It's not the sort of thing you can do inhouse.

You might be interested in the Uniform Fare Standards Program effort that the APTA Standards program is working on

by J on Oct 30, 2008 11:54 am • linkreport

A couple years ago I was living in Columbia Heights and dating someone in Court House. I would do the same thing from Metro Center to Chinatown (take the 11th st metro Center exit to the 9th street Chinatown escalator). It's a gamble late at night and half the time you end up on the same train home for all your effort, but during rush hour it easily saves 5 minutes or more.

And at night time if you make it by a few seconds and save a 20 minute wait on the platform it's like you won the lottery.

by Alex on Oct 30, 2008 12:07 pm • linkreport

So there is not timesaver advantage to this situation, but the effect on crowding at Metro Center seems to be the real long-range goal.

by The King of Spain on Oct 30, 2008 12:16 pm • linkreport

Building the actual tunnel would save more time, and even this saves some time, but yes. WMATA's main reason for planning a real tunnel is insufficient transfer capacity at Metro Center.

Alex: They also want to build a tunnel from Gallery Place to Metro Center, by the way.

by David Alpert on Oct 30, 2008 12:19 pm • linkreport

A tunnel between the two Farragut stations seems like a no-brainer. Was disappointed when I got here and discovered it doesn't exist.

by Froggie on Oct 30, 2008 5:45 pm • linkreport

Perhaps there is an issue regarding this ecologically sensible yet rashly canceled project:

IIRC, it was to pass through the high ceiling area of the WMATA vault; perhaps the two tunnels conflict?

by Douglas Willinger on Oct 30, 2008 8:39 pm • linkreport

I remembered and found this:

(alas the graphic links are dead and Way Back Machine is no help)

John R Cambron

More options Oct 13 2001, 3:49 pm

Newsgroups: dc.urban-planning, misc.transport.urban-transit, misc.transport.road

From: John R Cambron <**>

Date: Sat, 13 Oct 2001 15:32:13 -0400

Local: Sat, Oct 13 2001 3:32 pm

Subject: Re: Ped. tunnel between Metro Ctr., Gallery Pl; Blue Line to run on Green Line?

Reply to author | Forward | Print | Individual message | Show original | Report this message | Find messages by this author

Thomas Smith wrote:

I think a tunnel between the Farragut North and Farragut West stations would be more practical. I know it was initially planned that the two stations would be initially linked, but the National Park Service balked because it would mean digging up Farragut Square (although if they stayed along 17th St. NW. this wouldn't be a problem). Also, this would have interfered with the proposed I-66 along K St. NW. (part of the ceiling at Farragut North station was lowered for this reason). They did construct "knock-out" walls in both stations should they change their minds.

No doubt. however you have it wrong about the National Park Service objection. the location of the pedestrian tunnel would in fact run under 17th Street. Knockouts for this connection are in place in both Farragut stations. Farragut North (A02) south end paid area mezzanine opposite (west) side of train hall from existing surface entrance

passageway, Farragut West (C03) east paid area mezzanine opposite (north) side of train hall from existing surface entrance passageway.

The bottom of this section of the Farragut North (A02) station train hall shows the outline of the K Street freeway easement;

This plan of the south mezzanine of the Farragut North (A02) station shows the location of the knockout;

This cross section at the center line of the Farragut North (A02) station south mezzanine surface entrance passageway also shows the location of the knockout;

As for the pedestrian tunnel interfering with the K street Freeway easement this might be that big of a deal as there is a difference of about 20' between in elevation of the mezzanines of the stations.

The change in elevation could be made just outside of the Farragut North (A02) knockout with the pedestrian tunnel running above the K Street Freeway easement.

I also have similar drawings of the Farragut West (C03) station.

It would be more difficult to go between Gallery Place and Metro Center because it is twice as far between those two stations than the two Farragut stations (900' between Gallery Place and Metro Center compared to 450' from Farragut North and Farragut West). The configuration of the track makes this difficult, too. The tracks along that stretch of the Red Line are too close together to build between them. You would have to go parallel to the tracks risking damage to the buildings above. I don't know how deep the tracks are here, but I don't think they are very far down (about 30' from track bed to street level).

Such a tunnel would likely be built using cut and cover type

construction. Shoring piles used to hold the excavation open

during original construction are still in place and would likely be used in any pedestrian tunnel construction. Most of the buildings along G Street have similar shoring piles. How WMATA would make this a paid area connection will be interesting. Building it as a free area connection would be very easy, rexcavate above the tunnels, construct pedestrian tunnel above existing rail line tunnels connect pedestrian tunnel to surface entrance passageways in Metro Center (A01) (east) and Gallery Place (B01) (west).

by Douglas Willinger on Oct 30, 2008 9:02 pm • linkreport

For those that want to see the drawings that are linked to in massage Douglas Willinger posted above, here they are:

by Sand Box John on Oct 31, 2008 9:42 am • linkreport

Anyone who buys a Metrorail rail pass rather than using a Smartrip card or a pay-per-ride farecard can make a free walking transfer today.

by Steve Strauss on Oct 31, 2008 10:06 am • linkreport

Thank You John Cambron.

I was afraid the graphics were lost.

Good to see that you are still around.

by Douglas Willinger on Oct 31, 2008 11:01 am • linkreport


Less than 8,000 riders currently use a rail fast pass. It's probably because they're inconvenient compared to Smartrip, which doesn't have passes, and they don't end up saving an average commuter much money unless you ride from somewhere far like Vienna.

by Michael P on Oct 31, 2008 11:55 am • linkreport

I read my own post wrong, it's actually 6,275 combined weekly passes and 4,056 daily passes for rail. Compare this to more than 700,000 rail trips per day. It's somewhere on the order of 2-3% of the ridership.

Bus is significantly higher, with 19,572 estimated passes. Then again, the bus pass is actually a good deal for a commuter, since it's priced ($11.00) at less than 10 single rides ($12.50 on Smartrip, or $13.50 cash). According to this report, 27% of bus riders use a flash pass for the 16th Street bus line, the 2nd highest ridership line in the District.

by Michael P on Oct 31, 2008 12:13 pm • linkreport

The original poster stated that the transfer time from train to train between the two Farraguts was approximately 5 minutes.

I wonder how long the travel time between the two Farraguts using the existing transfer at Metro Center. My guess is that it is quite a bit longer, since you have to go several blocks out of your way on the train.

The best way to test this out would be to get two people to travel together with stop watches from Foggy Bottom to Dupont. They start at the same time boarding on the same inbound orange train. One makes the transfer at Metro Center, while the other makes the 17th Street transfer. See which one gets to Dupont first.

by mrs-man on Nov 2, 2008 8:14 am • linkreport

The Chicago Transit Authority uses Cubic for its farecard services and has had "invisible" transfers since 1999, when paper transfers at State/Lake between subway and elevated trains were phased out. Our farecard equipment doesn't require that you "exit" the system, either -- it only deducts fares upon entering the system.

You'd think it would be even easier to program a routine that says "anyone who exits at Farragut West can, within 20 minutes, enter at Farragut North for $0." Farragut Square shopowners should be pushing for this, too.

by Payton on Jan 5, 2009 9:22 pm • linkreport

Payton: That's a good point, but I think what's going on is that WMATA does not control its own destiny when it comes to Smartrip software. Implementing such a feature may be easy, in fact it may even be trivial, but in WMATA's case it still would require asking their Smartrip programming contractor, Cubic Transportation Systems, to change the software and perform testing, something that can't be done for free because it's not covered under the existing contract. My guess is that such a change to the contract would cost somewhere on the order of $10,000 based on my experience with similar government technical support contracts (about 100 hours of time billed at $100 per hour?).

It's different if WMATA has a "level of effort" contract with Cubic, though I don't know if they do. That way, as long as the level of effort contract still had some time left on it, WMATA could ask Cubic to do this software change and bill it against the hours in the LOE contract. Special projects like the upgrades anticipated next year could be done under a separate fixed-price (or cost plus fixed fee) contract like they are now.

WMATA's fare system is incompatible with entrance only billing. Unless we all went to flat fares, which would kill the central city short riders at the expense of the longer suburban commuters. It's a delicate balance now and would be terrible policy to try to change it now. My guess is that such a flat fare would be something like $3.00 per ride, might even be that high during non-peak too.

by Michael Perkins on Jan 5, 2009 10:08 pm • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

You can use some HTML, like <blockquote>quoting another comment</blockquote>, <i>italics</i>, and <a href="http://url_here">hyperlinks</a>. More here.

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.


Support Us