Last weekend I went to check out the contra-flow bike lanes on New Hampshire. I like them a lot. A friend of mine was telling me about them (My wife has informed me that I have to pretend to not know about these things when people try to tell me about them) and she was very excited since she already rode the wrong way on New Hampshire anyway. Now there was space for her.
I did find the bike signal hard to see when going north, and it wasn't entirely apparent that I could continue into a bike lane where it became one way, but my biggest problem was that I don't think the sensor sensed me. Maybe I wasn't in the exact right spot, or maybe it was the wet pavement, but I really thought I was doing it right.
Some people have complained about the "Get Behind It, The Bike Box" sign. The sign is becoming standard nationwide where cities are installing bike boxes. As I recall, DDOT had to modify the original design that Portland uses to address the unique diagonal entry point for this bike box. And the signs are smaller than Portland's because Portland mounts these on overhead mast arms but DC’s use of mast arms is limited by the overhead wire/viewshed protection issues, and so DDOT had to mount them on vertical poles adjacent to the intersection (they are 22” x 34”).
And it's important to note that this is an official FHWA experiment. DDOT is phasing in changes, such as green paint in the lanes and in the bike boxes. This is being done so that DDOT can measure and record behavior with each design element. Drivers are encroaching into the bike boxes now, but DDOT needs to document it over a period of weeks prior to adding new elements to the intersection. For FHWA’s purposes, it’s very important to understand the incremental effect on behavior of each element.
As an FHWA experiment, it means DDOT has to work with FHWA and sometimes design things the way FHWA wants them. FHWA insisted on the double yellow divider line, for example. DDOT wanted to do a single 6” white or double 4” white line. FHWA sees it as a lane of opposite moving traffic, and thus the MUTCD calls for double yellow. DDOT was concerned about the drivers having to cross a double yellow to park. It's FHWA's experiment and so they won that one.
I look forward to the future phases.




The lanes are great. Love them.
I think they might have the lights/signals off. They were working last week, but I gotta say, horribly timed. They sent bikers across 16th street when drivers on U St got their yellow. This led to me (going south) almost getting hit by a bus trying to beat his yellow light and make a left onto 16th.
When they do get the lights going again, I hope they will allow bikers to start crossing 16th at the same time pedestrians do. Making us wait until cars are finally able to make lefts/rights is a horrible idea.
Posted by: Frequent User | August 18, 2010 at 09:58 AM
Thanks for the even-handed explanation of the FHWA MUTCD experimentation process. Too often, the dominant narrative is just about it stymieing innovation.
Posted by: darren | August 18, 2010 at 10:25 AM
Yes, the sensors don't work as they should. This might be the District's first stab at sensors for bikes.
The second day after the new signals were operational, I got to the intersection via NH Ave northbound to find another cyclist stopped in the painted zone. The signal failed to give us our turn and he grumbled. He told me that he was a consultant who had designed the intersection.
I lived out in Portland, OR for a few years and they have several intersections with sensors for bicycles. Theirs work a bit more reliably. They also mark the center of the sensor with a small target, so you know exactly where it is, rather than painting the whole area as done here.
One suggestion: while you're stopped on your bike, look for the seams in the pavement where the loop coils are installed. Try to get your pedals/chain directly over this spot. Especially if you have a composite frame, it can really help to lean the bike down--almost on its side--to get as much metal as possible near the loops.
Posted by: Pauly Bolson | August 18, 2010 at 11:53 AM
This article discusses Washington state. This is also how it's done in Oregon:
http://bikelaw.wordpress.com/2009/05/02/breaking-news-in-washington/
Posted by: Pauly Bolson | August 18, 2010 at 11:58 AM
can we learn nothing? the dutch use loop detectors on most of their bicycle routes throughout the country. especially the fully separated ones. Why can't we reverse engineer or something if the requirement is built/final assembly in USA? seriously reinventing the bloody wheel every time!
Posted by: John_in_NH | August 18, 2010 at 09:11 PM
Ive used bike sensors in california and they work flawlessly. From what I can tell, theyre just like car sensors, but instead of a circle, a dense series of crisscrossed lines inside a square. You can see the lines in the picture above, and DC seems to have gone with a very simple triangle....possibly not strong enough.
The double yellow was certainly the right choice, although I would have striped it.
Posted by: J | August 18, 2010 at 09:53 PM