This group of videographers were blessed with well-behaved drivers stopped on 16th St. When U St. gets a red, cyclists still have a red. Usually, the northbound drivers on 16th edge their way into the bike box, blocking the space that cyclists are meant to occupy, creating a dangerous situation.

As the narrator notes, the green bike signal lasts barely long enough to cross the street. By the time a cyclist enters the bike box, there's maybe one or two seconds before 16th St. gets a green, an alarmingly short interval given that drivers are already creeping forward, feet poised on their accelerators.

The intersection might work better once DDOT ends the experiment and paints the blasted facilities green already. Signage might help. I'm a little baffled by the light timing - cyclists have a red light when parallel-moving pedestrians have a walk signal.

For now, I think this signal is a menace, and I'm going to start using the crosswalk.

by David R. on Jul 11, 2011 11:04 am • link • report

What a confusing mess that lane is! I believe though it was last week (or the week before) that we saw plans for redesigning this intersection ... and it appears that the bike lanes are being left out of the configuration. Apparently DDOT has already realized that this experiment in counterflows isn't going to work in an intersection where so many roads come together. Like the video shows, cyclists are better off just either using the walk signal or driving in the road.

by Lance on Jul 11, 2011 11:11 am • link • report

Agreed, making it an actuated signal is dumb. It's not as if you're taking time away from anyone by giving the bikes a green. The bike-green should be in sync with the U street crosswalk crossing 16th street. It says walk, you roll.

by MDE on Jul 11, 2011 11:13 am • link • report

@Lance:
There's nothing wrong with contraflow lanes in general, nor in this case.

The problem is that the signal timing isn't quite right and that the bike detector is either broken or needs to be calibrated.

If cyclists had, say, 20 seconds to get into position on 16th Street (which should be fine, given that pedestrians crossing 16th have the walk for much longer than that), the intersection would work fine. And it wouldn't require taking signal cycle time away from vehicles, either. It might require a "left turn on arrow only" treatment for U Street, though.

by Matt Johnson on Jul 11, 2011 11:15 am • link • report

The bike signal takes so long I thought it was broken. I had never used it ans just follow the normal traffic signals instead.

by Adam L on Jul 11, 2011 11:16 am • link • report

Given the sociopathic behavior of drivers turning left off of U St., a "left on arrow only" control would be welcome in a number of ways.

by David R. on Jul 11, 2011 11:29 am • link • report

They should paint at least the boxes green so that they are more visible.

Overall, the idea behind the boxes is good - it tries to formalize what people would already do (use the crosswalk to cross and then continue on 16th to NH).

The cyclists in this video completely missed that the signal is right in front of you when you're stopped in the lane:

by MLD on Jul 11, 2011 11:31 am • link • report

What a mess.

by TGEoA on Jul 11, 2011 11:32 am • link • report

I go out of my way to avoid this intersection on my bike, at least when traveling on NH Ave. Really poorly designed and confusing. If you have to post an entire placard with instructions on how to safely and legally traverse the intersection on a bicycle, you've designed it wrong.

by Phil on Jul 11, 2011 11:39 am • link • report

Well, I was wrong, they did notice the signal in front of them eventually.

It isn't a totally confusing mess of a failure though. Improvements:

1. Paint bike boxes green.
2. Change bike signal so it turns green when the crosswalk signal says "walk" across 16th. Every cycle (no in-ground BS required). Then you just cross to the bike box, and wait until 16th turns green. Right now they're trying to time it so you don't have to stop - 16th turns green right as you would get into the bike box.

by MLD on Jul 11, 2011 11:46 am • link • report

I've come through this intersection almost every day for two months and not once have I seen the bike signal turn green. Usually I just scoot across 16th Street into the bike box when U Street has green. There's always a little platoon of other bicyclists doing the same, so I never feel uncomfortable crossing in front of oncoming traffic. This may not be a wise choice, however.

by dan reed! on Jul 11, 2011 11:53 am • link • report

I have little faith that the city will continue to do the tweaking necessary with new infrastructure like this to make it more usable for those intended. Still waiting for DDOT to assert itself under the new administration, though the initiative encouraging "stepping out of the box" appears to not be there currently.

by Geoffrey Hatchard on Jul 11, 2011 11:55 am • link • report

Lance, the drawings pre-date this configuration. When the intersection is rebuilt it will include these contraflow lanes.

by David C on Jul 11, 2011 11:59 am • link • report

Matt, this is one area that left me totally confused.

I was traveling west on U street and was about to turn onto New Hampshire and really was stumped. I didn't know if that left turn was for cars or bikes and ended up turning @the next block. It wasn't until after I passed and saw cars turning that I realize I could have turned as well.

Yeah, totally confused and could imagine how a cyclist could be hit.

by HogWash on Jul 11, 2011 12:05 pm • link • report

I'm not sure I understand the very first scene in the video. In order to avoid a "right hook" hazard, the bicycles blow through a stop sign without stopping? How is that supposed to help them avoid a "right hook" hazard?

by Alan on Jul 11, 2011 12:30 pm • link • report

This is such a busy intersection with motorized and non-motorized traffic that I think U St. left turns should be protected turns. I've waited to turn left there, saw a break in oncoming traffic, was about to turn, and then saw peds crossing.

Does it make sense to have cyclists follow the motorized traffic pattern in this area by turning right onto T and then left onto 16th? It would decrease the chance of the right hook and the T/16th street light could be a delayed green for motorized traffic while peds and bicyclists could start earlier. Since T St. is one way bicyclists wouldn't need to worry about oncoming traffic as they turned.

In regards to the parking sign allowing the car in the bike lane. I've noticed that when bulb outs are constructed the signs aren't always moved to correspond to the changes in infrastructure. This issue also affected Adams Mill Rd as one approached Calvert St.

by Rob on Jul 11, 2011 12:33 pm • link • report

P.S. That the point that I stopped watching. If they want to complain about poor lane design, but they can't even be bothered to stop at a stop sign while they are videoing themselves, why should I bother to watch the rest

by Alan on Jul 11, 2011 12:33 pm • link • report

That intersection is bad for pedestrians - worse i think than it is for bikers or cars.

There is no walk signal to cross NH once you have crossed U and are on the median. Bikers and cars come zooming south off 16th and turn on NH without pausing at or looking for pedestrians corssing. Cars taking a left from U do not look for pedestrians crossing.

It is a dangerous mess, and I cannot wait for it to be redone.

by greent on Jul 11, 2011 12:42 pm • link • report

He is not stopped in the magnetic reader so the stop light doesn't know he is there. I've made the same mistake myself.
http://imgur.com/sfXWY
This is a picture of him riding over the strip, and the second shows him waiting out of it's range. I bike through this light everyday across New Hampshire. You need to be directly on the magnetic strip at a certain time during the light cycle or it won't read you. I've tried to use it, but I just following the walking signals it 10x faster.

by My Favorite Word Is on Jul 11, 2011 12:50 pm • link • report

They should just use a button like they have on crosswalks to activate the signal. That seems like it would work a lot better than the magnetic detectors.

by Andrew S on Jul 11, 2011 1:07 pm • link • report

They should just use a button like they have on crosswalks to activate the signal.

Maybe. Button walk buttons are generally placed in locations difficult for a cyclist to reach unless they dismount and step up onto the curb. We don't require drivers to leave their cars to trigger lights.

Maybe, in this instance a pole could be placed in the bike box.

I have also found that the buttons for getting a walk signal are often not working.

by JeffB on Jul 11, 2011 1:17 pm • link • report

Why have it be on a sensor at all? Why not just make it another part of the light cycle? Is it really that much of an inconvenience?

by David C on Jul 11, 2011 1:20 pm • link • report

@Matt ...

"There's nothing wrong with contraflow lanes in general, nor in this case."

I'd argue that there are fundamental problems with contraflow lanes in general. John S Allen does a nice write up here.

http://tinyurl.com/ogx98

But if you carefully look at the research record in Europe, you'll find similar problems with contraflow and parallel segregated lanes as well. Mind you, as with most studies with few real controls, interpretation is a messy art. But sweeping all of this under the rug is glib.

Similarly, the cite of something fundamentally flawed as the Lusk paper without qualification is "shaky". http://tinyurl.com/42axxxq

In short, we really have to think about what an improvement is and what precisely we're trying to improve. Are we simply trying to get more riders? Are we trying to improve safety? Are we trying to increase convenience? Too often, I find that the literature confuses so many issues and tortures the data to simply confirm our own biases that it is hard to get anywhere on any topic. For instance, picking on facilities at the moment, if the number of collisions/deaths post installation goes down did the facility make cycling safer? Maybe ... maybe not. If the facility attracts slower riders -- or if conditions result in everyone moving slower -- then simply comparing collisions or collision rates would lead to a biased result. Consider the intersection Lee Hwy and Lynn St by the Key Bridge -- http://tinyurl.com/6fkywbk. I've never looked at statistics for the intersection. But few collisions might take place there because everyone seems to know how nutty the ped-bike-auto traffic is there. Nonetheless, it would be silly to call it "safe" because it fails our mental and empirical models of the cognitive/observational limitations of people in traffic while maintaining efficient/convenient ped-bike-auto traffic. Stated another way for emphasis, the data by itself often fails to identify true casual effects. We can only interpret data through models to get at things we really care about ... which I'll guess is something about risk, convenience, and pleasure associated with cycling.

Mind you, whether something is "safe" versus makes something "safer" are two different statements that get conflated all the time. So perhaps we should be willing to make changes that make the cycling experience better for many people if the change in risk is reasonable. But, IMO, we should be explicit about these trade offs and the uncertainty associated with what we call knowledge instead of sweeping it under a rug.

by Geof Gee on Jul 11, 2011 1:37 pm • link • report

Having pedaled through this intersection for the last 10 years nearly every day, you should know that the contraflow lanes and bike boxes are a vast improvement over the previous design (which had nothing for cyclists). The bike signals are tricky to trip, so I've given up and cross with the ped signal. I think the bike signal & phase should be dropped. The contraflow lanes and bike boxes are a great improvement for bicycling safety and comfort. Let's put this in the context of doing a lot of thing right! Yes there are imperfections, but this is so helpful for every day riders. I saw 2 kids and a dad riding NH Av. the other morning. Continuing to improve these facilities will help more kids and less confident riders ride. Keep it up DDOT!

by ccort on Jul 11, 2011 2:48 pm • link • report

I pretty much ignore the ridiculous bike instructions and just dismount and turn into a pedestrian, crossing U St. on the east side of 16th and then crossing 16th, both times on crosswalks, mounting the bike when I'm back on New Hampshire.

Works every time.

Same for the useless dedicated bike signal at 15th and Pennsylvania. I'd still be waiting there now if I followed those signals.

by Ward 1 Guy on Jul 11, 2011 2:55 pm • link • report

MLD, its no surprise they missed the nearside signal. Look how high its placed! They took infrastructure designed for cars and pretended its for bikes.

Bikes arent cars. Its dumb to use car infrastructure for them. That height is for trucks to see for 500 feet away...not bikes stopped at the line.

Look at this bike signal in France. There are two signals mounted on the pole, one higher up for further away cyclists and one RIGHT THERE below it at bike height

http://maps.google.com/maps?q=le+havre,+france&hl=en&ll=49.492921,0.114081&spn=0.000586,0.00142&sll=37.0625,-95.677068&sspn=46.543597,93.076172&t=k&z=20&layer=c&cbll=49.492939,0.113875&panoid=RTsZCIcdxqIMBPwbHWurzg&cbp=12,88.8,,3,1.99

And another one
http://maps.google.com/maps?q=le+havre,+france&hl=en&ll=49.493064,0.112553&spn=0.000586,0.00142&sll=37.0625,-95.677068&sspn=46.543597,93.076172&t=k&z=20&layer=c&cbll=49.493064,0.112553&panoid=7q-geHESd2w5soB1R9I3Gw&cbp=12,325.82,,2,4.42

Thats what DC should use.

As for the induction sensor...looks like incompetence on the installers part. I go through induction sensors for bikes every day in California, and they work 100% of the time.

This pattern seems to be highly effective

Circle is for cars, square detects bikes better

http://maps.google.com/maps?q=93611&hl=en&ll=36.801513,-119.654999&spn=0.000728,0.00142&sll=49.493064,0.112553&sspn=0.000591,0.00142&t=k&z=20&layer=c&cbll=36.801372,-119.654999&panoid=0vfHSAP_R0HEyv-H_IQ2aA&cbp=12,225.38,,0,47.85

(note theres one on the left turn lane as well)

by JJJJJ on Jul 11, 2011 7:13 pm • link • report

Oh and protip, if a sensor is clearly ignoring you, move your bike over it back and forth....and then call the city and tell them its not working right.

by JJJJJ on Jul 11, 2011 7:14 pm • link • report

Alan: "P.S. That the point that I stopped watching. If they want to complain about poor lane design, but they can't even be bothered to stop at a stop sign while they are videoing themselves, why should I bother to watch the rest"

I stopped wathing after they said "On the approach to U and 16th streets, raised barriers threatened to topple the bicyclists into the oncoming lane."

FFS, those are so oncoming cars going around that curve don't enter the bike lane. Which they weren't in.

by Bob See on Jul 11, 2011 7:56 pm • link • report

Oh wow, the hypocrisy in these comments is absolutely amazing.

Complaining that they didnt stop?

WATCH THE VIDEO!

Neither the car in front of them OR the one behind them stopped!

(It appears that the car in front of the one in front of them did stop)

by JJJJJ on Jul 12, 2011 12:42 am • link • report

I live on the 1900 block of NH Ave. The lane/signal improvements help, but remains a clusterfuck of cars, buses, bikes, people, and Engine Co. #9. What about an omni-directional "scramble" crosswalk at NH/U/16th?

Priority Fix #1 - Getting pedestrians safely across New Hampshire on the south side of U St. It takes Jedi skills to divine the intentions of southbound traffic.

Still, good job DDOT et al. Keep planning, keep improving. Thanks.

by Andrew Broyles on Jul 12, 2011 9:06 am • link • report

I live in this neighborhood and traverse the intersection on foot daily. One thing about the bike lanes that consistently amazes me is that the bike lane headed south from NH to U Street, which cuts diagonally across part of 16th, is often confused by pedestrians as a crosswalk. I see it at least twice a week and always have to fight the urge to yell at my fellow pedestrians while I watch motorists' confusion and frustration.

by dcmazzie on Jul 12, 2011 10:40 am • link • report

I'm one of the cyclists in the video. First of all. I did stop at that stop sign. I didn't put my foot down, but I paused a good deal longer than any of the motorists did. It does appear that John didn't stop, though I can assure you it was unintentional. With all the other weirdness there —the bike lane crossing paths with the right turn only lane—one could be forgiven for neglecting to notice the stop sign on the first time through (especially since motorists were rolling it).

My first approach to that intersection was from the north. My initial impression was positive. I really liked how the bike lane was to the left of the general use lane and fed into the contraflow lane. I also liked the back-in parking. It appeared to be a creative bike blvd solution. But I thought the bike signal was too short to get established in the bike box and the bike box was way too shallow in front of the right lane.

On the northbound trip, I was appalled at the bike lane to the right of right-turn only traffic onto T street. That's unconscionable and should have been changed immediately upon creation of the contraflow lane. This kind of a conflict would NEVER be created for motorists. If it was, accidentally, I'm sure it would be re-striped within days. We should expect no less. To me, that demonstrates a lack of regard for cyclists beyond chest-thumping over symbolic bike-friendliness. But I'm cynical like that. I live in a city where we have 2-block-long bike lanes so a former mayor could add them to the mile-count. The door zone sharrows north of U St are also inexcusable.

I made numerous other trips through that intersection over the few days following this video. I also spent an hour or so on a couple occasions standing nearby and watching how the cyclists dealt with it. I haven't published my observations or videos yet, but I'll share 2 here:

1) On many occasions, motorists encroached into the bike box just as I was heading to it with the green bike signal. It seems they are watching the U street signal and when it turns red, they drift forward in anticipation of a green light. This is incredibly disconcerting, and was downright frightening when a metro bus did it! I noticed that most cyclists pull forward of the bike box, into the crosswalk, or even move over to the U street crosswalk. That makes sense in context, though I imagine it's problematic for pedestrians.

2) Cyclist behavior appeared random to me at first - with some going clockwise and others counterclockwise. But as I observed, the pattern became clear. Cyclists appeared to be taking the cue from the U street signal as they approached and going the way that provided the least delay based on whether U had a red or green. Again, totally understandable, though risky—I saw a number of close-calls.

In general, I like the intent of the installation. The concept of creating a bike blvd sort of a treatment to accommodate a cycling route is good, and from what I've seen DC could use a bit more of that. But as you said, this thing has some serious flaws that need to be addressed.

Hopefully your readers have also read John's first post, based on the rendering. He approached this prepared to like it, certainly with an open mind. As did I.

by Keri Caffrey on Jul 12, 2011 4:54 pm • link • report

JJJJJ : "Neither the car in front of them OR the one behind them stopped!"

2 wrongs does make a right!

And lol, "hypocrisy"? If this vid was about driving there and filmed with car-mounted cams, and they didn't stop, I doubt anyone would rush in to defend it. But here you are.

by Bob See on Jul 13, 2011 7:06 am • link • report

The one time I tried to use the bike signals I almost got hit. I just use the crosswalks now.

by M on Jul 13, 2011 11:06 am • link • report

I am the cyclist in the video, and I edited and narrated it. It is not my practice to run stop signs, as some readers imply. Consider that I could have very easily started the video a few seconds later and left out the approach to T street. Why do you think that I left it in?

Answer: I ran the stop sign because I did not see it. I left that in to show how that can happen.

The stop sign was at first hidden behind a tree, then as I merged out of the poorly-paced bike lane into the stream of traffic, I had to look back for traffic. Once I was in the stream of traffic, the sign was not where a bicyclist -- or motorists -- who also did not stop, as shown in the video -- were likely to notice it, especially when paying attention to other traffic in the intersection. Also, the distinctive shape of stop signs at the opposite entrance is not visible because they are back to back with Do Not Enter signs. As this intersection has a four-way stop, the stop signs facing T street give the impression that New Hampshire Avenue traffic may keep moving.

What happened here is very much like the famous and widely celebrated break-dancing bear video. It can happen to me, and it can happen to you, and especially at an intersection with as many confusing cues as New Hampshire Avenue and T Street.

by John S. Allen on Jan 4, 2013 1:39 pm • link • report