Observations from Europe: Near side traffic signals reduce crosswalk blocking

By Steven Vance On September 24, 2012 · 25 Comments

The driver of a Chevy Equinox blocks the crosswalk at North Avenue and Oakley Boulevard in Wicker Park. If the only traffic signal was on the near side of the intersection, she wouldn’t drive into the intersection as she wouldn’t be able to see when the signal turned green. But with far side signals, she can still see the light change.

It took me a while to see what was happening. I think I first noticed that people driving their automobiles were never blocking crosswalks while waiting at a red light. And people on bikes were doing a good job at respecting the crosswalk boundaries, too. I next realized I was doing it, too: waiting behind the crosswalk. I’d do this at intersections with hundreds of pedestrians and intersections with none. I then became aware of where the bike signal was: at the edge of the intersection, before you entered the intersection. And there wasn’t one on the other side.

Welcome to traffic in Germany, where traffic signals are mostly installed on the near side of intersections and rarely on the far side. The effect is simple but pleasant and profound: people stop at the stop bar, before the crosswalk. If you didn’t stop there, you wouldn’t see the signal and you wouldn’t know when it turns green. The near side signal also means fewer signal heads to install. Where in Chicago, a lot intersections have 3-5 signal heads, many German intersections I cruised through had 2. The intersection of Milwaukee Avenue and California Avenue has 3 signal heads for each direction, although there is only one lane in each direction.

This photo shows the effect of near side signal heads. It’s labeled to show where the signals are, and to whom they are directed. The Google Street View below gives you another view of this intersection in Munich, Germany.

View Nymphenburger Straße and Dachauer Straße in a larger map. This Google Street View is from the point of view of the driver of the silver BMW station wagon in the above photo.

This intersection, of Nymphenburger Straße (“stross-uh”) and Dachauer Straße, has cycle tracks with bike lane crossings between the intersection and the crosswalk. Near side signals keep automobiles out of both crossings, then. Eastbound Nymphenburger Straße has three lanes, one of which is for left turns. It has 4 signal heads, all on the near side. Two signal heads are for left turns: one is low, for drivers waiting at the stop bar, and one is high for approaching drivers. Two signal heads are for through movements and right turns: again, one is low, and one is high.

This wasn’t a tool mentioned in the pedestrian plan, and I’ve not heard of it being a feature anywhere in the United States, but I’d love to experiment with removing far side signals and using only near side lights at intersections. Pedestrians would have a much easier time crossing the street.

N.B. Attorney Brendan Kevenides, a sponsor of Grid Chicago, has requested that we discuss in the future features of transportation we experienced in Europe that we disliked. I’ll get right on that as soon as I can figure out what they were. I’m kidding, I have a few in mind.

AdamHerstein

Much of Evanston has these types of lights installed. A lot of intersections just have a signal on the near side on the right. Driving, I always found these an annoyance, since they are hard to see, but I can definitely see now that they make the intersection safer. Google Street View: http://goo.gl/maps/B1LlF

Edit: Upon closer inspection, I noticed that there is in fact a light on the far side on the left. But the fact that there are no overhead signals seems like a good start at least.

John

Keep looking, there is also one on the far right.

AdamHerstein

Ah, I see it now. I do think that these kind of signals look better than the overhead ones. I’m not sure if they are as effective as just the near-side signals, though.

John

I think they’re harder to see, especially on streets like Ridge Ave with a thick tree canopy.

nonya

when I read the headline, I thought, “Steven’s going blind, we have near side signals everywhere”. But no, he means *only* near side signals? That’s really, really smart.

I wonder if the pedestrian part of it was planned or if it’s just a happy side effect of German efficiency saying only near side signals are needed.

Chicago should try this at some high ped/car accident locations and soon.

http://www.stevevance.net/ Steven Vance

Ha. Yep, I mean *only* near side signals. It’s what I encountered 90% of the time when I was biking around Munich and Augsburg for 5 days.

http://twitter.com/aka60643 AKA60643

I’d *LOVE* to see this tried here – less intersection and crosswalk blocking, less visual clutter, and energy savings, too.
jen b

I think these lights are only effective on narrow streets, and even then I feel like unless you already know that light is there, one could easily not see it during the daytime and blow through the intersection. I do appreciate the idea that it would get more cars to stop behind the stop bar, but then again, isn’t that something we all learned in driver’s ed and should be doing anyway?

http://www.stevevance.net/ Steven Vance

Here’s another situation (in a series of situations I’ve been promoting) where engineering is self-enforcing.

As you see in the Google Street View of the intersection in Munich, there are signals at about eye level (if you were standing) and signals high above the roof of the car, so you could see them from afar.

John

“This wasn’t a tool mentioned in the pedestrian plan, and I’ve not heard
of it being a feature anywhere in the United States, but I’d love to
experiment with removing far side signals and using only near side
lights at intersections.”

Signal head placement is governed by the MUTCD, meaning it’s required by law. A local or state jurisdiction would have to submit a request to experiment to the FHWA. Seems like an interesting idea though.
http://mutcd.fhwa.dot.gov/condexper.htm
AdamHerstein

Whenever a car is blocking a crosswalk, just make sure you walk full-force into the driver’s door and exclaim “sorry, I didn’t see you there!” That should solve this problem once and for all. Bonus points if you manage to make a dent.

Mark Twain

Until they get out of their car and beat the shit out of you because you touched their little moving island in the city.

Eric Fischer

The sad thing is that the US standardized on farside signals basically by chance rather than from the result of any research. The 1935 MUTCD committee chose farside just because there were enough more existing farside installations than nearside that it seemed like a trend.

http://www.stevevance.net/ Steven Vance

Has there been any research since?

Have you seen this video of Jeffrey Tumlin vs. AASHTO? http://www.youtube.com/watch?v=XOYpjPT3BGI&feature=youtu.be

I haven’t watched it yet.

Eric Fischer

I don’t know if there has been more research, but I will keep an eye out for it as I look for other things!

I started to watch the video earlier but stopped because I decided it would just be painful to watch people argue.

http://www.stevevance.net/ Steven Vance

Did YouTube transcribe it? I prefer to read than watch.

http://stopandmove.blogspot.com/ Jass

There is some logic to farside signals

The placement is so drivers look straight, instead of looking UP. That way, the driver can see the people in the crosswalk (assuming hes not blocking it) instead of obsessing about looking straight above him and missing what was directly in front of him.

Thats the case even at pedestrian only crossing. Note the spacing between the stop bar and the light.

http://goo.gl/maps/ss7Jv

Of course, this reasoning goes out the window when you consider that European countries supplement their overhead light with one at eye level, which pretty much hits your side window.

Other countries, like Brazil, also use near side signals. Likewise, it does a good job of keeping crosswalks clear.

http://goo.gl/maps/Rs8q7

Goodluck changing it. As someone else mentioned it, it’s in the MUTCD, the traffic engineer bible. It’s not a ” may” it’s a MUST.

As for the talk about signals on the side (far or near) vs overhead signals…I think a recent (2009?) law or policy change was made requiring overhead signals. For example, every time Boston redoes an intersection, they put in overhead lights where none existed previously. Someone in government told them the feds were requiring it.
Pingback: 1400 words worth of time-wasting bike links to ruin any hope of Tuesday pre-High Holiday productivity « BikingInLA
Mark Twain

What’s the cost of one signal? Could the money saved be put into more left-turn signals to get traffic moving? I’m all about less cars, but left-turn signals would really make a difference at various intersections in congested areas.

Eric Fischer

Rhode Island’s price list http://www.planning.ri.gov/transportation/srts/estimated_costs.pdf quotes $8000-$11000 for the mast arm and $1000 for the signal head, plus wiring, controller, installation, etc. Regular poles are apparently much cheaper ($1500) than mast arms.

I don’t know how left-turn signals are going to reduce congestion, though, since those extra signal phases are going to consume time that would otherwise be available to vehicles and pedestrians moving straight. If the problem is that vehicles waiting to turn are blocking ones wanting to go straight, could you take away some parking spaces to make a turn pocket for them to wait in? If there is no turn pocket, they will still block the other vehicles while they wait for the green arrow.

http://www.stevevance.net/ Steven Vance

There are a few issues at play with left turns in Chicago.

Consider a 4-way intersection with at least two left-turn lanes on the same street. A left-turn lane doesn’t always come with a left-turn signal (and you pointed out something important about the presence or absence of a left-turn phase). Chicago has many short blocks and lots of on-street parking. Both of these affect the length of the left-turn lane.
At southbound Milwaukee and Ashland, there is a left-turn lane (without a left-turn signal). The lane is very short and there may be more people who want to turn left here than there is space available for their cars. The length of the absence of parking is shorter than the length of the left-turn lane. People who want to move straight who otherwise could have done so in this phase may have to wait for the next phase because there were “too many” people wanting to turn left. If there was no parking, drivers could bypass the left-turn queue on the right, against the curb.
To further complicate things here, there’s a bus stop where people queue to turn right. The waiting time of the bus is affected by this bus stop placement.
See a Google Street View: http://goo.gl/maps/sDlZM

Pingback: Near Side Traffic Signals prevent crosswalk blocking | Law, Library, Privacy, Technology, and Other News
Pingback: News & Notes | Greater City Providence
USbike

Another consideration is the right-turn-on-red. That would also somewhat limit the effectiveness of the near side-only traffic signals if the drivers in the right-most lane move forward into the pedestrian crossing (or bike lane if those exist) wanting to make the turn. Personally I really dislike allowing cars to make turns on red, as a pedestrian, cyclist and driver. It’s one more uncertainty in the equation that makes it unpleasant for non-motorists at best, and potentially increases the risks of collisions for everyone at worst.

Not to mention that when >95% of a city’s intersections allow it, that dinky little “no turn on red” sign does squat to prevent many motorists from illegally making right turns (sometimes they may not even realize it wasn’t okay). That very uncertainty of whether people will actually turn on red or not makes me more stressed at such intersections than the standard ones when biking or walking.
Pingback: No Turn on Red | Critical Transit