This post continues the discussion in the comments below my previous post about a crash on the Pinellas Rail Trail.
So, first, because images can’t be embedded in WordPress comments, here is my photo of an installation on the Capital City Trail in Madison, Wisconsin, USA (photo taken 2002). The diamond-shaped lines are saw cuts in the pavement, into which electrical wires have been laid, followed by epoxy filler. The installation works as an upside-down metal detector, sensing bicycles which travel over it. A clever feature is that electronics detect which loop of wire is activated first, to switch the traffic signal in the background of the photo only for bicycles traveling toward the crosswalk. The loops are set back from the intersection for advance detection. Above the white plaque on the traffic-signal pole is a pushbutton, so pedestrians can also switch the signal. This is the location in Google Maps: https://goo.gl/maps/mk4aI
The difference between bicyclists’ and pedestrians’ operating characteristics was a major issue addressed in the comments on the previous post. As discussed in the comments on that post, a traffic signal may be used at a crosswalk, as at this location. The red-yellow-green traffic signal is for traffic in the street and there is a pedestrian signal for the crosswalk. The pedestrian signal was the only option available under national engineering standards at the time of this installation.
More recently, bicycle signals have been added to the toolbox, and they are useful at a crosswalk because of bicyclists’ different operating characteristics — for one thing, to allow bicyclists, who travel faster than pedestrians, to enter the crosswalk later. Experimentation by former Denver bicycle coordinator James Mackay showed this approach to produce no change in bicyclists’ behavior, but to make it legal, see https://web.archive.org/web/20101212050451/http://www.atssa.com/galleries/clean_titles/031408_BicycleTraffic.pdf
Your post was written in the context of my comment in the previous blog post regarding FDOT’s decision to incorporate a number of design elements I had requested at a new trail crossing it constructed adjacent to a signalized intersection. I had written: “The DOT also eventually agreed to my request to install a push button upstream of the crossing so bicyclists can request what I call a ‘protected green crossing’ before they get to the crosswalk.”
You responded: “I would have liked to see an electronic detector that switches the signal without requiring the bicyclist to – hmm, stop – and push a button. Vehicle detection can also determine the speed of the bicyclist and adjust the signal timing accordingly. This is not my idea and it is not novel: see photo I’l attach to my next post.”
The engineers with whom I first discussed my suggestions were open to many of my recommendations, but they warned (in carefully couched language) that their bosses might not be so open minded. Which is precisely what happened because I ended fighting their bosses for almost three years.
Initially I had requested passive detection so bicyclists would not have to push a button, but they raised one objection after another. As far as they (and I) knew, in-pavement inductive loop sensors can not detect bicycles with carbon fiber rims, so they would not detect all approaching bicyclists. We discussed proximity sensors, but they claimed the technology was not mature enough for traffic detection purposes (a claim I found hard to believe).
We discussed other methods of detection, but in each case they objected, maybe knowing that the chance of getting any type of advanced detection would be an uphill battle. I eventually gave up, reasoning that an advanced push button was better than no advanced detection at all.
I had also requested bicycle signal heads for bicyclists, even going so far as to discuss it with the state bike/ped coordinator. He opposed my request because he claimed FDOT would not provide different types of signals for the same facility. (In other words, FDOT only designs these types of crossings for pedestrians—the very different needs of bicyclists are ignored because they obviously do not know how to design for people riding bicycles.)
Also, the MUTCD’s Interim Approval for bicycle signal faces prohibits their use where “bicycles moving on a green or yellow signal indication in a bicycle signal face are not in conflict with any simultaneous motor vehicle movement at the signalized location, including right (or left) turns on red.” (If only the MUTCD would place that same restriction at all signalized crosswalks.) At that time, FDOT refused to implement my request to forbid turning movements across the trail while it has green, something I fought for tooth and nail. But it eventually relented and now forbids these types of conflicting movements.
And finally, the DOT installed typical pedestrian crossing plaques at the push buttons, parallel to the bicyclists’ direction of travel, maybe so the absurdity of the 20 plus word instruction on how to use the push button and crosswalk wouldn’t be so obvious. How many bicyclists who are riding at, let’s say, 5 to 10 mph, have the time to read these instructions? (It’s another example of the MUTCD’s failure to understand not only the conditions bicyclists actually face when they ride on separated facilities, but even the needs of pedestrians.) Actually, the signs’ orientation was to comply with the ADA requirements that the face of said signs are parallel to the direction of travel.
Of course, bicyclists who are not familiar with that crossing will never notice the advance push buttons because the DOT refused to do something as simple as install Push Button to Request Green signs facing the bicyclists as they approach the crossing. Many of them just ignore all of the signs and signals and make their decision to cross based on traffic conditions, like they do at so many other crossings.
Here two more players enter into the mess; the Federal Government, through the Consumer Product Safety Commission, and/or Department of Tranportation, and the manufacturers and distributors of carbon-fiber bicycle wheels.
A simple loop of thin wire weighing next to nothing, embedded in the rim, would make carbon-fiber rims trigger electromagnetic detectors as well as metallic rims do. The Federal Government has ignored/neglected this concern, failing to regulate. But also, carbon-fiber rims represent only a tiny percentage of all rims, and so this is a lame excuse even lacking Federal regulation. A wheel with a carbon-fiber rim and metallic spokes also might trigger the sensors, but I haven’t seen any test of that.
Proximity sensors using ultrasonic sound have been tried, but the one installation of which I am aware was not very reliable. The one in this photo and in the video in the following page was not working a year later. http://john-s-allen.com/galleries/well_nick/slides/IMG_0426crossalert.html
Optical sensors using video cameras or, better LIDAR, are coming into increase use, and work for all modes of travel, but they are rather expensive, as are signal heads.
I do not intend to denigrate the legal aspect of the crash. You do good work in pointing to bicyclists’ having no valid legal requirement to stop, but in the real world, I often do have to stop, because my life and health depend on it.
Regarding my decision to not push for in-pavement inductive loop sensors because they do not detect carbon fiber rims, you responded: “carbon-fiber rims represent only a tiny percentage of all rims, and so this is a lame excuse”
Although I can not grasp why someone wants to ride on such exotic rims, it is not my place to impose my values on them. Their need for a safe crossing is as important to me as that of the next person who comes riding along on aluminum or even steel rims, on the latest racing bike, or a beat-up Walmart special. No, this was not a lame excuse at all, but rather a reflection of my policy to advocate for the safety all of people who choose to ride a bicycle, for whatever legal reason.
Also, remember that I was fighting for something I had never seen in practice, had never seen published in any literature, and something these engineers claimed had never been tried, at least not by FDOT. By agreeing to the upstream push buttons I knew that the underground conduits would be in place if better technology became feasible in the future.
You wrote: “Proximity sensors using ultrasonic sound have been tried, but the one installation of which I am aware was not very reliable. The one in this photo and in the video in the following page was not working a year later.”
Actually, the type of proximity sensor I discussed with FDOT’s engineers needed to detect bicyclists but not pedestrians (because of the relatively short green interval for the trail I had requested, a new concept to those engineers), so I wanted a sensor that would be mounted at the edge of the pavement which would sense only bicyclists, and only when they were riding toward the crossing. To accomplish this, I envisioned a proximity sensor that would detect a person’s hand when it was passed within a few inches of the sensor, rather than a push button which required physical contact.
I should also note that this happened in the context of the MUTCD, AASHTO, and FDOT’s failure to provide guidance for the design of sidepath crossings at signalized intersections for the intended users. All FDOT wanted to do was slap down a typical pedestrian crosswalk despite it’s own study revealing that pedestrians comprised less than 5% of all users. So, they wanted to design a crossing for the wrong type of user and would then have acted surprised when the typical user ignored the controls that weren’t designed for them. Because I could not find any official guidance, I was on my own. (I see that NACTO’s “Don’t Give Up at the Intersection” hints at the problem, but my fight with FDOT occurred a number of years before that document was released.)
We had discussed something akin to the Cross Alert System’s passive sensors because I was familiar with them, but if I recall correctly, I was told their system was not compatible with the existing controller. And besides, it would also detect people afoot, something I did not want. Regarding your comment about their reliability: there is a Cross Alert System about 3 miles south of this crossing on the same trail that has been working reliably for at least 5 years, although it’s possible it has been briefly down a few times.
You wrote: “Optical sensors using video cameras or, better LIDAR, are coming into increase use, and work for all modes of travel, but they are rather expensive, as are signal heads.”
Well, when it comes to motorists’ safety, it seems money is no object because FDOT had signalized the nearby intersection in response to concerns about an “unacceptably” high number of motorists whose proclivity for smashing into each other raised eyebrows. But, the crash rate after the signalization actually increased, although the severity of injuries may have decreased slightly (it’s hard to figure that out from the limited information at my fingertips).
Traffic engineers know that signals do not always reduce crash rates, but they went ahead and installed them anyway because it demonstrated they’d done “something” (with public funds). Instead of addressing the root cause(s) of the high crash rates at the intersection, they just threw money at the problem and thought they could walk away. Except that in the process they created new hazards for trail users and told them they just needed to be more careful. That all too familiar attitude did not sit well with me, which is why I decided to fight for a safer design.
So, when it came to trail users’ safety, it’s not so much that they tried to skimp on costs (although they tried that as well), it’s that they just didn’t have the foggiest idea how to design a crossing for bicyclists (and probably still don’t). This was also clearly demonstrated when FDOT’s state bike/ped coordinator told me he opposed bicycle signal heads at this crossing—another example of an FDOT official who doesn’t understand the needs of people riding bicycles on separated facilities.
You wrote: “. . . in the real world, I often do have to stop, because my life and health depend on it.”
I often have to stop as well, for the same reason, but I suspect your approach to these crossings is very different than mine, and consequently you probably stop far more often than I do. A DOT official once told me that I am a “confident crosswalk user” because I approach trail crossings at a speed which clearly signals to motorists that I plan to cross…and I expect them to yield.
Of course, many motorists are not paying attention, but depending on the sight lines and speed of traffic, I tend to force the issue, and you might be surprised to learn how often they will yield once they are challenged and get the message that I do actually plan to cross (because I’m not bluffing—I intend to cross unless it becomes apparent a collision is in the offing). If they yield, I wave a “thank you” if I feel the situation is safe enough to do so, but if they refuse to yield, I’ve been known to verbally abuse them.
Of course, I am prepared to hit the brakes as hard as necessary to avoid a collision—I’ve done so numerous times—but they don’t know that beforehand, do they.
Having said that, I admit that I am not as forceful when riding our tandem with my wife for a couple of reasons, including the fact that having to suddenly stop and restart our recumbent tandem is far more difficult than doing so on my single recumbent bike. However, I equipped our latest tandem (trike) with a NuVinci continuously variable hub which allows changing the input/output ratio even when the tandem is stopped, so that is less of an issue than in the past. But even so, I ride more defensively when we’re on our tandem because I don’t want to create stressful situations for my wife.