One unsung area where Portland is doing some very cool stuff for bicycling is with traffic signals and “ITS” — which stands for Intelligent Transportation Systems. The field of ITS encompasses all sorts of high-tech ways to make our streets smarter. From sensing vehicle patterns with RFID, to software that manages complex signal systems.
Around Portland, some of the most innovative examples of traffic engineering fall under this category. ITS is how PBOT managed to improve bike access on the NE 12th overcrossing without upsetting nearby freight-dependent businesses. It’s also how they dealt with the notorious right hooks on Broadway at Williams (with bike-only signals).
PBOT is quietly becoming a leader in this field. Did you notice the recent press release from Bikes Belong about their Green Lane Project? The head of that project said, Portland’s, “attention to detail to bicycle operations at intersections and other transition points is unmatched.”
This morning I spent some time investigating PBOT’s latest experiment in making our streets more intelligent: a traffic signal loop detector indicator light specifically for people on bikes.
On NE MLK Jr. Blvd at Morris, there’s a small, blue LED light facing traffic headed westbound on Morris.
Morris is a designated route for bicycle traffic to cross MLK (a high-volume arterial and state highway). Like many intersections in Portland, PBOT has installed “loop detectors” in the pavement that tell the signal a vehicle is present. These loops are triggered by bicycles and they tell the signal a bike rider is present; but the problem is, not everyone uses them. In order to encourage folks to use these detectors, PBOT has stenciled bike symbols in the location where you should wait, and they’ve also begun adding signage that reads, “To request green, wait on [symbol]”.
Here’s a photo of the pavement marking and the sign (these are at numerous intersections throughout the city)…
Those measures help; but there’s still no way for people to be confident they’ve triggered the light (and many people on bikes still roll over and push the crosswalk button). That’s where the new blue signal comes in. At Morris and MLK, a bright blue light immediately comes on when the loop detector is activated.
I asked PBOT’s Division Manager of Signals, Street Lighting, and ITS, Peter Koonce, to explain more about it:
“The problem that we’re trying to solve is this: there are times when a person on their bicycle reaches a traffic signal and aren’t sure if they are detected by the sensors. This is particularly problematic with new riders that haven’t learned to look for the stencil that we mark on the pavement or for experienced ones where we haven’t maintained the [stencil] marking.”
Koonce says the indicator light is necessary because it “provides information to the user” and that information, he says, is a good way to prevent users from violating the red light. The thinking is, if the person knows their presence has been triggered, they’re more likely to wait for the green — especially in situations where a small side street intersects a major arterial where auto and transit traffic is prioritized.
“We’re going to see if this helps with compliance,” says Koonce, “and get feedback on whether these should be standard practice.”
These type of indicators are commonly used in the Netherlands, and Koonce saw them first-hand on a recent trip.
From my observations, the new indicator light at Morris and MLK works well. It’s very sensitive and it came on for all the bikes I watched roll up (it also works for cars too of course). To test the sensitivity, I stood on the stencil without my bike. It didn’t come on. Then I grabbed my bike and it lit up immediately.
If you expect the light to turn green for you right away, you might be disappointed. The sensor might know you’re there, but NE Morris is not as important in the eyes of the system as NE MLK. PBOT says they have to sync the signal with others on MLK, and they put more weight on auto and transit traffic on MLK than a few bikes on Morris.
Even so, this type of innovation is key to building a network that respects bicycle traffic to the same degree it respects auto traffic.
Have you noticed this new light? What do you think?
— For more background read, Bike Science: Making sense out of traffic signal sensors.
If you have questions or feedback about this site or my work, feel free to contact me at @jonathan_maus on Twitter, via email at firstname.lastname@example.org, or phone/text at 503-706-8804. Also, if you read and appreciate this site, please become a supporter.
And (to extend a theme from an earlier post) should NOT be considered a specific bike infrastructure expense.
I’ve noticed this sort of light elsewhere but like this one it is unlabeled and cryptic in its purpose.
Why will new cyclists that don’t know to look for explicit surface stencils be able to decipher a single blue light?
If you are anywhere near the symbol and the light comes on, you can’t miss it. It’s very very bright and it’s facing directly into your line of sight.
I think what he’s trying to say is, if a cyclist doesn’t already know that a particular stencil is supposed to trigger the light, how will s/he know what the purpose of the light coming on is? That is, for anyone who doesn’t read this post, or doesn’t have a friend tell them, how will anyone know what the blue light means? Is there going to be an educational campaign or…?
This specifically is what I question, not the concept.
I think it is a great idea in need of tweaking, modification and, most importantly, standardization.
In this implementation it resembles too much a nearly identical blue light affixed to a traffic light structure at SW 19th Ave and W Burnside.
I often walked and biked past this intersection for 5 years and not ONE person knew what that little blue light was for.
Uninformed, why should we and PBOT expect that anyone will know why a particular light amongst a clutter of well know and labeled signals has come on suddenly?
Some attempt will be needed to educate the entire general public and sadly that is not BikePortland.org. Tweak this idea so that it in some what accesses the common sign linguistics of the MUTCD so there is less confusion amongst the general public.
I ride through this intersection regularly and figured this light was *something* bike-y. But it’s not intuitive at all (though I suppose after awhile I might figure out that the light only turns on when I ride onto the sensor).
Since a lot of people still don’t even know/understand how to trigger the sensors, having a mysterious blue light won’t increase their knowledge. Maybe the signage should state how to request a green light AND a message about the blue light.
Or PBOT could add to the stencil something like ‘it takes from 3-24 sec for signal to change when parking yourself here, depending on time of day–or something more concise.’ That would be vastly cheaper and perhaps just as ‘smart.’
What’s the best way to educate people on the signal detection marking? I tell people about it when I see them using it improperly, but many people really have no idea that the marking means anything.
How about telling people when you roll up next to them and see them failing to put their bike over the loops? I do.
Very cool. Someday maybe we’ll have cheap electroluminescent paint – and the whole biker symbol on the ground can light up and strobe when it activates – but until then this is a good move forward.
Unless it snowed, flooded, covered with debris or a car. On a pole everyone in line can see it.
Is there one of these at Burnside and 41st, or is that blue light for something else?
I believe that one is part of PBOT’s study for the HAWK signal there. It doesn’t appear to be the same thing.
The blue light seems like a good idea. I’ve explained to numerous cyclists that they need to put their bikes on the stencil to “trigger the signal. The usual response is a blank look. I’ve also been challenged with “How do YOU know?” and “That’s dumb.” A few have thanked me. By the way, PBOT has been VERY responsive when I’ve called to complain about not being detected.
Good work, Peter.
Yeah, it’s funny. I’ve had the same response from others.
I think this is also a good thing to know about, since when the stencil inevitably rubs off, and it’s a place I don’t frequent that often, I can be certain that there is still a sensor there, instead of trying to remember where the stencil was placed-ie.: “was it more toward the curb, or out in the middle of the lane?” This will help take some guess work out. Or, hopping on to the sidewalk and hitting the walk button.
The loop detector at NE 12th and NE Irving facing west is wildly inconsistent. I have about a 50% record of triggering it every time I go through. The only way I know for sure it’s triggered is if there is a car sitting there. I would love some kind of confirmation that I’ve actually triggered the light. Frankly, I think a lot of the loop detectors are garbage. I wish they worked better.
We’ll take a look at NE 12th & Irving based on your complaint. PBOT is committed to making our existing infrastructure work for people on bicycles, so please let us know if there are signalized intersections in the City we need to investigate further.
Peter, almost a year later, I’m be curious to learn how the stats for vehicle movements at different intervals around the 12th overcrossing–and overall–compare to the numbers collected before the changes.
I keep hearing anecdotes that, since the changes (both signals and lane configurations), when some people drive this route they now feel that movement through this area during peak hours has been slowed.
But I would guess that even if the intervals have actually improved overall, that people behind the wheel have the perception of increased delay because the backups start much earlier than they used to–or to put it another way:
the congestion (storage) has been successfully shifted off of the bridge and further back into the approaches
by virtue of that shift, congestion that was formerly concentrated on the bridge itself has now been lightened by being shared among the four main approaches to the bridge
safety and movement on the overcrossing itself are greatly improved
BUT perception when driving may not match the statistical reality, since cars start backing up further away from the bridge than they used to do
Anyway, I’m curious.
Thanks Peter. Appreciated.
since you asked, would you be able to look at the loop at the intersection of SW River Dr and SW River Parkway (leading from Riverplace across the streetcar tracks heading to moody and the south waterfront.)
In years past this loop used to work wonderfully, to the point where a rider could expect the E/W travel on river parkway to be getting a yellow as we rolled up to the intersection and get our green shortly thereafter.
Lately, if we can get the loop to sense us at all we’ll see the Crosswalk signal count down to 0 and then just flash to full walk again.
Time and again the bike users who actually follow our “sharrow” directions to this intersection are simply forced to wait for traffic to lessen and often just run the red light because it just won’t ever turn for us anymore. The other option is go down to moody, but then we are crossing 5 lanes of differing traffic and the turning streetcar tracks, rather than being able to cross them perpendicularly at the intended intersection.
Here’s one — the left turn from southbound SE 39th/Chavez to eastbound Woodstock. The signal only registers my bike if I come to a complete stop on the sensor, which means that if I’m at the end of a line of cars, it turns orange right before I enter the intersection unless I’m tailgating the car ahead of me. More frustratingly, if I’m still rolling up to the red arrow just before the phase when it would go green, it doesn’t register my presence and I have to sit on the sensor through the entire cycle. Could you increase the sensitivity of the back few loops so that when approaching cyclists roll over them, we get the same green we’d get if we were approaching in a car? If you could, that would just make my day.
Neato! They should add some signage explaining what its for though. I cross that intersection most days and figured the blue light had something to do with knowing a bike was there, but I wasn’t sure if it was itself a sensor, if it was signalling me, or if it was a cyborg about to unleash a hail of lasers if it failed to identify my retina.
This is very cool! Definitely might need some education, but it’s a great, smart idea. Do those loop detectors work for aluminum frames?
Yes. Actually, they mainly detect your aluminum rims, not the frame, due to induced currents in the circular wheels: http://www.humantransport.org/bicycledriving/library/signals/detection.htm
All of the burried loop problems are pushing municipalities to use video camera based traffic detection systems. I find these work quite well in Beaverton.
Except at Millikan & Cedar hills EB on Millikan… The trick to that one is to either have a car go ahead of you, or drift wayyy to the left when coming around the bend from Hocken, then drift back into the bike lane at the actual intersection.
I would love to see something like this done in Beaverton. So many of the loop detectors aren’t marked with the bike stencil, and even though I know that about 1/3 of the radius (for circular) away from the edge is the most sensitive area, the rate of detection is pretty hit-‘n’-miss. There is one left turn signal on my route to work that I have to run pretty consistently unless I wait (and wait, and wait) for a left-turning car to arrive–even though the detector loop(s) are marked with fading orange paint (but no bike-specific symbol; they are just outlined with paint).
Actually … you can cheat that exact signal with a helmet light with a 100% success rate. All you do is aim the light directly at the sensor; barring heavy Cedar Mills traffic the light turns green for Millikan within 5 seconds.
This works best with a steady / non-strobing light and only very noticeably bright lights will work during full daylight.
These camera sensors are dumb; they are unable to recognize objects. They are only able to distinguish a threshold level of pixel change within a certain area for a minimum trigger time.
Lens flare is your friend 🙂
D’oh! I don’t wear helmet lights. Although, I have used that trick at night before by just lifting my handlebars up enough to shine my light at the camera. Maybe my wide drifting across the lane creates just enough of a pixel disturbance to accomplish the same thing in daylight.
do they have the same thing eastbound at that intersection?
Nope, only going west. I’d love to see it for east-bound cars as well.
Bikes! East-bound bikes!
A technical question: I’m assuming the loops are electro-magnetic sensors, so it would seem carbon and aluminum framed bikes would have trouble triggering them. Competitive level road bikes really only have steel in a few bolts and possibly the bottom bracket. I realize this is only a small percentage of users, but many of us use our race bikes to get to work on as well.
My understanding is that the induction loops pick up your wheels (which are closer to the pavement) as much as they do your bike frame. So if you have a carbon frame with aluminum rims, you’ll still get detected.
Only problem would be if you have a carbon frame and carbon wheels. All I can say about that is if you choose to use an elite racing machine on the public streets (whether it’s a lamborghini or an ultra-high-end bike) you’re welcome to do it, but don’t expect the transportation system to bend over backwards for you.
Yup, that’s right. You just need metal wheels. http://www.humantransport.org/bicycledriving/library/signals/detection.htm
And from your link:
“…as of summer 2004, this signal no longer detects bicycles. Apparent changes in the pavement surface around the loop wire installation suggest that the sensor wire has been replaced and the circuit consequently adjusted to lower sensitivity by the DOT. Ironically, this intersection is on a signed recreational bike route.”
I rarely if ever trigger loop detectors with my ~1700 gm commuter wheel sets. I treat induction loop intersections as yields.
Thanks to Joseph E for that link. Very informative.
As for tripping the loop with a carbon frame / carbon rim bike, lay the bike with the the crank and chain side toward the sensor. It works for me frequently when I’m at a recalcitrant sensor. Granted I ride a steel bike, but the principal should be the same altogether.
Keep the rubber side down!
” Some” competitive level bike’s that is. According to the post by Joesph E. just above yours, more than likely your carbon rims will not trip it. If they have an aluminum braking surface, perhaps.
I learned that I was supposed to stop on the symbol to activate lights from one of the free safe biking for kids booklets we picked up somewhere. It was probably in the kids’ goody bags at the Buddy Walk last fall. Much thanks to whoever donated those to NWDSA.
We could use those here in Washington County. SW Allen and King comes to mined.
This little light could make my days soooooo much better! There are numerous spots where I don’t trust the signals to work 100% of the time, and having some means of confirmation is brilliant! Even if it’s not totally intuitive, it is still a step in the right direction. I hope to see more (and use more) of these soon!
Back before they started marking the loop symbols, some PBOT staffer told a neighborhood meeting how to line up on the circular loop cuts to trigger the light.
I was at Ladd and Division and a woman told me she had been waiting for the green light but no car had showed up to trigger it. I told her that bikes could trigger it and she asked how since we don’t weigh enough.
It’s not weight, It’s an electro-magnetic circuit and I showed her where to line up the wheels. The light almost instantly changed to green.
This is very cool! I think they should put one of those lights on Chavez and SE Taylor by the Belmont Library. Those poor buttons get pushed about a thousand times as people wonder if the light knows they are there. Maybe they should put the light on the sign that tells you to wait on the symbol–with the words “lights when activated” or something. How long would you stand at an elevator if you pushed the button and it didn’t light up?
This is precisely why we need bicycle training classes for the general population. And here I thought everyone knew how the loop sensors worked…
Maybe this explains all those bike ninjas running through red lights!
I think there is one of these at the bike signal at 41st and Burnside, but I never could figure out what it was. I’ll have to check.
Blue lights not to be confused with the backside of school zone flashing lights, or in some places, the back side of red stoplights so traffic cops can be sure if a red light has been initiated.
Please explain the function/reason of the blue lights on the back of school zone signs. Additional awareness ? Is that the purpose?
I think it is for enforcement? A cop can sit behind the sign and know for sure that it is on before he issues that $400 ticket.
The blue lights have traditionally used at a few select locations for enforcement purposes. This new concept of a “bike detected” indication (we’re talking about NE MLK & Morris, we also deployed these at SE 87th & Division) is the same device, but aimed in a way so it would be hard to miss this.
We have enforcement lights at just a few locations and to reduce confusion (please remember this is an experiment) we’ll likely develop a separate form factor for the light if the results suggest that it would be worth the investment. Your comments are helpful in encouraging us to work with the Federal Highway Administration Team that determines what the devices should look like. The intent is to develop something that is intuitive, if that’s possible.
Education is key, so please keep talking with the people on bikes that you meet that aren’t over the stencils on the pavement. I am always sharing during my travels by bike when I see someone in the wrong place. A larger education piece is needed for a lot of our transportation system elements. Finally, don’t expect these blue lights everywhere in the near term, continue to look for the bicycle stencils (that are part of the federal approved traffic control devices in the Manual) and for further information use this post from 2010 – http://bikeportland.org/2010/09/27/bike-science-making-sense-out-of-signal-sensors-39517.
Thanks to everyone for your comments and for Jonathon for the forum.
Ah, …enforcement. I’ve wondered what those blue lights were for on either side of the Alameda School crosswalk on NE Fremont.
Noticed that in the old post, Jonathan wrote:
“Personally, I would love to see a small light bulb turn on when you have triggered the sensor to let you know that you are correctly waiting in line.” How’s that for service!
The strategy I usually use if I can’t find the symbol (not covered in that article that I could tell, although I read kind of fast) is to put my bike toward the right or left side of a circular loop, so that the wheels are touching the circle in front and back (making a geometric chord of the circle). There are a few lights I can’t trigger this way, but not many. This is usually where the bike symbol is if there is one, and it seems to work pretty well.
Also, instead of a single mysterious light, how about if it’s a small sign that lights up “Sensor Activated”?
I’d rather have the city save cost on a few blue LEDs versus a sign, and then use that savings to roll this out in several other intersections. JMO. People will get the hang of it in time (as with the stencils).
I would LOVE to see this at some left turns I make. My commuter bike triggers the lights but my carbon bike doesn’t always (even when I lay it down). I sometimes have to wait a cycle and then run it because it’s safer cutting across oncoming traffic than the heavier flow behind me (at one particular troublesome light).
Good job PBOT! I encourage people in other cities to point things like this out to their bicycle advisory committees.
This is an improvement! As others noted above, needs more education and signage – as do the loop sensors in general. I have to say I’m skeptical of infrastructure that requires specialized user knowledge like this, given that there is no “bicycle license test.”
Why can’t we just go back to timed traffic lights that turn whether a vehicle is there or not? This kind of user-actuated infrastructure seems to me to prioritize traffic on arterials (who get a green light by default) over side streets (who now essentially always have to wait; with timed signals you might get a lucky green!). Given that arterials are overwhelmingly used by motor vehicles, this looks to me like another part of our infrastructure that favors motor vehicles over other modes.
I would imagine that a lot of fuel is saved by using detectors. There are still a few timed signals around town, and I see cars sitting at them idling when absolutely no one is crossing. Seems like a big waste of time and fuel.
A lot of fuel is also wasted because people use cars for trips that could easily be made by foot or bike if the infrastructure were more oriented toward non-motorized modes. Who knows if one set of wasted fuel is bigger than the other set. I do know that making biking and walking more convenient and pleasant through timed signals would have add-on benefits for health and livability.
Same thing happens at side streets when a vehicle just misses the signal as it defaults back to green for the arterial. I see this scenario All. The. Time:
– Signal is green for side street, with no approaching arterial traffic.
– Vehicle approaching on side street watches green light turn yellow/red by default just as they approach: side traffic stops.
– Now there is traffic approaching on the main arterial, but since side traffic has tripped the sensor, the light turns red for the main arterial traffic: Arterial traffic stops.
– Light now turns green for side street: side traffic proceeds
– Light then turns green again for main arterial: arterial traffic proceeds.
This scenario turns these intersections into exaggerated 4-way stops. The only difference is that all parties have to wait longer for the “smart” signals than they would for a stop sign.
This is a great concept but will need all the education that others have noted. Temporary explanatory signage at such intersections (similar to the signs placed for the new PSU Broadway Bikeway) would be most likely to reach the target audience.
And this intersection should be on my daily ride but i go one block south to Stanton as the time delay for the signal at Morris is way too long and traffic on MLK is very light in the early morning.
I live right near this intersection and have chatted with PBOT about trying to reduce the time it takes to change the light. they claimed they couldn’t make it work as frequently as, say, Tillamook for reasons that were related to the sequencing of the lights along MLK. I don’t really buy that but anyhow, that’s how its going to be
Let’s not forget the arrangement by which PGE paid $533,000 to re-time the lights on Portland arterials (39th, Division, Powell, etc.) in exchange for carbon offset credits. Ha. The logic was that if automotive traffic flows slightly faster through improvements to light timing then we use less fuel. Ha.
What I think was less central to this strategy was how pedestrians and people on bikes who might wish to cross those arterials would fare. How did the PGE carbon offset algorithm account for those of us not using gas? I asked PBOT about this back when they did this in 2007, but was never terribly satisfied with the answers I got.
Oh, is that why the waits for side streets on Powell are crazy long? All the reinforcement that “Yes, this light will turn eventually” in the world won’t do a whole lot if “eventually” means “in five minutes.” I’ve started using the unsignalized, marked crosswalks to cross Powell instead because it’s so much faster.
I’ve always thought it had to do with Powell being a state highway. I tell people all the time about the loop sensors, especially southbound at 43rd & Powell (where pushing the crosswalk button will not give you a green anyway, since the pedestrian crossing is synched to the northbound crossing.) I usually get answers like, “I tried it and it doesn’t work”, so lately I’ve been mentioning that it will take up to 75 seconds after detecting you before the cycle starts to change.
I live very close to this intersection, and I love this new blue signal. Thanks so much to PBOT.
It was a bit confusing the first time (I wasn’t sure if the loop was broken and if the blue light was a detector or indicator), but I figured it out by the second or third time. Because the Morris light can take so long, it’s especially helpful to have this signal; otherwise, I never know if the if the detector is actually working.
I’d love to see one going in the other direction, east-bound. I would also love one for NE Wheeler at N Williams at the Rose Quarter. There are some long cycles there, and it’s impossible to know if your bike has triggered anything.
And related to an earlier BikePortland story about the new bike crossing on the Springwater at SE Johnson and Bell in Clackamas County (http://bikeportland.org/2012/05/22/springwater-corridor-gets-a-new-bike-only-signal-72143): I was there a few weekends ago, and cyclists had no idea how to trigger the new bike signal. A cyclist told me it was broken, as he had waited for 10 minutes. I was a bit confused at first, too, and I know to look for the painted bicycle. I wonder if folks will figure that out or if Springwater users in that area might not have the knowledge to look for the loop.
Anyone have the number / website we use to get the stencils painted? I regularly find myself jumping up and down on Interstate @ Rosa Parks (N to W-bound) trying to trigger that signal. Since learning about those, and seeing them proliferate, I’ve been a LOT more willing to wait for a light!
I would guess that sensor is more than likely affected by the Max line. There may be an explanation if you were to contact PBOT. Or perhaps someone here may know.
Plus, I cheat there and walk/paddle through the cross walk, remount and scurry away. Not super stylish or correct, but i beats waiting 2-3 light cycles.
503-823-CYCL, press 1.
This is excellent. I’ve waited crossing MLK on Tillamook headed East and watched the countdown timer cycle back to walk signal three times before [legally] rolling through the light. And this is on a rather substantial cargo bike. In fact, there are few bikes heavier than this machine. Seems Westbound sensors work quite well but East are problematic. And I place wheels right on the circular cuts where the wires are buried.
Here’s a thought about explaining what the light means: on the “to request green wait on [symbol]” sign, add a line that “blue indicator will light when you are detected”, or something to that effect.
These should be installed everywhere that has a traffic light.
You can also look at the crosswalk signal perpendicular to the one facing you to fine tune your induction loop positioning skills. The countdown will start if the light has been red long enough and your technique is correct. Maybe we just need an education campaign about this.
I believe in the blue light.
now I can run red lights even faster… no more having to wait an entire cycle before I know that I’m not seen… just look around at the poles close by and if there’s no blue light shining at you then proceed when safe…
This is the intersection I am referring to (heading south into left hand turning lane):
I guess it is River Dr into Moody.
We’ll take a look at that location. I know there has been some construction in that vicinity. If anyone has problems you can call them into the PBOT Dispatch at 503-823-4000 or 823-CYCL.
There are a few places out in Beaverton where I’d love to see these indicator lights. In particular, there are at least a couple of intersections using optical rather than induction loop detectors, which aren’t always good at detecting bikes and you can’t tell if waving your arms over your head triggered it or not.
FWIW, even on my bike with lightweight wheels I RARELY encounter an intersection in the city of Portland where I don’t trigger the loop detector. Guess it helps to know how to do it right, and obviously there needs to be more education about this. Even in Beaverton most of the loops I use are pretty good. Most of the problems I have are at intersections with state highways (e.g., Scholls Ferry) which may be under ODOT jurisdiction.
I just noticed this light last night riding home. I happen to live on this street, and I can’t tell you how many times I’ve crossed on the red because of the long delay before it seems to notice I’m there. Kudos, PBOT, for teaching me patience!
I’ve ridden through this intersection every work day for about 8 years. It’s always been a pet peeve of mine to have other cyclists punch the ped button, because the loop has always detected me. The wait time for the signal change is variable though and I’m not 100% sure about why that is but suspect it might have to do with the band width assigned to MLK here. Another advantage of not punching the ped button is that if the signal change is triggered by the detector, the green phase is only 5 or 6 seconds long (which is plenty for a bike or car to cross) but if it’s triggered by the ped button the green phase allows a minimum time for a ped crossing, which is much longer. This makes no difference to the traffic on Morris but does to the traffic backing up on MLK.
Hope I’m not too late to the game here. One way to possibly help the uninitiated on the purpose of the blue light would be to have it in the shape of a little bicycle rather than just a round light. As soon as someone stops on the pavement marking, a little bicycle indicator lights up across the street. That might help folks know what’s up.
And maybe the word “WAIT” underneath it, until the light changes.
A great way to make this blue light easier to understand would be to make it smaller and position it just to the right of the red light on the standard 3 light traffic signal; if at all possible it should be part of each traffic light so that the little blue indicator is as close to the red light as is visible.
Placing the “side traffic noticed” signal immediately adjacent to the red light puts it where the general public is already trained to look while waiting for a green light.
When all road users collectively observe that an unoccupied side street’s red light suddenly gets a blue dot only when no one was there before, followed by their eventual green cycle, it will be easy for people to intuit that the two are directy related without any extra education.
Solving the educational part is easy, implementation will require some skill.
The biggest problem with my preceding idea is that it makes a standard 3 light traffic signal in to something with 4 lights is not compliant with MUTCD, FHWA nor any standards body nor rule set. Secondarily it adds extra expense in that it has that obvious extra failure point; planners will view it as increasing the failure rate of the signal fixture by 33% over a normal 3 light signal by the simple addition of 1 extra light.
Ultimately to get my idea rolled out without the existing laws will require the scrapping of the BLUE light idea. In fact having any light to the side (any side) of the red light puts it in potential conflict with other signal lights on more complex “turn lane only” signals.
It further dawned on me that this CANNOT use the green or yellow lights at all NOR can it blink the red light in any way; these operations would lead to inevitable user confusion and accidents.
So how about this: given a red LED traffic light that is made up of any array of several dozen individual LEDs (this is the norm for now) … Before any drivers show up at the signal on the secondary street the red light displays at 100% intensity.
After the first driver arrives the red LED light splits in to 2 halves (upper and lower hemispheres), one of which remains at 100% intensity while the other dims to 50%~75% intensity.
The two halves then begins to rotate slowly clockwise to roughly simutate a timer.
This has some major advantages:
() it will look exactly like a regular signal
() the only extra expense is a slightly more complicated controller card for the red LED part of the fixture. Much of this could be done in a programmable software/firmware module similar to the sub $25 Arduino controllers
() the user can be confused as to the new signal behavior, ignore said change and still be compliant with the law; this provides needed backwards compatibility with road users knowledge of the rules of the road
() the new “side traffic noticed” behavior could easily be remotely disabled by engineers and set to “fail safe” (to traditional operational mode) so that it fully complies with all existing laws and design codes in the event of a failure.
() the rotating animation will closely resemble the rotating hourglass of both MS Windows and Apple computers thus utilizing a “please wait” animation convention that has thoroughly embedded itself in to our collective subconscious.
You’ll notice that the new blue lights are mounted adjacent to the red lights. We just placed one at N Interstate & Lloyd.
The city of Fort Collins Colorado has been testing a similar type of indicator at the intersection of Lemay and Stuart. We are still waiting to hear what conclusions the Traffic Dept. have drawn from this test. The biking community here seems to agree that this is a great idea. If you know you are detected, you are more likely wait for the signal rather than run a red light.