"Flashlight Etiquette" for Cyclists

[swhs]

Sure, no doubt about the angular characteristics and the red color, everyone should keep those in mind when thinking about relying on retro-reflectors to get seen. In addition, low-mounted reflectors tend to get covered with snow, my pedal-mounted reflectors often come home sprayed in mud, and rack-mounted reflectors, even if they stay clean, tend to be situated above the automotive passing beams' cut-off at a distance if the road is even slightly bent the "wrong" way. IMHO, thinking how to look similar to a well-lit motorized vehicle seems a better idea than overestimating the effects of choosing wider retroreflexivity characteristics and skimping on active, visibly large area lighting.

I look at it differently. I consider reflectors an addition that can work well (the StVZO mandated low mounted reflectors on the rear fender make sense as indeed rack mounted reflectors are often too high), and which is needed especially in case there's a problem with the batteries or the cable or whatever that causes the taillamp not to light up.

@swhs: any inside scoops on Philips pulling out of pedal-cycle lighting?

No, do you think that might happen?

They didn't even tell me they changed the Saferide 60 and 80 to neutral white... I received no inside info the last 2 years. Well, not really inside info: I was told about budget cuts.

Despite these budget cuts, it seems to me more logical to continue as it fits in their business of lighting.

 

[Flamingtaco]

{automotive}


Flamingtaco, feel free to take your ideas to the general transportation lighting forum if you think you want to cover automotive as well as non-automotive. This bicycling section here tends not to get excited about lumen-killing projection lamps or dubiously reliable retroreflective devices.

This thread appears to me to be a perfectly calm discussion about lighting etiquette for cyclists. I did presume the discussion includes any bicycle lighting system. Is this thread for flashlight users only?

Not sure why you think this should be in the general transportation lighting forum as my comments regard the need for road appropriate lighting systems for cyclists, and what we are up against in obtaining them.

Confused about your comments on my enthusiasm. Maybe you mistook thoroughness for excitement?

David

 

[lectraplayer]

One thought though: since automotive lighting systems are critical to night cycling, as we have to be illuminated by them, I would say we may want to kick them around some here. So far this thread appears to be a perfectly mature discussion regarding bicycle lighting.

 

[-Virgil-]

LED light sources (coming to a car near you!) as they may produce so little light in the red spectrum as to not reflect any useable light at all.

There's plenty of red in the output of LEDs used for automotive headlamps. There has to be, otherwise stop signs and other important red signs, as well as red retroreflectors on vehicles, would not be adequately conspicuous.

I've noticed I cannot see the red part of the conspicuity stickers on the trailers at work with my bike lights.

Bicycle headlamps are unregulated in North America, and who knows what the SPD might be of the LEDs in yours. Moreover, it's entirely possible the conspicuity retroreflectors on your work trailers are of poor quality and/or past due for replacement.

Now that diffuse reflectors are extremely efficient (consicuity tape

"Conspicuity tape" is actually retroreflective sheeting cut into strips. Retroreflective sheeting, except for the bottom/dimmest three grades, is generally microprismatic. The bottom three grades tend to incorporate glass beads. None of them could accurately be called a "diffuse reflector"; they're all retroreflectors. The various grades of sheeting have different performance in terms of intensity of reflection and range of incident/observation angles through which they're effective.

The DOT's given lip service on the bad headlamp specs for three decades

Not sure what you have in mind with this. On its face, it doesn't make much sense in context of this discussion. What part of the headlamp specification do you regard as "bad"?

 

[-Virgil-]

it is very expensive to design and build a light that successfully utilizes all light from the source and does not overwhelm opposing traffic

This hasn't been true for decades. In the past, when only parabolic reflectors were available, the only way to create a sharp beam cutoff was to block off the lower portion of the reflector from "seeing" the light source. With the advent of non-parabolic complex-shape reflectors about 25 years ago (actually first developed in the early 1970s, but not widely commercialized at that time), it became not just possible but rather easy to use the entire reflector/lens area and still have a beam with a sharply-defined cutoff. Likewise, when projector-type headlamps evolved from simple ellipsoidal reflectors to complex-shape reflectors, the light losses due to the cutoff shield shrank considerably. As computers and software used to design optics have grown more powerful, this has all grown easier, not harder. And LEDs have made it easier still, as there are no 2nd- and higher-order reflections (as off bulb walls in the past) creating difficult-to-control stray light that would tend to cause glare above the cutoff.

Low-beam reflector headlamps have to block all light exiting the lens directly from the bulb as this is too bright a source.

You don't seem to properly understand the function of the bulb shield. It's not primarily to block glare directed towards other drivers, but rather to block upward stray light that would tend to create backscatter/self-glare in fog, snow, and rain. Whether a bulb shield is needed depends on the architecture of the specific lamp in question, but bulb shields generally do not significantly reduce the efficiency of the lamp.

the issue remains with bicycling to produce a system that is compatible with being on the road and providing enough light to be safe.

Numerous manufacturers are doing a very good job of it, and the products just keep getting better. The main reasons why there are poor-performing lamps on the market is that people buy them, and no regulation prevents them from doing so.

getting mfg's to even consider complying with DOT (NHTSA) rules is a very long shot due to the costs involved.

What specific NHTSA rules do you have in mind for manufacturers to comply with?

Yes, Phillips has done somethig, but they are a HUGE international company involved in ALL aspects of lighting and they own the automotive lighting realm.

No, they don't. They're just one of many companies involved in automotive lighting. They have a good chunk of market share in automotive light sources, but they don't manufacture automotive headlamps.

A golden savior might come in the form of optics designed for automotive use

I don't agree that an automotive type of headlamp would necessarily be optimal for bicycle usage.

we really need to use effective lights.

Quite true!

I do suggest they run TWO rear lamps, one solid (brightest), one blinkie, because having a solid lamp provides better depth perception and tracking

That is a very good recommendation. I agree.

The big, central problem not (yet) mentioned in this thread is that everyone's got their own ideas about what kinds of lights work to make a bicycle effectively conspicuous in nighttime traffic. And everyone's sure their idea is right because they can make what they consider a good case based on "common sense" to argue in favor of blinking lights, flashing lights, pulsing lights, moving lights, lights mounted here, lights mounted there, lights of this color, lights of that color, lights in a triangle pattern, etc. There's not a scrap of science behind any of this, just hoary anecdotes that amount to nothing useful.

We are far away from being able to specify anything usefully, whether it's a wholesale adoption of the German StVZO specifications or devising an all-new spec; what's needed is basic research on what bicyclists need and how best to specify and provide it. That research has not yet been done, and I don't hear much interest in it among the community of researchers and scientists qualified to do it. They've only just (finally) gotten around to looking at those questions with respect to motorcyclists, who have -- like bicyclists -- had inadequate and non-optimal lighting for many years.

 

[zespectre]

I haven't had a lot to add, but I've been really enjoying the information and discussion that has come about in this thread!

 

[Steve K]

<...snip...>

We are far away from being able to specify anything usefully, whether it's a wholesale adoption of the German StVZO specifications or devising an all-new spec; what's needed is basic research on what bicyclists need and how best to specify and provide it. That research has not yet been done, and I don't hear much interest in it among the community of researchers and scientists qualified to do it. They've only just (finally) gotten around to looking at those questions with respect to motorcyclists, who have -- like bicyclists -- had inadequate and non-optimal lighting for many years.

The existing rules for bicycle lighting in the USA are usually just the state's DOT laws.. something along the lines of "must be visible from 500 feet". It dates back to the era when people would use a small incandescent light powered by two D cells or something similar. The potential for creating glare or blinding oncoming traffic was zero (or less! ).

Now that we have lights that can throw over 1000 lumens out the front, it's reasonable to discuss what is appropriate. Given the current state of our national legislature, I don't expect anything being done pro-actively. Without a serious incident that creates a public outcry, there won't be a national standard. The best case scenario would be that people become aware of the benefits of a well designed beam (possibly through the League of American Bicyclists or other advocacy group), causing people to value a good beam, and purchase lights of this type. With a public demand, manufacturers would offer a wider selection, do some R&D, and we would see improved products.

The more likely outcome is that the marketing folks at bike light companies continue to advertise "we have more lumens than the next guy!", and will resist doing R&D since that requires money. The established manufacturers such as B&M, Cateye, and perhaps Philips, that already make StVZO compliant lights can easily expand their offerings into battery powered lights and grab whatever market share exists.

The subject of establishing a standard for rear lights is an interesting idea. It seems like there would be a benefit to people seeing a small red light ahead and being able to say "oh.. that's a bike and not a motorcycle. He'll be traveling slower, so I might need to move left to get by him. Better start checking for vehicles on my left side and find a hole".

 

[inetdog]

One of my friends once commented: " I don't want the driver to know I am a bicycle. I want him to think I am a semi, because he will not want to hit that."

 

[-Virgil-]

One of my friends once commented: " I don't want the driver to know I am a bicycle. I want him to think I am a semi, because he will not want to hit that."

This kind of glib, har-har-har "common sense" that leads nowhere productive.

The subject of establishing a standard for rear lights is an interesting idea. It seems like there would be a benefit to people seeing a small red light ahead and being able to say "oh.. that's a bike and not a motorcycle. He'll be traveling slower, so I might need to move left to get by him. Better start checking for vehicles on my left side and find a hole".

This is the productive way forward.

 

[inetdog]

I think that distinguishing a bike from a stationary hazard like a barricade is important, and is a good argument against a flashing light alone.
But the assumption that a bicycle is slow moving is not necessarily valid.
I can see the value in two or more light sources and/or reflectors to provide depth indication.
And I definitely agree that research is far superior to anecdotes and common sense.

 

[Steve K]

that brings to mind the work being done on autos, to have a system for them to communicate with each other and ... know how fast each other is going?? Shoot, I wasn't paying attention to the press releases. I think they were planning to use the RF spectrum for this communication, but I always thought it would be great to have the info encoded in the lights themselves. i.e. just by picking up the photons from a taillight, you would get the vehicle's speed. Maybe there's a way to ping the taillight and get a response, and thereby measure the distance? There's something similar used in aircraft for radio navigation (in TACAN, I think). That's definitely going to increase the cost of a bike taillight, though!

 

[Marcturus]

Instead of relying on a multitude of legacy bicycle lamps and generator - wheel size - wheel slip - pwm circuit - lamp combinations to indicate actual ground speed, I'd expect the kinds of active and passive sensors available to "enhance" automotive driver assist functions to be preferential means of determining presence and relative speed of other vehicles.

I doubt that Steve's other idea stands the test of reality. Just assume that others have gathered extensive experience about motorists' reactions to "pedal cycle" vs "some kind of vehicle."
SWHS, among others, has pointed out reasons why one "small red light" is not a good idea. What I dislike about his most favored, diffused taillights that are friendly on the eyes of closely following traffic and provide mentionable 2D area: they don't offer a high-luminance mode for fog and similar conditions.

One example of research (government sponsored, providing scientific backing for 1990's retroreflector policy) by automotive lighting experts failing to account for real-life behavioral (and for active bicycle lighting) components:
http://books.google.co.uk/books?id=XwKkMQAACAAJ
The overemphasis by "consumer protecting" governments and legal/marketing interests on bicycle retroreflectors makes a considerable portion of cyclists wrongly assume that these are at least a reliable fallback mode.

 

[-Virgil-]

Steve K, what you're talking about is called V2I (vehicle-to-infrastructure) and V2V (vehicle-to-vehicle) communications. Bike lanes and intersections will eventually converse with vehicles and their decreasingly-relevant drivers to minimize the likelihood of a collision. In the short run, the most promising idea for protecting bicyclists involves using DSRC and/or WiFi Direct to use pedestrians' and cyclists' smartphones as beacons perceptible to cars' driver assistance systems.

 

[-Virgil-]

One example of research (government sponsored, providing scientific backing for 1990's retroreflector policy) by automotive lighting experts failing to account for real-life behavioral (and for active bicycle lighting) components: http://books.google.co.uk/books?id=XwKkMQAACAAJ

The book's title translates, according to Google, into "Identification of bicycles in night traffic". What does it say that leads you to cite it as an example of bad policymaking?

 

[Steve K]

Reality?? Who said anything about that?

There's little chance that bike taillights will be changing, but... I think it would be great if there was a way to have either IR or visible data transmission from the bike to cars approaching the bike. I think the cars will eventually use radar or LIDAR to detect objects and track the relative speed. I think there is still value in the bike transmitting a code that says "bike". Perhaps there can be a CAN bus parameter ID assigned to this?? The cost to have a small ASIC or uC embed this code into the strobing of LEDs would be minimal.

Anyone know the range or sensitivity of the short range radars on the newer cars? I think they are used to set the follow distance while cruise control is engaged, among other functions. This means that in addition to optical reflectors on bikes, it might not be a bad idea to have a metallic corner reflector on the bike to ensure that the radars pick you up.

 

[Steve K]

Steve K, what you're talking about is called V2I (vehicle-to-infrastructure) and V2V (vehicle-to-vehicle) communications. Bike lanes and intersections will eventually converse with vehicles and their decreasingly-relevant drivers to minimize the likelihood of a collision. In the short run, the most promising idea for protecting bicyclists involves using DSRC and/or WiFi Direct to use pedestrians' and cyclists' smartphones as beacons perceptible to cars' driver assistance systems.

I generally don't like the idea of making non-motorized traffic incorporate active methods of protection, but perhaps it could be incorporated into bike lights?? I assume the power is fairly minimal and the antennas small. As far as whether I trust the car's sensors and software to not run me over... well, it can't do much worse than marginally sentient creatures currently operating the motor vehicles.

 

[Lampbeam]

I drive a bus and got blinded by one of these cyclist a couple of weeks ago. It was too bright. Bikers use noise pollution to be heard. "Unbaffled pipes save lives!" they say, while cyclist resort to light pollution.

 

[lectraplayer]

One thing I haven't seen here but have seen in bicycle discussions, is that bicyclists feel they don't get the respect from motorists with low brightness "be seen" lights without blinding everyone on the road. Some insight from the cyclists' POV may be seen here:
http://www.bikeforums.net/showthrea...t-use-more-lights-to-stay-safe-use-psychology

 

[Flamingtaco]

There's plenty of red in the output of LEDs used for automotive headlamps. There has to be, otherwise stop signs and other important red signs, as well as red retroreflectors on vehicles, would not be adequately conspicuous.

I found http://www.lrc.rpi.edu/programs/transportation/TLA/pdf/TLA-2005-02.pdf that discusses the results of tests on LED lamps meeting the SAE J578 specification. Page 8 has the SPD chart.

I also found http://www.ecse.rpi.edu/~schubert/Light-Emitting-Diodes-dot-org/Sample-Chapter.pdf that discusses eye sensitivity,

and http://zeiss-campus.magnet.fsu.edu/articles/lightsources/tungstenhalogen.html which covers the spectrum of halogens.


Without going into too much fluff:

The eye is less sensitive to red, making it a good color for tail lamps.
Halogen bulbs peak well off into the infrared and provide double the output near red as near blue, green, and yellow.
LED's are nearly opposite, with red near half of the other colors.

To prevent undue glare from white reflective surfaces mixed with red (like stop signs), I believe this means the red output in the LED's (above) would be about 1/4th of the output of the halogens (above). CLARIFICATION DESIRED

Is it required that the background color of a sign be visible at night? CLARIFICATION DESIRED

Bicycle headlamps are unregulated in North America, and who knows what the SPD might be of the LEDs in yours. Moreover, it's entirely possible the conspicuity retroreflectors on your work trailers are of poor quality and/or past due for replacement.

As warmer temp LED's have become available, bike lights have followed suit, but I feel that is neither here nor there as the aforementioned results show a glaring gap in the red spectrum for all white LED's. If the NHTSA spec is for tyext and symbol clarifty rather than color rendition (CLARIFICATION DESIRED), then LED lamps producing no red spectrum color would be in spec as long as they meet the output and direction guidelines.

CREE is a popular LED for bike lights. Of all the lights I've made and bought, both cheap and not, the old ones now in the hand of my boys are R2's, with my current selection being XM-L and XM-L2 in cool white and warm white. I feel the warm white is a great color for on-road cycling, but it is not adequate for signeage as even that bump you see in the Cree document in the warm part of the spectrum peaks just past yellow, deflates by half near orange, and only left with 20% total power by red.

The conspicuity tape on our brand new trailers is made by 3M. This was my reference as we did not have any trailers with the tape until this summer (that I have seen).

Not sure what you have in mind with this. On its face, it doesn't make much sense in context of this discussion. What part of the headlamp specification do you regard as "bad"?

The main problems are the amount of light permitted to into the oncoming lane, and a lack of a relation between lamp height and aim angle. Both of these contribute to the glare suffered by oncoming traffic, both can be improved upon for both motorized and non-motorized vehicles, and both (I believe) were addressed by the ECE a very long time ago.

 

[Flamingtaco]

This kind of glib, har-har-har "common sense" that leads nowhere productive.

I got a seriously good chuckle out of this, I think a good laugh now and then in a serious conversation provides great relief, but I do not moderate this forum.

That is somewhat the angle I am currently taking. Not to make drivers believe I am something in particular, but to confuse them as to what I am. I can echo the comments about riders getting no respect when using low watt lamps on the front, and feel the same applies for small blinkie lights on the rear. Having my 2W Solas on full burn definately gives them more pause than when in blinkie mode, but the three feet of LED flex tape I've added around the pannier has really sent them for a loop. When they do not know what they are dealing with, it puts them in an uncomfortable place, and they slow. With the LED tape and the Solas on full or ramping, they still slow. With the LED tape and the Solos blinking like a regular blinkie, I get treated like a cyclist with a blinke. They know that small red blinkies mean cyclist/runner/walker.

Transmitting data to cars... I don't think this would be a good thing for cyclists considering our vulnerability compared to motor vehicles. For this to be a viable system it needs a range of a few blocks, and at that strength you can build a scanner to track a cyclist a mile or more.

I'm thinking along the lines of a rear lamp system that incorporates a dual ultrasonic tranceiver to discern angle of approach and approach speed, and alerts with a low-power, high speed strobe during the day, and by overdrive, high-speed flashing the RED LED's at night (normally constantly on at night) for the driver, and audible for the cyclist, when it detects movement that falls within the parameters for a collision.