SweetLou

Ok? Your setup looks nice, wouldn't work for me since 1.5 hours is not long enough for my commute, but I am glad it works for you. But again, this has nothing to do with my question about research on which method is most effective.

chandltp

After and hour and a half, the charger still thinks the battery has a full charge. That means the run time is way longer than 1.5 hours.

chandltp

What are you running for a battery? If you said it, I missed it.

Transformer

Ctrl-f Niteflux Red Zone 4 not found.

cellery

When I drive, the thing that catches my attention most to cyclists on the road is bright/reflective clothing. Of course a blinky is necessary to catch driver attention, but bright clothing helps drivers to pick out the profile and distance of a cyclist immensely, as it is difficult to judge the distance of a flashing light.

mechBgon

Well, my subjective impressions from my various taillight testing is that if the point source is bright enough, it's quite effective. And no one wants a mammoth light on their seatpost, it would hit their thighs when pedalling, hence the slim compact designs you see on the U.S. market. I recall one of our BF members, possibly 10Wheels or 2manybikes, also posting comparision shots between that huge Foxfire light and a DiNotte (TKO victory for DiNotte, the Foxfire was barely visible in what appeared to be dwindling daylight). Perhaps the "bloom" effect of a bright point-source makes up for its small physical size?

Is it? I seem to have no problem with it. The motorists on the highway with me seem to have no problem with it either. But it never hurts to load up on reflective stuff, of course. Reflective tape can be ironed or sewed onto nylon panniers to further increase your profile, if you're using panniers:

mprelaw

I agree---it's really hard to not see a flashing light. I have a light with multiple flash options. Random flashes, from an intuitive standpoint, seems to be the logical best option.

SweetLou

Thank you, I misread what he wrote.

SweetLou

I guess no one knows of any research done on this subject. I was hoping to find some scientific studies.

CommuterRun

I don't know of any research done on this, but there are taillights that meet your criteria of having a wide viewable angle. The Cateye TL-LD1100 is a light that is always on my bikes. With six LEDs to the rear and two on each side, coupled with Cateye's OptiCube lens technology, this light is a very bright light that is viewable over more than 180 degrees. From one side all the way around the back of the bike to the other side. It also has two independently operated banks of lights, and 4 modes. At night I run one bank in a flash mode and the other bank in the constant on mode. The constant on mode is so that a motorist approaching from the rear can tell that the lights up ahead are moving, which is impossible when viewing only flashing lights. When riding on overcast days I run both banks in flash mode. By doing this on a <7.50 mile commute to work, one way, the 2 AAs needed to power this light last more than a month.

Falchoon

For most bicycles it isn't possible/practical ro put a car sized tailight on them. As others have posted they make up for lack of size by having multiple smaller lights and having them on random flashing modes.

SweetLou

I don't have a criteria and I don't really care about any one light over another. I am just interested in learning the most effective way to use the light and power at hand. For example, say a company decides it wants to produce a light that has 8 hours of run time on a battery. They come out with a circuit that produces a lot of light at little cost. Great! Now, they have to decide how to focus this light. Most lights I see today, are made with a single focused point. This will concentrate the light in a cone that is very bright.

Another way to do this is to diffuse the light. This will be less bright directly behind, but have more coverage. This is how B&M does it on some of their lights. They have a single bright LED and use mirrors to get more square inches, brighter at more angles.

The single point of light is great for selling lights. The customers pick up the light and shines it in their eyes and blinds themselves. But is this the best use of the light? Sure, if you rode straight roads and were directly in front of the cars, the bright light would go where needed. Where I live and ride the roads are curvy and hilly. A bright point of light will often not be pointed at the cars. Would a diffused light be be better? If I were trying to light up the road to see, I would want a focused light, but a taillight is not to see, it is to be seen. So, does a focused light make sense? I'm not sure, that is what I am trying to find out.

So, please stop recommending lights, this is not a "Which Light Should I Buy" thread. I have enough lights now, I probably won't be buying any new light in the near future. I was hoping for a more academic discussion on the benefits of each type, what is the best way to use the light output for out riding on roads, What viewing angle should a light be able to be seen, etc.

wsbob

4" by 6" is a size that may work for many people. Depends where on a bike the light is mounted. People mount tail lights on racks and bags. A small, Trek branded bag hangs from my saddle, attached to the seatpost; it's back end is very close to being 4" by 6". It could make sense to have the entire back end of the bag lit. I mounted a Hotshot to the top of the bag. Works fine, but it has occurred to me that at least in some riding situations, a full 4" by 6" panel of light could be advantageous.

If the Hotshot's battery and LED emitter were used to evenly light a 4" by 6" panel tail light, I wonder what level of brightness could be achieved.

Looking at peoples various improvised solutions is interesting. Walking by the big, new mega-Goodwill store the other day, I see an older, decent quality and cared for mountain bike. Rack mounted over the back wheel. Mounted on the rack was one of those little fiberglass trunks used on...like, Vespas...rounded, streamlined with flush mount tear drop shaped tail lights incorporated into the side of the trunks body. Didn't look too odd in the daylight. Curious what the bike's owner uses for power, lamps, and how it looks like lit, however he's set it up.

rekmeyata

Research? We don't need no stinkin research. All one has to do is look at tail lights and you can see with your own two eyes all the research you need.

Obviously the idea behind tail lights is for others to see you and not for you to see, I can't remember the last time I saw someone driving in reverse down a street at night using only their tail lights to see by!!

And it's very true that the bigger surface area of tail lights on cars, and their headlights, show up better to the observer then small puny bicycle lights. So we have to make up for the lack of size by finding ways to attracted attention to us. That's why on my bike I currently use the Blackburn Mars 4 2 watt led with amber side lighting as my main light attached to my seat tube, that light I leave on steady because I think there is a slight advantage of the steady light aiding the observer with distance relationship better then flashing. BUT I combine that Mars 4 with Soma Road Flare bar end lights and a Cateye LD600 on the helmet, and those flash because I also know that flashing attracts attention, not to mention saves batteries.

With the flashing attracting attention idea in your mind I do the same with the headlight system. The main light for me is a Cygolite MityCross 480 (lumens), this light remains on steady of course, it can flash but there's no point to that if you want to see at night with it, it could be useful on foggy or rainy days though. Then I added an old Vistalite Xenon flasher that has an amber lens to the front under the headlight so that the flashing will attract attention. I also added a older BLT 100 lumen helmet light to use as way of aiming the light into the eyes of drivers in their cars, and to project a flood light on the ground and to see street signs with.

Brightness of the light does matter especially if your using only one light, but how bright do you need to be in the rear? I don't buy into buying the most expensive tail light you can find business. I think if you get one significant bright light like the Blackburn Mars 4, or the Cygolite HotShot, or even the Light & Motion Vis 180 as your main light initially (if money is an problem you can buy lights in steps); then add the Soma Road Flare bar end lights which are the brightest bar end lights on the market, so bright they show up in the daylight as does the 3 lights I mentioned; then add a helmet tail light. With multiple tail lights you almost make up for the lack of size, then add flashing into the equation and you will get noticed pretty fast.

The Blackburn Mars 4 and the Light & Motion Vis 180 both have separate led's on the sides that light up amber lens, the Cygolite does not but it's more focused the the others and has a brighter appearance from the rear. However I tend to lean toward having some side lighting so that's why I have the Mars 4. Also my Cateye LD600 when mounted vertically has extremely good side lighting, as does the Soma Flares to a bit of lessor degree but highly noticeable from the side. Combine the rear lights very effective side illumination with the Vistalite front amber flasher that can also be seen readily from the side a person would have to be blind not to see me approaching from my side.

Also you can combine the lights with reflectors, but I'm not real big on them due to the fact their passive and the headlights have to be shining directly into the reflector to be effective. Still though I wear reflective clothing, reflective leg bands, reflective band on the seat bag, and reflective tape on the helmet; but I do not use any reflectors on the pedals or on the bike or wheels. Not saying you shouldn't go crazy with reflectors, just saying I don't. I rely on my lights to make me visible.

CommuterRun

I did not say that you need to run right out and buy this light. Being that I am not in the business of selling lights, I don't really care what light you use unless you want to tell me why you choose a particular light. What features does your light have that you like?

Single point of light? Why would you, or anyone for that matter, choose to purchase a light that only has a single point of light when better is easily available? The light I like has 10 single points of light in a single package. Two banks of 5, independently controlled as to mode. The only drawback I have seen to this light is that it is pretty heavy for a clip-on application, but it works great mounted on a seat post. For a clip-on I use a Mars 3.0 on the back of my helmet. It's a good light, but I wouldn't make it my primary. A Cateye TL-LD500 mounted on the back of my rear rack rounds out my taillight arrangement unless I'm towing a trailer, and covers me for the legally required CPSC certified reflector in addition to an active light as required by law in this state. When I am towing a trailer I have the same light arrangement on the bike, and a TL-LD500 on each rear corner of the trailer.

This is in no way to be misconstrued as saying these are the lights and the number of lights that you need to run. You need to use what you feel comfortable with. I saw a cyclist Monday morning with no lights, only a reflective vest. Not legal, but if that's the way he wants to roll, anything that happens, good or bad, is not my problem.

My only point is that the individual cyclist need to shop around if he/she wants better lights. They are available. Nobody needs to, nor should they, make choices for someone else.

Now, I will also take your "wide viewable angle vs. brightness" discussion one step further and state that very bright lights are not always necessary. The darker the road, the less light a cyclist needs to be seen. The more ambient light, the brighter the taillight(s) need to be. This is also true for headlights.

mechBgon

I can name some more drawbacks to the TL-LD1100... with two banks each doing their own thing, the attention-getting punch of firing everything at once is diluted most of the time. It also rapidly loses intensity as the batteries lose their initial freshness, and loses output across the board with NiMH rechargables due to the lower voltage. Unless you use lithium primaries, it never really hits its potential for long. I've had the "electric pickle" myself, and would rate the SuperFlash as the superior light in every way except visibility from the side.

I'd add the caveat "provided there aren't competing lights to blend in with." See ~6 minutes into this video: https://www.youtube.com/watch?v=JsdniW92GFQ It's plenty dark, but the cyclists don't show up too well against the traffic and strip-mall lighting, particularly with the glare from the wet road... I was there, and very glad to be equipped with a DiNotte.

Going back to the original subject, a reality check here: I don't think lights necessarily have to have a large emitter to be effective, judging from everything I've seen, filmed and photographed. I think power trumps surface area. I believe it was 10Wheels who posted comparative photos of that huge Foxfire 4AA taillight and a DiNotte 140, and it was no contest.

In what real-world scenario do I need super-wide-angle coverage from my taillight, though? Someone who's that far abreast of me has nearly passed already, they're not in a hover-car that can move straight sideways. If it were a helmet-mounted taillight where the aim can't be maintained precisely, maybe I'd value the shotgun approach. For actual side visibility, I have amber side blinkies clipped to my rear panniers on the main commuter, covering my front and side approaches.

My take: the best argument for wider-angle beam patterns is that so few cyclists actually aim their taillights properly. As an LBS mechanic, I find them aimed in all sorts of random directions, but usually downwards.

Chris516

That is exactly why I wear a reflective windbreaker, along with lights on my seatpost, back, and helmet, all set to blinking mode. Add to that, lane position.

njkayaker

The point of the flashing light is to be noticed from a far distance. Once noticed, drivers can use other cues to distances when they get close.
Flashing things are more noticeable and allow the light to be brighter for the same energy use.

rekmeyata

But there are arguments about this. In Europe and England it's against the law to have a flashing tail light because their studies have found it's safer to have a steady light on the rear. Whereas American studies are mixed but the law says flashing. That's why I use my main rear light on steady light for distance cue, and the bar end and helmet light flash to attract attention. Not sure if my idea works, but I've never been hit from behind either.

Giro

There is a fair amount of research out there on conspicuity which sometimes deals with lights. However, the studies are not without problems and is not easy to prove something actually works. See threads on BikeForums such as:
Scholarly literature on bike safety? including my post #17 with more than a few references. The Cochrane review Interventions for increasing pedestrian and cyclist visibility for the prevention of death and injuries. is particularly helpful in discussing the limitations of research in this area.

You could also run the following searches on Loughborough University's Institutional Repository in the UK:
Pedal Cycle
Bicycle (the results of these two searches are similar).
These will retrieve the links to free, full text online reports on methods of increasing bicyclist conspicuity for the purpose of avoiding accidents. For example, Motor vehicle and pedal cycle conspicuity: part 1- vehicle mounted warning beacons. Summary report.

If you want to go further afield than strictly bicycle (note, often "pedal cycle" is the term used in research literature), there are publications such as the National Highway Cooperative Research Program's Report 624, Selection and Application of Warning Lights on Roadway Operations Equipment which is available online for free in PDF format HERE. The appendices to this report have more details and references to the cited research papers; these are also available online for free HERE.

I've read most of the above. It makes for some interesting reading and helps somewhat in deciding on a variety of measures to not get hit by cars, etc.

As I noted above, it is difficult to do good research in this area. The question about controlled simulations is how well do they correspond to real life? The ideal of a randomized trials is unfortunately unfeasible because of the expense of doing a large enough experimental trial to have enough statistical power to detect a (fortunately) relatively rare event.

Compounding this difficulty is the phenomena of "risk compensation". This is thought to be in part why, for example, early experiments using New York City taxi cabs equipped with the now standard third rear brake light did not reduce rear-end collisions as much as expected when they became standard equipment on automobiles in the USA.

After reading some of the literature, you may be more knowledgeable to use your own best judgement. As Samuel Butler is quoted, "Life is the art of drawing sufficient conclusions from insufficient premises."

njkayaker

While it is illegal to have flashing rear lights in Europe, it's not clear why (I've never seen studies that show solid works better). I'm going to guess that the reason they aren't legal is because they think they might be confused with emergency uses of flashing lights (and lights flash on emergency vehicles to make them more noticeable). It's also possible that lots of flashing lights such as you'd see in a place with many bicycles, might be annoying/ugly.

One reason flashing might work better is that (in the suburban US) there are a fair number of solid red lights/reflectors. Flashing lights avoid being confused with them (in the US, little flashing red bikes mean "bicycle").

The more the better. Anyway, very, very few people are hit from behind and those that have been are likely not in a posting mood.

CommuterRun

Yes and no. The firing of the two banks is not in same time. So, because of the timing, it appears to be a steady light when they fire alternating; a flashing light the rest of the time; and a very bright flashing light when they are in sync.

As worded for dramatic affect, this is simply an incorrect statement that has no bearing on practical application. If it is true, then it is true of all battery operated lights.

This is true. However, I have never seen an application where having this light hit and maintain its highest potential was necessary.

I've must admit that I have never used the SuperFlash. The lack of side visibility is why I rate this light as inferior.

Ummm, thank you. That is exactly what I was saying about brighter lights being needed when in the presence of ambient light; or other lights, if you prefer.

However, a bright light that covers a wider viewable arc is going to be better than a single light that shines only straight back regardless how bright that single light is.


What about another vehicle approaching from a side street or driveway?

I agree. But, personally, I would never rely on only my helmet mounted taillight.

So you use three lights to cover what my one primary taillight does. That's good though. I do agree that side visibility is important.

Agreed. I aim my seatpost mounted and rack mounted lights straight back on a level plane. But I have also seen them get knocked askew from something as simple as transporting the bike in a car. Perhaps that is some of the lights you see in the shop? However I do think it is a definite step in a positive direction that at least these cyclists are using lights.

Giro

I periodically see the statement that few bicyclists are hit from behind. As a absolute number, that is true. However it is actually the most common serious motor vehicle - bicyclist crash-type.

The often cited major US study is Bicycle and Pedestrian Crash Types of the Early 1990's (FHWA-RD-95-163, all 9 MB of it available online). Of their 85 crash-types, 5 (6 including "Type unknown") are in the "Class D: Motorist overtaking bicyclist". They compared their data to the 1977 Cross and Fisher study and for this class of crashes:

1977 study, Total Fatal=166 of which 37.8% were motorist overtaking, Total Nonfatal=753 of which 10.5% were overtaking
1990's study, Total Fatal=41 of which 29.3% were motorist overtaking, Total Nonfatal=2453 of which 9.8% were overtaking.

Unfortunately, things have not gotten that much better. The 2010 NHTSA FARS data now has PBCAT crash-typing. For dead bicyclists with known age 18 years and older the 13 most common crash-types are:


The most common group of fatal crashes is the "Motorist Overtaking" group with 13.3 + 7.7 + 4.1 = 25.1% for the most common three types of "Motorist Overtaking".

These fatal crash-types tend to be relatively more frequent in lighting conditions classified as Dark-Not Lighted, Dark-Lighted, and Dark-Unknown Lighting compared to Daylight due to "Undetected Bicyclist" (Dawn and Dusk fatalities not tallied here):


Finally, these are relatively more frequent in rural than urban settings and are an increasing proportion of crashes in both settings (at least in North Carolina where the following data were collected on over 10,000 bicycle crashes from 1997 through 2009, 2008 and 2009 not graphed here). Motorist overtaking increased to about 1/3 of rural crashes and 1/10 of urban crashes:

njkayaker

And how many of those hit were using lights?

The point I was making is that not being hit isn't really a good test of whether flashing or solid lights are better. It's like arguing that meteor repellent has kept your house from being hit by a meteor.

Giro

We agree. Unfortunately, I do not think either the FARS data fields nor the subset of PBCAT data fields that NHTSA records have a field with any information on the bicycle lighting (will check when I get the chance). Even if they did, it is not a randomized trial where half the bicyclists in the USA were randomly assigned a standard brightness flashing rear light and the other half a standard non-flashing light. Instead, the cyclists who use any sort of rear light are more likely to take related safety enhancing choices in terms of route selection, retroreflective use, defensive biking, etc. I am doubtful any multivariate analysis of such data would be able to tell if flashing vs steady is better in reality. And as often noted, there is no good data on bicyclist exposure to risk in terms of hours or distance or even numbers of bicyclists in a particular environment.

The best you can do with the FARS PBCAT type data is determine what are relatively common crash-types in a particular bicycling setting. For example rural roads with a high speed limit at night vs city intersections in daylight. Then you can make better informed decisions on what are plausible ways to reduce the more common crash-types in that setting. On a rural road with a high speed limit at night, some sort of bright rear light, retroreflective things on the bike and you, an ear and an eye to the rear (mirror?) and consider taking an alternative route on a lower traffic and/or lower speed limit road. In daylight in the city, there would be some different considerations.

Based upon the various simulation studies of conspicuity of bicycles, pedestrians, motorcyclists, and other vehicles I think reasonable things to do at night are both a flashing and a steady rear light, retroreflective measures, choosing lower traffic lower speed limit routes, and being aware to the rear on straight sections and increase awareness to the sides and front at intersections.

A look into the night sky reveals that commercial aircraft use a strobe to provide a strong intermittent off-visual axis signal that our visual system detects. Once detected, there are steady lights which are easier to track in our narrow angle of high-resolution vision with a standardized color scheme (red left, green right) that helps determine the observed aircraft's orientation.

I suspect it would be better if, for example, bicycle tail lights were standardized on some distinctive flashing pattern combined with a less intense continuous light. Probably the closest we have to this in the USA are the biomotion of retroeflective pedals. Lots of other signals and signs on the highway are standardized for just this sort of reason, from red-white-red retroreflective DOT markers on trucks to highway sign shapes and colors to road markings.