All of these, including the shape of the beam and intensities in different directions is quantified by German law :)

http://swhs.home.xs4all.nl/fiets/tes.../index_en.html

Some of it doesn't translate so well into English, but this guy gives it a shot and does complain that it's not precise enough.

I was on the train the other day and another guy had his bike with him. As he was taking it down from the head high peg that MAX trains have tor you to hang your bike from (by the front wheel) I noticed that his front brake couldn't possibly be working. I looked closer and sure enough. His front brake was trashed. He was riding around town on a bike with seriously degraded braking performance. No one is going to give a rats patootie unless he is creamed in an intersection that he couldn't stop for. If they even bother to find out what made an otherwise reasonable human being just ride out into rush hour traffic. Car lights in America and car and bike lights in Germany are regulated... I don't know about Europe but I can tell you that you can have the most scientifically calibrated low beam cut-off that automotive lighting science can devise but it all goes to hell if the mouthbreather driving the car, uses his high-beams in town at night. And many do. It makes little difference where the cut-off of the low beam is, if the mouthbreather driving the pick-up has lifted it so high that children and older adults riding with him need supplemental oxygen. Even low beams that high off the ground are going to blind oncoming traffic, and they do. So what does all that care and concern about beam width, cut-off, intensity, etc. mean in the real world of lax (non-existent) enforcement of roadgoing equipment?! I honestly don't think German style attention to bicycle lights will ever be a reality in the U.S. Not in yours and my lifetime.

H

This is a strange diatribe that pertains little to the discussion at hand.

I detect some connection to the discussion.
Germany has regulations and specs on bike lighting, including specs on cutoff.
The vendors can meet these specs and therefore sell to the market.
But, it all breaks down if the user aims it incorrectly or ignores others.
If the person does not care about others, there is probably little to be done, other than giving them some feedback.

Another point is that if there were lights available that were both bright, and had a good cutoff, some of us would buy them and aim them correctly.
I made my own collumator from a plastic lid, to help reduce light intensity where it's not needed, but it's not ideal.

For those riders with very bright lights coming towards me in pitch darkness, I have my visor, ....

But both, with the proper information, can give you an idea of how they will work for you.

What you say is partly true. Lumens are for the luminous flux in all directions...if the light isn't constrained. For bicycle lights all of them have some kind of reflector to constrain the light to a certain path.

Light doesn't just "fall off with distance". Light intensity decreases in proportion to the square of the distance. So if the light has an intensity at point A of 1, if you measure the light at point B which is 2 times the distance of point A, the light isn't half as bright but the intensity decreases by 4 times. That's the problem with stating that only the lux of a light should be used for comparison. If you don't know how far away from the source the measurement is made, you can't really determine how much area is being illuminated. With lumens, you can calculate a lux...which is a useful measurement...at any distance. That will tell you a lot about how bright the beam will be. Without the distance measurement, you can't back calculate the lux for varying distances.

On any beam, you can assume that the beam shape is going to be roughly round. Even with cutoffs, the beam will eventually assume a round shape since it is going to be spreading out in a cone from the source. It may take a little while to assume a conical shape but it will get there.

Hence my point about not riding on bike paths at night. If you are really concerned about "blinding" a fellow path user, don't do it. Even then "blinding", i.e. the complete loss of visual acuity, is too strong a word. You might cause someone looking at your lights discomfort but they aren't going to lose their eyesight completely.


I disagree. Go stand a few dozen feet from in front of a car with an abrupt cutoff. The light is coming at you in a cone. You get a lot of upward spillage as an observer and the light will appear conical. If you are lower than the light source, the intensity of the light will increase. The world isn't flat and there are plenty of times when you might find yourself above an observer. A helmet light...something that I use...is even worse because the source is almost always higher than someone coming at you.

As for light brightness, I have to disagree here too. LED lights are currently about where high end halogens were around 2000. You can force about 700 lumens from an MR11 halogen by overvolting it. LED emitters are currently putting out about that amount of light. If you go to an MR16 halogen, you can overvolt it and get out 1500 lumens from a single source. You can gang up several LEDs and get that kind of output but the throw of a single source 1500 lumen lamp is much, much further than that LED. Back when I used halogens, I could throw a coherent beam across Crown Hill Park lake and illuminate the trees on the other side of the lake. That's a distance of almost 0.5 miles. Haven't been able to do that with any LED I've used so far.

Just to show that I'm not always disagreeable, I wholeheartedly agree with you here.

I guess I should be happy that someone agrees with at least one thing I say....

I agree with you that lux and lumens are good to know and compare.
And agree that the distance from the source is required to give any meaning to the claimed light intensity.
Of course, there is marketing... we do not always get that info. Others may disagree.

I recall that the inverse square law applies to a point source of radiation.
I'm pretty sure light falls off less rapidly when well focussed... (extreme case of shining a laser spot on the moon)...
Any light source, at a sufficient distance, does behave as a point source (divergent beam)
Others may disagree.

Overall bike light brightness: I was refering to, then and now, the brightness with "off the shelf" bike lights.
I think they are (on average) brighter now, due to LEDs and batteries. Others may disagree.

It's probably true that a bike light cutoff would work best if all roads and trails were flat,
so at 50 feet, the light intensity created in the first 2 ft above the ground was bright, and much less bright above 2 ft.
Since the world is not flat, having a bike light cutoff is sometimes no help at all. Others may disagree.

I like the idea of having a cutoff, I'm not saying to make them mandatory, just a choice.
Others may disagree.

cyccommute, I agree that lux ratings don't mean anything without specifying distance. I'm going to bet that bike light makers are assuming some sort of standard distance such as ten meters, which means that the ratings might very well have meaning.

Lest anyone believe that there is nothing here but arguments from irreconcilable opinions, I eagerly await a 1200 Lumen Cree XML T6 bike headlight (Amazon) to replace my Ultrafire, although the flashlight is performing acceptably. It was on someone's recommendation on one of these threads, I think cyccomute.

I kind of believe the claimed 1200 lumens even, or close to at least, and even though it's vague for describing how well a light will illuminate the area of most interest, I'm pretty confident that illumination volumes are similar enough between headlights that a higher number (if it is realistic) will be relatively brighter in practice. So I am optimistic and all is not as dark as it may seem here.

The inverse square law applies to all kinds of phenomena. Light happens to be one of them. The light propagates in a spherical wave front which increases in area as you move away from the source. The size of the wave front is related to the size of the aperture of the beam. A laser stays well focused because the radius of the aperture is very small so that the wave front stays small for a longer distance.

Niteriders of around a 2000 vintage...as "off the shelf" as you can get...were as bright or even brighter than the single emitter LEDs currently available. They were a MR-11 that was overvolted (about a 4.5 V lamp run at 6 V, if I recall correctly) so they pumped out more light then at nominal voltage. There were other 12 V systems available which put out 400 lumens at nominal voltage.

Although no one used an MR16 in an off the shelf unit, there were plenty of home brew systems around that put out 800 lumens at 12V and a few tinkerers that made lights that put out the 1500 lumen lights. Halogens are power hogs, especially the MR11 but an overvolted MR16 has about the lumen per watt output as is currently available with LED. LED has the potential to go to a much higher lumen/watt output but they aren't there yet. I currently get about the same run time on the same amp-hour battery with LED as I did back in my days of halogen.

I agree that the distance is probably 10 meters but that's only speculation without further information. To state that the lux is the measure of light to use without knowing what the distance to the target is and further state that lumens is meaningless is a fundamental misunderstanding of the issue at hand. Yes, lumens are often inflated. But we know that and, from measurements that other people have performed as well as the specifications of the emitter, we can short through the chaff and get to the wheat.

Lux is a measurement that could, potentially, be even more inflated than lumen measurement. As I showed above, I have a million lux light if I measure in the right place. Because the lux measurement is so distance dependent, even a little variation in the distance can throw the measurement off. For example, Telly says he has a 60 lux light. Assuming 10 m distance, that means that the light is pumping out 1300 lumens. I find that value a bit hard to believe. If the lux measurement is made at 5m, the lumen output of his light is 330 lumens...a value that is more believable.

I went over to Peter White and looked at some of the Lumotec lights. He has a chart on the Luxos lamps. The claimed lux is 70 or 90, depending on model. At 10m, that's a lumen output for a dynamo lamp of 1500 lumens and 1971 lumens, respectively. I kind of doubt that you can get that kind of output from a dynamo. At 5m, the lamps have a lumen output of 380 lumens and 500 lumens, respectively. Both of those lumen values are more in the range of what I've seen for dynamo outputs.

This just goes to show that assumptions can be tricky. They are often wrong.

I have 2540 lumens between my Niterider Pro 1800 and Cygolite Turbo 740. The NR is mounted on my helmet on flash mode, the power is amazing...it has the capability of freezing cars in time and space if I want to.

That said, i am looking for a brighter 3000+ light in flash for the helmet, so that the Niterider can join the Cygolite on the handlebars to luminate the runway for me.

Does the lupine betty have flash mode?

haha, I have thought the same thing! The brighter the better I say.


haha, 600 lums hurts your eyes? Playing in traffic is for the big boys and girls, where there are real dangers, with real lights coming from every direction. It looks like you can't handle it, ben.

cyccommute, that's interesting what you say. I'm going to see if the lux ratings are supposed to be done from a standardized distance. I certainly hope so, but you've raised doubts. And because of the inverse square law, the closer they measure, the bigger the potential for mismeasurement and fraud.

I did not read this whole thread so I'm not sure if it has been mentioned. They do have the Ansi F1 standard for flashlight.

http://flashlightwiki.com/ANSI-NEMA_FL-1

Many of the top name flashlight manufacturer uses this standard as a rating of their flashlight under using this standard. It also worth mentioning that reviewer of flashlight over at CPF also test the flashlight they reviewed using this standard.

I noticed bicycle light manufactor has yet to catch on. Don't get me wrong, I know there will be differemce when comparing a bike light with cutoff vs one without cutoff when using the Ansi standard, but it does comes in handy when comparing light with similar design.

Here is another interesting site, they look only at lumens (measured using an integrating sphere).

http://reviews.mtbr.com/2014-mtbr-bike-lights-shootout

I think there is such a thing as too much....
... but what that is keeps changing over time.

http://www.youtube.com/watch?v=dL9_Tldmrhs
http://www.youtube.com/watch?v=gBuiPZm6hK4

For the Lumotec lamps, I think they are measuring them at 5m which is about 1 car length. At 10m (about 2 car lengths), the lux would be around 30 for a 500 lumen and 40 for a 600 lumen light. My old halogens would be rocking a lux of 97 at 10m;) I may have to dust those off again.

Really? Why would that be? What lumens would that be at 10m? MagicShines are rated at 1000 lumens and most testers downrate them to around 500 lumens. It takes four 3.7V 18650 lithium cells to feed the LED's thirst for amp/hours. Do tell the wattage of halogen bulb(s) you were rocking in that obsolete light, and the other pertinent details of battery capacity and run time or your post becomes nothing more than suspect.

H

1. Lumens are a measure of the luminous flux of visible light from the source and aren't distance dependent.

2. Magic Shines and other single emitter lights claim 1000 lumens. Their actual output is dependent on the emitter used. A Cree XM-L T6 emitter can put out around 700 lumens under the best laboratory conditions. The actual output is probably lower. I've seen integrating sphere data that puts Magic Shines in the 600 to 650 lumen range,

3. The number of 18650 cells required to run an LED is 2 to get to the required voltage. You could run a Magic Shine on two 18650 cells for about 1.5 hours while 4 cell 2S2P packs tend to run for 2 to 3 hours in my experience. The charge capacity of 18650's depends on a number of factors but most of the ones I've seen for LED runs around 4.4 Ah.

Finally, you can find details on the old halogen lights I ran here. I still have several sets but I have to redo my battery packs to run the packs in series instead of parallel or get 14.4 V packs. I used to run 3.3 Ah to 5.2 Ah packs and got from 1.5 to 2.5 hours.

I suppose this subject has been flogged.

But here is an interesting article:
http://en.wikipedia.org/wiki/Luminous_efficacy


The halogen bulbs are incandescent, so black body radiators, and when you "over amp" them, they get brighter and their color shifts towards the blue.
That means more energy emitted nearer to yellow light where human eyes are more sensitive.
Because of the color shift at higher filament temperature, their efficacy increases. And these bulbs can be overdriven with only some loss of lifetime.

LEDs cannot be easily overdriven because their efficiency drops as temperature goes up, and there is also the danger of thermal runaway leading to possible destruction. LED's are usually driven by constant current to get constant intensity, incandescents can be driven with a voltage or current since they are resistive.

Based on that data, for a given light duration, the halogens should require a greater battery capacity compared to LEDs at the same light intensity,
because they would use more power. Also, halogens emit much more energy than LEDs out of the visible region, for example, as heat.

A long time ago, I knew someone who had a 12V Diehard battery (on a rear rack) running a ~50W halogen headlight, yes it was very bright and very heavy.

My take on this thread is that although many of us think our lights are more than bright enough already, others want more.

Although volts, resistance and amperage are related, we used to run halogens at a 20% higher voltage when we overvolted halogens. The light output is proportional to the voltage cubed and the luminous efficacy is is proportional to the voltage to the 1.3 power. However the life of the bulb is proportional to voltage to the -17. The loss of lifetime is on the order of 90% when running at 20% of nominal voltage. Since most halogens have about a 5000 hour life time, you still get 500 hours of useful light. And halogen lights are cheap.

The problem is that LEDs have the capability to put out 150 lumens/watt but they are running more in the 60 lumen/watt range. Overvolting a halogen gets you more lumens/watt. A 20 W MR16 run at 12V puts out 42 lumens/watt. Bumping up the voltage to 13.2V, gives you 50 lumens/watt. Bumping up the voltage to 14.4V, gives you 62 lumens/watt. This is a better efficacy than HID and on a parr with current LEDs. LEDs will get better but it's going to take a while.

An 20W MR16 run at 14.4V will put out just about as much light as a 50W bulb run at nominal voltage. And, from my experience, you don't need a car battery to run one. Lead batteries really aren't that good as a power source.

I agree with all that you said.

And I agree that measuring lux at 10m would be a good standard method for bike lights.
That would mean, 10m on axis, with very little reflection from other surfaces

BTW, this link shows how MTBR measures lux in a 10' x 10' room
http://reviews.mtbr.com/httpreviews-...ights-shootout


They get a direct conversion from lumens to lux shown here:
http://reviews.mtbr.com/2013-bike-li...-lux-vs-lumens

I've been reading (and have posted) in this thread from day 1, and IMHO, I believe that you are missing the point on bicycle lighting. To explain this, I prefer to give a real world example with a couple of different light I already own and use on my nightly commutes.

I recently bought a MagicShine clone supposedly rated at 1800 lumens which had a very bright spot beam which only a very small portion was utilized for viewing the road (city streets). After a fellow forum member suggested, I went ahead and purchased a diffuser lens which replaced the original clear lens and threw a much wider beam across my path, with a big hit on the distance ahead, but overall a major improvement.

And for the comparison; I have a B&M CyoT (properly aimed via the instructions) driven by a dynamohub on my other bike which is rated at 60 lux, with my guesstimate of around 200-300 lumens (compared to other quality lights I've seen).

Even though the MS has a more powerful light source, the CyoT is a far better light for commuting since the optics on it almost 100% utilize the light with almost no side spill and total vertical cutoff. I have never been told of having the light shine onto cars or pedestrians, or fellow cyclists, which is not the case with the diffused MS.

At least in the urban environment, more is NOT better if you have the majority of the light shining on everything except where it should be focused on.

That's my 0,01 cents and I hope I didn't come out condescending.

Good point. At one time lights were usually mounted on the fork crown or even further down the fork. Now most people have them up on the handlebars, which is higher than the position of all but the largest trucks and 4-wheel drive vehicles. So that's going to make even my 350 lumen light seem more dazzling than it really is.
Hard to say where the happy medium is.

+1

people don't have an understanding if they haven't used an excellent dynamo powered light built to the standard provided within the Straßenverkehrszulassungsordnung (StVZO).

http://www.gesetze-im-internet.de/st...067910012.html

I don't total disagree but for different reasons. However, a couple of issues to address first. It is helpful if you use the "Reply With Quote" button when you reply to posts. Using lets everyone know to whom you are addressing your comments.

It took me a long time to understand why the "1800 lumen" claim on lights was wrong. I'm not an electronics guy but I can learn. The Cree emitter that is used on these lights has a maximum output of around 700 lumens. The claims of 1800 lumens is highly inflated and I wish that the manufacturers wouldn't do that but they do. We just have to live with the claims and understand that the output is more realistically in the 600 to 700 lumen range. That, by the way, is about what a car provides per lamp on low beam. Now on to the lights.

I think you are misunderstanding what you are seeing for the clone. Magic Shine and their clone lights use a reflector that is about a 35 degree flood light. That means that the light intensity across the beam is greatly reduced. It will have a central hot spot but all lights have that. The rest of the light will be spread out over a large circle on the ground. When you add a diffuser, you are increasing the diffusion of the beam to something similar to a 50 degree flood light. If the light were evenly spread over the beam, a 35 degree 600 lumen light has a lux of 77. Diffuse the light over a wider area with a wider angle and the lux drops to 35. Going to a diffuser or even a wide angle flood light shots more of your light off into directions that you don't want it to go.

Going to a tighter beam puts more light in a smaller space, which is what the CyoT does as well as other tighter beam spot lights do. All you've done is purchase a narrower spot light and noticed that you have a better beam. I use a spot light...not a Magic Shine clone but another type of LED...which has a reflector angle of about 20 degrees. Back in my halogen days, I use a beam that was 12 degrees and a 7 degree spot. Because of the very tight angle, you don't get much spillage outside of the beam's hot spot and glare is reduced. In fact the 7 degree beam was too tight. You need a little spillage outside of the beam or you end up with tunnel vision.

I agree that more isn't better if you diffuse the light all over everywhere. But I don't do that. I want...and get...a bright coherent beam that lights up what I need to be lighted while still providing enough illumination to be seen against a sea of lights in an urban environment.

But that's my point, that the CyoT isn't a spotlight, and it definitely isn't as wide as a MS with diffuser (clone or not). There's a specific pattern emerging because of the optics which covers the needs of a commuter (at least mine).

Here's a beam pattern taken directly from the manufacturers site which shows the pattern (not conical as a spotlight would be) which resembles a car's headlight without the center line cutout which exists on cars so they don't blind the drivers coming in the other direction.

http://www.bumm.de/typo3temp/pics/0adbb1199c.jpg

honestly, i think that most won't understand it until they use one.

it only highlights what's necessary, which is very nice when commuting in darkness, rather than everything, which i find completely useless as one's eyes don't adjust to the dark as well.

more is not better when it comes to most things, including bike lights.

Look again. That's a spot light and a pretty tight one at that. It illuminates a "spot". Compare it to the Magicshine MJ 816E (a very floody light) or the Lupine Betty 12 (a tighter beam light) on Torchy the Battery Boy website

Your eyes don't "adjust to the dark" if you are using any kind of light other than a low intensity red light. Even the light above will completely saturate the rod cells in your eyes and render you night blind. You can see a little outside the main beam of the light because of the normal light spillage but if you were to turn off the lights and tried to walk around in complete darkness, it would be difficult. It would be impossible to ride a bike at any speed above walking speed. Don't believe me? Try it. Get away from any ambient light and see how long it takes before your rod cells relax enough for you to actually see enough detail to navigate.

The picture is also playing fast and loose with the photo set up. If they had pulled back and shined the light on the black part of the pavement, you'd see that less isn't necessarily better, either. Without that white surface to reflect the light, the black pavement would have swallowed the light and given us a better representation of what happens when you go to less light.

I think my Niterider 1800 is at 1800lums or very close to it. Its defintely not a 800 lum lamp.

The brain processes contrasts, which is exactly what the photo presents, so it is gaming the system slightly.

I don't know about you, but I don't ride anywhere without huge textural contrasts, usually in the form of water pooled on the side of the path/road, just like in the photo, or where the surface is painted, or where individual textures are interlocked.

I can't remember the last time I peddled anywhere that had a uniformly solid dark, darker or darkest background so I think your point is moot.