Headlamps: What is needed for a good beam, measurements vs. experiences and much more

On the main page:

• Beam shape of headlamps & lamp height
• Amount of light on the road from a circular beam
• Lamp height: Putting a lamp meant for 0.75 m at 1.05 m
• What do you need in a headlamp?

I'm thinking about moving those sections to this page as the main page is getting too big again (I will then put a summary of those sections on the main page).

See also: Cameras and beamshots: Calculations, interpretation issues.

The facts, not fairy tales

From the start of these webpages, and even before that with my preliminary reviews of bicycle lighting and mini-pumps on the forum of the magazine 'Fiets', it was my intention to give the facts, discredit nonsense talk by others who don't know what they are talking about yet give the impression that they do, and give explanations so all readers of these pages not only know, but understand the issues. This is for example why I made the comparison beamshots in the garden with various LED light colours, as this shows why neutral white is better than cool and warm white. There were lots of opinions, but I never saw comparison pictures such as I made. It's like that with everything on my website: I analyse, make pictures where useful, and explain.

Nobody else seems to do the effort to examine the facts and properly explain them, just as with vibrations from dynamo hubs. In 2007 this wasn't mentioned in any reviews, yet they are annoying (depending on the bike) and therefore a part of my reviews. Many people do notice but get told rubbish that this disappears by tightening a skewer, or that they are whining. It doesn't (generally) disappear and they are not whining.

Some manufacturers such as Supernova have the gall to talk about lux ratings as irrelevant, fairy tale material even, when in fact they are relevant. They just say that because they (or rather the optics firm doing the design for them) can't make a lens with high lux rating.

Then they further have the gall to give lumen values that are so far away from reality (which for their dynamo lamps are nothing more than lies as they can easily measure current through the LED and thus know their claimed output will not be reached even using datasheet values for the LEDs and then not deducting losses from the lens), that's it's in fact Supernova who is the fairy tale producer.

The rest of this page is about what makes a headlamp good or bad, and what the importance of a high lux rating is. These are again issues I have not seen discussed elsewhere. They come from analysing all the headlamps in lots of situations, but also from just thinking about what is needed, by analysing all factors that seem of importance in bicycle lighting. I have found a few similar comments to my views on e.g. the tunnel-vision problem which some headlamps give, but no emphasis on such issues for review criteria.

Reviews and measurements/beamshots

With light output measurements (in lumen or lux) and beamshots you should keep in mind that they are NOT the final word in a review. Beamshots and measurements of light output are only a small supporting part of a review. This is because with beamshots you cannot convey what you actually see while riding, and light output/lux doesn't tell you how good the beam shape is, whether there is overexposure of the near field, etc. The actual experiences, and comparative experiences in particular, of headlamps and taillamps, are far more important. I already wrote about this when I started my website in 2008, but when I started to seriously expand the lighting section of my reviews (since summer 2010) and started reviewing a lot more lights, the importance of this issue got even more clear.

Being seen: Light to the sides, light above the horizon and/or DRL

Light to the sides is useful for safety or added awareness of motorists from a long way away, in the area of intersections. I don't think DRL (daylight running lights) are needed as every headlamp gives a lot of light above the horizon anyway, but if a headlamp has DRL, then the same principles of good taillamp apply here: Shining in a cone is bad, and you want almost-collimated light for long distance visibility and diffuse light from a large surface or large width for distance estimating (see Theory for taillamps). This is part of the reason why the DRL light in the Cyo RT is an extremely bad design. It's visibility by being annoying because the high luminous intensity annoys, but especially at short distances... And that's not really when you need to be noticed, that needs to be sooner...

Optimal beam shape: Size, strength, light colour, artefacts

There are multiple factors involved to determine what is a good beam:

• Physics: Lighting up the road, so you need:
  • A certain beam width, preferably one that lights up the entire road. Seeing the sides of the road is especially advantagous in case of rain. On smooth asphalt with heavy rain you barely see anything from the road itself, and seeing the sides is good so you see in case you go too far left/right.
  • Beam length: Depending on speed you need a certain beam reach.
  • Beam shape: Wide is also good for cornering (more on this to be added).
  • Light colour that shows the colours you see on the road, or on the trail. Neutral white LEDs are better than cool white LEDs because you see more with them (yellow, brown, not all greens becoming a single green blob). This disappearing of a range of colours in cool white also results in a loss of depth perception. This can clearly be seen by looking at the photo's I made in the garden using cool and neutral white LED light: With cool light the scene looks flat. This is also why up to a few years ago some people were still mountainbiking with Halogen headlamps. See also LED light colour, CRI and experiments.
  • 2013-12-28: I realised a while ago that I missed talking about the evenness of the beam on this page, well here it is:
    That the beam is even is important to be able to see changes in the road's surface, or even objects on it. How do I know when riding, that a darker section is caused by the beam or by the road being of different asphalt? Or even how do I know a spot beside a hotspot/hotsection is darker from the hot section ending, or from a hole in the road for example? Well the answer is that it takes time... You need to analyse all the time seeing the beamshape perform over various surfaces and possibly see how you can overlook defects in the road because of it. But this costs time and continuous effort of the brain. It's quite clear that it takes away especially of the ability to see far on the road. I think this is the final part of the puzzle why no other headlamps despite fairly high lux rating come close to the Philips LBL/Saferide 80 (of course don't forget that what I wrote about the lux rating needing to be maintained at a high value...)
• Psychology: If the beam shape has sharp formed edges/corners, and/or a hard transition from beam to darkness such as the Cyo R/RT it will/can give a feeling of driving in (and being trapped in) a tunnel of light. At least in case of beams which are not very wide. Hotspots in the beam and sharp edges/corners attract the rider's attention which is bad. You will spend some time while riding looking at those corners/hotspots instead of what's on the road. These are factors why the beam shape of the Cyo R/RT is so bad. Artefacts outside the main beam are not so bad as artefacts within the main beam, they can in fact prevent the feeling of being trapped in a tunnel of light. If a beam has very sharp bright-dark edges and no artefacts outside the beam, but is very big, say 10m wide and 70m long, then I suppose this effect of feeling trapped in a tunnel of light is not present because the beam lights up such a large area.
• Physiology:
  • Beam strength: Should be strong enough to see on unlit roads, and on unlit roads with cars on a parallel road, and the worst case: On a low lying cycle path next to a road for cars.
  • Light colour (neutral white is easier on the eyes, cool white gives more eye strain).
  • Overexposure of the near field results in that you can't see well other sections that are less well lit up, this usually means you can't see at long distance. This is caused by the fact that the eye accommodates to the brightest light source. This is also why you have problems seeing anything when a bright light shines into your eyes (e.g. scattered light from car headlamps is problematical for this reason). This is the same reason why an even illumination of the road is very important.
  There are more factors that I will write up soon (ageing, amount of light needed, clarity of the lens diminishing, and someone on the Philips lighting day said this makes scattered light even more problematical as it also causes hazy bits in ones vision. I'll have to investigate that more).

A good beam for seeing far: The StVZO lux rating, why it matters

The lux rating as measured by the method StVZO prescribes (which is done at a distance of 10m) has a definite purpose: The maximum lux rating measured is taken to be the top of the beam. From there 3.4° above that position (at 10m that means 59.4 cm), the rating should drop to below 2.0 lux so that oncoming traffic doesn't get blinded.

If you look at the case of a headlamp and a camera lens they are very similar: The camera lens, or better its sensor can be compared to the front lens of the headlamp (even when the lamp uses a reflector, then you should take the lens on the front). If you look at a picture from a camera you will see the stuff that is at further distances on a road will get very close together in the picture. This is the same for a headlamp: The section of the beam on top gets spread out over a large area, the section on the bottom not so much (this assumes for the lamp that the light is almost collimated which is not quite true but near enough for it not to matter).

2013-1-10: In a real reflector (such as the one I'm designing) this is not necessarily the case at short distance. A wallprojection at short distance can give a wrong image (if you transform it to show what the illumination will be on the road) of the real intensity on the road surface... but how I described why the lux rating matters is good enough to give an impression without going into a lot of detail.

2013-12-28: I noticed that I left out something essential here which I mentioned elsewhere on my website: Reflection back to the observer (cyclist) depends on the angle... (more on this with calculations to follow).

Thus you need a lot of intensity at the top of the beam, or rather you need a difference in intensity, so the intensity must go down on a wallshot, otherwise the near field will be overexposed.

I will add some pictures of the camera and headlamp later showing this along with some calculations.

Pictures and example calculations to be added.

Note about how far you can see with a lamp: When I talk about how far I can see with a lamp, I mean how far I can see what's on the road surface, not objects higher than the road surface such as cars (esp. with reflective parts), nor hedges etc. Axa claims with their Luxx 70 plus that with it you can see 95 metres, and that might be possible to see reflective parts on cars, but anything else not unless you aim it too high, and you certainly will not be able to see the road surface illuminated clearly up to that distance unless you have the vision of an eagle or a cat :)

Is it so hard to make good optics for headlamps (and taillamps)?

I quote here what I wrote to someone wondering whether it is hard to make a lamp with good lighting pattern, slightly edited for clarification:

Well, it is apparently. For taillamps the funny things is that the old incandescents had good visibility, esp. a fairly collimated beam (the taillamps all have a lens...). Newer LED taillamps don't except for the Line plus and the Lumiring. Peter White notes on his page about LED taillamps that they put out plenty of light and thus don't need optics, which (again) shows he doesn't properly test as most taillamps' visibility is poor at long distance (and too bright at short distance). But this same phenomenon of thinking 'LEDs need no optics' occurs with the manufacturers! They (at least the Dutch manufacturers) made good incandescent taillamps, so why not good LED based ones? I think it's a case of luxury: LEDs give plenty of light so they assume visibility is good just from that. And what about being annoying? I'm not sure how that comes through their own reviewing, as I said it seems either they don't test, or they think "wow, that's bright! That's really visible! It's good!" when in fact it's a really poor taillamp. The people who did the optics and design for incandescent taillamps probably are all gone and working elsewhere or retired so they won't disagree with the views of the designers and 'testers' of the newer lamps who put little effort into actual optics.

Last modified: Mon Oct 14 05:16:46 CEST 2013