Which light is brighter?

[gaz]

It didn't say they were the top ten, just ten popular lights.

GC

Which is my point. They just cherry picked which ones they wanted and not necessarily picked lights that people are actually using on the road.

 

[GrasB]

I see someone else has highlighted the HC provision on dazzling others with lights.

The Sunday Times ran a piece a couple of weeks ago on cycle light brightness:
http://www.thesundaytimes.co.uk/sto/news/uk_news/National/article1238554.ece

"The brightness of 10 popular light-emitting diode (LED) bicycle lights was tested at a photometric laboratory. On their lowest setting and with angular adjustment to account for pointing down at the road, eight of the lights exceeded the legal maximum level for dipped car headlights. At their maximum brightness setting, nine of the lights were brighter than a car headlight."

I very much doubt that's what they actually found but putting what they did find, that non-focused beam bike lights have a higher intensity than a car headlight, into an understandable form by all non-technically inclined people is rather difficult. The general public don't want to deal with the difference between brightness & intensity, however this is what light manufactures are constantly having to play with.

Good see-by bike lights often come out at around 1-4lumen/mm2, but have 300-80mm2 of surface area, respectively, they emit light from. My Alfa nominally has a surface area of around 4400mm2 per headlight but a surface brightness intensity of around 0.6-0.65 lumen/mm2 above it's cut-off (and some people consider that objectionable), that's a whopping 2750 lumen x2... not counting the light that's thrown out bellow the cut-off where the apparent intensity is much higher!

 

[GrasB]

Which is my point. They just cherry picked which ones they wanted and not necessarily picked lights that people are actually using on the road.

Actually I think they're just dumbing down the results to their audience. LED bike lights often have a much higher intensity (aka surface brightness) than car head lights. A 150 lumen Lezyne mini-drive has a apparent light source intensity to rival most HID/Xeon car head lights.

It's not the brightness that's the problem, it's the size of the reflector being used. This is why I have so many problems with badly aimed bike lights, my dark sensitivity is right up at the top end of the scale. Car head lights are bright enough to make my pupils contract reducing the amount of after image to virtually0 even after a main-beam pass, it just takes a second or two to regain my night adaptation. Bike lights however aren't bright enough to induce significant pupil contraction leading to a trailing after image which lasts several minutes.

 

[outlash]

CCD/CMOS sensors are more sensitive at longer wave lengths (towards the red end of the spectrum). So on my D70, for instance, an Osram deep red LED looks brighter than it's normal red counter part, however a human's eye says that the normal red LED is brighter! Also before anyone can make that call from those photos people need you to take a photo of the lights which aren't fully saturated (your first picture). Then we need the following information:
The nominal LED wave length
The LED spectrum spread
The IR cut off filter wavelength
The CCD/CMOS sensor in the camera
The red quell balancing algorithm of the camera

Only then does anyone have a chance of really answering your question. It would also be nice to have the luminosity plot of the lenses used on those lights. A tight centre beam may be hiding a higher luminosity at extreme angles - throwing 30% of your lumen out in the +/-80 to 135 deg band will significantly reduce the centre spot brightness compared to a lens which has 0 illumination beyond +/-90 degrees.

Sorry to take this off topic, but that's only sort of true. Whilst certain sensors are pretty sensitive at the red end of the spectrum, DSLR's have an IR cut off filter placed in front of the sensor. Problem is that the filters almost universally cut into the visible red wavelengths which can dramatically cut down their sensivitiy. Hence DSLR's are modified by removing the filter for astrophotography because a lot of the light from emission nebulae is red.

I think in this case you'd be better off using a light meter or putting them at the end of the garden and forming your own opinion .


Tony.

 

[GrasB]

Sorry to take this off topic, but that's only sort of true. Whilst certain sensors are pretty sensitive at the red end of the spectrum, DSLR's have an IR cut off filter placed in front of the sensor. Problem is that the filters almost universally cut into the visible red wavelengths which can dramatically cut down their sensivitiy. Hence DSLR's are modified by removing the filter for astrophotography because a lot of the light from emission nebulae is red.

Exactly is why you need information I asked for:
The nominal LED wave length
The LED spectrum spread
The IR cut off filter wavelength
The CCD/CMOS sensor in the camera
The red quell balancing algorithm of the camera

This basically tells you how the red response of the camera & all the filters to normalise the photon capture across the spectrum is applied to the LEDs you have.

 

[Crankarm]

Given that so few bicyclists actually use lights on their bicycles excessively bright lights are hardly a widespread problem! When I do come across them they are generally seen on expensive road bikes ridden by over weight Mamils tanking it along who don't give a crap about any other road users.

 

[fossyant]

Given that so few bicyclists actually use lights on their bicycles excessively bright lights are hardly a widespread problem! When I do come across them they are generally seen on expensive road bikes ridden by over weight Mamils tanking it along who don't give a crap about any other road users.

Mamils dont usually take their expensive bikes out at night. weekend warrior's

 

[outlash]

Exactly is why you need information I asked for:
The nominal LED wave length
The LED spectrum spread
The IR cut off filter wavelength
The CCD/CMOS sensor in the camera
The red quell balancing algorithm of the camera

This basically tells you how the red response of the camera & all the filters to normalise the photon capture across the spectrum is applied to the LEDs you have.

Perhaps, but TBH you could take an image which isn't over exposed and run it though say, Maxim and get the A/D values at the centre of each light and tell which one is brighter that way, but isn't that a little bit fussy? Maybe just using your eyes might be a bit more practical.




Tony.

 

[glasgowcyclist]

Which is my point. They just cherry picked which ones they wanted and not necessarily picked lights that people are actually using on the road.

Well, you don't know that so rather than guessing at it, why not write to the journalist and ask him which lights he submitted to test and how he arrived at his choice? (It's highly unlikely he'll rake through the comments below the piece.)

I'd be interested to learn what they were too.

GC

 

[GrasB]

... Maybe just using your eyes might be a bit more practical.

as long as you've actually got the lights in front of you, yes.

 

[RiflemanSmith]

Any one else who has got the see and be seen front lights cable outter worn away where it goes into the unit?

 

[Crankarm]

Mamils dont usually take their expensive bikes out at night. weekend warrior's

Well the ones in and around Cambridge have them on whether it's night or day, flashing as well. Muppets.

 

[RiflemanSmith]

What's wrong with having them on flashing?