Bicycle lighting question

Quoted message said:

Dear Tim,

What's developing in LED lights? That is, what's changed in
the last five or ten years in light emitting diodes?

Have LED's gotten significantly more efficient, putting out
more light (or longer) for the same electrical input? Any
figures?

Or have they become significantly cheaper at the level
needed for bicycle lighting? Any prices?

Are LED's reasonably expected to expand beyond the small
niche of bicycle accessories, traffic lights, and guide
lights?

I'm willing to believe that there's been the kind of
progress that people seem to be assuming in their
predictions, but I'm looking for some figures, such as a
150% increase in light output per amp/hour since 200x or a
75% reduction in cost per unit since 199x.

Somebody needs to tell Carl about Google! ;-)

--
Frank Krygowski [To reply, remove rodent and vegetable dot com.
Substitute cc dot ysu dot
edu]

Steven M. Scharf said:

Matt O'Toole said:

And you do? My, you're feeling grand today. Please spare us, for once!

A great deal. Especially about thermals and leakage. I probably spend
more time at work dealing with this than anything else. If you don't
want to learn anything about the subject then just don't read my replies.

Quoted message said:

They are now, but they'll come down in price. Demand for these for
architectural lighting, etc., will be high. Ask an architect or property
manager who's been sued by someone who tripped on a sidewalk, because a
light
bulb was burned out.

This is the great thing about the fluorescent replacements, or even
those PAR-90 replacement lamps with ninety 1 watt Luxeon LEDs. They last
a very long time.

But using high power LEDs (3W-5W) presents a lot of problems. You can't
cool the bulb with natural air flow because the fixture has to be
sealed. Collimating the beam is inefficient.

Quoted message said:

Cateye has no trouble with this -- check out their 5 LED design. Its beam
pattern is fine, it just doesn't put out enough light. LED or not, it's
still
only a watt or so.

Scaling up to provide enough light is not easy. You don’t just replace
the LEDs with higher power LEDs, you need to redesign the whole thermal
solution. It’s a very difficult problem because unlike a regular PC
board you can’t simply position the components wherever you want in
order to have the best thermal solution, the LEDs MUST be in specific
positions. You could get fancy with heat pipes, but this gets expensive.

Quoted message said:

I look forward to the new 3W LEDs hitting the mainstream, for battery and
generator powered lights alike.

3W is more manageable than 5W take three 3W LEDs, collimate the beam,
figure out a way to manage the thermals, and you'll be close to the
brightness of a 5W quartz-halogen bulb. Not saying it’ll never happen,
but that it’s more of a challenge than you realize.

Dear Steven,

Am I right in summarizing your points like this?

At more than a few watts, individual LED lights face serious
cooling problems.

Multiple low-watt LED lights spread out on a board can
combine to replace a diffuse light (like street lamps or
turn signals), but don't do as well for a beam light (like
headlights).

That is, cooling and focus are problems for LED lights.

Carl Fogel

Quoted message said:

I'm willing to believe that there's a bright future for LED lights
(sorry, couldn't resist it), but I haven't noticed much rise in volume--
the ones that I see are in new traffic lights, on keychain flashlights,
and blinking on bicycle accessory lights.

A little homework for you, Carl. Find an old (i.e. 10 yrs) mass market
catalog. JcPenney might be a good example. Look for LEDs. Then do the same
in a current catalog. Report your findings.

The same exercise with, say, a Radio Shack catalog of >10 yrs vintage might
also be interesting.

I'm sure you'll find more LEDs in the newer catalogs. I don't know the
magnitude of that change.

--
Dave
dvt at psu dot edu
Fri, 17 Dec 2004 14:30:42 -0400

=== Posted with Qusnetsoft NewsReader 3.1 Unregistered

Quoted message said:

Dear Frank,

How much have LED prices "certainly dropped in the past
three years"?

Sorry, I don't have figures at hand. Try Google. But looking at
consumer products, prices have dropped significantly.

Quoted message said:

And why won't the price of conventional lights drop just as
fast or faster?

In Edison's day, they probably did. But by now, the technology of
production is mature. It's very unlikely that some massive breakthrough
will occur for incandescents.

Quoted message said:

I'm not arguing against the claim that LED lights will
become much cheaper and more popular. I just want to know
what I'd say to someone who asked why he should actually
invest in an LED light company.

I wouldn't know, either. Investing is another matter entirely.

Quoted message said:

For the same electrical input, do LED's produce as much
light as conventional lights?

From what I've read, IIRC, they do in red (where white incandescent
light is actually filtered to red, with heavy losses). For white light,
LEDs are just now passing ordinary incandescents. Halogens were still
ahead of white LEDs in the lumens/watt race last time I looked, but
halogen efficiency versus time has been flat for many years, while white
LED lumens/watt is still rising significantly.

What do they cost to produce

Quoted message said:

compared to conventional lights--including expected lifetime
and replacement costs? Should we expect to see LED lights
replacing car headlights, street lamps, and the familiar 100
watt bulbs and fluorescent tubes in buildings?

They're already out there in ever-increaseing applications, and
appearing in more and more places. Look at truck and bus taillights,
and taillights of luxury cars, and high-mounted stoplights of most cars.

Car headlights will probably take quite a while. Street lamps, closer
to "never" since high voltage sodium lamps are much more efficient (and
nobody cares much about the terrible color rendition). I'll bet some
room lighting will be LED in just a few years, but it'll be specialty
stuff. Flourescents are better for that.

You might like http://ledmuseum.home.att.net/ledleft.htm

--
Frank Krygowski [To reply, remove rodent and vegetable dot com.
Substitute cc dot ysu dot
edu]

Quoted message said:

Dear Steven,

<snip>

Quoted message said:

That is, cooling and focus are problems for LED lights.

I believe this to be the case.

The Lumileds web site has an extensive application note dealing with
thermal issues. These issues are non-trivial. The way they get increased
brightness is to increase the voltage and current. This increases the
temperature, and necessitates better thermal solutions. Heat is the
enemy of incandescent type lamps as well as of LEDs. It’s not often not
practical to use active cooling (a fan), you have to depend on
convection cooling. But convection cooling is also difficult when you
want to have a light with no vents. Unlike an incandescent lamp, where
the heat radiates out from the bulb and can be dissipated through the
lens and the sides of the enclosure, the heat from the LED is more
toward the rear, where the junction is located.

One person used to harp on the “white hot metal” filament in
incandescent bulbs as a reason why they just could not be efficient,
while failing to understand that junction temperatures and leakage of
semiconductors correlates as a problem in LEDs. It’s a common
misunderstanding by non-engineers, who don’t understand how power and
thermals scale in semiconductors.

I think that the collimating issue can eventually be solved, it'll never
be as good as a point source in the center of a matched reflector, but
it will be good enough, and the intensity and quantity of LEDs can be
increased to compensate for the less efficient optics. But this is a
compromise on efficiency, which is of great importance to bicyclists.

The LEDs in traffic lights are not 1W, they are closer to 100mW.
Typically 121 LEDs are used, with a total power of 14W (116mw per LED).
And of course it's easier to make red, green, and yellow LEDs than it is
to make white ones.

Realize that the use of LEDs in traffic signals and rear car lights is
NOT based on the initial cost (which is much higher than using a
filament bulb), or on the intensity (which is equal to that of a
filament bulb). It’s based on maintenance costs and safety (no burnout).

Thermodynamics and electromagnetic field theory are the two classes that
turned out to be the most relevant for most of what I do these days.

Steve
http://bicyclelighting.com

--------------------------------------------------
"If you can't answer a man's argument, all
is not lost; you can still call him vile
names."
Elbert Hubbard
--------------------------------------------------

Tim McNamara said:

Quoted message said:

Or have they become significantly cheaper at the level needed for
bicycle lighting? Any prices?

Beats me, ask an electrical engineer.

The low power LEDs are a commodity item costing a couple of cents each,
if that much. The very high power LEDs (3-5W) are still very expensive
due to yield issues. The mid-range LEDs, 1W or so, are around $3 each in
large quantities. Note that traffic lights use 100mW or so LEDs, which
are around 20 cents each in quantity.

Quoted message said:

Presumably the manufacturers are looking for ways to enlarge their
market into other areas- automotive headlights, in particular since
that'd be millions of sales per year.

For "being seen" applications, and low wattage indicator lamps, LEDs
rule. For applications where a beam needs to be projected, it's only
when a weak beam is adequate. Headlights are moving to HID. It is not
likely that LEDs will ever be used for motor vehicle headlights.

Quoted message said:

How the hell would I know? I'm a psychologist! :-D

Ah, that explains a lot. Why didn’t you admit this earlier?

The fabrication process for the low power LEDs is static. For the higher
power LEDs, it's all down to figuring out a way to manage thermals.

Quoted message said:

Quoted message said:

Quoted message said:

As a single product, no. As a class of products, hub gears were

The only figure that I see in your post is a 90% figure--you firmly
believe that the old 3-speed is the optimal gear system for about
90% of the public. Since (at least in the U.S.) 90% of the bicycle
public shifted (sorry, couldn't resist it) to a different gear
system years ago, that would raise some questions in any intro to
economics class, wouldn't it?

The 3 speed hub is woefully inadequate in my area. Even the multi-use
paths have hills where a 3 speed rider would be doing a heck of a lot of
walking.

Quoted message said:

Sure. The amswer is the market often buys (1) what's marketed most
heavily and (2) what's available. Whether or not it is the best
product for their individual use is a different question. Look at the
SUV market, for example, with only a tiny minority of buyers having an
actual practical use for the things.

Actually, where I live, a majority of buyers have a practical use for it
about 0.1% of the time. If you ask 95% of the people in my area why they
bought a 4WD SUV, the answer will be so that they don't have to put on
tire chains when they go to the mountains in the winter (tire chains are
often mandated by the highway patrol, mainly to slow people down, even
when they are not needed for traction, but 4WD with snow tires are
usually exempt).

Quoted message said:

However, the consumer may be
waking up to reality with sales of the biggest Ford SUVs being down
20% over the past year. Ford's current strategy is to try building an
even bigger version, apparently thinking that the consumers feel the
current product is too small...

This is not their strategy. They have just announced that they have
discontinued their behemoth Excursion due to poor sales. There are no
plans to build anything larger than the Excursion.

Steve
http://bicyclelighting.com

--------------------------------------------------
"If you can't answer a man's argument, all
is not lost; you can still call him vile
names."
Elbert Hubbard
--------------------------------------------------

Steven M. Scharf said:

Quoted message said:

Dear Steven,

<snip>

Quoted message said:

That is, cooling and focus are problems for LED lights.

I believe this to be the case.

The Lumileds web site has an extensive application note dealing with
thermal issues. These issues are non-trivial. The way they get increased
brightness is to increase the voltage and current. This increases the
temperature, and necessitates better thermal solutions. Heat is the
enemy of incandescent type lamps as well as of LEDs. It’s not often not
practical to use active cooling (a fan), you have to depend on
convection cooling. But convection cooling is also difficult when you
want to have a light with no vents. Unlike an incandescent lamp, where
the heat radiates out from the bulb and can be dissipated through the
lens and the sides of the enclosure, the heat from the LED is more
toward the rear, where the junction is located.

One person used to harp on the “white hot metal” filament in
incandescent bulbs as a reason why they just could not be efficient,
while failing to understand that junction temperatures and leakage of
semiconductors correlates as a problem in LEDs. It’s a common
misunderstanding by non-engineers, who don’t understand how power and
thermals scale in semiconductors.

I think that the collimating issue can eventually be solved, it'll never
be as good as a point source in the center of a matched reflector, but
it will be good enough, and the intensity and quantity of LEDs can be
increased to compensate for the less efficient optics. But this is a
compromise on efficiency, which is of great importance to bicyclists.

The LEDs in traffic lights are not 1W, they are closer to 100mW.
Typically 121 LEDs are used, with a total power of 14W (116mw per LED).
And of course it's easier to make red, green, and yellow LEDs than it is
to make white ones.

Realize that the use of LEDs in traffic signals and rear car lights is
NOT based on the initial cost (which is much higher than using a
filament bulb), or on the intensity (which is equal to that of a
filament bulb). It’s based on maintenance costs and safety (no burnout).

Thermodynamics and electromagnetic field theory are the two classes that
turned out to be the most relevant for most of what I do these days.

Steve
http://bicyclelighting.com

--------------------------------------------------
"If you can't answer a man's argument, all
is not lost; you can still call him vile
names."
Elbert Hubbard
--------------------------------------------------

Dear Steven,

Aha! Details and figures!

Somehow I got the impression that there were street lamps
using arrays of 1-watt LED lights. Thanks for clearing up my
misunderstanding--someone must have meant traffic signal
lights and enough ~0.1 watt LED lights for the 14 watt
design that you mention.

Thanks again,

Carl Fogel

Quoted message said:

For the same electrical input, do LED's produce as much
light as conventional lights? What do they cost to produce
compared to conventional lights--including expected lifetime
and replacement costs? Should we expect to see LED lights
replacing car headlights, street lamps, and the familiar 100
watt bulbs and fluorescent tubes in buildings?

they are equal to halogen lights watt for watt. but they are far more reliable.
well right now the 5 watt LED has problems. but the 3 is a really good deal.
I doubt they will replace streetlights but they do use less power.

--
Knight-Toolworks & Custom Planes
Custom made wooden planes at reasonable prices
See http://www.knight-toolworks.com For prices and ordering instructions.

Steve Knight said:

Quoted message said:

For the same electrical input, do LED's produce as much
light as conventional lights? What do they cost to produce
compared to conventional lights--including expected lifetime
and replacement costs? Should we expect to see LED lights
replacing car headlights, street lamps, and the familiar 100
watt bulbs and fluorescent tubes in buildings?

they are equal to halogen lights watt for watt. but they are far more reliable.

For headlights, where you want a focused beam, LEDs are not a great
choice. Either you have to use a bunch of low power LEDS to get enough
light, and then you have to collimate them into a beam, or you have to
use the high power ones which are expensive and don't have a long lifespan.

Even the 3W LEDs are still quite expensive. The 1W ones are reasonable.

Remember that most of the low power front LED lights (<3W) are more
about being seen than actually projecting a useful beam.

Also, a 5W incandescent bulb will provide a much better beam than a 5W
LED. It's not just about wattage, it's about how much of the input power
is wasted due to the filtering to get the proper wavelenghth for white
light, as well as the inability to focus as much of the light that is
produced into a beam.

Quoted message said:

well right now the 5 watt LED has problems. but the 3 is a really good deal.
I doubt they will replace streetlights but they do use less power.

They do not use less power, for the same amount of light.

O

Quoted message said:

Quoted message said:

they are equal to halogen lights watt for watt. but they are far more reliable.

For headlights, where you want a focused beam, LEDs are not a great
choice. Either you have to use a bunch of low power LEDS to get enough
light, and then you have to collimate them into a beam, or you have to
use the high power ones which are expensive and don't have a long lifespan.

there are two companies that make nice LED lights for bikes using three 3 watt
LEDS with separate reflectors. they work pretty well.
I think the 3 watt ones are about 14.00 and they have a long life. the 5 watt
ones have a poor life span and the generate way too much heat.

Quoted message said:

Also, a 5W incandescent bulb will provide a much better beam than a 5W
LED. It's not just about wattage, it's about how much of the input power
is wasted due to the filtering to get the proper wavelenghth for white
light, as well as the inability to focus as much of the light that is
produced into a beam.

I have not heard of anyone filtering the new LEDS and there are good lens out
there for the new LEDS though that's one of the most important part is getting
the lens right. the advantage of a LED though is it is small so you can use a
smaller reflector. this is why HID lights can be so small.

--
Knight-Toolworks & Custom Planes
Custom made wooden planes at reasonable prices
See http://www.knight-toolworks.com For prices and ordering instructions.

Steven M. Scharf said:

For headlights, where you want a focused beam, LEDs are not a great
choice. Either you have to use a bunch of low power LEDS to get enough
light, and then you have to collimate them into a beam, or you have to use
the high power ones which are expensive and don't have a long lifespan.

I think the optics for LEDs are getting there. I'm more familiar with the
LED headlamps for hiking than bike lights, but I'm impressed with the new 1
W LED Princeton Tec Eos that I got last month. It projects a beam pretty
well. I would say it has good light to 20-25 meters in the main part of the
beam. On a dark trail, it is bright enough to turn blackness 70 m away into
recognizable tree shapes... not very useful at that distance, but it is a
lot more than I expected coming from a 3 small LED headlamp. It also has a
good squarish spill that is very useful. I find this headlamp to be enough
for 20-30 mph descents down a dark twisty trail on cross country skis. Very
impressive for a light that takes 3 AAAs. I don't bike in the dark
nowadays, so I can't tell you how it works on a bike, but I think the skiing
is comparable to biking at least where I live in the winter (light in summer
is not an issue). The white surface of snow is brighter than most roads,
though most bikers here ride on some of the same snowy trails/multiuse paths
in the winter. I wonder how the optics in this light compare to the 1 W
LEDs on "to be seen" bike lights? I wonder how this would compare to
generator lights if you found a way to mount it low on a bike?

Princeton Tec is coming out with a 3W headlamp next spring, hopefully with
similarly good optics. I think the 1 W LED on the Eos is already bright
enough for winter biking/skiing/running on the snowy trails where I live, so
I can't wait to see how bright the 3W model is.

Here's a review of the Eos:
http://www.flashlightreviews.com/reviews/princetontec_eos.htm
A thread that I started on the Eos last month: http://tinyurl.com/5mxob

Andrew Lee said:

I think the optics for LEDs are getting there. I'm more familiar with the
LED headlamps for hiking than bike lights, but I'm impressed with the new 1
W LED Princeton Tec Eos that I got last month. It projects a beam pretty
well. I would say it has good light to 20-25 meters in the main part of the
beam. On a dark trail, it is bright enough to turn blackness 70 m away into
recognizable tree shapes... not very useful at that distance, but it is a
lot more than I expected coming from a 3 small LED headlamp. It also has a
good squarish spill that is very useful. I find this headlamp to be enough
for 20-30 mph descents down a dark twisty trail on cross country skis.

I like night hiking and XC skiing too. I use a Petzel 3AA (also uses
some sort of proprietary single battery) headlamp with a Krypton bulb.
It provides more light than I usually need. Got caught out at night
doing a descent in Yellowstone once, which convinced me to always bring
it along.

I think I like the new hybrid headlamps with one regular bulb and then
some LEDs. You can choose long run time with the LEDs or a bright beam
with the regular bulb. But they do have one big drawback. One very
useful feature is an adjustable spot-to-flood beam (Mag-Light,
Streamlight Strion, some Petzel's). You can't get this with an LED
flashlight, or even a hybrid.

begin quoting Nubo <[email hidden]>:

Quoted message said:

difference in the response of motorists. It's one thing to say that a
light is visible from 1 or 2 miles (hell, a candle is visible from a
mile) and quite another for a motorist to actually recognize it and be
concerned about right of way.

[...]

Quoted message said:

rudely bright. But bright enough. And, this is in measure due to the
significant sidespill light which others refer to as wasteful.

How? I mean, unless you're just about to round a corner, someone 1 or 2
miles away on the same road with a line of sight to you is in the central
beam of a headlight.

.... for that matter, how do you determine that a motorist recognises you
two miles away? Even at 60mph closing speed, it's two minutes before they
reach you. They're just a pair of headlights; at 2 miles, you'd be lucky
to hear their engine slow, or anything like that.

And unless you ride somewhere with flying cars, I don't see that up/down
overspill helps much.
--
David Damerell <[email hidden]> Distortion Field!

Steve Knight said:

there are two companies that make nice LED lights for bikes using three 3 watt
LEDS with separate reflectors. they work pretty well.
I think the 3 watt ones are about 14.00 and they have a long life. the 5 watt
ones have a poor life span and the generate way too much heat.

Are you sure about three 3 watt LEDs? That's 9 watts, and 4 AA batteries
would have a hard time powering such a light. $14 wouldbe very cheap for
such a light, considering the cost of the 3W Luxeon LEDs. Or are you
saying that each LED is $14 to replace?

Frank Krygowski said:

Are there any LED headlights that use properly shaped beams - that is,
beams with flat upper cutoffs, that throw their light onto the road?

(Anybody seen a dead horse to beat around here?)

Frank, I was tempted to change the subject to something like "Another
preconceived notion dies hard," but figured that wouldn't be very
descriptive.

My daughter came home this weekend, and I noticed she had a 4-LED
light on the front of her bike. Tonight, I decided to test it against
my "bigger" lights. I have to say I'm impressed.

I have a CygoLite with 6W and 10W bulbs. Her Planet Bike Super Spot
LED light pretty well matches the illumination and the well-lit
footprint of the 6W halogen. (The 10W lamp is brighter and aimed
higher.) This LED light has a rectangular beam, with generous bleed
on the bottom, a sharp upper cut-off, and not much at all on the top.
There's a bit of light coming off past the upper cut-off, so it should
be visible by oncoming (or on-waiting side) traffic, but most of the
light hits the road.

To be honest, I was planning to show her how bad her light was, and
emphasize she should hurry home with this light warning oncoming
traffic before it was too dark. I guess I'm slow to accept new
technology, and I impressed my wife with the big lights because I was
SO visible. But I think I could commute with this thing. I won't,
since I've already got the bigger light, but I could.

The only thing that really worries me is the battery life. Planet
Bike's propaganda says 3 AAAs are good for 70 hours. I suspect that
means it's good for 35 hours, tops, before it's too dim.

Pat

Email address works as is.

Patrick Lamb said:

I have a CygoLite with 6W and 10W bulbs. Her Planet Bike Super Spot
LED light pretty well matches the illumination and the well-lit
footprint of the 6W halogen. (The 10W lamp is brighter and aimed
higher.)

Good information, thanks. I haven't yet bought an LED-based headlight,
and was planning to do so (assuming Santa doesn't deliver one). I'll
check that one out.

--
--------------------+
Frank Krygowski [To reply, remove rodent and vegetable dot com,
replace with cc.ysu dot edu]

Quoted message said:

Are you sure about three 3 watt LEDs? That's 9 watts, and 4 AA batteries
would have a hard time powering such a light. $14 wouldbe very cheap for
such a light, considering the cost of the 3W Luxeon LEDs. Or are you
saying that each LED is $14 to replace?

14 each if I bought them. with three you need a rechargables system. plus you
need the power regulated and controlled for LEDS.
but give it time and they will go way down in price.
I don't know what all is involved with the circuity to get them to run well
but you can't just run a battery lead to them. they have to be regulated.

--
Knight-Toolworks & Custom Planes
Custom made wooden planes at reasonable prices
See http://www.knight-toolworks.com For prices and ordering instructions.

Patrick Lamb wrote:

<snip>

Quoted message said:

My daughter came home this weekend, and I noticed she had a 4-LED
light on the front of her bike. Tonight, I decided to test it against
my "bigger" lights. I have to say I'm impressed.

I have a CygoLite with 6W and 10W bulbs. Her Planet Bike Super Spot
LED light pretty well matches the illumination and the well-lit
footprint of the 6W halogen.

Your daughter’s light is not the Planet Bike Super Spot.

The Super Spot is a single 1W LED light. It's received adequate reviews,
but the beam projects only 25 feet. It is rated for 30 hours run time
with 4 AA batteries.

The Sport Spot uses 4 lower power LEDs (I’m pretty sure that they are
so-called “high intensity LEDs, which provide about 3.5 lumens per LED
(14 lumens total), versus 30 lumens for the 1W Luxeon), and is rated for
100 hours flashing. I think that this must be the light your daughter
has, as it’s the only headlight from Planet Bike with four LEDs.
It sells for $18 on closeout from REI outlet.

The review of the Sport Spot on the Planet Bike web site is not
encouraging: "It gives enough light that you could hike by it, or even
ride slowly. The Sport Spot isn’t intended to replace high dollar
halogen lights, but it works very well as a backup. It also works well
as a commuter light, when you are more concerned about being seen that
seeing, i.e. dusk or on well lit streets."

If it matches your CygoLite with the 6W bulb then this says that your
Cygolite has some serious problems.

She’d be better off with a Planet Bike Super Spot, or some other 1 watt
Luxeon light, or better yet, a Streamlight Twin-Task 3C flashlight on a
TwoFish Cycloblock. The Streamlight Twin-Task 3C is 70 Lumens (xenon,
with a run time of 3 hours) and also can use three or six LEDs in “being
seen” mode (see http://streamlight.com/tasklight_specifications.htm).
It’s $27.50 on Amazon.com
(http://www.amazon.com/exec/obidos/tg/detail/-/B00008VE4I/104-4937282-0690333).

Even the Streamlight Twin-Task 3AA is 46 lumens, and costs about $20 (I
bought one on sale at Fry’s for $20, it’s $25 on Amazon).

Steve
http://bicyclelighting.com

In article <[email hidden]>,
[email hidden] says...

Quoted message said:

begin quoting Nubo <[email hidden]>:

Quoted message said:

difference in the response of motorists. It's one thing to say that a
light is visible from 1 or 2 miles (hell, a candle is visible from a
mile) and quite another for a motorist to actually recognize it and be
concerned about right of way.

[...]

Quoted message said:

rudely bright. But bright enough. And, this is in measure due to the
significant sidespill light which others refer to as wasteful.

How? I mean, unless you're just about to round a corner, someone 1 or 2
miles away on the same road with a line of sight to you is in the central
beam of a headlight.

... for that matter, how do you determine that a motorist recognises you
two miles away?

I know a motorist recognizes me at a distance when he dims his high
beams instead of blinding me until he's close enough to make out pedal
reflectors. And I find that a narrowly-focussed beam works best at long
range -- the 3W Lumotec that came with my Schmidt is at least as
effective as my 8.5W NiteRider narrow spot.

--
[email hidden] is Joshua Putnam
<http://www.phred.org/~josh/>
Braze your own bicycle frames. See
<http://www.phred.org/~josh/build/build.html>

begin quoting Patrick Lamb <[email hidden]>:

Quoted message said:

The only thing that really worries me is the battery life. Planet
Bike's propaganda says 3 AAAs are good for 70 hours. I suspect that
means it's good for 35 hours, tops, before it's too dim.

This is one of the nice things about the Smart LED/halogen switchable
headlight I have. When the batteries need replacement, the halogen doesn't
work, so you aren't stuck looking at it going "hmmm, how bright is
bright?"
--
David Damerell <[email hidden]> flcl?