[BC] Cree XR-E meets Schmidt E6
Guenther Schwarz
The availability of the Cree XR-E with it's high lm/W ratings did motivate me
to look again in the FL/2 design to mate LEDs with the well known optics of
the Bisy/Schmidt reflector.
The basic idea is simple: cut the reflector vertically and mount a LED in each
piece pointing to the side. The tricky bit is to find the correct focus
point. The reflector is designed for a halogen filament which is smaller and
has a different shape as compared to the diode's light emitting surface. This
somehow compromises the arrangement.
A quite elaborate setup:
<http://www.enhydralutris.de/Fahrrad/LEDWerfer0402/p2120003_s.jpg>
(all pictures and data by Olaf Schultz <webm...@enhydralutris.de>)
As I don't have the tools and skills for this my approach was a more basic
one:
- take two sheets of 1,5 mm alloy. These should be larger than the
reflector to serve as a heat sink and also for mounting the light
to the bike.
- grind away the contacts on the backside of the LED
- solder wires to the top contacts. Make sure not to contact the
metal ring holding the lens.
- glue one LED on each sheet
- cut an E6 reflector in two pieces. The reflector is available as a
spare part, so there is no need to buy the entire light.
- glue three nuts on each piece. These serve for adjusting the height
relative to the LED. Thanks to R. M. for this idea.
- connect the LEDs to a current source
- point the arrangement to a wall and try to find the focus point
- the line of symmetry is easy. For and aft are much the same as with
the halogen bulb. The hard part is to find the correct
height. For me it was close to 3 mm above the metal base. I used
a halogen light for comparison:
<http://www.xn--gnther-schwarz-gsb.de/Pictures/cree-tmp.jpg>
(top is left; the halogen light has low voltage; the picture is not
for quantitative comparision but just for comparing beam shapes)
- use lots of epoxy and glass fibre to fill the gap between the reflectors
and the base
- glue the two parts back to back with some spacers in between in order
to allow for air flow
The resulting light is quite nice. While the E6 reaches something like 20 lx
in the central spot at 10 m the LED version is well above 30 lx. It still
shows the asymmetrical pattern of a low beam light:
Schmidt/Bisy with 3 W halogen bulb:
- 22 lx max.
- total appr. 20 lm
- 2 lx in 'forbidden zone' 3.4° above horizontal axis
- 6.0 V x 0.47 A = 2.8 W
<http://www.enhydralutris.de/Fahrrad/Goniometer/Bilder/gox2004061223_Wand.png>
Modified light with two Cree XR-E:
- 48 lx max.
- total appr. 90 lm
- 7 lx in 'forbidden zone' 3.4° above horizontal axis
- 6.6 V x 0.4 A = 2.6 W (appr. 3.7 W with rectifier at 0.5 A)
<http://www.enhydralutris.de/Fahrrad/Goniometer/Bilder/gox20070214_Wand.png>
So lab tests and also first test rides show that this light is much brighter
than any other dynamo powered halogen one. It also outperforms the DLumotec
and Inoled which both employ a single 1 W LED.
For climbing the lm/W ratio of LEDs increases with lower currents
while halogen bulbs get dim very rapidly. In my setup (Schmidt SON XS laced
in a 622 rim) the light reaches almost full power above 10 km/h. At lower
speeds it starts to flicker. This can somehow be damped out by adding a
rather big capacitor (e.g. 4,7 mF) in parallel to the LED. With a small wheel
or with a standard hub dynamo in a 622 wheel this limit will be shifted to
lower speeds.
As it is the light does not meet German standards for bike lights as the
spread caused by the big emitter results in too much light above the
horizon. The legal limit is 2 lx as compared to about 6 lx for two car
headlights following European standards. Light is also spread more to the
sides as compared to the E6 which is welcome.
So there is still much room for improvements with optics designed specifically
for high power LEDs. But at the moment this is about as good as can be done
with parts available.
Bike light companies might also hestitate in spending money for reflectors
that can be sold for a very short time only as new generations of high power
LEDs almost always are not compatible in mechanical dimensions and optical
properties with prior versions or devices from other sources. It seems to be
about time for a standardization process. Maybe the emergence of LED
headlights for cars will trigger this.
Günther
--
To subscribe to this list send a BLANK email to bikecurren...@topica.com
To unsubscribe from this list send a BLANK email to bikecurrent...@topica.com
--^----------------------------------------------------------------
This email was sent to: bikecurrent-g...@googlegroups.com
EASY UNSUBSCRIBE click here: http://topica.com/u/?aVxiBc.aBSdFg.YmlrZWN1
Or send an email to: bikecurrent...@topica.com
For Topica's complete suite of email marketing solutions visit:
http://www.topica.com/?p=TEXFOOTER
--^----------------------------------------------------------------
John Clary
> As it is the light does not meet German standards for bike lights as the
> spread caused by the big emitter results in too much light above the
> horizon. The legal limit is 2 lx as compared to about 6 lx for two car
> headlights following European standards. Light is also spread more to the
> sides as compared to the E6 which is welcome.
You might want to try a reflective "eyebrow" at the top of the light
to reflect the stray light back down to the road. It's a bit unweildy
and slightly ugly, but I have had very good results in getting a sharp
horizontal cutoff by polishing up a chunk of Al angle and gluing it
over the top of a LED cluster. It takes a bit of math to get the right
length.
--
Opus
My gas is up to $.99 a burrito, $5.99 for premium, and I'm only
getting 20 miles to the regular burrito.
Guenther Schwarz
John Clary wrote:
> On 2/22/07, Guenther Schwarz wrote:
> > As it is the light does not meet German standards for bike lights as the
> > spread caused by the big emitter results in too much light above the
> > horizon. The legal limit is 2 lx as compared to about 6 lx for two car
> > headlights following European standards. Light is also spread more to the
> > sides as compared to the E6 which is welcome.
>
> You might want to try a reflective "eyebrow" at the top of the light
> to reflect the stray light back down to the road. It's a bit unweildy
> and slightly ugly, but I have had very good results in getting a sharp
> horizontal cutoff by polishing up a chunk of Al angle and gluing it
> over the top of a LED cluster. It takes a bit of math to get the right
> length.
4cm/tan(4) = 57cm, quite long :-)
This is a light for road use which means it is aimed almost horizontally. Up
to now I did not have complaints. But I will have to be aware of possible
difficulties on trails and narrow streets. On normal two lane roads it does
not seem to be a problem.
John Clary
>
> John Clary wrote:
> > You might want to try a reflective "eyebrow" at the top of the light
> > to reflect the stray light back down to the road. It's a bit unweildy
> > and slightly ugly, but I have had very good results in getting a sharp
> > horizontal cutoff by polishing up a chunk of Al angle and gluing it
> > over the top of a LED cluster. It takes a bit of math to get the right
> > length.
>
> 4cm/tan(4) = 57cm, quite long :-)
> This is a light for road use which means it is aimed almost horizontally. Up
> to now I did not have complaints. But I will have to be aware of possible
> difficulties on trails and narrow streets. On normal two lane roads it does
> not seem to be a problem.
>
> Günther
>
working with I could get right next to the 5mm LED so I only needed
about 2.5 cm of polished aluminum. to get most of my light on the
road.
Ian Boehm
> - 6.6 V x 0.4 A = 2.6 W (appr. 3.7 W with rectifier at 0.5 A)
G'day Günther
This looks like a tasty solution - thanks for sharing it with us. Your design is in danger of being my new lighting system for PBP.
Can I assume from the above that you are driving this with nothing more elaborate than a bridge rectifier?
Cheers
__o
_`\<,
...(*)/(*)
Ian
-------------------------------------------------------------
Ian Boehm e-mail: I.B...@bom.gov.au
Guenther Schwarz
John Clary wrote:
> On 2/26/07, Guenther Schwarz wrote:
> > John Clary wrote:
> > > to reflect the stray light back down to the road. It's a bit unweildy
> > > and slightly ugly, but I have had very good results in getting a sharp
> > > horizontal cutoff by polishing up a chunk of Al angle and gluing it
> > > over the top of a LED cluster. It takes a bit of math to get the right
> > > length.
> >
> > 4cm/tan(4) = 57cm, quite long :-)
> > This is a light for road use which means it is aimed almost horizontally.
> > Up to now I did not have complaints. But I will have to be aware of
> > possible difficulties on trails and narrow streets. On normal two lane
> > roads it does not seem to be a problem.
> How close to the center of the LED can you get?
Not so close. The reflector is about 8cm in diameter. I could try to put
something *inside* the light just next to the LED. Similar to the low beam in
a car headlight with dual beam. But as a consequence only a quarter of the
reflector instead of one half would be effective.
So someone skilled in tiny mechanics might set up a small mirror inside the
light that pops up for switching. But that will be a level of sophistication
I'm not up to.
> The light I was
> working with I could get right next to the 5mm LED so I only needed
> about 2.5 cm of polished aluminum. to get most of my light on the
> road.
With a small light this will be possible.
Guenther Schwarz
Ian Boehm wrote:
> > - 6.6 V x 0.4 A = 2.6 W (appr. 3.7 W with rectifier at 0.5 A)
>
> G'day Günther
>
> This looks like a tasty solution - thanks for sharing it with us. Your
> design is in danger of being my new lighting system for PBP.
It is nearly ideal for that purpose. Actually it is on my bike set up for
Brevets. But then I'm not feeling up this year for the very long distances.
Check if they allow for LED lights this year. As far as I know only halogen
lights did pass the technical checks at PBP last time.
In total the light just weights about 500g: 300g for my wheel with Schmidt
XS100 as compared to my normal one with an Ultegra hub, about 100g for the
light, and something for wires, a small piece of ciruit board, and the
bracket that mounts the light at the dropout. Not too bad for a light with
unlimited runtime and a very pleasant beam.
> Can I assume from the above that you are driving this with nothing more
> elaborate than a bridge rectifier?
It is a home made Schottky bridge rectifier with about 0.8V drop. Right now
I'm experimenting with capacities in parallel to the LEDs. It needs more than
1mF to smooth out flickering at low speeds. But too big a capacity like a low
ESR Goldcap results in the LEDs not lighting up initially or when going slow.
4.7 mF seems to be reasonable and is not too bulky. For a standlight a
Goldcap with 1F/5.5V together with a small resistor is useful. But I still
have to determine the optimal value for this resistor. It will be some tenth
of an Ohm up to two Ohm or such.
This takes some time as lab testing is nearly useless. I have to go out and do
test rides on my favorite climbs in order to find the optimum.
Günther
Ian Boehm
> Check if they allow for LED lights this year. As far as I
> know only halogen lights did pass the technical checks at PBP
> last time.
Thanks for all that Günther
The extract from the regulations in the PBP2007 brochure on my desk says:
"Cycles must have a lighting system both at the front and the rear which should be sufficiently bright to be easily seen for at least 100m at night for the front lamp(s) and 150m for the rear lamp(s)". The regulations go on to say no flashing rear lamps, all securely fixed etc. There is no mention of LED lights (or any others) being verboten. My view is that poor lighting adds to ones fatigue load and I will take the best lights that I can afford, not the worst lights I can get away with.
I'm off to buy an E6 reflector at lunch time.
Cheers
__o
_`\<,
...(*)/(*)
Ian
-------------------------------------------------------------
Ian Boehm e-mail: I.B...@bom.gov.au
Francis Cooke
GS> Check if they [PBP] allow for LED lights this year.
They do.
Should be "visible from 150 metres" which raises interesting images of
the bike scrutiny procedure ...
--
Best regards,
Francis mailto:fra...@aukadia.net
Ian Boehm
>
> They do.
> Should be "visible from 150 metres" which raises interesting
> images of the bike scrutiny procedure ...
Les binoculars s'il vous plait Pierre...
The cynic in me says that it will be as cursory and random as last time
with some of the most derisory lighting systems (e.g. D cell torches
taped to forks) getting a nod. At least those of us with decent systems
will not be short of company in the long marches of the night.
Cheers
__o
_`\<,
...(*)/(*)
Ian
-------------------------------------------------------------
Ian Boehm e-mail: I.B...@bom.gov.au
John Laidlaw
Of course, we all realise that the rule has more to do with rendering
ourselves visible, rather than being able to see the road, and the hazards
thereon. if we cannot see what's there, then it's our own responsibility to
improve the lighting, or slow down. I recall an intensive little session in
my motorist's training bout the hazards of 'over-driving' one's lights. I
grant, this was long before the days of halogen or other high-intensity
bulbs, but I noted, immediately, it's relevance to what I also experienced
on my bike. Pity was, in those days, there was little available (at least to
the average kid, in Canada) that was any better than one of those D-cell
rigs, and the cells died, usually, about ten minutes after one left on the
way. It was a long, dangerous ride out and back again. The standard
(garbage) dynamo sets were not that much better, as they tended to drown out
in the slightest damps. Besides, a stand-light was unknown, here.
How things have improved!
John Laidlaw
> Internal Virus Database is out-of-date.
> Checked by AVG Free Edition.
> Version: 7.5.432 / Virus Database: 268.15.29/608 - Release Date:
> 29/12/2006 8:22 AM
Guenther Schwarz
Ian Boehm wrote:
> > Check if they allow for LED lights this year. As far as I
> > know only halogen lights did pass the technical checks at PBP
> > last time.
>
> Thanks for all that Günther
>
> The extract from the regulations in the PBP2007 brochure on my desk says:
>
> "Cycles must have a lighting system both at the front and the rear which
> should be sufficiently bright to be easily seen for at least 100m at night
> for the front lamp(s) and 150m for the rear lamp(s)". The regulations go on
> to say no flashing rear lamps, all securely fixed etc. There is no mention
> of LED lights (or any others) being verboten. My view is that poor lighting
> adds to ones fatigue load and I will take the best lights that I can
> afford, not the worst lights I can get away with.
Last time the Cateye HL-EL300 was state of the art for LED lights. So it seems
to be a wise decision not to approve it. Things have changed obviously. But
it might still be wise to ask them or simply to take a halogen light to the
event just in case.
> I'm off to buy an E6 reflector at lunch time.
I might find a little time for some photos and such during the next weeks. I
will also built a second version. For this light the workshop at my institute
will cut the reflector for me. With their more precise work I might be able
to give the proper height for mounting the reflector which will make further
copies much easier. But even with the proper figures at hand I want to
recommend strongly to check the beam shape while assembling. A second
reflector or a complete Bisy/Schmidt with a halogen bulb is ideal for this.
Günther
Joe Gross
On Fri, Feb 23, 2007 at 01:20:58AM +0100, Guenther Schwarz wrote:
>
> The basic idea is simple: cut the reflector vertically and mount a LED in each
> piece pointing to the side. The tricky bit is to find the correct focus
> point. The reflector is designed for a halogen filament which is smaller and
> has a different shape as compared to the diode's light emitting surface. This
> somehow compromises the arrangement.
This is truly an awesome combination.
I took a look at my own E6 pair and have to wonder if it would work
just as well to mill a mount out of some Al in the shape of a bulb so
that I could simply replace the bulb with the LED mount. This way I
wouldn't have to cut up a reflector and it would fit perfectly back on
the light mount.
I'd have to build a rectifier into the bulb-shaped mount but there
should be room for that.
Could heat be an issue? If so, I'm curious as to why the 3W halogen
bulb doesn't have a problem.
Do you think you could post more pictures of your device?
Joe
Steve Kurt
Joe Gross wrote:
>
> I took a look at my own E6 pair and have to wonder if it would work
> just as well to mill a mount out of some Al in the shape of a bulb so
> that I could simply replace the bulb with the LED mount. This way I
> wouldn't have to cut up a reflector and it would fit perfectly back on
> the light mount.
>
> I'd have to build a rectifier into the bulb-shaped mount but there
> should be room for that.
>
> Could heat be an issue? If so, I'm curious as to why the 3W halogen
> bulb doesn't have a problem.
hi Joe,
regarding the first issue: mounting a LED in a headlight.. it can be
done, but it won't produce the same beam as an incandescent bulb. The
bulb radiates light over nearly a complete sphere, including to the
rear of the bulb. The led only radiates forward. This is why people
are cutting headlights in half (down the middle, usually), so that the
led can radiate into a half of the usual optics/reflector assembly.
There are pictures somewhere of how this works... (don't recall where,
tho).
the second issue: heat.
Bulbs work *because* they are hot. LEDs will die if they get hot enough
to glow. They really do best if they don't get warm at all.
Unfortunately, a 3W led will produce nearly as much heat as a 3W bulb.
Gotta get that heat away from the led, and out to the surrounding air.
You can still mount a Luxeon or Cree emitter on an aluminum base, and
rig up something, perhaps. I've got a 5W Luxeon attached to a threaded
aluminum rod, which is mounted within the reflector of an old Union
dynamo headlight. The beam isn't great, but it works well enough for
me. The aluminum rod has an aluminum plate attached just behind the
reflector, and this gets rid of the heat.
regards,
Steve Kurt
Guenther Schwarz
Joe Gross wrote:
> On Fri, Feb 23, 2007 at 01:20:58AM +0100, Guenther Schwarz wrote:
> > each piece pointing to the side. The tricky bit is to find the correct
> > focus point. The reflector is designed for a halogen filament which is
> > smaller and has a different shape as compared to the diode's light
> > emitting surface. This somehow compromises the arrangement.
>
>
> just as well to mill a mount out of some Al in the shape of a bulb so
> that I could simply replace the bulb with the LED mount. This way I
> wouldn't have to cut up a reflector and it would fit perfectly back on
> the light mount.
Like this (again stealing from Olaf's web pages):
<http://www.enhydralutris.de/Fahrrad/LEDWerfer0402/p1220858_s.jpg>
<http://www.enhydralutris.de/Fahrrad/Beleuchtung/node249.html>
Fig. shows three typical dynamo lights: Nova, Schmidt (FL), and Lumotec.
The focus point is at the position of the bulb filament. Light is collected by
the mirrored reflector. Most LEDs emit preferencally in the forward direction
missing the reflector. This is why some bulb replacements like the Terralux
point the LED to the side directly towards the mirror:
<http://www.ultraleds.co.uk/product_info.php?cPath=35&products_id=791>
One could copy this design with the Cree XR-E. But with about 1.6 mm for the
LED between the mounting and the emitting surface and some 0,5 mm metal sheet
in between the two LEDs this results in a 'filament' of 3,7 mm lenght which
is simply out of focus. This and improved cooling is why it has become
popular to cut the lights into two pieces. More freedom to find the optimal
position. But your approach might result in an useful light also.
Another reason for me not to try a replacement bulb is that I do not know how
to work with a lathe. Producing a bulb shaped object is more complicated than
glueing a LED on a flat plate and then mounting half a reflector on top of
it.
The Everled uses a side emitting variant of the 1 W Luxeon which gives another
possibility:
<http://www.leddynamics.com/everled.html>
bumm employs this also for their DToplight variants. But this LED is only
available as a 1W version. My arrangement is also more compact than having
two complete lights.
Compare to a specific LED optic like the MOBDAR:
<http://www.xn--gnther-schwarz-gsb.de/Pictures/mobdar01.jpg>
(This picture was not taken for documentation. I was just playing with some
sticks made from fluorescent plastic.)
Here the lens sits directly on top of the LED. It collects most of the light
and guides it to the front.
> I'd have to build a rectifier into the bulb-shaped mount but there
> should be room for that.
>
> Could heat be an issue? If so, I'm curious as to why the 3W halogen
> bulb doesn't have a problem.
Steve Kurt already gave an answer to this. I only want to add that a bulb
produces it's heat as infrared radiation which mostly radiates away. A LED
device will heat up because of leakage currents and such. The difference
becomes obvious in winter when a halogen light at least partly melts away
snow from the front lens. With an LED light the front stays cold and has to
be cleaned by hand from time to time. Does anyone have a dynamo driven screen
wiper? Might be nice for my glasses also :-)
> Do you think you could post more pictures of your device?
Not right now, but stay tuned.