E3 triple
pamela blalock
Last Friday, we got an email from Elton (at Harris Cyclery – http://www.sheldonbrown.com/harris/index.html ) saying that a long-awaited box had just arrived from supernova. We got to the shop before he had even opened the box! It’s handy to live just a couple miles away. Actually it’s quite dangerous for our bank account, but that’s another story.
With great ceremony Elton opened the box and pulled out an E3-triple (http://www.supernova-lights.com/shop/index.php?cPath=56_65 ) for us to take home (well once we put yet another big dent in the bank account).
We stopped on the way home for a cappuccino and espresso and a couple of pounds of coffee beans from Taste Coffee House (http://www.tastecoffeehouse.com/).
Once home and fully caffeinated, I pulled out my soldering iron, attached the spade connectors, and taillight and mounted the light on my winter commuter. Then I had to wait hours for darkness. I had a few espresso drinks while I waited.
It’s bitter cold here and we have a bit of snow on the ground. But we suited up after dark and took the new light out for a spin. Given all the dire warnings about not using the triple on the roads, I expected to melt snow in my path. The light is brilliant, but I did not melt pavement or even snow. I did not permanently blind John as I rode toward him, nor did any oncoming motorist swerve and run over an innocent pedestrian after being blinded by my light.
We did a few ride-bys, switching rider and pedestrian roles, so we could each see the light in action from both perspectives. Then I grabbed another bike with a standard E3, and we rode along side by side to compare.
The standlight is the brightest standlight I have ever seen. At slow speeds, only one of the three LEDs is on. As you get up to speed, all three come on, and they just get brighter as you go faster. Slowing down for a stop, I notice the switch down to 1 led, and then when I stop, it switched to standlight mode which is actually brighter than slow speed. I did also notice when just walking with the bike up the driveway that the triple does not seem to come on at this slow walking pace, like the standard e3 does. But slow speed *riding* does produce great light. My impression is the slow speed light is about the same as the slow speed standard e3, but at higher speeds (when all three leds are on) it is much brighter.
The bike I initially installed the triple on is a fixed gear with studded tires, racks, fenders, etc set up primarily for commuting. For this reason, I have a somewhat low gear and I don’t go terribly fast on it. But thanks to the studs, I can test the light in our current conditions. I do ride fast enough to have all three LEDs on 95% of the time. I just don’t get over 25mph.
The next two days of riding, it was like riding in broad daylight. Oh right, that’s because I was riding in broad daylight, and somehow didn’t find the motivation to go out for a long ride in the dark just to try the light out further.
But then Monday arrived, and it was back to work. I forced myself NOT to leave work early, so I got a fully dark commute test. This involves some dark residential roads, a short section of bike path, some more quiet residential roads, and then a bit or urban ugliness. I also have a couple of big hills, so I get to test the slow speed light while climbing with studs, and the slightly higher speed light while descending on fixed with studs. Due to the melted snow, I get to test the light on wet pavement, one of the most challenging conditions, and then with snow reflected back off the sides of the road. Part of the bike path has been cleared, put there are still icy patches, and I only stayed on it for a short while to avoid a busy town center. (When it isn’t icy, I use it a bit longer, but then have to ride across a field, which is covered in snow and ice, so I passed on that).
Leaving work, I first have to climb up a very steep incline from our parking lot. I mentioned that when pushing, it doesn’t seem to light up, but the light did come on and provide reasonable light on at my crawling pace out of our parking lot, something a bulb light would not do. I then started down the road, and the light just got brighter.
My commute is about 15 miles and takes me a bit over an hour. In this time, I encountered many cars, a few pedestrians, many potholes, and a few patches of ice. No one high beamed me. No pedestrians yelled. I did SEE all the potholes and patches of ice. I did the commute again last night, so at this stage have the brief ride on Friday and the two hour long dark rides.
So this in my impression after these short tests. At speed the triple is definitely brighter than the standard. But at both low and high speed, the standard is actually plently bright for me. The standlight on the triple is the most impressive I have ever seen. It is not merely a be seen light. It illuminates the road quite well. The standlight mode on the standard is also amazingly bright, so no complaints there.
It is said to create more drag, but I can’t tell any difference. Heck I can’t tell if the light is off or on in daylight. I do notice the difference between a studded tire and a non-studded tire, so I am not totally oblivious to drag!
I am quite happy with my standard light, so we will likely put the triple on the tandem where it will be more appreciated on those high speed dark descents.
For oncoming visibility, it is definitely visible, as is the standard. Folks will not say they didn’t see you. I have my light aimed very slightly to the right, not so much that it hits pedestrians on the sidewalk, but so it lights up more of my lane than an oncoming lane. On most roads, it is not going to be lighting up the oncoming lane. So while I am visible to oncoming drivers and cyclists, it does not blind them. This applies to my setup for both the standard and the triple.
So in summary, I still think the E3 standard is brilliant and have this light on my commuter and brevet bikes, and will continue to recommend it. And just because we have it, we will put the triple on the tandem where we can take advantage of the additional light with the higher speeds we reach on descents.
If you are a person who can’t get enough light, and wanted dual e3’s, then this is the simpler more elegant solution than two separate lights. The triple is about 50% more expensive than a standard e3, so it’s also cheaper than setting up duals, but expensive than a single standard. So if you must have more go for it.
We have not tried the asymmetrical light. As I’ve stated before, I like illuminating street signs and don’t have plans to ride at night in Germany, so I don’t need the asymmetrical light.
Pamela Blalock
Jake Kassen
> We have not tried the asymmetrical light. As I’ve stated before, I like
> illuminating street signs and don’t have plans to ride at night in
> Germany, so I don’t need the asymmetrical light.
>
As it turns out, I just got back from Germany, so I now have some
practical experience riding a bike for commuting in and out of the
mid-sized city Karlsruhe and suburbs. Here are my impressions in brief:
Where I was (and I believe this is common) there are indeed hundreds of
kilometers of bike paths. That's the good news. The bad news is that
these paths twist and turn and are frequently just extensions of the
sidewalks. There are a million stops, obstacles, turns, etc. Frequently
you need to cross tram tracks. Sometimes the paths are lit, other times
you're riding on an isolated dirt road. Sometimes the path just aburptly
stops and you need to lift you bike over a barrier and continue on the
street. The paths were designed to accommodate someone riding a bike at
10kph -- about as fast as jogging pace. I saw few people riding faster
then this. I prefer to ride ~25kph (15mph) when commuting. Few people
commute more then a few miles and many ride to the store, load up the
bike, and then walk it home.
For the record, I prefer riding on the Boston streets.
I was using a Lumotec light with a Shimano Dynamo. Nothing fancy. What I
found was that the asymmetrical beam lit the road well enough to see,
but it didn't light all the obstacles well enough for me to be prepared
or comfortable at moderate speeds. I sorly missed by homemade
symmetrical beam light that I use for commuting and brevets in the US.
For all that is said about German law, I was surprised to see (or not
see) half the bikes at night with absolutely no lights whatsoever. Rim
and hub generators were common but frequently not used. About 1/3 of the
bikes completely lacked lights. Many people biked slow enough to
illuminate their lights but not fast enough to illuminate the road.
People seem to take German bicycle laws about as seriously as we take
our own bicycle laws in the US.
One more thing I'll add is that *everyone* has a bike. Many don't lock
them to anything, even in the city. I never saw a SON hub and I was
looking.
Anyway, thanks Pamela for the good light review. I'm more convinced then
ever that I too prefer a symmetrical light and think they would be
preferable when riding in Germany.
Jake
pamela blalock
I’ve said before and will say again that it really is a matter of personal preference.
Not too long ago my preference was for the asymmetrical beams, because light output was so limited that I wanted all of it on the road. As LEDs have improved dramatically to the point of being better than bulbs now, I have switched gears a bit, if you will. There is now enough light on the road that I can again have light above the horizon on street signs and such. And so now I can see the potholes AND the street signs thanks to my symmetrical beam. I also prefer the tapering of light at the sides, rather than the binary on/off of some of the asymmetrical lights.
If you are primarily riding busy narrow bike paths, it is only courteous to pick/position a light that is not blinding fellow users approaching you. Of course, users of busy bike paths could also show the courtesy of not using flashing lights (front and rear), but that’s another rant. The difference between bike path and road use is really width and where the light shines side to side. On a bike path the oncoming person is much closer (side to side) and much more likely to be in your beam. Even on narrow twisty New England roads, it is unlikely that my light is shining OVER THERE.
I am always conscious and aware of fellow bike path users when I am using a bike path. I do use one for a few km almost every day as part of my commute, but the path is still a small part of my commute (and an even smaller part of all my night time riding) and I often don’t encounter many other users at night when commuting. If using my very bright symmetrical lights, I will often cup my hand around the light to really be sure not to offend. But I will also mention that the only time I’ve ever had a bike path user say anything was with an edelux! I had a loaner for a while, and found that it cut so sharply at the horizon that I didn’t see people walking until I was very close, so I had it aimed a bit higher. I also had trouble spotting gates and bollards until really close – again, the path was well lit, but things above it weren’t like gates. The edulux beam is wider close in than the e3 and really needs to be aimed straight ahead. Anyway, it was when using this in a way that I didn’t run into the gates on the path that a runner yelled at me about my edelux – ironic, eh!
For those who didn’t see my original report on the edelux… I did like it. It is much much better than the first few versions of asymmetrical LED lights. But between the two, *I* prefer the e3.
It’s funny I’ve heard conflicting reports about enforcement of lighting rules in Germany. It seems in some places, the crime rate must be really low and there is more picky bike light enforcement. In practice, one can probably ride with whatever he/she likes, you are just restricted as to what you can BUY In Germany. It is cool that bikes are treated as tools and are used by everyone.
Pamela
prestonjb
the hub that the regular E3?
pamela blalock
prestonjb
Now that would be some serious light for riding downhill at night!!!
Dark Horse
emitters, or four, or more. I'm finishing a quad-die light now, and
the next will have 5 emitters. There are people running 6. It's all a
question of how much you want to push.
If you're going to run dual E3's, then you could run a triple and an
asymmetrical, electronics permitting.
Dark Horse
prestonjb
in parallel with the headlight? So basically if I can huff down on
the pedals to keep the current up then I can drive up to what... the
limits of the wires in the windings and the wires going to the power
taps?
What I was assuming in my question about the tail light was that the
triple was still within the "power band" of what we consider "typical
and normal effort" to drive @ 6V & 6W
prestonjb
assume 6V then some of them in series would still only consume 3W and
so the power is to get the system to output 6V@3W... then a second
tree could be the tail light or a second headlight.
But again I'm also interested in trying to prevent being a Dark.. Um I
mean "work" horse or slave to the lighting system...
Currently dual E3 single LEDS are still OK so the Q is how much "pedal
push" are we adding with triple,quad, quint LEDs?
albert meerscheidt
Thanks,
Albert
albert.me...@gmail.com
cbcr...@cascade.org
webm...@seattlerando.org
Dark Horse
the AAA-Blinky sort.
What I do know is that the SON is effectively a constant-current
dynamo above a relatively low speed. It provides ~550Ma from 4-5mph,
IIRC.
The rated output is officially 3w@6v, which is about right for an
E6. The thing that varies with speed and load is the output voltage.
Someone on Candlepower measured the open-circuit voltage at 50mph and
got over 50v. At that same 550ma. The hub will generate 1V/mph or 1A/
2mph, if it has the appropriate load. If you wire 4 (for instance)
emitters in series, then the Forward Voltages of the emitters add up,
and the light will only reach full intensity when the hub is giving
~13v@550ma. Since P=VI, you're pushing 7W into that hub to maintain
that light. The actual effort is higher when you correct for hub and
circuit efficiencies, but that's the idea. Since a current-gen emitter
is throwing 80-100 lumens/watt at 550ma, that 7W is buying you
560-700lm of output, less optical efficiency.
Confused yet?
More things that I do know. My light throws ~400lm on high-beam and I
can only feel the increased effort when I'm really tired. Like 600k-
level tired. Maybe. That's 5.2 watts of output. The 4-die light I'm
building will need about 7W to throw ~600lm.
Supernova makes some very impressive output claims for their lights,
but even the E3(3) takes only 10.5V to hit full brightness. Add an
asymmetrical E3 and a taillight, and you're only asking the hub for
14V plus the taillight. And the taillight is nothing. That's well
within the capabilities of that hub. There is a guy in Australia
running 6 Cree XR-E's on his 29'er racebike. He does a couple of
things with capacitor choice to get over 1000lm out of his dynohub.
Look towards the end of this thread:
http://www.candlepowerforums.com/vb/showthread.php?t=89607
And scan this one:
http://www.candlepowerforums.com/vb/showthread.php?t=172636
If that's not waaaay TMI, look here:
http://www.pilom.com/BicycleElectronics/DynamoCircuits.htm
BTW, if you have an E3(3) and you feel like opening it up and taking a
pic or two of the circuit board, do not let me stop you. I am
extremely curious about their electronics package.
</TMI>
Dark Horse
Dark Horse
With 5 emitters, you could have multiple beam shapes in the same
light housing, accessible by microtoggle. Or, if you had enough game
to design it, an autoswitching circuit reading AC Freq.
Dark Horse
WillemJ
for lights to become more effcient: the same amount of energy input
produces more light. The new generation of led lights has been a
quantum leap in this respect, and we may expect more such gains in the
future. The second route is to find ways to allow you to work harder
to produce more light. This was the principle of the dual E6 setup,
and I think this is probably also the principle behind the Supernova
triple. The extra series capacitor in the secundary E6Z pushed this
principle even further, by tweeking more light out of the system, by
making you work even harder. So, in the end, it comes down to the
question do you want to work harder for more light, or do you think
you have enough light already, and would rather go faster. An example
of the latter choice is Jan Heine's decision to use a SON20. He has
less light at lower speeds (but with leds that is now not much less),
but less resistance. I have done this on my wife's touring bike with
an Edelux, and it works like a treat, I think. But, it is up to the
rider to choose where his or her priorities are.
Schmidt have some German discussion on their website about the various
possibilities, and I have translated this here. It may help decision
making. I have also included a link to a slightly earlier German
article in Fahrradzukunft by their tech man. The text is in German,
but the graphs are perfectly clear, even if they relate to the IQ Fly
rather than the Edelux. The logic is the same, however.
First, about the combination of an Edelux with an E6Z:
c) Kombination mit einem E6-Z:
Der Halogen-Zusatzscheinwerfer E6-Z funktioniert prinzipiell auch (in
Reihenschaltung) zusammen mit dem Edelux. Der Serienkondensator im E6-
Z erhöht ab etwa 18 km/h die Leistung des Edelux. Das führt auch dazu,
dass das Licht des E6-Z erst oberhalb 25 km/h überhaupt wahrnehmbar
ist, aber im Lichtfeld des Edelux fast verschwindet. Wenn man diese
Kombination betreibt sollte man den Lichtkegel des E6-Z als Fernlicht
nutzen und dazu etwas höher ausrichten als den des Edelux.
A fairly free but accurate translation would be:
In principle it works (connected in series). The capacitor in the E6Z
increases the performance of the Edelux from about 18 km. Therefore
(i.e. because the Edelux itself produces more light than it would do
otherwise), the light of the E6Z becomes only visible from about 25
km, and even then almost disappears in the light of the Edelux. If you
use this combination, you should use the beam of the E6Z as a distance
light, and direct it a bit higher than the Edelux.
Combination with a second Edelux:
a) Kombination mit einem weiteren Edelux:
Prinzipiell könnte man zwei Edelux parallel- oder in Reihe geschaltet
an einem Nabendynamo betreiben. In der Praxis bringt dies aber nur
wenig mehr Licht. Die Parallelschaltung ist bestenfalls für Menschen
interessant, denen der Lichtkegel des Edelux noch zu schmal ist und
die beiden Leuchten V-förmig mit 10° seitlichem Versatz montieren.
Reihenschaltung bedeutet zunächst weniger Licht bei langsamer Fahrt.
Erst ab 30 km/h erreicht man wahrnehmbar mehr Licht als mit nur einem
Edelux.
A brief free translation:
In principle you could use two edelux, in parallel or in series. In
practice this does not produce that much more light. Parallel
connection is best for those who want a wider beam and want to use two
edelux with a 10 degree offset. Series connection gives less light at
lower speeds. You only have visibly more light from 30 km and more.
So far for the quick and dirty translation. The difference between a
second edelux and a secundary E6Z is that the latter has the important
capacitor, to tweeze the last bit of performance out of the system
(but also make you work a bit harder - there is no free lunch). They
also say elsewhere that you cannot combine a IQ Fly and an Edelux,
since the latter works at a slightly higher voltage.
The general principles are beautifully illustrated in their article in
Fahhradzukunft: http://fahrradzukunft.de/fz-0704/0704-05.htm It is
in German, but the graphs will speak for themselves. Also note the
table with the extra energy input you will have to provide! My
summary:
parallel works better at lower speeds, series better at higher speeds.
For the latter, you really want the capacitor. My intuitive advise if
you want more light:
if you already have an E6Z, use it, and only switch it on during fast
descents.
Willem
Dark Horse
> since the latter works at a slightly higher voltage.
>
> The general principles are beautifully illustrated in their article in
> Fahhradzukunft:http://fahrradzukunft.de/fz-0704/0704-05.htm It is
> in German, but the graphs will speak for themselves. Also note the
> table with the extra energy input you will have to provide! My
> summary:
> parallel works better at lower speeds, series better at higher speeds.
> For the latter, you really want the capacitor. My intuitive advise if
> you want more light:
> if you already have an E6Z, use it, and only switch it on during fast
> descents.
> Willem
Parallel-wiring allows emitters to light at lower speeds, but at the
cost of less maximum output. Series-wiring is exactly the opposite.
The piper must be paid in both cases: if you want more light you work
harder, so saith the conservation laws.
There are two assumptions that I find questionable. First, that
you're going to use a manufactured headlamp. BUMM, INO, Schmidt, or
something similar. Second, that you are bound by the StVZO lighting
standards.
There really are people out there who are running 3,4,5 and 6
emitters from a single SON. The low-speed/high-speed/output discussion
has been thoroughly gone over in builders' forums by some very
competent people. There is, for intance, a simple circuit that
incorporates a Greinacher doubler and a bridge rectifier with a manual
mode switch. There is also an autoswitching variant of that same
circuit. The doubler gives very good light at low speed with multiple
emitters, and the rectifier works at higher speeds. Capacitor choice
places the output peak where you want it. That Australian builder I
refer to has measured over 1amp from a SON.
I can't speak for serial/parallel Edeluxes, since I'm not familiar
with their electronics. However, for emitters in series, output and Vf
are additive. Two XR-E's in series will throw twice as much light as a
single, as long as their Vf requirements are met. The relationship is
direct. Not so if they're wired in parallel, since current per emitter
varies inversely with the number of emitters. The first gets 550ma,
the second 550/2ma, the third 550/3ma and so on. Emitter efficiencies
increase as current decreases, but output decreases also. A parallel
arrangement must also take into account that Vf requirements are not
additive. Supernova certainly claims that the E3(3) throws almost 3x
the light of the single E3.
The 5-Emitter system I casually referred to would look something like
this:
5 Cree XP-E emitters, R2 bin, WH tint. Mounted to 10mm pcb's. Rated
114lm@ 350ma. Almost exactly 100lm/Watt
http://www.cree.com/products/pdf/XLampXP-E.pdf
2 Ledil Lisa XP optics, 19deg FWHM.
http://www.ledil.fi/datasheets/DataSheet_LISA-XP.pdf
2 Polymer Optics 6x25deg collimator lenses
1 Polymer Optics 6deg spot
http://www.polymer-optics.co.uk/Cree%20XP-E%20Optics%20Range.pdf
Driver would be circuit #10 from here:
http://www.pilom.com/BicycleElectronics/DynamoCircuits.htm
Simple enough to add switching to power the emitters 2s/2s/1,
giving multiple beam options.
Switches from L&M and/or Grayhill.
Components from Digi-Key.
This is not complicated, and can all fit into a 1 5/8" aluminum tube
95mm long.
Mountings from Cateye, connectors from Molex, although you could use
Deutsch, Fisher, or Lemo if you wish.
Maximum output would be in the 800lm range, less optical
efficiencies.
Rider effort would be roughly 14W, for not quite 9W of output.
The Edelux, INO, and IQ-series are using single emitters, whose
specifications are published. Unless they are getting very cute with
electronics, their raw output is less than 200lm. Probably much less.
A cree XR-E from the R2 bin is giving ~175lm at 550ma, for just over
100lm/watt. The emitters that BUMM, Schmidt, and INO are using are
likely to have efficacies in the range of 80lm/watt or so at that same
550ma. Their optical beam control is superb, but the raw output is not
going to be more than that. Supernova's sales suggest that many people
would like more light than that.
I would be very interested to see what those manufacturers could do
with the idea of multiple beam shapes from a single source. The
positional difference between an automotive lo-beam element and a hi-
beam element is less than a centimeter, in a reflector that is inches
across. I see absolutely no reason that something similar couldn't
be done in a bicycle light.
Peripherals:
There is a lot of source material that's only available in German. I
know a lot of people who would like to read it, but don't speak or
read German. I read the Babel translation of that article, and it made
my head hurt.
It would be pleasant if Schmidt would publish the electrical
characteristics of the SON. Impedance, inductance and the like. Driver
design would be far easier if people didn't have to measure and deduce
source characteristics.
If you're building your own, you can pick the tint of your emitters
to suit your own preferences. Oehler refers to the cool-white
spectrum, but you can get emitters all the way down to a CCT of <4k.
Dark Horse
WillemJ
I know a handful of European languages, including German. So that is
why I thought I might make a contribution. I am sure homebuilt stuff
can be ahead of the developement curve, and can be more precisely
designed for the most desired mix of properties. It seemed, however,
fitting to distinguish analytically between on the one hand efficiency
gains, where the same amount of muscle effort produces more light (or
where you can get the same amount of light with less effort), and on
the other hand systems that allow you to work harder and translate
that effort into more light. It is the difference between what an
economist calls shifts of the production function and shifts along it.
Personnally, I am more interested in the former. From what I
understand (but Supernova are not very clear) the E3 Triple is
primarily an example of the latter.
best wishes,
Willem
prestonjb
and the Edeluxe (though the Fly were ahead of them) are the former...
I was hoping that the E3(3) was not the latter and still not
convienced it is totally railed to that category...
The question is "how much" it is so we can understand the "worth" of
it over the wait for a more efficient design..
I wasn't even thinking about these levels of details with my dual E6
lights... I guess having started with them I had no reference to how
much leg work I was doing to push those lights...
Even so I really don't feel enough draw from them (or now my dual
E3's) to "feel" that I am overworking to generate the power (system
weight is more noticeable...)
So it now becomes a question... Is an E3(3) going to be felt on my
legs more than a pair of E3s? What about two E3(3)s?
These numbers are not published in a athletic-power-format that I'm
aware of so I can't like get on my Cyclops P300 trainer and dial up an
extra X watts to see if it is too much for my comfort.. Currently my
only choice is to buy/build the lights... Both of which either consume
lots of money or lots of time...
Not sure what I will do to answer this but I think what we need from a
cyclist perspective is to understand the "efforts required" for
various hub generator "component combinations" too...
hughgs
It seems that all you really need is to measure the current and voltage to
the lights. That's the power needed and you have to generate at least
that much extra voltage.
I thought I saw somewhere in this discussion that one of the generators
was putting out 550 mA. You just need to figure out the voltage from the
generator as a function of RPM (which conceptually shouldn't be difficult)
and you're most of the way there. Maybe multiply your numbers by 1.5 to
account for circuit losses and you've got a rough of idea of the power
needed by your lights.
261 Hudson Annex
(302) 530-9335
hug...@duke.edu
Dark Horse
Your production curve analogy is perfect. Wish I'd thought of it
myself.
I think that there are several sources of confusion here.
First, there are two quite different efficiencies here. Andreas
talks about the efficiency of transforming effort into electrical
power delivered to the light head. I expect that any gains in that
area will be incremental rather than dramatic. Dynamos are a well-
understood technology with known behaviors.
The second sort of efficiency is the transformation of electrical
power into light. That IS undergoing dramatic improvements right now.
Halogen bulbs throw ~20 lm/W, and they were the state of the game 4
years ago. Cree's new XP-E emitter throws slightly over 100 lm/W. LED
efficacies have increased from ~60 to over 100 lm/W just in the last 3
years.
The improvements that Andreas is talking about are incremental
increases in the electrical efficiency of the dynamo/electronics
system. The improvements that the LED manufacturers are giving us are
not incremental.
A pair of E6's throws somewhere between 120 and 150 lumens for their
6W draw. No matter how good the optics, that's the raw output. An IQ-
series light throws something under 175lm for whatever its draw is.
My Cree setup throws 400 lumens (high-beam) at a 5.2 Watt draw, and
200 lumens at half that. the optics aren't as good, but it lights up a
lot of road. Flipping back and forth between high and low, I do not
feel a difference in effort.
That, I think, is the sort of efficiency gain you're looking for.
Unfortunately, Schmidt, INO, and BUMM are locked into the European
lighting standards, and that limits what they can build.
I wonder how much light you could get out of an IQ-series light with
a different emitter, and what shape the beam would take. At some point
I'm going to wire an MC-E in place of a stock emitter and find out.
</TMI>
Dark Horse.
Dark Horse
On Jan 11, 7:21 pm, hughgs <hug...@duke.edu> wrote:
> It seems that all you really need is to measure the current and voltage to
> the lights. That's the power needed and you have to generate at least
> that much extra voltage.
>
> I thought I saw somewhere in this discussion that one of the generators
> was putting out 550 mA. You just need to figure out the voltage from the
> generator as a function of RPM (which conceptually shouldn't be difficult)
> and you're most of the way there. Maybe multiply your numbers by 1.5 to
> account for circuit losses and you've got a rough of idea of the power
> needed by your lights.
>
characteristics. The efficiency of the dynamo/electronics isn't going
to change much from one light to another. Andreas' numbers suggest
that the greater the load the greater the efficiency.
Assume a constant current of 500Ma and a Voltage Forward per emitter
of 3.4V. The voltage isn't exact, but it's within a couple of tenths
of any of the high-output emitters. That yields 1.7W per emitter.
Assume further that the electronics in the E3 and E(3) are 90%
efficient. It's probably more, but I'm being conservative. That gives
5.6W drawn by the light head at maximum output. Multiply as necessary,
and apply Andreas' numbers from the page Willem linked to.
I flip back and forth between a 2.6 and a 5.2W draw quite often during
a ride. I cannot feel a difference in effort.
One of the reasons I would really like to see the inside of an E3(3)
is that the output they're claiming is above the range for a 500ma
operating current. If they're boosting 500ma up to 700-750, I want to
know how they're doing it.
I don't think you're going to see any great efficiency gains in either
the hubs or driver electronics. There may be more large gains
available in the Lm/W area, but even that is maybe starting to peak.
90 lm/W at 500Ma is pretty d*mn good. The next significant advancement
(I hope) is going to be optical design.
Dark Horse
WillemJ
since I can easily feel it when I switch on even one light, let alone
a second. Thus my interest in using some of the efficiency gain of
modern lights for a reduction of the power I have to put in. With
halogen lights Jan Heine's trick with the SON20 only worked at such
high speeds that for me at least this was not a good road to follow.
However, since leds come on at much lower speeds, this has suddenly
become quite feasible. I have tried it on one bike (SON20 and Edelux),
and it works very well. As long as you do not try to run a second
light, you can thus now also use the much lighter SON20R (if you are
not a super heavyweight or ride very rough roads).
Schmidt at one time designed a more powerful 12 volt hub, to go with
the plans for a German 12V standard. It produced a high output already
at low speeds (as required by that standard), but it was a lot bigger
and heavier. There is no free lunch, and with rapidly improving led
technology I don't think they will ever produce this monster SON.
The future is indeed likely to bring more efficient emitters, so the
SON20(R) route will be even more attractive, I think. I am convinced
that in three years time we may have as much light from a SON20 and
one light, as you can now have from a SON28 and e.g. two lights. There
is also gain to be made in another direction: quality. LED's don't
have a continuous spectrum, and I would gladly trade some total output
for a better spectral range. In fact, this explains why the Edelux is
a bit greenish. Schmidt say on their website that they carefully
considered the different colour options, and this was the best they
could find for most cycling conditions. The new B&M Cyo with reflector
also trades some of its extra output in a different direction again.
The reflector spreads the light a bit more than the original lens, and
thus gives more light immediately in front (and addresses a complaint
from some), and it also has slightly more stray light above the upper
cut off, to make it easier to see signs etc. The price for this is, of
course, less light directly ahead than you have with the Cyo Sport
(only as much as with the original IQ Fly). Schmidt may also bring out
a version of the Edelux with this reflector. At least, that is what
Andreas Oehler vaguely suggested on a German forum. It may be a neat
way to use the extra output from the next bin up.
Willem
Mike Sturgill
The SON hub can indeed deliver more current than 500 mA. With 3 emitters
(Seoul P4s), I was able to get 1.1 A out of the hub. When connected to 6
emitters, the maximum I could get was 600 mA. The maximum numbers are
based upon speed and driving circuit. Many circuit designs may be found
here: http://www.pilom.com/BicycleElectronics/DynamoCircuits.htm
As has been stated, LEDs are more efficient (lumen/watt) at lower drive
currents. So even though the maximum current drops with 6 vs 3 LEDs, the
efficiency of the LEDs allow more light output with greater numbers of
LEDs. This is as would be desired, I assume. I have not tried more than
6 LEDs. For my setup and my eyes, 6 LEDs is perfect. I'm definitely in
the "more light is better" camp. And for the record, I have never been
flashed by oncoming cars.
I have a combination of narrow (5 deg) optics and oval optics (5x20
deg). This provides great peripheral lighting as well as lighting signs
and road furniture a half mile up the road. I've tested it side by side
with 3 different HIDs on dark back roads, and in all cases it put out
more light than the HIDs.
Also, in no case can I feel the drag. I can't speak for the E3 or other
commercial designs since I don't have those. I could feel the drag when
I ran my dual E6 setup.
YMMV.
-Mike
saltytri
If one were to adapt an existing LED flashlight to run on a Schmidt
hub, would it be necessary to add a voltage regulator after the full-
wave bridge? Specifically, I wonder whether it would be practical to
substitute a simple bridge, cap and regulator circuit (say, a 7805)
for a 18650 lithium battery in a light that uses this pill:
http://www.lighthound.com/Lumens-Factory-D26-LED-Special-36-volt-250-Lumens-Lamp-Assembly-for-SureFire-C-P-Z-Series-Select-Tint_p_2750.html
I'd appreciate your thoughts on this approach.
Thanks,
David
Dark Horse
On Jan 12, 6:33 am, Mike Sturgill <mikesturg...@cox.net> wrote:
> I have a number of home made LED designs and have measured all of them.
> The SON hub can indeed deliver more current than 500 mA. With 3 emitters
> (Seoul P4s), I was able to get 1.1 A out of the hub. When connected to 6
> emitters, the maximum I could get was 600 mA. The maximum numbers are
> based upon speed and driving circuit. Many circuit designs may be found
> here:http://www.pilom.com/BicycleElectronics/DynamoCircuits.htm
perfect for what I want a light to do.
I have modified it to allow a 2/4-die option with an MC-E emitter. It
lets me visibly drop my light for oncoming, and provides good light at
really low speeds.
When you got 1.1A out of a SON, were you doing that by capacitor-
matching? This is the main reason I wish Schmidt would publish source
characteristics for their hub. I don't know enough yet to accurately
deduce them from observation.
D@mn steep learning curve out here.
>
> As has been stated, LEDs are more efficient (lumen/watt) at lower drive
> currents. So even though the maximum current drops with 6 vs 3 LEDs, the
> efficiency of the LEDs allow more light output with greater numbers of
> LEDs. This is as would be desired, I assume. I have not tried more than
> 6 LEDs. For my setup and my eyes, 6 LEDs is perfect. I'm definitely in
> the "more light is better" camp. And for the record, I have never been
> flashed by oncoming cars.
Lights
Brakes
Traction
This I believe.
>
> I have a combination of narrow (5 deg) optics and oval optics (5x20
> deg). This provides great peripheral lighting as well as lighting signs
> and road furniture a half mile up the road. I've tested it side by side
> with 3 different HIDs on dark back roads, and in all cases it put out
> more light than the HIDs.
switch it 2/4/5 for low/high/descent.
There's a thread going on Candlepower forums right now, if you go
there. Nominal topic is beam shape for road bikes, and what people
would like that to be. There is some interesting thought there if
you're curious.
>
> Also, in no case can I feel the drag. I can't speak for the E3 or other
> commercial designs since I don't have those. I could feel the drag when
> I ran my dual E6 setup.
>
> YMMV.
> -Mike
>
>
>
> Dark Horse wrote:
>
> > One of the reasons I would really like to see the inside of an E3(3)
> > is that the output they're claiming is above the range for a 500ma
> > operating current. If they're boosting 500ma up to 700-750, I want to
>
> - Show quoted text -
Beezodog
them in say parallel, this would be preferrable to running a single
unit on my bike?
WillemJ
translation of what Schmidt themselves wrote about it.
Willem
Beezodog
currently run dual E6s on both my bike and my wife's. I want a setup
that is at least as bright as that setup. I am having a difficult time
imagining that a single Edelux LED lamp can outperform (i.e. outshine)
a dual E6 setup. Am I mistaken in that assumption?
Dark Horse
photons..."
I am having a difficult time
> imagining that a single Edelux LED lamp can outperform (i.e. outshine)
> a dual E6 setup. Am I mistaken in that assumption?
Go to White's beamshot page if you need to check.
http://www.peterwhitecycles.com/headlights.asp
Or here:
http://www.nabendynamo.de/produkte/Edelux.html
On Jan 13, 1:06 pm, Beezodog <Beezo...@gmail.com> wrote:
> My biggest concern is in settling on a single light configuration. I
> currently run dual E6s on both my bike and my wife's. I want a setup
> that is at least as bright as that setup.
single, once you get above a certain speed. I'll guess that speed at
9-10 mph, based on what I do know of dual-emitter configurations. A 2S
setup does not do as well as a single at low speeds.
Those two Edelux/IQ/E3 in parallel will not give you twice the light
of a single. Thumbnail calculations give you an optimistic 20%
increase over a single light. 2P setups will, however, do better at
low speeds.
If you don't feel like building sometihng, then the sensible approach
is to run two of something in series, and add or include a switch to
take one of them out of circuit. A setup like that would yield 2-3
times the light you're getting now. At least.
Dark horse
pamela blalock
edelux and now the e3triple. The edelux and e3 both outperform the e6 dual
setup, with the one caveat that the single light works best mounted
centrally, not in the side of the fork. Mount it on the side of the fork,
you will have a tire shadow.
Now if you want to spend more money, and expend a bit more energy you can
certainly go with duals and the e3 triple, but my subjective, and I believe
honest opinion of both the edelux and e3 is that the light output is better.
Recall the fire sale I had a while back on e6 lights. I was happy enough to
replace them all, with single lights.
But let's pump some more money in the economy, buy everything in pairs :)
Pamela
-----Original Message-----
From: ran...@googlegroups.com [mailto:ran...@googlegroups.com] On Behalf Of
Beezodog
Sent: Tuesday, January 13, 2009 4:07 PM
To: randon
Subject: [Randon] Re: E3 triple
Michel Gagnon
> My biggest concern is in settling on a single light configuration. I
> currently run dual E6s on both my bike and my wife's. I want a setup
> that is at least as bright as that setup. I am having a difficult time
> imagining that a single Edelux LED lamp can outperform (i.e. outshine)
> a dual E6 setup. Am I mistaken in that assumption?
I have experience with the Lumotec IQ Fly. Even though it is less bright
than the Edelux, it is brighter than two E-6.
If you insist and want to go with two lights, I would definitely suggest
you make your own switch box to short-circuit one at low speed.
One idea. Has anyone used dual LED headlights? Considering the
advantages and drawbacks of the symmetrical vs German-regulated design,
I was wondering how the following design would work :
– Symmetrical beam like the regular E-3, preferably used at all times
– Asymetrical beam like the Delux, ideally short-circuited so it is used
at high speed
P.S. I'm not really interested in buying a pair -- or even one -- of
these. Not that I have a problem with those lights, but simply I usually
ride on well-lit streets.
--
Michel Gagnon
Montréal (Québec, Canada)
WillemJ
I think you are indeed mistaken - fortunately. Even a single Edelux
gives an amazing amount of light. For me, that is more than enough. My
wife has one (on the fork crown of a 26 inch wheeled bike) with a
SON20 for even lower resistance, and that is what I would recommend to
anyone who does not already have a hub. However, a second light on a
SON20 is unrealistic.
In your case, the situation is different. You and your wife both have
SON28's, and you also still have the secundary E6Z's. So I would just
keep those, and only replace the primary E6's with an edelux. As I
translated above, the secundary E6Z's will not make much difference,
and I would therefore not buy them new anymore. But since you have
them, why not keep them for fast descents, where their added
resistance is no problem, and where from 18 km the in built capacitor
will also tweeze the last bit of extra light from the primary edelux.
Just try - I would be curious to hear of the results.
If you are not into serious tinkering and you really wanted to combine
two led lights, and with different beam patterns, I think the best
combination is probably two Cyo's, one with the reflector as a
primary, and one sport model without for the distance. But as Michel
also said, these lights really are very bright - all of them.
Willem
Ingle, Bruce
> sweat into photons..."
Or, for many of the lights discussed here, turning dollars into
photons...
- Bruce
RUSA2691
(most of it) but thought I'd give my non-technical two-cents:
I have a pair of eDeluxes, mounted on the low-rider braze-ons, halfway
down the fork blades, wired in parallel.
As to the arguments against purchasing an additional light:
I've spent much more on much less (like one month of bus-commuting
into NYC or seven parking tickets in town). One crash-causing pothole
avoided is worth it.
As to the placement:
I agree with the previous comment that lower is better for throwing
the brightest light out at the optimal distance.
I'd predict that much lower would produce too diffuse a light and much
higher would mean the brightest light being too close to the bike.
As to series vs. parallel:
I wouldn't know how to wire in series (a) and (b), I tend to ride on
the slower side and grew tired of switching my secondary E6 off and
on.
It's been stated repeatedly that the benefit of dual lights is a
'wider' pattern. For me, the advantage is depth, not width:
I've angled each light in a wee bit so both are centered in front of
the bike.Then the second light is angled slightly higher on the
horizon, giving a deeper patch of brightest light plus some spill-over
up onto signs, guard-rail reflectors, deer, etc.
This is what I've found to be a good setup; I hope I've been helpful
to someone.
One more note: I've found that very bright, concentrated light, like
from halogen/battery types I've used, dim my peripheral vision too
much. The eDelux are bright enough for me; I'm not sure I'd want a
light any brighter, at least in any one concentrated spot.
-Paul
Beezodog
Evidently the eDeluxe lights are at least twice as bright as an E6. By
some accounts three times as bright! In which case you actually get
MORE light with a single eDeluxe than with a dual setup of E6s.
As for wiring in series, the E6s are evidently always wired in that
fashion. I did the soldering for ours and while you have to take care
to get the spade connector on securely it is not too difficult a job.
I did got out and buy extra heavy duty shrink sleeves to ward off the
effects of a Chicago winter.
On a separate note I have read in the past that some brevets require a
dual light setup, or more precisely that there always be redundancy
built into the lighting setup. So I am curious that most of the folks
here don't see the need for secondary lights. Or is this a case of
ride organizers bowing (as I now have to do) to the advances in
technology that make superfluous the redundancy requirements?
WillemJ
technology have bypassed them. For I can think of one solution, but it
will only appeal to some. B&M still do a 12 volt taillight for their
now obsolete 12 v set. If you use two edeluxes in series, and have
both of them on at all times, you could use this 12 volt taillight.
But really, it is time for Bumm to catch up. What I do is I have a
bumm generator taillight on the fender, and a battery bumm XS
reflector plus taillight on the rack ( you could do it the other way
around, but this seemed neater with the wiring). It also gives me
redundancy. But indeed.
Willem
On 14 jan, 17:34, Jan Heine <hein...@earthlink.net> wrote:
>
> >Thanks for you anecdotal recount, it makes all the difference.
> >Evidently the eDeluxe lights are at least twice as bright as an E6. By
> >some accounts three times as bright! In which case you actually get
> >MORE light with a single eDeluxe than with a dual setup of E6s.
>
> the useful beam is at least 3 times as wide, and the transitions from
> bright to dark are more gradual. The latter is a very important
> consideration for me, as I don't like the tunnel vision I get with
> highly focuses lights. I want to see the deer standing by the
> roadside before they run into the beam of my headlights!
>
> After using the Edelux, riding with the E6 is like peeping through a keyhole...
>
> When the E6 came out, I wrote that it was one of the best at the
> moment, but that I was looking forward to better lights. I did not
> think long before replacing the E6 with the Edelux.
>
> With the Edelux, I can see how the beam pattern could be improved
> slightly, but I doubt I'll buy a new light just because it is
> slightly better than the Edelux.
>
> I think with these new lights, bicycle lighting finally has become a
> mature technology. Now if we only could get decent generator-powered
> taillights...
>
> Jan Heine
> Editor
> Bicycle Quarterly
> 140 Lakeside Ave #C
> Seattle WA 98122http://www.vintagebicyclepress.com
Andrew Karre
a decent market for generator headlights because you can add them to
almost any bike with a minimum of hassle. Unless you have a custom
bike, though, you're left with extremely unappealing solutions for
wiring the taillight. Aesthetics aside, the length vulnerable wire is
enough to make me uninteresting in generator taillights until I have
the resources to buy a custom rando bike.
(Home theater manufacturers have come up with some extremely
interesting flat, surface-mounted wiring solutions for speakers and
components. I'd be interested in seeing someone adapt that technology
for bike wiring.)
Andrew
WillemJ
wires run zip tied together with the hoses for the hydraulic brakes or
the brake cables. It is neat and tidy, and gets them almost where they
need to be. The last bit is in a thin plastic sleeve along the stay of
the fender, to the rear of the fender to reach the fender mounted
light. I have never had a cable or contact failure on these bikes, and
aesthetically it looks fine to me (the frames are black, admittedly).
I don't like internal wiring because you cannot see what is happening,
and because I do not like to perforate my frame. Nor do I like the SKS
fenders with in built connectors. Had them: they fail. My experience
with a stable full of family bikes that are used daily in often bad
Dutch weather is that the generator lights are the most reliable
lights, and of them, the Schmidt stuff is by far the best in terms of
reliability (and so it should be at that price). No B&M failures so
far either, even if they are somewhat less well made. I am sure,
however, that there must be other good stuff.
Willem