Look through Peter White's website. He has more information in one place than any I've seen. I personally prefer the tire driven BM Dymotec for several reasons, but I don't commute in the dark everyday. I would go for a Schmidt dynohub if I was riding in the dark on a routine basis.

Marc

I have the Shimano Afline Dynohub (dsh501) and it has about 10,000 miles on it. Commuting, touring, singletrack and some epic 55 mph downhill runs in Colorado. Recently (last 500-1000 miles), it developed a pretty strong humming noise when I get to about 25 mph but it still works. The right cone nut came poorly adjusted from the factory/wheelbuilder and fouled out at about mile 5,000. I replaced it, which was quite difficult. No shop was able to order it locally, I resorted to importing it from the UK.

Also, overhauling the disc side bearing is a real PITA. It is much better sealed than the right side. I've changed the balls/grease 2x since I got it about 2 years ago. The hub has survived being submerged in FL swamp water and mountain stream water. It's not perfect but for most riders who aren't too demanding it's priced just right. I will probably get a SON hub in the future if/when this one kicks the bucket. Loose ball bearings are tiring to maintain.

You might want to do more involved googling "research" and find http://www.bikequarterly.com/images/...Resistance.pdf. Among other things, it shows good dynohubs have less than 1 W draw when not running the lights. Google a bit more and you'll likely find that good dynohubs have less rolling resistance, with the lights off, than not-so-good hubs. I.e., there's no practical downside to rolling resistance from a dynohub.

The electrical Output is nominally 3 watts, 6 volts..

light for bikes have long been on this standard.. when bulbs were used
the 3w load was split, 2.4 for the front, 0.6 for the rear..

Not sure such a tone was necessary, but I think it was clear in my post that I was talking about when the light is turned on, in which case dynohubs output 3W, drawing closer to 6-7 (suppose I didn't get the number right the first time, but that's further evidence against your point). Of course they have an off setting in which case a dynohub performs just like a regular hub, but I would think that's more than self-evident.

but as a human mass of muscle you put out several hundred watts.

not including the freebies, going down hill ..

While some hubs cost $100+, many don't -- dynohubs are a mainstream commuter/utility item in countries where bike commuting is common. A $30 Shimano dynohub is a perfectly fine entry point for commuting.

The new Luxos U is an exception -- its standlight stays on for 10+ minutes, and is easily bright enough for changing a flat. It has a much brighter standlight than most on the market, using an actual battery instead of a supercapacitor.

Somebody's got to ask; in your opinion, is it worth the cost to upgrade? And what did you upgrade from (assuming here it's not your first dyno-driven light)?

I can't answer for Josh. I'd like to hear his response.

Myself, it depends on where you're coming from. I own a 2009 Schmidt Edelux and a 2013 B&M Luxos-U. Both are the high-end light for their respective model years.

If you're coming from another high-end dynamo headlight, it's not worth upgrading at this time. If you're coming from a low to mid-range LED dynamo light (10-40 lux) it's certainly worth it, and if you're coming from an old halogen one, what the hell are you waiting for?

In a nutshell, in the five years separating my lights, there hasn't been significant improvement in the amount of light that high-end dyno headlights produce.

The difference between the Edelux and the Luxos are that the Luxos steals some photons from the central beam and spreads them into a wider pattern. Over 15 mph, I prefer my Edelux--it has better "throw". Under 15 mph I prefer my Luxos, especially when slowly picking my way through icy ruts and semi-plowed streets. It's also far easier to wipe collected snow off the lens of the Luxos--there's no bezel ring to trap the snow.

The Luxos also has a nice daytime running light mode, and the U model uses a battery for the standlight instead of a capacitor, and offers USB charging at the handlebar.

They're both fine lights, I recommend them both, if you choose the right one for your circumstances. If you have one, it's not worth "upgrading" to the other.

How do you get a decent lighting system if you only have 3 watts to deal with? I run 5 watts just on my tail, and I certainly would not be willing to reduce that. Even with an efficient shaped beam I'd think you'd want at least 3 watts up front as well. My current light draws 8.5 watts, I imagine a shaped beam could get that down to 5 and still seem reasonable, but I'd think 10 watts would be a minimal amount of light I'd want to commute with in total.

I'm not trying to diss dynos. I'm getting involved with the discussion because I'm interested. Just not sure if I'm interested enough to drop $250+ and build a wheel in the middle of the winter, rather than just continue to use the $30 light that I'm reasonably happy with now.

And 10 watts of light is going to use about 20 watts at the wheel- dynamo's aren't very efficient. That's about a fifth of your total output when cruising along at 15 KPH. It's gonna feel like pedaling through mud.

That was kinda what I was thinking. Glad you were able to verbalize it better than me :)

What light are you using on your tail light? A quick search for 5 watt tail light only found a 0.5 watt tail light for me (admittedly, a very very quick search).

To your wattage question, I own a long list of lights, from very expensive (Seca 1400), to middling expensive (2 Dinotte 400L's), to cheap (a Dinotte 200l AA), and a Lumotec Cyo dynamo light.

The weird thing is, on my road bike I put 2 Dinotte 400L's on it, so 800 lumens total. The difference between low, medium, and high power is doubling the lumen output at each stage. But one of the weirdnesses of light is that turning up the power only gives me a perceived increase of about 10% more light. Believe me, I tried it again and again and again, and I just couldn't believe it - making the light twice as bright only helped a very little bit.

The shaped beam on a dynamo doesn't *just* avoid wasting light off to the sides. One of the problems with my big battery lights is that you up the power, then your adjust adjust to more light, losing some of the effect. When I moved to a dynamo light, one of the advantages was that I could still see off the the sides much like I could when the light was off. If I have my dynamo light I can see everything around me just like as if I didn't have the light on, but I can also see better in front of me where the light is actually hitting. (Note: This is also likely because I live in the city where there's some peripheral light, if you lived in the country things might be different)

You mentioned in another thread that you just bought and returned a Phillips Saferide, other than the abysmally crappy battery life, how did you like the light from the light itself for riding? A dynamo light will be similar to that (they also make the Saferide in a dynamo version).

first off I must apologize, with the up coming holidays I haven't had a chance to follow the board and this thread . I also got my winter tires in the mail and had to get them on for the weather we've been having. thanks for all the info. I should have included a little info about my plans. My commuter / tourer has about 9000 hard miles on the current rear wheel and unknown miles [more] on front. I commute 24+ miles round trip daily year round and need lights every morning , early mornings and afternoons when inclement weather.
I was considering a new bike, but I really love my poprad, so I've been collecting parts to replace the drive train , and its time to do the wheels. I'm planning on tearing the bike down in the spring and rebuilding , possibly changing out the fork and installing a disk brake.
So thats where I'm at. Again, thanks for all the info, a lot to ponder over the cold winter months.

I bought the IQ2 Eyc. I wasn't expecting much and I certainly didn't get much, and according to peterawfulwhitecycles's overinflated beamshots, the Eyc just has a narrower beam than the Luxos. Needless to say, I took it off and and used my diy lights instead.

A proper diy circuit can drive a dynamo light beyond 10w, enough for 1000 lumen. I have to say, what's on the market for dynamo specific lights disappoint me unless you're spending north of $300.

Incorrect.

Magicshine taillight.

Watts are a measure of power consumption, not of light output. Using watts to compare light output works only in an apples to apples situation.

If you're using watts to refer to light output, then I'm guessing you're using incandescents of some sort? (Halogens are also incandescent.) Remember that incandesents are very, very, inefficient. It helps to think of them as heaters that give off light as a waste product. Most of the power consumption goes into heating the filament, in much the same way as tube amps use most of their power to heat the tubes.

My first real light was a 10-watt NiteRider TrailRat halogen. Its light output was roughly equivalent to the pair of 200-lumen DiNotte 200-L-AA lights I replaced it with. They draw less than three watts each.

The current crop of LEDs are very efficient. Both my dynamo lights are easily 2x to 3x brighter than that pair of 200-lumen DiNottes, so, their 2.4 watts of power consumption is producing light output easily in excess of a 10 watt halogen--at least judging by my old eyes and memory.

The German lights all rate their output in lux, which is a measure of how much light actually falls on a specific target at a specific distance. It's a much better comparison than lumens, which is a raw measure of light output, not necessarily what ends up on the road.

Same goes for taillights, BTW. The pair of LEDs in my B&M Toplight Line Brake Plus draw a half-watt together, but are easily visible in daytime. This past summer I chased a bike on a sunny day to find out what taillight he was using in the daytime--a B&M Toplight Line Brake Plus. That's why I bought one.

I agree that Peter White's beamshots leave something to be desired. While they don't come close to representing what I see on the road with my lights, they do provide a pretty good way to compare the different lights Peter sells. Just be aware that what you'll see IRL is about a third of the brightness seen in his photos.

IMHO, the problem is the ISO setting he uses, 3200. Higher ISO settings reduce the contrast between light and dim areas, bringing up the dimmer areas to meet the brighter ones. You can't make the fully-exposed areas any brighter, so you bring up the dimmer ones, which effectively, is reducing the contrast between them. (EDIT: This is why even after "darkening" his photos due to customer complaints, Peter's photos are still a long way off from reality. He needs to re-shoot them using a lower ISO setting.)

I used to shoot a lot of existing light photos back in the day when we used silver instead of silicon, but the same principles apply. I found an ISO of about 800 seemed to come closest to what I saw with my eyes. To this day, I still use ISO 800 when I shoot existing light.

http://www.brucew.com/images/wallpaper-linux-thumb.jpg

Clicky to embiggen (Caution mobile users: 2MB file.)

Not an upgrade, an addition -- my other good dynamo light is a Philips Saferide, which now lives on my touring bike.

Touring bike has a first-generation Schmidt SON dynohub, commuter has a Sanyo, no noticeable difference in light performance if I swap wheels.

The Luxos U has a wider beam pattern, especially at lower speeds -- its beam pattern changes with speed, accessory flood light LEDs on the sides that come on at lower speeds. I don't have the facilities to do a scientific test, but in use, it feels like the Luxos should be significantly more visible to cross-traffic at intersections, as well as being more visible when stopped.

Luxos U has slightly higher total output, not enough to be worth upgrading from one to the other, but double or better what older LED dynamo headlights put out.

Luxos U has a battery-backed standlight instead of capacitor, so it holds a standlight charge longer (over the weekend or longer) and can also run all output on high-power mode, 90 lux vs. 70 if I remember correctly.

Considering the Luxos U costs less than I'd spend on gas for a month of commuting by car, I can't see any reason not to buy the best dynamo headlight on the market. I think that's the Luxos U at the moment, but the Philips Saferide is still right up there.

Agreed, 800 is a reasonable speed for existing-light shooting. Here's the Philips Saferide at 800, haven't taken photos of the Luxos yet.

http://farm9.staticflickr.com/8498/8...61092851_z.jpg
Philips SafeRide Bicycle Headlight Beam Pattern by joshua_putnam, on Flickr

And here's a different view of the Saferide,
http://farm8.staticflickr.com/7119/8...a14ce6f0_z.jpg
Headlight Beam Pattern, Philips Saferide by joshua_putnam, on Flickr

Hard to say which is a better rating, unless the lux ratings take several targets into account. Otherwise you end up with a low-output, highly-focused beam being the same 'rating' as a high-output, wide one.

Certainly there are ways to game any system. There's no question that manufacturers can and do game the system (and sometimes outright lie) to make thier product seem better than another. Nothing's perfect.

Yes, I realize that. I did take shaped beams and their better efficiency into account when I mentioned a 5 watt headlight instead of my current 8.5 watt one.

However, in the case of taillights, the whole point of them is to smear light all over the place. I run a 3 watt main light and two 2 watt secondary lights. I don't always run all three lights so I figured 5 watts is a good number. I actually probably would only run the 3 watt on the dynamo since the whole point of having multiple lights is that so you can have an entire system fail and still have a light. For the same reason, even if I had a dynamo front light I'd still have to carry at least a hand torch and a lockblock. I've been stuck 4 miles from anywhere in the middle of a snowy, potholed gravel road in the middle of the woods at 4 in the morning before with a dead light, and it's NOT FUN. In that case I had no backup and wound up slowly making my way home by holding my taillight on steady mode in my hand. Also not fun.

As far as new lights being efficient, I suppose the Philips Saferide counts as a new, efficient light. It comes with 2300 mah batteries rated at 2 hours on high. That's 1.15 amps draw at 4.8 volts nominal, which is 5.5 watts power draw. A touch more than the 5 watts I mentioned above (where I was, admittedly, just pulling numbers out of a place).

In any case, if I continued to use a battery powered 2nd taillight, that's 8 watts of draw.

I did think of something else last night on my way home. A lot of the time in the winter (the only time I really need a light), I'm plowing through poorly plowed, rutted mush. I was doing so last night. When the conditions are like this, I'm only going about 7 to 9 MPH at best, sometimes 4 MPH, steering like mad and skidding the front wheel all over to muddle through the criss-crossing car wheel ruts, for 20 minutes at a time. And that's the time when I need the most light, because it's almost impossible to see the potholes through the rutty snow even with a bright light. Also the tail is doubly important on a road where even the cars are mooshing around all over, around bends and hills in the black with full tree cover.

Is a dynamo going to be able to deliver the power at 6 MPH to keep a 1000 lumen light and a 3 watt taillight going indefinitely? 20 minutes is too long to expect a standlight capacitor to hold.

I continue to run battery lights in addition to the dynamo ones, simply from a belt-and-supenders standpoint. For the record, when I ran strictly battery lights, I also ran two on each end, all powered separately. I've had lights die unexpectedly in a commute, and been glad I had the second.

I'm also of the opinion that standlights on dynamo lights dim too much and then are easily lost among the background lights of the city, my primary habitat. The last thing I want is to start from a light, having someone turn left in front of me because the standlight disappeared among the background lights and they forgot I was there.

Dynamo systems (both hub and lights together) that are in compliance with the German StVZO regulations must achieve full power (from the hub) and brightness (from the lights) by 10 km/h, or 6.2 mph. A homebrew system like you're describing is not going to be in compliance, so there's a very strong YMMV with that.

I have a similar situation on my commute--picking my way through rutted streets and trying to assess, "will the hardpack give way and sideslip or not?" and "will the sides of that rut let me move over to the next one?"

I've found the Luxos-U running in its higher power 90 lux mode is sufficient in those conditions right up until low-speed flicker sets in at around 5 mph. Also a reason I run an auxiliary battery light.

There is no perfect system at this stage of the game--either battery or dynamo. The best you can hope for is to put together a compromise.

My diy lights light up brighter at lower revolutions than my Eyc, so much for "German regulations". At my stage of the game, I don't have to compromise. I can get get battery level brightness with dynamo power without ever having to worry about the disadvantages of battery lights, namely charging them, how they function in the cold, and their limited runtime. I also never carry battery lights for short rides anymore, there's no point, I have all the lights I need.

Bubble is not in Germany so their lighting minimum standards are Moot ..


Eyc on my Brompton is fine.... I dont need to blind others, or descend at 50mph..

I'm getting one of those 12V 6W bottle dynamos, and putting together my own lights from a Costco 5W Cree LED flashlight ($20 for 3) and some custom electronics. The flashlight head goes into an old Royce-Union headlamp with a small rechargeable battery pack, for that vintage look. It will have a USB port, too, for powering i-phones and suchlike.

Dude, I was willing to read your dissenting opinions and give them consideration, but then you said "breaking". I'm out.

I'd like to hear more about this. I'm running a cheap-azz flashlight mounted with interlocking hose clamps. I really like how bright it is (CREE XML T6, allegedly 1600 lumens, actually I don't know, but it's like a car headlight), I'm content with recharging 18650s (weekly, and I have a backup pair in the seatbag that will get me home if they fail early). I'm interested in maybe someday upgrading to dynohub, but are there super-bright CREE dynolights out there? Or starting to reach the market?

If you have not already found the following thread you may find it usefull. http://www.bikeforums.net/showthread...nerator-system