|
|
Thread Tools | Display Modes |
#81
|
|||
|
|||
New B&M 100lux headlight.
On Sun, 10 Dec 2017 12:07:54 -0600, Tim McNamara
wrote: On Sun, 10 Dec 2017 10:58:56 +0700, John B wrote: On Sat, 09 Dec 2017 19:58:49 -0600, Tim McNamara wrote: The piece that you and Steven don't seem to get is that "good enough" is good enough. Better than "good enough" rapidly becomes multiple redundancy and enters diminishing returns. I don't need a beam suitable for going 100 miles an hour because I never will go that fast on my bike. But I descended on unfamiliar roads with my old halogen setup at 25 mph; I could descend faster with my current LED light and expect that will be even more true with the incoming eDelux II. When is enough enough? When one succeeds in recreating daylight? Fiat Lux! Damn, probably 50 years ago I rode back and forth to work with a bottle generator and a headlight with an incandescent light bulb. And, part of the ride was across small "dikes"perhaps 10 - 12 inches wide that divided a rice paddy. Slipping off the path there ended up with the cyclist being knee deep in very smelly sludge. Amazingly I rode that bike nearly every day for a year or so and never had a problem. I'd have probably ended up in the rice paddy in broad daylight. ;-) Back in those days the Japanese used human feces for fertilizer and rice paddies had a rather "rich" odor. My guess was after you fell in once you'd have gotten off and pushed from then on :-) -- Cheers, John B. |
Ads |
#82
|
|||
|
|||
New B&M 100lux headlight.
On 08/12/17 19:52, Oculus Lights wrote:
On Tuesday, December 5, 2017 at 9:35:43 PM UTC-8, James wrote: On 04/12/17 11:28, Oculus Lights wrote: On Tuesday, November 21, 2017 at 2:34:34 PM UTC-8, James wrote: https://www.bike24.com/p2144878.html Is there a power rating? One can safely assume it will work with any normal 6V/3W dynamo. 100 lux at 10 meters, as the STVZO test requires, is exceedingly bright. I'm hesitant to state they "must" be drawing at least so much power, but my gut feeling is that its in a range that a single LED can't handle. Depends on how the light is focused. Anyone can rate a light without stating the distance. My single LED 325 lumen measures 33 lux at 10, 500+ lumen measures 50 lux, and the best of the others on the market, such as Supernova's 205 lm that's standard equipment on many e-bikes, measure 25 lux at 10 meters, at most. My B&M IQTec Premium is rated at 80lux. It also works with a 6V/3W dynamo. The light is focused to a very bright band just before the cut off, so that you can aim the light well in to the distance and achieve a relatively even illumination of the road surface over the entire distance. If yours is only reaching 33 lux, it is less well focused and more of a flood light. See images here. http://www.bentrideronline.com/messa...d.php?t=131473 Those "if" statements followed by unsure speculation from conventional wisdom are always fascinating. There was nothing unsure or speculative about my reply. -- JS |
#83
|
|||
|
|||
New B&M 100lux headlight.
On 10/12/17 10:40, Oculus Lights wrote:
On Tuesday, November 21, 2017 at 2:34:34 PM UTC-8, James wrote: https://www.bike24.com/p2144878.html My ray traces shown on the Learn More page on the Oculus website have literally no center spot, virtually no red dots at all, ... https://www.barrybeams.com/learn-more.html On the Learn More page I see a screen shot of a "computer simulation". I assume that is of the light projected horizontally against a vertical surface. Given that, it will not illuminate the road surface evenly at all. I also see a picture of a supposed NASA Rover with an Oculus flood light, that shows a bright spot just in front of the vehicle. Is that what I was supposed to see? -- JS |
#84
|
|||
|
|||
New B&M 100lux headlight.
On Mon, 11 Dec 2017 13:31:17 +1100, James
wrote: On 10/12/17 10:40, Oculus Lights wrote: On Tuesday, November 21, 2017 at 2:34:34 PM UTC-8, James wrote: https://www.bike24.com/p2144878.html My ray traces shown on the Learn More page on the Oculus website have literally no center spot, virtually no red dots at all, ... https://www.barrybeams.com/learn-more.html On the Learn More page I see a screen shot of a "computer simulation". I assume that is of the light projected horizontally against a vertical surface. Given that, it will not illuminate the road surface evenly at all. I also see a picture of a supposed NASA Rover with an Oculus flood light, that shows a bright spot just in front of the vehicle. Is that what I was supposed to see? I think the rover photo is intended to show how the Oculus Light improves illumination at a distance. I don't see that on the photo. I find it very difficult to judge intensity from an image consisting of different white levels. To be sure, I converted the image to false color: http://www.11junk.com/jeffl/pics/bicycles/Front-Light-False-Color/ Original image: http://www.11junk.com/jeffl/pics/bicycles/Front-Light-False-Color/slides/occulus-rover.html False color using the "16 colors" LUT (look up table): http://www.11junk.com/jeffl/pics/bicycles/Front-Light-False-Color/slides/occulus-rover-false-color.html Note that the false colorizing assumes that the ground is uniform and of equal reflectivity across the viewing area. I'll spare you my opinion of the beam pattern because I'm not certain that my criteria is the same as NASA's. You can make your own false color images using a Java program called ImageJ. Instructions lifted from my previous posting at: https://groups.google.com/d/msg/rec.bicycles.tech/wPP-6f7mJOE/75h2lv1aAwAJ https://imagej.nih.gov/ij/ Load original image. Convert to gray scale with: Image - Color - Channels Tool Set it for Grayscale, which should also set it for only Channel 1. The image should now appear in gray scale. Convert the image to false color with: Image - Lookup Tables Pick an LUT (look up table). I suggest 16 colors or BRGBCMYW. -- Jeff Liebermann 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558 |
#85
|
|||
|
|||
New B&M 100lux headlight.
On Tuesday, November 21, 2017 at 2:34:34 PM UTC-8, James wrote:
https://www.bike24.com/p2144878.html -- JS Biased? Its hard to be impartial all the time, but I do work on not drinking my own KoolAid, at least not too much. Everything in my lights is my own design and final assembly. Circuit boards are assembled in Scott's Valley from bare boards made in Illinois. The collimator is a standard part number, but from a custom run made by the manufacturer with tighter specs and tolerances than their standard part. The LEDs on the 1800Ultra are from a special kit code Cree's west coast rep got arranged for me. The 3000Extreme LEDs, so far, are from a specific reel that DigiKey has after I ordered from a few reels to see what variation there would be within a wider color temperature range, then picking what one particular trusted long term tester liked best. The generator vs battery debate, like schrader vs presta, tubular vs clincher, has no best solution for all riders. My extended battery life in a self contained unit lighting the road better then other lights for a given burn time does give options that only generator lights could have solved before. Note, generators can be used with Oculus. Oculus has input holes on the main board labelled for external power inputs.. You can connect a power converter giving 3.0 - 4.2VDC into these holes. Oculus will warranty such usage. But AC pops the circuit quicker than you can say "blackout". Oculus also works in a hybrid battery + generator mode. Do this with a common generator to USB 5.0 - 5.1VDC output plugged into the charging jack, and run the light nearly endlessly at the third brightness, while also keeping the battery charged for when you need or want the higher brightness. Use the USB charging adapter cord listed on the Oculus website, or equivalent part sold by many electronics outlets. |
#86
|
|||
|
|||
New B&M 100lux headlight.
On Sunday, December 10, 2017 at 7:34:16 PM UTC-8, Jeff Liebermann wrote:
On Mon, 11 Dec 2017 13:31:17 +1100, James wrote: On 10/12/17 10:40, Oculus Lights wrote: On Tuesday, November 21, 2017 at 2:34:34 PM UTC-8, James wrote: https://www.bike24.com/p2144878.html My ray traces shown on the Learn More page on the Oculus website have literally no center spot, virtually no red dots at all, ... https://www.barrybeams.com/learn-more.html On the Learn More page I see a screen shot of a "computer simulation". I assume that is of the light projected horizontally against a vertical surface. Given that, it will not illuminate the road surface evenly at all. I also see a picture of a supposed NASA Rover with an Oculus flood light, that shows a bright spot just in front of the vehicle. Is that what I was supposed to see? I think the rover photo is intended to show how the Oculus Light improves illumination at a distance. I don't see that on the photo. I find it very difficult to judge intensity from an image consisting of different white levels. To be sure, I converted the image to false color: http://www.11junk.com/jeffl/pics/bicycles/Front-Light-False-Color/ Original image: http://www.11junk.com/jeffl/pics/bicycles/Front-Light-False-Color/slides/occulus-rover.html False color using the "16 colors" LUT (look up table): http://www.11junk.com/jeffl/pics/bicycles/Front-Light-False-Color/slides/occulus-rover-false-color.html Note that the false colorizing assumes that the ground is uniform and of equal reflectivity across the viewing area. I'll spare you my opinion of the beam pattern because I'm not certain that my criteria is the same as NASA's. You can make your own false color images using a Java program called ImageJ. Instructions lifted from my previous posting at: https://groups.google.com/d/msg/rec.bicycles.tech/wPP-6f7mJOE/75h2lv1aAwAJ https://imagej.nih.gov/ij/ Load original image. Convert to gray scale with: Image - Color - Channels Tool Set it for Grayscale, which should also set it for only Channel 1. The image should now appear in gray scale. Convert the image to false color with: Image - Lookup Tables Pick an LUT (look up table). I suggest 16 colors or BRGBCMYW. -- Jeff Liebermann 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558 @Jeff, do you want the original NASA Roverscape instead of what you pulled off the website? Might get more even false coloration. The NASA setup is a special layout of multiple beams to create an even spread above a needed threshold, to fill a desired goal downrange and for a given width. The initial evenness of the Oculus beam allows them to be 'woven' together to get that level of evenness over a larger field. Combining multiple round beams leaves hot spots, and also needs much more power than the rover's power budget to create for the needed brightness for that large a light field. |
#87
|
|||
|
|||
New B&M 100lux headlight.
On Tuesday, November 21, 2017 at 2:34:34 PM UTC-8, James wrote:
https://www.bike24.com/p2144878.html -- JS Biased? Its hard to be impartial all the time, but I do work on not drinking my own KoolAid, at least not too much. Everything in my lights is my own design and final assembly. Circuit boards are assembled in Scott's Valley from bare boards made in Illinois. The collimator is a standard part number, but from a custom run made by the manufacturer with tighter specs and tolerances than their standard part. The LEDs on the 1800Ultra are from a special kit code Cree's west coast rep got arranged for me. The 3000Extreme LEDs, so far, are from a specific reel that DigiKey has after I ordered from a few reels to see what variation there would be within a wider color temperature range, then picking what one particular trusted long term tester liked best. The generator vs battery debate, like schrader vs presta, tubular vs clincher, has no best solution for all riders. My extended battery life in a self contained unit lighting the road better then other lights for a given burn time does give options that only generator lights could have solved before. Note, generators can be used with Oculus. Oculus has input holes on the main board labelled for external power inputs.. You can connect a power converter giving 3.0 - 4.2VDC into these holes. Oculus will warranty such usage. But AC pops the circuit quicker than you can say "blackout". Oculus also works in a hybrid battery + generator mode. Do this with a common generator to USB 5.0 - 5.1VDC output plugged into the charging jack, and run the light nearly endlessly at the third brightness, while also keeping the battery charged for when you need or want the higher brightness. Use the USB charging adapter cord listed on the Oculus website, or equivalent part sold by many electronics outlets. |
#88
|
|||
|
|||
New B&M 100lux headlight.
On 12/10/2017 8:01 PM, John B. wrote:
On Sun, 10 Dec 2017 12:30:16 +0100, Sepp Ruf wrote: John B. wrote: On Sat, 09 Dec 2017 19:41:16 -0600, Tim McNamara wrote: On Sat, 9 Dec 2017 10:34:13 +0100, Sepp Ruf wrote: Because you generally want to shield from a flattish \ line of approaching lights, but not shield your view to the sides, the flatter brims work better. We're talking about cycling caps, not baseball caps. A cycling cap bill's curve would shade the center of the field of view while allowing peripheral vision. True. Just use what works best for you, short of Yehuda Moon style. Our beloved champions aren't strict either, at least after a bad (styrofoam) hair day: https://ekz-crosstour.ch/wp/wp-content/uploads/2017/10/RSP_1321-1024x682.jpg Just like the curved cutoff in the awful OculuStvzo engineering sample doesn't work... There's no equivalence between the shape of the brim and the shape of the cutoff. The cutoff is designed to manage outgoing light, keeping it where it is useful and not shining it where it is not. Not functionally equivalent, of course. You did see the video he linked? There are only two choices with a reversed crescent cutoff: Either aim it low enough that it does not glare directly opposing traffic, or aim it high enough to keep the 90 and 270 degree fields of vision for fast curves. black-tinted rear window; auto-dimming mirrors; manure-spreading trailer. I have an auto-dimming mirror in my new car. It verges on useless as it does not dim anywhere near enough. I'd rather have a manual mirror so that I don't have to reach up and point the damned mirror at the ceiling so that I'm not blinded by the vehicle behind me. My wife's small Honda has a manual tab on the bottom of the mirror. Flick it with your finger and the "guy behind"'s lights are focused on the ceiling :-) Still leaves the side mirrors vulnerable. Okay, so the trailer will need to be wide enough to block the view, but not as wide as to require the mounting of extra side mirrors on the Honda. I'm not sure about side mirrors. I have them on my pickup and my wife's little Honda has them But I don't remember ever having a problem with them. I sometimes have the problem, mostly on freeways. My wife even has that problem sitting in the passenger seat. It occurs mostly when truck drivers (pickups, box trucks, tractor-trailer rigs) have high headlights and insist on using high beams. It's more MFFY behavior. Our current car has electrically adjustable side mirrors. If speed differentials are small enough that the problem persists for more than a few seconds, I point the mirrors downward until the offender is out of range. If they pass quickly, I just shade my side mirror with my hand. -- - Frank Krygowski |
#89
|
|||
|
|||
New B&M 100lux headlight.
On Sun, 10 Dec 2017 23:46:48 -0800 (PST), Oculus Lights
wrote: @Jeff, do you want the original NASA Roverscape instead of what you pulled off the website? Might get more even false coloration. Sure, I'll give it a try. Email address is in the message signature. Using ImageJ is easy enough and using JVA should run on Windoze, Mac, Linux, etc. The program came from the NIH (National Institute of Health) and was used primarily for analyzing medical microscope photos. There is a HUGE selection of plugins to do weird things with photographs. https://imagej.nih.gov/ij/plugins/index.html False coloring is only one plugin in the programs bag of tricks. The hard part will be calibrating the intensity to false-color levels, so that the images make sense and can be compared. I have some ideas on how to do it, but so far, have done nothing. Also, I tried to setup a color LUT (look up table) that will give a finer intensity resolution in the area of interest, but the results were awful. The NASA setup is a special layout of multiple beams to create an even spread above a needed threshold, to fill a desired goal downrange and for a given width. The initial evenness of the Oculus beam allows them to be 'woven' together to get that level of evenness over a larger field. Combining multiple round beams leaves hot spots, and also needs much more power than the rover's power budget to create for the needed brightness for that large a light field. I can guess(tm) why they need a large even beam directly in front of the rover. There are videos of how they "drive" the machine. https://www.jpl.nasa.gov/edu/learn/video/mars-in-a-minute-how-do-rovers-drive-on-mars/ It's not like an automobile, where everything is done in real time. The driver (or whatever he or she is called), plots the intended path of the rover on a computah screen. The rover then follows the dotted line. The idea is to avoid driving into any pitfalls or running over any obstacles. Stereo cameras give them a 3D view of the ground, which highlights these holes and rocks. That might explain why the lighting footprint in your photo shows lots of even light directly in front of the rover, while everything in the distance is barely illuminated. They need to see terrain details. From the video, the programming seems to done when Mars is dark. If they want to program the rover to go a longer distance, they need to see further forward. My guess(tm) is uniform lighting at a distance is what they're asking. You might want to modify your description under the photo on your web pile. "To re-light where the moon has been dark for three billion years, NASA's Rover can see 60% farther with Oculus optics." is not what's happening. I suggest "NASA's rover can see details and obstacles better with Oculus optics uniform lighting". or something like that. Lose the 60% or someone like me might ask you to explain why it's 60% further than what unspecified lighting system. As for power budget, you might ask NASA if they can synchronize their cameras to the PWM (pulse width modulation) dimming of your light. Instead of a 100% duty cycle, the light would be on full for perhaps 10% of the time, thus drawing 1/10th the power. Essentially, it would act like a strobe flash. If the camera frame rate is synchronized with the strobe, it sees full brightness for a short interval, and then nothing until the next pulse. On a real time TV, it would look awful, but on a system designed to capture single video frames, each image would be complete and look like it was photographed at full power. I hate to admit it, but this is my current idea of fun. Good luck. -- Jeff Liebermann 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558 |
#90
|
|||
|
|||
New B&M 100lux headlight.
Frank Krygowski writes:
On 12/10/2017 8:01 PM, John B. wrote: On Sun, 10 Dec 2017 12:30:16 +0100, Sepp Ruf wrote: John B. wrote: On Sat, 09 Dec 2017 19:41:16 -0600, Tim McNamara wrote: On Sat, 9 Dec 2017 10:34:13 +0100, Sepp Ruf wrote: [ ... ] black-tinted rear window; auto-dimming mirrors; manure-spreading trailer. I have an auto-dimming mirror in my new car. It verges on useless as it does not dim anywhere near enough. I'd rather have a manual mirror so that I don't have to reach up and point the damned mirror at the ceiling so that I'm not blinded by the vehicle behind me. My wife's small Honda has a manual tab on the bottom of the mirror. Flick it with your finger and the "guy behind"'s lights are focused on the ceiling :-) Still leaves the side mirrors vulnerable. Okay, so the trailer will need to be wide enough to block the view, but not as wide as to require the mounting of extra side mirrors on the Honda. I'm not sure about side mirrors. I have them on my pickup and my wife's little Honda has them But I don't remember ever having a problem with them. I sometimes have the problem, mostly on freeways. My wife even has that problem sitting in the passenger seat. It occurs mostly when truck drivers (pickups, box trucks, tractor-trailer rigs) have high headlights and insist on using high beams. It's more MFFY behavior. Sorry, I have to call Joerg on that one. Unless Ohio differs surprisingly from the rest of the US you are not seeing semi truck drivers with their brights on behind traffic. In my experience they are really quite careful about that -- the only group of drivers who sometimes behave that way are in the Usenet demographic, have deteriorated night vision, and probably shouldn't be driving at all. You are seeing low beams at roughly eye level, and I'll grant they are obnoxiously bright, much brighter than they used to be. Our current car has electrically adjustable side mirrors. If speed differentials are small enough that the problem persists for more than a few seconds, I point the mirrors downward until the offender is out of range. If they pass quickly, I just shade my side mirror with my hand. -- |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
The last headlight you will ever need | somebody[_2_] | Techniques | 115 | April 28th 14 02:12 AM |
Headlight | Tom $herman (-_-) | Techniques | 16 | August 17th 12 03:43 AM |
FS: L&M ARC HID headlight | Zach | Marketplace | 0 | February 13th 08 07:25 PM |
LED Headlight | HughMann | Australia | 12 | August 30th 06 11:51 AM |
Headlight | Bruni | Techniques | 8 | August 31st 03 06:27 PM |