Mountain Bike headlight with LEDs

Empowertorch -- Trn on flw ctrl!
wink.gif
 
Originally posted by Entropy:
5W whites ARE out, but VERY difficult to come by. Your cost estimate for them is way off - The ElektroLumens group buy came to $22 each. 5W coloreds are more readily available - I've seen quite a few people using cyans, which are $23 in single quantities from Future-Active. (Not sure about availability of the cyans, but it seems better than the whites).
<font size="2" face="Verdana, Arial">What are the chances of us getting 5W whites from Future? I know they're listed on their website, but described as "Call for availability" and I think I saw somewhere that the group buy pple aren't getting theirs until March...

And they're listed on the page as costing over $40 last I saw
shocked.gif
Coloured ones are cheaper though.
 
I just saw some news on the group buy that I think they're getting them near the end of this month/beginning of next month.

Could've sworn the group buy was for $22 each... And I'm not surprised that in low quantities the whites are more expensive. White 1Ws are $11-13 or so from F-A, while coloreds are under $9.

I think the chances of getting 5W whites outside of group buys is highly unlikely - They're RARE.
 
I was quite a bit off, total cost with a lens is about $26 a pop, but even so if you want to rival 40 watts your going to put out near $200 in LED's alone. Add to that circitry, a good housing, proper beam aiming, heat sink, switches, a serious battery and smart charger and you will spend more than the $300 you can get a HID unit for. Not that HID is a perfect system cause it's not by far. However right now I think it's by far the best bang for the $.

Along with the fact that if I remember right they are in the 3/4" dia range so putting many of them in one area is going to have two problems first in size of the housing just to hold all the LED's and second in that if you are going to try to overdrive them at all a serious heat sink is going to be required. Along with the fact that if you put 4 or 8 of them together in close quaters more heat sinking will be needed. Along with the fact it seems the current 5w offerings are only a 1000 hr life LED, which is worse than most good quality halogen MR11/MR16 bulbs even when overvolted 10%. Most HID bulbs are in the 2000 hour range. Though it sounds like in late 2003 a longer life version will be released.

Show me a 5 watt LED in a small flashlight in the size of a e2 that gets me excited, show me a cluster of 5 watts trying to beat a HID bike light in the same size, output, runtime, and cost and I'll be very impressed but I don't see it happening in the next year or two.

I do think they could make a very nice headlamp, a single 5 watt or dual 5watt with good lenses, it would be small, compact, and bright, and durable. I just don't think for awhile they can rival HID bike lights in size and performance for the same cost.

If all a person wants on their bike is that 5-10 watts for bike path and slow trail riding they will probably work nicely, but if you really want some light I don't think a cluster of 8 of them is very practical or cost effective.

Todd
 
I recently built a pretty cool light with a 100wh (20 18650's arranged in 5 parallel packs of 4 serial charged with a 16.1 volt 2.2a laptop charger) lithium ion battery pack and 55 watts of halogen bulbs. I didn't understand why it was so stinking bright until I read the post above about excess power.

I was tweaking with the light last nght on a walk and it only ran for about one hour. I was dismayed to say the least as I was expecting an hour and a half. Then I found the calculations about light and the way I was over-driving the lamps etc. I honestly could not believe how bright it was. Anyway it turns out that the total draw of my 55 watt of combined bulbs was really about 85 watts, and the equivalent light output was about 152 watts (about 2x an HID headlight). We were walking the golf course around the neighborhood, and if you stood at one tee you could see the green..... 300 yards away........



Is there some sort of self help group for people like me?



Chad
 
Chad,

I think that you are walking on a wire here, above the pits of flashaholism. You have developed a love for bright light and the feeling of having such a light that you can hold and carry everywhere.

At the early stages flashaholism will only cause you to have the urge to own and caress such a powerful device, but some such as yourself will attempt to construct one of their own.

Once you fall into the pits of flashaholism, you will not be able to climb out. Your web browser will have its startup page set at www.candlepowerforums.com, you will have McGizmo, dat2zip, Lambda, Brightguy, etc, etc's site on your favorites list. Your wallet will mysteriously become thinner and thinner. You will begin to sleep during the day in order to wake up in the deep dark hours of the night. The light switches in your house will have spider webs on them.

It is imperative that you try to stay away from here before it is too late. I urge you to take the first step to the cure by sending me that monster you have created.

grin.gif


-YC
 
CPF: Feeding the addiction since ???

I think I'm pushing $60-70 put into my MiniMag now... It's nothing like the original light. All of its strengths, none of the weaknesses...
smile.gif


Chad: If you see the words McLux: CLOSE YOUR BROWSER WINDOW.

Trust me - The McLux is such a powerful light that it can instantly burn a hole in your pocket just by reading about it.
smile.gif
 
Originally posted by Jonathan:
Originally posted by Soul-tez:
[qb]If you really want to operate a halogen lamp properly, take a look at Willie Hunt's Light Bulb Voltage Regulator

Essentially the LVR is a PWM dimmer that measures the input voltage and adjusts the dimming so that the lamp sees the exact desired running voltage. Then you start with a battery that is too high a voltage. The LVR regulates the voltage down to what the light needs, and as the battery voltage drops the the LVR dims less and less, resulting in constant output.
<font size="2" face="Verdana, Arial">Don't Willie's LVRs have a setup as well?

-john
 
Originally posted by John N:
</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by Jonathan:
The LVR regulates the voltage down to what the light needs, and as the battery voltage drops the the LVR dims less and less, resulting in constant output.
<font size="2" face="Verdana, Arial">Don't Willie's LVRs have a setup as well?</font><hr /></blockquote><font size="2" face="Verdana, Arial">Nope. The output stage is a simple PWM chopper with no filtering at all. The output is _not_ DC, but instead a duty cycle modulated square wave that goes from zero to the full supply voltage, with an 'on' period selected so that the RMS voltage is the desired output. The output is 'smoothed' by the thermal time constant of the filament.

With no filtering or energy storage components, the LVR cannot do step-up.

-Jon
 
Originally posted by Jonathan:
Nope. The output stage is a simple PWM chopper with no filtering at all. The output is _not_ DC, but instead a duty cycle modulated square wave that goes from zero to the full supply voltage, with an 'on' period selected so that the RMS voltage is the desired output. The output is 'smoothed' by the thermal time constant of the filament.

With no filtering or energy storage components, the LVR cannot do step-up.

-Jon
<font size="2" face="Verdana, Arial">I know that Jon knows this but others reading this may not. Since this thread appears in the LED forum it should be pointed out that a PWM voltage regulator without output filtering is not suitable for driving LEDs unless the source voltage is already very close to what the LED needs. In this case, a linear current regulator would produce light more efficiently and be easier to implement.
 
Originally posted by Jonathan:
</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by John N:
</font><blockquote><font size="1" face="Verdana, Arial">quote:</font><hr /><font size="2" face="Verdana, Arial">Originally posted by Jonathan:
The LVR regulates the voltage down to what the light needs, and as the battery voltage drops the the LVR dims less and less, resulting in constant output.
<font size="2" face="Verdana, Arial">Don't Willie's LVRs have a setup as well?</font><hr /></blockquote><font size="2" face="Verdana, Arial">Nope. The output stage is a simple PWM chopper with no filtering at all. The output is _not_ DC, but instead a duty cycle modulated square wave that goes from zero to the full supply voltage, with an 'on' period selected so that the RMS voltage is the desired output. The output is 'smoothed' by the thermal time constant of the filament.

With no filtering or energy storage components, the LVR cannot do step-up.

-Jon
</font><hr /></blockquote><font size="2" face="Verdana, Arial">Sorry, I know this is getting OT, but I'm confused. So this means I have to have a power source higher than the lamp in order to maintain something that looks like constant output?

So if I want to run a 6V lamp, I have to have greater than 6V battery supply so as the battery runs down I can still supply 6V?

-john
 
Since the "on" voltage is not filtered and you need to use higher voltage power supply, I would assume that mean the bulb gets intermittant bursts of voltage much higher than the bulb is rated for. Isn't this hard on the bulb?

Also, doesn't this mean there is some maximum input voltage that is bulb specific, rather than LVR specific?

Thanks,

-john
 
Originally posted by John N:
Since the "on" voltage is not filtered and you need to use higher voltage power supply, I would assume that mean the bulb gets intermittant bursts of voltage much higher than the bulb is rated for. Isn't this hard on the bulb?
-john
<font size="2" face="Verdana, Arial">Not if the "on time" is short relative to the time it would take the filament to heat up by an appreciable amount.
Originally posted by John N:

Also, doesn't this mean there is some maximum input voltage that is bulb specific, rather than LVR specific?
-john
<font size="2" face="Verdana, Arial">As a practical matter, No, provided that the pulses can be made short enough to meet the criterion above.
Somewhat off topic, this discussion brings to mind a couple of seldom heard words that you can try out at your next cocktail party:
Incandescence [note spelling]: the time for an incandescent lamp to go from 10% to 90% brightness when power is applied.
Nigrescence: the time for an incandescent lamp to go from 90% to 10% brightness when power is removed.
For a given lamp, these times are usually not equal. Nigrescence tends to be shorter.
 
Yes, because the on voltage is not filtered, the bulb is getting bursts of voltage that is considerably higher than its rating. However, as long as the PWM frequency is high enough, this is _not_ a problem.

The way a tungsten filament lamp works is as follows: an electric current flows through the filament. The filament is a resistor, and the resistance to the current flow causes some electrical energy to be converted into heat. The filament heats up. As the filament heats up, more and more energy is carried away by conduction of heat and by radiant energy (visible light, (mostly) IR, and a bit of UV). Eventually a point is reached where the total energy carried away is exactly equal to the energy being converted into heat, and the filament will reach thermal equilibrium at a given temperature.

If you increase the applied voltage, then more current will flow, more heat will be generated, and the filament will have to heat up in order to dissipate the increased heat. If the filament heats up too much, then it will melt and break.

In the case of the PWM system, the instantaneous applied voltage is substantially greater than what the filament could stand continuously. During the 'on' period, the filament temperature starts going up, because more heat is being developed in the filament than is being dissipated. Then during the 'off' period,t he filament temperature starts going down. The net result is that the filament temperature is not _constant_, but is rather a 'sawtooth'. The higher the frequency, the smaller the teeth of the sawtooth, and the smaller the temperature variation.

In my experience, small halogen lamps (such as used on bicycles) are quite happy with 400Hz PWM (easy to do with a PIC) even when the supply voltage is twice the rated lamp voltage. In this case, the power dissipated in the filament during the 'on' periods is _4_ times the normal rated power. I currently use 6V Vistalite head lamps with a 12V NiMH battery and a PWM regulator. No problems.

-Jon
 
One other thing about using LEDs instead of other types of lamps, its not the power consumed by the device thats important, its how much light is being put out. A larger wattage device with lower luminous efficiency can be dimmer than a lower wattage device with higher luminous efficiency (Think flourecent tubes vs incandescent bulbs).

Theres a chart on this page that lists the luminous efficiency of various light sources. LEDs I'd guess are just a little less efficient than halogens at the moment.

That said I believe somebody here posted link to a Japanese company a while back, they do headlamps for bikes and had one LED based solution.
 
Originally posted by Lucien:
One other thing about using LEDs instead of other types of lamps, its not the power consumed by the device thats important, its how much light is being put out. A larger wattage device with lower luminous efficiency can be dimmer than a lower wattage device with higher luminous efficiency (Think flourecent tubes vs incandescent bulbs).

Theres a chart on this page that lists the luminous efficiency of various light sources. LEDs I'd guess are just a little less efficient than halogens at the moment.

That said I believe somebody here posted link to a Japanese company a while back, they do headlamps for bikes and had one LED based solution.
<font size="2" face="Verdana, Arial">Don't forget the one thing that has given LEDs a big advantage in flashlights - As the battery runs down, they merely dim, rather than becoming orange/yellow. (And at this point, their efficiency is less since a larger portion of their emissions are IR).

LEDs might be slightly less efficient than halogens, but in terms of usable battery life, their life is much longer because of the fact that they continue working when incandescents would be nothing more than a heater. (Esp. LED lights with regulators)
 
Top