Electronics, Lighting, & Gadgets - DIY Generator LED light

Bikeforums.net is a forum about nothing but bikes. Our community can help you find information about hard-to-find and localized information like bicycle tours, specialties like where in your area to have your recumbent bike serviced, or what are the best bicycle tires and seats for the activities you use your bike for.
arrasmithf
10-11-07, 01:35 PM
I love that homebuilt LED light that runs off of the bottle dynamo, but although cheap, seems a bit complex. Has anyone ever tried to use one of those hand crank LED flashlights with a bottle dynamo. I saw some nice ones at Target, when looking for the cheap-o dynamo. It looks like the lights are bright enough, and I think all the electronics would be there to run off a dynamo. I'm not sure the difference between a hand generator or a tire run generator. I think all you would have to do is pull the generator and steal the leads to the circuitry. Any thoughts?
ModoVincere
10-11-07, 03:22 PM
Could be wrong, but I think those hand held lights are 3 volt, not 6 volt so you would fry them with a bottle dynamo.
arrasmithf
10-11-07, 03:26 PM
even better, could you wire two in series?
ModoVincere
10-12-07, 08:46 AM
I just bought a light from Peter White Cycles...its a lumotec 3 watt light and its plenty bright and works just fine with my cheapo bottle dynamo.
LED would be really nice, but is more effort than I'm willing to do right now.
I love that homebuilt LED light that runs off of the bottle dynamo, but although cheap, seems a bit complex. Has anyone ever tried to use one of those hand crank LED flashlights with a bottle dynamo. I saw some nice ones at Target, when looking for the cheap-o dynamo. It looks like the lights are bright enough, and I think all the electronics would be there to run off a dynamo. I'm not sure the difference between a hand generator or a tire run generator. I think all you would have to do is pull the generator and steal the leads to the circuitry. Any thoughts?
I tried doing exactly what your talking about and it worked fine. The batteries act as both voltage and current regulators. All you need to do to modify it is remove the two wires from the flashlights internal dynamo and replace them with the bottle dynamo wires. The batteries in the flashlight should be 3 1.2-volt NiMH connected in series for a total of 3.6 volts. If you connect a voltmeter you will find the voltage will not exceed there rated recharging voltage. When the voltage of the dynamo reaches that point the batteries will load the dynamo preventing it from producing excessive voltage. A bicycle dynamo is a constant current device and will produce about 500mA at rated driving speed. The batteries will absorb excessive current not used by the LEDs. This is where you will run into problems using that "wind up" flashlight. Your bicycle dynamo is going to make much more current than the tiny dynamo in the flashlight so you can easily overcharge the 3 small batteries in the flashlight. It has 3 20mA LED's so your bicycle dynamo is going to be pushing 440mA recharging current into them. You will need to keep the dynamo disengaged until the batteries are discharged before engaging the dynamo for a minute or two to recharge them.
I did some experimenting and would up with the following circuit.
http://i143.photobucket.com/albums/r154/n4zou/all-standlight.jpg
My headlight uses a Luxeon 1-watt LED with matching Fraen elliptical beam pattern holder and lens. The 4.9-ohm resistor is found experimentally with the battery pack used so that LED draws 350mA with fully charged batteries. The taillight LED draws 30mA with a 100-ohm resistor. This allows 120mA-recharging current while the LED's are illuminating. The 30mA taillight was used so current would be 120mA as this is the suggested slow recharging current for NiMH batteries. Opening S1 allows 500mA current to recharge the batteries. That is the suggested rapid recharging current for NiMH batteries. You must still be careful not to overcharge the batteries.
Otherwise this dynamo and battery system keeps your head and taillights illuminating at all times unless the battery pack is completely discharged. If you manage to completely discharge the batteries you should open S1 at stops to prevent deep discharging of the batteries thus reducing their lifespan. When your moving again you can close S1 again supplying illumination and slow recharging. Optionally an LED driver can be used between the LED and batteries such as this example circuit.
http://i143.photobucket.com/albums/r154/n4zou/dynamo-battery-1.jpg
The LED driver chip will automatically disconnect the batteries when there discharged and you stop. The chip will automatically turn the LED back on when you start moving again and the dynamo is supplying enough power for the LED. I've never had a problem discharging my batteries. My time is spent mostly on an MUP with few stops so I must be careful not to overcharge them.
arrasmithf
10-12-07, 01:10 PM
sooo, like i said, a bit complex. I consider myself to be a fairly technical person and i'm getting lost in your wiring diagram. I was just trying to find out if feasible to re-purpose something and re-wire, as opposed to circuit boards and solder. I think what you did was great, much of us are just as willing, maybe just not as able to do this.
If you have the time, do you think you could eloborate with some photos of the work, unless you are trying to patent it or something, which is fine also.
I am going to try here in the next few weeks to do some experimenting myself, and maybe try to find a simple hack that can get me running, and i'll try to get it on www.instructables.com with some photos. I actually love the idea of fully charging the flashlight in a few minutes, and then enjoying the fruits of your labor with some quiet runtime.
How is the quality of light those wind-ups put out. Is it close to what the battery, or expensive dynamo lights put out?
Hand cranked dynamo flashlights typically use 14,000 MCD 5mm LEDs. That’s only good enough to make it a "be seen" light. A 1-watt Luxeon or equivalent will produce enough light to actually see where you're going. You're never really going to be happy trying to use a hand crank dynamo flashlight on your bicycle for serious night riding.
You don’t need any circuit boards and minimal soldering to DIY a system like mine. Here is a photo of my headlight.
http://i143.photobucket.com/albums/r154/n4zou/Hpim0312.jpg
It's built into a 1-inch PVC pipe coupler and cost less than a dollar. I used child proof prescription bottle caps with the center parts pried out to hold a clear cover over the lens on front and an aluminum cover cut with scissors from a soft drink can to cover the rear. I used an emitter already soldered to a star circuit board.
http://www.electronics-supply.com/admin/images/lxhl-mwec-l.jpg
My heat sink is cut to just fit inside the PVC pipe coupler from copper sheet metal scrap. I drilled holes for the lens holder mounting holes. I just used hot glue to hold the heat sink to the lens holder as it's easy to remove just incase I ever burned out my emitter. Here is a photo from the Luxeon showing a lens holder, star mounted emitter, and very large heat sink with many lens holders attached.
http://www.electronics-supply.com/admin/images/fhs-heb1-ll01-h-l.jpg
The heat sink for a 1-watt emitter only needs to be the same size as the lens holder but can be larger if you like.
Here is a photo of my test setup.
http://i143.photobucket.com/albums/r154/n4zou/Hpim0307.jpg
All parts from the circuit except the headlight and taillight can be seen in this photo. The plastic terminal strip has both resistors. The 4 diodes are soldered to it's own terminal strip and can be seen between the switch and plastic terminal strip. Again, no circuit board is used or required, just simple terminal strips with screws to hold the components and wires. I have updated the system for a cleaner look as well as moving the dynamo to the front fork and tire so I may engage/disengage it much easier. The battery pack is still mounted to the rack. You may obtain the emitter and matching lens from this link.
http://www.luxeonstar.com/index.php
Alternately you can find DIY flashlight parts and accessories here.
http://www.dealextreme.com/
New emitters are available that can be operated with variable current of between 350mA to 800mA. Some are just hooking these emitter boards directly to the diode matrix + and - connections with a 470mF capacitor to remove ripple. The dynamo can only supply 500mA so it can't burnout the emitter with too much current. Light output is reduced but it's still greater than a 1-watt LED. Check this forum for information.
http://www.candlepowerforums.com/vb/forumdisplay.php?f=86
At least that doesn't look very complicated ... LOL!
lectraplayer
10-13-07, 08:03 PM
I imagine that running a flashlight dynamo off of bicycle movement would make it difficult to pedal. Those things have an overdrive gear ratio, which gives it resistance to turning. I would suggest robbing a DC motor from some sort of dead device (maybe a tape player, drill, or whatever) and putting some sort of friction wheel on it to go on your rim. Motor RPM's, that way, will be way faster than you'd expect and you can easily get 12V out of a drill motor, and good current too. I would suggest putting a powerful light to go with it (at least a one-watter). More is better--I like light. I would also consider adding a small NiMH or NICD storage battery to keep the light on when you stop. A blocking diode on the output of the dynamo would keep the battery from being discharged through the dynamo. Current limiting shouldn't be a problem either.
Lt.Gustl
10-13-07, 08:39 PM
Yikes! n4zou try to put that stuff in at least a hammond box... or just go to the electrical section of your home depot/lowes and get some kind of enclosure... if you're going to get good service life from the whole circuit encase it in epoxy, but leave the switch serviceable. Otherwise great job! Thanks for sharing
bikingbrit
10-14-07, 08:25 AM
n4zou,
Every time I see your schematic there is something which bothers me: The dynamo is shown with a floating i.e. not grounded output. Most bottle dynamos have only one actual terminal and use the case ground to complete the circuit. The hub dynamos like the Sturmey Archer had a floating output with two terminals and a bridge rectifier/battery was available as an accessory to keep the lights on while stopped.
In your circuit though the bridge rectifier will not work as expected. Ground the bottom of the dynamo winding and the negative terminal of the bridge rectifier in your schematic. Now it is obvious that D2 is shorted and will not operate. D3 will conduct when the dynamo output is positive, and produce a useful output. D4 will always be reverse biased and thus inactive. D2 will be loading/short circuiting the dynamo on negative half cycles but will not produce a useful output. I expect this would increase the physical drag the dynamo produces.
Basically it looks like the circuit would function but will have less useful output and more drag than expected. You might wish to try insulating either the AC or DC side of the circuit from ground and see if it works any better.
n4zou,
Every time I see your schematic there is something which bothers me: The dynamo is shown with a floating i.e. not grounded output. Most bottle dynamos have only one actual terminal and use the case ground to complete the circuit. The hub dynamos like the Sturmey Archer had a floating output with two terminals and a bridge rectifier/battery was available as an accessory to keep the lights on while stopped.
In your circuit though the bridge rectifier will not work as expected. Ground the bottom of the dynamo winding and the negative terminal of the bridge rectifier in your schematic. Now it is obvious that D2 is shorted and will not operate. D3 will conduct when the dynamo output is positive, and produce a useful output. D4 will always be reverse biased and thus inactive. D2 will be loading/short circuiting the dynamo on negative half cycles but will not produce a useful output. I expect this would increase the physical drag the dynamo produces.
Basically it looks like the circuit would function but will have less useful output and more drag than expected. You might wish to try insulating either the AC or DC side of the circuit from ground and see if it works any better.
You are correct about bottle dynamos having only one terminal. The second terminal shown is taken from the bottle dynamos ground, which becomes a common AC ground throughout the entire frame of the bicycle. If you attempt the ground the negative side of the rectified DC circuit to the bicycles frame you will short it out and nothing will work. The DC side of the circuit must float and not come in electrical contact with the frame or any other parts of the AC side of the circuit. Some hub dynamos have two isolated terminals so AC current is never connected to the bicycle frame. All modern hub dynamos are made this way so the bicycles frame can be used as a DC ground. The circuit I show can be used for Bottle, BB, and Hub dynamos.
As for the diodes and their connection, it is a standard bridge rectifier circuit and all 4 diodes may be purchased in the form of a single electronic component such as this.
http://rsk.imageg.net/graphics/product_images/pRS1C-2160490w345.jpg
The reason for using 4 individual diodes is the use of high efficiency 1N5818 diodes, which drop .8 volts instead of silicon rectifier diodes, used in silicon bridge rectifiers drop 1.2 volts across each diode. Using a bridge rectifier configuration will multiply the rectified DC voltage by 1.4 so you actually gain a small voltage increase even with the loss of .8 volts across each diode. The diodes you think are connected backwards are actually turning the negative going AC voltage into positive (ripple) DC voltage; otherwise it would just be blocked and lost as heat in the diode.
Drag is the same from bicycle dynamos no matter what the load. All bicycle dynamos are constant current devices and will push about 500mA of current into the circuit so when you engage a Bottle or BB dynamo the added drag is going to be constant. Hub dynamos are a little different with very little drag with the switch open but as soon as you connect a load like a filament type bulb or an LED you will feel an increase in drag and that drag will be the same no matter how much the current the circuit was designed for. The dynamo will push that 500mA across the components if no current sink like a battery is there to absorb any excess current required by the components. The original design mandated by the German government in the 1950's required a 3-watt filament bulb system consisting of a 6 Volt-2.4 watt headlight bulb and a 6 Volt-.6 watt taillight bulb. Consequently the dynamo magnets and coil are designed so that exact load will regulate the output voltage and current of the dynamo for 6 volts and 3 watts.
That’s the standard we are all living with to this day. Count yourself lucky here in the USA where there are no mandated bicycle lighting systems so you can pretty much do as you please.
Yikes! n4zou try to put that stuff in at least a hammond box... or just go to the electrical section of your home depot/lowes and get some kind of enclosure... if you're going to get good service life from the whole circuit encase it in epoxy, but leave the switch serviceable. Otherwise great job! Thanks for sharing
I know it's ugly! I was testing and wanted to be able to switch out diodes and resistors as I experimented with different battery and LED configurations. The dynamo was mounted to a seat stay so I could crank the rear wheel with the pedals thus driving the dynamo by the rear wheel on that stand while watching my current and voltmeters. My dynamo is now mounted on the fork where I can get at it while riding. The switch is a water proof type mounted on the handlebar, and the batteries are hidden in a small blowout bag under the seat. The resistors are spliced into the wires leading to the headlight and taillight with heat shrink tubing covering and sealing them. The diodes now reside in epoxy permanently sealing them. There's not much to see of it now that I have finished testing and "locked in" the design.
I imagine that running a flashlight dynamo off of bicycle movement would make it difficult to pedal. Those things have an overdrive gear ratio, which gives it resistance to turning.
Modern Bottle and BB type bicycle dynamos don't induce very much drag like the old ones from my youth. After a few minutes you forget it's engaged and away you go happily pedaling along while illuminating the path ahead. The old vintage dynamos were a different story! They induced lots of drag making you feel like you were pedaling up hill all the time even when going down a mild grade! Hub dynamos induce much less drag and don’t rub the tire or make noise. They cost considerably more money however! I can't find one that costs less than $100 US dollars here in the states. I've found them much less expensive in Europe bike shops but after you pay shipping to get it across the Atlantic Ocean your cheap imported hub dynamo is going to cost as much as a quality dynamo purchased from a dealer here in the states. I just gave up on obtaining one for my own use and will keep using my cheap bottle dynamo. It's simple and works.
I would suggest robbing a DC motor from some sort of dead device (maybe a tape player, drill, or whatever) and putting some sort of friction wheel on it to go on your rim.
This is more complicated than you think. Been there and done that! Use a stepper motor as a substitute bottle dynamo if you're going to try and go that route. Here is a site that goes into detail on selecting and building power supplies from stepper motors.
http://www.reuk.co.uk/Stepper-Motor-Basics.htm
When your finished you'll find a real bicycle dynamo was much less expensive and much easier to mount and use.[/QUOTE]
If you want more power than 3-watts download this PDF file so you know what you're getting into. It makes very interesting reading so go ahead and save a copy to your hard drive.
http://www.minesactioncanada.org/competition/2002_2003/PedalGeneratorBatteryCharger_1stPrize.pdf
Motor RPM's, that way, will be way faster than you'd expect and you can easily get 12V out of a drill motor, and good current too. I would suggest putting a powerful light to go with it (at least a one-watter). More is better--I like light.
My headlight is using a Luxeon 1-watt led fitted with a Fraen elliptical lens. That LED draws 350mA and my taillight LED draws 30mA-leaving 120mA to recharge the NiMH batteries. My lights do not go out when I stop or when climbing steep hills. My problem is a tendency to overcharge by batteries! I try to remember to disengage my dynamo for 15 minutes of every hour while keeping the LED's burning so I don’t overcharge the batteries.
I would also consider adding a small NiMH or NICD storage battery to keep the light on when you stop. A blocking diode on the output of the dynamo would keep the battery from being discharged through the dynamo. Current limiting shouldn't be a problem either.
The bridge rectifier consisting of 4 1N5818 diodes block DC voltage from the dynamo. There is a tiny amount of leakage however so if your going to be storing the bike for several days or more you should disconnect the batteries from everything. As for current, a dynamo of any type using magnets is going to force current into the circuit, which is 500mA for all bicycle dynamos. This requires a circuit design that will draw almost the exact amount of current as that supplied by the dynamo. Adding batteries gives the excess current not consumed by the LED's a place to go, they become current storage devices. Otherwise excessive current flowing across them would burn up the two LED's. The solution would be to add more LED's until the total number of LED's equals 500mA so the current is balanced. I just like the option of using batteries as they keep the LED's illuminating constantly.
ModoVincere
10-15-07, 08:20 AM
Silly question, but if I wanted more light than the 1 watt luxeon puts out, how hard would it be to rig up a system where the batteries supply the 700ma(?) 2 lux led's would draw, or perhaps, the additional 200ma if the other 500ma is supplied by the dynamo?.
Seems like a set up like this could provide a nice illumination pattern, and when your not using the lights, then you can charge the batteries.
Silly question, but if I wanted more light than the 1 watt luxeon puts out, how hard would it be to rig up a system where the batteries supply the 700ma(?) 2 lux led's would draw, or perhaps, the additional 200ma if the other 500ma is supplied by the dynamo?.
Seems like a set up like this could provide a nice illumination pattern, and when your not using the lights, then you can charge the batteries.
You'll need a high quality hub dynamo like a Shimano DH-3N70 or Schmidt's Nabendynamo to do that. These hubs are available here.
http://www.peterwhitecycles.com/lightingsystems.htm
http://www.peterwhitecycles.com/images/products/Lights/DH-3N70.jpg
You will also need to build a circuit board as well like this one.
http://www.pilom.com/BicycleElectronics/LM2675design.jpg
This link will take you to this circuit and other electronic circuits as well.
http://www.pilom.com/BicycleElectronics/DynamoCircuits.htm#High
vBulletin® v3.7.3, Copyright ©2000-2009, Jelsoft Enterprises Ltd.