Bicycle dynamo LED - adding a "super cap" to keep it lighted ( after stopping )

Xenagor

New Member
Hey there !

Seems like a very nice forum !

I never posted anything about electronics, I usually figured my things out even if I rarely do projects like this.

Basically, it's a very simple project, but I thought the advice of people who know a lot in electronics in a place like here would be wise - before ordering all my pieces !

What I wanna do is very simple. It's to basically replace my normal front bulb light on my bicycle by a prebuild LED light ( something like this ):

DealExtreme: $6.90 9-LED Bike Head Light (3xAA) I got some pretty good knowledge in electronics, so feel free to elaborate. First here's a link of the circuit I wanna build for my dynamo LED light for my bicycle. It seems like a pretty good base to start, and also to optimize it - I'll probably start with something like this website shows : Dynamo LED Light Systems for Bicycles (electronic circuits) There seems to be interesting options in the evolution of the circuit shown, but I was thinking about starting with Circuit 2. Here it is for quick view of it : I'll most probably be adding some stuff to it, but let's start with this one for my main question .... The role of C1 ( 1000uF - 10000uF ) in this one is supposed to be only for making sure the light is constant according to the author on the website. I was wondering however, if I replace that cap with a super cap, something like 1 Farad for example - would the cap act as a power buffer also ? Like - without modifying the circuit at all ? This is for, when I stop completly, the light can stay on for couple of minutes ( that is after charging it a little while of course ) I found some cap for sell on ebay ... I guess I can link it ... : 1 Farad GoldCap, 5.5 Volts Memory Backup Gold Capacitor - eBay (item 110361406586 end time May-07-09 12:45:21 PDT) Like you may have seen, those are 5.5V max ... my bottle dynamo on my bike is a 6V generator. I thought it would still be okay, since with that circuit there's at least 1 diode in the way, dropping the voltage of about 0.7V - making it 5.3V or less - am I right ? So this wouldn't be a problem ? Or does the fact that the dynamo is going faster on high speeds will get the voltage much higher and therefor needing more resistant components ? ( I didn't check the power output of it with a meter I gotta admit ) I was thinking of using a 3 X AA battery flashlight anyway - which is 4.5V if I'm not mistaken ( like linked above ). Also - out of curiosity - if I just stack the 1 Farad cap, right in parallel with C1, would this increase the duration of the "power buffer" - therefor increasing the length time of lighting after stopping ? Or something bit more special needs to be done ? So ! How does this sound til now, will this work ? Or do I need to do something special to get a decent amount of time of charge ? I'll keep it at is for now I don't wanna make a to big post to start - wishing to increase the number of people posting advice lol If you need any additional info to comment just ask, I'll be glad to post them, even though everything I could thought of useful. Thanks in advanced ! I'll keep an update of what I come up with, whatever happens. Hey there ! Seems like a very nice forum ! I never posted anything about electronics, I usually figured my things out even if I rarely do projects like this. Basically, it's a very simple project, but I thought the advice of people who know a lot in electronics in a place like here would be wise - before ordering all my pieces ! What I wanna do is very simple. It's to basically replace my normal front bulb light on my bicycle by a prebuild LED light ( something like this ): DealExtreme:$6.90 9-LED Bike Head Light (3xAA)

I got some pretty good knowledge in electronics, so feel free to elaborate.

First here's a link of the circuit I wanna build for my dynamo LED light for my bicycle. It seems like a pretty good base to start, and also to optimize it - I'll probably start with something like this website shows :

Dynamo LED Light Systems for Bicycles (electronic circuits)

There seems to be interesting options in the evolution of the circuit shown, but I was thinking about starting with Circuit 2. Here it is for quick view of it :

I'll most probably be adding some stuff to it, but let's start with this one for my main question ....

The role of C1 ( 1000uF - 10000uF ) in this one is supposed to be only for making sure the light is constant according to the author on the website.

I was wondering however, if I replace that cap with a super cap, something like 1 Farad for example - would the cap act as a power buffer also ? Like - without modifying the circuit at all ?

This is for, when I stop completly, the light can stay on for couple of minutes ( that is after charging it a little while of course )

I found some cap for sell on ebay ... I guess I can link it ... :

1 Farad GoldCap, 5.5 Volts Memory Backup Gold Capacitor - eBay (item 110361406586 end time May-07-09 12:45:21 PDT)

Like you may have seen, those are 5.5V max ... my bottle dynamo on my bike is a 6V generator.

I thought it would still be okay, since with that circuit there's at least 1 diode in the way, dropping the voltage of about 0.7V - making it 5.3V or less - am I right ? So this wouldn't be a problem ?

Or does the fact that the dynamo is going faster on high speeds will get the voltage much higher and therefor needing more resistant components ? ( I didn't check the power output of it with a meter I gotta admit )

I was thinking of using a 3 X AA battery flashlight anyway - which is 4.5V if I'm not mistaken ( like linked above ).

Also - out of curiosity - if I just stack the 1 Farad cap, right in parallel with C1, would this increase the duration of the "power buffer" - therefor increasing the length time of lighting after stopping ? Or something bit more special needs to be done ?
So !

How does this sound til now, will this work ? Or do I need to do something special to get a decent amount of time of charge ?

I'll keep it at is for now I don't wanna make a to big post to start - wishing to increase the number of people posting advice lol

If you need any additional info to comment just ask, I'll be glad to post them, even though everything I could thought of useful.

Thanks in advanced !
I'll keep an update of what I come up with, whatever happens.

Hey there !

Seems like a very nice forum !

I never posted anything about electronics, I usually figured my things out even if I rarely do projects like this.

Basically, it's a very simple project, but I thought the advice of people who know a lot in electronics in a place like here would be wise - before ordering all my pieces !

What I wanna do is very simple. It's to basically replace my normal front bulb light on my bicycle by a prebuild LED light ( something like this ):

DealExtreme: $6.90 9-LED Bike Head Light (3xAA) I got some pretty good knowledge in electronics, so feel free to elaborate. First here's a link of the circuit I wanna build for my dynamo LED light for my bicycle. It seems like a pretty good base to start, and also to optimize it - I'll probably start with something like this website shows : Dynamo LED Light Systems for Bicycles (electronic circuits) There seems to be interesting options in the evolution of the circuit shown, but I was thinking about starting with Circuit 2. Here it is for quick view of it : I'll most probably be adding some stuff to it, but let's start with this one for my main question .... The role of C1 ( 1000uF - 10000uF ) in this one is supposed to be only for making sure the light is constant according to the author on the website. I was wondering however, if I replace that cap with a super cap, something like 1 Farad for example - would the cap act as a power buffer also ? Like - without modifying the circuit at all ? This is for, when I stop completly, the light can stay on for couple of minutes ( that is after charging it a little while of course ) I found some cap for sell on ebay ... I guess I can link it ... : 1 Farad GoldCap, 5.5 Volts Memory Backup Gold Capacitor - eBay (item 110361406586 end time May-07-09 12:45:21 PDT) Like you may have seen, those are 5.5V max ... my bottle dynamo on my bike is a 6V generator. I thought it would still be okay, since with that circuit there's at least 1 diode in the way, dropping the voltage of about 0.7V - making it 5.3V or less - am I right ? So this wouldn't be a problem ? Or does the fact that the dynamo is going faster on high speeds will get the voltage much higher and therefor needing more resistant components ? ( I didn't check the power output of it with a meter I gotta admit ) I was thinking of using a 3 X AA battery flashlight anyway - which is 4.5V if I'm not mistaken ( like linked above ). Also - out of curiosity - if I just stack the 1 Farad cap, right in parallel with C1, would this increase the duration of the "power buffer" - therefor increasing the length time of lighting after stopping ? Or something bit more special needs to be done ? So ! How does this sound til now, will this work ? Or do I need to do something special to get a decent amount of time of charge ? I'll keep it at is for now I don't wanna make a to big post to start - wishing to increase the number of people posting advice lol If you need any additional info to comment just ask, I'll be glad to post them, even though everything I could thought of useful. Thanks in advanced ! I'll keep an update of what I come up with, whatever happens. RODALCO Well-Known Member You need a resistor in series with the LED. even at 5.3 Volts it will fry a 3 Volt white LED. If you bike fast the voltage will easily go up to 7 or 8 Volts. Also with a LED the loading on the dynamo is less and may increase the voltage even further. Try to keep you max LED current at about 20 mA for standard 5 mmØ LED's RODALCO Well-Known Member By re-reading your post again I see that the output voltage of your bridge rectifier may be too high for your 1 Farad supercap. These are rated at 5.5 Volts and that is the maximum voltage. Any more will damage the cap. You need to measure the maximum voltage or, at least fit 1 or 2, 1N4001 diodes in series with the positive output too keep the bridge rectifier voltage below 5.5 Volts. Last edited: Monkey New Member I did it long time ago, using a lead accumulator instead of the supercap. audioguru Well-Known Member Most Helpful Member The original circuit uses the wimpy generator to limit the current to 0.5A for a high power LED that limits the voltage to about 5V. Ordinary LEDs will blow up with 500mA and if a resistor is used to limit the current then the voltage from the generator will go very high when the bike goes fast. A supercap might power an ordinary dim LED for 10 seconds with it dimming. Xenagor New Member Thanks for the replies. I know that a LED would need a restriction to the voltage if I was building that part, well I had an idea it would. However by choosing a prebuild flashlight I thought it would solve that problem ? That flashlight I linked is using 9 LEDs, I have no idea how they are plugged but I highly doubt they are all in parallel ? ( giving them the max voltage on each ) - so it should be alright if I apply the same voltage as the battery would right ? ( 4.5V ? ) Correct me if I'm wrong at any point. Also I'm gonna check the voltage output tonight, if and how much it increases when I'm going fast. I was wondering if it would be instead the max current capacity that would increase with a faster speed ? I never experimented too much with such dynamos.. I'll check tonight what it says - writing this from my job right now. If the voltage go too high, what would be the easiest way to diminish it so it stays always at 4.5V or so ? I guess there are some limiters ? ( I'd like to avoid using a power source chips for heatsink problem and all .. ) Well ! I gotta go now. I'll check it some more tonight. BTW - sorry about my first post LOL I have no idea why it is writed 3 times in a row audioguru Well-Known Member Most Helpful Member That flashlight I linked is using 9 LEDs, I have no idea how they are plugged but I highly doubt they are all in parallel ? ( giving them the max voltage on each ) - so it should be alright if I apply the same voltage as the battery would right ? ( 4.5V ? ) 3AA cells are 4.5V only when they are new. Their voltage quickly drops to 3.6V. The white LEDs might be 3.0V and each one probably has its own current-limiting resistor. But we don't know their voltage and we don't know how they are connected. Also I'm gonna check the voltage output tonight, if and how much it increases when I'm going fast. I was wondering if it would be instead the max current capacity that would increase with a faster speed ? I think both the voltage and current from the generator increase when you go fast. You must use a suitable load to measure the voltage and current. If the voltage go too high, what would be the easiest way to diminish it so it stays always at 4.5V or so ? I guess there are some limiters ? ( I'd like to avoid using a power source chips for heatsink problem and all .. ) Use a low-dropout 5V linear voltage regulator IC. A diode in series with the output will reduce the voltage to 4.2V. The regulator might need a small heatsink. Xenagor New Member Hmm ... WT* ? I'm trying to plug my multimeter on the ground of the dynamo, and the +6V, both exactly where my wires are connect, and I get absolutely no voltage ? I mean, no voltage at all, even when my wheel is turning and the light is well visible ... ? My meter works, since I just tried it on a 9V battery, and it does show exactly 8.96V .. so it does seem to work Seriously, am I missing something ? How can I have 0 Volt difference between the ground ON the dynamo and the +6V ALSO ON the dynamo - while it's running at good speed and my lights are lighted ?? I don't get this one .. ! I tried it like 5 times, always making sure there's contact each time of course. Boncuk New Member If you take a close look at a factory built bicycle headlight (OSRAM, PHILIPS) you'll be amazed by the number of components. Under no circumstances you must exceed the maximum allowable LED forward current. Supercaps can be considered bombs if charged beyond their nominal voltage. WIMA manufactures supercaps with the capacitance of 500, 700 and 1,200F. You could chose one of those and make sure that they won't be overcharged (voltage) using a regulator circuit. Also you must take care that the cap won't discharge into the dynamo when it's not moving. Refer to the schematic when planning your headlight circuit. For D1 use a Schottky-Diode capable of 1A current flow. It prevents discharging of the supercap into the voltage regulator. R1 limits the charging current into the supercap. When it's uncharged it is like a dead short. As charging current decreases the voltage across the cap will reach 5V. Boncuk Attachments • 9.3 KB Views: 4,473 Last edited: Boncuk New Member Hmm ... WT* ? I'm trying to plug my multimeter on the ground of the dynamo, and the +6V, both exactly where my wires are connect, and I get absolutely no voltage ? I mean, no voltage at all, even when my wheel is turning and the light is well visible ... ? My meter works, since I just tried it on a 9V battery, and it does show exactly 8.96V .. so it does seem to work Seriously, am I missing something ? How can I have 0 Volt difference between the ground ON the dynamo and the +6V ALSO ON the dynamo - while it's running at good speed and my lights are lighted ?? I don't get this one .. ! I tried it like 5 times, always making sure there's contact each time of course. DVMs are designed to measure AC and DC. The AC-measurement will only be correct if measured at the frequency they are designed for (true RMS measurement). The dynamo has a much higher frequency and you can only read approximate voltage values. Don't forget to switch your meter to AC. Boncuk Xenagor New Member Oh wow ... I didn't even know that dynamos were an AC source of power heh - it shows how much I know about them ugh ... I do read from 0 to 6V AC on my meter, and even more probably if I would have give it enough indeed " I thought the bridge was to avoid problems, IF ever the wheel was turning the other side. I should have remember a diode bridge kinda means AC power errr It's a bit odd though that they make it so that the "-" ( what I thought to be a ground ) is connected to the frame through the mounting device .. I'll have to remember to isolate the ground when I build that circuit, so it doesn't touch the frame at all. Thanks for the nice tip and schema seriously, it is very helpful. Still got some thing I'm wondering though .. ( obviously ! ) Not like I want you guys figure it out for me - but hey that's what a forum for uh ! . Well a few things ... but I'll start by this. I didn't know they were making SO HUGE cap ! 1200F sounds so crazy, but I'd like to get the smallest cap possible( physically ) - without compromising a goal of something like at least 4-5 minutes of light after stopping ( when fully charge ). If possible even a 30 minutes would be pretty damn nice - allowing a rest on my dynamo for a while. A supercap might power an ordinary dim LED for 10 seconds with it dimming. Really ? In fact it seems your right - I just found that website and gave it a shot. Super Capacitor Calculator - Maxim I used those values. Max Vcap = 5.5V Min Vcap = 3V Vbat = 5V Cap = 600F IbatTyp = 675000000nA ( 675mA - which should be around that for a 3W 3 X AA battery flashlight, if I'm not mistaken ) That seems pretty good, it gave me around 45 minutes of light, does that sound right ? Another question I can't figure .. for the safety of my circuit, and my own person ( I know a cap can be pretty destructive if not used well heh - I've only seen a 10nA blow and I got the picture ). So yeah ... WIMA seems very nice manufacturer. However, I can't find anywhere some 5.5V, there are only either 2.5V or 12V ... so HERE is my noob question !! I aint too sure how a cap react, since I never got to experiment with them ... If I use a 12V cap ... will it stop charging at the voltage of my power source ? ( 5V ) Therefor not charging it at is full capacity - and reducing the time duration of my light when the dynamo is stopped ? OR, it will continue to charge, until it gets full ? But then, ain't it gonna throw 12V in the ass of my circuit at the few first minutes and possibly do some damage ? ( unless I, again, add some things in the circuit to fix that also ? ) Either possbility, if any of them is right, seems to be bad. I was wondering if it's because I don't quiet understand how to use a cap ? ( naaaaah ... ) Well .. so many details ! I know my post are long heh and the answers are much appreciated. In fact I already made a huge step in trying to figure how I'll do it. ( dynamo is AC .. DUH !) Oh I was curious ... I know there's always obviously a lot of things to think when building a circuit. However I like it the more simple it can be ... I was looking at this one: Electrical schema At the start, I was thinking making something similar, with maybe 1 more diode to avoid cap discharge and a resistance if needed, but this guy circuit seems to work fine ( according to him ), and he got pretty much nothing of all that. Would it work if I do that ? ( Replacing the cap with a 600F ) Again, thanks for reading. Once my done I promise I'll post my noob project in the project section !!! tcmtech Banned Most Helpful Member Given the cost of a super cap and all the voltage limiting and what not needed you may very well be further ahead to wire it up with some small Ni Cad batteries. All you would need is a diode and voltage regulator on the line coming from the generator and one diode in the line coming from the battery source. When you are moving the generator output will be at a higher voltage than the batteries so it will supply the power however when you go slow or stop the batteries will take over. a automatic battery charging circuit powered from the generator could be added if needed also. Xenagor New Member I had thought already about that possibility honestly. In fact, I did a very similar system for a end of college project, where I was feeding a basic GPS with a solar panel, and that was charging my battery at the same time. It's easy to do for sure, but the first thing I don't like is the need of changing those battery eventually. Especially since a NiCd battery isn't made to be charge partially, their life duration is increase greatly by charging them full everytime you charge them ( from what I know. ) With a cap, that's not a problem at all, and it will basically last forever if I do it well ! This is kinda why I want that, because once I'm done I'm gonna seal the circuit well in a solid casing, so I don't have to think the water or dust getting it, and so I don't have to open it again and mess with it. Now like mentionned in the post below, I'm trying to figure how a cap works while discharging ... I know it's a basic question - but I couldn't find a CLEAR answer anywhere on the net. This is using a 16V cap ( they are the only available on WIMA website - apart the 2.5V ) If I charge that cap ( 16V max ), with a constant 5V. Is the cap gonna stop as soon it reaches 5V too ? ( which would sound right to me ) - Therefore not been a problem for it to work, but reducing quiet a lot the duration of the discharge, since it will never be fully charge ? Basically, I could just ask, how does a cap discharge Voltage is determined ? audioguru Well-Known Member Most Helpful Member The circuits you posted use a high power LED to clamp the voltage from the dynamo to 5V. The LED flashlight is not high power and it will not clamp the voltage. So the voltage from the dynamo will go as high as it can which might blow up the LEDs and the supercap. A capacitor charges to the supply voltage. The voltage of a discharging capacitor drops exponentially. The voltage drop is very fast at the beginning so you will see the LEDs dimming. The voltage from a discharging battery remains the same until it is dead then it drops quickly. Ni-Cad batteries are obsolete. Ni-MH cells are much better and have replaced them. Their capacity is 4 times or 5 times higher. tcmtech Banned Most Helpful Member Let me get this strait. You ride a bicycle. You cant possibly be so lazy that a several hundred dollar ultra cap plus all of the control circuits justifies the time and money 20 seconds or so and the under$7 battery cost investment for batteries takes up!
If time is a factor you must be the wealthiest man on earth! And if so why are you riding a bike?
Cant you pay someone then to change the batteries for you?
You don't want to use Ni Cad or a battery of some sort due to the fact that some day you may have to change them out. Right? A ultra cap of the same capacity of a simple set of 5 AA batteries is several hundred dollars! And is many times larger.

A cheap set of Ni Cads on a LED garden light get solar charge and rarely see a full charge every time but they still last for a year or two.
But you don't want to go that way because you may have to change them out some day. Is changing a set of batteries every year or two that hard?
And I know for a fact the environmental impact of the batteries is less than that of a ultra cap. So the enviro nutter argument is out!
Plus I am pretty sure a ultra cap based power system is going to take far more than the theoretical two minutes of time you would spend over the next five years changing batteries.

And how much time do you spend riding after dark any way?

I know I am a dick some times, but I just could not resist it on this one! Sorry!

audioguru

Well-Known Member
Most Helpful Member
The cheap old fashioned Ni-Cad battery cells in my Chinese solar garden lights fail after 2 years because their thin steel cover rusts away.

Xenagor

New Member
I had a good laugh reading this useless post ... alright let me answer this first ...

You ride a bicycle. You cant possibly be so lazy that a several hundred dollar ultra cap plus all of the control circuits justifies the time and money 20 seconds or so and the under \$7 battery cost investment for batteries takes up!
Your solution uses battery. I didn't want to be a dick but let ME but straigth.

Why would I come on this forum, wasting so much effort to make a system that charge my BATTERIES, when the goal is to remove them completly? You don't install a dynamo on your bike if you're gonna use batteries .. that really sounds dumb to me.

Several hundreds dollars ... ? The 600Farad 2.5V cap cost less than 10bucks ... As for the circuitry, it's not that expensive, a few diods and maybe a 7805 ... no big deal.

If time is a factor you must be the wealthiest man on earth! And if so why are you riding a bike?
If time was a factor and I was rich, I would buy this:

Supernova Lighting Systems - E3 TRIPLE

Unfortunatly it's not the case ... and the time I will and have put already into this surpass greatly the time to change stupid batteries .... This is a tech forum - I don't see your point at all.

I ride a bike to get back in shape btw, but why are you asking?

Cant you pay someone then to change the batteries for you?
Wow that comment was so stupid ... I'm posting here to get help with a little LED project and someone answers me this. Just Wow ...

You don't want to use Ni Cad or a battery of some sort due to the fact that some day you may have to change them out. Right? A ultra cap of the same capacity of a simple set of 5 AA batteries is several hundred dollars! And is many times larger.
Who is talking about the same capacity ? I said several minutes, not several hours ... But of course the more time I can get is always nice to have.

Again this is a tech forum, you're no help at all heh.

This project is a challenge to me, the goal would be to not have to touch it again once it's installed ... But yes it's a fun challenge for me to try to figure a way more innovative than stupid batteries - never thought it would be so hard to understand on a technical forum like here. But it seems I need to explain.

But you don't want to go that way because you may have to change them out some day. Is changing a set of batteries every year or two that hard?
And I know for a fact the environmental impact of the batteries is less than that of a ultra cap. So the enviro nutter argument is out!

What's wrong with been enviro friendly ??

Again changing a set of batteries once in a while would be a MUCH more simple solution. But I ain't looking that, or I wouldn't be here obviously ...

Plus I am pretty sure a ultra cap based power system is going to take far more than the theoretical two minutes of time you would spend over the next five years changing batteries.
Yes of course, who cares. You obviously but that's pretty much it. It's not the goal here.

And how much time do you spend riding after dark any way?
Not that often, but yes I do sometimes. Why would you care ?

There is absolutely 0% of help in your last comment, I just see someone that think he knows it all, and that has the only suitable solution. However you're totally off the track.

I'm glad I had other people answering me before ya heh, because I would think everyone here is a dork or somthing..

Alright ... I think that close the subject for that. I hope !

Audioguru

Thanks for the info, I didn't know about those power LEDs I shall check them out.

However I just realized in fact that the only available high capacity ( 100Farads or more let's say ) are 2.5V max.

I thought there were 16V, but these are normal low capacitor. The only one I could have enough "juice" in it would be in the 2.5V cap.

I'll have to figure this out, I wonder if I could get a good light with only 2V of power, even though I know some LEDs can work with that.

Maybe if I use cap for each LEDs I wanna light ? Or maybe if I boost the voltage ....

I'll check what I can figure about it.

Any useful comment or help is always appreciated. ( Which doesn't apply to tmctech's last comment.. )

audioguru

Well-Known Member
Most Helpful Member
Kids who ride their bikes on roads at night in the dark?
Stupid road-kill. Only a few of them have dim lights and the rest have none. SPLAT!

Boncuk

New Member
What's wrong using a SuperCap?

They're maintenance free, don't require a charging circuit like batteries and give a reasonable operating time for an LED. Calculate the time knowing that the capacitance of 1F means current flow of 1A for 1 second.

A SuperCap with 600F has these dimensions: lenght=48mm, width=26.5mm, height (including 6.3mm terminals for slip-on connectors) 87mm.

I contacted WIMA today to get a quote about SuperCaps. Will let you know where to purchase them and what they cost.

Boncuk

tcmtech

Banned
Most Helpful Member
My apologies. The last bicycle I owned was when I was around 13. As soon as I figured out engine power, that bike went in the the scrap pile and was never looked at again!
I am not against your building a super cap based system. It just seems like a waste of time and money for something that you will only use a little bit.
I do commend you for at least trying to design a system that keeps you lit up even when your not moving! Thank you!

But I do Side with AudioGuru on th stay of the street in the dark with a bike. Every year there are at least 10 -20 people locally get hit by cars while riding after dark! And at least one or more of them gets killed every year.

If it was up to me I would repeal the seatbelt law in the united states and enact a bicycle restrictions law! You may very well be the one that is a conscientious rider but around here there are too many that think they own the road when they are on their Huffy!

Again, Sorry. Cyclists just push my buttons the wrong way!

Loading