Electrical engineer ninjas needed for dyno hub circuit- Mtbr.com
Results 1 to 18 of 18
  1. #1
    Man up
    Reputation:
    Join Date
    Sep 2010
    Posts
    485

    Electrical engineer ninjas needed for dyno hub circuit

    I'm interested in a circuit like this, particularly with size: Dynamo LED Light Systems for Bicycles (electronic circuits) but something with this functionality: Regulator for a Hub Dynamo

    Ideally, I would want a Son28 hub (29errrrr) to power an LED light directly, any excess voltage produced would go into batteries, if the batteries are full it doesn't charge them. Basically LIGHT-OFF-CHARGE/NO LIGHT functionality with auto-switching at low-speed to change from AC to battery for the LED. This battery will also feed my helmet light as well as GPS.

    What do you electrical weenies think? I wish I knew more about this, but alas they only briefly touch on circuits in ME curriculum...

    Jesse
    '08 GF Montare
    '10 Rockhopper SS
    Semper Fidelis
    02-10 SSGT of Marines
    (Most) always down for a ride in the Gville area

  2. #2
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    752
    The circuits you referenced are fairly "old school", in the sense that they are primarily analog electronics. There's nothing wrong with that, but it's easier to do a lot more fancy logic with a micro controller based designs. This is particularly true now that lights are LED based and typically driven off switching current regulators. LEDs have very different characteristics as opposed to incandescent bulbs.

    If I were building something to meet your description, I would build an MCU controlled buck-boost input stage. That would go between the dyno and the battery pack. Then I would always drive the LED downstream from the battery pack using a buck regulator controlled by the same MCU. Then the input stage would optimize the load placed on the dyno to always get max power out of the dyno. That power would then be split between driving the LED and charging the battery. Anytime the battery is fully charged, the load on the dyno would be reduced. You could even customize the drag created by the dyno based on speed through the MCU programming.

    Designing such a circuit is no small feat. Not likely something I'd do in my spare time unless I was planning to use it myself and I don't even have a bike with a dyno.

  3. #3
    Man up
    Reputation:
    Join Date
    Sep 2010
    Posts
    485
    Quote Originally Posted by MtbMacgyver View Post
    ...If I were building something to meet your description, I would build an MCU controlled buck-boost input stage. ....Then the input stage would optimize the load placed on the dyno to always get max power out of the dyno. That power would then be split between driving the LED and charging the battery. Anytime the battery is fully charged, the load on the dyno would be reduced. You could even customize the drag created by the dyno based on speed through the MCU programming.
    What does this mean, particularly the input stage statement?

    Why not just rectify the AC from the hub with an overvolt/overcharge sensor upstream of the battery? Super loss of efficiency through the circuitry and battery?
    '08 GF Montare
    '10 Rockhopper SS
    Semper Fidelis
    02-10 SSGT of Marines
    (Most) always down for a ride in the Gville area

  4. #4
    mtbr member
    Reputation:
    Join Date
    Feb 2009
    Posts
    169
    The first thing is what kind of dynamo? What kind of voltage and power? Six volts I imagine, right? The particulars do matter here.

    I would be inclined to not deal with all the complication of the switch between half wave and full wave rectification. That's why you have a battery, correct?

    I'm thinking a bridge rectifier, a 4.1 volt shunt regulator, and a li-ion. Then a one or more 7135's for current control. Have a three position switch. The three positions charge, dynamo, and battery. In charge the dynamo is connected to the battery to charge it to that 4.1v. The second dynamo position the battery is disconnected, and the dynamo powers the LED. The light is not as much as the complicated circuits at low speeds, but that is better than nothing. In the third position the battery gets connected. When the dynamo is making power it lights the LED and any extra charges the battery. At low speeds the battery will run the light.

    Just be sure to put the switch in the dynamo position when the bike is parked or there will be a small drain on the battery.

  5. #5
    Man up
    Reputation:
    Join Date
    Sep 2010
    Posts
    485
    The particulars of the hub are in the hub itself, but yes, it is 6v/3w.

    So you are saying have the LED powered by the battery or the hub, wouldn't that require a light that is capable of DC and AC? Or would there be a rectifier upstream of the LED on one circuit path and the battery would T into it post-rectifier?
    '08 GF Montare
    '10 Rockhopper SS
    Semper Fidelis
    02-10 SSGT of Marines
    (Most) always down for a ride in the Gville area

  6. #6
    mtbr member
    Reputation: ktronik's Avatar
    Join Date
    Dec 2006
    Posts
    424
    I made a basic unit, that charges a battery via the dynamo, and outputs to charge stuff...have a look at the below face book link, for a detailed write up...maybe that will help with a few ideas...

    K

  7. #7
    Man up
    Reputation:
    Join Date
    Sep 2010
    Posts
    485
    I didn't find a detailed writeup, just some pics. No info on your website either. You have a more direct link so I'm not searching your entire Facespace?
    '08 GF Montare
    '10 Rockhopper SS
    Semper Fidelis
    02-10 SSGT of Marines
    (Most) always down for a ride in the Gville area

  8. #8
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    752
    Quote Originally Posted by orion_134 View Post
    What does this mean, particularly the input stage statement?

    Why not just rectify the AC from the hub with an overvolt/overcharge sensor upstream of the battery? Super loss of efficiency through the circuitry and battery?

    It's all a matter of how fancy you want to get. The buck/boost input stage actually includes rectification, but it also allows for two capabilities that are difficult to achieve with a straight bridge rectifier.

    - It will allow for full power charging of the battery at slower speeds. To get the full power out of a dynamo at slower speeds, you have to boost the voltage which can't be done with a straight rectifier.

    - Once the battery is partly or fully charged and the light is drawing less power than the dyno can actually produce, the buck function of the input stage will let you draw the power you need without fully loading the dyno. This results in less drag on the rider. This is also something that's difficult to accomplish with simpler solutions because it uses the buck function of the input stage to reduce the voltage from the dyno with a constant power conversion. Most simpler circuits regulate the voltage from the dyno to prevent battery overcharging by burning up the excess power as heat.

  9. #9
    Man up
    Reputation:
    Join Date
    Sep 2010
    Posts
    485
    So the concept sounds awesome; how do I do it?
    '08 GF Montare
    '10 Rockhopper SS
    Semper Fidelis
    02-10 SSGT of Marines
    (Most) always down for a ride in the Gville area

  10. #10
    mtbr member
    Reputation:
    Join Date
    Jan 2007
    Posts
    752
    Quote Originally Posted by orion_134 View Post
    So the concept sounds awesome; how do I do it?
    I'd love to design something like this, I've always been interested in this area. But I would estimate it's a couple of months of design and testing. Too many projects going on right now that pay for all the toys. Maybe in the future when things calm down some.

  11. #11
    Man up
    Reputation:
    Join Date
    Sep 2010
    Posts
    485
    I can do the testing, I just need to know what to buy.
    '08 GF Montare
    '10 Rockhopper SS
    Semper Fidelis
    02-10 SSGT of Marines
    (Most) always down for a ride in the Gville area

  12. #12
    mtbr member
    Reputation: ktronik's Avatar
    Join Date
    Dec 2006
    Posts
    424
    Hey...

    let me help...

    dyno-bob wheel

    and my FB post..

    Shannon is riding 5000km around Australia and needs power on the road.. The dynamo hub will re-charge a kLite 'Dyno-Batt' and that in turn charges his Laptop, phone GPS and camera...good luck to ya, you will need it!!


    Well Shannon, I think it works!!! just did the 'proof of concept' test...really the most important one, the one where I see if my mad ideas that are in my head, really work in the real world...with one spin on the dyno wheel, I got 480mA in to the battery, as charge current...more than happy with that.. I used the high speed bi-polar cap (100uf) and will test at slow speed, to see what the charge curve looks like...can add the extra cap if not enough at low speed

    dyno-batt

    dynobatt

    A quick iphone pic of the new dyno-batt prototype. Charges from a shimano dynamo hub, and outputs to the kLite charging station, which in turn offers, USB power out and 12v, built in charger for AA/AAA and other batterys.

    Part 2, the k-lite charging station...more soon.

    testing

    Dyno-Batt test1, 0-70km/h E-bob driving Dyno-bob, sweet spot in power curve is 18km/hr @[email protected] 70km/h no lower than 550mA.

    Dyno-Batt test1, 0-70km/h E-bob driving Dyno-bob - YouTube

    #
    Here's the trade off, we can have more power down the lower part of the power curve, but less peak at that speed, and will be drawing when you going slow, and suffering. if we leave it, as is, the big charging peak will be when your going a bit faster, taking the load off you when going slow...guuna try the slow speed curve and report back.
    June 26 at 11:29am LikeUnlike
    #
    kLite.com.au
    OK, just done the 200uF (low speed power curve) test...0-10 its has a tiny bit more power, but less @ 10-20km/h with a peak of 490mA, but very flat curve to 70km/h...IMHO, not worth it if you riding at 10-20km/hr most of the time. Due to th...e 16" wheel RPM roll out, we are better to leave it at 100uF tuning cap, for higher peak power @18km/hr and more power present down to 10km/hr...so for the Dyno-bob high speed curve is the go, as its in a 16" wheel and does higher RPM for a given speed than a 26" wheel that the dynamo is made for.See More
    June 26 at 11:55am LikeUnlike
    #
    kLite.com.au
    Real road test 1: hooked up the Dyno-BOB for a quick spin, with the current meter in the front of the bob...I rolled up to my touring speed, looked back to the meter, and it was 620mA, fantastic, great result, even climbing I was reading 2...00mA easy...looks like I could charge this 50Whr battery in 6-7hrs. We could then charge 18AA batteries, with 50Whr. (14.4v/3.5Ahr NIMH battery) We have really maximized the output power, by tuning the hub to the riding speed.


    its a very simple circuit...fullwave recitiver, AC input tuned via a 100uf bi-polar cap, for max input power, then sent to the nimh pack...


    final protection:

    dynobatt protection module

    has a loud settable alarm, and output to trigger a relay to disconnect the load.


    hope that helps

    K

  13. #13
    mtbr member
    Reputation: ktronik's Avatar
    Join Date
    Dec 2006
    Posts
    424
    The circuit is made this way so that the output of the dyno batt, drives the klite charger, to charge AA's / lipo or any battery type + 5v USB output...the cool thing with shannons kit, is that he can also use the same klite charger to charge the main dynobatt battery, from 240v at a powered camp site...this takes a little load off the legs for a day or two after he rolls through a town...big spaces here in OZ...thus the 14.4v / 3500mAh pack...as he's going to the moon, we added a few extra things.. swapable tuning caps, just plug in another or different value...and everything is made bypassable, in case something goes wrong, its 'plugged' so you can leave a bit out, what ever...and it will still run..its full proof system with backup redundancy measures...

    for another customer we are making a 7.2v version with only USB output...smaller lighter...better for that customers app...

    K

  14. #14
    mtbr member
    Reputation:
    Join Date
    Sep 2011
    Posts
    12
    Quote Originally Posted by orion_134 View Post

    Ideally, I would want a Son28 hub (29errrrr) to power an LED light directly, any excess voltage produced would go into batteries, if the batteries are full it doesn't charge them. Basically LIGHT-OFF-CHARGE/NO LIGHT functionality with auto-switching at low-speed to change from AC to battery for the LED. This battery will also feed my helmet light as well as GPS.
    Jesse
    I built a dynamo powered lithium ion charger a few summers ago. It was rudimentary (in that it used a linear regulator and no fancy switch mode stuff and would probably fry the charging IC once the battery was fully charged and the lights were off!), but it worked for a long time until some dork (to put it mildly) stole the rear light.

    I documented the process, but being a new member, can't post links (bummer, cuz I wanted to tell all of you about how much $$$$ I make working from home. ).

    In lieu of a direct link, try this tiny url: 4xoem5m

  15. #15
    mtbr member
    Reputation: znomit's Avatar
    Join Date
    Dec 2007
    Posts
    1,961
    Quote Originally Posted by minisystem View Post
    In lieu of a direct link, try this tiny url: 4xoem5m

    bicycle lighting: dynamo lithium ion battery charging

  16. #16
    mtbr member
    Reputation:
    Join Date
    Jul 2010
    Posts
    55
    My solution would be to use the following circuit to give me a constant souce and build from there , ie run usb or keep batteries charged etc.

  17. #17
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    2
    I've been fiddling with this sort of thing for a while, decided that charging a battery was a dubious proposition on a bicycle. IF I were to do such a thing, Li-ion batteries (or LiFePO4) would probably be my choice because you can read their charge from the voltage, and stop at about an 80% charge (safety margin, longer battery lifetime).

    I got my information from pilom.com, particularly BicycleElectronics/HubDynamo.htm

    What I built (latest-and-greatest) was two PCBs, one rectifying, doubling, and smoothing the dynamo output, also running it through a switching current supply with a supply-dependent brightness control, and the second PCB watching the wheelturns and connecting a battery when they are too low.

    Since it's all run by software, it's possible to connect the battery when you want to charge it, too.

    Standlight here at my wordpress blog, standlight-for-power-leds

    Rectifier/Doubler/Smoother/Regulator at same place, lighting-control-circuit-fixed-working

    (I'm not allowed to post links)

    The two most important things to be aware of are (1) vibration, it is not your friend, and (2) do not design around "6 volts". With a hub dynamo, failing to pull all the current it offers results in soaring voltages. A little of this gets you more power, a lot of this gets you cooked electronics. I have cooked electronics. Part of the regulator circuit is a shunt, designed to get rid of excess power at high speeds (or when a wire breaks, or when the battery charger decides the battery is full and quits drawing current).

    I'm still fiddling with this sort of stuff, and there is a certain air of Bullwinkle-moose-this-time-for-sure (see cooked electronics), but it's a fun hobby. Silicon switches make more sense than what I did in these circuits; it also makes sense to include a switcheable doubler, and a 5V supply capable of running a USB device. All this drives the cost up.

  18. #18
    mtbr member
    Reputation:
    Join Date
    Sep 2011
    Posts
    12
    Quote Originally Posted by dr2chase View Post
    I've been fiddling with this sort of thing for a while, decided that charging a battery was a dubious proposition on a bicycle. IF I were to do such a thing, Li-ion batteries (or LiFePO4) would probably be my choice because you can read their charge from the voltage, and stop at about an 80% charge (safety margin, longer battery lifetime).

    I got my information from pilom.com, particularly BicycleElectronics/HubDynamo.htm

    What I built (latest-and-greatest) was two PCBs, one rectifying, doubling, and smoothing the dynamo output, also running it through a switching current supply with a supply-dependent brightness control, and the second PCB watching the wheelturns and connecting a battery when they are too low.

    Since it's all run by software, it's possible to connect the battery when you want to charge it, too.

    Standlight here at my wordpress blog, standlight-for-power-leds

    Rectifier/Doubler/Smoother/Regulator at same place, lighting-control-circuit-fixed-working

    (I'm not allowed to post links)

    The two most important things to be aware of are (1) vibration, it is not your friend, and (2) do not design around "6 volts". With a hub dynamo, failing to pull all the current it offers results in soaring voltages. A little of this gets you more power, a lot of this gets you cooked electronics. I have cooked electronics. Part of the regulator circuit is a shunt, designed to get rid of excess power at high speeds (or when a wire breaks, or when the battery charger decides the battery is full and quits drawing current).

    I'm still fiddling with this sort of stuff, and there is a certain air of Bullwinkle-moose-this-time-for-sure (see cooked electronics), but it's a fun hobby. Silicon switches make more sense than what I did in these circuits; it also makes sense to include a switcheable doubler, and a 5V supply capable of running a USB device. All this drives the cost up.
    Nice work! I'm looking at doing something similar with an Attiny85 (or similar). So far I've got a prototype working using an Arduino.

    I too, have had my fair share of things going up in smoke (although frustratingly, they don't literally go up in smoke, which prevents you from easily identifying and replacing said smoking party).

    Going to dig through your schematic and code. Thanks for linking - your blog has not turned up in my countless searches for such things!

Similar Threads

  1. Are BMX bikers ninjas?
    By SWriverstone in forum General Discussion
    Replies: 6
    Last Post: 04-28-2011, 07:24 AM
  2. 3 Cree/R2 dyno light + circuit re-drawn for newbies
    By ktronik in forum Lights DIY - Do It Yourself
    Replies: 43
    Last Post: 07-28-2009, 10:30 PM
  3. Mountain Biking Ninjas!
    By Chumba15 in forum Downhill - Freeride
    Replies: 6
    Last Post: 07-25-2009, 05:41 PM
  4. KCNC Ti Pro Lite - CNC machine/engineer needed!
    By elephant in forum Weight Weenies
    Replies: 15
    Last Post: 05-24-2007, 06:19 AM
  5. Ninjas?
    By Johnny M in forum General Discussion
    Replies: 3
    Last Post: 10-12-2004, 07:45 PM

Members who have read this thread: 2

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

THE SITE

ABOUT MTBR

VISIT US AT

© Copyright 2020 VerticalScope Inc. All rights reserved.