Quick and dirty regenerative braking question.

Discussion in 'General Electronics Chat' started by rasputin666, Jan 16, 2009.

  1. rasputin666

    Thread Starter New Member

    Aug 22, 2007
    The controller on my electric bike recently died and the cost of sourcing a replacement was a little more than my wallet could bear, so I sorted out a cheaper solution utilizing a PICAXE micro controller.

    I also recently picked up 100k μF worth of electrolytics (2 x 30k and 2 x 20k) for next to nothing. So I've gotten to wondering if there is a quick and dirty way to capture the braking energy with one or more of these and then feed it back into the motor, switching over to the battery once the charge is depleted.

    Ideally I would like to do all switching control and sensing with the PICAXE and software.

    My first thoughts are back to back P & N-type MOSFETS between the motor and capacitor(s) and simple resistive dividers connected to an ADC input to sense the charge level. However I have a strong suspiscion that it can't be quite this simple.

    If anyone has any ideas I'd be most appreciative.
  2. kubeek


    Sep 20, 2005
    You need to realize that capacitor is not a smaller battery. Battery holds voltage fairly constant before totaly empty, but voltage on capacitor drops linearily through time, if the current draw is constant.

    You would have to use some switch-mode power supply, capable of operating on input voltage say 1V-50V and producing the voltage your motor needs.
    Ideally it should be able to work both ways, so that you can also charge the cap to full voltage, from the back EMF the motor generates on breaking.

    And anyway, 100mF is not too much for a cap bank, but I guess it is pretty large. Even if charged to 50V, you can store only 625J of energy. That is for a vehicle + person weighting 100kg, energy to change speed by cca 3.5m/s (13km/h).
    Average decelleration of public transport vehicles is about 1 m/s^2, so you need power of 625/3.5=180W.

    The convertor needed for charging and discharging the cap is possible, but will be quite big and complex, in order to be ideally efficient.
  3. rasputin666

    Thread Starter New Member

    Aug 22, 2007
    I found this, http://www2.ing.puc.cl/power/paperspdf/dixon/42a.pdf, and I think I get the basic principles of how it works, but most of what I know about electronics is digital, and LC circutry has always been way over my head.

    I'm not looking for ideal efficiency, just a way to claw something back, that is not too complex or expensive.
  4. kokkie_d

    Active Member

    Jan 12, 2009
    Hi. Interesting find. I did not have that paper yet. It's a very interesting design but I would use a more power leg design, see attachement.

    The equation to calculate the size of the inductor can be found on wikipedia:

    Now use the attched design per power source (1 for battery and 1 for capacitor) and control the mosfets through your controller (PWM). You might want to include a form of current control.

    100mW is not extremely large but might give you a bit extra power. The great thing about the design as shown is that you can also charge the capacitors from the battery and power the motor from both sources at the same time thus getting boost power for acceleration.

    I hope this helps.