I am a mechanical engineer therefore please correct me if there is anything that I got wrong. I am working on a regenerative braking system for an IC motorcycle and I am looking for the best way to improve its performance. The current theoretical system uses a 48V, 2kW BLDC motor and a bank of ultracapacitors. The problem is quite complex since the voltage produced by the motor/generator depends on the angular velocity which drops with the speed of the vehicle. Charging the ultracapacitors up to 48V would therefore require either braking at high velocities or a DC/DC converter. There is also another issue of controlling the braking force of the motor which is beyond my understanding of electrical engineering and mechatronics.I think that breaking torque is related to current? Should I control current for smooth breaking?

Looking for a simpler solution I had an idea of having a bank of ultracapacitors which could be electronically modified by changing the number of cells connected in series. At low speeds, the electronic system would charge low number of ultracapacitors connected in series at once, monitoring the voltage and switching between different units until the bank is fully charged taking advantage of low speed braking. If more braking power is required or the velocity of the vehicle is higher, the system would connect larger number of units in series to increase the overall voltage of the bank. Once the harvested energy is used to power the vehicle, the power and torque of the motor could be controlled by modifying the number of ultracapacitors in series. Obviously, the complexity of the system means that a number of monitoring devices would be required to prevent overcharging of single cells and to switch between the ultracapacitors at the right time. Would my proposed cap switching system be more complex than a DCDC converter?How many steps would I need for reasonable smoothness?

I understand that I am quite likely over-complicating things for myself therefore if there is a simpler way please let me know.

  • \$\begingroup\$ I'm sure you don't have just a braking system, I suspect you have something to brake as well, like a car? What does that use to power the motors from its batteries and regen caps? Probably a DC-DC converter. Does this have 4 quadrant capacity, or only 2? And no, adding flying switches to a capacitor bank, while sounding easy, would complicate the otherwise relatively simple design of the other 2 quadrants of control of the existing DC-DC converter that powers the car. \$\endgroup\$ – Neil_UK Jan 3 '16 at 9:02
  • \$\begingroup\$ The regen braking I am working on is for a motorcycle and because most of the braking force is available at the front I'm thinking about using a motor hub to harvest the energy. However this will take away the possibility of using a gear or similar to increase the angular velocity of the motor hence low voltage output. My concern is that the DC/DC converter might not be capable of brining the voltage up enough to charge to the ultracap. Do you know much about the performance of the DC/DC converters? \$\endgroup\$ – M Rogoz Jan 3 '16 at 10:25
  • \$\begingroup\$ DC-DC converters can be made very efficient, >90%, and can happily be designed to go to any practical voltage. 48v is a particularly sweet spot, as far as available power devices is concerned. Use a motor hub, don't use gears, you will have to have electronics anyway to control charge, whether you switch series parallel or not. Have the switcher do the whole job in one go. When the caps are charged, what are you going to do with the energy? Put it back through the front wheel, or the back wheel? \$\endgroup\$ – Neil_UK Jan 3 '16 at 12:40
  • \$\begingroup\$ Powering the front wheel would be quite risky and a lot of research (mostly primary through experiments) would be required to make it safe therefore I would use another motor to power the rear wheel. My original system used a motor/generator connected to the rear wheel but now I'm investigating different options viable for mass production. I'm only doing theoretical work (masters project) but I do not want to end up with a wish list of things that cannot be accomplished. Do you think that using an ultracap with a small Li-ion battery would be easier because It would definitely be cheaper? \$\endgroup\$ – M Rogoz Jan 3 '16 at 13:50
  • \$\begingroup\$ I'm maybe just not asking the right question. Does your bike have a battery pack on it? Can that pack accept a charge, and if so, at what rate? What do you expect the braking power output of your front wheel hub generator to be. In other words, do you need ultracaps as a temporary power buffer between your braking and your main battery pack? The energy storage of the buffer, and its input/output power ratings are far more important design parameters. DC-DC converters are mature enough that you can almost ignore specific input/output voltages as a difficulty. \$\endgroup\$ – Neil_UK Jan 3 '16 at 14:35

While what you propose can be done in theory, it seems impractical compared to alternatives. The various switches, and controlling them, to combine the capacitors in different combinations won't be easy.

You said yourself a DC to DC converter is another option, but ignored it. That seems like a simpler system, and most likely you will want such a converter there anyway for other purposes, like charging the battery from long downhill slopes.

This also begs the question why you are using capacitors in the first place. Why not have the motor charge the battery during braking or long downhills? The energy density of capacitors is less than that of batteries. Since the batteries have to be able to produce the maximum current into the motor, they should also be able to handle about the same current in reverse charging them. What do you think the advantage of a mixed capacitor/battery system is as apposed to just a battery? I suppose they are more efficient short term storage, but at the cost of considerable bulk and weight. Of course that weight decreases overall efficiency and the extra space either makes the vehicle less usable or presents more wind drag at high speeds.

I haven't done the calculations, but I'm skeptical that adding enough capacitance to store the kinetic energy of the vehicle makes sense.

  • \$\begingroup\$ Well, it all depends on the application. I only need a relatively small amount of energy stored in the bank to run the generator for about 20 seconds. The battery, although has much better energy density (200 Wh/kg) while ultracapacitors available on the market have anything between 5-10 Wh/kg, the power density and the life span of the second is much better. Another reason for using capacitors is the potential in the technology. At the moment there are many claims made by scientist that in laboratory conditions they can match the performance of Li-Ion batteries so maybe its worth working on. \$\endgroup\$ – M Rogoz Jan 2 '16 at 22:39
  • \$\begingroup\$ Whether capacitors are used is primarily dependent on the rate of energy transfer; in large scale applications the batteries cannot handle the current being generated, and we have to throw away the energy thatr cannot safely be used to charge the battery. Ultracapacitors are used simply as a reservoir to avoid wasting that energy, at the cost of (as you rightly point out) extra weight, complexity and a loss of some efficiency. I have some ultracapacitors in a closet at the office capable of sustained charge / discharge in 100s of amps and a 1 second rate in excess of 1000A. \$\endgroup\$ – Peter Smith Jan 3 '16 at 14:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.