Need help with a Kenetic Energy Recovery System

Thread Starter

bentatts

Joined Nov 17, 2011
2
Hey Guys,

I'm looking for a bit of help on a project I'm building at the moment. I've designed a Velomobile (a mix between a bike and a car) but due to its weight I need a bit of help when it comes to acceleration. This thing is seriously cool and looks like a mini Le Mons car. Made of full carbon fibre it's pedalled like a recumbent bike but due to it's light weight and aerodynamics this bad boy should be able to cruise at about 40-50km/h (around 30mph for our American friends) on just 150 watt of power, what the normal human puts out when they go for a ride on their bike.

The idea is to create a KURS system similar to what the F1 cars use to draw wasted kenetic energy from braking, store it when the vehicles stopped then release it with a simple steering wheel mounted switch.

I'm a bike mechanic by trade so I know a lot about mechanical problems and nothing about electrical. After doing some research I've come up with this plan:

When you want to slow down or stop the vehicle, apply the KURS (a paddle behind the steering wheel) and instead of using a normal disc brake, the KURS system engages and starts to slow the car (theres also a disc brake system on the other paddle so the driver has more control over braking).
The KURS would be an electric motor that is on a clutch system, when the paddle is engaged, the clutch is let out and the motor engages with the rear axle gear and starts generating power (the motor would be a reasonably strong brushless model). The motor draws the energy from the drive shaft, slowing the car and the power generated by the motor would be stored either via a battery (a really light one) or a flywheel (the power would transfer to a second motor that would spin the flywheel).
When the driver needs to accelerate, they simply hit a button on the steering wheel which reverses the cycle, the flywheel motor draws energy from the flywheel and feeds the main motor power so it spins the transmission system linked to the rear axle. The flywheel only needs to store enough power for the motor to run for 10 seconds or so which should be enough for the velomobile to accelerate away accompanied by the driver pedalling it.

The problems I have with this system are:

  • Can an electric motor be both a generator and a motor and swap between the two?
  • How would I get them to change eg. If the driver hit the KERS release button I want the clutch to engage on the rear axle but have the motor driving, not generating and slowing the car
  • Flywheel vs Battery? I like the idea of a flywheel but it requires another motor generator and changing from Mechanical force to electrical all the time I'll surely loose a lot of the energy?
  • AC/DC problems? Power regulator?
  • Weight - this whole project banks on being light, I need an electric motor that can push 100kg (vehicle plus driver) from a standing start to a reasonably acceptable speed (20km/h?) so the lighter the system, the less powers required to push
  • What would I need? I was thinking a motor capable of around 600 watts of power and an energy storage source capable of powering that motor for anywhere from 5-20 seconds?
  • And finally, Is this even possible?
I'd really appreciate any help you can offer, I've got no idea about electric motors or circuits (the only knowledge I have is thanks to wikipedia). The aerodynamics and mechanical side of things are all taken care of and we just need to find a way of incorporating this KERS system into the vehicle to aid in acceleration and give it a burst of speed when required.



Best Regards,
Ben
 

Thread Starter

bentatts

Joined Nov 17, 2011
2
A quick note that's already come up, the electric motor would be applied before the transmission system so the driver can start in an easy gear and shift down thru the gears as the vehicle accelerates. I'd imagine moving 100kg on a big gear would cook an electric motor before it moved anything.
The easiest way to explain the gear system is to think of it as the rear gears of your mountain bike, if you start at the top (the biggest cog) you'll be doing wheelies but not going fast and by the time your at the bottom you can hardly spin your legs.

The diver will also be assisting during acceleration so the motor won't be doing all the work.

Cheers,
Ben
 

Georacer

Joined Nov 25, 2009
5,182
How many gears does your vehicle have? Does it have two sets of gears or one?

I 'm sorry I can't help you on this, but I think we might have some guys here with the adequate knowledge.

Some pics would be awesome too!
 

wayneh

Joined Sep 9, 2010
17,496
  • Can an electric motor be both a generator and a motor and swap between the two?
  • How would I get them to change eg. If the driver hit the KERS release button I want the clutch to engage on the rear axle but have the motor driving, not generating and slowing the car
  • Flywheel vs Battery? I like the idea of a flywheel but it requires another motor generator and changing from Mechanical force to electrical all the time I'll surely loose a lot of the energy?
  • AC/DC problems? Power regulator?
  • Weight - this whole project banks on being light, I need an electric motor that can push 100kg (vehicle plus driver) from a standing start to a reasonably acceptable speed (20km/h?) so the lighter the system, the less powers required to push
  • What would I need? I was thinking a motor capable of around 600 watts of power and an energy storage source capable of powering that motor for anywhere from 5-20 seconds?
  • And finally, Is this even possible?
1. Yes, definitely.
2. The easy part is the manual switch operated by the driver. Harder part will need help from experts here.
3. Both are problematic, and heavy by nature. Biggest problem with a battery is that the charge-discharge cycle is not very efficient. The flywheel is probably more efficient end-to-end, but requires weight and moving parts. My hunch, a battery wins.
4. Not sure what the question is, but most of your system will be DC if a battery is involved. The generator might be an (AC) alternator, requiring you to rectify the output.
5. Yes that's another problem with the whole concept: When you add up motor and battery weight, which must be carried ALL the time, it's hard to make up for that with the little bit of power it might capture since that is only available rarely. When I ride, I work very hard to never use the brakes except for emergencies.
6. Makes sense.
7. Possible? Yes. Practical? I'm skeptical. I was thinking about doing something similar - using a generator as part of the braking system - but only to power a lighting system. That only needs a small percentage of the rider's power.

You might learn a lot by studying the electric-assisted bikes that are out there. They've put design effort into the battery, motor and gearing issues.
 

John P

Joined Oct 14, 2008
2,025
Electric vehicles don't usually have variable transmissions, as an electric motor can produce torque over a wide range of speeds.

I wonder if anything as light as a glorified bicycle would benefit from something this complicated. As Wayneh said, it's hard to justify it in terms of the weight to be carried. It might make sense if the vehicle is to be used in mountainous terrain, but just coming to a stop and starting again doesn't seem worth it. Also, there is the issue that the battery presumably gets totally drained every time the vehicle restarts, and then has to totally charge again when you come to a stop. That's a tough way to operate a battery.

Given the amount of energy to be stored, could you wind up a spring as part of the braking system, then apply the force to the wheels to restart the bike?
 

GetDeviceInfo

Joined Jun 7, 2009
2,192
I was thinking a motor capable of around 600 watts of power and an energy storage source capable of powering that motor for anywhere from 5-20 seconds?
Considering the description of your vehicle, that's a substantial amount of power, both in terms of wieght, and of storage capacity. What your describing is regeneration, unless you plan on bringing a charge onboard. So, what you put in, is what you get out. In regeneration, the maximum is what energies can be accumulated in braking the vehicle.

but here's your problem, and I'll use a mechanical analogy for your consideration. Consider a balloon as an energy storage device. The problem is how do you continue to blow up the balloon with dwindling air pressure? Another way of looking at it from the flywheel point of view, where the mass is fixed and speed accumulates, is how do you add speed (energy) to the flywheel from a decelerating vehicle?
 

wayneh

Joined Sep 9, 2010
17,496
It might make sense if the vehicle is to be used in mountainous terrain...
That's a tempting thought, but I think it's the reverse situation, that it would be a big pain on hills because of the added weight. It HAS to be less than 100% efficient overall, and I don't think the small amount of energy it captures on the downhills would make up for the extra energy you'd need to haul it up a hill. The hills would actually accentuate the problem.

I think on level ground, and in an environment with lots of unavoidable braking (city?), it might make more sense. Still, you'd have extra mass to accelerate after each stop. It'd have to capture more energy than just the amount needed to re-accelerate its own mass. That should be possible, since it needs to decelerate a relatively heavy rider and bicycle carrying lots of kinetic energy.
 

Feign

Joined Mar 30, 2009
50
I believe a flywheel would be inappropriate. I wouldn't want to push much less pedal a sturdy enough vehicle to hold that kind of power.

If you really just want a helper, here are some very light ones you may of already seen. They do not however have your primary requirement, regenerative braking, since they are using off the shelf RC control.

http://www.recumbents.com/wisil/tetz/e-assistmetric/

http://www.recumbents.com/wisil/shumaker/

These guys are using a 110F/16V Capacitor from Maxwell, for a electronic flywheel. Mousers lists the price at about 650$.

http://web.mit.edu/first/kart/

http://web.mit.edu/first/kart/everpres.pdf
Vehicle Specifications:

Weight (No Driver) 250lbs (113kg)
Ground Clearance 1"
Drive RWD
Wheel Size F:10"/R:11"
Gear Ratio 3.2:1

Battery Voltage 39.6V
Battery Capacity 40Ah (1.6kWh)
Ultracapacitor Voltage 16V
Ultracapacitor Capacitance 110F
Ultracapacitor Energy 14kJ (3.9Wh) (19hp-s)
Peak Controller Current 300A

Peak Mehanical Power 10hp (7.5kW)
Peak Torque @ 0RPM 40lbf-ft (54N-m)
Peak Acceleration 0.6g
Top Speed (est.) 40mph (64km/h)
Way overpowered for a velo-mobile. Just scale it back. 750 watts in most of the states, is the maximum legal.
 

Georacer

Joined Nov 25, 2009
5,182
I 'd like to point that the system doesn't need to be more efficient than the bike without it, in terms of rider stress. It just has to carry its weight and then some, to be worth it.

It will be cumbersome in long outdoor climbs, but in general in an environment with a lot of downhill slopes it might be worth it, regulating the descent speed and charging simultaneously.
 

wayneh

Joined Sep 9, 2010
17,496
I 'd like to point that the system doesn't need to be more efficient than the bike without it, in terms of rider stress. It just has to carry its weight and then some, to be worth it.
Your distinction is lost on me - I don't see how you get one (efficiency) without the other (carry its weight and then some).

Bottom line is rider energy needed to traverse from point A to B. I call that efficiency. Personally, I ride in the (flat) country and go for hours without ever touching the brakes. No way this thing would help me.

I'm coming around to your way of thinking though, that this technology might be useful any time the rider is using the brakes frequently, including hills or city. It draws energy from the the combined kinetic energy of the rider, the bike, and its own weight. The only energy penalty is accelerating or raising its own weight, which is relatively much smaller. So as long as it captures more than enough to move itself around, it should be a net plus to the overall system.

Since a rider on a bike is already the most efficient mode of transportation known, it'll be interesting to see if any significant improvement could be achieved.
 

Georacer

Joined Nov 25, 2009
5,182
We should keep in mind that what comes up, must go down, though. In an uneven and hill(y) environment, there would be energy gain, while going downhill, but when going up the hill, the extra weight can bring the biker to his knees. Battery charge must be present there to help the rider climb.

Looking backwards, I indeed see that what I said in the previous post doesn't make much sense. Thinking again over this, the system would be beneficial if it operated in an environment where it would capture otherwise lost breaking energy; also that energy should overcome the encumbrance its weight causes.

So our "equation" should be:
Breaking Energy>System Carrying Work

I 'm not sure that in a bike that would be profitable. The bike weighs very little, and therefore the kinetic energy captured by the KERS system would be minimal.
 

THE_RB

Joined Feb 11, 2008
5,438
...
In regeneration, the maximum is what energies can be accumulated in braking the vehicle.

but here's your problem, and I'll use a mechanical analogy for your consideration. Consider a balloon as an energy storage device. The problem is how do you continue to blow up the balloon with dwindling air pressure? Another way of looking at it from the flywheel point of view, where the mass is fixed and speed accumulates, is how do you add speed (energy) to the flywheel from a decelerating vehicle?
In DC motor regeneration (with synchronous rectification) you simply change the PWM up to increase current going from battery to the motor, or reduce PWM down, which will increase current taken from motor and delivered to the battery.

And at the "coasting" point the PWM will balance motor EMF vs battery EMF so no current flows either way.

All it needs is a PWM driven half bridge and it will be a motor controller with regenerative braking. It's done in mobile robotics as standard.
 

wirednuts

Joined Nov 13, 2011
20
In DC motor regeneration (with synchronous rectification) you simply change the PWM up to increase current going from battery to the motor, or reduce PWM down, which will increase current taken from motor and delivered to the battery.

And at the "coasting" point the PWM will balance motor EMF vs battery EMF so no current flows either way.

All it needs is a PWM driven half bridge and it will be a motor controller with regenerative braking. It's done in mobile robotics as standard.

wow, i never knew that. its such a simple idea... why is this technology so expensive in cars? the money really is going into battery safety isnt it? i thought the electronics to do all this new braking energy recovery was where the money was going, but if this is how they do it then thats a really cheap solution.

my only other guess was the batteries... and even though they might only cost $5k to make a pack, the price of an electric car is $20k more simply because all of the safety precautions from RnD to physical hardware is whats so expensive.
 

GetDeviceInfo

Joined Jun 7, 2009
2,192
In DC motor regeneration (with synchronous rectification) you simply change the PWM up to increase current going from battery to the motor, or reduce PWM down, which will increase current taken from motor and delivered to the battery.

And at the "coasting" point the PWM will balance motor EMF vs battery EMF so no current flows either way.

All it needs is a PWM driven half bridge and it will be a motor controller with regenerative braking. It's done in mobile robotics as standard.
help me with this, I must be having a seniors moment (or is it like this for the rest of the ride). Current will flow from the higher voltage to the lower, so at start, the battery will deliver current to the motor until EMF counters. On decel, how does a slowing motor generate the EMF to overcome battery voltage and subsequently send current to it?
 

wirednuts

Joined Nov 13, 2011
20
im not sure i understand the question? the motor can output power at whatever it is designed to... which is just above battery voltage typically. that way when its in charging mode it can charge the battery, and during acceleration youre just giving it the batterys voltage which is lower but still plenty to run it.
 

wayneh

Joined Sep 9, 2010
17,496
On decel, how does a slowing motor generate the EMF to overcome battery voltage and subsequently send current to it?
Your question is essentially, how does a generator work? Shaft work applied to the motor forces the coil windings through the magnetic field established by permanent magnets. Moving a conductor through a magnetic field creates an EMF proportional to the number of windings and the rate of rotation (actually, rate of change of the magnetic field). When the EMF exceeds the battery voltage, current will flow in a direction to charge the battery. With current flowing through the conductor, it's harder to force it through the magnetic field, so braking is applied to the vehicle - the motor resists turning - as energy flows into the battery. If the battery EMF exceeds the shaft work input, the motor will behave as a motor.
 

THE_RB

Joined Feb 11, 2008
5,438
help me with this, I must be having a seniors moment (or is it like this for the rest of the ride). Current will flow from the higher voltage to the lower, so at start, the battery will deliver current to the motor until EMF counters. On decel, how does a slowing motor generate the EMF to overcome battery voltage and subsequently send current to it?
A PWM half bridge is a SMPS circuit. It has two power FETs which operate alternately which mimics the operation of a SMPS buck converter when driving the motor. The two FETs always work like a FET and a rectifier as in a SMPS.

Driving the motor; works like a buck SMPS; high voltage low current in, low voltage high current out to motor.

Regen braking; works as a boost SMPS; low voltage high current supplied by the motor, converted to high voltage low current and supplied back to the battery.

The only real functional difference between motor drive and motor regen is how the PWM is adjusted, ie if the PWM average voltage is higher or lower than the motor generated EMF.

Like I said, it's done all the time in mobile robotics where the motor is attached to the wheels with no clutch (no coasting).

In electric vehicles (especially lightweight bikes etc) the best efficiency usually comes from coasting.
 
Top