Ebike regeneration system. Advice needed.

[rickey1990]

Recently i brought myself a ebike, Recently in the UK the law has changed regarding twist throttles, meaning someone without a licence can only ride a pedal assist. luckily any bike brought into the UK before January 2016 is exempt. it helps taking stuff up and down my allotment with my bike trailer. So most of the time i am using the throttle.

long story short, I brought a front wheel with a built in dynamo, AC/ 6v. I am looking for a way for the dynamo to kick in automatically when there's no power being sent to the motor, So it will detect that the throttle and the pedal assist isn't in use. It will be used for going down hills and when im slowing down. I didn't want to have it on 24/7 as there's going to be resistance on the front wheel from the load of the battery I want to charge. its not a big issue, but i don't want to be using power from the main battery to be basically powering a smaller one for lights, phone charger ect.

I did come up with a circuit that used a hall effect switch, op-amp, transistor inverter, optocoupler, resistor to drop the output down to zero volts. Then i came across the TLE4935L, A bipolar (Zero volts when off) hall effect switch with pretty much everything built in. I looked at Hall effect loop current sensors which go around a wire. £25. So i would also like to know if anyone has ever made a DIY one, i have a plan of getting a ferrite toroid, cutting a slot into with a dremel, there needs to be a small air gap, so im thinking of using some thin plastic strips around the outside of the hall effect sensor. place it over the wire leading from the controller to the motor, then heat shrinking it?.

Also to make it feel abit more like i actually achieved something, a no inverting signal booster, to get the signal from on end of the bike to the other, if there's any another idea's please recommend them as i want to do it the best way.

[uncle_bob]

Hi

You hopefully realize that what you are doing is an electric brake. You will slow down as soon as the generator kicks in. How much you slow down depends on how much power you pull from the system.

Unless you can run things at their optimum speed, efficiency will be poor. Figure that 30% of the braking force will go into your battery. The rest will just be braking.

Next up is the battery it's self. You only get a finite number of charge / discharge cycles. The exact number depends on the depth of discharge, the load, the charge rate, and the particular battery you have. In a braking situation, you have little control over how much energy you put into the battery. You also need a fairly fancy circuit to make sure that things do not go wrong.

Yes, this could go on a bit further ....

Bob

[NiHaoMike]

The right way to do it is just use the motor as a generator. PWM the low side transistors and the motor acts as the inductor in a boost converter.
http://techno-fandom.org/~hobbit/cars/boost-hack/

[max_torque]

generally, regen braking on a push bike (low mass) isn't cost effective.  IE the extra wear on your batteries outweighs the tiny saving in electricity recovered ('lecy is cheap, you can buy a whole kWHr for about 15p in the UK......)

On  a push bike, which tends to be ridden at a constant speed and typically coasts to a halt rather than brakes to halt the vast majority of energy (95%) goes to rolling friction and aero drag

[NiHaoMike]

There's little reason not to include regen when it's just a matter of a little code to implement it.

[Siwastaja]

IE the extra wear on your batteries outweighs the tiny saving in electricity recovered

This is a myth. Regeneration has been shown to increase, not decrease battery life, for example, in this paper: http://www.a3ps.at/site/sites/default/files/downloads/evs28/papers/A3-01.pdf

Of course, the effect will be quite small in a bike, but a brushed DC motor, "brushless DC" (permanent magnet AC) and induction AC motors all can implement regeneration in control software only, so there is little reason not to do it. It's easiest to start developing without it and implement it later.

[Halvmand]

That dynamo. Is that the kind that is used for the lights?
If so there vil be nothing useful to be gained using it as a generator. It won't generate enough power for you to notice. I guess they're in the order of 1 - 5 w or something.

[eneuro]

Next up is the battery it's self. You only get a finite number of charge / discharge cycles. The exact number depends on the depth of discharge, the load, the charge rate, and the particular battery you have.

Yep, with KERS http://www.magnetimarelli.com/excellence/technological-excellences/kers  you ar enot limited by finite number of chrage/discharge cycles, but... it is another story how to make ebike KERS