Regenerative braking on an e-bike.

Electrically assisted bikes, trikes, etc. that are legal in the UK
User avatar
Cugel
Posts: 5430
Joined: 13 Nov 2017, 11:14am

Regenerative braking on an e-bike.

Post by Cugel »

Having acquired a hybrid car that employs electric motor drag as the brake as well as a generator of battery juice, I was wondering if a bicycle version was possible - a motor that's also a dynamo, I suppose; employed to generate enough charge to slowly fill a battery from both downhill braking and flat cycling albeit a bit slower for one's pedal pushing efforts than normal.

Some clever Spaniards have come up with one:

https://ebiketips.road.cc/content/news/ ... -just-13kg

The ladywife has a raceyish Focus Paralane2, which gets her up the hills at my pace: a bike for bringing the slightly less fit or able up to the abilities of the more fit, then. No regenerative braking, though.

This Spanish e-bike seems to take a different approach. It's made to be efficient but obviously is no racing bike. However, as a means to level out the required efforts over hilly terrain, for those otherwise unable to manage the steeps unassisted, perhaps it has a place?

Wot U fink?

Cugel
“Practical men who believe themselves to be quite exempt from any intellectual influence are usually the slaves of some defunct economist”.
John Maynard Keynes
Marc
Posts: 89
Joined: 18 Aug 2016, 6:03pm
Location: Hamburg, Germany
Contact:

Re: Regenerative braking on an e-bike.

Post by Marc »

Every direct drive hub motor with the right motor controller can do regen braking. Its used for ages (BionX assist systems had it from the start).

However, compared to a light geared hub motor with internal freewheeling (no regen brake possible), the slight cogging of direct drive hub motors are a pain to pedal at some speed without power. Its kind like riding with a semi flat Marathon Plus. Definitely no fun to ride without battery.
I got various cycles with various different hub motors, so its easy for me to compare.
MikeDee
Posts: 745
Joined: 11 Dec 2014, 8:36pm

Re: Regenerative braking on an e-bike.

Post by MikeDee »

Some folks on the electric bike review.com forum that have used regenerative braking say it doesn't work that well because, on a bike, you are just not using the brake enough to make it worthwhile.
Cyril Haearn
Posts: 15215
Joined: 30 Nov 2013, 11:26am

Re: Regenerative braking on an e-bike.

Post by Cyril Haearn »

Not for you Cugel, you like to go fast downhill
I am a very timid descender, limiting my speed to 15 kmh or mph would store a lot of energy to get up the next hill, even if the machine is inefficient it should be possible, main thing is to be grid-free, should be a bit of energy over to heat my tea & soup too
Another advantage: extra weight slows one down most of the time :wink:
Entertainer, juvenile, curmudgeon, PoB, 30120
Cycling-of course, but it is far better on a Gillott
We love safety cameras, we hate bullies
User avatar
Cugel
Posts: 5430
Joined: 13 Nov 2017, 11:14am

Re: Regenerative braking on an e-bike.

Post by Cugel »

Cyril Haearn wrote:Not for you Cugel, you like to go fast downhill
I am a very timid descender, limiting my speed to 15 kmh or mph would store a lot of energy to get up the next hill, even if the machine is inefficient it should be possible, main thing is to be grid-free, should be a bit of energy over to heat my tea & soup too
Another advantage: extra weight slows one down most of the time :wink:


Ha ha - you're right - such a machine is not for me. But as a pure utilitarian mode of transport, I can see the appeal. A lot of people are put off cycling by hills but don't mind pedalling along quite slowly on the flat. Many are like you and fear the rapid descent, perhaps for good reason. Hawthorn hedges! Bandit tractors with huge hay bale prongs!!

This Spanish bike has a combined rear hub motor & battery. The battery can be charged with a plug to hold about 30km worth of go before regenerative braking or pedalling must be used to supply charge to power the motor. For many utilitarian journeys, even that small battery will do. The regenerative aspect will merely eke out the initial charge by recovering some rather than wasting heat via the brake, with the flat-pedalling regenration (one would hope) being a last resort to store up some go for the final 20% hill, to get up the last half mile home. :-)

This bike is like the hybrid car, which has a battery that holds about 30 miles per full charge - enough for most journeys here and there - but has a petrol engine to charge the battery if the journeys are longer and there's no rapid charging point available. In the bike, the cyclist is the petrol engine, sort of. (The cyclist is also the cyclist, of course).

Cugel
“Practical men who believe themselves to be quite exempt from any intellectual influence are usually the slaves of some defunct economist”.
John Maynard Keynes
PH
Posts: 13106
Joined: 21 Jan 2007, 12:31am
Location: Derby
Contact:

Re: Regenerative braking on an e-bike.

Post by PH »

Your link is a bit short on detail, it gives no indication of how much extra effort is required or how much assistance that offers. Research I read said regenerative braking is already pretty efficient, capturing around 70 - 80% of energy that would otherwise be lost. That research showed a car in typical driving conditions increasing it's range by up to 30%, but for an electric scooter it was low single digits 4 - 5%, pedal cycles were not included. Pedalling a bit harder on the flat for a bit of assistance on the hills sounds appealing, but I suspect it's more than a bit harder if the assistance is going to be meaningful. That 30% increase for a car is significant - it allows for a smaller battery, a reduction in charging costs and the convenience of longer journeys without charging. When it comes to bikes where the larger batteries weigh 2 -3kg adding 30% capacity before you start wouldn't be significant and the current range is already enough for most users or carrying an extra battery is a possibility in those circumstances it isn't.
BrightonRock
Posts: 153
Joined: 4 Apr 2019, 7:37am

Re: Regenerative braking on an e-bike.

Post by BrightonRock »

I am genuinely disappointed how slow the tech advances for battery and things like solar and regeneration are concerned. What's the problem? Is there simply not enough profit in ebikes as compared to electric cars?
Brucey
Posts: 44519
Joined: 4 Jan 2012, 6:25pm

Re: Regenerative braking on an e-bike.

Post by Brucey »

MikeDee wrote:Some folks on the electric bike review.com forum that have used regenerative braking say it doesn't work that well because, on a bike, you are just not using the brake enough to make it worthwhile.


this is more or less correct IMHO. If you think it might be worthwhile for your riding and you want to do a 'sanity check' on it then just analyse your braking habits. Make a note of

1) every full stop you come to and the speed you were going at beforehand

2) every time you have to drag the brakes going down a hill

In the latter case try to estimate what the gradient is, how long the brake is on for and what difference in speed the dragging brake is making.

You can then calculate how much energy you have available to 'harvest'. Note that if the motor is rated at (say) 500W peak, that is the most you can reasonably expect to harvest with it too. Heavy braking can dissipate a lot more than 500W (briefly) so don't expect to be able to get all the energy back again.

Big hills appear to offer a large 'harvest'; a 100m hill and a 100kg bike/rider has ~100kJ of PE. If you drag your 'e-brake' down the hill you might expect (because aero drag and rolling resistance still eat energy whenever you move) to be able to harvest about half of that (~ 50kJ) which is not to be sniffed at.

However you get losses in this process too. For example;

a) turning KE into electricity will probably be about 90% efficient at best (and could be a lot less)

b) charging batteries is usually about 70% efficient (volts x amps in vs volts x amps out)

c) the motor drive could be ~90% efficient (but might be a lot less).


So whatever 'crop' you might be able to harvest needs to be multiplied by 0.8 x0.7 x0.9 = ~0.57 before you get a realistic estimate of how much benefit you will get from it. If the motor/alternator is inherently lossy then it might be closer to 0.7 x 0.7 x 0.7 = ~0.34.

So you sit at the top of a hill with 100kJ in hand and you can expect to get about 27kJ 'back again' with a good system and ~15kJ with a less efficient one. In other words at best you are going to be able to go up to about a quarter of the way back up the hill again (actually much less than this because of the usual rolling resistance and aero losses) .

Another way of looking at it is that 27kJ is enough to run a 250W motor for just over 1-1/2 minutes. 15kJ is enough to run a 250W motor for one minute. It is something rather than nothing but in terms of how long it is going to take you to climb up the next 100m high hill (between six and ten minutes, I estimate) this isn't going to be a very big difference.

The same considerations apply to cars too. But a typical car with occupants is the thick end of two tonnes, which means that there is ~ x20 more energy available to harvest.

Solar panels on bikes need to be huge (huge enough for the bike to be a sail when the wind is blowing) in order to be useful. And the sun needs to be shining. Unless you are riding all day in the sunshine, it is probably better just to plug the bike into a (static) solar panel installation when you are not riding it.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
User avatar
Cugel
Posts: 5430
Joined: 13 Nov 2017, 11:14am

Re: Regenerative braking on an e-bike.

Post by Cugel »

Brucey wrote:
MikeDee wrote:Some folks on the electric bike review.com forum that have used regenerative braking say it doesn't work that well because, on a bike, you are just not using the brake enough to make it worthwhile.


this is more or less correct IMHO. If you think it might be worthwhile for your riding and you want to do a 'sanity check' on it then just analyse your braking habits. Make a note of

1) every full stop you come to and the speed you were going at beforehand

2) every time you have to drag the brakes going down a hill

In the latter case try to estimate what the gradient is, how long the brake is on for and what difference in speed the dragging brake is making.

You can then calculate how much energy you have available to 'harvest'. Note that if the motor is rated at (say) 500W peak, that is the most you can reasonably expect to harvest with it too. Heavy braking can dissipate a lot more than 500W (briefly) so don't expect to be able to get all the energy back again.

Big hills appear to offer a large 'harvest'; a 100m hill and a 100kg bike/rider has ~100kJ of PE. If you drag your 'e-brake' down the hill you might expect (because aero drag and rolling resistance still eat energy whenever you move) to be able to harvest about half of that (~ 50kJ) which is not to be sniffed at.

However you get losses in this process too. For example;

a) turning KE into electricity will probably be about 90% efficient at best (and could be a lot less)

b) charging batteries is usually about 70% efficient (volts x amps in vs volts x amps out)

c) the motor drive could be ~90% efficient (but might be a lot less).


So whatever 'crop' you might be able to harvest needs to be multiplied by 0.8 x0.7 x0.9 = ~0.57 before you get a realistic estimate of how much benefit you will get from it. If the motor/alternator is inherently lossy then it might be closer to 0.7 x 0.7 x 0.7 = ~0.34.

So you sit at the top of a hill with 100kJ in hand and you can expect to get about 27kJ 'back again' with a good system and ~15kJ with a less efficient one. In other words at best you are going to be able to go up to about a quarter of the way back up the hill again (actually much less than this because of the usual rolling resistance and aero losses) .

Another way of looking at it is that 27kJ is enough to run a 250W motor for just over 1-1/2 minutes. 15kJ is enough to run a 250W motor for one minute. It is something rather than nothing but in terms of how long it is going to take you to climb up the next 100m high hill (between six and ten minutes, I estimate) this isn't going to be a very big difference.

The same considerations apply to cars too. But a typical car with occupants is the thick end of two tonnes, which means that there is ~ x20 more energy available to harvest.

Solar panels on bikes need to be huge (huge enough for the bike to be a sail when the wind is blowing) in order to be useful. And the sun needs to be shining. Unless you are riding all day in the sunshine, it is probably better just to plug the bike into a (static) solar panel installation when you are not riding it.

cheers


A nice bit of maths to illuminate the subject, that.

How much regenerative braking on an e-bike is worth having will perhaps depend largely on the typical terrain rode and for how far, as well as the style of riding. I've been comparing a few months experience in an electric car with the experience of riding with the ladywife when she's on her e-bike. Here are some observations:

Careful driving and judicious selection of the amount of regenerative braking in the e-car (there are 6 levels inclusive of a "none") can make a big difference to the range of the battery. A 12 mile round-trip going up from 100M to 400M and back down again into Brechfa Forest to walk the collies sees the miles-left go from 30 to 14 on the way to the top and back up to 21 to 23 miles-left by the time we go back down to the start point.

The trick is to use the regenerative braking for all necessary braking on the way down, with the disc brakes used only to come to a stop from 2-3mph at any junctions. This also requires the max speed on the way down to be kept low, as above a certain speed a steeper bit would also require disc braking on top of max regenerative braking to stop the car accelerating under gravity to a speed too fast for the road.

****
On her e-bike, the ladywife accompanies me over similar climbs and descents. The descents are often quite long, continuous and complete with steep down-bits as well as the odd blip up for a few metres. She is often braking for significant periods to keep the speed safe. Were the bike to have regenerative braking as sophisticated as that of the car (i.e. with variable levels of braking, including "none") I believe she could get down with little or no use of her disc brakes - as long as the top speeds were kept relatively low to avoid run-away and hence a need to apply the disc brakes.

If such regenerative braking was similarly efficient to that of the car (and I presume it too suffers efficiency losses as you describe above) she would obtain a significant increase in the range of her battery. As it is, a lot of energy goes in heating the disc brakes and cooling the air they pass through, instead.

***
So perhaps the question is, how efficient and sophisticated could e-bike regenerative braking be made without it becoming horrendously complicated and expensive? There seems litte doubt that it would be worth doing if similar battery-range extension to that of the car could be obtained.

At least, it would in the very hilly terrain we ride here. On a flatter landsacpe, perhaps such braking would be redundant? I have no experience in the e-car to judge by as everywhere here (everywhere) is full of large hills.

Cugel
“Practical men who believe themselves to be quite exempt from any intellectual influence are usually the slaves of some defunct economist”.
John Maynard Keynes
Brucey
Posts: 44519
Joined: 4 Jan 2012, 6:25pm

Re: Regenerative braking on an e-bike.

Post by Brucey »

Cugel wrote: …..Careful driving and judicious selection of the amount of regenerative braking in the e-car (there are 6 levels inclusive of a "none") can make a big difference to the range of the battery. A 12 mile round-trip going up from 100M to 400M and back down again into Brechfa Forest to walk the collies sees the miles-left go from 30 to 14 on the way to the top and back up to 21 to 23 miles-left by the time we go back down to the start point.....


There are a few issues that may be clouding this picture;

1) the proportion of energy that is actually being used to climb (rather than just overcome rolling resistance and aero drag) and is therefore 'available' to harvest, and

2) what 'range' really means anyway.

On the latter point the 'range' calculation is made using some algorithm which takes an average of the previous miles driven. Sometimes it is an average of all the miles since the last reset, sometimes it is a rolling average.

So if you drive six miles to the top of the hill and the range reduces by 16 miles that might be a reflection of the range under the average conditions (of the last fifty miles or something) or the range if you carry on using energy at the same rate as for the last five miles. The same logic may apply going down the hill too, so the 'range' may be (rather optimistically) predicting that you may continue to use energy at the same rate as the last few (downhill) miles. Thus the 'range' indication may be saying more about the way it is calculated rather than how much energy is actually harvested.

One way of finding out a bit more about how the range calculation is done is to see if (on your 'hill route') the range changes in the same way regardless of the number of miles since the trip was reset or the battery was charged.

if you don't understand how the 'range' calculation is done, the thing that is most likely to be meaningful is the actual reduction in range. To determine this accurately it may be necessary to drive in a controlled fashion on the same (flat) route to establish the baseline range on a charge, then include hilly elements in the middle of the battery discharge, so that the miles are racked up on the flat, then in the hills, then on the flat again as the battery is discharged. That way the cutoff point in the battery discharge ought to be reached similarly in each case.

BTW if your average speed is different on the flat than in the hills, this may have as big an influence on the rate at which energy is used. Car tyres have (vs bicycle tyres) very poor Crr figures; @ about 50mph each tyre can have a good fraction of 1kW being dissipated in it, and in many cars the aero losses are about equal to the rolling resistance losses at some speed around 50mph. The aero drag force goes as the square of the speed, thus just driving more slowly greatly reduces the total aero losses in any one journey.

FWIW modern electric cars are necessarily overweight and un-aerodynamic; folk won't buy them unless they look like cars they are used to, have similar crash safety etc. This means that even a smallish EV hatchback is liable to weigh the thick end of two tonnes, because it will have half a tonne or batteries in it if it is to have any real range. Massive changes in the efficiency of the vehicle can be made if it is made lighter and more aerodynamic; most of the energy is used currently in shifting about tonne and a half of extra stuff around the countryside, rather than the passengers.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
User avatar
Cugel
Posts: 5430
Joined: 13 Nov 2017, 11:14am

Re: Regenerative braking on an e-bike.

Post by Cugel »

Brucey wrote:
Cugel wrote: …..Careful driving and judicious selection of the amount of regenerative braking in the e-car (there are 6 levels inclusive of a "none") can make a big difference to the range of the battery. A 12 mile round-trip going up from 100M to 400M and back down again into Brechfa Forest to walk the collies sees the miles-left go from 30 to 14 on the way to the top and back up to 21 to 23 miles-left by the time we go back down to the start point.....


There are a few issues that may be clouding this picture;

1) the proportion of energy that is actually being used to climb (rather than just overcome rolling resistance and aero drag) and is therefore 'available' to harvest, and

2) what 'range' really means anyway.

On the latter point the 'range' calculation is made using some algorithm which takes an average of the previous miles driven. Sometimes it is an average of all the miles since the last reset, sometimes it is a rolling average.

So if you drive six miles to the top of the hill and the range reduces by 16 miles that might be a reflection of the range under the average conditions (of the last fifty miles or something) or the range if you carry on using energy at the same rate as for the last five miles. The same logic may apply going down the hill too, so the 'range' may be (rather optimistically) predicting that you may continue to use energy at the same rate as the last few (downhill) miles. Thus the 'range' indication may be saying more about the way it is calculated rather than how much energy is actually harvested.

One way of finding out a bit more about how the range calculation is done is to see if (on your 'hill route') the range changes in the same way regardless of the number of miles since the trip was reset or the battery was charged.

if you don't understand how the 'range' calculation is done, the thing that is most likely to be meaningful is the actual reduction in range. To determine this accurately it may be necessary to drive in a controlled fashion on the same (flat) route to establish the baseline range on a charge, then include hilly elements in the middle of the battery discharge, so that the miles are racked up on the flat, then in the hills, then on the flat again as the battery is discharged. That way the cutoff point in the battery discharge ought to be reached similarly in each case.

BTW if your average speed is different on the flat than in the hills, this may have as big an influence on the rate at which energy is used. Car tyres have (vs bicycle tyres) very poor Crr figures; @ about 50mph each tyre can have a good fraction of 1kW being dissipated in it, and in many cars the aero losses are about equal to the rolling resistance losses at some speed around 50mph. The aero drag force goes as the square of the speed, thus just driving more slowly greatly reduces the total aero losses in any one journey.

FWIW modern electric cars are necessarily overweight and un-aerodynamic; folk won't buy them unless they look like cars they are used to, have similar crash safety etc. This means that even a smallish EV hatchback is liable to weigh the thick end of two tonnes, because it will have half a tonne or batteries in it if it is to have any real range. Massive changes in the efficiency of the vehicle can be made if it is made lighter and more aerodynamic; most of the energy is used currently in shifting about tonne and a half of extra stuff around the countryside, rather than the passengers.

cheers


That's all meaningful. The estimated mileage left in the battery that's displayed to the driver of the car is known amongst the owners of such cars as the guessometer - for good reasons, many of which you describe.

But over dozens of similar up & down trips of the same route to the start of various collie-walks in Brechfa Forest, the guessometer does give consistent estimates. Moreover, subsequent journeys tend to confirm that it's guesses are not out by large amounts, even if the journeys vary a bit. Typically another, later, out & back somewhere else on the same day will give the mileage the guessometer suggests is left +/- about 15% before the battery runs out and the petrol engine starts.

****

But this to the side. The point I was trying to make was that e-cars - despite their high weight and drag from tyres and air, can be configured to put significant amounts of charge back into the battery from regenerative braking down significant amounts of hills. Why can't it be the same for an e-bike?

As I mentioned, there may be cost and complexity issues for an e-bike of making them to provide regenerative braking as sophisticated as that of an e-car. There may be other reasons too, such as a high weight penalty for all the necessary gubbins.....? No one seems to have yet made such a sophisticated regenerative system for an e-bike. But is anyone trying, I wonder?

*****
Today the ladywife et moi have just done a 30 miler from near Lampeter to Brechfa village, up and over the top of Brechfa Forest twice: just over 1000M of climbing and descent. The lanes are narrow, twisty and often awash with loose gravel, even though they are generally smooth tarmac underneath and pothole-free. There are many very steep descents (and ascents!) that are also quite long. One must do lots of braking.

I don't know how much heat we both made with the disc brakes but we did spent a significant amount of time doing so. There are nine black arrows on the route. There are two long, long descents of around seven kilometers each (from 360/380M to 72M respectively). I can't help feeling that regenerative braking on her e-bike would put a useful percentage back in the battery, on such a ride. And save brake shoe, disc and tired hands as well.

Cugel
“Practical men who believe themselves to be quite exempt from any intellectual influence are usually the slaves of some defunct economist”.
John Maynard Keynes
rfryer
Posts: 809
Joined: 7 Feb 2013, 3:58pm

Re: Regenerative braking on an e-bike.

Post by rfryer »

I think that the number of cyclists that might benefit from decent regenerative braking may be so small that it's not worth implementing.

I live in an abnormally hilly area, but it's still the case that I very rarely need to brake unless approaching some kind of obstacle such as a tight corner or junction. The vast majority of the time, almost all of my potential energy is consumed by wind resistance.

I can imagine an effective system of regenerative braking where the bike harvested power whenever you tried to exceed 15mph. This would dramatically increase the opportunities to recharge, and I'd expect to see a measurable increase in range. However, I feel it would steal all the joy out of cycling for me.
User avatar
Cugel
Posts: 5430
Joined: 13 Nov 2017, 11:14am

Re: Regenerative braking on an e-bike.

Post by Cugel »

rfryer wrote:I think that the number of cyclists that might benefit from decent regenerative braking may be so small that it's not worth implementing.

I live in an abnormally hilly area, but it's still the case that I very rarely need to brake unless approaching some kind of obstacle such as a tight corner or junction. The vast majority of the time, almost all of my potential energy is consumed by wind resistance.

I can imagine an effective system of regenerative braking where the bike harvested power whenever you tried to exceed 15mph. This would dramatically increase the opportunities to recharge, and I'd expect to see a measurable increase in range. However, I feel it would steal all the joy out of cycling for me.


There are so many modes of cycling, though. Despite what some think is the speed-inducing nature of an e-bike, most e-bikers potter about at lowly speeds on 40-50llb beasts. They also want to potter down the hills, rather than fly down like thee & me. :-) Regenerative braking would suit many, I feel, as they could also avoid tiring their wee mits.

ALthough the ladywife is improving her downhilling all the time, she is still very cautious on the many slopes at which I could and would go down at 40mph. After all, she doesn't have my near 60 years of practice. When I go out with her, I match my donhill speed to hers and that means a fair amount of btraking - typically to stay at a max of 25mph instead of doing that 40mph. She could be recharging.....

********
There are other issues. For example, a hub motor in an e-bike is presumably relatively easy to employ as a regenrative brake - like a rather draggy (but variable) hub dynamo. But what about BB motors; or the Fazua system and similar that has the motor in a bike frame tube driving the BB? You can't use such motors to directly brake the bike. You'd still need some form of hub brake that generated e-juice to the battery. That's more weight and complexity, as you's still need the friction brakes even if they get used less.

Cugel
“Practical men who believe themselves to be quite exempt from any intellectual influence are usually the slaves of some defunct economist”.
John Maynard Keynes
kwackers
Posts: 15643
Joined: 4 Jun 2008, 9:29pm
Location: Warrington

Re: Regenerative braking on an e-bike.

Post by kwackers »

The energy you'd get back from a bicycle even in a hilly area might be enough to get you up your drive after several tens of miles of riding.

In start stop traffic heavy cars benefit fairly well from it, but unless you're the sort of person that is always dabbing at your bike brakes then you'll get nothing and if you are that person then learn to read ahead and use brakes only as a last resort - saving the energy in the first place is far better than recouping it inefficiently later.
Mykidsdad
Posts: 11
Joined: 1 Jun 2019, 11:04pm

Re: Regenerative braking on an e-bike.

Post by Mykidsdad »

My Radwagon electric cargo bike from the USA (allegedly) has a direct drive rear motor and regenerative braking..
Post Reply