E-bike Generator Challenge

General Discussion about electric bicycles.
Post Reply
energyi   10 W

10 W
Posts: 86
Joined: Oct 16 2012 5:30pm

E-bike Generator Challenge

Post by energyi » Oct 27 2018 12:33pm

Challenge your thinking some by envisioning a generator at the bottom bracket of bikes. No chain, no sprocket, no derailer, no shifters, no cassette. Only an electric motor, battery, controller (with high-efficiency regenerative braking) wires, throttle, and a generator. The E bike would evolve in an interesting way.

Is anyone doing this already? Thanks in advance for any discussion.
HS3540 20" on Terratrike Tour Recumbent, LifePO4 48V 20 amp (was 12S2P Lipo), Greentime controller w/regen. Denver Area.

docw009   10 kW

10 kW
Posts: 901
Joined: Aug 02 2015 7:43am
Location: Chicago area suburbs.

Re: E-bike Generator Challenge

Post by docw009 » Oct 27 2018 3:37pm

You got the footloose. Not for me.

http://www.fullycharged.com/Mando-Footloose-IM

dustNbone   10 kW

10 kW
Posts: 713
Joined: Mar 13 2016 9:59pm
Location: Vancouver, BC

Re: E-bike Generator Challenge

Post by dustNbone » Oct 27 2018 4:00pm

It's not technically complicated at all, but it's a terribly inefficient way to transfer power from pedals to wheel. We have already invented far better systems for that.

energyi   10 W

10 W
Posts: 86
Joined: Oct 16 2012 5:30pm

Re: E-bike Generator Challenge

Post by energyi » Oct 27 2018 6:51pm

dustNbone wrote:
Oct 27 2018 4:00pm
...it's a terribly inefficient way to transfer power from pedals to wheel.
I would have to disagree. A generator can be 97% efficient, transfer the power directly to the electric motor or store it in a battery if not immediately needed (storage has lower roundtrip efficiency). 100 watts of power optimized for the humans peddle rpm and ergonomics makes lots of sense. This makes as much sense as a mechanical gear, cassette and chain scenario. The cost reduction should be significant, weight reduction marginal. Optimization of rpm for exact cadence will increase overall system efficiency. For people who can't pedal, but use a hand crank, this is a no-brainer better scenario than the current expensive mechanical solutions available.

energyi
HS3540 20" on Terratrike Tour Recumbent, LifePO4 48V 20 amp (was 12S2P Lipo), Greentime controller w/regen. Denver Area.

AHicks   10 W

10 W
Posts: 86
Joined: Jul 24 2018 10:53am

Re: E-bike Generator Challenge

Post by AHicks » Oct 27 2018 7:22pm

As one that agrees regarding the inefficiencies of a system like that, but smart enough to know better than say it can't be done, I'll be waiting for your proof of concept model. Hopefully on display prior to the day I end up pushing up daisies.

User avatar
wturber   100 kW

100 kW
Posts: 1481
Joined: Aug 23 2017 8:52pm
Location: Fountain Hills,AZ
Contact:

Re: E-bike Generator Challenge

Post by wturber » Oct 27 2018 9:44pm

energyi wrote:
Oct 27 2018 6:51pm
dustNbone wrote:
Oct 27 2018 4:00pm
...it's a terribly inefficient way to transfer power from pedals to wheel.
I would have to disagree. A generator can be 97% efficient, transfer the power directly to the electric motor or store it in a battery if not immediately needed (storage has lower roundtrip efficiency). 100 watts of power optimized for the humans peddle rpm and ergonomics makes lots of sense. This makes as much sense as a mechanical gear, cassette and chain scenario. The cost reduction should be significant, weight reduction marginal. Optimization of rpm for exact cadence will increase overall system efficiency. For people who can't pedal, but use a hand crank, this is a no-brainer better scenario than the current expensive mechanical solutions available.

energyi
97% seems unlikely. Heck, 98% is pretty much the best case scenario for a crank and chain. So the mechanical aspects of your generator scenario would probably cost ya 3% before you ever move an electron. A quick look online shows generator efficiencies peaking at around 91%. Converting the generated electricity back to torque would have its own inefficiencies. Typical motor efficiencies are around 85%. I seems really unlikely that you'd be able to match a simple chain and sprocket setup for efficiency. Certain you wouldn't be able to match by cost - otherwise most bikes would be built this way.

There is an ES member who could probably give a lot of practical information on this. I forget his name, but his project is the Electrom. That bike uses a generator only pedal drive after the bike reaches 6 mph. Up to 6 mph, it uses a mechanical coupling as well as the generator. I'm betting he'd be a good resource on net efficiencies.

http://www.electrom.ca/

User avatar
amberwolf   100 GW

100 GW
Posts: 25982
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: E-bike Generator Challenge

Post by amberwolf » Oct 27 2018 9:51pm

There are a number of bikes and trikes that have been made this way here on ES, and others elsewhere. Sometimes it's called a Serial Hybrid Cycle.

The first one I recall, Lowracer, unfortunately no longer posts here, and all of his videos are gone but many of the pics are still there.

Justin_LE's solar challenge trike does this at the rowing end, IIRC.

There's a few that show up in this search, but there's more that don't:
search.php?st=0&sk=t&sd=d&sr=topics&key ... l*&start=1

billvon   10 MW

10 MW
Posts: 2272
Joined: Sep 16 2007 9:53pm
Location: san diego

Re: E-bike Generator Challenge

Post by billvon » Oct 27 2018 9:59pm

energyi wrote:
Oct 27 2018 6:51pm
I would have to disagree. A generator can be 97% efficient
It can be, but bike generators won't be. You get those 97% numbers for utility scale generators at one operating point. These monsters operate at one speed all the time, and are only 97% efficient at one power level (usually maximum.) They can get there because weight and size (and sometimes even cost) are no object.

For a bike generator, you have to gear it up to run at a reasonable speed. That's another loss. Or you can run directly off the pedals, but then you are designing a VERY low RPM generator with huge magnets, and your eddy current losses go way up. You would be lucky to hit 80-85% efficiency with one.

Next you have to convert AC to DC. You can do it with diodes, easy. Diodes guarantee a loss of between .5 and 1 volt. On a 48 volt system that's another loss. Call that 98% efficient.

Next you have to charge batteries. Lithium ion batteries are very efficient coulombically - for every amp hour you put in you can get an amp hour out. But they are not efficient energy wise, because you are charging at 4.2 volts per cell and discharging at 3.8 or so. So that's another 90% efficient step.

Next you hit the motor controller. You can do pretty well with these - 95% would not be unusual for a good, well-matched controller.

Finally, at the motor, you can see 85% peak efficiency. But for a bike ridden in the real world of hills and changing loads, your average efficiency is going to be closer to 75%.

So your overall efficiency is: .80*.98*.90*.95*.75 = 50% efficiency. Now, you could improve that with custom designed (and expensive, and heavy) components. But let's say you're rich, and do an awesome design job and get that to 70%. Do you really want to throw away a third of your pedaling power? I don't know about you, but I need all the legpower I've got when I am on a conventional bike.
For people who can't pedal, but use a hand crank, this is a no-brainer better scenario than the current expensive mechanical solutions available.
Justin did a pretty good job with one for not a lot of $$. The commercial ones out there run $1500 and up - but your solution would likely be a lot more expensive.

User avatar
wturber   100 kW

100 kW
Posts: 1481
Joined: Aug 23 2017 8:52pm
Location: Fountain Hills,AZ
Contact:

Re: E-bike Generator Challenge

Post by wturber » Oct 27 2018 10:28pm

amberwolf wrote:
Oct 27 2018 9:51pm

Justin_LE's solar challenge trike does this at the rowing end, IIRC.


If you mean the Sun Trip trike, I don't think so. The rowing mechanism connects directly to the main crank drive. I recall Justin talking about trying to match the drum diameter of the pull cable so that the rowing motion would tend to match up well with the standard crank cadence.


User avatar
amberwolf   100 GW

100 GW
Posts: 25982
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: E-bike Generator Challenge

Post by amberwolf » Oct 27 2018 11:04pm

Ah, well, originally he'd planned for a rowing-end generator; its' even still in the title. ;) I didn't realize (or at least, didn't remember) that it didn't work out that way.

User avatar
MadRhino   100 GW

100 GW
Posts: 5601
Joined: Sep 03 2010 5:28pm
Location: Montreal QC Canada

Re: E-bike Generator Challenge

Post by MadRhino » Oct 29 2018 4:43am

I recall one ES member had done it on a faired recumbent, 2 wheel velomobile. I agree cranking a generator is terribly inefficient as compared to a bicycle drive train, not only in energy conversion losses. The extra weight has to be part of the equation too.

Yet, I have seen worse ideas. Like this guy who wanted to regen the front wheel to power the rear. :mrgreen:
Make it fool-proof, and I will make a better fool.
Current bikes
Street: https://s20.postimg.org/ewrvugywt/Session_04_2015.jpg
Dirt: https://s20.postimg.org/lbqwr55ml/IMG_0157.jpg

User avatar
dogman dan   100 GW

100 GW
Posts: 34543
Joined: May 17 2008 12:53pm
Location: Las Cruces New Mexico USA

Re: E-bike Generator Challenge

Post by dogman dan » Oct 29 2018 7:16am

It does have a few limited advantages, but efficiency is not one of them.

You could charge the battery some while stationary, like at many stop lights, or keeping the DJ powered up at a party, or charging a house battery off grid, at night.

There are several versions of e bike, that can also charge a battery (just not at each stop light), it just requires lifting the wheel off the ground, not a bottom bracket made into generator.

energyi   10 W

10 W
Posts: 86
Joined: Oct 16 2012 5:30pm

Re: E-bike Generator Challenge

Post by energyi » Nov 02 2018 2:17pm

Great discussion, thanks for all the insights. The Mando Footloose and the Electrom are great examples of what I was requesting. Thanks. ( http://www.fullycharged.com/Mando-Footloose-IM and http://www.electrom.ca/ )

These designs will evolve and bottom bracket generators will become more efficient. I like that the Electrom is beginning to utilize the advantages of the evolution to a generator, lower profile and fairings. Eventually I would think that the ergonomics for the most efficient positioning for a constant wattage generation, for the distance, will be a big driver for overall system efficiency. Again, this would be for a bike or trike that already has an EV motor; already has a motor controller, battery, throttle, wires. And I would envision the generator to be DC, not AC and that the DC would go directly to the motor when it is needed eliminating storage losses.

Thanks again for the discussion.

energyi
HS3540 20" on Terratrike Tour Recumbent, LifePO4 48V 20 amp (was 12S2P Lipo), Greentime controller w/regen. Denver Area.

billvon   10 MW

10 MW
Posts: 2272
Joined: Sep 16 2007 9:53pm
Location: san diego

Re: E-bike Generator Challenge

Post by billvon » Nov 02 2018 2:24pm

energyi wrote:
Nov 02 2018 2:17pm
And I would envision the generator to be DC, not AC and that the DC would go directly to the motor when it is needed eliminating storage losses.
All generators are AC. You can convert them to DC via a mechanical commutator, electronic commutator or diodes.

Note that bicycle motors are AC as well. The common term BLDC stands for brushless DC - but requires an external controller to convert the DC battery voltage to the variable frequency AC that the motor windings actually need.
--bill von

energyi   10 W

10 W
Posts: 86
Joined: Oct 16 2012 5:30pm

Re: E-bike Generator Challenge

Post by energyi » Nov 02 2018 5:13pm

Permanent Magnet DC Generators. My hub motor regens DC.
HS3540 20" on Terratrike Tour Recumbent, LifePO4 48V 20 amp (was 12S2P Lipo), Greentime controller w/regen. Denver Area.

User avatar
eee291   10 kW

10 kW
Posts: 734
Joined: Nov 05 2015 4:35pm
Location: Austria, Burgenland: Güssing

Re: E-bike Generator Challenge

Post by eee291 » Nov 02 2018 5:23pm

energyi wrote:
Nov 02 2018 5:13pm
Permanent Magnet DC Generators. My hub motor regens DC.
In this case, the Carbon brushes (mechanical commutator) rectify the AC Voltage.

billvon   10 MW

10 MW
Posts: 2272
Joined: Sep 16 2007 9:53pm
Location: san diego

Re: E-bike Generator Challenge

Post by billvon » Nov 02 2018 5:36pm

energyi wrote:
Nov 02 2018 5:13pm
Permanent Magnet DC Generators. My hub motor regens DC.
That's because there are either commutators or rectifiers converting the generator's AC to DC.
--bill von

User avatar
neptronix   100 GW

100 GW
Posts: 13167
Joined: Jun 15 2010 5:56pm
Location: California refugee living in Utah, USA
Contact:

Re: E-bike Generator Challenge

Post by neptronix » Nov 02 2018 6:54pm

A motor that is 97% efficient is very expensive. Ever seen what those kinds of motors cost? :shock:
Then you're gonna lose a few % in the controller both ways unless you use an expensive controller.

So this is a very expensive way to have less efficiency than a string of metal hooks and pins..
If you eliminate the bike chain entirely then you need another motor on the rear.. and since you're not using gears, then you need an oversized motor to take hill climbs and efficiency will be nothing like 97% on the drive side. More like 80.. 90%

Now you have a ~$3000 bike chain that's far less efficient than a real one.

A better option for a 'hybrid' bike is a tiny DD hub with 0.2mm laminations and ultra ultra tight windings so that it only gives you ~1nm of losses from the iron lamination stack.. pair it with the most expensive low friction bearings.. you can build this mythical tiny DD for about $500.

..or you build a geared motor with an electrically actuated clutch so that when you hit the brake, you get regen. When you hit the power, you get power.. otherwise you are freewheeling.

The expense of the custom motor could be offset by using a smaller battery since this is a hybrid bike, right?
Allow the user to add a bigger one on..
My first major build: 8T MAC motor on a Trek 4500.
The new all-arounder: Leafmotor 1500w on a Turner O2 full suspension.

billvon   10 MW

10 MW
Posts: 2272
Joined: Sep 16 2007 9:53pm
Location: san diego

Re: E-bike Generator Challenge

Post by billvon » Nov 02 2018 8:23pm

neptronix wrote:
Nov 02 2018 6:54pm
A motor that is 97% efficient is very expensive. Ever seen what those kinds of motors cost?
Yep! I worked with a Sunraycer team for a bit. $5000 for a motor isn't unreasonable when it means the difference between second and third place - but it's not something that most people are going to be interested in.
--bill von

User avatar
neptronix   100 GW

100 GW
Posts: 13167
Joined: Jun 15 2010 5:56pm
Location: California refugee living in Utah, USA
Contact:

Re: E-bike Generator Challenge

Post by neptronix » Nov 02 2018 9:03pm

Sounds like a very fun project.
It's amazing what's required to squeak those last few percentage points out of an electric motor.

Lucky for us, there's emrax motors today and they don't cost a bundle and a half.
http://emrax.com/wp-content/uploads/201 ... l_data.pdf
But making a small super efficient bike sized motor is apparently a huge and expensive challenge.
My first major build: 8T MAC motor on a Trek 4500.
The new all-arounder: Leafmotor 1500w on a Turner O2 full suspension.

User avatar
dogman dan   100 GW

100 GW
Posts: 34543
Joined: May 17 2008 12:53pm
Location: Las Cruces New Mexico USA

Re: E-bike Generator Challenge

Post by dogman dan » Nov 03 2018 7:03am

The limited efficiency is still worth it, in specific applications. Example, pedal your e bike to charge a small house battery, that gives you one light at night, and charges your cell phone or other small battery devices. This works well in the third world, where house power might not exist at all, or cannot be affordable yet. Presumably in the am, you might be able to charge the e bike at work, or some other place away from home. Pedal gen might suck balls, but not as much as missing that important call, like you get to work today.

StuRat   100 W

100 W
Posts: 121
Joined: Mar 15 2018 7:18pm

Re: E-bike Generator Challenge

Post by StuRat » Nov 03 2018 3:12pm

dogman dan wrote:
Nov 03 2018 7:03am
The limited efficiency is still worth it, in specific applications. Example, pedal your e bike to charge a small house battery, that gives you one light at night, and charges your cell phone or other small battery devices. This works well in the third world, where house power might not exist at all, or cannot be affordable yet. Presumably in the am, you might be able to charge the e bike at work, or some other place away from home. Pedal gen might suck balls, but not as much as missing that important call, like you get to work today.
This is known in the industry as the 'first watt'. The first watt is the most valuable to the consumer, and also the seller. Like an addictive drug, 'the first one's free' then they're hooked, they want/need more. We are all power addicts. We're always trying to squeeze out the final few watts... in the name of efficiency.

User avatar
MadRhino   100 GW

100 GW
Posts: 5601
Joined: Sep 03 2010 5:28pm
Location: Montreal QC Canada

Re: E-bike Generator Challenge

Post by MadRhino » Nov 03 2018 6:03pm

To pedal for charging is only making sense when you need the energy for some other purpose than moving. Why would one want to pedal more energy and carry more weight than watts required for the ride? Masochism or physical exercise addiction?
Make it fool-proof, and I will make a better fool.
Current bikes
Street: https://s20.postimg.org/ewrvugywt/Session_04_2015.jpg
Dirt: https://s20.postimg.org/lbqwr55ml/IMG_0157.jpg

User avatar
wojtek   10 kW

10 kW
Posts: 965
Joined: Nov 29 2009 3:52pm
Location: Luxembourg
Contact:

Re: E-bike Generator Challenge

Post by wojtek » Feb 17 2019 1:41pm

I think it is great idea for those who are on high powered ebikes who still want to have good excersice.

Personally i am looking for a system that i could install in my bike using standard bottom bracket in my velomobile. The chain is so long and after fitting hubmotor in the rear wheel it is hard to make a smooth chainline so it doesn't rub anywhere making a lot of noise...

Any suggestions would be welcome. I don't give a damn about efficiency, just want to have more or less natural resistance when pedalling the generator.
WIRES AND CONNECTORS
http://www.endless-sphere.com/forums/vi ... 31&t=30176
http://endless-sphere.com/forums/viewto ... 31&t=32244

KMX Cobra with Xlyte 5403
Go-One Evo with Cromotor
BATMOBILE - Astro 3220 / Nuvinci in progress
Giant CRS Alliance - yes, just pushbike! frock you..
3Element Espire - SOLD
Custom built Specialized BigHit with BBS02 - SOLD
Velokraft e-VK3 SOLD
Steintrike Mungo Sport HS3540-SOLD
Scorpion FS HS3540-SOLD
Kona Dawg BMC v2-SOLD

markz   100 GW

100 GW
Posts: 6572
Joined: Jan 09 2014 11:38pm
Location: Alberta Canada

Re: E-bike Generator Challenge

Post by markz » Feb 17 2019 9:32pm

If you are going long distance, I would try to find a 500W mini generator running off of gas. Find a generator that will be sufficient for the ac to dc charger, I know my MW HRP wanted 4.22A from the 120V wall. They probably run 35cc motors, or could be 50cc. There is the added weight of the mini generator but knowing as you are riding, the gas geny is charging up your spare battery. This sure beats trying to find a power outlet on long distance rides. You could easily dim the sound by having a box, with the open end facing backwards. On another idea, there are generators that are built to run quiet, like the Honda i-series that others have adopted to indicate their quietness.

Hey its just an idea, might be worthwhile for someone thinking about doing a highway e-ride.


For city riding, build a battery that can handle lots of charging amperes and buy appropriate charger. The LiFePO4 26650 cells can handle huge currents. The limit is the 120V outlet and the current pumped out of the charger, and hopefully you dont trip the circuit breaker on the outlet you are juicing up on.
This means you can stop off at a Starbucks, juice up in 30 minutes while having your fancy Starbucks name sugary diabetes drink.

Post Reply