Free Wheeling or Not in ebike?
- Thread starter The7
- Start date
Rassy
1 MW
Free-wheeling: Less wear on brushes and/or bearings when not using motor, such as coasting, or pedaling only. Easier to coast or pedal when not wishing to use the motor.
Non free-wheeling: Normally simpler and cheaper to produce. Allows for regenerative braking.
And of course the positives for the one are negatives for the other. There are probably other issues?
Non free-wheeling: Normally simpler and cheaper to produce. Allows for regenerative braking.
And of course the positives for the one are negatives for the other. There are probably other issues?
Agreed. The coasting speed could be much faster than the no-load speed of the motor. However, mechanical braking is required to keep the bike to stay at the no-load speed when going down slope. And there is extra wear in the free wheeling part.Rassy said:
Reversing is possible.Rassy said:
Even without regerative braking, there will be an regenerative effect when the ebike tends to move faster than the no-load speed. This regenerative effect will charge back the battery through the internal diodes of the FETs when the bike speed is higher that of the no-load speed. There is a probability that this regenerative current is too high to burn out the FETs.
Any other pros and cons?
xyster
10 MW
The7 said:
Unless I'm missing something, regenerative braking is possible with a hubmotor with an external freewheel; but not possible with a chain-drive with a freewheel.
xyster said:
Sorry. This topic should be referred to the free wheeling with respect to the motors.
For example, hub motors with internal free-wheel (Puma hub motor and IZIP hub motors).
Or non-hub motor with external free-whee.
There is always a free wheel for the pedals but may not for the motor.
Rassy
1 MW
I agree with The7 that reverse is also a possibility with a non free-wheeling motor. But you almost confused me xyster. Aren't all production hub motors either internally free-wheeling or not internally free-wheeling? And unless you are Matt Gruber or Safe (No pedals) you would always have a free-wheel attached for your pedaling if you put a rear hub motor on your bike. So the issue of regenerative braking has nothing to do with the external "pedaling" freewheel but with the internal design of the hub motor.
However The7, you raised the issue of a regenerative effect when the ebike moves faster than the no-load speed. I was of the belief that there was no feedback allowed through my basic WE BD36 controller. I have only coasted down hills that push me about 5 MPH over no load speed and never noticed any effect. Can anyone shed any light on this issue? Would it happen if I opened the throttle while coasting?
However The7, you raised the issue of a regenerative effect when the ebike moves faster than the no-load speed. I was of the belief that there was no feedback allowed through my basic WE BD36 controller. I have only coasted down hills that push me about 5 MPH over no load speed and never noticed any effect. Can anyone shed any light on this issue? Would it happen if I opened the throttle while coasting?
xyster
10 MW
Rassy said:
I could have worded that better...
Logically, all hub motors are, like you said, either internally free-wheeling or not internally free-wheeling.
No, not necessarily. There are non-freewheeling, fixed rear sprockets available that are popular in some quarters. I haven't seen anybody use one with a hubmotor, but there's no reason I know of that one of these fixed sprockets wouldn't work, though the pedals would always be spinning.
http://en.wikipedia.org/wiki/Fixed-gear_bicycle
I'm having some trouble visualizing the operation, but it appears to open another option for regen -- a small brushed motor on the chain, connected to the chain by a freewheel, while the rear hubmotor is attached to the chain by a fixed sprocket. The brushed regen motor could have it's own little controller to PWM the voltage back to the batteries.
Rassy
1 MW
Thanks xyster. I actually thought of this option but dismissed it as not really applying to e-bikes. I think they are called fixee's and are popular with bike messengers. They sometimes have no brakes and stop by pedaling backwards, like a little kids trike. A gal in Portland recently got a ticket for no brakes and was fighting the ticket by trying to prove she could stop quicker with leg power then with brakes.
xyster
10 MW
Rassy said:
I'll remember that, since I ride around Portland.
Do you know why she was stopped? Did she convince the judge?
Rassy
1 MW
Lowell's spelling of fixie is correct, thanks Lowell.
Google "bicycle brakes fixie portland" and you will get a bunch of entries such as the one below.
http://www.cyclelicio.us/2006/07/portland-judge-brakeless-fixie-is-bad.html
Google "bicycle brakes fixie portland" and you will get a bunch of entries such as the one below.
http://www.cyclelicio.us/2006/07/portland-judge-brakeless-fixie-is-bad.html
Rassy said:
Assume the ebike has a dc motor without free wheeling.
No-load speed is the speed when the drive wheel off the ground with full throttle.
Max speed on flat is about 80-90% of the no-load speed.
WE DB36 is a bushed motor unit.
If you are coasting down hill with a speed faster than the no-load speed, the back emf (E)generated by the motor is higher than the battery voltage (V).
The back emf will force current (I) back into the battery (charging the battery) against a resistive path (R).
This resistance of the resistive path is composed of the winding resistance , the resistance of the internal diode of the FET and internal resistance of the battery.
Then the current I = (E-V)/R.
So this charging current I depends the over-speed.
Note that this charging current remains the same even if you turn on the throttle because the back emf E is higher than the battery voltage V. This means that there is partically no change in the over-speed.
It would be speculated that <20% over-speed would give a charging current within the safe limit.
What are your no-load speed and max speed on flat?
If your no-load speed is 25 mph, then 5 mph should not be a problem.
If it is more than 20% overspeed, it is wise to use the mechanical braking for personal safety as well as electrical safety.
Note that if you turn off the battery by mechanical switch, your downhill speed would be much higher. I guess that it would be at least 10 mph more.
Note that you could also prevent this charging current if a power diode (20-35A) is connected in series with the battery and the controller. But this will increase the forward volt dop by about 1 volt and decrease the max speed.
xyster
10 MW
Rassy said:
Thanks Rassy. I read the posted testimony. Interesting case.
Rassy
1 MW
Thanks The7, I appreciate the information as I don't know much about the inner workings of controllers, etc. The no-load speed (20" tire) is about 23 MPH and top speed on the flat is just over 20 MPH. Current top coasting speed has been just under 28 MPH, which is fast enough since the trike, 3 hub motors, trailer, batteries, and me weigh in at about 430 pounds. I have a couple of short hills that I do coast at 27+ MPH, and next time I'm in that area I'll put my power/brake toggle to the neutral position (which opens relays between the motors and their respective controllers without closing the braking relays) and see what happens speed wise. So if someone had a lighter bike with a motor like mine and wanted to pedal faster then the no-load speed for a while the back emf would start to work against him more then the slight magnetic resistance?
mi7d1
100 W
xyster said:Rassy said:
I'll remember that, since I ride around Portland.
I don't want to hijack the thread to long but xyster, since you ride around Portland, at least occasionally, if your going to be in the SE area PM and I'll be happy to show you the velomobile. I live around SE 82nd and Foster rd. Same for you Rassy, if your ever in my neck of the woods PM me. The7 not sure if your in BC or WA but the welcome mat is out for you also
I have an electric bike shop a few blocks from my hours where I bought my electrics from but doing business there sometime feels like pulling teeth.
xyster
10 MW
The7 said:
The Currie is a little bit easier to pedal at speeds over 10mph -- I can pedal the Currie about 1-2mph faster. There may be other factors involved besides the hubmotor's small bit of cogging torque:
-The X5's rear tire is 1/2" wider, and made of softer rubber
-The 25lb hubmotor is more than 20lbs heavier than the Currie's tiny motor, and of course as it's also a spinning hub, like a flywheel there's also extra angular inertia to overcome when accelerating.
Below 10mph, the lower gearing on the X5 makes it easier to pedal -- the Currie's lowest gear (haven't counted teeth) is about 12mph at the fastest cadence I can maintain; the X5's lowest gear about 9mph.
With the batteries switched off, I can see on my voltmeter the X5 regenerates about 10-15 volts (don't remember the exact number). There's enough regen current with the batteries off to power a 12 volt light I imagine, though I haven't calculated that or tried it.
Rassy
1 MW
xyster, when you said your X5 regenerates about 10-15 volts. If you put a light on it, wouldn't the lights requirement determine how many amps the motor would generate, while the speed determines the voltage? I.e. if you put a big load on it I think you would also get a big drag.
xyster
10 MW
Rassy said:
That's my understanding. I was more referring to my ability to pedal-power the light without it creating noticeable drag.
Rassy said:
If the hub motor generates 10-15V during pedalling, this 10-15V will supply small current to the controller with the battery switched OFF (or disconnected).
The controller is actually active and could draw a quiescent current (say 50 to 200 mA)
If this current is 200mA, the power generated is 15X0.2 = 3 watts.
This 3 watts power will provide an extra drag brside the normal cogging of the magnet ploes and stator poles.
xyster.
So if you disconnect the wiring connection between the controller and the motor, you could have lesser drag in pedalling the X5.
Wonder if you could do this experiment?
A dead controller (with burnt and short-circuited FET) will give a very big drag in pedalling for non-freewheeling hub motor.
Up to now:
Freewheeling motor provides less drag for coasting and pedalling.
Non-freewheeling motor provides regenerative energy in braking.
Wonder if there is any simple design to a motor capable of freewheeling in coasting/pedaling and non-freewheeling during regenerative braking !!???
Freewheeling motor provides less drag for coasting and pedalling.
Non-freewheeling motor provides regenerative energy in braking.
Wonder if there is any simple design to a motor capable of freewheeling in coasting/pedaling and non-freewheeling during regenerative braking !!???
xyster
10 MW
The7 said:
But this quiescent current is drawn from the batteries, not the motor's EMF. With all system power turned off, I'm getting about 15 volts generated by my pedaling at ~10-12 mph, registered by my analog voltmeter in parallel with the main power. Which correlates to a peak no-load speed at 80 volts of about 50mph -- sounds right since the bike goes 44mph.
I leave the controller's power button 'on', and use a power relay on main power from the batteries to switch the controller off and on.
I'd try it, but the drag is so minimal as to be almost unnoticeable -- perhaps about as much as the extra rolling resistance from the wide rear tire.
Similar threads
