Crankin' It - Mid-Mounted Crank Drive

DrkAngel

1 GW
Joined
Dec 15, 2010
Messages
5,300
Location
Upstate-Western-Southern Tier NY. USA
For years now, I have been considering a mid-mounted freewheel crank drive.
Running a geared reduction motor through the 7 speed freewheel sprocket, or freehub cassette.
Most any bike could be converted, adding less than 10lbs of motor hardware. (Battery additional.)

Stage 1.
Typically, this involves using a geared reduction motor connected through the front chainring.

A big improvement over most any type eBike drive!
With a 11-34T 7spd freewheel you can have the equivalent of a 7 speed motorcycle, with +300% the torque at low speed.
Great low speed torque and high speed assist ... right to 30mph(?) ... from the same motor!

Stage 2.
An important option is a 2, or 1, chainring freewheel crank.
This allows motor power without pedals being turned ... much safer!

There is the variable of matching motor to pedal speed.
The size of the 2 chainrings can be varied, and / or voltage adjusted, to match your rhythm.

Stage 3.
The next stage option is a freewheel sprocket, instead of a fixed sprocket on the motor, this allows easy pedaling when motor not engaged.

My project motor is the Unite MY1018z gear reduction motor.
I've poked-prodded-tested and tried this model for ~20,000 miles.
As a demonstration,
I graphed the speed and torque available from running the motor through the 7 speed sprockets.

MY1018z motor @24V w/9T sprocket
At 24V Low gear (34T) will give you great power-acceleration till 7mph, great for the worst hills.
Then, there are a variety of optimal hill climbing, or acceleration, choices till about 16mph.
Of note is that a 34-13T would limit maximum speed to the "legal" 20mph.

Note - Speed-Torque lines are for gearing ratio comparisons and do not represent "torque curves".
file.php



I just finished purchasing the individual components and will continue to post build reports.
 
Previous graph is for the MY1018z 24V motor with the OEM 9T sprocket.

Stage 3.
In order to operate pedals without using the motor the use of a freewheel sprocket on the motor is required.
The smallest compatible freewheel sprocket I've found is a 13T.
Of course, this will vary the motor applied speed noticeably.

MY1018z motor @24V w/13T freewheel sprocket

Note - Speed-Torque lines are for gearing ratio comparisons and do not represent "torque curves".
Crank Drive - 24V 13T.jpg
Sadly, this loses much of the low speed torque advantages of a multi-speed mid-drive, might be OK on a 20"er?
On the other hand...
In lowest gear as I am accelerating from a red light, I hit 10mph at my max pedal speed.
So I would be shifting right at maximum torque of motor and pedal assist, maybe ... not so bad.

Personally, I could be happy with a 9T sprocket with motor always engaged and pedal assist.
Optimally shifting at a more casual 60-70 pedal rpm.
A larger chainring from the motor can offset some of this loss.
With an intermediate jackshaft, most any gear ratio can be accomplished.

Stay tuned for more developments ...
 
Crank Drive Build #1 - Stage 2

Components:
Huffy dual-suspension mountain bike
Dual chainring freewheel crank (40T/40T will change after testing)
33.3V 25.92Ah 9s12p homemade recycled Lipo battery
Unite MY1018z 24V (450w) gear reduction motor w/9T sprocket (624w@33.3V)

MY1018z motor @ 33.3V though DNP 34-11T 7spd Freewheel

Note - Speed-Torque lines are for gearing ratio comparisons and do not represent "torque curves".
Crank Drive - 33.3V.jpg

I think this to be nearly ideal for my 1st trial
90rpm pedal speed looks to match optimal shift points at full throttle
Partial throttle can move optimal shift points towards a more casual 60rpm.

Low gear torque is excellent, and top motor only speed looks to be ~26-27mph
Lowering seating profile or moderate pedal assist should make 30mph possible.
 
Stage 4. See - revised reasoning-variation
Important, IMO, is the survivability of the drive train!
Shifting under throttle will produce a (un)healthy "clunk" ...
an impact on all drive train components: motor > chain > chainring > chain > 7spd freewheel > hub

This "damage" can be moderated by not shifting while under throttle, release throttle before shifting then gradually re-engage.
This works but kinda degrades performance ... substantially!

So, I propose the addition of a heavily sprung idler roller.
Place behind the chainring, pushing chain down, roller should be ~ 50% sprung under full throttle.
Instead of a shift inducing a straight line immediate impact the "impact" will be spread over the unsprung distance the idler provides.
"Impact" might be reduced 90%(?), providing a great increase in durability-reliability!

- $4.74

See - revised reasoning-variation

Surprising to most, there is minimal abuse or wear on the Bottom Bracket!
With motor positioned in front of the BB, motor to BB abuse is evenly offset by BB to wheel abuse.

The exception being the crank freewheel bearing which must endure a constant twisting, worse under heavy throttle.
This "twisting" abuse could be largely eliminated.
The use of an intermediate jack shaft positioned directly behind the BB, (if there is room), inside the wheel chain path, would even up the stress on the freewheel crank bearing.

But, of course, doing this would apply greater stress on the BB bearing ...
 

Attachments

  • Roller.JPG
    Roller.JPG
    11.5 KB · Views: 11,526
I have followed some of your other posts on this and another site. You do some interesting projects, I am interested in this work for a few reasons. I am currently building my first electric trike and using two 120 watt brush motors with a friction drive, I expect to upgrade to something more efficient in the future. What you are doing here might be right for me. So I was wondering if you had an opinion about efficiency of brushed vs brushless motors at speeds in the 15 to 20 mph range, I am thinking that at moderate speeds there may not be much difference.

Also, do you know if the gear drive is noisy, does it have plastic gears.

I like your idea to reduce impact on the final chain, I built a recumbent gas motor bike a few years ago with a chain idler that could be modified to include your idea. In the photo, the chain slack is adjusted by bending the link attached to the idler sprocket. The idler arm pivot is bolted tight. This could be changed to allow the idler arm to pivot and the link to act as a spring. At some time I may convert this bike to electric power and it would be easy to use the same gear motor as you.

41ChainIdler_zps3663decf.jpg
 
As long as you go Permanent Magnet (PM) there is not a lot of difference, about 80% peak efficiency for brush and 90% for brushless.
Though, brush is less complex and usually cheaper.

The MY1018z has straight cut steel gears and is noticeably noisy.
The later XYD-16 model has helical cut gears and is nearly silent. (More expensive and (perhaps) a bit less efficient )

Placing the idler arm etc. on the pedal chain is preferable. The derailleur picks up the chain slack.
 
Some Components:

SunRace 13-34T MegaRange - $16.99 shipped


DNP Epoch 7spd 30-32 to 11T Freewheel - $34
Upgrades EZip w/44T Front Sprocket to near 30mph pedal assist (90 crank rpm)

More sizes - 11- 28,30,32,34T - $32

Quality ACS Freewheel for freewheel crank! $16.50
BCD not specified ... might bolt directly to certain chainrings?

Also available _ more components, from Staton-inc.com

Freewheel Cranks - Choices ... >$32



450w 24v, gear reduction motor, standard 1/2" x 1/8" chain (MY1018Z) - $95 + shipping


The "silent" motor (well ... not real loud)
XYD-16 motor w/mounting bracket - $108.89 - Will try to find working link


Bikemotive.com
Sickbikeparts.com
 

Attachments

  • Crank Freewheel.JPG
    Crank Freewheel.JPG
    23.8 KB · Views: 11,252
450w (24V) Gear Reduction Motors (Brush) - thin profile
25.9V = 485w
29.6V = 555w
33.3V = 625w ... partial throttle at low speed recommended to be more efficient and to reduce damaging heat
37V = 695w ... reduced amperage controller (or throttle restraint) recommended to reduce low speed heat damage
44.4V = 835w ... reduced amperage controller (or throttle restraint) recommended to reduce heat damage
48.1V = 900w ... reduced amperage controller (or throttle restraint) recommended to reduce heat damage and prevent shearing the sprocket key

XYD-16
- Confirm with seller! White label States XYD-16 and "C" on cover


Monsterscooterparts.com

Eledtricscooterparts.com - 7th item down

Thesuperkids.com

MY1018
Triangle on Cover
Monsterscooterparts.com

TNCscooters.com
 
Stage 5 - Under Development

Next evolution?
More complex ... but with great advantages!
The use of 2 freewheels, bypassing the need for a freewheel sprocket on the motor.
Will most likely involve an intermediate jack shaft, allowing swappable gear ratios.
 
Miles said:
So, what will the RHS freewheel do? :)
With the LHS freewheel allowing pedaling without motor engaged ...
The RHS freewheel allows motor use without the pedals turning. - Under Review
 
DrkAngel said:
Mid-Drive Resources

ElectricBike.com - 10 mid-drives info & reviews!

Mid-Drive Kits
GNGeBike.com
Lectriccycles.com
Middriveebike.com
eBay

More Mid-Drive Components
GNGeBike.com - Try to combine shipping and negotiate shipping price
Bikemotive.com

Bonus: Very Interesting!


I installed a Bafang Mid Drive on my Hardtail. This kit is really nice, and simplifies everything. May not be much challenge for the custom builder/modder, but it sure performs like a dream.
 
Hi DrkAngel,

Sorry to stop you in your tracks! It is possible to achieve the set-up you're proposing but it's definitely not a trivial task!

I made a start at designing such a system, as have others.

http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=35846

http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=37224

file.php
 
Yeah,
I'm leaning more towards a dual freewheel on the pedal side ... very tight fit ...
But more likely an intermediate jack shaft.

Of course I am merely speculating, as my current objectives, (1st build), will not include a motor freewheel.
With 33.3V 31.2Ah & 43.2Ah homemade batteries I have plenty of power to continuously idle along.
 
Freewheel cranks

Got 1, then got a spare.
These are some unknown items with original source unknown, so don't ask me.

Got a double chainring, then a discount on a single.
All 3 chainrings are 36T, ... OK for initial trials.
They have an offset so that there is a nice spacing, but with a single, flipped over, chain is positioned directly over the center of the freewheel. This should eliminate the "twisting" stress common with the doubles and greatly extend usable bearing life.

file.php

I believe I will try the single ...
Positioning the motor sprocket in front of and below the chainring, I will stretch, or lengthen the chain around a single path, ensuring a ~120 degree motor sprocket contact.

file.php


Freewheels are the Pheasant brand ... the equivalent of what we would call a turkey ... but I have an ACS Crossfire for when it fails.
 

Attachments

  • Crank1.JPG
    Crank1.JPG
    168.3 KB · Views: 1,996
Single Chainring Advantages - Disadvantages

Advantages
Eliminates the twisting (off-center) stress-wear on the crank freewheel = greatly increased reliability-usable life
Reduced cost! "Standard" crank freewheel with single centered chainring will survive longer or as long as super duty freewheel with multiple chainrings

Disadvantages
Negates the possibility of shifting between multiple front chainrings
Eliminates the variance of chainring sizes to match motor to pedal speed.
Doesn't look as complex = doesn't impress the simple minded as easily

Perhaps not obvious ... It is still possible to modify motor and pedal speeds

Motor speed can be varied either by:

Changing the size of the motor sprocket

The Currie motors, MY1018 & XYD-16, have 8T,9T,11T,13T available (mod-able 15T also found)
Adapter is available for 13T-14T-15T small freewheels
Additional adapter is available for the "standard" 16T - 22T freewheels

Changing the voltage of the motor
At 24V, gear reduced (sprocket) speed is ~400 rpm
22.2V = 370 rpm
24V = 400 rpm
25.9V = 432 rpm
30V = 500 rpm
33.3V = 555 rpm
37V = 617 rpm
40.7V = 678 rpm
44.4V = 740 rpm
48.1V = 800 rpm

Then ... rear sprocket can be adjusted
14-34T to 11-28T to 21-26T 7-speed freewheels (or cassettes) available

Front Chainring
Now you adjust your pedal ratio by changing the single chainring ... 32T - 62T+.
This has no effect on the motor speed ratio! ...

Motor speed is now determined by motor sprocket to rear wheel sprockets ... not by chain ring.
Chainring is effectively an idler sprocket for the motor contribution, varying the size has no relation to motor speed.

I do not see any reason to use dual chainrings of equal size ... ?
I have 2 - dual 36T and 1 - single 36T.
Advantage of chain contacting 20T rather than 10T?
I can't justify that as any reasonable advantage ...

So, I will likely post up, at least, 2 36T chainrings with "standard" 5 hole - for ACS CrossFire etc. - Appear new - steel.
Or swap for compatible 44T 48T ... other?
 
DrkAngel said:
Single Chainring Advantages - Disadvantages

Advantages
Eliminates the twisting (off-center) stress-wear on the crank freewheel = greatly increased reliability-usable life

It may or may not reduce it but I don't see how it can eliminate it? Input and output are not aligned, therefore..... Yes?
 
Miles said:
DrkAngel said:
Single Chainring Advantages - Disadvantages

Advantages
Eliminates the twisting (off-center) stress-wear on the crank freewheel = greatly increased reliability-usable life

It may or may not reduce it but I don't see how it can eliminate it? Input and output are not aligned, therefore..... Yes?

Double Chainring - With motor pulling the outside chainring forward ... wear is on outside rear edge of crank freewheel bearings.
Inside chainring pulled rearward stresses front inside edge of the crank freewheel bearings.
Hence the twisting damage.

A single chainring has center load which evenly loads the inside and outside edges-bearings ... the way the freewheel was designed to be used.

It eliminates the excessive twisting at the crank freewheel ...
Yes, there is still the comparatively minor off-center pull to the rear sprockets, but bearing load is only on the front of the freewheel, one side slightly less load.

Oh, and the motor dos not put any pulling stress on the freewheel, merely a center weighted idling stress, pulling stress is on the rear sprockets.
 
Sorry, you've lost me...

You have a typical set-up with a chain from the motor going to one chainring and the output to the wheel on an adjacent chainring?
 
Miles said:
Sorry, you've lost me...

You have a typical set-up with a chain from the motor going to one chainring and the output to the wheel on an adjacent chainring?
Single chainring, turned, to center chain on freewheel.
Single chain, lengthened, pulled forward, lower quadrant below chainring - around motor sprocket.
See pics and 1st thoughts - Got double and single chainring crank freewheels

Similar to spinningmagnets pic except ...
with motor sprocket forward-below chainring.
 
Back
Top