non-pedal GNG 4kw

Joined
Aug 22, 2012
Messages
265
Location
Falun, Sweden
Inspired by bzhwindtalker:s great builds and the incredible power deliverance from the GNG 450 motor i'm aiming for a very simple setup.
The Specialized bighit will be used with the BB as a jackshaft, no pedals and footpegs a bit in an mx-style.
There will be a single reduction from motor to BB and another reduction from BB to derailleur.
55/11 at the motor and 34/9 from BB to cassette. (slowest gear)
GNG450 motor.
24S lipo. Lyen controller.
The round paper at the BB symbolizes the 55 sprocket.
new_GNG.jpg
 
Now this might be cool!
 
This may help when making a cardboard disc to simulate the sprocket so you can check for fit before you order parts using real money.

Sprocket diameters:

#25/#219____3/32" Bike chain____inches__mm
__80T__________40T___________ = 6.4___163
__84T__________42T___________ = 6.7___170
__88T__________44T___________ = 7.0___178
__92T__________46T___________ = 7.4___188
__96T__________48T___________ = 7.7___196
__100T_________50T___________ = 8.0___204
__104T_________52T___________ = 8.3___211
__108T_________54T___________ = 8.6___219
__112T_________56T___________ = 8.9___226
__116T_________58T___________ = 9.3___237
__120T_________60T___________ = 9.6___244

http://www.rbracing-rsr.com/calcsprocketdiam.html
 
nice project! the 9t bycicle sprocket will kill your final drive chain in minutes ;)

why not use a more adequate sprocket with 16t?

4kW is a lot for the bycicle drivetrain. you should use no sprockets smaller than 16t, also not on the wheel.

with 4kW you get incredible hill climb even with only 200% gear spread
 
"9t bycicle sprocket will kill your final drive chain in minutes"

That's an interesting point - do you know why that is and is it a function of chain pitch? I'm working on an ebike with 10T driving a 32 tooth 1/2 x 1/8 chain which from your statement
is going to be a killer. I notice that #219 chains are typically available with very low tooth counts, I assume that I'd be better off going #219.
 
there have to be enough teeth that are engaged. You can use 10T sprockets if the angles are right. As a rule of thumb, the full 2200N chain force can be used if there are at least 6teeth engaged. The chain wears still significantly slower if 8 or more teeth are engaged. This is just my personal experience. But if one looks closely at commercial available motorcycles, bikes and other machines, it looks like i'm not the only one who cares about these "rules of thumb":

What one should aim on when designing a chain drive:
-8t engagement
-max chain force is 1/4 of the chains tensile strenght, so about 2200N for a bicycle chain

Of course you can also use smaller sprockets with smaller angles, but with reduced the power capacity/max chain force.

A normal bicycle drive train with 12T wheel sprocket and derailleur can take 1200W max if geared to human speed. With 2000W the same drive wears chains very quickly. If you gear it faster you get lower chain force, but also less torque at the wheel.

When pushing 4kW through the drive train, one should do both, using large sprockets and the right motor reduction, keeping chain force down, keeping in mind that ~2200N chain force should not be exceeded
 

Attachments

  • tooth engagement.png
    tooth engagement.png
    3.5 KB · Views: 11,122
i did the math... with 9T you can only push 2.6kW into the chain... with 16T it would be 4

still there are 2 more flaws:

1st, the GNG is an 8-pole motor that has 0.5mm lams, so max RPM are ~4000, but you get ~5500rpm@24s lipo (didn't I tell you that earlier?)
so the max efficiency should be at 18s lipo, better push more amps instead, both your battery and your motor will like that better ;)

2nd
11T sprocket on the motor is good for 2500rpm, because of the polygon effect, you should use at least a 13T motor sprocket.
 

Attachments

  • low_power_sm.png
    low_power_sm.png
    35.8 KB · Views: 11,005
  • adequate_power_sm.png
    adequate_power_sm.png
    34.6 KB · Views: 11,005
Thanks for those calculations! What worries me a bit is the jackshaft torque. The #25 chain is choosen for the primary reduction and rated 140 lb, 64 kg.
When the 71,43 Nm is converted to kg it must also be divided in some way by the 55 sprocket diameter to get the Nm load on the chain? However, that #25 chain will be stretched...
9 tooth on the second can be switched to a bigger one. But i wanted this to be small because i dont want the 55t first reduction to be any bigger, it will make it hard to make the footpegs fit.

24s feels great, really exciting on my GNG/kona bike. The controller is set to 40, 60 and 100% on the power levels. I have only dared to try it on 60% so far and it feels like the motor want to rip the bike into pieces. Acceleration is brutal. 40% is nice in slow trailriding and match perfectly with the 48t BB-chainring.
 
Dont get me wrong, but 18s should still be more efficient with more amps. We once calced this using PeakEff. This motor CAN make use of 55amps battery current. You have to compare 24s@45amps with 18s@55amps, I doubt that one can feel any difference :?

You may do what you want, but be told that you waste expensive battery stored energy this way, without any performance gain.


You could use your 9T sprocket, if you get enough teeth engaged, but I doubt that this works:

chain.png

I guess that there will be the problem that the upper part of the chain may touch the swing arm, since it was not designed for such small sprockets at the BB :? maybe you should check?

Also, you will still waste your chain, since this drive exceeds chain specs at 4kW. This is just too much chain pull. A 24" rear wheel would help to keep the torque down, then you could use at least a 12T motor sprocket. Still a 12T sprocket is too small for 3000rpm+, you will get massive chain noise, your whole drive train wont like.
 
Thanks very much for the explanation crossbreak.
I haven't seen that spreadhseet you used for calculating chain forces before, looks excellent, is it available somewhere?

One question - do #35 and #219 chains work well at small tooth counts because the rise-fall time of the links is short due to small pitch?
Is there power handling ability a function of pitch and hence # of teeth engaged for their small drive sprocket size or are they just constructed with Kart abuse in mind?
This chain discussion is clearing up a ton of confusion for me, thanks again.
 
Hi Crossbreak. Great point. Polygon effect is significant in the smaller sprockets. Not so noticeable off road, but very lumpy feeling / sounding on road. Agree 11 tooth is pushing it.


D757P478_1.gif
 
I don't know the engineering behind this, but many builders have mentioned that an 11T or more is noticeably quieter than a 10T or less. Also, thicker sprockets are quieter (if noise is an issue).

Christerljung, if you are happy with the way this project is running, ignore my suggestions, but...If you wanted the #25 sprocket to stay as small as the 55T but it would be nice to have more reduction, like crossbreak said, you can use a smaller rear wheel. You can use a 20-inch, 24-inch rear wheel, and also a Pirelli ML75 moped tire on a 20-inch rim is 22-inches.

Since #25 is very cheap, you can run dual side-by-side sprockets for more power survivability?...

UB_Eurobike_2010_Conway_e-racerCIMG5163UB_Eurobike_2010_conway_e-rider_klein.jpg
 
sry, I can't post the chain force calculator spreadsheet, I will post it as soon as i'm home again.

What came to my mind eventually is, that the reduction ratio issues get worse with more voltage, which the GNG isn't made for anyway.
So that's another plus for using 18s and with more amps, I will post the full DriveCalc when i'm home again.

@spinningmagnets: I also thought if I should mention smaller wheels, both wheel would have to be changed, otherwise the bike wont stop wheelying :p if only the rear wheel is a smaller one. A longer swing arm or elongated dropout are things worth thinking about. Only one inch more wheel base can make a huge difference.
With 26" it's almost impossible to push 4kW through the normal derailleur drive train reliably. If I try to keep chain force within spec, top-gear-max-speeds of 90kph+ are the result. With 24" or 22" things look diffrent.

how fast do you wanna go in the end, christerl? Is 90kph the aim? Then your gearing make sense a lot :p

@blitzwagen
Is there power handling ability a function of pitch and hence # of teeth engaged for their small drive sprocket size or are they just constructed with Kart abuse in mind?
That's how one can explain this phenomenon :p Very small sprockets mean one less reduction stage, which is good for people who want to build things that might not have to last very long ;)
 
crossbreak said:
nice project! the 9t bycicle sprocket will kill your final drive chain in minutes ;)

why not use a more adequate sprocket with 16t?

4kW is a lot for the bycicle drivetrain. you should use no sprockets smaller than 16t, also not on the wheel.

with 4kW you get incredible hill climb even with only 200% gear spread

I have a dual 2400w Cyclone setup, both motors have 7 teeth and I've put 1860km on my bike.

My old motor chain lasted about 1500km, with me neglecting the chain lubrication for the first 1,100 and then using vegetable oil for the final 400km.

My chain going to my 26" wheel is the original 6 speed chain I got with the bike, also neglected lubrication wise, but I have proper chain oil now.
 
This is a great thread! Christer, very creative and interesting concept for your GNG adaptation. I want to ride it! Thanks, Crossbreak for the in-depth technical information on chains and sprockets. I think all of the time I've spent on E-S has been good for me. I actually understand what you're talking about. :D
 
that looks very professional :) allways great to make others work :p sry I'm still on holiday, next week i can make the complete calculation. any updates about your sprocket choice? which speed do you wanna go? how did you calc your first 11/55/9t reduction choice?
did you check if the 9t can fit your frame/swing arm the way suggested?
 
Made a calc for 3 versions:

-original GNG without cranks with 18s lipo
-your suggestion
-my suggestion (with larger sprockets and only 18s lipo)

All have about the same reduction ratios. I used 15-34T rather than the normal 13-34T cassette for all 3 calcs, I just use only 6 of the 7 gears available. 6-speed-Grip-shift works fine with 7-speed cassettes like the HG-50 13-34T. Or just never use your 7th gear like i do :p I just blocked it using the derailleur-adjustment screw. Without blocking it there is the risk of accidentally using it, which the chain wont like.

The used 13T output-gear I suggest should be fine if you use an additional guide sprocket the way i suggested above. With this drive your chain should survive more than 1000km if lubricated. You can use 3/32inch ingle speed gear for using it as an output sprocket. Hyperglide gears (from the cassettes) are not suitable as output gears IMO.
You should limit phase current as close to battery current as possible. this way you can squeeze the most out of your chain drive, motor and controller.
 

Attachments

  • comparison.png
    comparison.png
    52.8 KB · Views: 10,370
  • christeljung_GNG_pedalless.ods
    23 KB · Views: 170
non-hyperglide sprockets: http://www.bike-mailorder.de/Bike-Teile/Ritzel/Ritzel/Singlespeed-Ritzel-EXTRA-STRONG-3-32-Zoll.html

MS-Excel version of the calc:
 

Attachments

  • christeljung_GNG_pedalless.xls
    22.5 KB · Views: 204
Great spreadsheet, thanks! :D
I do like your gearing options better as from a pure mechanical aspect. Gears will survive longer and probably create a lower noise.
But: The 80t sprocket on the BB-shaft gives no good option to fit the footpegs. It would be a strange U-shaped construction to get under and around the 80t sprocket.
The idea is to bolt the footpegs mount to the Lighteningrods custom adjustable BB, but other suggestions are welcome 8)
 
Back
Top