Dropouts, torque-arms, flats and sides

[TylerDurden]

I see comments here regularly about how much contribution the "tightness" of axle-nuts contributes to keeping a hubmotor from spinning in the dropouts.

I think the amount of grip the axle-nut and flange have on the dropouts is minor. The flats are the real support, IMO:

• The moment-arm of the nut/flange are miniscule compared to even a 20" wheel.
  Even the tightest axle-nuts have allowed axles to spin.
  Coaster-brakes have always had torque-arms.
  Torque-arms have resolved spinning issues; not tighter axle-nuts.
  Flats of the axle-nuts are used to tighten the axle-nuts.

Your thoughts?

 

[slayer]

Hi Tyler,
I am not very good in english but if i understand you correctly you say that you need to build a ¨torque arm¨ that just thightening the bolts are not enough

me i took my course as a auto body technician and worked in a body shop for 10 years ...there are some specifications on a car that says how much torque you should put on a nut and all the nuts are also rated by marked on the bolts that is going to tell yonutu the quality of that nut (how much torque it can take ...and dont forget that between the nut if it is tight enough there is a bearing so the motor torque is not all on the nut ... there are some torque wrench that tell you exactly how much torque you are putting on the nuts and i would estimate that 150 foot pound inch would be enough

 

[CGameProgrammer]

Please cite examples where an axle has spun even when the axle nuts were very tight. I have never seen or read of this happening; always, the axle nuts loosen before the axle spins. You're underestimating how much force a strong clamp can withstand.

 

[Mathurin]

Well, drum brakes also have torque arms, frames with disk mounts have slightly more weight in the proper areas and often you can see a brace that connects the seatstay and chainstay in the rear triangle close to the caliper mount. However, brakes can apply rediculous amounts of torque, to the point of locking a wheel while the bike is moving. If you look at gearhubs, they basicly multiply or devide, and when you have a clydesdale standing up to mash with >500% gear range there's quite a bit of torque applied to the axles, yet the drops apparently survive that. Some gearhubs have nothing but axles and bolts, some have anti spin washers with a tab thing on them made of something fairly soft, so the washer strips out instead of ruining the frame if something bad happens. Even the ridiculously reliable BionX setups use washers with tabs. Unless it's an e-moped or alu or something, I wouldn't worry.

 

[rkosiorek]

knoxie had a BMX bike with a front PUMA motor on 72V. link to a video was posted somewhere here on this forum but i cannot locate it at this time.

anyway he had a 10mm wrench on the axle flats and a couple of hose clamps to keep the arm of the wrench against the forks. usual axle nut clamping wrench head against dropout.

torque of the motor broke the wrench and when the axle turned it routed out a nice round hole in the dropout.

personally i have had a 4011 motor in an aluminium dropout, break the dropout on the front fork. and yes i did have it nice and tight.

so this is not just an urban legend. it can happen.

rick

 

[slayer]

rkosiorek

he probably did not apply enough torque on the nuts (ft lb in)...other than that i don t really understand about the 10 mm wrench with the hose clamp ...we would have to see it in person make a judgemnt ...and it would be just a assumption...i like sometimes to look at accidents in the shop and we discuss how it happened and why ...twisted metal to say this happened because of that ...and disandat...lol

 

[TylerDurden]

Please cite examples where an axle has spun even when the axle nuts were very tight. I have never seen or read of this happening; always, the axle nuts loosen before the axle spins.

Please cite examples of how anyone could tell which happened first.


For that matter, how could a nut hold an axle if it can't hold itself in the first place?

:lol:

 

[CGameProgrammer]

Well I once had a steel dropout rip apart even though it covered the axle flats, but the axle nuts were not tight; I'd tightened them only with small pliers, which obviously doesn't work very well.

From what I understand, when a nut is tightened to a given torque, nothing less than that torque will loosen it. So it needs to be tightened to a greater torque than what the motor will exert, which is around 30-70 lb-ft for most people with an X5.

 

[Mathurin]

[...]I'd tightened them only with small pliers [...]

headdesk.jpg

Don't blame it on the bike...

 

[HAL9000v2.0]

Here you can calculate how much torque you need.
http://www.tribology-abc.com/calculators/e3_6b.htm

For an example my x5 have M14x1 thread on both sides. This bolt have maximal strength when is tightened with 105Nm.
But this is not actual value that this bolt is holding. Inner wider side of the axle is holding about 60% of the torque
only by friction with inner side of the dropouts also we must add surface (where bolt is flattened on 10mm)
that pressing against dropout. Force is not equally spread on this surface.
Unfortunately highest is on the highest radius. It is easy to calculate that on, lets say, 65Nm of motor torque there
is (r=6mm, biggest radius on flat surface on bolt) pressure of 10833N or 1ton.
As normal steel have tensile strength of 720-760 N/mm^2 it is easy to now how thick have your dropouts should be.
For the "holding surface" you can use half of flat part on the bolt, cca 4mm x width of dropouts.
I'we made additional "wrench-like" 8mm thick torque arm and both of my dropouts are 4mm thick. All this holding with any
visible damage torque of 132Nm (by ebikes.ca simulator)
Pardon my Metric system
Also you may need converter...

 

[slayer]

thanks hal9000 that is going to be very useful conversion program

 

[rkosiorek]

since my experience with the Aluminum fork - no more aluminum forks for me. it wasn't a cheap fork either.

but i think hal9000 has a point. steel and the thicker the better is my moto for the dropout.

rick