Preventing Broken Dropouts: Torque Arm Info

[hjns]

My 9C axles are 10/14mm. The change to 10/12mm is apparently new. 12mm axles are much harder to keep from spinning. They are not as suitable for high power.

Well, I do not have the skills to replace the 12mm axle with a thicker one...

I would consider limiting front axle power to a lower level, perhaps half or so of the rear power. So set rear at 45A and front at 25A. Especially for initial testing, perhaps even lower for early tests. Consider setting the phase current lower for less multiplication that causes high torque at low speed. Don't want the front wheel to slip.

Good point. Actually, I choose the 2810 based on a 70% power delivery as compared to my rear HT3525 at the same voltage. However, I can certainly limit front amps even more. I still have the CA-LDP, and I can choose to which controller it is connected, so it is easy to limit the front controller "on the fly".

I would also consider stacking two of the torque arm sides to make the axle part thicker. The other piece, the torque strut, is not as stressed, so a single one there is fine.

Good idea. I already have asked DocBas about his TAs. I will probably end up epoxying these to the dropouts. At the same time, I will await the response from Rheinhard about his TAs.

Consider this possible scenario:

Running 75V, 45A controller settings
Phase current limit set for 2.5x so 110 amp phase current max

Starting out the throttle is opened slightly. The motor has no back EMF since it is not rotating, so the voltage is very low and the current goes to 110 amps at a few volts, developing maximum torque. Power is only a few hundred watts, not kilowatts.

This is the condition that blows controllers and breaks torque arms and dropouts. Max torque, very little throttle.

OK, thanks. I had not thought of it that way, and it certainly makes sense. I guess I have to pedal up to a certain speed (or have the rear motor do the first 5mph) and then allow the front to kick in. Maybe I can program the controller to accept a throttle signal after a higher threshold. Alternatively, I can just add a resistor/variable pot to the throttle signal, to achieve the same.

Thanks for the new ideas! Certainly a lot to try out!

 

[hjns]

Your broken fork is still good.
Here's an updated version for the front of a 2wd using moto hubbies and a fork with aluminum lowers. This one has clamping dropouts, so it's superior to the gen1 torque sleeves from 4 years ago. Note that the tubing is stiff enough and long enough that silicone does the trick to hold both gen1 and gen2 on the fork and gives you time to get everything aligned with the wheel on before it cures.

I like it very much. I also like the idea of not having wasted a good fork. However, I do not have the skills nor the equipment or even the outside room to do any welding (like to learn though, another time). Also, there are no obvious places here around Basel to have this kind of stuff done.

I will keep it in mind for the next E-bike 2WD implementation and for when I have a real house with garden and room to do some Tungsten arc welding.

Alternatively, if I measure the diameter of my fork, could you make them for me from some nice shiny strong metal and send it to me?

 

[Doctorbass]

From now, it wil l be a bit complicated to manage many size of torque arms for sale.. I dont say NO.. but at the moment i preffer staying the rear hub motor department.


Some info about my desing choice:
The one i make are made with DOMEX 100 wich is the most hard steel you can get and are 10mm wide, making any deformation impossible to happen with a 10mm x 14mm axel.

The reason why i choosed these parameters have been examined carefully to accomodate most of the people on the forum and it does work. I sold 300+ set from now and upgraded the desing 4 times since the revision A

I keep the desing simple to reduce the cost and allow people to get 2 torques plate at a lower the price of one of the other availlable.

With the 10mm width, you can certainly grind some parts on it to make them to fit on your dropout if needed because there is alot of material thickness surrounding the slot so drilling holes or cuting some part should not weak them if you make it in the right way.

If you keep them as is, they are impossible to bend with an axel. The axel simply break ! and that's your real max torque limit, not the Torque plate

They are just unbreakable with a 10mm axel motor period !

I was tired of broken Torque arms subject on the forum.. and now.. these subject suddenly began to dissapear :wink:

That's why i can say they are bulletproof... and get so popular

I want people having zero doubt about their torque arm strengh and enjoy each of their E-ride with a better safety feeling! 8)

Btw.. Next batch arriving in the next week .. same most prefered revision with 10.1mm slot, same Nomex 100 carbon steel.


Doc

 

[sn0wchyld]

Hjns, Hah! No soup for you! <snicker> No worries friend.

Alan B, force is force and it doesn’t make a bit of difference which side of the shock the torque arm is mounted so long as there is one and fastened down in a manner that offers good protection against axle rotation.

However I will concede that my preference would have been to mount them on the same side as the brake caliper for appearance, protection from the elements, and for aerodynamics. As such, it wasn’t convenient for several visibly-apparent reasons so I flipped them to the forward side.

As far as which side matters (or not), one has only to look at how bridges are constructed (especially in Portland!) to know there are twenty different ways to skin a cat, er… build a bridge. Cantilever above or below? The only people that care are the one’s paying the bill. Same with the torque arm: front or back? Either way, the links will be in compression or in tension, with regen-braking manifesting opposing force. So you are darned if you do and darned if you don’t

Taking this further, if Angle-A was at the axle, and Angle-B at the shock-mount, with Angle-C connecting the two links, the best that physics can provide is an equilateral triangle truss having all angles = 60°. My T.A. @ C is more low-profile, relaxed, and well… less expensive. If I were MadMax, we’d want them out farther with little pointing bits at the end… but I digress.

Soup to nuts & bolts, KF

I think what he was saying is if you picture the torqe force going anti clockwise in the above picture, it effectively tries to push 'down' on the bolt attaching the torque arm to the supporting rod (the torque arm is trying to spin anti clockwise relative to the hub). if you have a torque force going clockwise, you can see how it pushes 'up' on this same bolt, therefore providing an opposing 'down' force on the axle.

Im not entirely sure why calipers are mounted in such a way as to provide a 'downward' force on the axle (out of the dropout), but I'm guessing that the resultant downward force is minimal relative to other forces experienced by that section of the bike.

 

[hjns]

I think what he was saying is if you picture the torqe force going anti clockwise in the above picture, it effectively tries to push 'down' on the bolt attaching the torque arm to the supporting rod (the torque arm is trying to spin anti clockwise relative to the hub). if you have a torque force going clockwise, you can see how it pushes 'up' on this same bolt, therefore providing an opposing 'down' force on the axle.

Im not entirely sure why calipers are mounted in such a way as to provide a 'downward' force on the axle (out of the dropout), but I'm guessing that the resultant downward force is minimal relative to other forces experienced by that section of the bike.

Hmm. Actually, with a normal wheel, instead of a motorized one, during breaking there would be a huge force coming downwards from the fork to the dropouts onto the axis. In my view this is most likely be much more than the force from the disc brake calipers and the resulting pushing of the axle downwards out of the dropout.

It's a matter of leverage. The direct downward force from the bike + rider keeps the axle in the dropouts. With a motorized wheel, the vectors are different. Especially during acceleration, there are even less downward forces from the bike + rider. So I noticed...

 

[hjns]

From now, it wil l be a bit complicated to manage many size of torque arms for sale.. I dont say NO.. but at the moment i preffer staying the rear hub motor department.

Yeah, yeah, I got the message. I want 4 of them. Two front, two rear. Happy to wait for the new production batch.
Maybe in addition to what Rheinhard can produce. 8)

 

[dogman dan]

Man, if that axle is really only 12mm, you have a major problem now. I've run the 9c 2810 on front hubs for years at lower power and never had a problem. Two ebikes CA torque arms.

But when I tried to use the clamp on torque arms for 3000w, using a 2807 front motor on the race bike, I got a similar result. When the power was applied, the clamps simply failed to secure the arm good enough, and the axle was allowed to rotate, dragging the torque arm along with it.

I solved the problem by welding tabs to the steel fork, and securing the arms second piece with two bolts. No longer able to slide down the fork, two torque arms was plenty to hold more than 3000w.

But this was using a 14 mm axle with wide flats. No way the axle was going to rotate in the torque arm like yours did. I have always felt 12 mm axle flats were a joke, even for 1000w.

You have simply got to go to something that is going to pinch the axle to apply real power. Something like the dropout sleeves. I can't believe anything else will work with the lame flats on a 12 mm axle.

 

[John in CR]

If the axle is a problem, I've got a fix for that too. That is my torque axle nut solution with larger flats at a greater radius. It even solves partially stripped axle thread.



Essentially you make an axle nut that is an inch or longer, and put it on all the way down to the hip on the axle so it can't turn any more. The flats on it become the new axle flats. Keep in mind for forward motion the right side is the important one, because the axle twisting force is to tighten it, and for regen the one on the left holds back the torque.

I didn't use loctite or anything on mine, because to get the covers off requires removing this torque axle sleeve/nut. I used that in conjunction with my gen1 fork sleeves above, because not only were my AL fork dropouts history, but the high torque motor had those small flats too and between the idiot seller using nuts with the wrong threads, the spinout stripped more of the axle threads. Almost 4 years and those new nuts are still going strong. The only tricky part is if using regen, you want to wait until after test fitting to cut the flats on one of the nuts so the two end up on the same plane when tightened to the axle hip.

You get the side benefit of a stronger axle too, so you can torque down the clamping torque arms to the new flats without so much worry about crushing the hollow axle on the wire side.

John

 

[hjns]

Thanks Dogman and John.

I am on a business trip and will continue working on this next weekend.

 

[dogman dan]

Brilliance from John again. Love the way you look at things, and come up with obvious and simple solutions.

It would certainly be possible to weld that nut to the axle shoulder too, after pulling wires out of course.

 

[Cyclebutt]

Excellent thread, and only two pages. I'm in the process of putting a Bafang on the front of a Mongoose Beast. I never thought I'd mount a front drive (actually a rear hub on the front, 130mm dropouts, fat bike tires/wheels..(4" wide rims), hey, it's an experiment!) but this factory design makes the front tire fold under with tight turns so I went front drive to have it PULL itself through those turns. That should cure the folding under issue..hope...

Anyhow, thanks for the torque arm info and that spun axle fix is pure gold! You guys rock! Never delete ANYTHING!