Torque arm design help

E-HP

10 GW
Joined
Nov 1, 2018
Messages
6,201
Location
USA
Looks like it's raining all weekend, and into next week over here, so I have some time to think about converting my Cove Stiffee FR downhill/freeride frame for e-bike use. The frame triangle will snugly fits my hardcased triangle pack, so although I haven't decided whether the final bike will be mid-drive or rear hub drive, I'm going to set it up to inherit my current rear hub and battery. That way, I can convert to disc brakes and use my much more plush Fox Talus (130mm) fork, sooner than later.

As part of the conversion, I need to beef up the dropout(s) for using my rear hub motor. The design of the rear dropouts, while beefy, are contoured in a way that I have to plan carefully on torque arm design. The motor is only 1000 watts @48V, running on a 52V pack, but if I decide to go with rear hub with a little more power, I want the torque arm(s) to be able to handle it.

Here's what I have to work with. As you can see, the dropouts are flat for about an inch, then start tapering out. The left side has two triangle openings, and the right has one. Just under 1/4" thick at axle end. I will be using pins that will fit into the openings of the frame for strength, but using Devcon or other high content steel epoxy to fit within the contours of those openings. I'll be using release agent so that the arms can be popped out after the epoxy cures. The Devcon and pins, which will conform to the openings, will provide a tight fit with no movement.

My current thinking is to design an arm for the left dropout, so I don't need to work around the derailleur. My plan is to leverage the contours of the frame as much as possible to create the anchoring points for the arm. I want the arm(s) removable, since I still may go with a mid-drive at a later time. I will be using, at minimum, 1/4" stainless flat bar.

iBZaKOYPdythZA5YwA4CsMv86koFkgwg-OaIMc9Ga4ompz8tShsrxJtdKKDWTaY8lRARaZFK4RVo3G87QB6gXDfp0lIzCbRQ_CWcRrUtNY6HycYbfGI0W4RYfwC759SKiHu_qN_hp1hYMedOvOFjrkJlK7Qtw_J3KUOvaiz0gf5AY3IJSAzdAnp3EcQEyDWRLG2vFetHGDSfQ_zSEcsH5SBa8rd-AY2MKl2yTMX1ra8LXSO5n0n0_Q7OWEDO5X-cjOZhPg8HWjmgh3L_6g8i-8BCt94kB1V9X1rUMlBr-lgbhQ9osqsDI4qHjYfEs2cVV_6xXnmxYUYwmuvpOi40bC0NH9HYDsSTuAUly5GzVrFlJn7kKXq0UUfd3bFU8Ht21RJeO4_uh9ZQgZbU-i75u8a1LQ3WDS9hAai5DQB3aEX4qyn1pvZCXYFIzo6suUcnVDwnRxsvKRRw9HLHYMUB4mSXzEYjK4LgyPUgu3CtMLTGZwBhGXFC9mxMnH7gpuAhi_Ub2EF9EhDxQF7wQeEvi6ZnuhVWlTNz7mzw0ZeBHaIGORNj7v0fJSTSPI3Qd2ViiKF5LAGBF-_oOYBGJ3l2nuglh8GFh8zj77EPssr6LmIatIP8KptAA2pET2sMJRaAis6Zu1KQcieKXeqFLDmNTT_L=w1137-h852-no


This is the first design, which I think will provide enough support for the current 1000 watt motor, and using regen. The green .
arrows are where the forces will be contained when torque is applied for both rotational directions.
TiapGw9FHxRCWLHhyoBG6jGjPDKGwPknqZpqOOoeNKHsKBrbIU6c-vorSMyBAJQ--cNExPSPq9nMdst6XpinbMYQ_ir_FPCXtID_0xHXSlazqhe_qZwDqWA48lI1VTaJUms6FbaFirBUqKN6UjTuCJJsJ8ZNQrvMu7h9a7CpS5sGPv2KGBMHew4ZveqVnT5dG2lKxqygiLRfhEKmFyrII6UsKjMCIZyBjr_pBZwedvBM2OpdVIxfbQHF4HcOpYh9VQ-ApMJjpW4ftn4VZ50uCINHsLDt1-v8SHKpt8Q-7F8RfxbCu4btWw0VspiBoGPKjPFFaY5CNQidv9MOtTp4Su5FuFnvF80K_cMwhuJkZ7FZf3SkTPQOAjk4zuKLIjqhrc80c7Gje9KzmDR-UJdXI2tGrBSc8HlYkm7aExFRZid7krnaFCL4CAHGjzevst2fuJamFekaTqwxaCcoxS0WMX6Bp-iTyAusWUWgOX3ombM0xv6q5uCtJ6KHwiuWrNnVku5ih5eTKhPeiamC3uPwwwgvkSqwiHMSc5_Oud5Bd3WfitJbm9VUVa6H6JXdIQmf-uZpm9a4nXEalFGI39AWn27F1jzAWvxxHf22K7f9a546tNaF_8ITC_xib3Ns7AKIy3O1x0cAZBYaE4L5xplzR8RG=w1137-h852-no


The second option extends the torque arm to the second triangle cutout. I will need to mill/file/grind a channel to clear the first triangle section, but this will provide additional support, further from the axle. Because of how the frame tapers out, there is more support along that furthest contour from the axle. Again, the green arrows are where I think the opposing forces will the contained.
xTH4swrAaqSH0N03Kb52JmEM60_eaozd1_3YaGNvqMUZZ4YkHgf1MDevikdIIsdpMR45uGM8K77QLgi15SRj5_I4_nUL3sXSdOd9kkaWBEH6VB5whBFyum-aOIr5MGEVhrGgGDMIeCxlOVTOWmoxeBIVTD9ZxPK6CKM0_cwWDuK79J5A_0SCmfUrbtJRAhqkIxCQsWr0jyKigTjtECXEKAsBcFHGsoEPuCPX8e327P1fhb8H8i3HTy2UKpmGAKIm-qjTFJO-A46KfasA5RPl2HZZg-ckGSQwxFHJF6Q3DOlG9oIy3kcU7GE7FVx-8oXyQL8YK3sGlp21dRHXx1pp4VBIKTe1i5zG9x3WcmxVl1zWU3j6kBqdZWyZsOLaB-MdUCMYv8uHIojviF7I2smqeHdJ6XSBR75HY5XMuZbGCb2DSP8y3mOTpO38UHW9KA5oM9swuOzEQPZpNrK_bKuR15ORFOARXOOO6dZ2WIBg5Ft-4HRYwcKjUjshpsjG-flxcbGJCs2jrXgXRkthCicw5RYQnt-dP3vPjc_GyHwFYHt8GsDG89GskP-siH8cvo69xpSsB4TQFkAvUF8O3tCc27qe2WQ55iAMZIRQmWMGJ6sb5d6FJoKYqhJJBwbdRVjI82ivgDVknWq7lkgSeVIzLqEZ=w1000-h750-no


If I decide to go with two arms, this is how I would approach the right side dropout. More straightforward that the left side, but I need to fit the derailleur and make sure everything clears.
Hp5Bnl5us8aHBy0CyaQufodjsy73OCRQPk34scJpZnwHaCTyILVJhe-4kUqbYq_oMEhsNb4wOZ6VoxQ79sZ5fWCHIbinT0Ef7HSa2iixiuFnnvW8CILC3pSPZOeK_cxh09WeVGeN2TRW_1Qj6Y6oS5dssbahsckaGpEeD9sB5uZvV0TB5LpoO-j9U5Ia86iGnj_MjueTxh5rKcBRpD85zygMaPG1GEYhib4KxQZHRUawiZcOykuoCsVN-qQt_Jl-B736Uz6mlBqWagqVHbE7gFTpXxHVwMCbwwR7oINskRz-940xAgem4TviBVPHQW-ERhpFAbqrkg9P9yK6telbSRTbKulX7kuBVPcBLa7TcxN5mEFnGgGTnoxUhCuqaINKhS5K-u0hsx0FEirmPmbtm-iwwbGo73ooeiBKovoNwSnLuV9jOjIMECTIc6mDfxR-z6729TAO-vakoxRfqFw-jI2yFqPwcHsy-y-YuhK8ta4FG428fbBu91Ko7mjHYJazkizoWrUhoaZ3iEdDjEVK9-biMBcxgdG2fyZS74sW81w81nqqAPV-PKvj4EvteyB7wcyoHn3ckcKUwbliTTgKivJaEgmYdrxQ-IVcz1X49XvGlcBPQEdjDVMj89ilP8YNGta8fAuXsTbTZO8fnMnJN3IF=w1000-h750-no


Any feedback would be greatly appreciated, since I haven't ordered the stainless flat bar material yet. Thanks ahead of time for your advice and expertise.
 
I suggest not using slot for the axle, but instead use a hole. If your motor wire won't fit through a hole, then use a slot on the side with the wire and a hole on the side without.

I'd look to anchor your torque arm to either the seat or chainstay. On the disc brake side, I'd think about using the brake mounts. Just get longer bolts of the appropriate grade. The irregular shapes of the holes you are looking to use make me a little concerned about whether you can get a firm connection without damaging that part of the frame.
 
wturber said:
I suggest not using slot for the axle, but instead use a hole. If your motor wire won't fit through a hole, then use a slot on the side with the wire and a hole on the side without.

I'd look to anchor your torque arm to either the seat or chainstay. On the disc brake side, I'd think about using the brake mounts. Just get longer bolts of the appropriate grade. The irregular shapes of the holes you are looking to use make me a little concerned about whether you can get a firm connection without damaging that part of the frame.

I think I can do the hole and shape it to make things stronger. I was originally thinking if I went' thicker, like 3/8", then I'd have sufficient strength, or use a cross bolt design to clamp the axle. For the arm, I was looking at something similar to this, where the frame is utilized to capture the arm. I figured the stainless pins, maybe 1/4" diameter, and the Devcon, which I believe would hold until the whole aluminum frame broke off, would be enough. It's pretty strong with respect to compression, between the pins and frame. (it's used for bedding high power big game rifle actions).

https://west-coast-electric-cycles.myshopify.com/collections/torque-arms/products/custom-torque-arms-for-gravity-bullseye-monster-comp-fat-bikes.

Maybe adding a threaded anchoring plate on the opposite side of the dropout would help as well. I'm not anticipating anything more than 3kW at the very most, but maybe closer to 2kW. The longer version is close to 3" or so, with about 2 1/2" wide butting flat against the chain/seat stay junction, where they meet the dropout. I may need to get out the calculator and estimate some of the forces that would be applied at the various points.
 
I don't think the shape of the area you are dealing with will make a fit like the one you linked to work. You have shapes that are complex along all three axes. But that gave me an idea. Maybe use some epoxy and fabric, fiberglass, carbon fiber filler to fill that complex shape. Once cured, you can drill through it and clamp from the other side.
 
wturber said:
I don't think the shape of the area you are dealing with will make a fit like the one you linked to work. You have shapes that are complex along all three axes. But that gave me an idea. Maybe use some epoxy and fabric, fiberglass, carbon fiber filler to fill that complex shape. Once cured, you can drill through it and clamp from the other side.

Yup, that's exactly what I'm planning on. The Devcon will fill both of those triangle sections, but with the stainless pins reinforcing the areas where the most force will be applied. I'll use clay to keep the Devcon from going too far into the triangles to avoid forming a mechanical lock, which is always the thing to worry about. Maybe I can do a mock up with wood and putty to make it easier to visualize. I can even use Devcon for the clamp on the opposite side, which will lock that into position as well.

Mainly, I don't want to alter or damage the frame. It's an old frame, but it's still sort of legendary as far as hardtail freeride frames go. And the super beefy Easton RAD tubing in the triangle will make the triangle battery pack look like it belongs there. I'm picturing the final product, and I think it will look and perform great. I'm patient though, and don't want to mess up.

Thanks for the help on this.
 
You'll probably want to coat the frame with something that will act as a mold release. You don't want to bond the epoxy to the frame.

Hopefully others will wade in and tell us what's wrong with these ideas. :^)
 
wturber said:
You'll probably want to coat the frame with something that will act as a mold release. You don't want to bond the epoxy to the frame.

Hopefully others will wade in and tell us what's wrong with these ideas. :^)

Yup, I've used Kiwi neutral shoe polish with great success as a release agent. Never has failed so far, but I get knots in my stomach right up until it's time to break it loose. I've tried other release agents, that are supposedly formulated for the job, but they don't work any better than Kiwi. I light it up with a lighter and it liquifies so it coats the surface without any voids. I'll rough up the stainless plate, where the Devcon will bond to it.
 
I would just buy an ebikes.ca torque arm and build an adapter to it that hooks it up to the disc brakes.

On the other side, do the lame hose clamp thing.

Should be good enough for a 1000w build.
 
neptronix said:
I would just buy an ebikes.ca torque arm and build an adapter to it that hooks it up to the disc brakes.

On the other side, do the lame hose clamp thing.

Should be good enough for a 1000w build.

Ya that would work. The picture is a rough sketch. I'm going to see if I can just blend it in with the contour of the frame so it's not obvious. Like an e-bike version of the frame. I have another idea. Instead of one 3/8" thick torque arm, sandwich two 1/4" pieces, like two arms on one side. The 2nd and 3rd picture combined, and bond then together. The first is machined/ground to the contour of the first triangle. The next arm matches the contour of the bigger triangle and blend into the frame. Then painted to match.

I need to measure my axle to see if that would work. I don''t have a lot of tools, just the usual garage stuff, but I think it should be pretty easy. But maybe not.
 
I did a bunch of measurements and it looks like the double/sandwiched design will work and fairly easy to machine/shape. It is fully supported along the edge of the frame where the dropout meet the chain and seat stays, and the main forces for acceleration (yellow arrows) will transfer to the chain stay, and the forces for regen (blue arrows) will transfer to the seat stay. If I make an are for the right side, it will virtually be the same or mirror of the left side. One cross bolt (bigger red circle) will anchor it, but won't be subject to any of the forces. I'll put two pins between to two arms, but they'll also be bonded together with the steel epoxy. The thicker red and green lines in the drawing are captured/supported by the frame pieces. The part of the frame on the left, at the junction of the seat and chain stays is about an inch deep/wide, so pretty beefy. Both the left and ride sides have just enough axle to accommodate the 1/2" width of the torque arms. Looks like the stainless steel plate will be about $18-$20 in material costs.
 

Attachments

  • Torque Arm Left.jpg
    Torque Arm Left.jpg
    97.2 KB · Views: 774
Make an L shaped torque arm, that will bolt to both those disk mount holes. then bolt up the brake with longer bolts, and if needed, washers to bring the TA and the disk mount to the same plane.

The other side, looks to me like the derailleur hanger is a pretty effective TA, when used with a fabricated TA on the other side.
 
dogman dan said:
Make an L shaped torque arm, that will bolt to both those disk mount holes. then bolt up the brake with longer bolts, and if needed, washers to bring the TA and the disk mount to the same plane.

The other side, looks to me like the derailleur hanger is a pretty effective TA, when used with a fabricated TA on the other side.

I must say I agree with this suggestion. The L shaped TA avoids all the complex geometry to work around and provides plenty of leverage. This is the best solution. I just went through this process for my 1000watt hub and I was lucky to have simple flat surface dropouts on the rear. The hole rather than the slot in the TA is also a great safety idea. I did that as well. I used 3/16" 304 stainless cut on the mill and hand filed to a perfect fit. I believe it is plenty strong at that thickness. 1/4" certainly would be enough.

Epoxy is not as strong as you may think and it has a different modulus of elasticity from the steel. This will promote differential motion and rapid failure.
Robert
 
Back
Top