OpenEbike - open source frame design

conrader

1 mW
Joined
May 17, 2016
Messages
10
Hey all,

That’s my first post here - so I’m sorry was not yet properly introduced. Me and my colleagues are in the ebike business for more than two years - building batteries, assembling ebikes, selling parts. We’ve come to a conclusion that there is hardly a choice for us for an e-bike frame. We have some 3D printer engineering background so we figured we’re going to design our own.

The design guidelines were simple:
- modular and customizable (no welding required) - you just bolt in parts
- ultra light - we use honeycomb structure to reduce mass
- full suspension
- effective turn radius
- not more than 90mm thick but with an option to change the thickness
- interchangeable batteries

sSmz2Wa3G0RLSCeISAZEDFyoxKKfC9KN0hMSUm3nSDArMkOoSDeaI9YTBLTofDBzUqxHbghRmdqS2ma72iX8TzQaQwb1_ICrszBFO24clCnXsFKjXsaUVO9VNeS6_yK1k4KDxUcOYzsY13StPshvxh24TaJNugMM7UNU2GxnmRTBJ3tfWYyr1UUX2OkC9buiUfxq3AaoG0dAORNa7t5Xs9RQjGrkiQUOUWlEFzjIJeKG80fPmCFW5aZh84lxSrGjNf3LyvixZ3scvmBLLpKfA1_T6_laP8w0rXaJaVPVTBlgXjPgku_k-aPvpWg3Ph4zq8QYTaXxohC7lMxKeRjJqnrzEfx-IwI53h0Xvhbxil7qu9n5UFb-mGbVoOsmMB1ovDso7rQYe3_HI3N57bKcPyjvVobTDkoWMGPAV1zkB_DVp0_Dw8o5bEPvza3kt_q8Gf5b3UhqrlYp6393NSEn8vxkv5ynJA8CmNsca4IuN-d2tECRJVY0W46SUk1ckZfODrzupbZmC31AqsRSRk-kzv6pKBcjwwKXMNgT00d24XCwUGA9iaIVNzXZ2YOS1MJ7Om5xW6Zaz6-SLThpZAaRbXq6lepkYmRnuaQ8VDgDNOZU5BRk_VHNETU2Vd5_DKRvnOqJDOyPD_o16Gqy2cAojLwdp6cnuJAjtJo=w1115-h699-no


As there is no welding required everyone can basically assemble such a frame from flat parts. The design itself does not put too much stress on bolts. It also allowed us to mix steel and aluminum parts together - effectively reducing mass and strengthening critical parts. Once you will own a frame like this you do not need to commit to certain shape or parameters. Just by changing a few parts you can basically alter frame properties at a very low cost.

We would like this to be more than a frame - we’re thinking about it as a platform to design and customize an ebike specifically for the needs of each user.

We hope we could have a working prototype on Christmas.

The best thing of all is that this project will be open source. You will be able to download project files and cut your own parts or help us with further project development. Soon we’re going to setup a website where we will share project files.

As for now, I think we’d like to have some feedback and perhaps some encouragement. It would be great to know there is also someone who is ready to participate in this open source project.

We can’t decide if the project should be called OpenEbike or osEbike. Maybe you could help us make the right choice? :)

We'll be grateful for any feedback!
Conrad
 
conrader said:
- ultra light - we use honeycomb structure to reduce mass

Those honeycomb spacers might look cool; but they are not triangulated and will flex in plane and could shear under high stress or impact.
 
Buk___ said:
conrader said:
- ultra light - we use honeycomb structure to reduce mass

Those honeycomb spacers might look cool; but they are not triangulated and will flex in plane and could shear under high stress or impact.

Yes. I thought that the lightweight honeycomb structure used in aerospace were typically sandwich structures. The thin top and bottom sandwich layers resist the shearing forces.

And while 3D printing is cool, isn't there a strength issue with the layering? From what I've seen parts are frequently strong in one direction but weak in another as compared to homogeneous casting/molded parts because the parts have been built in layers. So you'd surely need to specify materials and printing technology/procedure to ensure strength.

Frankly, I'd be pretty hesitant to ride a 3D printed bike at any kind of speed until it had been well proven. If you are committed to 3D printing technology for this project, you might find this of interest.

https://www.3ders.org/articles/20170109-mit-researchers-use-3d-printing-to-develop-material-20x-less-dense-10x-stronger-than-steel.html

or this ...

https://engineerdog.com/2015/03/08/3d-printing-a-3d-honeycomb-infill-concept/
 
wturber said:
Buk___ said:
conrader said:
- ultra light - we use honeycomb structure to reduce mass

Those honeycomb spacers might look cool; but they are not triangulated and will flex in plane and could shear under high stress or impact.

Yes. I thought that the lightweight honeycomb structure used in aerospace were typically sandwich structures. The thin top and bottom sandwich layers resist the shearing forces.

The honeycomb in aerospace/F1/auto laminates runs the other way.
HTB1Zh.iLVXXXXclXpXX760XFXXXY.png


That is to say, the open cells are closed top and bottom by their bonding to the top and bottom surfaces; and that bonding closing the cells is critical to the strength.

It also puts the rectangular sections that form the cells 'sides' in combined compression and tension -- like the web in an I beam -- and in three directions and dimension. And as with the I-beam, by holding the two tension members apart, and preventing 'crush', it gives a much higher tensile and compressive strength than a solid beam from the same amount of material.

But, the honeycomb spacers shown in the CAD image are open cell and have low cell wall depth. Effectively like a bastardized planar truss, but without the triangulation. Not good.

Edit:Also
And while 3D printing is cool, isn't there a strength issue with the layering?

Although he mentioned 3D printing experience, the mentions of "flat packing" and "no-welding" led me to think he was talking about assembling from parts laser/waterjet cut from flat sheet/plate rather than 3D printing.
 
Ok, so first of all - this frame is not going to be 3D printed. Steel and aluminum laser / water cutting only. Second of all - this is an early drawing. I'm going to put some more of them soon but the honeycomb structures are going to be covered with 1 mm of flat metal sheet to reinforce it.
 
conrader said:
... the honeycomb structures are going to be covered with 1 mm of flat metal sheet to reinforce it.

Just covering the honeycomb will do almost nothing for its strength. Those coverings would need to be permanently bonded to the top and bottom edges of the cells to give any extra rigidity.

In a manufacturing plant, that is usually done using: meticulous jointing surface preparation; robotic/mechanical delivery of controlled amounts of specialist epoxy to the joint surfaces; mechanical pressure (press or rollers) applied to the top and bottom skins; low pressure autoclave to ensure bubble removal from epoxy; sometimes heat while the epoxy sets up.

Reproducing that as a DIY project is fraught with problems.
 
conrader said:
Ok. Well, let's see what we can achieve with just bolts.

Here's the problem. Without triangulation or the cell ends being braced by continuous closure, there is very little material to prevent this from happening:HexagonSpacerShear.jpg

Also, producing hexagons from sheet -- through a subtractive process:laser cut/water cut/punching -- is an inherently expensive and wasteful process; with more material you've paid for being discarded than used.

Please note: I'm not trying to discourage you -- it sounds like a cool project with great goals. I'm just giving you a heads up that you will need to get some structural engineering skills on board earlier rather than later.
 
Fortunately those honeycomb sections on the frame are NOT paper thin like the examples presented later. They have considerable thickness to them, but are strategically hollowed out where strength and rigidity are not compromised. Also that picture is a computer rendering. Who knows what the final results will look like.

Keep up the good work. I hope you produce flat pack frames that can be bolted together out of a box full of flat parts. I'd buy one!
 
Conrad,

To make the project work for a larger number of buyers ,

1) Design the Frame around the more common, and much better priced forks and rear shocks.
In other words Design the Frame to use Forks with 140-160 mm of travel, same with the rear shocks 140-160 mm ,
Also it is getting harder each year to find a fork that has a straight head tube.
Most forks now days are 1 1/8 " top and 1.5" bottom. ( Tapered ).
So
Make the frame / head tube a Tapered one.

For Bolts use what Airplanes use , AN bolts.

A couple of models will be better as well, one for rear hub motors and one for mid drives, so that the rear swing arm can use a through axle 12mm x 143 mm , rear wheel.
 
Hello guys,

thank you for your extremely valuable feedback.

1) First of all please help me understand the term triangulation. I'm not an engineer myself and I will have a hard time explaining something to my designers if I don't understand it clearly.

2) Having the frame modular allows you to basically customize it for your parts. So you can change just the head to fit 1/18'' or tapered or you can change the rear damper mounting to increase or decrease the size of that shock. The main advantage is the ability to fit different replaceable battery containers. Either standard ones or custom made.

3) AN bolts? I will look it up.

We've made a considerable progress on the project since my first post. I will upload a few new images tomorrow.
 
ScooterMan101 said:
For Bolts use what Airplanes use , AN bolts.

Don't do that. AN fasteners are not extra strong or extra good; they're only extra expensive. And they are highly traceable and certified, which doesn't help us nearly as much as it does the aerospace industry.
They're also plated with super toxic cadmium, which is bad news.

Grade 8 industrial fasteners exceed AN fastener specs in every way, for a lot less cost. But they're specific to inch sizes, so don't do that either.

10.9 or 12.9 rated metric fasteners are what you should be looking at. Stainless fasteners are not as strong, but are tough and ductile and fully satisfactory in most applications.

As to the hexagonal mesh, just substitute triangular mesh and the specific deficiencies mentioned will be corrected. What won't be corrected is the fact that the frame as drawn has extremely compromised torsional integrity because there's little to prevent relative shearing and twisting in the side panels. Either think in terms of making a frame unit that's basically tetrahedral, or else you'll need to furnish interior bracing to accomplish the same thing.

The swingarm you've drawn... just don't get me started. Instead, reflect on why you've never seen a swingarm like that before. Consider the forces acting on it and what's necessary to resist those forces.

Anywhere you have a rectangular opening without a diagonal bracing feature across it, you have a built in weakness. That's the triangulation folks are talking about.

I think you have the gumption to do real mechanical engineering, but you're going to need to back up about seventeen steps and study basic principles before you get too deep into a project like this.
 
I understand what you mean Chalo. This swingarm is intentionally shaped like this and we have designed specific countermeasures for the forces that will affect it. This is an open swingarm which will allow us to use belt systems instead of chains - I've should have mentioned it in the project goals.

We have a steel hexagonal rod which should counter the forces you mention but only experimenting and prototyping will answer if this is a solution.
 
I like the idea of what you want to do, but I think you've got a steep learning curve ahead of you. ;)

And I get the feeling your design team is obsessed with hexagonal shapes. :?

You might consider going past that into first doing what's structurally required, and then perhaps add surface etching or embossing or similar (or just decals/paint) to give the appearance of hexagonal structure.

Hexagons are an efficient shape for storage (like honeycombs used by bees), but they are not structurally rigid by themselves.

You should look up some things like Truss and Trestle bridges, or almost any triangulated structure, to see what makes thing structurally strong.

There are also different parts of the frame that do different things. Typically, stuff along the top of the main frame is in compression, while stuff along the bottom is under tension, though that doesn't cover the torsional / twisting forces from front to back that occur while riding.

Similarly, the rear swingarm will have these forces on it. Additionally, I don't see anything to prevent torsion from twisting the swingarm out of shape, allowing hte rear wheel to twist sideways (for instance, top to the left and bottom to the right) vs the frame itself. It would be like riding a rubber bike, especially since there's nothing connecting the two arms together, so they would be able to pivot separately during this torsional movement and it would make the twisting worse.

You can make a swingarm that is still "open" for belt installation by having it bolt together as a modified "H" like many swingarms are (in top view), without it being completely open as separate arms like the only drawing uploaded so far shows. Then the arms can't move independently, and the entire assembly will be stiffer and less prone to twisting.



Made from extremely stiff and strong materials, the frame might work as-is, but I'm not sure what those materials would be.


(note that I'm not an engineer, but I have a fair bit of practical experience in building things that don't work as intended, and having to modify them until they do. ;) )
 
Again thanks for the input. The hexagonal shapes are just a matter of design rather than an engineering decision. We're kind of inspired by it so far.

I also believe it's very difficult to judge this design based on one poor quality screenshot - I'm attaching some more so you can look at it in greater detail. Keep in mind this is still a work in progress.

h5KPZ4DJexZXFJEpqi9_qVLA2rNStdObbuzKnabahFiVOQjOkynBYZ7ax6n-0w-k1t58TbH4ncbJH8iCEPpOhEWWbWr7GsQbooX_5t7tOhumi7Bb8H21dFr4BpUr7nDxZwaJvGap5BS47J36hLK2q2K5zLyP03FZYd-WaB7riKvx0dFDTYTF3jFo3jUHcFWbXNKyMcIz5PPD1khXS2Yih6Hzr6t5AwpW9K62aezGNCSUHn1buxpqHuBQ10z43hdleFPrHy1I7MnhdhM2Bq9HMEEloUizhA2B0ac2E8JshO4b9D45g47EzNAujPlphXTV5xj9wcQiIXD5HQjvHF_jKhwxufW444FLlSVQimJcb5R24VbOm4tPtuFzZRdHT022PvACAJ2QQsdLxdMnp-VE_LFbRH-qRFOp6NiQ8N9E0PKn8TKyWbeYmGNPT_55-rbzCPXri90HhzA_2ihB3GWHQzBdcWtNcrZhOO2SyoD_0sT0FWoBBaNBsqHXicDfh6flMJNCaYSYl_BudnJFYGtIiu39XzRjt6-9dL43PNjlUXxCljFzjr4ibCQ-h-thypc3YniptXdUHp94XshHhsy-1ExtJdXKE2mgxQRqfOgzvQ=w737-h650-no


D4sxB5NRbX9JfC2ZFjh0kp7mLtNDlraMeFXuv-CbaPgahhMv4tQx8_oDRRbyMgKnGBfmAhW7UV9lWmhk08AoR7fS31SHLERKjnnpaj4lYGIdBu2ZIQhl3ZcGfg64TtAOhDkdb27Rax6atwGx9h0phCjKgJNBR0Xsvnd2YktsgWJPOXsXGBK3gAh5DixlP_0TzjEiHvT8ojKcvCTvf8BkMkYt_PEyZMynck2bpGZ7o0R6j4HUOzNy4hNR9J426dWdA5ZMjXrie4JPUk9QOlLoRD4ae8Ym5BtUQcTT0PfgOmh2eE_582JXfGDPmPm4-sv85iPJl7CnkQFF9e8lnDwFHnXiN7H376KvkB8n_0kKC53xV7POYzt9xRhX1twT8JVutO6steRAjX37YObAoEcOvZRx6AgDZ2McKmgQyAt4xb0YxKhvD4eqZJOJAwnhsF0_BkNsYK4LzKhhhbqvHm_7Kw4dexqK5bLWCR2SRp9KFus3sc2W13j-ZZPHNmUeadobYugh52TiR_NGY-uoILpBWnFNtSCUP_GAIvrlGG70ZXbsOrop0kReXHXrPI-t9_zuhPxgoYan5r80klNxNZfCacpc5dGCNhqsStaaZ6ao93Q3JhDkCzijaWiNmC2ZCo8xW2GODqB_2q6-LfU-5EwzX-GKsTdatzJOwMA=w1096-h695-no


FGBxM3Ie01GIPCtF1fxJK1FutOqHHkxBZxnHmWoTEOvdEKmYLgVGmphjsZ1NHsZQDjF_cGyMwJzkv4NkFRzCNIRzTTfIS9qykju-Nw_Dd0SwN31Fl5bRT0_1b-1HhUoMQuZurO0IFT1wv6i4YqFrvRDvySqdpJoNxuEDJ_WyFyvlI9ECPxGUldaR4Ux5-KrtAU20L_Gs2vzUtYRXUMDgla8aCLxDPRkgDRCfP9frQ6WbSEKjClmjar3C5-4Nq7Ch-prTX7BVo77Zx6Ghxws7wMtI74ZIRre5cNFS5uW9h_Ccb3LL83hUV1jGdZbsb7HlE8661Ukc_tM_z8Zcu_a76z0_Zxs2ZuNUUFDAcEN44umhpci6pPzofmlyz9sdrcitEYTYg1Z0EuKVWpvsiXSm4GlQWOaEutb4RuekhU05_tDT4a_LOXWvjCbAiuWHwyUFid_Q4pM_NpeROpFDF8SUk90t5iz7nZyQ694u7xXLBEBBHqcava_0HHogojAWeVCI7Hh_fLUfbblWu1E5u9LOGyBEZhIorzxK_379KiyxJUkjg0OOnWqeoOs8mpV3-k7kASXbvsQNM4qFzF55w023-7WPL7SfVHF9HRHtHgVRfnHPClewIAn7qc6fmbhcoqMNBUU8MZlgrn7QGkG6DXX6DNfrm9a-KlWxQok=w1150-h695-no


mVF-W57rkSxqOEINpyL0orReLMEJ3xdkHWJPnSjbU4WlpKKs4XrFk1sW23fOkUMtwZSW1rM7X9jsOzbELXXfS8exsloNp5kkfEPITzI_faU5trBJjmkVNRlsgWrAM-W-zo510k-Bqf5R1JC43lQtd8X4V1NkV7ixpVEi5RT9DgZe6gdqitSH5TwIaXTTRg2ohI9osdrvfpWUPCIWxRofpS6KP1A6yt8IINieiA8UKke1688WrX2_aFIpgED70yLQ2vyBVxW0JQQfHO_6nLSPXHclbzveQ77gZZj7z-g3Or4pWW9SZ-FKNaKpQN4rHq9alW0uYLQmjOLqnWE2DeUoQZ2AR01M8z-oZtSUF1ZMjWzWOu6XtHfMu3NdaKviET0W6cdHtPjVjaYx0L3tKg1uz4WztM26uIX5vYiuZVDLkX-iAQaIH_gdNn6NJOBkFZPS2dIPKxh3A5MH6h810AozldMve70AgNEaWDxOXp4nzOJ6Lzf--7UyoThRi-KyFfBDQqsl63P1QDK7ucug8dbMXXo7xskGro8ZlDQPG-SRrUYcDXEgt1sw2HT_IHNX3rr6y6QCSmcfkkxEYDiIrG9dqbnOuniCTKTcwFq84QK2WkgirXqvD1gZKC13KgJ_TpTogLN3ghwZi73JRBz0QHcbC88-G-39MOKW5gM=w1084-h695-no


EcfqopYGa28HCkw9I7q5uLBhN8Gw9U_AoQQPAvoYfmtnyZLEDzZoDDYSGKNJf-6pW8Sx4O8tfLOu1hNAEjRJ8vEd0SlIgbJFQCLT2eymHYNFMRbLQI9W-laO9ScUYx7BdtJfVKz7xe-v13VHhJIXsF5k7gKQCepE8ucsIgRuPhRPcNW0npB9zmsNB_omsrW730FrqBZgJYzBB0xF0qOorr0DlpKruy0XfHn8XVpmJd1AAIOikNU_3II0gS_t8UJlz1JsbvRE9cF1ClT_0qoAL39YQywVeKbhMb8ekufWM6HGP8uk33ukDTT7_zzcx4ldS9dh4MHlsfMp1jzduaOle2SaHxM1BD2OZYpKx8NJ9NQRKOnqgPvP-HEJE88_luHBD5q36UCFbW8zvtZYinjOb0VPOwuIouX0FM58hWhlpXrS6VNoy_kveS_gnSbQGEoejbG8v1oZlFFqZ5Pg5KfK3O-XxQNW6hXdGRsw9dA3sAy0dgs3fMAJzalKiC2t-70JSWyB0Z5hco6vyeUoJo61qdhDdlQew37UcqPbmQ7jzXqepzWvVO_CbzBsoqVlKCsGTZA_5KWgfEJC6D0dRfdy3c21jkHIp-GpR_-dyLBglU8yXD9HXeKqal0Qu7EGzuDMruQlG4gih2lvyctW-PkUbLr73gbGUgjUGug=w1090-h695-no
 
None of the pictures show on the latest post. Also, I think the point made about the structure are not only generally valid, but also easily determined by looking at the drawing given. As drawn, the design has problems.
 
Perhaps try attaching the images to the post itself (uploading htem to the ES server) via the Attachments tab below the posting text box.

Then anyone that can read the post can see teh images, *and* they will be archived/backed up with the post itself, so they will always be there for viewers to see, even when the external site is long gone or charges money to view them (like Photobucket does now, breaking millions (possibly billions) of images across the entire internet).

Note that the original image in the first post, that is on the same external website as the latest unviewable posts, is visible.
 
Please let a real engineer "design" the frame instead of a "designer" who is "inspired" by honeycombs.
Or get a designer who likes triangles better than hexagons... :roll:
 
My bad. I forgot to change sharing settings on these images. I hope they're visible now. As for triangles vs hexagons, this is an opensource project so you're free to change these parts (as soon as we will publish the files).

BTW. we've made some structural durability simulations which we will share later. According to it hexagons work very well.
 
Here is the simulation of both structures. Triangles are in fact stronger.

Conditions of experiment
Dimensions of beams - 548x60x10 mm
Distance of support points - 524 mm
Strength applied - 1000 N
Weight of a each single beam - 1.005 kg (mass exactly the same)
Material for calculations - steel
 

Attachments

  • image (6).png
    image (6).png
    33.4 KB · Views: 3,282
Back
Top