Dani's mid drive (GNG motor) - CAD heavy pics!

[Dani]

happy to give food for thought. Rotating on slides to set the tension is a nice idea.. but I think it will take more space/material this way. Still interesting though.
I also secretly thought about the same setup, but instead of a huge drivebelt with high gear ratio.. why not use another gear? Real gears, made of metal. Mhhh, yummy. 8)
This would make it smaller overall, the gears can be placed between the sheets. Too much stuff in my head, so little time..

Well, how about this for a frame mount?

two small noses are bent from the aluminum sheets, two cheap clamps screwed on to it. Simple.
If your bikeframe is not wider than ~36mm this would fit to any bike out there (you need maybe 3 sizes of clamps and plastic half-tubings/whatev, which is easy to get in any size).

 

[speedmd]

I went with the #25 chain on the primary side and made eccentric jack shaft bearing mounts. Very compact. Narrow bb and two front road rings. Q is not much different than my road bike. Need a bit more independent secondary adjustment. Thanks, I figured out now how I will do that. It is noisy compared to the bbs02 setup for certain but the speed this thing goes it is good to let folks know you headed toward them in a hurry. Will try a belt some day on another build.

 

[Samd]

Hi Dani
You have me thinking that both chain adjustments can be accomplished in just the top bracket if you allow for that bracket to rotate about the motor a bit!

A few folks have suggested an alternator style swing down assembly in the history of this thread. Just need to get common tension with the primary and secondary. Both when new, and once the chains stretch in a little under use.

Hey Dani,
suggest you keep the underside enclosed to keep the mud and dust out, and use more cross braces for torsion.

I'm thinking of making a HDPE and 3d printed enclosure for mine.

 

[t0me]

Consider in your design, making the top mount of the system (to the bicycle frame) independent of adjustments to the belts/chains.
That is, that the mount points are fixed permanently and tensions can be adjusted without having to touch those.
I think that's going to be a key factor, especially in high power applications.

 

[Dani]

Hi Dani
You have me thinking that both chain adjustments can be accomplished in just the top bracket if you allow for that bracket to rotate about the motor a bit!

A few folks have suggested an alternator style swing down assembly in the history of this thread. Just need to get common tension with the primary and secondary. Both when new, and once the chains stretch in a little under use.

Hey Dani,
suggest you keep the underside enclosed to keep the mud and dust out, and use more cross braces for torsion.

I'm thinking of making a HDPE and 3d printed enclosure for mine.

I'll screw a light/thin aluminum plate on the ground. I cant make it one piece because i can only bend around 1 axis in that case, not allowing me to do a second brace this way. Even a 2mm full plate, fixed with 6-8 screws to the bottom should increase the stiffness by a lot. and secure the motor. Or maybe I'll find someone with a 3D printer.

Consider in your design, making the top mount of the system (to the bicycle frame) independent of adjustments to the belts/chains.
That is, that the mount points are fixed permanently and tensions can be adjusted without having to touch those.
I think that's going to be a key factor, especially in high power applications.

I think I got that right, no?

 

[Samd]

3d printing you say...

 

[Dani]

Damn! I should build the whole thing with a 3D printer.. that is impressive, I like it. Did you do it yourself?

Oh, maybe I should start my own topic instead of spamming the original GNG BB Drive thread?

 

[speedmd]

A few folks have suggested an alternator style swing down assembly

Yes this will work to a degree. It still will have issues with getting chain lengths correct and chain ring size changes become a nightmare. Ideal would be to have two separate adjustments. Upper plates that would rotate about the motor and also adjust spacing to jack-shaft looks to be the cleanest approach to date IMO. Once adjusted, if you could hold mounts firmly in place, a alternator swing mount on both mounting points of the upper plates, it could do this also. Can not picture it exactly yet what a good setup for that would be. :?

 

[t0me]

You've got it Dani, I hadn't seen your mount till after I'd posted that.

 

[Dani]

Here's an article that might give you some ideas: http://www.electricbike.com/cadcam-the-diy-builders-best-friend/

Thanks for the link, didn't saw that one, yet.I love that blog, man. Read many articles there over and over. I just found out about the Prusa i3.. I wish I had the time.

 

[ebike11]

@ dani...
Will you be selling these brackets also??

 

[Dani]

Hey ebike11,

as of now: no. One day: possible. I don't have any working prototype yet. I don't know the final costs. This will take weeks/months. But nice to know there is already demand for it

Thanks for your feedback, guys! I really appreciate it. This community rocks 8)

Edit:

I love the lasercutting.. so I thought why not do this for the pulley:

You'll screw them together. Improvements to be done (less material, maybe two starwheels with thinner "spokes". Yeah, that sounds awesome. maybe something cool like this

or this

or my very own logo. Hm...

This is made of 4 sheets of 6mm each, gets me to 24mm. I know the drivebelt is 25mm, but I'm kinda limited to 6mm of Aluminum where I want to manufacture it. Maybe I'll look for another company to lasercut thicker parts.

Can anyone give the the exact dimensions of this freewheel? I can't find them anywhere
http://bikemotive.myshopify.com/collections/bmx/products/custom-acs-crossfire-flanged-freewheel

how about that?

 

[spinningmagnets]

Here are the dimensions: http://endless-sphere.com/forums/viewtopic.php?f=16&t=56565

 

[scorpionice]

Awesome concept...good luck...Cheers

 

[Dani]

Thanks for the info, very handy.

I'm still looking for a good way to make my own GT2 pulley (25mm, 90 tooth). Lasercutting is expensive (4x6mm sprockets cost roughly 50€ plus screws..) and the lasercut quality may not be good enough. I get problems with those tiny radiuses. It might not be that importend to have a very exact geometry, but I expect more quality for that price.

The company suggested water cutting, which is 4-5x the price. No way. Maybe I can save money on making a square, one-piece-cut for all of the parts

Edit: Way to go. by splitting 2 sprockets into two pieces each and recombine them like this..

I can save some bucks.

Also I think I'll go with 8M instead of 5M. 8M will take high power easily and lasercutting a bigger radius results in better quality cuts.

 

[bandaro]

Subbed. You seem to know what you are doing, is that solidworks?

Looks good, will be interesting to see how it holds up. Have you considered using a thinner stainless steel instead of the aluminium? What thickness aluminium are you planning on? I'm working on a similar sort of this with an rc motor and 3 stage reduction, so keep us posted when stuff arrives and breaks!

 

[Dani]

Subbed. You seem to know what you are doing, is that solidworks?

Looks good, will be interesting to see how it holds up. Have you considered using a thinner stainless steel instead of the aluminium? What thickness aluminium are you planning on? I'm working on a similar sort of this with an rc motor and 3 stage reduction,

Thanks. Yes, thats solidwork (2012). I considered stainless steel, but it is heavier and more expensive. I'm going for 6mm aluminum.

so keep us posted when stuff arrives and breaks!

hopefully there will be no update on this :lol:

 

[Lebowski]

which lasercutting service are you using ?

I'm busy learning FreeCAD these evenings, want to lasercut the disks for a 6kW Axial flux motor

 

[Dani]

which lasercutting service are you using ?

I'm busy learning FreeCAD these evenings, want to lasercut the disks for a 6kW Axial flux motor

It will be www.fabtools.de I don't intend to do commercials, but this is the best I could find. It's "cheap" compared to other locals and you can upload .dfx files and get a quote almost instantly. I chose Lasercut HQ, aluminium. Although it is german, you can easily use it. The keywords are very close to english, in case you want to give it a try.

I can recommend solid works, it's free and simple to use. Lot's of tutorials included and many more online.

 

[Miles]

I can recommend solid works, it's free and simple to use........

Free???

 

[Dani]

I can recommend solid works, it's simple to get for free........

Free???

Is that better?
It's free at last for me. As a student, you know..

cross post over here
Right now I'm stuck at the pulley. I don't want the full 25mm width, it looks too bulky and costs too much in lasercutting. Also the lasercut is bad for tiny radiuses. Instead I try 8M with 20mm. Main problem is the ratio. I'll end up with ~11 Tooth on the driver pulley, which sucks. But I think it can be done. Any experience with that? Anyone?

5M vs. 8M

 

[Miles]

Is that better? ........... It's free at last for me. As a student, you know..

If it's provided for you by your college, that's great. Last time I checked, Solidworks still charged for the, 12 month time limited, student version, for which you need student ID etc..... SolidEdge, on the other hand does have a free version: https://www.plm.automation.siemens.com/en_us/academic/resources/solid-edge/student-download.cfm? Not quite so easy to learn, though... Obviously, almost all software is "acquirable" for free......

Anyway, interesting thread. Carry on

 

[Miles]

I'll end up with ~11 Tooth on the driver pulley, which sucks. But I think it can be done. Any experience with that? Anyone

I don't think that would work very well... 22t is usually the smallest recommended for 8M. I've used down to 16t but I wouldn't recommend going lower than that. Also, the idler is bad for efficiency.

 

[Dani]

I'll end up with ~11 Tooth on the driver pulley, which sucks. But I think it can be done. Any experience with that? Anyone

I don't think that would work very well... 22t is usually the smallest recommended for 8M. I've used down to 16t but I wouldn't recommend going lower than that. Also, the idler is bad for efficiency.

Hm.. thanks for the input. I couldnt find any source for someone trying such low tooth count. I have no experience with timing belts, yet. In the end I think it's easiest to just use a classic 3 stage gearbox. No huge pulleys, no belts, pretty silent and somewhat cool sound. Rocksolid.

I could easily have a reduction of 1:50

first a small reduction with a belt (1:3,3) then first gear (red ~1:2), 2nd gear (purple 1:2) and last but not least the chain reduction of 1:4. Alltogether 1:53. But this is a different concept for serious machines I don't own.

All this headache because I think ~45€ is too much for a 90T pulley. If I had a 3D printer I could print it for very little money. Maybe I should go that way.

 

[Miles]

We had a thread about DIY pulleys: http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=26122