Custom frame for the Cyclone Coaxial motor

Chambers

100 W
Joined
Nov 11, 2011
Messages
298
Location
Brisbane
Following on from this thread .. https://endless-sphere.com/forums/viewtopic.php?f=28&t=88921&start=225
--------------------------------------------------------------------------------------------------------------------------------------------------
EDIT: Lots of changes and refinements happening these are the current versions -
cyclone coaxial -26.jpg
EDIT 2: OK not quite the current version anymore :lol:

Follow on from here to see how I got to this point...
--------------------------------------------------------------------------------------------------------------------------------------------------
I'm considering a custom frame for the cyclone coaxial 4.8kw motor ( could also be for the dual drive set-up)

Here's where I'm at so far..coaxial -4.JPG

Note: The forks seat and rear end are from GrabCad - I just designed the frame in the middle.

Down tube is 85mm wide the whole way - tapering out vertically to 140ish at the motor mount.
Should easily fit 20s 6p - 7p or 8p if the whole down-tube was filled.

Thoughts for improvement?
 
If you have a non fatbike hub, the motor will have to be offset to poke out on the non drive side. Then it would make no sense to have the frame 140. Since on one side the motor will overhang, the other side the frame will. Just keep the frame 85mm or whatever.

If you are going from 0, dual drive motor would make a lot of sense, depending on power. If you have at least 4kw, 1 gear, geared for 65kph with field weakening (54 without) is just fine.

Also I don't see a spot for a controller.

Also this shape of battery is much better for center of gravity. You don't feel the weight if it is low as you lean in corners. IMG_20200608_101016.jpg
You can still have the top of the tube accessible (maybe from the top) as a storage compartment for tools or electronics.
Though bottom access is probably better for strength since is only one cut and more weather resistant.
 
@ Tommm,

I think you misunderstood - Or more likely I explained poorly....

The frame is 85mm wide - As in between your legs

The 140(ish) section is vertical connecting to the diameter of the motor mount - The controller could live in the down tube above the battery - Maybe.

As to the battery placement - that was my first thought also but I was struggling to get a decent size to fit and clear everything - But I have a few more things to try.

Thanks fro the input
 
Controllers monted in a box get hot very fast. Even controllers that aren't even warm to the touch in open air. You would need to get a good connection to the frame itself to use it as heatsink if you were going that way.
 
Good point - I was thinking of having a cut out on the underside for the bottom of the controller to stick through slightly.

But you're right - open air would be better.
 
Couple of updates.

Moved the down tube forwards to allow the battery to slide in similar to Specialized kenevo.
Usable battery internal space in this revision is 615L x 81W x 87H -

Also curved the down tube to add enough clearance for a 200mm travel shock at full compression.

coaxial -5 annotated.jpg
 
160 cells at absolute max - this is with 0.8mm gap between cells in an offset configuration.

So 140 with BMS should fit comfortably

160 cells.JPG
 
Chambers said:
160 cells at absolute max - this is with 0.8mm gap between cells in an offset configuration.

So 140 with BMS should fit comfortably

160 cells.JPG

What material are you planning on?

https://e4bike.ru/page/battery-shape-configurator?c=71920302009000110065003791433726539579461526429462387
rrrrr.png
I think you need way bigger space. You need a box and padding on the outside and inside of it. 18650 aren't exactly 18mm diameter, more like 18.5mm if you look at spec sheets + add space 0.5mm for glue or holder.

With 87mm box, you could fit only 120 with the same settings.
For 160 cells, which I highly recommend, you will need 94 x 633 mm, which if I look at your picture, should fit, as your pack ends at 615mm with a bit of space left at the top.

For BMS/Controller you would still need the shape I drew. If the 2 sides on the bottom were openable, you could slide the bms/controller out too. You can get nucular 12f, 86х43х125 mm, plug and play FOC and up to 200A battery, I would recommend it.
https://e4bike.ru/page/battery-shape-configurator?c=71920300000000110065003791433726539553540402453393347
rr22.png
 
What material are you planning on?

Probably 6061 Aluminium - Maybe Steel for the first iteration

With 87mm box, you could fit only 120 with the same settings.
For 160 cells, which I highly recommend, you will need 94 x 633 mm, which if I look at your picture, should fit, as your pack ends at 615mm with a bit of space left at the top.

Edited the down tube to 100x85 outer dimensions, so if Aluminum with 2mm wall that's 96x81 internal.
Also edited my cell circles from 18.8 to 19.2 -
The bottom corner few cells will end up in the diagonally opposite top corner.

coaxial -6.JPG

For BMS/Controller you would still need the shape I drew. If the 2 sides on the bottom were openable, you could slide the bms/controller out too. You can get nucular 12f, 86х43х125 mm, plug and play FOC and up to 200A battery, I would recommend it.

I already have a 24f on order from a previous abandoned project (became too much of a motorcycle) - I was planning to swap to the 12f. Controller placement is planned as per the image - Probably with some sort of shrouds to tie it in to the frame visually.

Anyone have the dimensions of a good 20s BMS?
 
How long ago did you order the 24f? I have some 12f's ready to ship, but shipping is delayed and I could really use another 24f. Maybe we can do a swap of place in the queue with Vasily where you get your 12f quicker and I get a 24f quicker. I'm unlikely to use all 6 of my 12f's that are ready any time soon. Changing one to a 24f might convince me to spend the extra on shipping. That will only work if your 24f is early in the production queue.
 
Another option? Different battery box shape - Kind of a combination of the first two.
Battery would have to be side load.

Any bigger than this and it becomes a lot less bicycle looking and feeling - But this would easily fit 160+ cells and BMS etc....

coaxial -7.JPG
162 cells pictured in a generic block - each circle is 19.2mm
coaxial -8.JPG

Thoughts?
 
Chambers said:
Another option? Different battery box shape - Kind of a combination of the first two.
Battery would have to be side load.

Any bigger than this and it becomes a lot less bicycle looking and feeling - But this would easily fit 160+ cells and BMS etc....

coaxial -7.JPG
162 cells pictured in a generic block - each circle is 19.2mm
coaxial -8.JPG

Thoughts?

Newest ant bms goes down to 15mm for 20s and 18mm for 24s. Old one was 25mm.
https://www.aliexpress.com/item/4000529723243.html

15*80*160mm in total


An idea
You could flip the battery 90degrees, so the downtube would be 80mm tall but 100mm wide. It would look like a bike again and the weight would be positioned lower. Making the box smaller also makes it possible for the rear suspension to have a oil reservoir piggyback.

For reference, a hailong pack is 90mm wide, 100mm in width is no biggie.

Then we could go back to a nice shape like this and only need to find space for the bms.
file.php

I still think the controller is better placed under downtube, it is 700gr and putting it 10cm lower is an easy gain, the nucular is ip67 when potted.

Do you have a plan to use bearings everywhere vs bushings?
Also do you have an idea for final head tube angle, rear travel, chainstay length, reach?
 
Newest ant bms goes down to 15mm for 20s and 18mm for 24s. Old one was 25mm.
https://www.aliexpress.com/item/4000529723243.html

15*80*160mm in total
Cheers


An idea
You could flip the battery 90degrees, so the downtube would be 80mm tall but 100mm wide. It would look like a bike again and the weight would be positioned lower. Making the box smaller also makes it possible for the rear suspension to have a oil reservoir piggyback.
Maybe - Will have a look

Then we could go back to a nice shape like this and only need to find space for the bms.
file.php

I still think the controller is better placed under downtube, it is 700gr and putting it 10cm lower is an easy gain, the nucular is ip67 when potted.
This is my preferred shape ATM too - BMS takes the space of 9 cells so if I can find a place for them that's a win. I will play around with the controller anf BMS placement tonight.

Do you have a plan to use bearings everywhere vs bushings?
Also do you have an idea for final head tube angle, rear travel, chainstay length, reach?
Definitely bearings everywhere - Chainstay will be as short as practical, Initially looking at 150mm travel (ish) front and rear, but I want it to be capable of 180/200 just by changing some parts (fork/shock). Head tube will be slacker than pictured but not 100% decided yet.
 
Quick render with a person for scale - I have flipped the downtube 90 degrees in this image so 85mm vertical 100mm between legs I think it does make it look visually smaller

Trail Bike.jpg
 
Chambers said:
Quick render with a person for scale - I have flipped the downtube 90 degrees in this image so 85mm vertical 100mm between legs I think it does make it look visually smaller

I think steel would be better material for the motor clamping part, less chance of a rip happening. You have to get it pretty tight or the torque will spin the whole thing.
Also I am working on a plan B in case the coaxial gives some problems, to go full moped. I have a good hunch the qs motor 2000w mid drive would fit right in. It is a beefier motor than the coaxial as it has no internal reduction, but the kV is lower. So you wouldn't need a humongous reduction, just 3:1 or similar (coaxial is fine with 1.5:1). Only change needed would be some foot pegs mounted somewhere. I already asked the diameter for the motor, and will report as soon as they respond. They only advertise stator diameter as 120mm.
https://www.aliexpress.com/item/4000076478471.html
 
I think steel would be better material for the motor clamping part, less chance of a rip happening. You have to get it pretty tight or the torque will spin the whole thing.

I have an idea for that but will have to wait until I have a motor to check the exact dimensions.
The plan is to put a flange on the circular mount to pick up on the motors bolts - Rather than relying on clamping force
 
To the watchers out there...

Am I on the right track with this? Or should I be looking at something closer to the FUTR beta? Or?...
I do like the simplicity of the Beta - It does have some shortfalls though....

is 160 cells enough? Too much?

coaxial -13.jpg

I'm chronically indecisive when it comes to e-bikes :roll:
 
Chambers said:
is 160 cells enough? Too much?
To determine that, you have to answer a few questions for yourself, and do a bit of figuring.

What specifically is the expected power usage? (wh/mile) If you're not sure, then...what's the terrain, speed, wind, weight, riding style, etc., to make a guesstimate of the power usage?

What is the expected range under those conditions?

1-- Those will determine how large a pack you need, as far as total capacity.



Then..what is the peak and sustained current you expect from the controller?

What is the specific cell being used, and what is it capable of? (for real, not imaginary vendor specs ;) or optimal-condition-lab-test-spec-sheet stuff)

2-- That will determine how many parallel cells you need.

What is the required nominal voltage of the system? (how fast do you need it to go, vs the kV of your motor, vs any gearing between it and the wheel, vs the wheel size)

What's the nominal voltage of the cells being used?

3-- That will determine how many series cells you need.


1, 2, and 3 will then tell you how many cells you must have, minimum.


Then you should add at least 20% to account for pack aging over time, and for detours, extra headwinds, etc., to ensure you still get the range required over the expected life of the pack. More, if you prefer extra insurance against running out especially as it gets older.




Some other points to consider, if you havent' thought of them already:

Stuff that gets hot next to batteries will change the performance of the cells it's close to, because it changes their temperature, while not doing so to the rest of the pack. It will also change their aging rate. The hotter they get, relative to the rest of teh cells, the greater this issue will be.

I would recommend some insulation between the hot things and the cells, especially the motor.


Regarding the taller vs wider battery, I'd go with one that's taller, and keep the pack as narrow as possible. It will help you with q-factor for pedalling, and make it simpler to work with and/or easier to find for common good-quality BB parts.
 
To determine that, you have to answer a few questions for yourself

Thanks AW -

Sage advice as always.

Regarding the taller vs wider battery, I'd go with one that's taller, and keep the pack as narrow as possible. It will help you with q-factor for pedalling

I agree, I made a cardboard model earlier today and the width just doesn't feel right - I'm reverting back to 100mm vertical
 
What size are the wheels in the comparison pic?
If you want to keep up with the newest trend, the front wheel would be 29 and the rear one 27.5.
A smaller rear is seen as an advantage for ebikes as you can have a big lesser reduction and the diameter of the wheel is the length of a torque arm acting on the drivetrain and motor.

My coaxial has 160 cells, in addition the weight is higher, inside the triangle, it is more like blasting through stuff rather than doing technical trail riding.
If you want light weight, don't worry, you can have a 60-80 cell 72v 60-80a pack from vtc6 or hg2 strictly for trail.
 
A few more refinements....

- Battery now inserts from underneath
- Down tube is 125mm x 85mm outer dimensions
- Working on the suspension linkage which will be easier to construct and allow a bit more travel if required, I will have to change the vertical tube to a twin spar arrangement for the shock to go through.
- Changed the forks on the model to 888rc2 simply because I know the model is dimensionally accurate.
- 29 front 27.5 rear.

coaxial -16.JPG
 
Chambers said:
- Working on the suspension linkage which will be easier to construct and allow a bit more travel if required, I will have to change the vertical tube to a twin spar arrangement for the shock to go through.
Why not position shock so it's about parallel to battery, with fixture next to where horizontal tube and battery compartment meet. With dual linkage (which is stronger than single-), so no need to split the downtube.
 
Back
Top