Suzuki eRM

[rsnoiser]

Hello everyone, here is the presentation of my new work: convert an old dirtbike into modern electric dirtbike.

After my first try on electric dirtbike with the "small beast" that I set down the power a lot for the kids: it was like a KX or YZ 80cc 2 stroke dirtbike but with the size of a small 50cc pocket bike!
It is fun for me but not very safe, even for my older girl who is 10!

https://endless-sphere.com/forums/viewtopic.php?f=12&t=111393

It was the time to do mine.

My attempt is building a dirtbike with the same specs as the KTM Freeride, at least.

It starts with a very cheap Suzuki 125 RM 1995 as a rolling frame in a really bad state.


For the powertrain, I choose to build a 72V 52Ah battery (20S13P) with molicel P42A cells to feed a mi-drive QS Motor.

After some cleaning job, it was the time to see what motor to choose between these 2 QS Motors:
- QS 138 3000W V3 with built-in reduction
- QS 138 90H 4000W with custom made belt reduction

QS motor told me that the 4000W will more convenient for me with a belt reduction stage so I choose it.

I bought some parts:

- a QS 138 90H
- a Votol em-200
- 265 molicell P42A cells
- a 300A Ant BMS


The Votol controller is totally unknown for me but I saw some conversion with this controller so I will give it a try.

First, I put the motor inside the frame to see the space left for the reduction.



Next time, belt drive reduction...

To be continued...

 

[j bjork]

It will be interesting to see how this turns out, I liked the other build too :thumb:

 

[rsnoiser]

Thank you!
Your builts inspired me a lot!

 

[litespeed]

Looks like a great project.

Tom

 

[rsnoiser]

Hello everyone,

I choose to make a belt reduction between motor shaft and rear output sprocket.
The QS motor have a 8M pulley but I wanted torque more than speed, it's a dirtbike and on the tracks, the average speed is only about 30kph.
With the motor shaft diameter, the smallest pulley that I can put on it is a 22 teeth 5M pulley with a 25mm belt and the bigger on the rear output is a 48 teeth pulley.
With a 10 teeth chain sprocket and the 50 teeth wheel sprocket I have a reduction of 10.9.

With the specs of the motor, it gives:
-about 650 Nm of torque at the rear wheel
-a max speed about 50kph without flux weakening

So, for me it's a good starting point.
I will see if it's rideable or not, and if a 5M25 belt will survive!

I designed in 3D CAD the housing of the stage reduction



Now, it's time to print in 3D to see if it's OK before milling those big parts

To be continued...

 

[rsnoiser]

Hello everyone,

After CAD modeling, I have printed these parts in 3D to see if alignment and dimensions were good.


There were small mistakes and I fixed them, now it's time to mill!

I think I will make a ton of chips!

To be continued...

 

[stan.distortion]

5M probably wont work out well, the standard 8M pulley is the first thing folks seem to be changing when pushing those motors hard as they shred belts with it. All the examples I've seen of that are on heavy applications though, much more vehicle weight. Hopefully it will work out ok with less weight and it looks like a triplex chain would fit in the same space if need be

Beautiful work, I'd bet that printed output shaft carrier would work as-is with some extra bearing support to spread the load.

 

[rsnoiser]

Thanks for your advice.

For the belt, I also have a doubt about the service life.
The belt installed on my mini dirtbike is a 5M 9mm and after testing, it seems to resist but it's a small dirtbike even if it can carry about 100kg (me and my daughter) without problem.
So I hope that a wider belt will suffice.

I will see...

The other more radical solution is chain but the ratio can no longer be the same.
Initially, I was thinking more of a chain. I choose belt for the noise.

About these 3D printed parts, you're right. With bearing and real shaft, it will do the job (see my first build as a example) but not with the power that I want to dispose!

 

[stan.distortion]

Idk, that's a fairly hefty assembly and it's really well braced with the plates going to the front engine mounts, I doubt that would have any problem standing up to the job with a heavy wall print and maybe a sleeve around the swinging arm bolt. I'd done some parametric designs recently using just the stress values for ABS prints on the flat plane, the design adjusts its self according to the load values and the results where shocking. I'd been using a 10mm X 3mm strip for pulling 2 parts together with around 5kg load and it worked out something like a 3mm x1mm strip would have done the job just fine with more than 100% safety margin. The figures for that came from here:
https://reprap.org/wiki/Mechanical_Properties_of_Components_Fabricated_with_Open-Source_3-D_Printers_Under_Realistic_Environmental_Conditions

I'd guess there are similar studies for strengths along the height but that would need a much bigger safety margin, compression should be good though and with the area you have there I'm betting it would be many times more than the chain load.

Just had a quick look at chain tensile strengths, the belt strain on the standard 8M x 22 pulley at 50nm torque works out at around 180kg while the tensile rating for 1/4" pitch single row chain is around 420kg. Duplex should give loads of safety margin even with a 10T driving sprocket (about 20mm diameter at the roller centres) and triplex or more isn't too hard to find. A messy and noisy pain in the neck using chains rather than belts but certainly doable if the 5M belt doesn't prove up to the job, fingers crossed it will hold up and it certainly opens up the options for other builders if it works out ok.

 

[rsnoiser]

The inside of the first part is made!


When I see all these chips and it wasn't the biggest job, I think one day I'll try molding the metal to reuse all that lost material.


Now I have to do the fixture plate to mill the outside and put the other side of the piece.



After serious work, some mistakes, and a lot of G code for my little cnc,
The first half of the output shaft housing is done!


Now it's time to do the other part...

 

[rsnoiser]

The housing is made! More than 11 pounds of chips lost to do these big hollow parts.



Now I think about the two supporting plates, the output shaft and the battery shape.

I regret to not having a perimetric frame, with this old one, it is more difficult to integrate batteries (260 21700 cells) and all electric stuff.

My idea is to do a three parts battery:
-a small one inside the void between motor and frame (in the middle).
-two parts on the left and right of the frame up to the maximum height allowed by fairing.


On the left, I will place the controller and on the right the BMS.
The ingeneering of this is not easy but it's fun to design.

 

[SlowCo]

Very nice project :thumb:
I'll follow this.

 

[rivvs]

50 kph is very slow for a full size dirtbike. On a motocross track I often get up to 90 kph.

 

[rsnoiser]

Around me, all MX tracks are small with not a lot of long straight lines. It’s not like US where a lot are very large with a lot of straight line.
I can’t imagine do a 90 kph on my little nextdoor track!
With my offroad skills, I will finish twisted around a tree!

 

[Elektrosherpa]

It's a matter of acceleration.
If your bike accelerates fast enough, you can reach 90 km/h also on a windy track-
just only for a second... :lol:
Of course makes sense only if the brakes match

 

[rsnoiser]

Here is the image it inspires me:

Sorry for the non-French speakers, this is a very good French comic.