YZF 2007 Electric conversion

[larsb]

+1 on that!

 

[BenBenBen]

Did you have bad feedback concerning that kind of Belt drive first step?

Usually for this application HTD 8M belts work just fine and are easy to find. Take a look to CONTITECH catalogue.
They have several extremely strong heavy duty type.

I don't have any personnal experience with it, but that's what I heard too. As long as they are used in correct conditions (Tension, alignement, clean atosphere without gravels ) they have the reputation to be better than chains in terms of vibrations, noise, wearness and torque management.

But I know they can be a disaster in situations like:

- Vibrations when too long segments
- Breaks when used on the rear wheel in Off Road (gravels going into)
- Impossible to use with variable geometry (when the sprocket axle is not aligned with the swing arm rotation axle)

But In this case, it should be a stable situation and in a closed enclosure, so shouldn't make any problem.

I'll strat with a "cheap" reference from Gates https://www.123courroies.com/courro...roie-dentee-560-8m20-gates-5412571323696.html and probably switch to a stronger one when I'll be 100% sure of the correct lenght.

I also found my pulley. It's a standard 40 teeth. Only requires a bit of machining to make it lighter and to fit the 25mm center diameter with keyway. https://www.123courroies.com/2242-a-usiner/nombre_de_dents-40/largeur_de_courroie-20

For the chain sprocket, I spotted that reference that is supposed to be used on some kawasaki ATV called ATC300.

They are 520 standard, available from 12 to 15 teeths and fit on a large shaft profile (diam 25/21mm), that should be relatively easy to machine on the custom middle shaft.

https://www.franceequipement.com/fiche/produit/64974///64975

Should do the job with a 520 chain and 52 teeth original YZF250 rear sprocket

 

[BigBlock]

I would try to go with a slotted design to tension the belt. I'm not a fan of tensioner pulleys. They tend to loose over the time.

Without the pulley you can also end up with a more compact design.

 

[BenBenBen]

I would try to go with a slotted design to tension the belt. I'm not a fan of tensioner pulleys. They tend to loose over the time.

Without the pulley you can also end up with a more compact design.

Yeah your right, that's a potential additional failure cause + maintenance part.

On the previous similar design I made for a Golden Motor 10kW (I posted pictures on the begining of this thread) that's what I did and it worked well. The Golden was pretty adapted to do it.

But the QS 138-90H motor is designed in a way that the mounting holes are extrelemy close to the centering diameter. No way to add long enough slots here. An other way to do it could be to split the support plate in two and rotate one VS the other. But it will make the global motor/reduction assembly with a variable geometry and over complicate the mounting + tension tunning. Also, I'd like to have it fixed to use it as lower support for the pack.

So I think I'll stay on this design for the moment

 

[BigBlock]

Are you sure to have got an updated 3D model of the QS 138 ? As far as I know it come with a dedicated bracket on top.

You really don't need 6 bolts for side to have it bolted to the frame. I have just two bolts: one fixed and one on slotted to manage the belt tension. Quite easy and straightforward to design.

 

[rivvs]

on my bike I used the bac 8000 at 600 phase amp with the 138 90h and a gear ratio of 6.21 :1 And it did not lack low end torque. take a look at my build trhead there is a video.

 

[BenBenBen]

on my bike I used the bac 8000 at 600 phase amp with the 138 90h and a gear ratio of 6.21 :1 And it did not lack low end torque. take a look at my build trhead there is a video.

Thanks for sharing experience. Seen your Instagram video link on the RMZ thread, you from France right?

In my case, I don't think I can go so high in phase amp with the Silixcon. 500A will probably be a big max and with good cooling.

What decided me to go with 2 stages is that I really don't want a giant rear sprocket. I'll be using this bike mostly for slow Enduro in places with many big rocks, trunks and other stuff that I love to jump on. I'm afraid I would bend it pretty quick... I really don't want to push the bike 2km and make a new custom sprocket every weekend

 

[TysonScott]

on my bike I used the bac 8000 at 600 phase amp with the 138 90h and a gear ratio of 6.21 :1 And it did not lack low end torque. take a look at my build trhead there is a video.

I'll be using this bike mostly for slow Enduro in places with many big rocks, trunks and other stuff that I love to jump on. I'm afraid I would bend it pretty quick... I really don't want to push the bike 2km and make a new custom sprocket every weekend

are there any circumstances where you would not do a 2 stage reduction?

 

[BenBenBen]

Hi,

I felt on some tech info about the new Storm Bee from Sur Ron. I think it was interesting do retro calculate what they did on a serial production bike. So I did my maths and here what I found:

I just assumed that this bike got the same rear wheel than my YZF 250.

It seems to have a 96V motor turning faster than a QS, about 8k RPM but with lower torque arround 56Nm.
Final gearing ratio should be arround a bit more than 1:9

Just wanted to share it.

The bike is 128kg heavy, so it seems in the average of the converted Mx bikes I've seen.

With a QS 138 pushed to 22kW and the gear ratio I plan with the 2 steps transmission, I should arrive to more or less the same level of performance.

I didn't tried this Sur Ron, but I imagine the frame, suspensions and brakes are much better on a Yamaha YZF. Or, at list, high qualitity after market parts should be easier to find...

I hope I'll be able to fit 5kWh, so the battery capacity should be comparable too.

But a huge advantage for this 9k€ brand new bike is that it is fully raod legal... Which won't be possible on mine and it's verry sad!

 

[j bjork]

I wonder if there are any benefit to go with the 90h if you are limiting it with 500pA?
That is if we are talking peak, rms is a different story.

I mean, with just 500pA you might as well go 138 70h v3 and save a lot of work.
But it will be a less interesting build thread :wink:

 

[BenBenBen]

I wonder if there are any benefit to go with the 90h if you are limiting it with 500pA?
That is if we are talking peak, rms is a different story.

I mean, with just 500pA you might as well go 138 70h v3 and save a lot of work.
But it will be a less interesting build thread :wink:

Héhé And I love to make over complicated stuff that I finnaly use a lot less than expected cause I'm building something new... (Just like most of the people here I guess)

I'm still not 100% sure that I'll keep the Silixcon long term. I got to test it, but I aslo got the BAC8000 in mind and maybe, someday, I'll make a definive (or not) swap

Attached a datasheet extract. From friends that used it in industrial application, they told me that the figure are pretty realistic.

 

[rivvs]

on my bike I used the bac 8000 at 600 phase amp with the 138 90h and a gear ratio of 6.21 :1 And it did not lack low end torque. take a look at my build trhead there is a video.

Thanks for sharing experience. Seen your Instagram video link on the RMZ thread, you from France right?

In my case, I don't think I can go so high in phase amp with the Silixcon. 500A will probably be a big max and with good cooling.

What decided me to go with 2 stages is that I really don't want a giant rear sprocket. I'll be using this bike mostly for slow Enduro in places with many big rocks, trunks and other stuff that I love to jump on. I'm afraid I would bend it pretty quick... I really don't want to push the bike 2km and make a new custom sprocket every weekend

Im a french canadian actually! I started with 450 amp and it felt slow so Upped the amp until It felt at least as fast as a 250 motocross bike. I was planning to slowly ramp up the amp to see what the motor was capable of but then the battery fire happened.

 

[BenBenBen]

Im a french canadian

OK got it now! Was like "he seems to be French, but nobody afround here waits for snow to melt before going to ride, cause there's just no snow that stays more than 24h..."

 

[BenBenBen]

Hi everyone!

I'm about to start the firts parts of the reduction assembly in fabrication.

Laser cud for the plates, 3D print for the rest. And then Machining after validation with 3D printed parts

Any opinion or remarks about the jack shaft assembly concept?

 

[larsb]

the grease port isn't needed, filling with grease will add drag without benefit. open bearings, oil and seals would be better if you want lubricated bearings and possibility to service it.

How do you fixate it at the front?

Sideways stiffness could be an issue, i cannot really se how the "halfmoon shield" around the sprocket is intended to be. This will be critical for sideways stiffness. Might be good to add a similar one on the pulley side.

How are the hub sideplates in the middle fixated?

Where do you intend to take up tolerances and any adjustments for the chainline? I guess sprocket side to be fixed and the rest floating to take up tolerances and some spacers somewhere for adjustments.

 

[BenBenBen]

Yeah oil would be cool, but for it I need to add a oring between the jack chaft and the spacer on the pulley side to have the cavity fully sealed. Also need to have a breather on the top. And ideally a way to check that there is still oil inside! A lot more complicated and risky in case of leaks.

The "moon" part it just solid plain 6061 aluminium with 5 M6 threads on each side.

On the pulley side, the plate will be welded on the jack shaft housing.

For the front, I don't really k ow yet. I only got approximative measurements, so I'll wait to see how the mockup fit see where I can hook it. Plan is 2 M8 screws on each side.

 

[larsb]

that's a big block of alu with mostly waste to do the moon. I'd look at either designing a square piece made from alu bar or even 3D print it instead.

I ordered a 3d print part in alu from craftcloud a while ago, it was only 64€. That's not bad, of course printed in China but anyway..
I'll get it in a week or two, can let you know how it turns out.

This part in 100mm dia would be 72€+shipping:

 

[BenBenBen]

Maybe even just tubular type spacers with long bolts could do the job. I don't know yet how this part will look exactly.

I got this part machined in 6061 for cheaper than this. China too. MP for suppliers name sharing if interested.

3d printed alu could be great for my belt enclosure. I''l design something more se, y than my first metal sheet version and ask for a quote

 

[larsb]

I think that'll be too flexible so if you can get it for such low price (name of manufacturer?) i'd just order it. Anyway, i you can fit a rectangular bar instead of the moon it would save material, mostly not thinking of costs but on unnecessary waste

 

[BenBenBen]

Won't be so much torsion force on the part at the end.

The radial charge on the jack shaft bearings should be about 4kN @max torque (concidering a 3cm lever of sprocket and 130Nm of max torque on the jack shaft), but most of this effort should be supported by the link going into the swing arm axle.

The role of the "moon part" there is more about avoid a twist of the plate on the pulley side, caused by the belt tension which I calculated to be arround 0.64 kN (28mm lever @72Nm of motor torque). Nothing to crazy to handle. Espacially in this case where the the motor will also be supported on the front.

I like it to be "circular" so it's also acting like a "housing" for the sprocket. What's your exoerience about mechanical quality of Alu 3D printed parts?

 

[BenBenBen]

Concidering lub by oil, il could look like this with the additional Oring to avoid leaks between the shaft and the spacer.

I also added a M10 thread on the top, with a gasket, for oil filling. Do you think a breather is required?

 

[BenBenBen]

For the front attachent, idea is to drill holes and weld "tubes" for screws to pass through in the front area. And add some spacers between the plates and the frame tube to compensate the gaps. Will see how close to this I'll find in the real worls.

 

[BenBenBen]

Hello All,

I started to receive parts for my reduction assembly, and I'm about to launch the machining of the jack Shaft.
Big pulley is a heavy steel version for the moment, but I'll make a custom 6061 Alloy one when I validate the size.

I decided to go with a 327-13 sprocket from JT sprocket. I can't find any "standard" for the end shaft tolerances. The area where the sprocket fits.So I did it by feeling, inspired buy some standards that look similar like the NF E 22-131. Do you have any experience to share about custom end shaft tolerancing before I launch?

Cheers,

Ben

 

[El Gaucho]

I went by the guess and check method when I did mine. Ended about .002" to .003" under actual sprocket dimensions. A snug fit, but still will come off without a press. Very nice CAD work!

 

[BenBenBen]

Thanks! Yes I guess that's what most people do.

But I'll have to send the drawings and wait for the parts this time, so I'd better send the good tolerances!

=> 5h10 for the spline is 5 +0 -0.048 for 5.00 measured on the procket

=> 20h6 for the diameter is 20+0 -0.013 for 20.04 measured on the sprocket

Should be fine I guess, a little looser than yours but still snug.