Giant Trance Belt drive with MESC @ few kW

[mxlemming]

Cool update. How does the sliding part for the extra needle bearing work? Do the slotted holes work well enough for belt tension? Are you running an idler for extra tooth engagement on the motor pulley? What's the thickness of the bracket?

Not all is well in the world of ebike.

Didn't put all 12 screws in, boosted the power... Dead pulley. This many screws was fine with the little red motor.

Unfortunately I don't have the big 3d printer to use any more so I'm back to the yellow htd8m pulley which is far noisier.



Hopefully this explains a bit better how the tensioner and idler works.

Bracket is 8mm 6082-t6 water jet cut. The slotted holes seen to be fine for belt tension.

 

[SlowCo]

Can't you put the brake rotor back on the pulley slightly rotated (few degrees) and drill new holes? And before mounting use glue between the surfaces.

 

[mxlemming]

Can't you put the brake rotor back on the pulley slightly rotated (few degrees) and drill new holes? And before mounting use glue between the surfaces.

I could repair it, putting the other 6 screws that didn't rip out in and then doing what you said ina few other places but actually i just opted last night to buy a new 3d printer with a 300x300 bed.

So I'll print a new pulley, 5m, sightly bigger and with more screws and reinforcement around the screw holes. I'm going to go for slightly less garish colors. Bit more stealth.

 

[liveforphysics]

Great project!

Fun progress, and having done a 2 x 80-100 motor ebike, I'm glad to see you jump to the 120-70. In converting electrical power into satisfying experiences, theres no replacement for more copper and iron.

Belt drives on car blowers can get a more quiet by drilling a few mm diameter hole in the corner of each tooth trough area. It seems to positively effect noise from the displaced air pocket from each tooth as the belt tooth climbs. Some acoustic damping material like dynamat or similar product applied to exposed sides of the sprocket might help too. Increase infill to 100% for at least 2.5 x bore diameter radius around your fastener mounting points and they might stay attached a little better.

 

[mxlemming]

Great project!

Fun progress, and having done a 2 x 80-100 motor ebike, I'm glad to see you jump to the 120-70. In converting electrical power into satisfying experiences, theres no replacement for more copper and iron.

Belt drives on car blowers can get a more quiet by drilling a few mm diameter hole in the corner of each tooth trough area. It seems to positively effect noise from the displaced air pocket from each tooth as the belt tooth climbs. Some acoustic damping material like dynamat or similar product applied to exposed sides of the sprocket might help too. Increase infill to 100% for at least 2.5 x bore diameter radius around your fastener mounting points and they might stay attached a little better.

Thanks. It's been a lot of hassle and work to get this far. Should have just bought the controller and been done with it :lol: would have been a 3 weekend project not 1.5 year that way.

I've found there are substantial differences in belts noise as well.

The 8m is definitely the loudest and whirriest, but it simply does not skip. Might try the drilling idea with the 8m pulley.

The htd5m has a lot less noise, but skips. Continental standard profile.
The 5mhp from optibelt is much quieter but skips more under braking. It has a special parabolic tooth profile and little grooves to let the air out like your drilling idea.

The elephant in the room is the 12070 motor itself which rings like a bell. Any ideas how to stop that? Same ringing whether bldc, VESC foc, my FOC... I think it's just the shell is... Ringy. I put an elastic band around it but at high speed it just pinged off.

The real clanger to my fun is that it's usually programmed down to 310W (motor takes 60 just to spin) and i only remove the limit for off road trails. Been doing a lot of that this weekend.

 

[mxlemming]

Reprinted the pulley. Added reinforcement holes, double checked the infill on the slicer, modified the tooth profile to better suit the newer parabolic belt teeth and increased the tooth count from 148 to 151. A minor change in ratio but brings the motor a cm lower which is more aesthetically pleasing. Also changed the idler from 35mm diameter to 45 which is much kinder on the belt.

Also fitted all 12 screws this time.

This means I can now run the megadyne isoran belt which is so much quieter than the conti or gates belt.




I set it up with 100A phase and 3kW on the controller and pedaled it unplugged to the farm tracks to play. It absolutely rips now. Sweet spot at about 1/3 to 1/2 of max speed where the front wheel just floats upwards gently.

Got the regen braking working again, and with the new belt and motor with 80A phase regen arrangement I get a quite substantial amount. No skids but reasonable deceleration.

Nothings getting even warm at these settings. The motor and controller I'd describe as pleasant to put your hands on on a cold day. If anything, the 12070 motor actually seems to stay cooler when you start driving it harder, need to look into this, there might be weird electrical effects. Might have to add another 50% to those settings.

 

[Barncat]

Hey mx- novel drivetrain you built yourself there. Seems like a big ask for the dropout and brake bosses, but since it's based off the axle it's workable. I'd keep an eye on the frame attachments.

And congrats on the success with your own controller. An impressive feat... small club of guys capable of that.

 

[mxlemming]

Hey mx- novel drivetrain you built yourself there. Seems like a big ask for the dropout and brake bosses, but since it's based off the axle it's workable. I'd keep an eye on the frame attachments.

And congrats on the success with your own controller. An impressive feat... small club of guys capable of that.

This has occurred to me quite a few times as a potential issue. My logic started at them being good for the discs which generate way higher torque than the motor ever could/ and I planned on using a 63mm motor which weighs like 800g. I also thought "well it takes my 80kgweight dropping off a meter high ledge to flat".

How wrong I was about the true capability of the 63mm motor. You can pretty much divide their listed power capability by 10.

So now it's a 2.4kg motor and I'm looking at it thinking hmmmm, not so sure about this.

I've got another one of those laser cut metal plates with a 10mm through hole and I'm eying up my downhill bikes rear wheel with 12mm through axle and a 12mm reamer bit.

 

[Barncat]

Of course you've considered the possible failure modes.
Aluminum weldments can be susceptible to fine stress fractures that are not obvious, and those stay tubes are also usually pretty thin wall, so I'd be more comfortable bolting the drive to steel. It may just keep working fine indefinitely... Trust but verify.

 

[mxlemming]

Of course you've considered the possible failure modes.
Aluminum weldments can be susceptible to fine stress fractures that are not obvious, and those stay tubes are also usually pretty thin wall, so I'd be more comfortable bolting the drive to steel. It may just keep working fine indefinitely... Trust but verify.

Just an observation that the perspective of the picture i posted before is deceptive, makes the motor look far bigger than it really is.

This is more representative.

 

[Barncat]

It is a clever and compact design, and quite light with that type of motor which we both like. Good that you've got your belt/pulley situation sorted. Perhaps a tad less efficient than a clean chain but way less maintenance, so a good choice for off road. It would be a good drive to bolt up to a custom stay and dropout.

 

[mxlemming]

This new motor is presenting all sorts of problems the previous one didn't, like I can run enough braking force to skid if I lean forwards a bit. And a few goes later it rewards me with a 20km pedal back home through the fields, reminding me that it is in fact very much still a functional pedal able bicycle!

That design idler did probably 1000 miles with the small motor with no issues

 

[Barncat]

Yeah, a 12 mile pedal is better than a 12 mile push.

I'd only trust a full aluminum idler wheel in your application.

 

[mxlemming]

Nice new printed wheel that should bea lot stronger. Better filament, extra wall layer and the cutouts force it to have lots of thick radial lines. Survived the first ride out at least.


Here's the motor and bike 5 mins after the ride. Had a bit ofa fight with the thermal camera. Motor was 32 degrees max. Battery was showing a max of about 25...

The hot strip is where there's a cut in the side of the case so you can see straight into the cells. I actually took the pic with the camera because i was wondering if there was a loose connection, cell p group 6 and 7 are dropping faster than the rest.

 

[Barncat]

Hope the new idler holds up. I have mixed feelings about printed plastic parts... obviously one can produce very complex objects, but the strength and durability relegate them to limited applications.

Could be a couple dodgy welds on groups 6 and 7... you have access to individual cell V right? Nobody wants a