#crossbreaks' midmotor CB120

awesome project!

cb, do you already have numbers for the torque (kT), maximum RPM and peak ETA?
 
hey mate, whats the width? (stator and shell)? is it a static axle like a hub, or with a 'boltable' section like the revolt?
looking promising mate, depending on outcome id be keen on at least 1...
 
the width of the motor is at least 68mm plus sprocket. it will be about ~80mm in the end. the sprocket must be further out since it must pass the bottom bracket, which is 68mm wide (or more). It will look similar to the RV120-Custom I tried to make (which did not work so well - i didnt even try it out in the end, plz read the first 6 pages of this thread for more info).

It will look very similar to the concept I showed on pages three and one:
https://endless-sphere.com/forums/viewtopic.php?f=28&t=80516&start=50#p1191267

The sprocket and chain/belt adds another 20mm to the 80mm as it stands out of the BB-distance (bottom bracket distance). So total width will be ~100mm plus bolt heads and mount sheet. At least 100mm total. Plz give me some time to make drawing. I try to keep it as narrow as possible for sure. Just like this:
file.php


I think think the HDT-M5x25mm is not suitable for this motor. I just offered this as an option to say that belt drive is an option. HDT-M8x15mm is more suitable.
 
you're welcome :)

sn0wchyld said:
is it a static axle like a hub, or with a 'boltable' section like the revolt?
Click to expand...
static, like a hub. The revolt needs large bearings that have a lot of drag...and loss. They are also not durable enough for a chain/belt drive...the hub design fits our needs much better. I used 6302 bearings that have much higher Cn than the 169xx series bearings that are used in outrunner motors that where disgned for planes/ propeller drive in the first place. We discussed that on page 3 https://endless-sphere.com/forums/viewtopic.php?f=28&t=80516&p=1191703&hilit=bearing#p1191703
 
exactly. 2kW/4kg motor goes out, 8kW/4kg motor goes in 8) but you may have problems with the width of almost 100mm. I try to make it as narrow as possible. I think my design skills are okay, but i have no idea how to make it even more narrow :/ ideas are welcome!!
 
crossbreak said:
I think my design skills are okay, but i have no idea how to make it even more narrow :/ ideas are welcome!!
Click to expand...

on the CAD it looks like there would be some space on the left side (between bearing and stator).
or is it necessary for wiring and sensors?
 
Great job, congrats. I am also searching for a "usable" mid drive. I started to rework the bht. See attachment. I implemented a servo shifted 2 speed gearbox and ended up with getting it between 150mm swingarms. 1st gear reduces 1:2.6, second is direct 1:1. The issue with that is, that I need to play around a lot with tolerances such that it shifts smoothly... which most of the time means to CNC a new part and try a 10th of a mm here or there. Pretty frustrating. So I started to get some info about how ALTA or KTM is doing the thing. KTM uses a PERM PanCake (axial flux) construction. They rev the thing up to 6000rpm and give it a non shiftable gearbox reduction. The ALTA uses a 15.000 rpm radial design (as crossbreak does, but as the put it in a motocross frame, they have plenty of width). So I started diving into the axial flux solution (search for Scolton LEAF for a very detailed essay about axials). As Scoltons solution does not use iron in the core, it is pretty lightweight AND narrow. The torque could be increased by using iron cores. The good story about the ironless core is, that there is NO cogging at all, if there is no voltage connected to...it spins absolutely free. Anyway an iron core would increase torque. In the LEAF one the iron core (coil form) uses pure ABS. Anyway I found a patent for a axial motor which uses a abs/ferromagnetic powder mix (melt the plastic, put the powder it, let it cure, done).
For me this means to skip my geared solution and focus on the pancake now :)
 
Nice work again Felix! I have a version of Dog clutch 2 Speed in CAD too but with chain drive.

The Alta uses an in-runner with internal watercooling to the frame, and the frame acts like a heat sink
 
those are built into the ktm http://www.heinzmann.com/en/electric-motors/pancake/synchronous-motor
 
hey, good to hear from you. If my CAD helps you with the LMX let me know by PM. No problem to share it with you incl the CAMs.Cheers ! :)
 
@maddin, i think i need the space for the phase wire connection. Time will tell if so much space is needed

sad thing about the BHT is the Km² copper loss figure. It is at around 0.5 (Nm/√W)²
The RV120Pro is at 1.6, the CB120 is at 3.0

for a given torque, you need 6 bht motors to get to the same copper loss as two RV120pro or one CB120. This is why i went this way to make a gearbox or high reduction ratio obsolete. A 3:1 reduction to the wheel does just fine.
This motor is basically small version of the JohnInCR Midmonster with optimized packaging
 
madin88 said:
on the CAD it looks like there would be some space on the left side (between bearing and stator).
or is it necessary for wiring and sensors?
Click to expand...
Actually, the space is only needed near the axle. the side covers could be cone shaped to save some space in the outer region and to make it more rigid. Even a steep cone is more rigid than a plate.
 
will this motor include hall sensors ? and hence will it run with a normal kelly controller even though its 15 pole ?
 
Some measurements done with the prototype:

No load consumption:
6.70 A @ 48.04V @ 2270 rpm
4.11 A @ 24.73V @ 1171 rpm

Terminal resistance: 9.93 mOhm

So if i calc right: kV = 47.3 rpm/V
Km² = 4.076
Peak eff: 92% @ 1800 rpm @ 28.3 Nm 5367 W PowerOut and 5792W PowerIn

Km² is 2.5 times higher compared to the revolt :) good that it does no cost 2.5 times as much :mrgreen: It turn better then i thought at 2270rpm. The prototype rotor is not balanced yet...anyway this may be a lucky shot
 
You must log in or register to reply here.
Back
Top