ES DIY Motor Challenge

[modern_messiah]

Post removed

[ron van sommeren]

Veryvery improved winding diagram 'calculator', it will give the winding diagram in character code and in a picture, English and German:
http://www.rcgroups.com/forums/showthread.php?t=1434363

http://www.powercroco.de -> English

[Miles]

Veryvery improved winding diagram 'calculator', it will give the winding diagram in character code and in a picture, English and German:

Wow! That's great! Thanks Ron.

[Dee Jay]

Oohh... I've got some unique electric motor ideas that could possibly be very powerful. but i gotta do some research to confirm it hasn't already been done...

[Lebowski]

I just found this forum

As a hobby project I've been building my own bicycle engine and engine controller (both from scratch
using of the shelf stuff like magnets, an iron bar and some sheets of aluminium and using only
tools like a simple drill and a jigsaw).

http://www.youtube.com/watch?v=dGfrcnu5J-I

(There's a second video where it spins up to around 900 rpm @ 67 V)

The engine is supposed to be build onto the luggage carrier of my bike with a chain going
down to a sprocket (attached to spokes) on the left hand side of the wheel. The axle of
the engine is actually a shimano deore rear-axle, this gives me bearings, freewheel, easy attachment
of a sprocket to the engine and I have a wide range of choises for the transmission ratio. The
rotor of the engine is attached to the axle by 32 screws going through the (drilled out) spoke holes.

The controller is build around a general purpose PIC 16F88 processor (a component which does not
have a motor controller in it) running around 300 lines of code (assembler). There are no
hall or optical sensors, its a sensorless design.

For smooth and constant torgue I am planning to drive the engine not from a voltage source but
from a current source using a DCDC converter and a (different from standard) control loop. Battery
will probably be relatively low voltage, 18V. I'm not planning to use PWM or sinusoidal PWM as found
on commercial bikes

As a small introduction, I am 39 yrs old, Dutch (so excuse my english) and live in Switzerland. I am an
analog / mixed signal design engineer by trade (basically an electronics engineer who designs IC's,
I mostly do ADC's, DAC, amplifiers and stuff like that).

[dynamo dave]

Hello great idea! I have been thinking about building a motor from scratch that would move me around at 15 mph for a long time.I find that with axle flux designit is a real possibility.I like thud's idea cheap. i started studying S.P.THOMPSON DYNAMO ELECTRIC MACHINERY two year ago . A lot of the new stuff is really old stuff. THUD if you you can get some time go to archives.com and check out the oblique winding in the book i just mentioned.I will start making a elctro-magnetic machine based on this design in 9 months after I retire from 30years as a railroad electrician. I have seen a lot of changes from carbon pile regulators beautiful clockwork devices to IGBT. Anyway great idea. People working together for a common goal is good Challenge is just a word.

[Miles]

Welcome Dave!

At the moment, the challenge seems to be to make anything at all...

[Thud]

Hello,
Just Chiming in to welcome the new blood & let you know I have not abandoned this trhead or my motor building ambitions.
being a consumate home experimenter, Lately i have been researching stator fabrication & may attempt some chemical machining of laminations with a silk screening process.

My little axial motor is patently waiting on the shelf for me to return to it.

DynamoDave, thanks for the referance material, I will definatly be looking for it.
T

[Dee Jay]

DAMMIT! this is EXACTLY what I had in mind! EXACTLY! Coils high on the frame and magnets on the wheel. But he didn't think of a variable torque system that I'm seeing...

I didn't even know it was called axial flux till now.

Is this guy an ES member?

Anyone see this bike in action?

[qwerty0]

DAMMIT! this is EXACTLY what I had in mind! EXACTLY! Coils high on the frame and magnets on the wheel. But he didn't think of a variable torque system that I'm seeing...

I didn't even know it was called axial flux till now.

Is this guy an ES member?

Anyone see this bike in action?

You might enjoy this former members efforts.
http://www.ebikehub.com/forum/viewtopic ... &start=135

[kevo]

Two years ago Miles wrote:

Maybe we could have another category for motors of, say, 3kg to 6 kg?

Given the years this thread has been open, is the higher weight option open?
Still very much in progress, but how about adding 50% plus regen.
muahahahahaha

[Miles]

What are you thinking of?

[kevo]

High density copper, coreless.

[Miles]

If it's a hub motor, that doesn't make sense to me (in the context of the challenge criteria). Which makes it all the more interesting.....

[Honk]

Have any of you guys ever considered a large & high density slotless powder core motor aiming ultra high efficiency > 97%?
http://powerditto.de/slotless_outrunner.html (If you can't read German "as me" then use google translate)
http://www.rcgroups.com/forums/showthread.php?t=972859

I believe such a motor could become highly efficiency when correctly designed with powerful neomags and compressed heavy duty litz.
Powder cores as MMP or Sendust has 5 times lower losses than regular Somalloy powder. The iron losses is almost non existent in MPP or Sendust.
That allows for extremely low no-load motor current as the hysteresis and eddy currents is not a problem as in regular electric steel laminates.
Using thick enough magnets would allow for a pretty large airgap that would be filled with epoxy soaked & compressed litz for lowest possible resistance.
The litz wiring is a must in a slotless motor to eliminate any eddy currents in the windings. With some aid from low friction bearings the end result would be an extremely efficient motor.

The size of the motor must be big enough for a low and usuable KV range, or else the extreme efficiency would be lost by any multistep gearbox.

[Farfle]

Aww bugger, I just looked up the challenge rules... my motor is a no go . If all my math is right, which I am NOT anything resembling an expert , heres what I got:

"Goal:

- Over 4Nm continuous torque per kg of motor weight. check!

Quoted from the build thread, Kt (Nm / A) is equal to 9.549/kv
The kv is around 26. so about 0.36 Nm/A .
To get the 4nm/kg I would need 4x13.6 or 54.6 nm continuous.
54.6Nm/.36Nm per A gives 151 phase amps.
The winding resistance is .014 ohms lead to lead, P=I^2R P= 151^2 x .014 that's only 319.2W in heat, easy.
Now heres where the WAG comes in, that's not Including hysteresis/eddy losses and those add up, but 4nm/kg should be more than possible.

Rules:

- Less than 3kg in weight. definitely not check, 13.6kg

- No energy input other than that to the motor itself.- check!

- Capable of practical use on an electric bike. - maybe check, would be a little rough on bike parts.
Miles"

[Miles]

Never mind... I was going to set another category for motors over 3kg.

The thing is that it's much easier to achieve higher torque densities with larger motors, so we have to raise the bar a bit

How about a category for 3kg to 10kg and then another one for 10kg to 20kg?

[Miles]

How about:

3kg to 10 kg specific torque needed 5Nm/kg

10kg to 20 kg specific torque needed 6Nm/kg

[Miles]

When one of the challenges is met, we could "raise the bar" for that category by 10%?

[Farfle]

It would be hard building a rig to test cont. torque in the 70+ Nm range.

[Miles]

Do you have access to a motorcycle dyno?

[Farfle]

Sadly no, I can check around and see if anybody has anything around town that would be interested enough to do a discount.

[Miles]

We need an impartial referee..... I'm hoping to participate, so.....

[Arlo1]

Do you have access to a motorcycle dyno?

I have one. You guys want to travel?

[Farfle]

There's a shop here in town that has one, I'll head down there on Tuesday and see if I can get a continuous torque out of it. There is a possibility that I can't push it that far with only 250 phase amps out of the controller, plus that's it's peak rating IIRC it's only good for 180 phase cont.