axial flux motor question !!

hmm...

Thin copper should be use for minimize skin effect. For high frequencies...
The Eddy current is for the stator and back iron not for windings (or not significant for windings) :roll:
 
dotrick said:
hmm...

Thin copper should be use for minimize skin effect. For high frequencies...
The Eddy current is for the stator and back iron not for windings (or not significant for windings) :roll:
Click to expand...
That depends on whether there's an iron core, or not. :roll: :)
 
Tanks to the very interesting thread : " doing the math" from Kingfish , I did some math today , changing some of my numbers , here is what it came out !!


single aluminium stator : 21 coils , drop shaped

double iron made rotor : 2 x 48 magnets ( disk shape 30 mm x 7 mm and 25 x 7 mm ) covering the coil shape

Air gap : 0.8 mm / magnet to magnet : 11 mm

Magnets at 163 mm from rotor center

motor rotation : 720 rpm

Torque : 200 Nm

Power : 15 KW

Angular velocity : 75.36 r/s

Motor Constant : 2.66

Kt: 0.88

Motor Frequency : 168 Hz

Volts : 96 v

I : 157 A

Conductor Length : 5.8 m with 0.65 T ( witch appear to be realistic !!)


http://www.casimages.com/f.php?f=14041912191853124.pdf

green is the stator and red the rotors


Few question :

1 / are my numbers correct ??

2 / is that lenght , the overal lenght or single coil lenght ??

3/ I choose litz wire 40 stands of 0.1 mm , is this size enougth for the amps

Tanks for your usefull comments :wink:
 
I took the formula from Kingfish where Kt : torque/angular velocity and finaly divide by 3 phases
 
it seems that the wire section is too small for 15 kw :? , I just got the book " axial flux permanent magnet brushlees machine" 2nd edition , for my understanding ,

and I start to understand a bit better the meaning of the all thing :wink:

there's an example in that book ( the case study 6.6 ) , which is very close to my design , and the wire used is another section :shock: , 6 x 1.2 mm :mrgreen:

and this motor already exist and run , so I may be carefull of the infos i can read somewhere :?
 
fdracing said:
3/ I choose litz wire 40 stands of 0.1 mm , is this size enougth for the amps

Tanks for your usefull comments :wink:
Click to expand...

I think, it's better to use square or rectangular wire than litz wire. Litz wire is for high frequency motor.

For your application with Litz wire 40 x 0.1mm --> Area 0.4mm2 --> for 5.8m/phase --> 0,31 ohm --> Copper losses @157A is about 7.7Kw.
For your application with square wire 0.63 x 0.63 mm --> Area 0.4mm2 --> for 5.8m/phase --> 0.25 ohm --> Copper losses @157A is about 6.1Kw.(-26%)

What is your target efficiency? :D
 
you may laugh , but my target efficiency is the more as possible , let say 92 %

I understand that the wire size i choose is wrong :? , I 'll be more around 4 to 6 mm2 , flat , square or litz , with low resistance winding :!: , any suggestion :?:

I also understand that raising the volt will decrease the cooper loss :wink:

I already decide to make the stators in Polyamide PA 6 GF ,expensive stuff but it will solve some eddy current there :wink:
 
fdracing said:
you may laugh , but my target efficiency is the more as possible , let say 92 %
...
I already decide to make the stators in Polyamide PA 6 GF ,expensive stuff but it will solve some eddy current there :wink:
Click to expand...

92 it's not a number... 95 it's a target! :D

If you want to improve the motor efficiency, you have to use laminations (M19 steel for example). For same stator size, the torque with stator lamination will be 2x than PA66. If you want the same torque, you have to double the turns for PA66, it means double the copper losses or you have to double the current, it means 4x (I^2) more copper losses.

Do you know FEMM software? It's a very good tool to design an electric motor... And Miles is a FEMM expert! :wink:
 
95 for the first motor !!! kind a dream :wink:

I have this soft ( FEMM ) on my computer , but being able to use it , it's another story :roll:

on axial flux motor it's difficult to use lamination , because of the design itself , but I may be able to fit some laminated iron core at the coils , it will help for sure :wink:

step by step my design is growing , i will have something nice to show you , soon :wink:

I thougth that raising the volt will lower the amps and so the copper loss :roll:
 
fdracing said:
I thougth that raising the volt will lower the amps and so the copper loss :roll:
Click to expand...
me too... :?
If you increase the turns by 2x, you increase the FEM and the torque x2. You can decrease the current by 2x. The efficiency would be better. You have to check if you have enough space to do this... :?
 
dotrick said:
You have to check if you have enough space to do this... :?
Click to expand...
Quite. If there is, you could have utilised it by using a larger gauge previously. :) There's always a presumption of equal copper fill. With that assumption, generating a given torque will create the same losses no matter what combination of Volts and Amps is used.
 
@miles : I'm not absolutely sure but I think generating a given torque will create less copper losses with more turns. If you make motor with more turns you have more FEM... But you have to choose a Controller with high voltage output (more than FEM) to manage you motor. :?

@fdracing : What is your max output voltage controller?
 
dotrick said:
@miles : I'm not absolutely sure but I think generating a given torque will create less copper losses with more turns. If you make motor with more turns you have more FEM... But you have to choose a Controller with high voltage output (more than FEM) to manage you motor. :?
Click to expand...
More turns means longer and thinner wire, in same cooper fill condition, so resistance/cooper loss goes up per given current. Right? :)
 
parabellum said:
More turns means longer and thinner wire, in same cooper fill condition, so resistance/cooper loss goes up per given current. Right? :)
Click to expand...

It's more difficult than I thought! :?

Example for a Slot area of 10mm x 10mm :
Wire diameter : 1.414mm / 2mm
Nb wires / slot : (7x7=49) / (5x5 = 25)
Lenght : 20 m / 10 m
R windings : 0.22ohm / 0.054ohm
Current : 50 A / 100A
Copper losses : 541W / 541W

same thing for copper losses... :? Miles was right... :oops:
 
Well if you spin your motor faster, you won't have to make as much torque for the same amount of power. Using thin wire like litz wire is necessary if you want your motor to spin fast. Fast here is relative and I don't know any hard numbers about how thin you have to go. Most likely you will just have to try different thicknesses of wire and adjust accordingly.
 
thepronghorn said:
Well if you spin your motor faster, you won't have to make as much torque for the same amount of power.
Click to expand...
Sure, but we're working to a specification of 720rpm and 200Nm....
thepronghorn said:
Using thin wire like litz wire is necessary if you want your motor to spin fast.
Click to expand...
For a motor with an ironless core, you also have to take into account the flux density that the wires are exposed to.
 
the voltage and so the controller will be choose going on the way , all the examples I have for this kind a motors is high voltage from 120 v to 400 v , all these motors are ironless coils , non conductive thin stators , and they all work at 92 + efficiency , so it seems to be possible that way :wink:

my problem remains the same ( as the song :mrgreen: ) ..............the coils :!:

I will start in a way and see what happens , make test and mesurement and so, make the corrections , it's quite empirical :shock: , but no way out , there's a time for speach , and now it's time for facts ......
 
fdracing
I believe you start from the wrong end if not that just want to get rid of a big litz wire spool being around for years.
You should pick controller first and know its switching frequency, volt and current you want to run it. Then make motor design target finally coming to conductor cross section area and only then you will be able to decide how reasonable it is to implement litz wire, again, if it is not just litz wire obsession. :D
 
it can be litz or what so ever , it does'nt realy matter :wink: , what matter is 200 Nm and 720 rpm , all the rest will fit with these numbers :wink:

today , after some hours of reading some good books , i understood that i will took me a bit more than expected :shock:

i love that job :mrgreen:
 
You must log in or register to reply here.

Similar threads

V
Axial flux motors
Replies
14
Views
1,369
Roookee
R
E
    • Like
Eldridge Axial Flux Motor/Generator
Replies
19
Views
4,160
Logic11
L
I
Axial flux motor not rotating but vibrating.
Replies
4
Views
1,675
imraj
I
E
    • Like
Homemade Axial flux PM motor
2
Replies
27
Views
11,995
Erik952
E
T
Multistage axial flux motor design and prototyping
2
Replies
25
Views
6,538
larsb
L
Back
Top