axial flux motor question !!

fdracing

100 mW
Joined
Apr 12, 2014
Messages
45
Hi,

I am making my own axial flux 15 kw motor ( for my co-axial small helicopter ) and I have a question for those who realy knows !!! ( I know there are some here !!!)


these motor are : 400 mm diameter , 24 coils / 2x 32 magnets ( diam 30 mm / 10 mm) ,


they will rev 720 rpm , should I use Litz wire ( 40 x 0.1 mm) as somebody told me or not !!!


Tanks for your answer , I will also take every good advice for axial flux motors serious builders ;)
 
Hi fd,
fdracing said:
they will rev 720 rpm , should I use Litz wire ( 40 x 0.1 mm) as somebody told me or not !!!
If you're not using an iron core then, using Litz wire will reduce the generation of eddy currents. See also, the attached PDF.

At 720rpm, your motor will need to generate 200Nm to achieve 15kW output. Good luck with that..........
 

Attachments

  • Stranded wire.pdf
    147.5 KB · Views: 100
I know about the torque :? it's my first goal :wink:

I am not gonna use iron core for the coils ( maybe some at the back of the magnets ) , because of space , coils are 8 mm thick and air gap will be a bit less than 1 mm for the two rotor plates , as you said torque is the point :mrgreen:

tank's , anyway :wink:
 
At 6000 erpm my axial flux motor takes 20 watt to spin unloaded. I use dual 0.8mm diameter wire, so no litze...
I don't know what your erpm goal is but at 6000 eddy current are not yet a big deal...
 
my rpm goal is 720 , so I understand that eddy current will really not be a problem for me !!!!

the fact is that Litz wire are expensive and not easy to do by yourself :?

dual 0.8 mm wire that's for big kw motors ??
 
fdracing said:
my rpm goal is 720 , so I understand that eddy current will really not be a problem for me !!!!
fd,

With 32 poles you're running at 11,520 eRPM (192 Hz). If Lebowski was running his motor at this frequency, his parasitic losses would be nearer 80 Watts. Scaling up by a factor of 15 gives you parasitic losses of 1200 Watts.....
 
Hello,

The skin depth for copper at 192Hz is 4.71mm.
I checked your design with OD=400mm, ID=272mm, 32Poles, 24Slots, 4 turns/tooth (96turns/phase), I find about 200Nm, and you can use a wire awg 6 (4.12mm) for a ratio tooth/slot of 0.5.
I didn't check with FEMM...

Did you find this values?
 
Miles said:
fdracing said:
my rpm goal is 720 , so I understand that eddy current will really not be a problem for me !!!!
fd,

With 32 poles you're running at 11,520 eRPM (192 Hz). If Lebowski was running his motor at this frequency, his parasitic losses would be nearer 80 Watts. Scaling up by a factor of 15 gives you parasitic losses of 1200 Watts.....

? how do you figure this ?

The 20W includes all losses, so also the 'air pump' effect and bearing losses. And where does the factor 15 come from ?
 
dotrick said:
Hello,

The skin depth for copper at 192Hz is 4.71mm.
I checked your design with OD=400mm, ID=272mm, 32Poles, 24Slots, 4 turns/tooth (96turns/phase), I find about 200Nm, and you can use a wire awg 6 (4.12mm) for a ratio tooth/slot of 0.5.
I didn't check with FEMM...

Did you find this values?

I dont realy have tooth , my design is axial , with flat air core drop shaped coils , my first idea for the winding was : 26 meter of 1 mm2 copper wire / its about 150 turn , calculation are from a friend of mine :roll:

coils will be delta connected , 3 phases , 48 v .
 
Lebowski said:
Miles said:
fdracing said:
my rpm goal is 720 , so I understand that eddy current will really not be a problem for me !!!!
fd,

With 32 poles you're running at 11,520 eRPM (192 Hz). If Lebowski was running his motor at this frequency, his parasitic losses would be nearer 80 Watts. Scaling up by a factor of 15 gives you parasitic losses of 1200 Watts.....

? how do you figure this ?

The 20W includes all losses, so also the 'air pump' effect and bearing losses. And where does the factor 15 come from ?
It's true that I assumed all the losses were eddy current losses. How would you apportion them?

Whatever value you decide, quadruple it, if the frequency is doubled. The factor of 15 was scaling up from a 1kW motor to a 15kW motor. Is that not reasonable?
 
Do you have a drawing of your motor? What is your motor configuration, 1 stator-1rotor, 2 stators-1rotor, 2rotors-1stator... :?: ironless... many possibilities ... :D
 
my motor is 1 fixed stator with 24 drop shaped coils ( no iron in the coils ) , 2 rotors with 32 magnets each ( magnets are 30 mm diameter , 10 mm tick ) 10 mm iron at the back of the magnets

stator and rotors made in 2017 A aluminium

stator diameter 400 mm , 9 mm tick , rotor diameter 340 mm ( center to center of magnets) , booth air gap will be set at 0.8 mm

I want these motor 200 NM at 720 / 780 rpm

for the drawing i dont know how to do with posting dwg files :?

what i dont understand is why MILES is speaking about 11.520 rpm , when i said just before that the motor will rev 720 rpm !!!! any explanations :? are your erpm same as my rpm ??? for me rpm mean round per minute

my problem for the moment is the coil winding ,number of turns ?, wire size ? , litz or not ?, coils wire lenght ?, etc .. nothing else ( witch is enougth !!)

and please, can we stay understandable , I am not a specialist like many here , I need just some help from those who have already made such job or information that I can convert in fact

Tank you in advance :wink:
 
Miles said:
Whatever value you decide, quadruple it, if the frequency is doubled. The factor of 15 was scaling up from a 1kW motor to a 15kW motor. Is that not reasonable?
The 20W was for my 3kW triple stator motor...

Erpm, this is the electrical rpm, which is not the same as mechanical rpm. You'll probably have 16*720 erpm, where the 16 is 32 magnets divided by 2.

OP, have you seen my triple stator axial flux motor build thread ?
 
fdracing said:
my problem for the moment is the coil winding ,number of turns ?, wire size ? , litz or not ?, coils wire lenght ?, etc .. nothing else ( witch is enougth !!)
There are several parameters, it's difficult to explain with few words...

If you want to increase the torque you have to increase turns number. If you increase the number of turns, you have to decrease the wire diameter...you increase the copper losses. If you want the same wire diameter, you have to increase the motor diameter.
Wire size --> increase the diameter, increase the efficiency
Litz or not --> Check with your max frequency and the wire size.(The skin depth for copper at 192Hz is 4.71mm --> It's your max wire size)
Coil wire lenght --> it's proportional to the number of turns
 
dotrick said:
There are several parameters, it's difficult to explain with few words...
If you want to increase the torque you have to increase turns number. If you increase the number of turns, you have to decrease the wire diameter...you increase the copper losses. If you want the same wire diameter, you have to increase the motor diameter.
Wire size --> increase the diameter, increase the efficiency
Litz or not --> Check with your max frequency and the wire size.(The skin depth for copper at 192Hz is 4.71mm --> It's your max wire size)
Coil wire lenght --> it's proportional to the number of turns

That's exactly what I need infos that can be turn in fact :wink:

if my Aluminium parts are anodizied , they will not conduct anymore ?? on surface , what do you think !!, and the stator will be cooled , in a way or an other :wink:

I did not seen your triple stator , Lebowski , any link ??

I understand that the matter is very complex , and step by step I get infos in order to make my first test motor , the goal is one : 200 Nm at 720 rpm , I have time , money and the will to understand , what else !!!
 
I don't understand.... :?
No steel for your rotor? only aluminium?
How is the flux path? any drawing? :wink:
 
dotrick said:
I don't understand.... :?
No steel for your rotor? only aluminium?
How is the flux path? any drawing? :wink:

the design is very close to the yasa motor as woolmer first designed it , the coils are just thinner , steel is at the magnets back , but I am also thinking to do the rotor in soft iron two :wink:
 
what do you guys think about using thin copper strip ( example : 0.1 x 9 mm) instead of standard wire , it will be best for copper fill , and what about eddy current ??
 
Back
Top