Information about back iron dimensions

[LazyYogi]

I have finaly built my first axial flux (outer rotor) coreless generator rated at 3 kW for grid conection. Very happy about it but some problems did occure....

The nominal power of the generator was measured at 2700 W rather than 3000 W that was found in my FEMM-Matlab simulation. After measuring the generator's dimensions I found out that the airgap was larger. I executed a new simulation and the results were very good (1.5% error in the EMF and about 4.5% error in output power). But the problem is that the airgap was not uniform. In fact, the two steel discs (outer radious 480 mm) were not flat and so the airgap was difficult to be set at the correct value. Any comments on how to avoid this problem in future generators ?

Another problem I encountered had to do with the thickness of the rotor discs. I had (and still have) no idea on how to calculate this. Just some advice (the thickness should be equal to the thickness of the magnets for example) that I found here and some research on Hugh Piggot's and other constructed machines in order to find some kind of equation between the force of attraction and the thickness of the iron. In the end I just chose the 10 mm thickness thinking it's a safe choice. But now, wanting to build a larger machine, with more powerfull magnets than the N40 I used, this knowledge is essential. So I am asking your help for any information in order to produce a tool of general use...

In addition, I think that there is also a heating problem in the generator... "I think" is becouse I can't measure the temperature now that the machine is built and while it is working and mostly becouse I don't have any clue of what the impact of the (physical) ventilation of the generator will be on the temperature... Just to be sure, I was thinking of drilling some symetrical holes in the rotor.... Any comments on how to control the temperature?

Thank you very much!

[electrondady1]

steel that is the same thickness is a very safe bet
 
around this site 1/4"steel for 1/2" mags

not much leakage with 1/8"saw blades and 1/4" mags.

a measuring tool called a dial indicator can detect if your rotors are straight.

a big hammer is a handy adjustment tool.

[Flux]

Mild steel plate is not really flat and rotors just cut out without any machining may not run exactly true. Unless you have a lathe big enough to turn these parts then it is just luck how accurate they run. You can sometimes correct things a bit with a few shims under critical places.

The demands on the discs magnetically are not that stringent and if you choose a thickness that is adequate to support the attractive forces of the magnets that is normally sufficient. Changong the magnet grade won't make any spectacular difference but if you use thicker magnets or larger diameter discs for more magnets then the strength will have to be increased.

I don't know of any information that will give you the thickness required, it's quite tricky calculating the deflection of a disc under the attractive force of magnets. Most of this stuff is just instinctively over engineered. If you keep within the normal range of things then discs the thickness of the magnets seems to work out well enough even when people cut holes near the centre for additional air flow.

You are not alone in not knowing what cooling there is on this type of generator up in the air under working conditions, I am not aware of any serious work being done on this. i still suspect the cooling is a lot less than you might imagine. Even in a 30 mph wind there won't be a lot of air movement near the centre of the prop and what really happens around the stator disc itself is even more debatable. The other thing that is very uncertain is the actual duty cycle of this type of alternator. It is really quite low, if you try a bench test you won't sustain the expected full load rating for more than a few minutes.
Flux