Halbach PM generator

thingiverse

After looking at https://www.thingiverse.com/thing:1693579 i decided i would try and come up with my own design, this is my results, using 44 15mmx3mmx2mm N50 magnets. the prints come out well, ill upload pictures later. my design differs in it actually gives you a case to mount it in using 3mm cap screws to secure one half of the stator winding to the stator cap then running 4 4mm all threads from one cap to another and using the remaining 4 screws in wither side to mount it to a bracket (make your own to fit your application.) Blender files are included. scale in cura is 0.1 this is a work in progress ill have field test in a few weeks when my magnets get here from china. another thing about my design is it uses irregular size bar magnets to test the real world effects that having different sized magnets may have on a design. i did this because magnets that are different sizes are cheaper than a perfet rectangulat magnet . i paid 4$ for 100 of the above mentioned magnets, which will allow me to create 2 of these PM generator heads. reading more on magnents i have discovered in the case of small magnets the entire 50 passes of wite crucially has to be within 1MM of the face of the magnet which means on outer frame will not work, i dont have a halbach version yet but i have a longer larger rotor that will work with a standard N-S magnet arrangement using my stator caps and a longer 10MM bolt or all thread as the shaft. this design uses 100mm x 10mm x 3mm N50 Bar magnets. Whats next: integrated cooling Halbach frameless for small magets the stl file 68mmfaceless is a design to perhaps help improve the efficiency of this design https://www.thingiverse.com/thing:1733875 it will not work with my stator assembly, however it will work in the aforementioned stator. by pusing the magnets to the outside and using CA glue you can minimize the space between the windings and the magnets thus yielding better results. i also figured that i would integrate a fan impeller in the design. when i get a moment i will print it. my intention for these is not low speed applications, in fact i want to put it in a plasic box and see how far i can go be fore it destroys itself. i figure if i use a fan impeller in the center to attach the hub i will beable to create some sort of ait flow that will reduce magnetic friction in the coils, a cool coil is a happy coil. Update 6/28/2017: after a few countless updates to the design files and a test i hereby deem this design useless overall. after changing about every thing to do with this design and assembling one the tests come back as such: 1. with 5 winds of 22AWG wire and spun at about 3500 rpm ( as fast as a DeWalt 18V drill will go ) i generate 0.489VAC on average 2. based on earlier tests i generate 0.150VAC at 3500 rpm with a single winding 3. based on research the current i am generating should be more than the voltage becasue of the diameter of the windings, however i am unable to test this as the current side of my meter hasnt worked in a few dozen years. 4. the faceless rotor is needed with such small magnets but i would expand this to all magnets for best results becasue the highest flux is as close to the magnet face as possible. 5. with 4 mentioned i will also mention that loctite and Neodymium magnets and pla are not the best of friends and about 20 of these magnets are still embedded in my wall from higher powered air compressor tests. (air compressor used to spool the turbine) 6. surprisingly the PLA never shattered not even when approching 20KRpM 7. based on my experience you would need to have about 25 windings on this to make it close to useful 100 would be better, however 100 22awg windings doesnt fit on this design. using smaller wire would give you more windings and less current which has very few applications. I plan to buy smaller wire and try it one more time before i junk it. 8. i suppose when you consider it is only the size of a coke can when assembled it works pretty dec/ent. however in later testing it turns out to make a better bipolar stepper motor than a generator. 9. learned a lot will post the one that makes 5V soon... 10. the one thing i did learn that was important here was that the rectangle magnets vs the square do just as good a job if not better of moving the NS magnet field when arranged in a Halbach array. there is definitely a doubling of the magnetic field on one side vs the other, until your magnets com flying out and you cant reassemble the assembly the same way as it was before and you mess up the order then you get funny results. 11. the designs for the integrated fan in the rotor have prove4d pointless and move very little ait instead i will be focusing on reduced wind drag in future designs. UPDATE 6/29/2017: this morning i unwound a transformer from a small switched mode power supply and rewound the generator in 5 winds of 0.25MM (30AWG) magnet wire. 5 winds is all the wire i was able to get from the transformer. the results are much better. with e the Drill i was able to generate 0.538VAC and at about 10KRpM i was able to generate 1VAC. the reason i feel these results are better is because the smaller windings will allow e to put more windings on the spool allowing for more voltage, however the less current will probably be a problem, i will in the end probably try for 200 windings and get or wind a 2:1 transformer that will allow me to reduce the voltage and double the current. I have just placed the order for the wire it should be here within the week. on another not i have yet again blown the magnets out of my rotor, i am going to print another rotor today and this time i am going to assemble it with the CA glue and run a single run of electrical tape around it or maybe heat shrink to secure the magnets for good but also minimizing the amount of space that is between the magnets and windings.