tag:blogger.com,1999:blog-8200098102909041178.post2867692373701442149..comments2024-03-24T09:12:10.872-04:00Comments on Shane Colton: Epic Axial Motor: Epic Axial UpdateShane Coltonhttp://www.blogger.com/profile/10603406287033587039noreply@blogger.comBlogger13125tag:blogger.com,1999:blog-8200098102909041178.post-57431952904311877772024-01-23T13:34:05.797-05:002024-01-23T13:34:05.797-05:00Shane, I have spent quite a bit of time reviewing ...Shane, I have spent quite a bit of time reviewing your Thesis work(s) as well as following your recommended reading from James Mevey. <br /><br />Studying the t = 2NIBLR was helpful and intuitive as I developed the conceptual concepts for Axial Flux Motors. Reviewing a sizing equation on pg 22 of https://web.mit.edu/kirtley/binlustuff/literature/electric%20machine/designOfAxialFluxPMM.pdf , you can work out all the same variable barring L and R.<br /><br />Numerically I end up a factor of 2 off.... I am using 0.0543790885 as the B value of flux density (following your number). 96 Active wires (N = 48) (4 coils per phase, 4 coils per turn, 48 total coils, 96 wires). A 4" Wire ( .1016M). Assume a 16" diameter motor with an active radius of 6.8" (.17272 M). RMS current of 102 Amps. Yielding 93.22 NM.<br /><br />The alternate approach involves integrating across the active region and yields 164 Nm. A factor of 2 or even sqrt(3) would be close.<br /><br />I have a spreadsheet laying this out well. Would you be willing to compare the two methods and comment on the difference in approach? As I read them, they should both yield near identical numbers as they follow simple Lorentz Law or BLi (BYi) laws multiplied by radius.<br /><br />I mostly wonder if I am missing a factor of 2 for the flux considering 2 rotors or if I misunderstand where to pull the radius number for or if I am misunderstanding that there should be a multiplication of sqrt(3) for a three phase system.... <br /><br />We use some Axial Flux Motors in my day job. My motivation is developing a deeper understanding, documenting a "rough" model that allows "rough" sizing, and eventually I plan to build a YASA or coreless system to study back EMF.<br /><br />Any input would be greatly appreciated!<br /><br />kai.justice@vanair.com <br /><br />Kai Justicehttps://www.linkedin.com/in/kai-justice-858bb895/noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-29021347787477696422012-01-15T09:44:51.111-05:002012-01-15T09:44:51.111-05:00notice SmCo magnets have higher flux density.
als...notice SmCo magnets have higher flux density. <br />also, your calculations about saturation seem correct... except you assume there is no current flow in the coils, contributing to the field.<br /><br />in practice yes, near-saturated motors work efficiently - at high rpm and low torque (and practice confirms this).<br /><br />but then they are not really efficient at stall or low torque/high slip , with efficiencies dropping to as low as 10% in such conditions. <br />for direct drive application like hub motor this sounds bad. <br />also rotational speed variance isn't that great, i.e. in high rpm mode eddy current losses contribute to efficiency losses and too long windings (high inductance) limit upper rpm of motor. <br /><br />given high frequency PWM (with freq's like 1Mhz for 100A which was not possible (so cheap) in 60-70's) is available to drive modern electric motors i do not see much point into increasing axial distance of windings (and their impedance). <br /><br />instead more of them can be ran in parallel , and more poles can be used. <br />the axial design of the motor, with two magnet plates perpendicular to eachother allow strong focus of magnetic field - and i think it is what axial motor is all about. <br /><br />few theoretical questions remain, like what is optimal number of poles and what speed range and stall torque it offers. it will ofcourse depend on motor diameter aswell.curioushttps://www.blogger.com/profile/12239217480104895956noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-89738238136610982822011-12-07T16:53:10.827-05:002011-12-07T16:53:10.827-05:00I later went on to make a coreless motor that was ...I later went on to make a coreless motor that was flatter as you describe. That would definitely be a better choice for an electric bike, since it would have zero cogging torque and the potential for lower eddy current loss, if the windings are done properly.<br /><br />NdFeB magnets have a remanent flux density of about 1.3T, and accounting for some leakage and the reluctance of the air gap, the average flux density in the air gap might be close to 1.0T. Silicon steel saturates at about 2.0T, so to not saturate the core, it would have to have at least half the area of the magnet disk at its narrowest point.<br /><br />Many motors run parts of the stator core saturated or near-saturated and still achieve high efficiency. Saturation does not necessarily mean loss. One benefit of the steel core is that it can focus a lot of flux through a long axial distance of windings.Shane Coltonhttps://www.blogger.com/profile/10603406287033587039noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-87840399002243635852011-12-07T01:17:07.822-05:002011-12-07T01:17:07.822-05:00hello. i have some doubts about core saturation an...hello. i have some doubts about core saturation and losses related - you seem to use neodymium magnets , and those give out enough field to saturate any type of steel core i am aware of. <br /><br />would it not be better to use no cores, flat coils, and increase number of poles (and magnets) - perhaps also increasing diameter of the motor ?<br /><br />this way magnets put on two sides of the design will be more close , and magnetic field inbetween their north and south poles will be stronger - creating less magnetic reluctance.<br /><br />also, motor will be more 'flat' which makes it better heat radiator. <br /><br />p.s. i try to design such motor myself, for use in electric bicycle - so low loss one (incl. low free spinning loss) - but i lack machining workshop and skills.curioushttps://www.blogger.com/profile/12239217480104895956noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-53763879119617455612010-05-21T12:35:15.684-04:002010-05-21T12:35:15.684-04:00Hello, I have a question about the magnet arrangem...Hello, I have a question about the magnet arrangement. How would arranging the magnets in a halbach array affect the LEAF motors performance? At one time, I had wanted to try an create a motor like this but in a much smaller foot print for use in rc vehicles. Unfortunately I have neither the skill or aptitude for this sort of thing.SteveJnoreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-48756536795776737242010-01-08T12:36:59.374-05:002010-01-08T12:36:59.374-05:00Thanks...Thanks...NeoRuleshttps://www.blogger.com/profile/08168847416178383162noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-7497552111378648682010-01-05T13:15:39.293-05:002010-01-05T13:15:39.293-05:00I think the eddy currents want to flow in a path t...I think the eddy currents want to flow in a path that the bolt does not short-circuit, but I'm not sure. The bolts themselves will see internal Eddy current and hysteresis loss, though.Shane Coltonhttps://www.blogger.com/profile/10603406287033587039noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-52850828529779836102010-01-03T21:09:50.994-05:002010-01-03T21:09:50.994-05:00How much problem does the bolt cause? wouldn't...How much problem does the bolt cause? wouldn't it short the lamination's or cause any hysteresis heating?Unknownhttps://www.blogger.com/profile/09350943739011052628noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-80853306938010386842009-11-26T00:10:34.579-05:002009-11-26T00:10:34.579-05:00These were made by Proto Laminations: http://www.p...These were made by Proto Laminations: http://www.protolam.com.<br /><br />They are laser-cut M19 steel, 0.025" thick.Shane Coltonhttps://www.blogger.com/profile/10603406287033587039noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-33485067650132853542009-11-24T21:22:52.646-05:002009-11-24T21:22:52.646-05:00Where did you get the rotor Laminations made?
Than...Where did you get the rotor Laminations made?<br />Thanks JCUnknownhttps://www.blogger.com/profile/09350943739011052628noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-84309750146102337762009-11-14T10:11:43.970-05:002009-11-14T10:11:43.970-05:00For now, something like 72-144V. I know that's...For now, something like 72-144V. I know that's a huge range, but I've been worrying a lot about the core losses at high speed. I'm working on a more thorough thermal model of the stator that includes both copper losses and core losses, as well as speed-dependent cooling. But like I said, heat transfer and fluid mechanics is not my strength.<br /><br />The copper losses are a little easier to predict, so I can say more confidently that it can probably run at 125A continuous current and 300A peak for three minutes, starting from 50C with 38C ambient temperature.Shane Coltonhttps://www.blogger.com/profile/10603406287033587039noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-24482813734022046862009-11-12T15:10:52.426-05:002009-11-12T15:10:52.426-05:00What is the design voltage?What is the design voltage?Electric_02https://www.blogger.com/profile/00081064670193199431noreply@blogger.comtag:blogger.com,1999:blog-8200098102909041178.post-83727246741856767822009-11-10T19:47:29.762-05:002009-11-10T19:47:29.762-05:00Man!!! You fabricate FAST!!Man!!! You fabricate FAST!!Electric_02https://www.blogger.com/profile/00081064670193199431noreply@blogger.com