APL's DIY axial-flux motor

[APL]

Finally got some time to take the core out of the mold. Turned out really good.



Ferrite is 29 grams, and steel is 63 g, so about half the weight of steel. Defiantly need to add more ferrite to
the core. I still have to weigh some laminations. I'm guessing, flux density and steel content go hand in hand.

Now I can test for magnetic strength, and eddy current ability's. The plan is to run AC current through them to see
how they react to eddy fluctuations, by measuring the heat they make in a specific time.

 

[APL]

Turns out that trying to pump current through 10 strands of 4 foot wire is no easy task! Ha! :?
Pretty much a solid copper rod. It's only when these coils are strung in series around the stator,
that they have a higher resistance. So I'll have to go single strand for the magnetic comparison.

The AC test was a bust, but no big loss, since this information is already known, and there's no since proving whats
already been proven anyway. The magnetic comparison is pretty much the same,..half the steel content will result in
half the magnetic ability,.. however , I will do a test to confirm this anyway. Can't assume everything.

It would be more useful to have some factory ferrite to test against the DIY stuff, but I'm having a hard time finding
some, close to the size I need, and since it's ceramic, machining it is very difficult. But I'll keep looking.

If I had a corporate funded lab, I would just have it made,..but this is DIY out in the garage.

After much thought, it's looking like thumbs down on the ferrite prospect.
It's not viewed as a loss, or a failure, on the inventors trail.
The knowledge gained from the search for information, and the experience gained from the experiments are more valuable,
and can be used down the road for other things. Like a needle on a compass, it just points you in the right direction.

I don't think the resin content of the part is the problem. I poured just the iron powder in the mold. and weighed it.
There was only one gram difference between the powder and the part. So it has to be the space between the particles.
Steel is twice as dense.

The good news is that I was able to find a supply of lamination material. Surprisingly hard to do, unless I wasn't looking in
the right places.

At any rate, I think I'm just going to have to 'Man Up' and make the sixteen million lamination parts. The bottom line is
that the outcome is known to work, and the effort won't be a waste.

 

[APL]

Back to the original aluminum stator plate idea. But I was thinking of making the rotors out of 1/8" steel, instead of
1/4" aluminum. Steel is three times heavier than aluminum, but also three times stronger.
(strength is defined in different ways)

Half the thickness for steel, is still a little heavier than aluminum, but I wouldn't have to make the back iron rings for
the magnets,..the magnets could be placed right on the rotor, saving the weight of the two rings.
(thats if 1/8" is enough material for them)
Plus, I could drill or machine a lot of holes and relief in the steel to lighten it further.
I could also turn down 3/16" plate steel on a lathe to get more back iron thickness if needed.

I fully expect this motor to be loud and noisy because of low material. One of the grate features of the Crystalyte motor
is their quietness, but that comes with a sixteen pound cost. Noise is something that can be dealt with later.
The primary goal is just to get something to spin!!

 

[APL]

The Lamination material didn't have a lot of info. on it. But after reading more on this stuff, it's CRNGO
transformer steel. Thats cold rolled in one direction which makes it up to 30% better transfer in that direction.
And from the looks of it,..it's fairly high test stuff.


View attachment 3


Calipers read .026" on each sheet. But after looking closer, I found that it was actually two pieces stuck together!

That makes each sheet .013", which is really thin. Good for the motor, but bad for me, because I have to cut,
and work with twice as many pieces now.

View attachment 1


I'd of been happier with .035" on this motor,.. since it's not exactly a high performance motor.

The steel is a little brittle, but can be cut with the snips. Not something I want to do though, with 24 magnets.
So, I tried a hacksaw, and it cut through a stack of them fairly well.





Still a lot of work, so I'm thinking 'band saw' at this point. I have a plasma torch, but I don't want to heat the metal.
Still an option though.

In manufacturing, this stuff is usually laser cut.

The big question I have is, can I machine it?,.. all the thin laminations are going to catch and bend with the cutter.
So that will be the next test. Otherwise I'll have to grind it,...which is not something I want to do.

I'm going to epoxy all the pieces together first, so it will stay together while I try.

 

[APL]

More information on lamination material, and manufacturing.


https://en.wikipedia.org/wiki/Electrical_steel

http://www.protolam.com/page7.html

.

 

[amberwolf]

If it helps, Big Blue Saw here on ES has a thread in the marketplace for waterjet cutting, and has sales now and then. I'd guess you would probably want a roll of lamination material, isntead of individual pieces, to make the process easier (and cheaper) for the cutter, but you'd have to ask them about that part.

 

[APL]

Thanks amberwolf, I'll look into that,..might be the way to go. Would sure make things easy.
I've got to beat it up on my machines for a few days, and see what I can do with it first. If it machines at all,
maybe I can plasma cut it larger, and then machine the heated part away. Maybe.

Having it done is a good option, considering how much time it's going to take to make all these.
Otherwise,.. I know that they punch press this stuff,.Maybe DIY one of those? I'd like to do it here in true DIY fashion,
but sometimes the work involved is just too much for simple tools.

I'll know more in a few..

 

[APL]

Some follow up on the magnetic test results. A comparison between the ferrite, and the solid steel cores.
I downloaded wtubers magnometer/smartphone idea, and devised an analog digital scale version as well.


View attachment 1





Using the same single strand wire on both, and the same duration, at the same temperature.

Steel core was 1700 with a drop to 1400 on the magnometer. 350 gram pull with a drop to 320g on the scale.
Ferrite was 680 with a drop to 540, and a 35 gram pull down to 33g.
Temperature was 63 degrees, and climbed to 83 degrees. (aprox.)

This is all 'comparative' info., and the numbers don't mean anything by themselves.

It was interesting to see the magnetic field drop, as the wire heated up,.and the resistance in the wire went up.
I wonder if anybody has tried putting 'dry ice' in their motor yet?

At any rate, it's another 'busted' on the home made ferrite idea.

The smart tool program is very simple to use, and works quite well. It's very sensitive, and even reads the earths
background magnetic field, and orientation.
Took a while to get it set at distance that was 'in scale'. I can see where this will be useful in the future.

Heres the app link again;
https://www.vieyrasoftware.net/physics-toolbox-magnetometer

 

[wturber]

The smart tool program is very simple to use, and works quite well. It's very sensitive, and even reads the earths
background magnetic field, and orientation.
Took a while to get it set at distance that was 'in scale'. I can see where this will be useful in the future.

Heres the app link again;
https://www.vieyrasoftware.net/physics-toolbox-magnetometer

Welcome to the Dark Side. ;^)

 

[APL]

The force is getting stronger...

 

[APL]

The machining went OK, I could tell that the cutter didn't like it, and was probably getting duller by the minute,
but it was doable. If I could cut the piles close to the shape,..and the machining was just a clean up, then it would
be OK.





I was surprised to find out that the laminations were all electrically connected. Running an Ohm meter across the
machined surface came out a dead short everywhere. I thought the laminations were supposed to be separated, but evidently
a surface short is normal. I checked two Bionix stators, and they were the same.
I have to assume that overall, the laminations are separated, and a surface connection is a low percentage.

I'm still looking for a better supply of 'silicon steel', but almost all of it is made overseas. I also looked at Big Blue Saw's
site, and didn't see any lamination cutting specifically, but I'll have to E mail them and see what they can do,
maybe they can do the stator plates as well.

I'm starting to think about going straight sided cores, and a few less poles for room, ..and expediency in building.
Once it's working, I can fine tune, or make new plates and add more. Hmmm..

 

[amberwolf]

I have to assume that overall, the laminations are separated, and a surface connection is a low percentage.

Commonly, the surfaces are not connected everywhere, but many lamination stacks for various things (motors, transformers, etc) have a weld across the stack somewhere that electrically connects them, or are pinned (or bolted) thru the stack in one or more places, etc.

 

[APL]

Thanks amberwolf, I've seen some of the welds, and have wondered about the rivets that hold the stators together.
Apparently, the mill cutter is smearing the edges together. I'll have to try and belt sand, or wire brush the surface,
and see if that helps. More R&D.

I got an E mail back from Big Blue Saw, and they have an extra charge for using my steel, plus the charge for cutting it.
So it looks less attractive at the moment. Cutting all these by hand isn't very attractive either.

I guess its time for the plasma cutter. If I keep the stacks short, maybe it will cut clean, and quick. Won't hurt to try

I knew laminations were going to be trouble, but I still have some ideas to try out yet.

 

[APL]

Found some time today to try out the plasma cutter. Fall is here in Wisconsin, and I'm busy getting things ready for the
'long cold white' these days.

I was able to cut 3/16" thick stacks easily into strips that can be cut to 35mm length's with the hacksaw.
So thats a good thing. Three of these strips are about one cores worth.
The bottom lamination will have to be discarded because of slag, but thats no big deal.

Things should go pretty fast once I get a method perfected.





I'm sure there will be a performance loss from using heat to cut them, but theres also a gain from using .013" laminations.
Either way I'm going to try them out and see what happens.

The industry's usual practice is to anneal the parts at about 1800 degrees, if they have been stressed, but it has to be done
in an inert atmosphere, in a controlled way. A bit beyond my reach at the moment, but perhaps something that I can research
should it be necessary.

 

[APL]

I've been busy cutting stacks of laminations, a very tedious job. Did I say I hate laminations yet? .. I do! :x
But it has to be done,..so be it.

I ordered out the 3" aluminum bar, for the axle, so at least I'll have that to show, while I'm doing the lam's.
Theres going to be a heap of lathe turning and machining on that.

Things are moving along, but it gets slow sometimes.


I also had an idea for the 'V4' bike a while ago,.. a fairly common cruiser design that I wanted to try and modify
for electric conversion. A cascading design, with a balloon bike feel, and a touch of board track racer.

Kind of a 'Muscle Cruiser".

After many try's, I think I've finally got it to work. I don't know if it will make the reality list yet, but it's looking
like a goer at the moment. The huge 12ah batteries are a bit hard to hide, (there will be covers), and wheel sizes
are still up in the air. I'm trying to leave more room for the seat to move back,..for larger folks.

The problem is, I've never built a 'bent tube' bike before. So I'm not sure that I can bend large tube chromoly in
to this shape yet. That stuff is made 'not' to bend! So I'll have to do an experiment.

Otherwise, if it works out, I'll probably start another thread, and show the build, start to finish, so it doesn't
clash with this motor build. I'm a lot better at building bikes than I am at building motors.





Let me know what you think of it,.. I'm blinded by designer-hood.

 

[spinningmagnets]

I love it!

 

[amberwolf]

You might look up AussieJester's Custom Cruiser build (trike and bike); he shows how he built each one, in steps; including the bending processes.

FWIW, I'd also go for a springer-style fork, so the front is curved like the rear (since it's really the only straight lines, and sort of looks out of place).

 

[APL]

Thanks Ron, and amberwolf, sometimes I stare at a design so long that I can't see the obvious.

Thanks ambewolf for the Aussiejester link, Ive seen those benders at HF, but always thought they were a bit
wimpy looking,.. good for small stuff. It's amazing that he was able to bend the large tubing.
He has amazing building skills! Beautiful bikes. Gives me hope.

I'm thinking that between one of those benders and a curved wood bending form, I should be able to do it.
The curves are pretty mild. The top and down tubes are the problem tubes, the small tubes are easy to bend.

I designed this battery case for the big Milwaukee 12ah batteries, but got to thinking that maybe I should take another
look at the 18650's option. Probably a lot cheaper than the packs, and possibly more Ah with single cells,..plus,
a lot thinner profile, at only 3" wide. I'll have to give up the one hour charge though.

I don't know a lot about them unfortunately,.. having spent all my time with drill packs, maybe you could suggest a
good choice for a cell,..Panasonic? Samsung? 3000ah,..3400ah? Whats the best value?
Meanwhile, I'll start searching the ES for battery pack builds like this.

Keep in mind, I'm just thinking about this.

I drew up a mock fitting to see about how many cells would fit in this case,.. so that I could keep the option for both,
and came up with about 150 cells. Perhaps I can't pack them that tight? Need room for end caps?





With the 3400ma cells.
At 37-40v, it would be 15 sets of 10 cells, 10S-15P for 51ah.
Or, at 55.5v it would be 10 sets of 15 cells, 15S-10P for 34ah.

Both of those would kick the drill pack's ass,.. any insight to all of this?
I know I have to tackle the BMS's yet. I'm just thinking out loud at the moment,..feel free to shoot it down.

 

[izeman]

The layout as shown would work, but only without spacers. I learned that it's recommended to have at least tristar shaped spacers between all cells, to avoid any possible shorts that may happen if the battery flexes and the regular heatshrink may fail.
The cells can be glued together and will be held together be the nickel as well. Still they need support as there should be no stress on the welding points.

 

[APL]

Thanks for the info. izeman, I re-read your Kona build, awesome bike! :thumb: Lots of great information there, and a
surprisingly similar battery pack to this one. I defiantly need some kind of spacers.

Well, as usual, I've got a lot of research to do on this idea. The power density, and price is way better than the
drill packs. But the pain in the ass factor is way up. The battery cell arrangement is easy enough, but wiring
up a bunch of BMS's and welding the cells ..isn't.

Well, this build is still a little in the mist yet, and I have to figure out the finances, and 'build timing' yet,.. where
I'm going to fit it in. When I do, (hopefully soon) I'll open a new thread in this category.
Have to say though, I'm liking this bike more and more!

I need to do some tube bending experiments first.

For now, I need to concentrate on this motor build, or it's never going to get done.

 

[APL]

Speaking of which,..I got the axle aluminum the other day,.. a solid chunk of 6061 3" wide and 6" long.

Not the 7000 series we talked about, but strong enough for a prototype.





Somewhere in there is an axle,.. screaming to get out!

Unfortunately, my little crap lathe isn't up to snuff these days, and I'm trying to locate a better one.
Trying to find a lathe on Craigslist is like pulling teeth! The right price, the right place, and finding an
actual functioning human being to buy it from, is near impossible!

But I will prevail!

 

[amberwolf]

Being unable to get the poster of an item to either respond in a useful way, or sell something to you is really frustrating. I've been around and around that with old mopeds and motorcycles, trying to source parts for my more extreme-load cargo stuff. :/

Keep an eye on your local Freecycle / ReUseIt lists, too.

I got an ancient 1930's? 40s? South Bend lathe several years back that way. Required some cleanup, but was usable as-is (though I barely know how to use it). Have done some small work with it now and then, but only with crappy HarborFright cutting tools (weren't any with it). Used it as a sideways drillpress a few times, too. Someday I'd like to get better cutting tools for it, and learn how to really use it, especially since it's possible to cut gears and threaded items with it too. (I think I have some of the gears to do certain threads, but I've yet to learn how to set it up for that). I could make my own axles if I ever got that far.

 

[APL]

Yea, the Craigslist experience is an acquired skill. I discovered the 'all' search mode at the top of the category selection
box, that searches all the category's at once. Theres a lot of people that don't put things in the right place, or even
call them the right thing. I have to get creative in how to ask for something.

But, it payed off finally! I just picked up an older South Bend, like yours, yesterday for only $400. About a third of
what they usually go for around here. It was in the 'general' category, instead of 'tools'. I had to drive a hundred miles,
and call the guy every hour, to keep him honest, and make sure he didn't sell to someone else for a higher price,..
which has happened to me twice now.

It's a bit crusty, but South Bend's are all excellent lathes, and made to run forever. Built back when people cared about
quality. Eventually I'll get it all cleaned up and looking sweet.
This one is a South Bend 9" 'Junior' model. Late 20's early 30's. 3 foot bed.

This is how I found it.




I don't know how to run a lathe either,.. so it will be a learning experience for us both. I used to share a shop with a
machinist so I have a little residual knowledge about them from that, but that's about it. It's a machine I've needed for the
last 30 years and I'm excited to finally have one. Between the mill and the lathe, there isn't to much you can't do.

Luckily, this axle is aluminum, so it shouldn't be to hard to do,.. just an awful lot of turning to pull a skinny little axle out
of a three inch bar! That will give me some practice. The bearing surfaces are going to be touchy though. :?

 

[spinningmagnets]

Start watching youtubes from "this old Tony", and the right column thumbnails will have other suggestions.

https://www.youtube.com/user/featony/videos

You are halfway to being ready for the zombie-pocalypse...

 

[amberwolf]

The South Bend Lathe website (or it's wayback machine version) has all their old instructional films up as video files, and they have the manuals for them, too.

EDIT: apparently they don't anymore (not that I can find) but there's a bunch on YT, including "how to run a lathe":
https://www.google.com/search?q=South+Bend+Lathe+videos&num=100&newwindow=1&prmd=ivns&source=lnms&tbm=vid&sa=X&ved=0ahUKEwiRr9Tu4sXeAhVGfbwKHQ4dCuEQ_AUIBg

There is also a forum Practical Machinist that has a lot of good info on these lathes, including restoration help.
https://www.google.com/search?q=forum+Practical+Machinist+sbl+site:www.practicalmachinist.com&num=100&newwindow=1&hl=en&sa=X&ved=0ahUKEwi3xrKg48XeAhVDTLwKHef_CU0QrQIIHDAA