How do I design an eletric motor to...

>>1214329
>making an electric motor stronger

Define stronger in terms of speed, torque, efficiency or something.

And of course wires that not intended to be connected have to be insulated by something.

File: Screenshot - 07232017 - 04:56:01 AM.png (77KB, 356x339px) Image search: [Google]

77KB, 356x339px

>>1214329

Pic related is for one particular voltage applied to a brushless DC motor like your picture. I assume that the outer ring in your picture is the rotor, which is unusual but works if that's what you need.

Increase the voltage and the line moves out parallel; more stall torque, more no load speed, more max power. You can apply any voltage to any BDCM as long as it does not overheat, break down the insulation, or demagnetize the magnets. Of these three, overheating is the main one that you might actually encounter.

If you make the magnets larger the only "simple" approach is to make the motor longer and that means the stator as well. This is a more efficient design because more of the copper is in the area where magnetic flux is and less is wasted in the ends. But if you are going to do this sort of thing then you need to be a motor designer and motor manufacturer and this is not the place to discuss that.

The only realistic change is to remove the windings and rewind it with heavier or lighter wire. This does not change anything about the motor or that speed torque curve except that the voltage and current will change, which might make your motor "stronger" for your particular drive electronics, but it does not change the motor's performance or efficiency in any inherent way. Second order effects might make a small difference, such as more efficient use of the slot area where the windings go, but in general there's nothing you can do to a particular motor to make it "stronger", whatever you meant by that.

It's an enormous pain, it's actually not overly difficult with modern software, there's just no decent open source software for motor design. The electric motor industry is exceptionally scummy, they sell motor design software for $15,000, any western company involved tries to keep electric motor production and design costs inflated and they are aggressive patent trolls.

If someone could get a few programmers to work on an open motor simulation and design program it wouldn't actually take all that much effort, never found anyone interested in helping though. With modern fiber lasers cutting motor laminations isn't too difficult for low production volumes. At the moment without paying loads of money your only options are doing 2d magnetic simulations and manually working out all the copper losses and other things.

The motor design software is so obscure I have never even been able to find a cracked version to try out. http://www.infolytica.com/en/products/motorsolve/ Maybe someone here might know somewhere that has cad stuff.

>>1214377
>It's an enormous pain, it's actually not overly difficult with modern software,

Has anyone ever made a homemade BDCM? I'm sure someone has, but as you said it's an enormous pain. Let's say you have access to a "modern fiber laser cutter" and you make a stack of lams for the stator, you wind the coils and impregnate them with something. Now you need to grind the OD and the ID; who has either capability? You also have to machine a decent rotor hub and bond the magnets to it, and then grind that OD as well.

It sounds like a lot of fun, just like making a home made motorcycle engine from blocks of aluminum and steel would be fun. It just seems impossible at the hobby level.

Designing motors is fun for various reasons. I did it for 9 years in the aerospace industry. Our biggest challenge was making a motor with a purely sinusoidal characteristic so that the satellite would not have to deal with vibrations when the Control Moment Gyro was spinning. Lots of freaky winding calculations and lam design features come into play. But just making a "better" hobby motor doesn't really make sense to me. The ones you can buy are well designed and perfectly good for the things we build.

But then I'm amazed at what people do. Making their own vacuum tubes and transistors in their basement, so don't let me discourage anyone from trying.

>>1214386
There's tons on youtube, it's not too hard to order the laser cut laminations. Coreless axial flux are popular as there's less metal needed.

The real thing holding them back is the lack of access to parametric motor simulation software. Building a program to do it isn't even that hard it's just building brushless motors is such a niche thing it's hard to find any willing to work on it.

This guy built a decent 3d printed motor using a halbach array to solve the issue of no steel backing for the rotor. Without access to simulation software his design is likely unoptimized though
https://www.youtube.com/watch?v=cDOunzr6lXo

>>1214390

Wow. That is amazing. I take back everything I said about it being beyond the hobby level. I just need to buy a 3d printer and go to work.

>>1214377
I found that program, version 4.1 (not the latest unfortunately) within 30 seconds on Google.

>>1214394
I wish I new how to make these windings, the simple design makes it easy for hobbyists to build. The steel tube it sits in is connected to the rotor, so it rotates with the field meaning there's no losses. Hobbyking only make it in one small size, as there's not much demand for sinusoidal motors from rc hobby people, couldn't find anything that shows who the oem is either.

Coreless axial is also good, but getting wedge shape magnets can be a problem.
https://www.youtube.com/watch?v=3j8JgDLrFYY

>>1214416
>I found that program, version 4.1 (not the latest unfortunately) within 30 seconds on Google.
Is the link legit though? I have only ever found fake sites or dead links.

>>1214355
OP here, I meant in terms of torque and speed. I was looking at 3d printed electric motors and was hoping that with a bigger printer like I have I could make some sort of electric bike motor.

>>1214329
OP here again, what about this software? https://www.emetor.com/file/open/

>>1214675
Can you 3D print silicon steel? If not, you'll never print a motor as efficient as one you buy. The strength would also be insufficient for high speeds and accelerations. It's probably easier to make an induction motor than a DC motor, so I'd look into that if I were you. But no matter what you're making, you'll still need a solid amount of electrical steel laminations to get any efficiency.

>>1214765
>Can you 3D print silicon steel?
Plastic impregnated with carbonyl iron powder might actually work

>>1214329
>making an electric motor stronger?
What type of electric motor? Different motors may need different things.

>>1214765
>Can you 3D print silicon steel?
Possibly but I would worry about the surface finish. Proper laminations have a coating on them for a reason:
>Coating. Laminations punched from semiprocessed steel are uncoated, while those punched from fully processed sheet are typically coated at the steel mill with a core plate coating. This coating insulates laminations from each other to reduce interlamination eddy currents, protects the steel from rust, reduces contact between laminations from burrs,

>>1214833
>Plastic impregnated with carbonyl iron powder might actually work
Again might work but at that point I'm not sure the laminations are doing anything. You might as well make the stator a solid piece.

>>1214675
What is it with nerds and 3D printers? Cutting metal isn't hard.

>>1214858
>You might as well make the stator a solid piece.
Have fun with your hysteresis. There's a reason why all modern transformer and motor cores are laminated; it helps prevent eddy currents from being generated by the changing magnetic flux and hence causing energy loss via core heating.

The laminations are insulated from one another and prevent any induced currents from flowing through the core, but I'm wondering if a 2D or 3D isolation (rods and cubes respectively) would work better at this than simple 1D layers. Putting particulate high-permeability alloy within a printing filament would be "3D isolation" and might be effective, but the effect probably couldn't trump standard laminations because of the average permeability being too low.

I think you wouldn't ever really need to use laminated cubes, because laminated rods in the direction of the magnetic field would stop pretty much all possible eddy currents, but the cubes might see used at the corners of a rectangular transformer if it would be easier than bending the laminated rods.

>>1214862
Definitely this.

>>1214878
>Have fun with your hysteresis.
Laminations do nothing about hysteresis. This is a fundamental property of the substance. They are there to reduce eddy current losses. And my point was that this impregnation process is likely to have its own problems that would be equivalent to iron losses in severity.

>>1214422
I'll download it when I'm back within Wi-Fi range and let this thread know, or I will make a thread for for it under motor design software

>>1214858
>Again might work but at that point I'm not sure the laminations are doing anything. You might as well make the stator a solid piece.
Well that's the point. The iron particles are very small and pretty much perfectly insulated from each other by the polymer binder, resulting in small eddy losses. You can just print the whole stator as a single piece in whatever shape you need (or mold it using epoxy). It will, most likely, have lower permeability than laminated one, but also lower losses at high frequencues.

>>1215109
Oh, solid pieces of 3D printed ferrite-plastic? I assumed he meant to just grab a chunk of roundstock.

>>1214329
Just increase the voltage. Add active cooling as necessary. Nothing else is practical.

>>1214329
for brushless use a delta wye switch and gear it up
for brushed just up the voltage till things melt