I had been wondering if this would work for a few years, and finally got around to testing it

[G+_Chris O'Riley]

I had been wondering if this would work for a few years, and finally got around to testing it. Not particularly useful, but you can make a brushless motor using a hall effect sensor (one that has a high enough switching frequency), a mosfet and an inductor acting as an electromagnet. Runs pretty well on 3 volts, and I tested it with a 2032 coin cell.

 

I'm going to make up a PC board so all the components can be soldered cleanly and 3D print the armature and top bracket. Should make a good soldering project for my son and his friends. I'll post all the files once I have it ready... I should be able to get that done before Google+ shuts down :(

 

I'm already thinking about a more advanced version that would use something like a Tiny85 to read the hall sensor, calculate the RPM, optimize the timing and activate the mosfet. I could also use a series of surface mount inductors around the circumference. Should be a good project once my son is up for surface mount soldering!

 

[G+_Chris O'Riley]

My son noticed something interesting when watching the motor at night - a ghosted image of the armature caused by the LED. While the LED appears solidly illuminated to the naked eye, it's obviously still strobing and matches the timing of the motor when the hall sensor is moved.

 

I love unexpected little things like this!

 

[G+_Ben Reese]

That's pretty awesome! I saw a similar project using a fidget spinner.

 

And that ghosting effect is cool. Totally makes sense that it would happen, but neat to see it in action.

[G+_Chris O'Riley]

Thanks! I think I remember seeing that fidget spinner. I've always liked making little motors. I used to make them as a kid from paperclips, enameled wire and shower curtain magnets!

 

Here's a PC board I whipped up for this. It'll mount on a double AAA battery holder. The white area is for my son's friends to be able to write their names.

 

Still need to model the armature and top bracket, should be able to get that done by the time I get the PC boards back. As soon as I verify that it works, I'll post all the files.

[G+_Chris O'Riley]

Just got the first PCBs back from OSH Park. Need to make some small tweaks just to make it easier to solder, and I have to find a cheap small SPST switch instead of my jumper, but it works pretty darn good! Nice and smooth. I have to figure out a way to determine the RPM, just to satisfy my own curiosity.

 

Next up will be a surface mount version with an ATTiny that'll optimize the timing automatically. Right now, I just bend the hall sensor to one side or the other and listen to the sound to get an idea if it's making it faster or slowing it down.

[G+_Chris O'Riley]

Playing around with the little PCB brushless motor to see if I can get more RPM by controlling the timing with an Arduino. They're both running around 5,000 rpm, with the Arduino controlled one a few hundred more, but that could very well just be from slightly less resistance between the armature and the board/top bracket.

 

The bouncing around of the ghosted armature on the Arduino one tells me the timing is constantly changing slightly. I'm using a hardware interrupt to read the hall sensor and direct port manipulation to turn on/off the inductor/magnet so the control is as fast as possible, but it's interesting that it's varying so much.

 

The non-Arduino one uses a different inductor and is running on 3 volts (2 AAA's), while the Arduino one is running on 5 volts. I'll have to swap the inductor so it's more even.

 

Not sure what my ultimate goal is, or if I even have one! I started just wanting to make a cool little soldering project for my son and his friends, and at this point, I think I've spent more time playing around with this thing than he has!

 

[G+_Chris O'Riley]

Two more updates... I added (well, hacked in) a Tiny85 to control the timing, which let me then randomly vary the on/off times once it got up to speed to light the armature up at different points. Took a bit of fiddling to get the code to run the motor as fast as possible. It has a very holographic look that doesn't come across nearly as well on camera.

 

And once I had the code working to control timing, that let me make an armature with 3 magnets instead of 2 where one is activating the hall sensor exactly when the other was passing the inductor. The Arduino monitors the time it takes for each magnet to pass the hall sensor and turns the inductor on about 1/2 of that time, when the next magnet coming around is passing. The 3 magnet version runs at about 8000 rpm on 5 volts!

 

 

[G+_Chris O'Riley]