Sciencemadness Discussion Board

Homemade magnetic stirrer troubleshooting

ficolas - 13-1-2017 at 10:25

Today, I arrived my home town from the city Im studying in excited because a PWM I ordered for my magnetic stirrer finally arrived home. After some floor-testing (Floor testing: The testing done at the floor because excitement doesn't allow you to set up a proper working space), it basically doesn't work at all.
The stirrer is basically a meccano motor with two big neodimium magnets on top. Pics:
I added a ceramic tile t̶o̶ ̶m̶a̶k̶e̶ ̶i̶t̶ ̶h̶a̶r̶d̶e̶r̶ ̶f̶o̶r̶ ̶t̶h̶e̶ ̶s̶t̶i̶r̶ ̶b̶a̶r̶ ̶t̶o̶ ̶m̶o̶v̶e̶ to insulate it a bit

When I turn it on, the voltage needed for it to start with those magnets on is quite big, so the speed is too much for the stir bar to begin moving.
When the speed is big, the box thing vibrates a lot, because the magnet-thingie isnt really symetric.
And it seems like the distance between the magnets and the stir bar is too big for the stir bar to correctly catch the movement. Reducing this wouldn't be hard.


What would I need to do to fix it? Will I have to use a gear box to increase torque and reduce speed, while at the same time separating the magnet from the motor? Should I use electromagnets instead of a motor? That looks quite hard for me, but at the same time cooler! If so, any recomended reading on this?

[Edited on 13-1-2017 by ficolas]

[Edited on 13-1-2017 by ficolas]

Melgar - 13-1-2017 at 12:11

Ok, first of all, don't be surprised if you get a lot of advice to scrap it and start over. Most magnetic stirrers use Alnico magnets, which have a much higher heat resistance than most magnets, and a much, MUCH higher heat resistance than neodymium magnets, so they don't need so much insulation between the magnet and the plate. They also use a metal plate to evenly heat the surface, with or without a ceramic coating to protect it. I've always just used aluminum or nonstick (teflon-coated) aluminum electric pans for the surface of hotplates. Have I ever spilled acid on them? Of course! But even hot acid takes longer than most people think to eat through aluminum, and if you're boiling acid, you should have ammonia or sodium bicarbonate solution handy anyway to neutralize accidental spills. Aluminum plates are fairly cheap anyway. Really, you need to put more effort into reducing that distance, because magnet strength falls off with the cube root of distance, so if you can half that distance, the magnetic field will be about eight times as strong.

For a motor, brushless DC fan motors are ideal; a lot of times you can get them out of hair dryers or broken box fans. There are ways of telling what kind of motor is in something; basically if it looks like the motor hooks up directly to AC, then it's probably an AC induction motor, and if you can see it hooked up to at least one integrated circuit or silicon device, then it's probably a brushless DC motor, assuming it runs a fan. It's getting to the point where the reduced copper needed by brushless DC motors actually makes them cheaper than AC induction motors despite all the extra control needed, which is good for us. But anyway, fan motors are really good, especially out of smaller things like air conditioner fans, because their speed and torque is optimized for a similar speed range as magnetic stirrers need. The motor you have right now probably isn't. Brushless DC motors are also good at providing a high starting torque; I did part of my masters thesis on measuring torque from a endodontal (dentist) drill, and those things can provide VERY high starting torque for their size. However, if you want to just have a DC motor in there working, while you wait on whatever part you order, broken inkjet printers are a great source of motors that are strong enough to turn a magnet that size.

Your problem is with starting torque, I'm almost certain. Try giving the motor a little spin with your finger as you turn it on and see if it doesn't stay spinning.

edit: oh, one more thing. try turning the shaft around and you'll notice that it tends to settle at certain positions. How many positions are there in one rotation for your motor? It's usually an even number, and probably ten or fewer.

[Edited on 1/13/17 by Melgar]

ficolas - 13-1-2017 at 12:25

Its a wooden box, magnet shouldn't suffer much from the heat, otherwise Im going to have bigger problems than the magnet loosing its magnetism.

Brojen inject printers... I happen to have a broken inject printer!

I'll change the motor to the motor I dispose from the printer, or to a pc fan if it gets complicated in some way (the fan has a very nice shape for this), and make a hole in the top cover to house the magnets, so that the distance gets way closer.
Its a wooden box because its just a "temporary prototype", the kind of temporary that last 3 years, but you know what I mean.
It is with starting torque for sure, it works when raising the voltage high, and then lowering it, or when spinning it with my fingers.

Thanks for your help :)

Edit answering your edit: 6, what does that mean?

Edit 2: I just tested a 12VDC fan with these magnets on it, it gets completelly stuck, at any decent voltage. With more spacing between the fan and the magnets, it goes veeeeeery slow, and it needs help to start (at 12V)

[Edited on 13-1-2017 by ficolas]

[Edited on 13-1-2017 by ficolas]

Melgar - 13-1-2017 at 17:40

Not computer fans, those are way too small, and much too high speed. I mean like something where the blades are 20-40 cm across, or something that pushes a lot of air, like a hair dryer. Also, better if it has speed control and is somewhat newer, otherwise it's probably an AC induction motor.

Interesting aside: you know how fans go Off -> High -> Medium -> Low, rather than off, low, medium, high, which would be more intuitive? It's because that way you have to have it on "High" for a split second to get it turning, before moving to a lower speed.

If you want to just add some circuitry to make your current one work, you could have it discharge a capacitor through the motor when you turn it on or something. run the whole thing on 12V nominal, then use an adjustable voltage regulator and potentiometer to change the speed. Then have your switch charge the capacitor when it's set to off, and discharge through the motor when it's turned on. You may need a DPDT (dual pole, dual throw) for that, but those are easy to find.

Also, try mounting the magnets on a round steel plate. That should keep their fields from interfering as much with the magnets in the motor and vice versa. The plate forces their fields to intersect at its plane, and spinning two fields that intersect on a plane doesn't require as much energy as two fields that are constantly interfering with each other.

Six is okay, but for high starting torque, the more you have, the better. Computer fans would only have four positions, for example, and some only have two.

ficolas - 14-1-2017 at 09:17

Ok, good news, I found the biggest issue with the starting torque... and bad news, im dumber than I thought. The magnets were touching wood on top, making it take a lot more than it should.
It alredy has a square steel plate below the magnets, so the magnet shouldnt be interfering that much?
Now the biggest issue, other than the distance between the magnets and the stir bar, that I can fix by making a hole in the bottom of the cover, and raising the magnets, is the vibration. The thing vibrates A LOT. And by a lot, I mean, it would separate big particles from small particles in no time if I placed a container on top of it. When on the carpet, that it has somewhat more friction with the floor, it moves at a rate of (very roughly) 1cm/second at max speed (max speed is overkill anyways, but you get the idea), I wanted to make a magnetic stirrer, not a car. And when on a table, its even faster. How can I make it vibrate less? I guessed it has something to do with the weigh simetry of the magnet-thing, however I have no idea how to know where there is too much weigh and where there isnt enough so that I can add/reduce it. And im not even 100% sure that is the reason, however knowing how ERM motors work, it seems to be exactly that.

Melgar - 15-1-2017 at 05:06

Well, obviously your magnet isn't well-centered on the shaft. Try holding the motor in your hand while it's spinning, alternating between holding it loosely and holding it tightly. While you do this, watch how the center of rotation changes for the spinning magnet. The rotational center of inertia will be at approximately the center of the spinning magnet when you're holding it loosely.

eesakiwi - 22-1-2017 at 07:19

I'd use a larger 6 inch + dia fan motor off a power supply or such, the sort with a metal housing and a fan centre dia of 2 or 3 inch.

Break off the fins and just use the centre part.

Mount your magnets with epoxy.
'Pull' the magnetic field into two spots where the stirbars ends are going to be by adding a couple of small neo magnets at those points.

ficolas - 12-2-2017 at 09:28

Quote: Originally posted by eesakiwi  
I'd use a larger 6 inch + dia fan motor off a power supply or such, the sort with a metal housing and a fan centre dia of 2 or 3 inch.

Break off the fins and just use the centre part.

Mount your magnets with epoxy.
'Pull' the magnetic field into two spots where the stirbars ends are going to be by adding a couple of small neo magnets at those points.

This weekend after finishing the exams I finally went home with time to work on the stirrer, I minimiced the distance between the magnets and the bar, and it works better than I expected! (The speed it takes is way more than enough, without the stirr bar flying everywhere)
The two magnets toguether are as big as my stirr bar, as (the size and placement of the magnets can be seen in a photo), would adding two small magnets still help? I can but a couple small magnets if that makes it better, because when I add the heating mangle + insulation I dont know if the distance will be toó big for the stir bar to move.