Any gripes with fan based stir plates?

I was sitting at work today extremely bored with my current project and had poly phase motors on the brain.

I am not discrediting the fan based stir plate. It is simple and seems to work wel; but, there might be another way to skin this cat.

First of all, does any body have problems with the fan based stirrers? Like thrown stirbars, dust, size, and noise?

I was thinking of maybe setting up a 3 or 4 pole system with 6 or 8 electromagnets arranged in a circle. Then cycle and swap polarities of the poles to spin the stirbar above it. The stir bar becomes the rotor essentially. It does sound like overkill, but there are no moving parts. Which I thought was pretty nifty.

What do you guys think?

I bet it would work really well. The fan solution is cheap but inelegant. Your solution is cheap (or could be), elegant, and perhaps overcomplicated. Maybe that's why I like it!

HMMMM???? I think you may be on to something. I just finished my fan style stir plate but I haven't made a starter yet. I like your idea.

would love to see your design.

drake

The only thing I wonder about is that you'll have to overcome the inertia of the stir bar so I suspect you might have to slowly build up the angular velocity when you start.

Sounds like the way a particle accelerator works... kinda.

I think that is exactly how higher end commercial stirplates work. If you build it I would love to see a circuit diagram.

I'm sure you could do something pretty straightforward with a 555 timer and a potentiometer.

weirdboy said:

I'm sure you could do something pretty straightforward with a 555 timer and a potentiometer.

Click to expand...

Yeah that could work, would require a bit of discrete logic too though. Im thinking of a cheap PIC with at least 4 Ouputs and an A2D for speed control input. I think the cost difference would be negligible. Push/Pull drivers for each pole and that would be all she wrote.

I am going to goto the craft store tomorrow and look for some thin plastic thread bobbins to hold the windings. Ordered some magnet wire earlier today too.

I just might be able to get the thing to fit into a standard CD case.

I think it's a brilliant concept and I see no reason why it wouldn't work. I would design it so that you could control both the sequencing speed and the strength of the electromagnets.

I also think you could sell a ton of these if the cost was low enough. The muffin fans work OK, but I think there may well be a better way.

This sounds like a slick idea...

Not only could you vary the "rotational" speed, (how fast you swapped polarities...or whatever)...but you could adjust the power of each electromagnet, right? I'm better that too would also allow you to fine tune your rig so you didn't throw stirbars...

Edit: Dammit, too slow again...today is not my day....and Catt22 said it so much better than I....

If you can make one these work on a 5 or 6.5 gallon carboy I would love one

Tough to get a single stirrer to work with the void on the bottom created by the curvature of the glass.

If you think its worth the hassle, making the ring of electromagnets variable in diameter will help avoid thrown bars when using bars of varying length. I've learned the hard way from my fan based stirrers that the magnetic poles of the stirbar should line up with the poles on the perm/electromagnets on the plate.

k1v1116 said:

If you think its worth the hassle, making the ring of electromagnets variable in diameter will help avoid thrown bars when using bars of varying length. I've learned the hard way from my fan based stirrers that the magnetic poles of the stirbar should line up with the poles on the perm/electromagnets on the plate.

Click to expand...

I agree that this would be a nice feature, but it would be much easier to simply design it for a 2 inch stir bar and leave it at that. A 2 inch bar should work with any size starter.

IMO, the critical factor in constructing a stir plate such as this would be the precise location of the magnets. Otherwise, the bar will likely rattle excessively and probably get thrown.

I like the OP's idea. Maybe add some leds to light up each time the phases change for a "disco yeast starter effect"; sure to be a hit at homebrew/rave parties!

Yuri_Rage said:

The fan solution is cheap but inelegant.

Click to expand...

Rational folks may disagree, but I think the fan based unit can be an elegant way to get the job done. Sure, some folks have made a regular Frankenstein's Monster out of their fans, but I think some versions are pretty elegant. Doubly so for mine because every piece I used would have been in a landfill had I not recycled them into a useful item. Just my two cents.

k1v1116 said:

If you think its worth the hassle, making the ring of electromagnets variable in diameter will help avoid thrown bars when using bars of varying length. I've learned the hard way from my fan based stirrers that the magnetic poles of the stirbar should line up with the poles on the perm/electromagnets on the plate.

Click to expand...

Well, I am going to start with a 1.5 inch diameter cause that's how long my stir bar is. I'll experiment with smaller ones, granted if I am successful. I bet it is nothing a few Teslas wont fix though!

Edit.. Or, I could make the cores about 1" by 1.8" and arrange them so they radiate out from center of the circle. Depending on how far they are from center you will have +/- 1 inch from center to work with. But that is if it is REALLY necessary to have the stir bar pass over the field directly.

This can be done rather cheaply using an existing simple Running LED kit (QKits Electronic Kits: FK115, LED Running LIght Kit 10 Dot Straight Pattern) at $5.95. The kit would be able to drive 10 electromagnets. Instead of driving the LEDS it could drive a simple NPN transistor which would switch an electromagnet. The electromagnets could be arranged in a circle using alternating poles making the stir bar spin.





You would have to calculate the DC resistance of the electromagnets so you can pick a transistor that can handle the collector current. A simple switching transistor would do the trick.

Only caveat to the above is that you'll want a diode on that load or else you'll get a current coming back from the coil when you switch it off.

You got half of it Sawdust.

Once the bar has spun 180* you need to reverse the polarity of the field. So a NPN and PNP channel transistor will be required on each side of the magnets. This can be simplified by putting each pair of magnets in series.

Yeah, I was going to put 2 leds by each magnet so you can see if it is active what polarity it is currently in. With the disco affect those yeast ought to be ready to party come pitch time.

Weirdboy, Yeah a diode in parallel across the coil will help with inductive kickback.

Hot Damn! Subscribed.

weirdboy said:

Only caveat to the above is that you'll want a diode on that load or else you'll get a current coming back from the coil when you switch it off.

Click to expand...

Ok gentlemen here is a circuit that will work much better, While driving home tonight I got to thinking an H-Bridge circuit used to reverse DC Motor direction may work. I don't believe in redesigning the wheel so I had a look on the web and came up with the circuit below. It has a low component count and can drive 150 mA.



The chip is a MOSFET Driver from Maxim (4424) and the diodes are 5817's. You would need a circuit for each electromagnet and a simple single level translator from 0 to 0.7 volts (LED ON) to 0 to +5 volts (Logic "1"). I hope this helps.

yeah, I had the L293B in mind but this requires 1 less control line. However, it can handle 2 at a time up to 1 amp per channel. Each channel drives TWO coilsoriented to have opposite poles.

2 drivers, 12 diodes, leds and a few resistors, voltage reg and a PIC. pretty small build budget.

the snubbing diodes don't have to be anything specific. 1n4007s are cheap and readily available. level translator for the led is a little overkill, resistor would be fine, 1.2kish depending on the current requirement of the led.

The logic stuff is pretty cake. I have a Pic Kit development package from a few years back from an old project.

rying to set up a jig to make the bobbins for the magnets. PIA!

Yeesh, that 4424 looks like it is going to be a little difficult to get your hands on it according to mouser any how

Im thinking I may just go old school and do it with discreetes, npn/pnps s are cheaper than dirt

If you use the LED board to drive the circuit I posted you will need to tanslate the 0.7 volts that turns the LED on to at least 2.4 volts. The TC4424 is available from Digikey real cheap.

Sawdustguy said:

If you use the LED board to drive the circuit I posted you will need to tanslate the 0.7 volts that turns the LED on to at least 2.4 volts. The TC4424 is available from Digikey real cheap.

Click to expand...

Ah okay.
Think I have found the first real obstacle though. The size of and/or the power requirement for a magnet strong enough to excite the stir bar with reasonable is going to be relatively large.

I made a 3600 At magnet last night and wasn't overly impressed. Not sure how much difference a soft iron core will make though. The only thing I had available was a framing nail.

CodeRage said:

Ah okay.
Think I have found the first real obstacle though. The size of and/or the power requirement for a magnet strong enough to excite the stir bar with reasonable is going to be relatively large.

I made a 3600 At magnet last night and wasn't overly impressed. Not sure how much difference a soft iron core will make though. The only thing I had available was a framing nail.

Click to expand...

I did a little searching with the googles and didn't turn up much in the way of small off-the-shelf electromagnets. Surely somebody, somewhere makes these things. The cost may make them prohibitive to use in this application as you would need about six or maybe eight of them to make it work. Four might do it if they were powerful enough.

I might be confused on your design, so bear with me, but would it not be easier to engergize closely spaced magnets sequentially to sort of lead the stir bar instead of switching polarity to make it a push-pull force? Just a passing thought. I really want to see you make this thing work.

Catt22 said:

I might be confused on your design, so bear with me, but would it not be easier to engergize closely spaced magnets sequentially to sort of lead the stir bar instead of switching polarity to make it a push-pull force? Just a passing thought. I really want to see you make this thing work.

Click to expand...

The problem with that is that once the stir bar has rotated through 180 degrees you will need an opposite magnetic field to exert the same force. Does that make sense now?

This does a fair job of explaining the principle.

Polyphase motor design : POLYPHASE AC CIRCUITS

I remember having a bunch of small 12VDC electromagnets scabbed off of some electro mechanical junk I had laying around. Dozens of them. Tossed them a few months ago thinking I was never going to use them.. Isn't that always the case?

The poles of the magnets do need to be in close proximity of the stir bar, especially if they are oriented vertical. If they are mounted at an oblique angle it may allow for modest deviation in the size of the stir bar. It would require a stronger field still and may introduce a whole new dimension of problems that I am not ready to deal with.

My second thought. More poles. The less distance between poles would mean that the electromagnets themselves could be smaller since they don't have to pull the stir bar as far. Could also do a push pull sequence with the adjacent poles. The sequencing wouldn't be difficult but I am not sure what the stir bar would should it be tossed.

Third thought, make a better stir bar... a stir bar made of a neodymium would easily be more attracted to the weak electromagnetic fields.

Starting to look like a $700 toilet lid. Not sure if it would work any better but, it would be different.

Any one else feel free to jump in and try too

dantodd said:

The problem with that is that once the stir bar has rotated through 180 degrees you will need an opposite magnetic field to exert the same force. Does that make sense now?

Click to expand...

I understand that part. I envisioned a multi-phase arrangement with maybe eight magnets instead of only two. Each pair (180 deg apart) could be switched on just ahead of the stir bar without the need to reverse the polarity of the magnets. IOW, instead of trying to rotate the stir bar through a full 180 deg, it might be better to only pull it through 45 deg before switching a pair off and the next pair on. You would need magnets that were small enough to arrange in a tight 2" diameter circle, yet powerful enough to pull the stir bar. That could be difficult to achieve.

Catt22 said:

I understand that part. I envisioned a multi-phase arrangement with maybe eight magnets instead of only two. Each pair (180 deg apart) could be switched on just ahead of the stir bar without the need to reverse the polarity of the magnets. IOW, instead of trying to rotate the stir bar through a full 180 deg, it might be better to only pull it through 45 deg before switching a pair off and the next pair on. You would need magnets that were small enough to arrange in a tight 2" diameter circle, yet powerful enough to pull the stir bar. That could be difficult to achieve.

Click to expand...

So you would use twice as many magnets and only use the + magnets when one half of the stir bar goes by and then use the - magnets when the stir bar has rotated 180 degrees and the other end is in that area? Seems that making the magnets reversible would be easier than doubling them.

I thinkyou guys are arguing the same point

Catt22, I had originally planned to use 6 magnets or 3 pairs if you would to move the stir bar 120 degrees at a time. each pair would be, as you say, 180 deg apart. My experiment was to see how the star bar would react to a home made magnet on one end. It wasnt very impressive but I was only using one pole of the stir bar.

Im crunching some numbers now on some coils to see what can be done.

dantodd said:

So you would use twice as many magnets and only use the + magnets when one half of the stir bar goes by and then use the - magnets when the stir bar has rotated 180 degrees and the other end is in that area? Seems that making the magnets reversible would be easier than doubling them.

Click to expand...

Actually, I think you are right. I forgot that the stir bar has two poles, so it seems it would be necessary to reverse the magnet's polarity to make it work. My reason for suggesting eight magnets was to make it operate more smoothly and more powerfully. Six or four might work just as well, but I don't think only two would do it. This is all just speculation on my part. Just thinking out loud.
It sure was a cool idea, but looks like it might not be possible to build one easily and economically. Suitable magnets would be the key I thiink.

The irony of it all is that you are trying to do exactly what the fan (BLDC motor) is already doing.....

Drake371 said:

If you can make one these work on a 5 or 6.5 gallon carboy I would love one

Tough to get a single stirrer to work with the void on the bottom created by the curvature of the glass.

Click to expand...

You make 5-6 gallon starters? WTF you brewin?

I am starting to think that with the amount of magnets needed, the current needed to drive them, the component count growing higher, this project is not practical at the DIY level.

I have a fan based stir plate. It looks neat, clean, it is absolutely silent (cant tell it is running unless the stir bar is on it) and it doesnt throw the bar.

To answer the OP

The Pol said:

I have a fan based stir plate. It looks neat, clean, it is absolutely silent (cant tell it is running unless the stir bar is on it) and it doesnt throw the bar.

To answer the OP

Click to expand...

Pol, you should know by now that us engineer types just can't leave chit alone. I am afraid after looking at it, that trying to DIY a Magnetic Stirrer without the use of a motor is expensive and not cost practical.

The Pol said:

I have a fan based stir plate. It looks neat, clean, it is absolutely silent (cant tell it is running unless the stir bar is on it) and it doesnt throw the bar.

To answer the OP

Click to expand...

Cool, thanks. I've come to the conclusion that this is a bit excessive.

Wihophead: where do you think I got the idea First thing I tried was to rip the fan off and give it a whirl .

At this point it has become mental masturbation.

For example, I have come up with the following to determine the resistance in a length of wire over a given cores Diameter, L inches wide, with x many layers using 30 gauge wire.

total resistance = Sigma 0 to x ((n*0.02)+D)*pi(L)1026.71

Ill probably build a fan based one and toy with this idea.

Sawdustguy said:

Pol, you should know by now that us engineer types just can't leave chit alone. I am afraid after looking at it, that trying to DIY a Magnetic Stirrer without the use of a motor is expensive and not cost practical.

Click to expand...

Exactly! It's not a matter of a did it cheaper or better but, Different! better and or cheaper is always a plus though

Something I thought of last night...

Electromagnets create heat. A pretty good amount of heat.
You put 8 together cycling on and off, and changing polarity constantly, you will probably create way more heat than your yeast starter should be exposed to.
This may actually be counterproductive.

That's a fascinating idea, turning the stir bar into the rotor! Polyphase motors with multiple poles in a size that would fit into a small box may be hard to find, though. Aren't most polyphase motors made for big pieces of rotating machinery to keep the torque constant? But it's an idea that would be cool to work with, if you have one less moving part that's an advantage.

For the DIYer, you can't beat the cheap, tough little muffin fan. I have done a locked rotor test on the fans I use for my stir plates and I can't get the things to fail. I simply stick a pencil in between the fan blades and the housing to keep the fan from spinning and wait to see how long it takes to fail. I have gone days with the rotor locked and when the pencil is removed, the fan just continues spinning. Short of running over them with a truck tire, they are pretty much indestructible.

Noise is not a factor. When they are enclosed in a box you can't tell when it's on or off. Vibration can be an issue if the drive magnets are not centered and spinning properly. And the heat generated by the electronics is minimal. With a fan running at 12 VDC and 100mA you might throw off 1 watt of heat, but with the fan turning that heat is dissipated quickly.