Electromagnets to stir water

I am building a heat exchanger and looking to add something to stir 10 gallons of water in a stainless steel vessel with electromagnets. I don’t want to use a small motor and prop as I want something hidden and out of the way. The vessel will have a 6000 watt element in the bottom center and a Stainless steel coil 4-6” off the bottom to pass my secondary liquid through. I want to use a series of electromagnets in a circular pattern about 8” in diameter around the heating element on the bottom. I want to be able to cycle the electromagnets in order to pass a ball bearing in hopes that it will stir the water. I am a jack of all trades master of few and having problems with initial design and feasibility. I have seen on another Arduino forum that someone wanted to use 3 EM’s to turn a small prop for a fish tank, but I don’t have the space really to turn a small prop and or I don’t want to mount anything in the bottom inside of the vessel. I was just hoping to have something small and discrete that the vessel will sit on.

Initial problems:
-Building a strong small electromagnet – I have lots of stranded copper wire; hundreds of feet between 6-12 gauge. I was thinking about stripping out the strands and using the strands in them, but what about insulation or shielding do I dip the EM in resin leaving the end of the core exposed.

-I think I want to use DC, how best to accomplish a small strong EM? My understanding of these magnets has to do with amount of current passed through and how may coils or wraps around the iron core.

  • I think an Arduino controller is the way to go, but I am not sure of which type or will it offer the best fit for my application. Also, thought about using the Arduino for control voltage on a bank of relays for higher current to get stronger magnets. I will only have about 1/8” Stainless steel between the EM and ball bearing.

-Can I cycle the electromagnet fast enough to effectively pass a ball bearing? And if so how to calculate how many I will need in relation to size of ball bearing and or path it needs to travel (spacing)?

-Am I limited to 11 electromagnets from digital output pins 2-13 of Arduino or can I write a program to create sub interfaces of some sort like in Networking(in hopes to add more EM’s)?

-Or can I wire two EM’s opposite of each other on the same output and fire both at the same time, having two ball bearings to create a more turbulent environment.
In the past I created a small stir plate out of a CPU fan and stir bar for a small flask and it works great, but I am limited here as I do not have a flat bottom or the center of the vessel available. I have a slight conically bottom shape vessel.
I have seen some examples and tried to assemble what I found, here is a basic wiring diagram for a small EM that Stefan Holodnick put together.

I somewhat understand what the Mosfet and the resistor are doing, but not 100% sure. I also want to change out the momentary push button for a push to make to kick off the program.
Also from the code I found on at http://arduino.cc/en/Tutorial/Blink
I was thinking my code would be something like this to loop and just allow the ball bearing to run in a circle. I know it takes time for the EM to energize and actually create a magnetic field, but I am not sure how long and guessed 1000 milliseconds would be enough.

int EM3 = 3;
int EM4 = 4;
int EM5 = 6;
int EM6 = 7;
int EM8 = 8;
int EM9 = 9;
int EM10 = 10;
int EM11 = 11;
int EM12 = 12;
int EM13 = 13;

void setup() {
pinMode(EM3, OUTPUT);
pinMode(EM4, OUTPUT);
pinMode(EM5, OUTPUT);
pinMode(EM6, OUTPUT);
pinMode(EM7, OUTPUT);
pinMode(EM8, OUTPUT);
pinMode(EM9, OUTPUT);
pinMode(EM10, OUTPUT);
pinMode(EM11, OUTPUT);
pinMode(EM12, OUTPUT);
pinMode(EM13, OUTPUT);
}
void loop() {
digitalWrite(EM3, HIGH);
delay(1000);
digitalWrite(EM3, LOW);
delay(10);
digitalWrite(EM4, HIGH);
delay(1000);
digitalWrite(EM4, LOW);
delay(10);
digitalWrite(EM5, HIGH);
delay(1000);
digitalWrite(EM5, LOW);
delay(10);
digitalWrite(EM6, HIGH);
delay(1000);
digitalWrite(EM6, LOW);
delay(10);
digitalWrite(EM7, HIGH);
delay(1000);
digitalWrite(EM7, LOW);
delay(10);
digitalWrite(EM8, HIGH);
delay(1000);
digitalWrite(EM8, LOW);
delay(10);
digitalWrite(EM9, HIGH);
delay(1000);
digitalWrite(EM9, LOW);
delay(10);
digitalWrite(EM10, HIGH);
delay(1000);
digitalWrite(EM10, LOW);
delay(10);
digitalWrite(EM11, HIGH);
delay(1000);
digitalWrite(EM11, LOW);
delay(10);
digitalWrite(EM12, HIGH);
delay(1000);
digitalWrite(EM12, LOW);
delay(10);
digitalWrite(EM13, HIGH);
delay(1000);
digitalWrite(EM13, LOW);
delay(10)
}

My thinking here is to give time for the circuitry to reenergize a new EM without possibly overloading something hence the delay(10). I don’t know if this is necessary or not. Anyway I know this is a lot, but I really like to tinker with this kind of stuff and if someone has any ideas please let me know. Much appreciated.

I think that magnetic stirrers usually use a magnetic object in the chamber which is attracted to a rotating magnetic field. Although you can use a non-magnetic object, that means you will only be able to attract it towards an electromagnet rather than align it with the field, which will reduce the torque you can generate. But, with that restriction you should be able to get it working reasonably easily by putting a circle of electromagnets on the base of the vessel vertically aligned, and just energize them in sequence. I imagine you will soon find out by trial and error how close the ball has to be for a magnet to be moved towards it, and this will determine the size and spacing for your coils. I think it would work quite well to energize multiple coils simultaneously, for example you could energize four at a time (spaced ninety degrees apart) so that three groups of coils could give you twelve coils in total. You would need to make sure that the shape of the chamber bottom was such that the ball(s) tended to settle close enough to the coils to be attracted towards them. The switching frequency needs to be governed by how fast the balls will be move, which will be defined by the strength of the magnets and the viscosity of the fluid. If you switch them very slowly the ball should just slowly follow the field round, but as you switch faster and faster you'll eventually reach the point where the ball can't keep up and falls out of sync with the magnets. You will probably find you need to set the speed very low at first so that the magnets can pick up the ball, then you can smoothly accelerate up to your sustained speed.

how best to accomplish a small strong EM? My understanding of these magnets has to do with
amount of current passed through and how may coils or wraps around the iron core.

You are correct. More turns (bigger) and more current (lower resistance or higher voltage) will give a stronger field. The problem with more current is that it means more heat and if you get too much heat the insulation will burn and the coil will short out causing sparks and flames.

Typically coils are wound with "magnet wire". This is wire insulated with a thin layer of enamel. The thin insulation allows more turns to be packed in the same space.

There is a trade-off between resistance and voltage. You reduce resistance by using thicker wire. If you use thinner wire you have to limit the voltage to keep from overheating.

You want to use electromagnet(s) to move something -through- a stainless steel vessel?

GoForSmoke:
You want to use electromagnet(s) to move something -through- a stainless steel vessel?

What's the problem?

When I was doing chemistry these Teflon coated stirring bars were much better than a direct motor driven. They might be better than a ball bearing. They pretty much stay fixed above the external bar magnet and you don't need a lot of logic to get them to spin and the other magnets stay on the outside. Since they are Teflon coated, they won't scale up or corrode in the boiling water. The bar magnets to drive the stirrers can also be purchased. I was always were stirring through glass, but you should be able to get enough penetration of thin stainless to pull it off.

Wow, I am surprise to have so many posts already, I have never posted anything here before. Ok, so since we have taken a turn on the EM’s. Theoretically how many can I power with the ardiuno?

I believe I saw the board is capable of 20 mA. What kind of Mosfet to use that will give me a beefier current?

Can EM’s pull things? as in will it even be able to pull a metal ball bear or “spherical stir bar” from one magnet to another? On my very small scale testing with a D cell battery I was able to pick up items, but it would not actually draw or pull the item to the core.

I really hope I am getting the terminology right here, please bear with me if I am not stating something correctly.

PeterH:

GoForSmoke:
You want to use electromagnet(s) to move something -through- a stainless steel vessel?

What's the problem?

Stainless steel is a conductor. The moving field will make lots of eddy current in the conductor and that is where the power will go. Better to use a glass vessel.

I believe I saw the board is capable of 20 mA.

That's a safe value per pin but from: http://arduino.cc/en/Main/ArduinoBoardUno
This is for the UNO, other boards are listed on the Hardware page.

DC Current per I/O Pin 40 mA
DC Current for 3.3V Pin 50 mA

Those are maximums, it is better to run 20 mA or less on any one pin at 5V, but total, you don't want more than 200 mA through that poor little chip.

Can EM’s pull things?

Yes, very. Look up Nikola Tesla and the induction egg, or just induction motors.

GoForSmoke:

PeterH:

GoForSmoke:
You want to use electromagnet(s) to move something -through- a stainless steel vessel?

What's the problem?

Stainless steel is a conductor. The moving field will make lots of eddy current in the conductor and that is where the power will go. Better to use a glass vessel.

So even though SS is not magnetic it won't pass a smooth EM field to the inside of the vessel? If this won't work then my whole idea is shot. I have to use a SS vessel that I can tig weld for the heat exchanger coil and hot water element.

It should pass through but I don't know how much will get through. You can try running AC through a coil outside a similar thickness stainless steel pot and use a Hall sensor to see what gets through. Also see if the pot does not get warmer.

Won't the convection currents set up by the 6kW heater produce enough stirring action for you. Alternatively have you thought about simplifying the problem by using a dip tube and "bubbling" air through the water. Yes, that doesn't answer your question but it might be a solution to your problem.

Stainless steel is not completely non-magnetic as there will be ferrite in the steel. So it might be worth confirming the properties of a particular steel.

You know the stirrers you get in labs. You drop a bar magnet into a beaker and it rotates and stirs the contents. Is the drive for the magnet not simply provided by having another bar magnet rotated by motor underneath the beaker? Would that not be a simpler way of getting the stirring action, and an easily variable rotational speed?

Is a glass or ceramic container not an option?

GoForSmoke:
Stainless steel is a conductor. The moving field will make lots of eddy current in the conductor and that is where the power will go. Better to use a glass vessel.

I'm envisaging the magnetic fields switching at a few Hz and I would have thought this would be low enough frequency for the inductive effects to be negligible.

jackrae:
Won't the convection currents set up by the 6kW heater produce enough stirring action for you. Alternatively have you thought about simplifying the problem by using a dip tube and "bubbling" air through the water. Yes, that doesn't answer your question but it might be a solution to your problem.

Yes the 6 kW element will agitate the water, but not enough from my research, I am running 50 ft of 1/2 tubing for heat exchange coil and need something a little more aggressive in agitating the water.

I have also thought about air bubbles as well, but I am leary that I may end up with super or overly oxygenated water. I intend to use this liquid at a later time and it may effect the end product if I have overly oxygenated water in the procress. I think I might be a little to picky on this one and may have to test it further though. Thanks for the idea!! And if you haven't guessed already I am home brewer and I am designing a new system with all the bells and whistles.

radman:
Stainless steel is not completely non-magnetic as there will be ferrite in the steel. So it might be worth confirming the properties of a particular steel.

You know the stirrers you get in labs. You drop a bar magnet into a beaker and it rotates and stirs the contents. Is the drive for the magnet not simply provided by having another bar magnet rotated by motor underneath the beaker? Would that not be a simpler way of getting the stirring action, and an easily variable rotational speed?

Is a glass or ceramic container not an option?

Yeah I have made one of these, I took a rare earth magnet out of a harddrive and epoxied it to a spare CPU fan. Added a potentiometer and mounted all of this is a small plastic junction box from home depot. Works great!! Also, I do not have the center of this vessel free and the bottom is not 100% flat.

I'm sure you will find that heating the water with your 6kW element will resolve any oxygenation-of-the-water problems. One of the best ways to de-oxygenate water is to heat it up. That's why many cafes cannot make decent tea - they use water that has been kept hot, rather than brewing from freshly boiled water.

cbhklh:

radman:
Stainless steel is not completely non-magnetic as there will be ferrite in the steel. So it might be worth confirming the properties of a particular steel.

You know the stirrers you get in labs. You drop a bar magnet into a beaker and it rotates and stirs the contents. Is the drive for the magnet not simply provided by having another bar magnet rotated by motor underneath the beaker? Would that not be a simpler way of getting the stirring action, and an easily variable rotational speed?

Is a glass or ceramic container not an option?

Yeah I have made one of these, I took a rare earth magnet out of a harddrive and epoxied it to a spare CPU fan. Added a potentiometer and mounted all of this is a small plastic junction box from home depot. Works great!! Also, I do not have the center of this vessel free and the bottom is not 100% flat.

This is why you should look up Tesla and induction motors. You don't need the fan or super magnet, which would not last in so much heat anyway.
If a picture is worth 1000 words, how much more is video?

The only moving parts are the switch and the objects rotated.

cbhklh:
I have also thought about air bubbles as well, but I am leary that I may end up with super or overly oxygenated water. I intend to use this liquid at a later time and it may effect the end product if I have overly oxygenated water in the procress. I think I might be a little to picky on this one and may have to test it further though.

Have you considered using pure N2 gas instead of air to provide bubble agitation? It might not be practical for your project, but I do know that it used in the food and beverage industry, sometimes as a replacement for CO2 in draft or fountain beverages and specifically to pressurize containers of beer.

This is why you should look up Tesla and induction motors. You don't need the fan or super magnet, which would not last in so much heat anyway.
If a picture is worth 1000 words, how much more is video?
http://www.youtube.com/watch?v=KomcIPrSJko
The only moving parts are the switch and the objects rotated.

Ok, so I think I am getting what you are saying finally. I lookup up the Tesla and his egg when you previously mentioned it, but I didn’t get what you were hinting at.
Basically if I have one induction motor the EM field will just spin the object, like a massive stir plate. I need to research this more, but wasn’t this Columbus egg run off a three phase motor?
Other than hand starting it off single phase is there another way of doing this? I don't have three phase as an option.

Far-seeker:

cbhklh:
I have also thought about air bubbles as well, but I am leary that I may end up with super or overly oxygenated water. I intend to use this liquid at a later time and it may effect the end product if I have overly oxygenated water in the procress. I think I might be a little to picky on this one and may have to test it further though.

Have you considered using pure N2 gas instead of air to provide bubble agitation? It might not be practical for your project, but I do know that it used in the food and beverage industry, sometimes as a replacement for CO2 in draft or fountain beverages and specifically to pressurize containers of beer.

Yes, I have considered inert gas of some sort, but this means another bottle I have to maintain and or worry about. Electricity is always there.