Manipulating Ferrofluid with Electromagnets

Thread Starter

danielj

Joined Apr 13, 2017
6
Ever since I saw this Ferrofluid clock I got fixated on building something with Ferrofluid. I'm somewhat of a beginner with electronics but this seems within reach.

So my idea was to build a grid of electromagnets that are controlled with a microcontroller to create shapes with the fluid. Let's say I have a bunch of electromagnets arranged in a grid. How do I connect them to an Arduino in a way that I can control with software whether each one is on or off? How do I overcome the fact that an Arduino only has a limited number of pins when I want to the grid to be at least 8 by 8?

Does the max current the Arduino allow suffice to drive this many magnets and strongly enough? (It seems the fluid requires fairly strong magnets).

My next question is what would be the best way to arrange the grid to achieve a strong, focused magnetic field to control the fluid. Will the magnets interfere with each other if they're close enough, to the point where they're not focused to give any control to manipulate the fluid?

If anyone has a better idea on how to go about this please share. It seems like the clock in the clip is using permanent magnets but I couldn't quite figure out how the whole mechanism works. I'm not expecting anyone to design this upfront for me, but a few initial ideas to help me get started and experimenting will be highly appreciated.
 

BR-549

Joined Sep 22, 2013
4,928
Let's say that you had a fascination with the main bearings on your auto engine crankshaft.

Do you think that you could change those bearings out.........just by asking a few questions?

What do you do for a living? Could a person off the street...........just by asking you a few questions.....do your job?

It's like asking how tho plot a course to Jupiter........to where Jupiter will be ten years from now.

It's a top level question. You will need previous knowledge........to understand the answers to your questions.

A lot of previous knowledge.
 

Thread Starter

danielj

Joined Apr 13, 2017
6
Let's say that you had a fascination with the main bearings on your auto engine crankshaft.

Do you think that you could change those bearings out.........just by asking a few questions?

What do you do for a living? Could a person off the street...........just by asking you a few questions.....do your job?

It's like asking how tho plot a course to Jupiter........to where Jupiter will be ten years from now.

It's a top level question. You will need previous knowledge........to understand the answers to your questions.

A lot of previous knowledge.
Who says I don't have the time and will to do so? I'm pretty proficient in programming and have built a few simple circuits with an Arduino before.

The guy that made that clock had no experience in electronics or mechanics, and look where he end up.
 

Thread Starter

danielj

Joined Apr 13, 2017
6
Let's say that you had a fascination with the main bearings on your auto engine crankshaft.

Do you think that you could change those bearings out.........just by asking a few questions?

What do you do for a living? Could a person off the street...........just by asking you a few questions.....do your job?

It's like asking how tho plot a course to Jupiter........to where Jupiter will be ten years from now.

It's a top level question. You will need previous knowledge........to understand the answers to your questions.

A lot of previous knowledge.
Who says I don't have the time and will to do so? I'm pretty proficient in programming and have built a few simple circuits with an Arduino before.

The guy that made that clock had no experience in electronics or mechanics, and look where he end up.
 

BR-549

Joined Sep 22, 2013
4,928
I didn't say that you didn't have the time or the willingness. I just said that you needed knowledge. And you don't have it.....or you wouldn't be here.

First you need to study basic dc circuits. Then ac circuits. And then microprocessors. You might be a programmer.........but you don't know anything about micro-controllers. Your questions show that.

I wasn't insulting you.......was just trying to explain the enormity of your question. It's a project question. Not a simple circuit question.

Anyone can takes off the shelf components.......follow directions and put it together.

But you don't want that. You want a custom project. To do that..........you have to know how it works.

And to understand and know how it works...takes a lot of study.

The thing that's really neat about ferrofluids, isn't that it can make cartoonist displays.

The really neat thing about ferrofluids............is that it shows the rotation of current.

And if you ever studied ac circuits.................you would know that this is much more important.

The rotation of current is neither recognized nor taught.

But I am a lowly crackpot..........and will leave you alone.

I wish you luck on your project.
 

Sitara

Joined May 2, 2014
57
Hello danielj,

I could not link to the hyperlink which you provided in post #1, I gave up waiting. So I Googled for Ferrofluid clock and got a link to the following video:
. This seems along the lines of what you're after. At the end of the video, a number of "references" or "sources of information" are given. They should be able to answer your questions. The video immediately after this one, concerns a Candle Lamp. That one was (for me), even more fascinating than the ferrofluid clock.

I strongly suspect that in pursuing the above-mentioned "information sources" you will realise that BR-549 spoke the truth in posts #2 & #5. I too am pretty good in programming, but electronics is a whole different ball-game. It takes not just years of study but TONS of mistakes: blown up transistors, burnt resistors, circuits on which you spent weeks to design & construct, which just sit there doing nothing when you first power up - all that effort, expense, sweat and time - down the drain. Only after these painful lessons and errors does one's proficiency rise. Electronics is as much a practical, workman-like skill like woodwork, as it is a theoretical discipline.

This ferrofluid clock project is not some easypeasy walk in the park, which will take a couple of weeks or even a couple of months. Its a hard project, and quite frankly, in my personal view, a waste of time, effort and resources. Much better clocks can be constructed with standard displays and techniques, in much shorter time & with much less effort.

But more relevant to your question is the following: taking babysteps in the shallows when you first start off in any new endeavour, is much easier, more likely to succeed and more certain to build self-confidence than to straightaway dive into the deep end.

Of course, you are free to decide for yourself. But if you fail, never be disheartened, because it IS a tough project. The next time, choose a babystep.
 

djsfantasi

Joined Apr 11, 2010
9,156
First, I agree with the previous two posters. This is a complicated project. And even though you have several questions, you have not provided enough data.

You ask about the current capsbility. Yet you don't provide the current drawn by the electromagnets.

Do the electromagnets need to be always on, or can they be multiplexed.

Just a couple of examples. I am sure there are more.
 

DNA Robotics

Joined Jun 13, 2014
647
I don't see that project as being so difficult. These 7 segment driver chips and logic level MOSFETs should operate any size magnets you want to make. They also have them for multiple digits and I2C so you only use a few pins.

CD4511BE or SN74LS47N IC 7-SEG LED DECOD/DRVR 16-DIP

Don't let these guys discourage you.
 

djsfantasi

Joined Apr 11, 2010
9,156
I don't see that project as being so difficult. These 7 segment driver chips and logic level MOSFETs should operate any size magnets you want to make. They also have them for multiple digits and I2C so you only use a few pins.

CD4511BE or SN74LS47N IC 7-SEG LED DECOD/DRVR 16-DIP

Don't let these guys discourage you.
No, but familiarity with decoding, latching and driving some sort of driver IS necessary.

There are multiple small problems to be solved. Calculate or document the digital IO pins, so you know how much current is recommended. This will allow you to specify the driver components.

The clock circuitry has many circuits available. So build and test your driver first., such that a logic circuit can operate it.

Then, add on a latch. The clock input can be inverted to get the proper signal for latching the display.

You never answered if the ferro magnetic. Electromagnetic could be multiplexed?
 

DNA Robotics

Joined Jun 13, 2014
647
Arduino is open source. How many hits would you get from a search for 7 segment clock sketches? From there it is just research & wiring. He may have to read a book or two. The fun is in learning & creativity & accomplishment.
Ferro-magnetic material is physical & viscous. I think it would stay where you put it while being strobed.
He could start by just getting a 7 segment display working. Then ask more questions here if necessary to get the MOSFETs & magnets working.
 
Last edited:

Thread Starter

danielj

Joined Apr 13, 2017
6
I don't see that project as being so difficult. These 7 segment driver chips and logic level MOSFETs should operate any size magnets you want to make. They also have them for multiple digits and I2C so you only use a few pins.

CD4511BE or SN74LS47N IC 7-SEG LED DECOD/DRVR 16-DIP

Don't let these guys discourage you.
Thanks, I ordered a few of these since they're pretty cheap.

I found an essay the guy who made the clock wrote. It has these pictures:




So it looks like there's a glass display where all the fluid rests and on the back an array of electromagnets close to each other. It honestly doesn't look that complicated to someone who knows what they're doing. :)

Here's the pdf http://isea2015.org/proceeding/submissions/ISEA2015_submission_246.pdf
 

Alec_t

Joined Sep 17, 2013
14,280
How do I overcome the fact that an Arduino only has a limited number of pins when I want to the grid to be at least 8 by 8?
If only one magnet at a time needs to be energised you could address an individual one using 8 row address lines and 8 column address lines (16 pins in all). How many I/O pins do you have available?
If more than one at a time will be energised then some latching arrangement will also be needed.
 

Thread Starter

danielj

Joined Apr 13, 2017
6
If only one magnet at a time needs to be energised you could address an individual one using 8 row address lines and 8 column address lines (16 pins in all). How many I/O pins do you have available?
If more than one at a time will be energised then some latching arrangement will also be needed.
I'll need to be able to energize all of them at the same time. Also to control the amount of current of each one.
 

DNA Robotics

Joined Jun 13, 2014
647
I'll bet you could hack an LED alarm clock to drive the MOSFETs if you could find one with access to the segment circuits.

I think with the right resistor / capacitor combination on the gate, a MOSFET would stay on & not even know it is being strobed.
 

Thread Starter

danielj

Joined Apr 13, 2017
6
Here's a clip of another one of these clocks. This one appears to be using stepper motors that drive permanent magnets. Seems a lot easier in terms of electronics:


Has the advantage that it doesn't consume any power while a shape is on display too.
 
Top