It’s more of a half-fail than a full fail, but [Basti] is accustomed to getting things right (eventually) so it sticks in his craw that he wasn’t able to fully realize his ferrofluid dreams (German, translated here). Anyway, fail or demi-fail, the project is certainly a lesson in the reality of ferrofluid.
We’ve all seen amazing things done with ferrofluid and magnets. How hard can it be to make an interactive ferrofluid wedding present for his sister? Where ferrofluid spikes climb up a beautifully cut steel heart in a jar? (Answer: very hard.)
First off, ferrofluid is expensive, and estimating the right amount for the project is tricky. [Basti] got 20 ml. Next, it stains everything in the way that only suspended nano-particles of rust can. Simulations, like every other tool, are useful but in certain strange applications (like moving ferrofluid around in a bottle) are only guidelines. The electromagnet ended up not producing enough field and the driver circuit got pretty warm. But he had first results.
Ordering a liter of ferrofluid solved the quantity problem, and significantly lightened his wallet. Rather than drive the magnet’s MOSFET in analog mode or get into high-frequency switching, he decided to just turn the field on and off. Finally, the ferrofluid would just stick to everything on the inside of the glass jar, muddying the view. The solution ended up being a little bit of detergent in the water, which helps the ferrofluid slide off the glass.
The end result is short of [Basti]’s dreams, but not too shabby either, so maybe it’s only a fail from a price-performance standpoint. Check out the video below and decide for yourself.
Anyway, the whole spirit of Fail of the Week is to help other folks by pointing out one’s own pratfalls. So thanks for that, [Basti]! And good luck selling off the rest of your ferrofluid.
We should probably add this to our list of hacker folk wisdom:
- All projects involving Peltier devices ultimately end in disappointment.
- Ferrofluid is trickier than it looks on YouTube.