Show HN: I spent a year making an electro-mechanical prototype of a liquid clock
hellorhei.com
hellorhei.com
Super mesmerizing to watch the ferro fluid making its way through the display.
Edit: I remember reading that the ferro fluid used in Ferrolic would begin to lose its magnetic properties a few months in -
Ferrofluids are fundamentally unstable. They're colloidal suspensions that rely on surfactants to keep the particles apart, and as the surfactants degrade, the particles start clumping together and falling out of suspension. - jpatokal [1]
Is the Rhei somehow designed to last longer? If so I would be really interested in purchasing one if you ever decide to get them into production.
Btw: Comes off a labor of love and insanely-awesome attention to design and detail. Outstanding. Kickstarter/indiegogo/lockitron selfstarter page yet? :)
KS, maybe one day :)
If the cost of making another example (parts, time, skill) can be made sufficiently below (say 30-50%) of a reasonable sale price, it may be business. The real "hacking" done by contract manufacturers is cutting corners to make something beautiful, functional and inexpensive to manufacture.
Challenge, should you choose to accept it, if not already:
Try to assemble a similar example with mostly scrap, for the least budget imaginable. (Penny wise & pound wise.)
Another possible path:
Patenting and licensing to boutique clock makers / electronics firms, which frees you from nearly all the headaches of dealing with advertising, manufacturing and distribution (after negotiations, assuming one would usually seek advice/assistance of trustworthy, relevant partners). Also, people may not like patents, but it's a double-edge sword necessary to create a survivable business... without that, the newbie wantrepreneur is often shoot themselves, in not just one foot, but reloads to see if that stops the fail. It's nice to broadly claim "patents are evil," but good luck making money sustainably in an environment where anything & everything is knocked off, without recourse (usually legal) of any measurable kind.
But not if you are constantly stirring them up. The Ferrorfluid used in loudspeakers for instance doesn't seem to degrade.
The slogan is geat too, I would just drop the "literally" at the end, I think it's unneeded. "Time flows" is perfect IMHO.
Can you explain how did the idea grows in your mind?
--- yeah its funny, the first time i tried to make this was long time ago, here is the video 3 years old: https://vimeo.com/42313987
The idea was pretty spontaneous. I saw CZferro toys online and wanted to play with it so much. I ordered one and played with it and after a while i was thinking why couldnt this be something concrete. I loved the idea of translating something thats abstract in a visual sense to a concrete thing and back.
After that a clock came as a logical object, since it changes non stop.
Thank you for the explanation. ;)
What was an unexpected problem you had? And how do you solve it?
There's a bunch of mechanical stuff going on. How much training did you have in that before you started?
What one thing will you remember for your next project?
Thanks a lot!
Wow... were to start.. Perhaps the biggest problem i had was preventing the ferrofluid to stain glass and aluminum frame i used for the screen. I tried to get some help on the faculty of chemistry here in Belgrade, but no real success. Finally after months and months of experimenting with different liquids, you start to feel and "undertand" the way ferrofluid behaves.. very strange :) Its still not perfect but im working on perfecting it.
The first time i sketched the project seemed very easy, i will make a matrix of electromagnets and simply attract the ferrofluid. It actually gave some results but i realized that having a system that requires a huge amount of electricity to be spent all the time to simply show a number thats fixed - is not elegant. Here is one of the videos i recorded: https://www.youtube.com/watch?v=Kvp7bWe-M94
So finally i decided to go with neodymium magnets and mechanics. Building this from scratch was a crazy adventure of lots of lathe machining and cnc machining and so on.
I had no training basically i had to learn all along the way. I used illustrator for the most of the construction, since i am UI/UX designer, and finally gave it up and learned solidworks so i can make a whole 3d model of the clock and assemble all parts in 3D.
For the next project i think i will have to curb the need to make it "perfect", i lost A LOT of time beacuse of this. For example each time i wanted to make a screen for this i had to buy a glass, cut it on waterjet, then drill a hole, then glue to the frame that needed to be cut on waterjet also and polished before adding adhesive and so on. And once it does not work, i have to do it all again...
This is amazing. Very inspirational.
e.g.: http://www.eclipsemagnetics.com/media/wysiwyg/datasheets/lif...
I know very little about them, other than that they're cool. The ones in that datasheet start about at about 2kg, so I don't know if you can get them small enough for your purposes.
Pretty awesome name for a liquid clock.
Was this funded by a grant, or is it a personal investment in hopes of selling completed versions of the clock?
Shooting took around 7 days because we were on a very tight schedule, trying to make it for the public announcement of the red dot results http://www.red-dot.sg/en/online-exhibition/rhei/
The good news is that i do have all the process over the year documented, if you want i can upload some photos but i am writing a making of presentation that i will place on behance and medium. I think it will be a really useful story for an industrial design student or someone who has good ideas but still not sure if he should start prototyping or not.
This was funded solely by me, which made it so much harder. I had to do freelance by day and work on this by night. Its difficult especially because prototyping means you have to make the same thing many times over and over again and its really time and money consuming.
At the moment its working but its not perfect. The biggest challenge i had was to make this really look like fluid and kill the ferrofluid spikes. I simply wanted to animate the fluid not make it obvious that its a ferrofluid. So i had to thin the ferrofluid with all kinds of different things, some work better some work worse, still testing.
What is the 'thickness'? I'm guessing only several millimeters??
https://www.dropbox.com/s/066c8502j2zwlwo/rheiani.gif?dl=0
Thats the reason why i had to work with ferrofluid and change its structure and composition, i wanted to have thicker numbers. Whole screen is 10mm thick inside, and numbers sometimes go up to 8mm of thickness. They are especially nice from the angle because the seem like they float and not simply stuck to the wall. I will make better photos definitely.
Really adds to the organic look and feel.
Well done!
However seconds would be doable but it would result in animation being constant since it needs about a second to transform, so i finally decided to go with HH:MM
Here is another video of this little thingy: https://www.youtube.com/watch?v=SoOMQ8TaTyE
you can see that the liquid is quite different, it was long time ago, but it basically acts the same but looks more blobby
https://www.youtube.com/watch?v=-drmRYsvUaM
You can really appreciate the difference between the gentle movement induced with mechanically actuated magnets, and the nasty instantaneous movement which would happen with electro.
One question: what sort of power consumption is required to maintain the digit positions inside of the enclosure?
On the other hand, you are right, it only draws power when it moves and it moves every other segment every few seconds which is enough to constantly animate the numbers, because the fluid needs time to readjust.
The motors are actually stepper motors, which made it all harder, because its not as easy to move them as it is to move servos, and steppers have no idea where they are which required an initial calibration sequence. The good news is that steppers are not loosing any steps, i guess the mechanics work very well.
https://www.dropbox.com/s/cjvngsya64m5bb7/rhei-prototype-num...
Damjan asked me to post a comment here, he says he is sorry but HN is not allowing him to respond to all coments quick enough. But stay assured he will anwer each and every question. :)
Cheers.
Is all the equipment yours or did you work in a Makerspace or something?
No mention of a particular Makerspace, but obviously a lot of collaboration with people with various tools/skills.
Because of the video, I can literally see through the box into the 7 segment magnet chunks moving with the servos when the time ticks over. I like the simplicity & the way the tick flows into shapes.
@damjanstankovic: can you throw up a Cinegraph/GIFV below the fold or something, so that the flow is clearly visible before we even head into the video itself?
Like a 1,2,3...9 sequence for the last digit, because the beauty of that should hit the next person who clicks this.
GIF on the way :) thanks a lot!