I'm trying to design a digital clock that will use several electromagnets and iron filings to show the time. Each digit should have 7 electromagnets that turned on and off to attract the metal shavings through white plexiglass (I'm also considering using immersion oil to make it easier for the filings to reform). However, I'm stuck on how to design the electromagnet itself. I initially started with a solenoid, however, I quickly learned that the shavings will mainly attract two ends, leaving the middle of the digit line with no definition.

I'm looking for advice on how this might be done. Will wrapping the wire vertically instead of horizontally be a better choice? Should I use a different shape altogether?

Thank you,


rotate your electromagnet by 90°, so that its axis is perpendicular to the plexiglass. That's kind of obvious – it puts only one of the filings-attrackting ends close to the glass.

If you wind your magnet around a rectangular-crosssection piece of iron, then you can make "straight lines". Notice that even these lines will be feathered – the filings will align along the magnetic field lines, and those form a convex shape from one end to the other

  • 3
    \$\begingroup\$ In an oil-filled, back-lit scenario, I think most of the filings in this scenario would line up in some degree of orientation that would minimize their effect on the light (something I thought about after typing my answer)...might actually work better in reverse. Use shielded cells filled with oil-suspended filings, magnet off=little light passage; magnet on=more light passage. ... Or add opposing poles at either end for "black," and the homopole for "white." \$\endgroup\$ – Robherc KV5ROB Apr 20 '17 at 22:38
  • \$\begingroup\$ uuuh I like that! \$\endgroup\$ – Marcus Müller Apr 20 '17 at 22:43
  • 1
    \$\begingroup\$ Almost a ferro-magnetic LCD pixel then...lol \$\endgroup\$ – Robherc KV5ROB Apr 20 '17 at 23:09
  • \$\begingroup\$ Very cool suggestion. I think that's probably the way to go. I'm having a hard time finding a 2"x2-4"x1/4" pieces of soft iron. Would anyone know where I may find this? \$\endgroup\$ – oneself Apr 21 '17 at 18:26

I see two workable options for accomplishing something close to what you're describing:

  1. Build the electromagnets with a flattened (as opposed to round) core, then place one pole against the underside of the plexiglass. This would attract filings fairly uniformly to that "segment."

  2. Build the electromagnets as small "horseshoe magnets" with both poles touching the underside of the plexiglass (and the windings around the middle). This would form poles at either end, and gaussian lines connecting them, which may prove to give you better "resolution" in getting your iron filings to line up in a recognizable digit.

CAVEAT: Both of these designs should work well for a 1-segment "test case," but there will be more problems to overcome when you start combining segments. I think option #1 may have less "intersegment interference," so long as all poles are equal and synchronised in-phase with each other. However, in practice there are a lot of ways for the electromagnets to interfere with each other when in close enough proximity to form a 7-seg. display. ... In the end, you'll have to do a lot of testing & adjusting, and may end up having to use magnetically shielded segments of some form.

  • \$\begingroup\$ I think #1 is probably the way to go. I'm having a hard time finding a 2"x2-4"x1/4" pieces of soft iron. Would anyone know where I may find this? \$\endgroup\$ – oneself Apr 21 '17 at 18:26
  • \$\begingroup\$ call pretty much any steel supply yard & ask for 2" flat bar-stock in 1/4" thickness. Structural steel is intentionally ver low-carbon steel & should work fairly well (use AC electromagnets to reduce "permanent magnetizing" of your cores. Also, get a saw+wafer wheel to cut it with (a cheap "angle grinder" works quite well); they won't accurately cut small pieces without charging a high processing cost. \$\endgroup\$ – Robherc KV5ROB Apr 22 '17 at 1:39
  • \$\begingroup\$ @oneself alternatively, a few pieces of 1/8"-1/4" wire/"rebar" lined up next to each other & epoxied would also work well, and can be easily cut to size with bolt cutters instead of the grinder (plus they'll have somewhat lower eddy-current losses." \$\endgroup\$ – Robherc KV5ROB Apr 22 '17 at 1:41

Filings are hard to manage, once you get them attracted to a pole piece they are actually hard to get off due to residual magnetic retention effects. You may have to use a high frequency AC signal to demagnetize them.

You might be better off using a Ferro Fluid. This is dense, but with very low magnetic retention.

Of course like all things novel.....it's already been done quite effectively. The researchers even wrote a paper showing the methodology and a good picture of the electromagnets. I've always fancied making one of these since they showed the prototype...but never got around to it. It looks like it would be fun manipulating the blobs of Ferro fluid to build characters.

Show pictures when you are done!!

  • \$\begingroup\$ Ferrofluid is a little bit expensive for any significant quantity. I was hoping to make something more cheaply. Although, it's very cool to see how they arranged their electromagnets. \$\endgroup\$ – oneself Apr 21 '17 at 18:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.