Monday, January 21, 2013

Rheoscopic coffee table

I've received a few questions about my coffee table, which I built in 2006. I used fairly standard woodworking techniques (floating tenons, breadboard ends) with maple, walnut, and bloodwood for contrasting colors. I didn't use any stains or colored finishes -- all of the color is from the wood's natural appearance.

The table has a 30" diameter disc that contains about 2 gallons of water mixed with Pearl Swirl fluid. This fluid is essentially very fine glitter that becomes suspended in the water, and will show flow currents as the water moves within the disk. I positioned the disc on a 24" diameter lazy susan bearing for low-friction rotation. When the disc suddenly changes rotational speed, there will be turbulence in the water, which causes the rheoscopic fluid to show fluid vortices and eddies that are part of the turbulence.

The outer metal ring was custom-built by a shop that specializes in metal rings (see sources below). All of the other parts were bought from McMaster, eBay, and Steve Spangler Science for the Pearl Swirl fluid.

I got the idea for this table after see a tornado visualization exhibit at the Carnegie Science Center in Pittsburgh. The Exloratorium has an orb filled with a similar fluid to show flow patterns similar to those on a rotating planet with an atmosphere.

The main pieces are: 3/8" thick glass top, aluminum ring with L cross-section, a 3/4" thick O-ring with square cross-section and a 1/2" thick PVC plastic bottom.  There are a bunch of screws that pull the aluminum ring down onto the glass, which pushes onto the O-ring, which pushes on the PVC bottom.  There is no glue used anywhere in the disk.  The overall diameter is about 32".

The disk holds about 2 gallons of liquid, and weighs a total of 80 lbs (just a good estimate -- I never measured it).  There are two threaded fill-holes with plugs on the bottom, and the whole thing sits on a 24" lazy susan bearing.

These are my sources:
Glass top
3/8" thick tempered   $85 shipped

Outer metal ring
Custom 2" x 2" x 3/16" thick, angle-in, "L" cross section, 32" outer dia
about $150 shipped

3/4", square cross-section, O-ring cord stock -- I made a scarf joint
to make this into a ring, about $40

Bottom plate
1/2" thick gray PVC plastic  $100

Get some food coloring too

Screws, cushioning rubber, stuff to wire-brush the aluminum ring:

lazy susan bearing
eBay -- search for "aluminum lazy susan"
12" will not work, you need to find a 24" dia bearing on eBay


  1. Thanks, I'll try to make mine now :)

  2. it is a knock of of a kalliroscope created and btw patented by Paul Matisse ( he is a genius artist also a grandson of Henry Matisse)

  3. Thank you for making this and for sharing.

  4. ¨very nice product and thank you for beaing so kind to share the details,.. I'll maybe try to make sth similar :) .. could you please post a picture of a scarf joint you used for the o-ring please?

  5. Anonymous, I don't have photos of the scarf joint in the table, but it is very similar to this:

  6. Hi,

    Huge fan of this - I am considering making something similar (minus the wood bit). Silly question, but does the whole assembly rotate (ie glass top too), or just the outer ring. It has been too long since I've played with one, and videos of it in motion are inconclusive. Thanks!

  7. Adam, the whole assembly rotates. It's only the inertia of the fluid that causes the flow patterns. This is neat because the table can be spun at constant speed, and the fluid will become stable; then when the disc is suddenly stopped, interesting turbulence patterns are generated.

  8. Hello Ben,

    Amazing table! I am interested in making something similar and was wondering if you'd be willing to answer a couple of questions regarding yours? I'm particularly interested in how you assembled the ring.

    From your post it sounds like the assembly is a sandwich of glass, o-ring and pvc sitting inside the L of the aluminum ring. To hold it together and compress the o-ring are you using screws into the pvc, or bolts with gaskets passing through the pvc? Did you consider using a thinner pvc backed with plywood for the bottom?

    For the scarf joint, did you use something to adhere the joint together, or was just the pressure of compression enough to keep it sealed?

    As far as filling the assembly, you drilled and tapped two holes in the pvc, correct?

    Also, now that you've had the table for over a year, any issues you've seen or design changes you'd make?

    Thanks so much for posting your design--one of the best table designs I've seen in a long time!

    1. Keith, yes the ring construction is done exactly as you have said. I hadn't considered a wood-backed thin sheet of PVC. That sounds like a good improvement. The 1/2" think PVC bows much more than I expected.

      I used adhesive in the scarf joint. It was probably E6000, but I don't remember.

      Yes, two fill holes drilled and tapped, with one hole right at the edge of the O-ring to let air escape.

      I haven't had any big problems with the table. The ring bearing is very noisy. If it spun more quietly, it would be more inviting for people to spin more often.

      Thank you!

    2. Thanks for your reply! I'll try to let you know how mine turns out.

  9. Dear Ben,
    Thank you for sharing your knowledge on the construction of your amazing table with us.
    I am considering to build one on my own, but would work with plexiglass instead of glass, because I think it is easier for me. I am not very talented in those thing...
    What do you think? Do you see any huge disadvantages?
    Thanks for your help,