Introduction: Ice Cube Clock ^3

The fluorescent glow tube nestled inside this polyester resin casting is a gas display tube from the Cold War era, made by our then-enemy, Russia. It has this cool retro look with its Nixie-ish blue glow.

My project is the result of my buying a clock kit from that utilizes the tube. I don't know how the Adafruit folks happened upon this tube, but clock kit is a great idea, me thinks. Then they housed their kit in a custom acrylic case and called it the "Ice Cube Clock", and you can see the fine product here:

Adafruit Ice Cube Clock

I gleefully assembled the circuit board from the kit, and with the board proudly built, realized I had lost the acrylic parts for the case. Probably absent-mindedly tossed them in the trash.


In my state of exasperation I brainstormed a hack of the Ice Cube Clock -- I would build on the "ice cube" concept by encasing the project in clear resin. I later refined this idea to encase only the tube so the circuit board would still be accessible.

My aim was to make the tube appear it was floating in the casting, so I have it connected with fine wires, with no other visible means of support.

And I'm calling my hack:

  • Ice Cube Clock ^3

(^3 being the mathematical nomenclature for "cubed")


  • Ice Cube Clock Cubed


  • ICC^3

My ICC^3 was for sure a learn-as-you-go project, particularly the dratted casting, which I've never played with before. I welcome suggestions for this or any aspect of the project for the benefit of others who might want to take on their own ICC^3.

Step 1: Pitfalls

Above, my three ICC^3 paper weights, one temporarily set up operating.

As I developed the project I ended up with three rejects -- three tubes encased in casting that were not what I wanted to achieve (screw-ups).

Luckily, I was able to get new tubes without having to fork out for the price of the entire kit, as the tube itself is available on eBay for about 10 bucks a pop.

But now I have these three odd-looking paper weights. Hopefully this I'ble will help others who want to do a ICC^3 avoid the pitfalls, where I tumbled headlong into the pit!


First of all we need parts, glorious parts:



Available from Clock Kit




One of these comes with the clock kit, but you will need two.

Something I just noticed, after all this time. The Adafruit project is not called the Ice Cube Clock. It is called the Ice TUBE Clock. Weird how I missed that all this time. Irregardless, my project will remain with its original name.


Environmental Technology Polyester Casting Resin, Clear, 32-Ounce.

This project requires only 16 ounces, however unless you are experienced with casting, you might consider to invest the bit more in this double dose size for in case you have to do a re-cast.

Or you can use left-over resin for another project, like casting your kid's baby teeth.

This is hyped as a clear resin, however I found that after casted it does have a very slight grayish tint.

ELMER'S GLUE, or the wood glue of your choice.


Craftech Industries did lab tests on adhesives for plastic and concluded that "JB Weld PlasticWeld" was the best overall adhesive for plastics. So I used this stuff.


Available at All Box

Don't lose the little rubber screw caps that come with this. They make nice feet to keep the box from scratching the surface it rests on.


Check with your local friendly hardware store

MOLD FOR CASTING, 5.7" x 2.3" x 3.1" i.d.

Available from Mold

This was the only mold I found that gave me the form factor I wanted for the cube. Experienced casters might be able to come up with a better one, or a way to make a custom mold.

This mold is designed for making soap, candles and cake; it is not intended for resin casting. You could bake some killer homemade Twinkies in this mold! Try the Organic Vegan Twinkie Recipe in this site.

The big problem with using this mold for resin is that it is made of silicone rubber and its sides are quite pliable. The weight of the liquid resin would cause the sides to bulge out.

Not a problem for your homemade Twinkies. But for my clear resin casting I wanted the sides to be flat to avoid optical distortion. Later on I show how to concoct a support for the mold to keep the sides flat.

But even with the mold support the sides did not come out as flat as I would have liked. A mold with rigid sides would probably make casting with flatter sides and not require as much sanding on the belt sander.

Aside from that, the mold works nicely with its silicon rubber walls, smooth as a baby's butt -- no mold release is needed. The casting just comes out with a little coaxing with hardly any sticking to the mold.









eBay sellers have sandpaper assortments at a decent price.


I used "Meguiar's Gold Glass Carnauba Plus Premium Paste Wax"



This is the style of thin loose-fitting gloves used by food handlers in restaurants.

Available cheap at Daiso in a box of 100.

JEWELERS ROUGE POLISHING COMPOUND. For use with the buffing wheel. Rouge at Amazon

Step 3: TOOLS

Soldering station

Laser cutter; or jig saw and file

Wire cutters

Needle nose pliers

Jars and sticks for mixing resin

Wood working tools

Orbital sander

Belt sander

Sanding belt cleaner

Sanding Pad: This is similar to a sanding block, but has a flat surface instead of curved.

Electric Buffer. The Ideal tool would be an electric buffing tool. Short of that would be a bench grinder with buffing wheel installed. And short of that (my method), an electric drill with buffing wheel attached.

Two Buffing Wheels for use with your method for buffing.

Step 4: Summary of Project

  • Build Ice Cube Clock Kit and test
  • Build mold support
  • Prep tube wires
  • Make wire holding fixture
  • Make standoff jig
  • Pour the resin
  • Finish surfaces
  • Prepare base
  • Final assembly
  • Voila!

Step 5: Build the Ice Cube Clock Kit

One departure from the Adafruit procedure is to attach the tube wires to the header at the very ends of the wires, away from the tube -- not up against the tube as shown in the Adafruit instructions. The reason is that the connection for now is temporary -- just for testing.

We will be removing the tube from the header, and we want to keep the wires long for suspending the tube in the casting.

When you first power up your clock, if all goes right the tube will glow -- but just barely. This is because the brightness is turned down by default. See the Adafruit instructions for increasing the brightness. On all the tubes I turned on, the right-most digit was slightly dimmer than the others.

Under the topic "Board Assembly" the Adafruit instructions state that the tube comes in 19-pin and 21-pin versions. All the tubes I worked with all were the 21-pin version, so I suspect these are the more common.

The Adafruit site goes on to state that on the 21-pin version, 3 of the pins are unused, and may be cut off. For the ICC^3 project it is mucho advisable to do just that. So, yes, please, do hack off the three wires accordion to the instructions in the Adafruit site.


Fully test the operation of the clock and all the controls, then take time to admire your glowing tube in all its blue serenity before you pull the power.

Step 6: Scrap the Tube Header

    The tube header is difficult to remove and re-use, and that's why I included on the parts list an extra header for use later on with the encased tube. So at this point snip off the tube wires at the header, then toss the header. Or...

    OR -- OR --

    If you have the patience, skill and desire you can unsolder the header for use later in the project. This will save you having to buy an extra header per the parts list.

    If you do this, FIRST snip the wires at the header to remove the tube. If you try to remove the wires before snipping there will be undesirable deformation and stress of the wires, not to mention general chaos.

    Once the wires are snipped, you can remove them from the header and use your fave solder removal method to clear the holes of solder.

    Step 7: Build the Mold Support Fixture

    • Slice a generous slab from the 3/4" plywood for the base of the fixture, at least 8-1/2" x 6-1/2".
    • Then cut small pieces to fit against the 4 sides of the mold: all 2-1/2" high, and...
      • Two 5-1/2" long.
      • Two 2-1/4" long.
    • Situate the mold in the middle of the base, then slop some Elmer's glue on one edge of each of the side pieces and fit them up SNUG against the mold, with the glue side down on the base.
    • Set a weight on top to provide clamping and leave the glue its spell to cure.

    Step 8: Prepping the Wires: Lengthen Wires

    • The tube's wires aren't long enough to support the tube from outside the mold, so we must lengthen them, thusly:
    • Cut nineteen 2" pieces of the 24 AWG bare wire.
    • Solder these wires to the ends of each of the wires.

    Notes for the wire forming operation in the following Steps:

    • If you are a male-type person you might want to enlist the help of a helpful female friend in doing this part for you, which requires neatness, finesse and good dexterity with small parts. This is a very delicate process and the gals are generally better than us guys at working at this miniature precise scale.
    • The other thing to mention here is nomenclature regarding orientation of the tube. When you set the tube in the mold, its wires will be protruding upward so they will project out of the casting. But in the final product these wires will be protruding downward into the base.
    • So ►FOR NOW◄ when I use the term "up", I mean it as when we are working with the wires pointing up as in the mold, in which case the tube itself is oriented upside-down.
    • Read this over till you are no longer confused.

    Step 9: Prepping the Wires: 90º Upward Bend

    • Use needle nose pliers.
    • Form the wires into two rows, one row is the 9 wires from the top circumference of the tube and the other row is the 10 wires from the bottom.
    • The 3 snipped wires will be one of the dividing points between the two rows.
    • The wires from the top will project back out from the tube base 1/8" before bending up.
    • The wires from the bottom will project out a little farther, 1/4" before bending up.
    • Avoid bending the wires right at the base of the tube to avoid stressing them and breaking one off right at the tube. And in general avoid any excessive bending or handling of the wires. They are not particularly fragile, but if one breaks at the tube, you're left with this odd phallic paperweight.

    Q. What is WRONG with the picture above?
    A. If you were to view the tube from its front side you would see that it is right-side-up. With the leads pointed up as in the picture, the tube should be inverted. I cast this particular tube this way before I realized my mistake, thus creating one of my paper weights. Had I gone on and made my ICC^3 clock with this casting I would stand on my head to read the time.

    Step 10: Prepping the Wires: Bend Top Wires Inward

    • Add another bend to the wires from the
      top circumference of the tube, so that they lean toward the opposite end of the tube.
    • Form a gentle bend in each of the wires around the midpoint. This will give the wires some flexibility to enable fine adjusting of the tube orientation later.
    • Spread the wires out. They should be spread out so that they will take up almost all the space between the sides of the mold, but not contact the mold.
    • Also the wires cannot extend too far back away from the tube, to avoid touching the back side of the mold.Make sure no two wires are shorted and that no wires are crossed.

    • Make sure your tube is properly oriented during this operation so that the wires get properly formed.

    Step 11: Make the Wire Holding Fixture

    This fixture will secure the tube’s wires and hold the tube in suspension during the casting process.

    • Hack off a 3" x 1" piece of the plated perf board.
    • In hacking, cut the board right through the lines of holes, slicing the holes in half.
    • Apply a thick slobbery coating of solder along the two long sides of the fixture.The tube wires will be peacefully nestled in the halved circles.

    Step 12: Connect the Tube to the Wire Holding Fixture

    In order to properly attach the wires to the Holding Fixture we have to rest the tube in the mold in its proper orientation. This requires temporarily placing a wood block at the bottom of the mold to support the tube in its desired vertical position.

    A 1-1/2" x 6" chunk of board works nicely. About board thickness:

    Sixteen ounces of resin will produce a casting of approximately 2" depth in this mold, so the center of the tube needs to be about 1" from the bottom of the mold. So the block of wood needs the proper thickness to properly position the tube. A block of 3/4" plywood will position the center of the tube slightly below the vertical center of the completed cube.

    If you want the tube to be closer to the exact center, go out and find some 5/8" thick board.

    • Put the mold in its support fixture.
    • Set the tube on the board in the orientation you want at completion.
    • Tape the Wire Holding Fixture to the top edge of the mold, between the two rows of wires from the tube.
    • Solder the tube wires to the fixture, in alternate holes, arranged according to their sequence on the tube. Neatness counts.

    The tube + Wire Holding Fixture will be referred to as the "Tube Assembly".

    Step 13: Take Measurement

    Measure the distance from the top surface of the mold to the top surface of the tube.

    This measurement will be used later.

    Step 14: Remove Tube Assembly and Adjust Orientation

    • Remove the Tube Assembly and board from the mold. Notice that when you hold the assembly by the Wire Holding Fixture, the opposite end of the tube dips down due to gravity. We are cool with this for now.
    • The gentle bends you obediently left in the wires previously will facilitate a fine adjustment of the tube orientation at this time. To bend the wires for proper orientation, grasp the Wire Holding Fixture firmly by the ends as you gently twist the tube toward its desired orientation, perpendicular to the Wire Holding Fixture.
    • The end of the tube will still be dipped down after you're done with this adjustment, and we're still cool on that one. Don't try to make the tube level.
    • Now is an opportune time to check carefully for crossed wires and shorted wires, and correct any problems now, before judgment day.

    Step 15: Make the Standoff Jig

    When the casting liquid is poured, the end of the tube that dips down now will suddenly float upward.


    We want the tube to be level!

    Enter the Standoff Jig, which will, if you please, hold the end of the tube level during cast curing.

    • Cut a 3" x 1/2" piece of perf board and a 3" length of bare 24 AWG wire.
    • Pass the wire through a center hole of the board so that 1/2 inch projects out one end, and solder in place.Using the measurement you trustingly took in Step 6, snip the long end of the wire to this length.

    Alternate wire: I found some rigid 30 AWG brass wire at Daiso, and I used this on my final casting. The 30 AWG is one-quarter the diameter of the 24 AWG wire, but brass is not solderable, so this had to be taped down to the perf board.

    Step 16: Final Preparation for Casting

    • Remove everything from the mold and wash the inside to remove dust and accumulated schmutz. Let it dry completely.
    • Set the mold in its support fixture. Set the Tube Assembly in the mold, orient it accurately; then tape the Wire Holding Fixture down to the mold.
    • Tape it securely so that it holds flat against the top surface of the mold. The end of the tube will be tipped down, but should otherwise be properly oriented. Eyeball this orientation carefully, this is you last chance for redemption.
    • Re-check that the tube wires are not touching each other or the edges of the mold.

    Step 17: Casting

    The uncured resin is quite aromatic due to some quite volatile petroleum distillates, and you will need a well-ventilated area for casting,...

    ...lest yon vile fumes breach thy nostrils and twist thy mind.

    Like Shakespeare said.

    Inevitably, for me at least, the process of mixing and pouring the resin results in sticky resin on my fingers, which then gets transferred to everything I touch. To avoid all this cross-contamination I use the polyethylene gloves. They are fast-on-fast-off gloves and I go through several pairs in one casting. Keep the box handy.

    Set the mold on a level surface.

    Mix and pour the resin per the manufacturer's instructions. I used 16oz of resin and 32 drops of catalyst at 75F.

    Note that once you pour the resin, the end of the tube now floats toward the top.


    Step 18: Immediately After Pouring the Resin: Attach Standoff Jig

    • Orient the Standoff Jig so that the wire you cut to the measured length pushes down on the end of the tube to keep it level. Tape the jig down.
    • Eyeball the tube and make sure it's indeed level.
    • If it's not, panic.
    • Once everything is in place, you can put a shield in place over the works to help keep airborne debris out. Believe it or not, gnats are attracted to the volatile fumes. I had to pluck one out of a fresh casting.

    Step 19: Removal of Standoff Jig

    The Standoff Jig must be removed before the resin is fully cured, lest it be stuck there forever.

    If you remove it too soon, the end of the tube will begin to float. If you remove it too late, it will leave a holiday in the casting.

    Check it frequently after pouring the casting and remove it when the casting is in a gel state. Mine came out about 25 minutes after the pour. Worked perfect.

    Good luck.

    Step 20: Remove Cube From Casting

    Be sure to give the casting ample time to cure. Minimum 24 hours. Sorry, you just gotta wait.

    There is very little adhesion to the mold, however it takes some coaxing to get the casting out, as this is a deep mold. Try turning the mold inside-out to pop the casting out.

    At this point the surface of the casting may be sensitive to handling and fingerprints may embed in the surface if you grasp it tightly, but these will be easily sanded out.

    All my castings ended up with alligatoring on the surface, i.e. the wrinkling you see in the picture. This is caused by uneven curing of the resin which is caused by too much catalyst. So to avoid this you can try less catalyst and/or curing at a lower temperature.

    However since you will have to sand the cube surfaces anyway, there is not a lot of need to alleviate alligatoring. Grinding down the alligatoring with coarse sandpaper is no sweat compared to all the fine sanding and polishing required later to get optically smooth surfaces.

    Before finishing the surface surfaces we will attach the header to corral the flailing wires and allow a quick re-test.

    Step 21: Attach Tube Header

    • Unsolder the Wire Holding Fixture.
    • Insert the wires from the tube into the header.
    • One method is to insert the 2 rows of wires into opposite ends of the header. You can leave the wires long or you can pull the wires through the holes till the header is at the desired length from the cube.
    • Long/short, front/back: The choice is yours.

    • Once inserted, solder the wires down. Avoid bending the wires right where they exit the cube to avoid stressing them and breaking one off and ending up crying profusely.
    • Double check that you have the header wired correctly. Triple, even.

    Step 22: Quick Test

    • Once the header is connected you can plug in the clock board for a quick re-check, and enjoy its display, in all its serene blueness. Listen and you will hear the angels sing.
    • If the display does not light up or if the characters are xxorked:

    a) Check for wires crossed. If you find that wires are crossed inside the casting, you can cross wires outside the casting to make up for this.

    b) Check for wires touching each other.

    c) Check that wires are connected to the proper header holes.

    • Once operation is confirmed, pull the power.

    Step 23: Finishing the Surface: Philosophy of Finishing 101

    Philosophy of finishing a casting

    Getting the resin casted right was a challenge; finishing the surfaces of the casting, likewise. Finishing a casting is somewhat of an art form, and it's a different process in different circumstances.

    I will direct you to examples of how it's done, but unless you are practiced in this process you will, to some measure, have to find your own way.

    I checked on other I'bles for guidance on finishing castings, and some make short shrift of it and were able to get nicely polished castings with fewer steps than I used, whereas others, one site in particular, described a very detailed procedure, extensive and expensive.

    You can use my instructions as a starting point, which are based on what I needed to do to get the finished product I wanted.

    And for additional perspective, check out these I'bles, and that will help you find your own way if you find my method does not work out optimally for you:

    So why the differences in methods? I'm thinking the type of resin would be part of the answer.

    In the first site I linked, they did all the sanding by hand, two people double-teaming. I used power sanders, and I found that polishing that way resulted in a finer finish than the same grit sandpaper used manually. That could be another part of the answer.

    Step 24: Finishing the Surface: My Method: Initial Sanding With Belt Sander

    First of all before getting into it, be advised:

    Up is down

    At this point, when referring to orientation of the cube, "up" will be with the tube right-side-up, and with the wires projecting downward out of the cube.

    I left the bottom surface of the cube unfinished since it is not exposed to direct view in the finished product. But now that my clock is finished I see that not finishing one surface does affect the overall appearance since the inside surfaces are reflective.

    So I would recommend that for best appearance you carefully finish all of the bottom that you can without impinging on the leads emerging from the cube.

    Sand the surfaces with 100 grit sandpaper, preferably with a belt sander. That will make quick work of the alligatoring. Harder will be to even out the uneven surfaces and leave a flat surface on each side.

    Sanding the polyester cube will tend to clog the sanding belt. You might avoid this by having the patience to sand slowly, then use the sanding belt cleaner when the belt becomes clogged. This is a great accessory for a belt sander in general, and will extend the useful life of the belt.

    If you sand the polyester cube too fast the polyester will heat up and melt into the sandpaper and hopelessly clog it. Wet sanding would help preserve the belt, should you have access to a belt sander that has the capability to inject water.

    Step 25: Sanding: Finer and Finer

    The last two of the pictures above are after sanding with 2000 and 7000 grit sandpaper.

    Next, sand the surfaces with 600 grit sandpaper. I used the orbital sander from here on. Keep the surface wet to help preserve the sandpaper, being careful to not get the electrical parts of the sander wet.

    Then sand with the 1200 grit. Give a quick sanding manually, then use the orbital sander. Use light pressure. After 2 minutes of sanding inspect closely and if you see any areas lacking. Touch them up lightly by hand the do a quick re-do with the orbital sander.

    Treat one of the surfaces with the scratch remover and polish to see if you need more surface preparation. Likely you will need to graduate up to 2000, 5000 and 7000 grit sandpapers. Use the same technique as with the 1200.

    Step 26: Final Finish

    Now for the electric buffing. Apply the white polish compound to the buffing wheel and buff the surfaces with light pressure. After a spell of this treatment, use the second buffing wheel without polish compound for a finer buff.

    Or you can try washing the compound out of the original buffing wheel and using it.

    Next, treat the surface with a few applications of the automotive scratch remover.

    Then apply a few coats of the automotive caranuba wax, buffing with a rag after each application. Finally buff with the clean buffing wheel.

    You should have a nice looking clear cube at this point, optically clear, but for the very slight grayish tint from inside the cube itself.

    Step 27: Prepare the Base for the Cube

    • Measure the length and width of the bottom of the cube where it will insert into the base. Cut a hole in the center of the top of the box to those dimensions.
      • Using Laser Cutter:
        • I used the laser cutter at the maker space "Crash Space LA" in West Los Angeles. Laser makes a nice sharp precise cut.
      • Using Jigsaw:
        • Suggest you cut hole slightly small so you can file the edges smooth.
    • In what will be the back of the base, cut out a 1-1/2" x 5" hole for access to the clock board.
    • Use the resulting cutout as a hinged cover for the access hole. Leave a slot for the power cord. Attach to the box with the brass hinges.
    • Mount the clock circuit board on the lid, locating it to allow mating of the respective connectors on tube header and the circuit board.
    • I put the following label on the hatch so a bloke will know the functions of the buttons on the clock board:

    MENU SET +

    • You may want to also put a "HIGH VOLTAGE" sign on the hatch. The clock board has a max of 70 volts, and though that level isn't likely to zap some schlep into eternity, a sign might still be a prudent caution. Also, it might serve to keep unwelcome fingers out of the works.

    Step 28: Gluing the Cube to the Base

    Use foam mounting tape to hold the cube in the proper vertical position while the adhesive is applied and cured. Dab the adhesive from the bottom side so that it won't seep up to where it would be visible from the top.

    The JB Weld PlasticWeld is hyped as a fast cure adhesive, however I found that in this case it takes about an hour. I'm guessing that the adhesive has a reaction with one of the components that prevents a faster cure.

    So be careful with the unit until you are sure the adhesive is solid.

    Once it does cure it is a good solid bond.

    I will call the Cube + Base the "Cube Assembly", which is little overmuch of me since we're almost done with the project.

    Step 29: Final Assembly

    • This is a tricky part: While holding the Cube Assembly at an angle over the lid, position the connector directly over socket on the circuit board. If you are a three-handed person, this will be a definite advantage.
    • While aligning the cube assembly to the lid, again make sure that none of the wires are shorted to each other.
    • Fasten the cube assembly to the lid.
    • Through the access hatch push the connector into the socket.

    • It wouldn't hurt to check the wires again for shorts.

    • Install the four rubber screw hole caps on the bottom of the lid.


    Step 30: Using Your ICC^3

    Plug in the power adapter and set the clock.

    The ICC^3 is at its best placed where there's a dark background so that the gentle cyanotic glow of the tube is most visible.

    Be sure to brag about this object to your visitors till their eyeballs recede into their foreheads.

    You got the right!

    Step 31: Addendum

    If you are environmentally conscious you may want to collect all of the items that have traces of uncured resin, i.e. container, mixing cups, gloves, etc. and put them in a labeled zip lock bag and take them to wherever you take your household hazmat.

    The Adafruit site has a mod to add a GPS Timekeeping feature to the ICC, which would be nice. Always on time and never the needing to set the clock. As for me, it's already been a lot of work and expense, so, "nope". Not for now, maybe someday.

    As a last caution, I thought I'd to mention the fragility of the tube.
    Yes, it's glass, but it's very thin glass and breaks readily. Handle with care!

    An interesting alternative on this project would be to make it in a cylindrical casting, mounted from the side. In this case the wires would go straight out rather than bending up. You would have to figure out how to finish the surfaces.