Dallas Makerspace Show & Tell - January 2019


That’s awesome @lukeiamyourfather! I love all things fractal and the Sierpinski triangle is one of my favorites.

I saw this video a while back (watched it around Halloween 2017): https://www.youtube.com/watch?v=kbKtFN71Lfs

I was blown away by it and implemented my own version just to confirm the magic. I can’t find the code at the moment (it’s pretty easy to write), but that this approach:

  1. Start with 3 points that form a triangle
  2. Start at a random position
  3. Choose one of the 3 points, move halfway towards it and put a point there, then repeat that process forever

produces this:

blows me away.

I tried it for some other polygons, and the pentagon looked cool as well:

Apologies for crappy cell phone pics as opposed to proper screenshots (I wanted to share it quickly with someone at the time). I also apologize for the state of my desk at the time (to be honest, it’s worse now :slight_smile: )


Wow that came out great!


Numberphile is awesome! Based on the explanation in that video I tried creating it in some software that I’m familiar with (Houdini). I’m pretty happy with the results.

When you get close you can see the individual samples. I started with 1,000,000 samples.

The five and six sided versions produce some interesting results.

When bumping up the samples to 10,000,000 the shapes get much more defined.

Thanks for posting that stuff @Kentamanos. It tickled my brain in just the right way! Scene file for Houdini is attached if anyone wants to mess with it.

fractal_v001.zip (9.2 KB)

import random

iterations = 1000000

node = hou.pwd()
geo = node.geometry()
points = geo.points()
divs = hou.node('../circle1').parm('divs').eval()
current = points[divs].position()

for i in range(0, iterations):
    index = random.randrange(divs)
    anchor = points[index]
    halfway = (anchor.position()-current)/2+current
    current = halfway


Kick ass :):grinning:

I tried the hexagon and got really boring results. Not sure what I was doing wrong (it just kept giving me hexagonal clusters of isolated hexagons).

I was wondering if a voxel approach could be used to generate a 3D version (similar to what you printed). It would be an absolute abuse of RAM and CPU for the achieved resolution :laughing:


Sorry in advance for the long post. I think some people are probably curious about the overall build process, but I should probably get a blog and just post a link :laughing:

For Christmas, one of my friends got me a little crossbow pistol. I felt like it really needed a rail on it to hold the bolts it came with and decided to design something that would attach under it. The length of it had oval shaped holes and I decided to use those. You can see them in this photo:

Those oval shapes were a complete PITA to attempt to measure with nothing but calipers (I looked around on Amazon for pin gauges today that would have made the job easier :smile: ). I decided to use the first and second slots near the front and did some test fit pieces to make sure I could get it correct:

My first attempt on the first one was really close. The second one took a lot more effort:

Once I felt like I had the slot dimensions good, I CAD’d up some pieces in FreeCAD (I like using it when I’m not sure if just “free as in beer” is good enough). The front pieces had a circular pocket for the head of the M3 screws while the back pieces had a hexagonal pocket for the nuts to fit into:

I then designed a rail (the piece that actually holds the bolts) to go between them:

A little time in “Prusa Control” (Slic3r) to generate G-Code:

The printed pieces looked like this (I pressed the nuts in with pliers. I probably should buy a small press or use a vice next time):

This shows the pieces (minus the rail that goes between them and the screws in place):

And finally the finished product with it in use (bolts attached):

Next time I might go with just slots to push the nuts through instead of the hex holes for nuts which were a PITA to clean out with support material due to printing orientation. It also makes me consider getting a multimaterial unit (MMU) and investigate the support filaments that dissolve in water.

I might try this again later with 3 total pieces (combining front and back pieces), but I didn’t want to worry about the exact length of that. I also probably need to fillet a few more sides where possible for comfort reasons.


Great post showing development!


I can add drilling holes in glass with the drill press to my list of super powers :grin:


Hi! I am Martha Myre. I am one of the Fiber Frolicers in the Creative Arts group.

This is a cover for a spiritual journal I made for a friend. Process: made a batt of the blue that I then “prefelted” by needle felting. I then created and felted in the pictures. Wet felting was next and then I finished by more needlefelting to compact and shape it more as well as adding details to the pictures and the words. Lastly, I steam pressed to make sure everything was snug.

The background and most of the pictures are wool, but with added bits of silk for texture and shine (like the flames). The words (and the chi rho) are felted silk.


Nice job. I’m surprised you didn’t have to use anything as coolant …


You bet ya there is coolant in there. You can not tell by the picture but the glass has a curve to it and there is water in there.


Ahh. You’re right. I just didn’t see it. Another approach is to put a snake of clay around the hole to retain the coolant. Alcohol works particularly well.


Anita and I have been working together and experimenting with diamond tip hole saws on glass. I’d bought a set and I’d say our observations are:

  • Have some sort of coolant/lubricant. On flat glass I used plumbers putty to create a small damn around area and used WD-40, water, 3-in-1. All seemed to work okay. Her’s was dished so it pooled around cutter.
  • Slow speed to medium speed I found worked best
  • We both agree: Light pressure. You can feel it cut and slowly go through.

When it finally breaks through they is some light chips around the edge on hole, it is small and sort of looks like a bevel in places almost. Anita used wood to back this up. I found with flat glass if my sacrificial piece underneath was another piece of flat glass that I put some water on and then “rung them together” so they were stuck together, when it broke through, much less chipping because it was supported.

Have not tried on tempered glass yet.


I had some putty but since it had a bowl shape it was not needed.


Terrain for RPG gaming. Created for a terrain-making contest online, and then to be given as a gift. The theme was “Nature reclaims”. It is scaled for 28mm miniatures.

Here are a few WIP shots. The basic piece is made from XPS insulation foam with a couple of 3D-printed parts.

I cut the foam on my homemade hot wire foam cutter (using a power brick from the give-away shelf). I textured the foam by rolling a rough ball of tinfoil over the surface.

The strangler fig tree roots were made by layering up Kleenex tissues with some wood glue and twisting into ropes. Additional glue firmed up the surface after the initial glue dried.

Some smaller details were added: lichen/moss everywhere, a palm with leaves cut from inkjet prints of palm leaves, and some mushrooms and bracket/shelf fungi on the trunk.


I want to give this 100 likes!


Seriously, that is awesome!


I know, right???


I had wood under my my piece, no issue with chipping. keep in mind it was 1/4 double glass fused, so it was thick.


Honestly, all are great but your board is my favorite. I hope you are very successful with this design and product! It’s truly awesome.


Twin pens - one with Cherry, one with Cocobolo. Both kits are antique pewter retro rollerball from Rockler.