Tuesday 13 October 2015

Let's get rolling! CAD design of custom embossed rolling pins.

From time to time my close friends and family (and clients, contacts, pretty much anyone) give me great ideas for items that would lend themselves to 3D printing. Being able to design and make stuff is only a part of the process - coming up with real-life uses for my design skills and 3D printing ability are often limited to things that I am interested in. 3D printer upgrades, robotic claws and house-hold objects which I will use myself to name a few.

Recently my brother saw a batch of cookies which had been embossed with a special rolling pin, The design was a skull image which was laser-etched into a plain-old rolling pin. This would leave a sheet of dough behind with the skulls embossed, ready to be cut with a plain-old cookie cutter, baked in a plain-old oven and presto! Cookies with little skulls all over them!

A short discussion followed with some of my female family members who were interested in having custom-made cookie rollers themselves, so I decided to take it on as a bit of a love project.


Using Autodesk Inventor as my software of choice, I initially came up with a basic shape for the roller. It would resemble a traditional rolling pin but to make it easy and cheap to manufacture I decided to use a piece of wooden dowel as the 'handle', that way I just needed to print a cylinder with a hole through it to put the dowel in and some embossing on the outer diameter.



I sat back and admired my work - first time I'd tried embossing a cylindrical surface with a continuous pattern. It turned out pretty good! I showed them and they liked it too. Great!

But while the artist in me (note: there isn't one) was happy, the engineer wasn't. I ran this model through my 3D printers 'slicing' software and found that it would take a whopping 13 hours and about a third of a roll of plastic to print it. They would be quite expensive to make, and if the girls wanted a few designs there would be a few rolls of plastic and many days of printing.

So I developed...

Sleeves!




The idea here is that I could print a central hub for them ONCE and then print much thinner 'sleeves' which would slip on over the central hub. The cost for a single design would be a little more than the solid roller, but once they had a second design in mind the cost came down significantly. Each sleeve would only use about one tenth of a roll of plastic. Great!

Add some handles for the ends of the dowel and we had a finished product! Here are the first two designs they requested, love hearts and paw prints.


Monday 3 February 2014

3D Design Project - Ikea Lamp Upgrade

3D Printed Lamp


I've just moved into a brand new house with my girlfriend, we're moving things into the house and I've come across a lamp base that we bought from IKEA. The lamp suited the small space it was in previously, but it looked a bit anorexic in our lounge room. I wanted it to be just a little more... "bulky"

This is a CAD render of the lamp base so you have an idea of just how simple this lamp was originally.

See... VERY simple, a square plate as the base and 12mm shaft as the 'stem'.

I decided to "Jazz" it up a bit by adding to it somehow. I have a 3D printer at my disposal so, why not use that to do something interesting! (I've wanted a nice big project to test it out on too!)


Designing:


The printer I have is not large at all. It's great for parts that fit into its 200mm x 200mm x 100mm build volume, but this lamp is at least 1.2m high. The only way to 3D print something this size would be to make it in sections. I measured the lamp base 'stem' with callipers and found it was 12mm diameter. I decided each segment of my design should 'clip' onto the stem. This seemed easiest. I sat down and sketched out a few ideas and came up with this design which I created in AutoCAD (note that I have added circular pads on the bottom which you can see in the pic below, these assist during printing and we can simply snap them off later).

"You know they got in trouble for the 3D printed gun right? You can't go making crossbows..." - My Brother

My idea is that I will have many of these clipped onto the stem in succession, and that I can then rotate each one slightly more than the last so that they sort of 'screw' up in a helix around the lamp base... or I could just have a straight vertical strip or a zig-zag pattern, or just randomly arrange them! I think it could look neat! I was confident it would turn out good, and so after designing it I decided I didn't need to do a final render of the entire lamp, I would just go ahead and see how it turned out! At least print a few segments and see how they looked.

Preparing the print:


There are 2 main plastics I use for 3D printing, PLA and ABS. PLA is renowned for being very easy to print with, but it also has a lower melting point (a VERY hot day could make a PLA part droop or distort) and it is less flexible than ABS - Since I was after a "snap" fit between each part and the lamp base ABS is in the lead, it will easily flex that little bit and snap onto the base. It is also plausable that the lamp would get a bit warm, either in-use or from the hot sun coming in a nearby window, so ABS was chosen for this project.

ABS DOES have some caveats though, it tends to contract as it cools, which in turn can make it 'curl up' and become un-stuck from the build platform. This results in errors in the finished part or completely failed prints. The circular pads on the bottom were incorporated to overcome this - the pads give the part more surface area and act like suction cups or wads of Blu-Tack that hold the part to the build plate and resist the warping that naturally occurs.

There are 2 programs I prefer to use for 'slicing' the 3D models and preparing them for printing. One is named Kisslicer and is my go-to slicer. It is VERY fast, easy to use, and gives the highest quality printed parts. Printing with Kisslicer is also usually completed quicker and the machine is smoother / quiter due to its intelligent path generation. Slic3r is a good backup tool - I will ALWAYS use Kisslicer first, but if Kisslicer has problems or the part requires some really fine-tuned tweaking to work, Slic3r will often do the job. Slic3r also has a 'brim' feature which automatically inserts a brim (like a top-hat brim) around the bottom of the part which has the same effect as the circular-pads I added to aid printing. (Note: both programs have come a long way since I wrote this post)

Kisslicer showing the first 'slice' or 'layer' that will be printed.

Results:


I decided on black material, made the first part and noticed that it was a really easy print, without much detail and big long lines / curves. I could run it a bit faster and sped the time up from around 42 minutes to just under 30 minutes per part. Before starting the second part though I snapped the circular pads off the first part and went to check how it fit on the lamp base.

Snaps on perfectly!
Great! So I continued printing and got into a 30 minute routine, I would go into the printing room, remove the freshly printed part and begin printing the next. After I had printed a few more pieces I tested out my design to get an idea of how the finished product would look.

It looks... ok? yeah.... I think? Maybe?

I really wasn't sure at this stage! But continued anyway - and as time went on it looked better and better. Here are a few pics while it was still not quite finished, but I was trying different configurations to come up with my favourite.



After trying different variations of twists, zig-zags and random configurations (changing a few times a day for a week or so) we decided that a nice gradual helix was most appealing. I also decided the 2 lamp shades I have are ugly on this now, so, I will either get a plain white cylindrical lampshade the same diameter as the lamp base segments, OR 3D print a lamp shade continuing the design of the base up around the bulb (perhaps in white plastic instead of black). When I decide I will update the post with pics!

From concept to reality:

My 3D Printed Ikea Lamp Upgrade!