Practical 3D Printing with the Anet A8: The Saga of the Toothbrush Holder

3D Printed Toothbrush Holder Anet A8 Final Result

Printing a toothbush holder on the Anet A8 Printing a toothbush holder on the Anet A8 Printing a toothbush holder on the Anet A8

Today I knocked out a quick 3D print for a practical purpose for once – a toothbrush holder. Not very exciting, but a good chance to get some practice in with Fusion 360. It was both a close failure and an accidental success, and a good lesson in tolerances/interferences.

Backstory

Our ensuite shower has been out of commission for about a month, waiting for me to get off my arse and fix a broken tap. I had been quite happy ignoring the problem and using the main bathroom shower, but my long-suffering girlfriend finally got tired of me working on every project but the important house repair ones, and so ordered me off to Bunnings for plumbing supplies.

The tap turned out to be an easy fix, but unfortunately a disused bathroom in a month of Australian summer lead to a little bit of gross mould build up in the shower. A spray of noxious bathroom cleaner soon sorted the shower tiles, but sadly our toothbrush holder was a write-off.

Aha! Thought I, finally a chance to use my 3D printer for a practical purpose. I not only could design a holder to replace the old one, but I could make it so it would fit perfectly in our wire shower caddy.

The Design

I measured up the caddy and came up with an idea for a design that would fit in the upper tier basket of the caddy, and also clip to the vertical arm to prevent falling over, a problem our old one had that irritated me to no end. Here’s what I came up with:

Fusion 360 3D Print Mockup of Toothbrush Holder

At 12cm, this was going to be the tallest print I’ve done yet with the Anet A8, so I was excited to see how the printer handled the dizzying altitudes.

I made the walls of the holder 3mm thick, which might seem slightly overkill for this purpose, but I had an issue with the last box-like structure I printed where I made the walls 2mm, then accidentally set Cura to do 1.2mm wall widths. The mismatch made for a weird result where two thin walls were printed side-by-side with no infill material between them, leading to some weird flexing, floppy walls. Also it didn’t print a ‘cap’ for the top of the walls, leaving a thin channel down the middle. If that makes no sense at all, see below for a pic.

So wary of this happening again, I increased the wall thickness. As it happens, I ended up remembering to changing the wall thickness to back 0.8mm in Cura anyway, so I guess it would have been all good either way.

I put a nice angular cut across the top of the holder for looks, filleted every corner that could stand to be filleted, and put some holes in the bottom for drainage. With all that done, it was time to export the STL, load in Cura, dump some GCode onto the SD card and off to the Anet A8 we go.

About four hours later I came downstairs to this:

3D Printed Toothbrush Holder

Looks pretty good. And yep, that’s definitely some pretty bad ghosting in the walls, no big deal for this print, but it’s definitely time to get some fibre reinforced belts to replace the stretchy stock belts.

I used the ‘triangular’ support setting for the jutting out clips on the back of the holder, just because I’d never used it before. It worked really well, the support structure was a bit harder than usual to break away, I had to use a craft knife to persuade it in some areas, but in the areas where it did snap away by itself it came off cleanly in one piece and didn’t leave much behind at all.

Triangular support structure on the Anet A8

Alright, cool. Time to chuck this in the bathroom where the world (or at least my girlfriend, or anyone else who happens to use my shower) can see a bright orange testament to how bloody clever I am.

Tragedy!

The fucking thing didn’t fit.

In my haste to whip up this design I didn’t consider tolerances at all. I made the length of the holder the exact width of the basket, and due to a slight taper in the basket and also probably the imprecise nature of 3D printing, the holder wouldn’t fit.

I found I could jam it in towards the middle of the basket if I flexed bent out the basket with my hand, but that negated the use of the vertical bar clips, and put the holder in a less useful spot, essentially voiding the main design elements of the holder. I could have just as easily just used any old narrow container, dammit.

Triumph!

However, just by coincidence, the height of holder turned out to be a centimetre or so less than the distance between the top and bottom tier of the shower caddy.

I realised (after some frustrating attempts to cram the slightly-too-large holder into other parts of the caddy) I could actually attach the bar clips to the lower section of vertical bar, turn the holder sideways, and voila! Fits perfectly, works a treat, and even looks like it was supposed to work that way! Hooray!

Actually, I got another lesion in tolerance/interference here. The bar clips were designed to use a very common friction-fit method of gripping the vertical bar – they are basically a cylinder with a section cut away, just enough so that pressing the clips onto the bar will cause them to flex slightly, they then jump over the wider part of the bar and settle into their original position once the bar settles into the centre of the cylinder (hopefully without breaking the clip in the process) and use that friction to hold on to the bar. This counts on the design leaving enough of a tolerance gap to push the bar through, but also leaving enough interference behind to grip the bar tightly.

Fusion360 Screenshot of Bar Clamp style fastener

I totally guessed at this during the design process. I don’t know if there’s a design process or method to calculate this (actually on thinking about it I’m fairly sure there must be, probably something taking material flexibility Vs  thickness of the bar Vs the width of the barrel of the clip into account) but as a result I just eyeballed it and got pretty close, but not quite.

With some quite hard pressure I could see the clip start to open up but not quite clear the bar. I was pressing so hard I got worried about snapping the clip, so I grabbed an emery board and filed down the edges on both clips a bit. After a few goes at pushing and filing and pushing and filing again, the clips finally clasped the bar with a satisfying ‘snap’.

3D printed Toothbrush holder clips

The final result of this rambling tale is I got a pretty nice looking custom toothbrush holder, learned some lessons about integrating tolerances into design, and got away with an accidentally useful alternate use for a print that technically didn’t work. Pretty good result I reckon.

 

The Anet A8 – The Little 3D Printer That Could

I’ve spent most of my spare time in the last two weeks playing around with my new favourite toy – The Anet A8.

The Anet A8 is a 3D Printer kit, an Prusa i3 clone to be exact. It’s made up of the cheapest components possible, and as a result comes with a rock bottom price tag, making it very attractive to those looking to get their feet wet with 3D printing.

I paid about $250AUD for mine, and that was just before Christmas 2016, so it’s probably even cheaper now. At the time it was cheapest on Gearbest, so that’s where I got mine, but you could check out Aliexpress, or even Ebay – which might feel a bit safer if Chinese dropshippers aren’t your bag.

In spite of the cheap parts (and a few well documented failure points) the A8 has a reputation for putting out high quality prints when properly configured, and it also has a thriving community around it helping out noobs and pros alike with troubleshooting, settings and improvements.

The Anet A8 arrived a week late, a cocoon of yellow tape containing a box of Styrofoam crammed with electronics, motors, wires, rods, screws and a wobbly acrylic frame. This is a kit printer, so you have to put it together yourself – so that’s what I spent the first few evenings of 2017 doing.

Anet A8 Electronics Test
Testing the electronics on the Anet A8 before installing into the frame.

There’s no printed manual, but there is an SD card included with PDF instructions and information and some starter models to print (this SD card later becomes your method of loading GCODE onto the printer) but the best way to put the thing together is to methodically follow along with the eerily silent YouTube videos posted on the company’s channel. The videos feature a faceless employee who screws the thing together while you frantically pause and rewind to read the English captions that display vital information on screen for approximately one millionth of a second each time.

Anet A8 Wiring
All the motors, hardware and wiring installed and wrapped, time to plug it all in.

It takes several hours to build, and once it’s all together it certainly looks impressive… just don’t lean on it or anything. You see, the main difference between this printer and one many times the price is the frame material – this is creaky acrylic, not machined aluminium or steel. Beggars can’t be choosers though, so you take a moment to bask in your handiwork, and then it’s time hold your breath and plug the thing in.

Once up and running, and with all the electronics checked and debugged (again by following along to a YouTube video, this one narrated in accented English with lilting guitar music in the background) after nervously observing the woefully cheap power supply with its hand wired cord for signs of explosive meltdown, I decided to throw caution to the wind yet again and just try to print one of the pre-prepared files on the SD card.

I levelled the heat bed with a sheet of paper as per the instructions, fed the extruder one end of the small sample spool of PLA that came in the box, selected the first file on the list, and hit ‘print’.

To my complete astonishment, it fucking WORKED.

Not only did it work, it worked really, really well! Obviously as you can see, the print didn’t finish – Just my luck, I randomly chose the largest object  in the test print folder – and there just wasn’t enough filament in the sample roll to finish it, but the level of detail and resolution are really quite impressive.

Now all that’s left to do is play the waiting game while some rolls of PLA are delivered, and then it’s time to start printing upgrades.

One of the coolest things about this printer is the ability print out parts that actually improve the durability of the machine and the quality of the results it produces. The community built around this particular model is gung ho about designing and implementing new parts and upgrades to reinforce the frame, improve the cooling and bed levelling, and fix or enhance just about every other aspect of the printer.

You too can get a glimpse into the fate of humanity and our future machine overlords as this seemingly benign piece of technology uses you as an agent to improve itself, as you find yourself up at 1am printing new and improved fan shrouds, mesmerised watching layer after layer of your own doom lay down on blue masking tape…

While I’ve had a great experience thus far with mine, it hasn’t all been roses in the short but eventful life-so-far of the Anet A8, and this should be pointed out at the outset to anyone looking to try their luck with this machine. Tales are rife throughout the community of cracked frames, burnt out boards, and even parts catching on fire. Remember, these are cheaply made components straight  from Shenzhen where the QA is light-to-non-existent, that you’ve whacked together in your garage while squinting at a Youtube video, and then happily plugged into the wall and run 240 volts into. That’s a lot of possible failure points.

Luckily though, just as prevalent are clever and resourceful owners coming up with fixes and DIY solutions to these problems, including adding cheap, easily sourced MOSFETs to the heat bed to stop boards and connectors from frying, adding fused switches and safety covers to the terrifyingly exposed power terminals on the power supply, and replacing boards, motors, belts, frame pieces, and just about everything else with higher quality parts should yours fail, or you just feel like upgrading.

Time will tell how well this printer can stand up to regular use, upgrades or no. For now, it’s running like a dream. Every hiccup I’ve encountered has been solvable with a small amount of googling or scrolling through the community facebook page, and fixing each issue has been a small lesson the minutiae of 3D printing tech.

This humble printer has captured the hearts of a huge community of Makers, from first timers to old pro 3D printer enthusiasts. For those with the time and the patience to build, maintain and upgrade this cheap and cheerful device it represents a perfect entry into the world of 3D printing – along with a crash course in the fundamentals of the hobby that probably won’t get with a more expensive or pre-built device.

I’ll keep writing about my adventures with this machine in the coming weeks, so stay tuned.