On a roll…

Two days ago I finally finished assembling my Filastruder and Filawinder. There were a few last minute tweaks that became necessary once I had decided on my final setup and positioning of the assembly in my studio, such as sourcing a longer cable for the Filawinder sensor and making a stand for the Filastruder to sit on. I had initially planned to go for a fully wall-mounted vertical setup, but the Filastruder is a reasonably heavy piece of kit, and as I don’t quite trust the strength of the walls in my studio, I ended up going with this 45 degree angle tabletop assembly instead, designed to fit the enclosure I am using. Going vertical should always be considered as the superior option, because of the way gravity aids the extrusion process, but you have to work with what you’ve got. Setting up the Filastruder on my workbench and the Filawinder on another workbench opposite gives me enough room between the two machines to drop the filament in a generous loop once extrusion starts, with the Filawinder sensor placed on the floor between them. The sensor cable that was included in the kit was not long enough for this to work (and as 3mm filament needs a slightly longer run before going into the Filawinder, I am guessing it would not be for most people extruding this diameter), but I sourced a 15ft version and the appropriate connectors, which should hopefully see me through all future eventualities.

With everything in place it was finally time to turn on the ‘struder and do the initial purge! Excitement mixed with apprehension as I turned on the heater for the first time and  watched the numbers on the display creep up, eventually reaching the target temperature of 171C. At this point it is really important to give the Filastruder enough time to heat up thoroughly – anyone who has ever used an enamelling kiln will know this as ‘soaking’. While the thermocouple might be displaying the target temperature, this is only measured on a tiny part of the assembly, and it can take anything from an additional 10-30 minutes for that temperature to reach other areas of the machine. Turning on the motor before everything has been thoroughly heated through can lead to disastrous results – in extreme cases even barrel deformation – as the plastic is not liquid enough to let the screw turn freely, putting strain on the motor and other mechanical components. After about 30 minutes I finally felt comfortable enough to turn on the motor. The PLA started shooting out of the barrel almost immediately, initially as very liquid blobs of hot plastic and eventually as filament. During the initial purge, this will be filled with metal particles and other debris, and it could take up to 8 hours to clean out the barrel completely. This gives you however plenty of time to adjust the temperature settings to suit your material, and really fine-tune the diameter of your filament. In my case, it transpired that 171C was far too high for the PLA I am using (Natureworks Ingeo 4043D) and eventually I settled on 155C which gave me a relatively stable output of 2.7-2.8mm filament.

Next: getting the Filawinder to work. 3mm diameter filament poses a further challenge for the winder, as it is a lot stiffer when it comes out of extrusion and thus harder to wind. My initial attempts to get it to work failed miserably, mainly because I had followed the assembly instructions to the letter and cut my length of PTFE tubing in half, making it too short to work in my setup. This resulted in increased strain on the spool which meant the motor was not strong enough to pull the filament and kept getting stuck. Luckily I still had the other bit of the tubing, and a bit of kapton tape later a full length PTFE tube means that the Filawinder is now working like a charm. This is really important for getting a consistent diameter – even moving the sensor during winding can mess up the fragile dynamics that exist between the extruder and winder and be the difference between producing excellent filament and something ready for recycling. After six hours my initial purge was complete and my first spool of filament all wound up:

FirstPLA01Of course, its contents will now be going in the bin as the contaminants mean the material is unsuitable for putting through a 3D printer or re-extrusion, but I’m still proud I made it this far and now have a working setup to extrude my own custom filament. More to follow soon…

Filastruder Build … Electronics

After receiving my hotly anticipated Filastruder kit, I couldn’t wait to put it together. Starting with the mechanical build, completed over the course of a few afternoons, things seemed to be going smoothly. I had ordered the enclosure kit to go along with the Filastruder, and after a slight hickup with the interior fit of the upgraded v1.6 motor that Filastruder creator Tim Elmore helped me sort out quickly and efficiently via email, my small-scale home extruder was ready to be imbued with electronic life.

This was the part I had been nervous about from the beginning, and even more so as I read the sparse instructions included with the kit. While I have accumulated some experience with soldering PCB assemblies over the last year, and my Arduino skills are improving all the time, doing wiring on a much larger scale is not something I am totally comfortable with yet. I think mostly I struggled with visualising what the end product was supposed to look like – how the wires are held together (soldering or crimping?), how long the wires are supposed to be, how everything fits in the case with the motor. I found very little on this subject on the Solidoodle Forum, the first point of call for any Filastruder owner, and what I did find looked positively lethal and not something I would want to run unattended in my studio. I read on the forum that the extrusion process takes between 8-12 hours or even longer, depending on how much material you are extruding, and I felt that I needed to be completely comfortable with running the extruder for that length of time without worrying about electrical fires. So what to do?

The first clue I found when I finally had the time to get stuck into the electronics assembly were a few large crimp connectors that had been included to wire up the various switches. If you have been following my blog you will know that I have recently developed a bit of a thing for electronics crimping – albeit on a much smaller scale. Crimps are a great and very reliable way to form an electrical connection, and my interest in the huge variety of colourful larger crimp connectors had been stirred already during a previous visit to Maplin. Learning from my previous forays into crimping, I decided to make another trip to pick up some more connectors as well as the appropriate crimping pliers. If you are going to get involved with crimping larger style connectors, a pair of ratcheted pliers is absolutely essential – and not the right place to skimp; your wrists will thank you for using a decent pair that exerts enough pressure to form a secure connection the first time round. Armed with my new tools and an excellent YouTube tutorial on electrical crimping, I decided to tackle the switch connectors first:

Filatronics04After a few unsuccessful attempts and failed connections, the results of my crimping efforts were starting to improve and the wiring was beginning to take shape. The Sestos controller has screw terminals, so I attached ring terminal crimps on the ends of the wires to create a better connection than I would have by just simply attaching the bare wires. After a few more hours and some wasted crimps, this is what I ended up with:

Filatronics01The thermocouple posed another challenge – its two large prongs did not fit well into the screw terminals, and leaving it sticking out straight would have interfered with the case. I resolved this issue by attaching two more ring terminal crimps on one end and two small spade crimps on the other (hidden underneath badly applied heatshrink in the image):

Filatronics02The final adjustments I made was to use a JST connector terminal soldered to a small bit of stripboard to attach the fan and add female header crimps with housings (taped together with their male counterparts with electrical tape for a secure connection) on the heater wires, as I wanted these two elements to be easily detachable if needed.

Filatronics03I also made the decision not to ‘hack’ the original 12V power supply cable as suggested in the instructions but to use a screw terminal barrel jack instead as it seemed neater and more flexible should I ever need to replace the power supply.

At this point I would like to add that all of these modifications and build strategies were my own and should in no way be taken as gospel – if you do things in a similar way as described here when building your kit it is entirely at your own risk. I am still feeling my way around electronics and figuring out the best way to do things. Similarly, if you feel I made a mistake or want to suggest an improvement, please get in touch as I would welcome any constructive advice!

That’s my Filastruder fully assembled now, including cramming the finished electronics into the case. I have not had a chance to try it out yet, but I can safely say that I have finished the job to the best of my abilities. The Filastruder kit was certainly one of the more challenging things I have done so far, despite great email support from Tim. Can’t wait to get going now!!!


This week I finally got the shipping confirmation for my Filastruder kit. After a bit of wrestling with Parcelforce and paying the appropriate customs charges (don’t get caught out by these if you order stuff from the US – at the very least you’ll have to pay UK VAT!), I finally took my little brick of a parcel home. Unfortunately, it could really not have arrived at a worse time for me – between speaking at the Handmade by Machines symposium last Friday and giving a paper at XcoaX 2015 next week. So, having no time to put the ‘Struder together for another week at least, I thought I’d do a small ‘unboxing’ photoshoot to get it out of my system and inspect the contents of the package in eager anticipation…


It might not look it, but this parcel packs a real punch in terms of its contents…


…most of which I have yet to identify. Most intriguing bit spotted so far: a huge drill bit that has been filed down in order to create the internal lead screw of the extruder.


And here are the contents laid out in all their glory. I ordered the whole enchillada, so what you see here are the both the Filastruder and Filawinder in bits, as well as a complementary pound of ABS pellets. I have already procured 12.5kg of PLA pellets for my research, and I can’t wait to feed these to the ‘Struder.

Another thing I was able to fit in between writing my paper for next week and doing the unboxing was to start printing the additional parts required to make the kit work. There are various ways to set up the Filastruder, and the design of the hopper depends on the way you decide to orient the extruder. My studio is starting to burst at the seams (especially since the Ultimaker entered my life) but luckily there is still a tiny bit of suitable wall space at the back, so I decided to print a vertical hopper mount that takes 2L bottles as hoppers from Thingiverse.


I chose ABS and 100% fill, as this part comes under a lot of strain and needs to be able to take quite a bit of weight – it might have been total overkill, but better safe than sorry. Can’t wait to put it all together…watch this space!

Enter the Filastruder…

So, my birthday has been and gone for another year. Knee-deep in my various PhD research projects, my parents asked me what I would like for a present this year to cheer me up in these stressful times. Imagine their faces when I told them that I would love a kit to build my own plastic filament extruder – probably not quite what they had in mind!

I first read about the Filastruder on Kickstarter about a year ago, before I even owned a 3D printer, and thought it looked really intriguing if a little dangerous. Kickstarter is a great way of finding out about brand new things happening in the 3D printing community, but even though I have been known to back the odd project or two, and so far have not been too unlucky (if you don’t count the month-long delays – Kickstarter is definitely for the VERY patient), most things presented there should be taken with a pinch of salt. Even vastly successful projects like the Form1 often benefit from a period of beta testing, and if your priority is reliability rather than ‘being first’ it often pays to wait a year or two for the technology to get more established and for kinks to be ironed out.

As I have started to experiment a lot more with my printer recently, two things have come to my attention. Firstly, even though PLA is not terribly expensive to buy, the cost soon adds up as I found myself going through filament at a scary rate in the first few months, especially printing out vital components for my research. Secondly, the amount of waste filament also builds up relatively quickly, either from support material or failed prints, which even despite careful planning can and do happen. So, enter the Filastruder. A relatively simple contraption, it is a build-it-yourself plastic filament extruder sold in kit form by the original Kickstarter developers in the US. Since its Kickstarter days, the kit has built up a solid following of DIY filament makers, with some, such as avid blogger Grayson Galisky,  documenting in great detail their filament making experiments on the Solidoodle 3D Printing Community. Of course, since the Kickstarter campaign other filament extruders such as the Filabot and more recently the Protocycler have come to the market – and I am definitely keeping an eye out for the latter. For a UK based option, the Strooder looks promising – compared to the earlier kits those newer machines are definitely heading away from the DIY aesthetic towards a more consumer-friendly look. But for the moment, the Filastruder is the most economical option actually available on the market (the Protocycler and Strooder are both still in their pre-sales phases with release dates estimated for late 2015/early 2016). I have also opted for the spooling kit, as I have run into tangles with loose filament in the past, and the spooler makes the whole assembly much neater altogether.

Making virgin filament is relatively straightforward – buy PLA or ABS pellets or powder in bulk, add a carefully calculated amount of ‘masterbatch’ colourant to the pellets in the hopper of the extruder, mix and start extruding! The pellets need to be completely dessicated to get a great batch of filament, so baking them in an oven for a few hours will really improve quality levels (and of course storing them correctly afterwards). The masterbatch colourant is either sold with the pellets (Colourfabb do this) or you can devise your own methods to make it up – as far as I can tell from the forums almost anything goes, including the addition of powdered metals, wood etc to make exotic filaments. The possibilities that await!!! And it is a lot cheaper than buying ready-made filament. Depending on where you get the raw materials, you could save between 50%-80%.

However, another major appeal for a lot of people will be recycling their failed prints into freshly extruded brand new filament. And this is unfortunately a lot more complicated. The main hurdle to this is really getting your failed prints and waste filament to become tiny granules again – to work in the Filastruder, they need to be less than 5mm in diameter. Plastic can be a really awkward material to work with, and anyone who has ever tried to polish acrylic will know that using powertools of any kind will lead to rapid melting of the material. So apparently will putting PLA in a food processor to chop it up. Industrial plastic granulators are not only prohibitively expensive, but also incredibly bulky – unless you have a massive garage or outbuilding and get really lucky on ebay or a used industrial equipment auction, this will not be an option for you. The second option is using a special plastic shredder. Filabot last year announced the development of their own mini shredder, the Filabot Reclaimer, which looked absolutely lethal in their original promo video and has since been revised to be operated with a hand crank. Within the EU, there is FilaMaker, with their hand cranked mini shredders which look amazingly robust but are expensively handmade to order. These are probably the only viable options for 3D printing enthusiasts at this point – the Protocycler promises to feature its own built-in mini shredder, which would be a massive boon, but who knows if it is still around in a year? If you live in a city with a large industrial base, you might be able to find someone who is willing to let you granulate your waste PLA/ABS. But failing that, it’s really a question I have no answer for as of yet.

Anyway, I will write updates about my new toy as and when it arrives…