The QuickFill comeback

By | Design, Software Updates | 4 Comments

QuickFill is currently one of the most underutilized features of the Robox. The idea is that the Fine nozzle prints perfect perimeters, whilst the unseen infill is printed fast using the larger fill nozzle. The original Robox single material head has been designed entirely around this idea, with the ability to quickly switch between the fine and fill nozzles, both being fed the same material from the same extruder.

2 owl

Why was QuickFill removed from the default settings?

The simple answer is that we were not happy with the quality of output from the Robox when QuickFill was enabled. Artifacts being left after the fill nozzle closed meant that the parts didn’t look particularly good at all. It’s always been possible to create a custom QuickFill profile, but we don’t want to create a default that locks people into rubbish prints! With recent changes to the post processor, improved control over the needle valves, and a little bit of Maths thrown in for good measure, we’ve improved print quality and subsequently made QuickFill a viable tool once again.


What are the benefits of QuickFill?

QuickFill can speed up a print without sacrificing quality. If we use Slic3r to create parts, we can utilize a feature called ‘infill every # layers’. This allows for us to only print infill on a specified layer, in most cases, this is every third layer. The amount of time saved can depend entirely on the size of the part, a 25mm (1 inch) cube will take approximately 2hours and 20 minutes using the fine (100 micron) profile and 1 hours 50 minutes using a QuickFill profile combination of fine (100 micron) perimeters with fill nozzle infill every 3 layers. When printing something slightly larger, and more complex, QuickFill really comes into its own. The owl pictured below takes approximately 12 hours and 10 minutes, using the QuickFill profile combination of fine (100 micron) perimeters with fill nozzle infill every 3 layers, the owl only takes 5 hours and 10 minutes!

1 owl

When will I be able to select Quickfill in Automaker? Will it replace the default profiles?

We’re hoping to release the QuickFill profiles in an upcoming version of Automaker. We don’t intend to replace the default normal and fine profiles, instead we want to implement a Quickfill on/off slider, much like the Raft and support selections to provide options to those who want them. If you want to help us test our current QuickFill profiles, or you want to create your own, head over to our QuickFill Community Testing page on our forums here.

Dual material Robox upgrades are close

By | Chris Elsworthy Design Blog, Design, Materials, News, Prototype Build, Software Updates | 28 Comments

It has been a little while since we last updated you on where we are with the dual material head, but I hope we can be forgiven for the lack of communication – especially as the real reason for this is that the entire CEL team have had their heads down testing the latest prototypes for faults, fact finding, and future proofing.

But we feel an update now is timely, as we are getting so close to being able to launch the new head to market; we really are almost at the end of what feels like a very long development journey.


Software Update

RBX01-EX_2_EXTRUDERSSoftware is the biggest challenge as there really isn’t a good dual material system on the market to use as a template and both the slice engines we currently use aren’t able to deal with dual material prints in a consistent manner. This means we essentially had to start from scratch creating innovative software for this particular hardware development.

Before we get deep into the dual material system progress it’s worth mentioning what we’ve done to the software in preparation for this momentous task.

For the release of AutoMaker 1.02 we spent a good amount of time restructuring and developing the post processor.  We gained a much better understanding of where and how to open and close the needle valves, how to integrate the material the needle ejected into the model and how this is going to affect dual material printing.

For the upcoming version of AutoMaker we’ve actually recoded the post processor from scratch, our new level of understanding was restricted by our previous work. This updated thinking has gone into the current version of AutoMaker and is the backbone of dual-material capable AutoMaker version 2.


Right back to dual material and where we are now….

There are a lot of new requirements needed to take a multitude of models and turn them into a multi material print. We wanted the ability to take a 3D model, break it into individual parts and Dual Material AutoMakerallow the user to easily select which part was going to be assigned to which material.

Because most models which require multiple materials will need two or more separate parts, we wanted to add Group/Ungroup and model splitting. The Group/Ungroup and model splitting all have to be combined with the original transform functions so that we keep the parts’ relative position.

Dual Material AutoMaker bAfter we started to look closely at the model files  AutoMaker was expected to handle we found a very high percentage had imperfections – not normally a problem as the slice engines do some simple repair work – but a bigger problem if you’re trying to bisect the data and then add back in the missing vertexes, polylines and faces. We’ve not fully finished this aspect of the software and currently only warn users of the imperfection in the models that may cause problems later.

We now have a software tool that allows us to perform all of the above functions and it’s helping us finish the development of all the hardware.

The development of the software will continue up to and beyond the date you receive your new hardware, but for the first version of AutoMaker which is capable of doing dual material prints we have tried to keep the functionality to a minimum so that we can ensure stability.



Hardware Update

After a few false starts we’ve now got working dual material heads and an innovative second reel holder complete with reader. We are also taking a very close look at the management of the Bowden tubes and cables between the x-carriage and chassis.


We were worried that the differing flexibility in filaments would affect the homing accuracy and thus distort the mapping of the bed giving poor first layers. We’ve made a fancy new cable chain to take the data and power cable and slightly constrained the Bowden tubes which now both reference the X-rails rather than the carriage.. (sorry if this is too geeky), basically we have more stuff to help improve probing and tidiness.

Another thing that only the people in the know would notice is that we have added a USB socket into the build chamber. It’s been there from the beginning – hidden away in the centre of the reel – but we’ve never got round to exposing it. We thought as we were designing a new PCB for the dual reel holder we would add the missing plug so you can start playing with it too.

The needle valve system has been extremely important in the quality of printing that Robox can achieve when talking about single material prints, removing all the stringing from one island to the next. When you use this innovative system with a dual material head it starts to become even more impressive.

All the dual material prints we’ve seen from other manufacturers have to include some sort of wipe tower, ooze screen, long cool down / heat up period, or purge sequence. None of this is necessary if your print head can stop the flow of material at its nozzle tip like the Robox can. Our very first print from the new system was one of the best we’ve seen. So now we’ve worked out how the system works, we’ve got working samples, and we are producing great print results.RBX01-DM_2HEADS_1OPEN

The final step is to ensure that all materials will flow through the new hardware smoothly, tidy-up the interface and test, test, test… I can’t tell you how big a chunk of work this last section is. We need to try every single combination of user input and output and not just once, on one machine. This needs hundreds of times on many printers by many different people to ensure that all the logic we’ve put into the software and hardware doesn’t crumble when given to a user who hasn’t been involved in the development.  With this in mind we’re now at the stage where we’re asking some of our customers who have placed pre-orders to beta test the first batch of hardware from the final production line.

So all in all, we’re really, really close.  If it sounds like we’re no further forward than last time – bearing in mind we finished our last update by saying we were “testing, testing, testing” – please be assured we are.  But after testing comes tweaking, and then more testing!



So the final date is……???

We can’t say for sure, at the moment things are looking great and we hope to be able to launch the dual material head end of November, as per our last post.  If not then, early December.  But to reiterate the sentiments we made last time, we will only release the head if we’re confident it’s going to produce excellent results for everyone.


And lastly….

THANK YOU once again for your unwavering support, stick with us, as in just a couple of months the Robox 3D printer will make its mark as something completely different from anything else out there on the market – and we want you there with us to enjoy the ride.


Oh you want a picture of a dual material print? Well….here is a sample of what we have to come.


Strong Shapes – A Robox 3D Printed Solution to a Vehicle Design Issue

By | Design, Robox User Blog, Stuff and Things | 2 Comments

Hi I’m Jack and I would like to share my experience of using my Robox, with regard to a component I designed and am now selling online. The guys and CEL were happy to hear my story and suggested I write a blog. Below is a short story of how I came up with the idea and what I did with it:

I first had the thought of the product when I was a passenger in a brand new Ford Fiesta, having a brainstorm of what things around the cabin I could 3d print. I knew that the vehicle was the U.K’s most sold car, and so there could be real demand If I came up with something significant. I initially focussed on a phone adaptor, located near the handbrake, but soon after noticed this very large gaping hole, situated just behind the handbrake lever. Essentially the plastic handbrake had a flange at the back, instead of a leather gaiter, which in the downwards position covered an opening. However in the upwards position this flange lowered into the car, leaving a ~5x3cm opening. This opening was situated approximately just down from the driving occupants top left hip, meaning if an item, such as car keys fell out of the persons pocket then it would fall down into this hole. Recovering items that were lost could not be done at short notice, requiring the centre console to be partly removed using specific tools, to gain access to the lower regions of the car. Imagine losing your car keys, what would you do!

At that point I had my Eureka moment! I could design a cover to stop this from happening. The next step was to confirm no one else had thought of the solution by having a look on the internet, using the car club forums. I came across a page, containing circa 100+ comments from people having lost valuables. Items from keys to lighters had been said to have been lost, causing real aggravation to customers. The section of the website had been viewed over 12,000 times, so at that point I knew there was demand for a solution.

Around work commitments I developed a concept over a 7 month time frame, using friends to trial my designs. I think to date I produced around 30 prototype designs, all designed on my home computer using a professional level cad program. This is where I really got to know my Robox and appreciate its capability. I opted for a third party filament, based on comments from the Robox community, choosing Colourfabb XT black and was impressed how easy it was to optimise through the Automaker software. I spent a long time making tweaks before coming up with a specific architecture, just optimising it around the printer, such as print time and material costs. Finally after some additional trialling I finalised my design and had a product ready for launch!


Though initially intending to sell the product on eBay I opted to create my own website. After coming up with a suitable name; strongshapes, FYI my surname is Strong, I set about creating a website that had a professional image to potential customers and also support them after purchasing. I’m really happy with the results and am continually developing it further. The website went live in July and I am happy to report it is a success, people are actually buying my product! I am currently looking at ways of advertising as I find sales are relative to social media. Ultimately though I think when I make the costs back from purchasing the printer I will have reached a personal milestone. I don’t think it will be too long and additionally I have a local university offering to 3d print my design, moving up from prototyping, all possible thanks to my Robox!

Finally just to say that if you know anyone who has a Ford Fiesta 2012+ to have a look at my website and see if they could use this product. Also If at least I could ask readers who are on social media to share my website, and help support this Robox enthusiast!

Thank you.


Mod: Flexible filament with Robox

By | Design, Prototype Build, Stuff and Things | 14 Comments

A modification which will allow Robox users to print with flexible filaments such as Ninjaflex. Bowden tube printers struggle with flex materials but the patented needle valves on Robox give excellent results.

In future we will produce extruders with this part as standard, this mod is for reference only. We are working on the details of an upgrade service which when purchased will allow your Robox to be updated with the latest available parts.

WARNING! This mod is not supported by CEL or it’s resellers, any changes to your Robox or changes which affect it will not be covered by warranty.

WARNING! Never remove or connect any plugs or covers while power is supplied to Robox via the mains plug or by USB.



This mod prevents flexible filament (tested with Ninjaflex) from distorting in the gears of the extruder. Details of the mod are deliberately sparse as this should only be attempted by skilled people familiar with the required tools and with Robox.

Do not open any casing or remove any cables while power is present. Leaving Robox completely unplugged – including USB and mains cables – for at least 30 mins will allow capacitors to discharge and reduce the chance of a shock.


It requires printing of the small part shown above (flex_material_part.stl – print some spares before removing your extruder) which fits between the 2 filament drive wheels without stopping the filament from being driven by these wheels. To fit the part some of the extruder housing needs to be cut away from 2 parts of the extruder housing.


Above: Top part is modified, bottom is original.

Below: It is very important that the filament is not trapped by the new part. Use a piece of filament put through the central hole to keep the new part in place while marking prior to cutting.


The depth of the cut is equal on each side. Take care to keep cuts square and neat, we used a grinder similar to a Dremel.


Remove small amounts of material and test the assembly often to ensure the fit is good.

Images below left for reference.


IMG_0984 IMG_0983  IMG_1004 IMG_1003    IMG_0994   IMG_0991 IMG_0990 IMG_0988

BBC micro:bit CAD Resources – Kitronik University

By | Design, Education, Stuff and Things | No Comments

An excellent resource and introduction to the BBC micro:bit by our partners Kitronik

See the article here:


This Kitronik University resource is part of the BBC micro:bit partnership and features FREE downloadable CAD files in relation to the BBC micro:bit. These downloadable files are available in three different formats and have been made using Autodesk’s Inventor Professional.


The BBC micro:bit CAD Files

The team here at Kitronik have created a CAD model of the BBC micro:bit. We will be using this model to produce many of the resources we will be creating for the BBC micro:bit. We’re sure these models will be useful for lots of applications so we are making them available completely free of charge, as we feel this maintains the spirit of what the BBC are trying to achieve with the BBC micro:bit project.

Students and teachers (and home users) alike are sure to find them a fantastic starting point for projects based around the BBB micro:bit.

The files are available in the following formats (see bottom of page for download links):

  • .iam (Autodesk Inventor)
  • .stl (which can be used in most CAD programs and for 3D printers)
  • .sat

This render was created using Autodesk Inventor Professional and shows the kind of images the files could be used to create.


Autodesk – FREE Educational Design Software

Clearly to use these files you will need some CAD software. Autodesk provide their professional software free of charge to Education and Home user which makes it an ideal choice.

Autodesk gave the following reason for providing such easy access to it’s products:

‘Closing the skills gap starts in education. Autodesk are tackling this by providing schools with common access to the same advanced technology being used by industry professionals today. Autodesk provides schools, students and teachers with free access to its professional 3D design software. This will enable educators to introduce design thinking into our classrooms allowing students to imagine, design and create a better world. Using these tools to learn how to solve real-world challenges in new creative ways will be the perfect preparation for our the next-generation workforce. Equipping them with 21st century skills to meet industry demands and advance economies worldwide.’

You can get a number of Autodesk software products for free, for both educational institutions and home use, check here for more information.



The above images are renders of the CAD files using Autodesk 3DS Max and Autodesk 360

How Kitronik Are Using The Design

We’re using this design to create a few resources which we thought you’d find useful.

One example is this cool poster highlighting the features of the BBC micro:bit (A4 download available below):


Another is this useful mechanical datasheet (download available below):


Using The Files When Creating 3D Printed Case Designs

Having the BBC micro:bit as a CAD object is incredibly useful when creating 3D printable (or laser cut-able) case designs. It means you can create your design with the knowledge that when you come to manufacturer the final design the BBC micro:bit should fit perfectly!

An example use of this would be when designing a case, like this one designed by Chris Elsworthy from CEL Robox to create this great 3D printed case design.


Having the renders available makes this job much easier, and ensures an accurate result. We will also be using the files in our own models and case designs.


You can see the full BBC micro:bit – Kitronik University Course here.

Where is my Dual Material Head?

By | Design, News, Prototype Build | 7 Comments

UPDATE! Check this one out


At the beginning of the year we estimated that we would be shipping the Dual Material Head at the end of Quarter 2, 2015, and many of you have pre-ordered your heads so you can take full advantage of all the extra functions Robox brings to 3D printing; for this we wholeheartedly thank you.

We are truly sorry for the delays so far, but development of new technology is an unpredictable business and we really want to make sure that we give you the best product possible when it is finally released.

What has been done so far?

We are over 90% on our way to seeing the Dual Material Head enter mass production, but as most of you will know the last 10% is the trickiest part of the whole process.

We have been working very hard to ensure that when the new technology is released it is as near to perfect as possible, to ensure our customers enjoy a smooth and faultless experience, so we are not about to rush the last 10% when we are so close to achieving something very special.

Even before we had released the first version of Robox we had already started on the dual material system ensuring as much of the work was included in the original unit as possible.  We had to ensure the Robox had the potential to be a multi-functional platform from the very beginning.  So we included the location features for the second extruder, a path for the filament to reach the head and all the electronics to drive the system.

Robox then exploded onto the market and we have spent the last 8 months making sure we could provide the necessary support to our customers, getting all appropriate systems in place and ironing out the few teething troubles which should be expected with the launch of any new piece of technology.  To say the last few months have been somewhat of a whirlwind, while very exciting, is an understatement.

We are now at the point where we think Robox is the best machine on the market, and feedback from our customers has been overwhelmingly positive.

During this time, work on the Dual Material Head has continued, and we are almost at the point where we can give the nod to the production lines to get started – almost, but not quite.

There are 4 distinct areas that need finishing before the Dual Material Head can be released; the head itself, Cable and Bowden tube management to the x-carriage, the second SmartReel reader and holder, and the firmware and software to handle all this.

  • The head; we’ve designed, manufactured and tested 2 new ways of sealing the needle valves into the head; we’ve temporarily abandoned one of these designs as although it showed great promise it requires a much longer development time; the second is a more subtle change to the existing system which is now being life-tested to ensure smooth continued operation.
  • Cable and Bowden tube management; this has been designed to ensure it is backward compatible. Stage 1 samples have been made and tested and we are now waiting for the production parts to be built and tested.
    Cable Managment
  • The second reel holder; on the surface this seems the easiest of tasks but has been the cause of some delay. The mechanics are simple enough but some of the electronic parts required have a very long lead time and even samples of them are taking a long time to procure. The factory is now making prototypes and we should see them at the end of the month.
    Dual Material Reel holder
  • Firmware and Software; the firmware only needed subtle tweaks and is all but complete. The software is more than 50% of the entire task, not only do we need an intuitive way of allowing the users to select and assign extruders to objects or tasks, but the structure to enable these to be processed into the final Robox commands. Our two senior software engineers Tony and Ian have been diligently integrating all the features required to support 2 materials into the interface, working closely to ensure that the gcode generator and post-processors have all the variables required to create the correct Robox commands. This has been completed to testing level and we now have a development version of AutoMaker with material assignment, Dual Material Head support. Ian has taken this opportunity to scratch-build the post-processor to improve its output and to handle multiple extruder support. This is the final part of the software development required to start multiple material printing.

What’s not finished yet?

Testing, testing, testing.  As you can see from the above, in essence the Dual Material Head is virtually complete and we are at the point where we could press the button on the launch of the head.  But we want to be at the point where we are very confident about performance and life, and as such we are spending the next 8 or so weeks undergoing a rigorous testing process.  This includes driving the heads with the software as a user would, testing the production samples of the cable management system, and testing the latest version of the dual reel holding designs.

In addition we will conduct final testing of the software to ensure we’ve covered all possible scenarios, and while the gcode creation is well into its development we need to make sure it has all the functions and stability essential for final release.

So what is the new ETA on the Dual Material Head?

So the good news is that while we’ve not met our early estimated date, we are confident the new head will be released to the public in November 2015.  But please note this is an estimate, what we don’t want to do is promise an exact date and then find we’re a few weeks out. As well as the immense amount of testing we have to allocate time for tooling and production.

Why is it worth the wait?

At this point you need to remember why you invested in the Robox in the first place – this all comes back to the reason why Robox fills a gap in the current 3D printing market.  Robox will be the ONLY 3D printer out there capable of additional functions such as dual material printing, paste extrusion, drag knife cutting, laser cutting and more. The Dual Material Head is the first of many heads we intend to release.

You’ve already bought a 3D printer which arguably produces better prints than anything else on the market in a similar price bracket, or even at a higher price point.  And soon you’ll be able to produce prints in different colours and with different materials which expands the range of what can be printed enormously.

Printing in different colours means you can print coloured logos, patterns, light pipes and more, the list is endless!
2 yoyo2

Creating parts with multiple materials which have different properties allows you to print items that have a hard core and a soft over-mould; for example, a bottle cap with a rubber seal or axles with bearing surfaces.

The Dual Material Head greatly expands the range of replacement objects you can print, such as buttons, phone cases, handles, chair pads, screws, plugs, watch straps, hinges, hearing aids, over-moulds etc.

And perhaps most importantly, users will no longer have to consider whether the design they have found is suitable for 3D printing – you’ll be able to download any model from repositories online and be confident of the final print.

Printing with two different materials – such as ABS and HIPS at the same time – means you could have support material which is more easily removed as it is different to the material used for the actual design.

Robox hardware doesn’t have the same caveats that others do; we don’t need to retract, cool or wipe the nozzles, we don’t have to build time-consuming, wasteful towers or shields to reduce cross contamination of materials.

The needle valve system ensures that only the material being printed is coming out of the nozzles. The needle valve system is good on single material prints as it allows multiple nozzle sizes but truly shows its potential during multiple material prints.

The user will not have to consider every aspect of a multi-material print, the interface we’ve designed is intuitive and the workflow simple to use; even handling situations where the material attached to the hardware doesn’t match the project. A simple change to the extruder which will be released very soon will introduce the ability to print a more diverse selection of materials, and as such, prints.

In conclusion, the Robox will be able to produce the best multi-material prints of any FFF printer in the market, development is well under way.

Thank You…

This all said, some of you have already placed your pre-orders for the Dual Material Head and are getting impatient.  We totally understand, and hope that this explanation on our part helps to put your mind at rest and retains your confidence in what we are trying to achieve. We’re extremely excited about what the Dual Material Head brings to the Robox – the wait will definitely be worth it.

UPDATE! Check this one out