Friday, 12 February 2016

Glass 3D Print build surfaces - are not all the same - and a solution for PETG

Glass damage with PET is not fun :-(

In this post I investigated some common glass build surfaces - Starting with a little history to see how we ended up using borosilicate glass and why that's not always such a good choice for 3D Printing.

Many years ago when 3D Printing material choice was more limited than today, enormous amounts of effort went into testing and experimentation of surfaces to print on. Take one look at the RepRap forum archives and you will see that people tried almost every type of material known to man as a 3D printer build surface.

The heated 3D Print bed solved many of these issues, and many people opted for a glass surface to print on.

Over the years I have always used standard mirror-glass as my printing surface of choice. I like that you get a reflection - it's easy to see the state of your nozzle. And I found that they heated up evenly and produced fantastic results.

Over time many people have started to specify the use of borosilicate glass surfaces in 3D printers. While they do have some good properties and can be ground and polished very flat, they also have some limitations that are not always great for 3D Printing.

One of the biggest issues of borosilicate glass is that is can scratch easily, you can dig out chips with a metal scraper and generally cause damage just from normal use.

This is a stunning example of exactly the problem with PETG - Image supplied by Ante Vukorepa

I had plenty of damage - even with small parts, borosilicate glass can get chipped easily.

Often the glass chips are also almost impossible to remove from the printed part.

Borosilicate glass is also starts to get fragile when it gets damaged, and if you decide to print in PET or many of the filled materials you can take massive chunks out of your glass bed.

This problem seems to be something that has not been addressed, other than coating the nice smooth glass surface with uneven gluestick, sprays, sheets of tape or films.

I'm at the point where I no longer want to re-coat build surfaces, mess with glue or generally do anything other than clean the glass every 10-15 prints. And for the last few years using the mirrored-glass surface and some standard window cleaner with vinegar that's worked out well.

My mission was to find a universal build print surface that can also handle more materials and work with IR detection systems.

I have used PEI sheets quite successfully, but they slowly deteriorate and do not seem to like wood-filled and other filled materials.

Tufnol was a fond favorite when the first round of Nylon materials were released back in 2011, Taulman Nylons stick well to natural materials like wood, Tufnol, paper tapes and even cardboard. But they all soon deteriorate too.

Copper FR4  - PCB material works quite well, but it's hard to keep clean and oxidises very quickly.

The only build surface that seems to carry on print after print is glass. But not everything likes to stick to 'normal' or toughened glass.

Recently various surface treatments (for glass and other materials) have made it much easier to print with many different materials, but again they all have a limited life and need to be constantly re-coated or patched up. Airwolf Wolfbite is a good example of a great surface treatment I use it on glass for ABS and PCABS.

BuildTak is another good sheeting material, and for some 3D Printing plastic's is simply too strong - NEVER, ever try to print with TPU materials onto BuildTak - there is a very good chance they will bond permanently to the sheet forever.

Until I had the BigBox I always used standard mirror-glass. Unfortunately because the BigBox used an optical Infra-red (IR) level detection system you can't use a mirror surface.

Adding glue, sheets or tapes does make a difference to the IR sensing Z offset, and a change of glue thickness or a missing patch can cause issues. I specifically wanted a clean sheet of glass to print with as many materials as possible - without damage.

And that's where the trouble started. My borosilicate glass sheet started getting damaged and with the use of PETG materials became quickly destroyed with missing chunks of glass.
It's also quite easy to dig a scraper into the glass and cause a nick or scratch.

The issues are not because of bad or low quality borsilicate glass, in fact E3D commissioned an optical instrument company to custom make the sheets of flat borosilicate glass for the BigBox 3D Printer. It's an expensive, and very good quality sheet of glass!

But I want a solution that does it all, and has no maintenance apart from a quick wipe down.

I made a video talking about some of the issues, and a solution using a tempered glass protector, seems to work exceptionally well - view below or in HD on Youtube here - 

To cut a long story short - I went around the same loop again I did in 2010 testing a lot of different materials for use as a surface - Copper clad PCB (FR4) material - glass fibre sheets, sanded Aluminium, Polycarbonate, Acrylic...

I was about to buy an A4 sized sheet of toughened glass (300mm x 200mm) and I found sheets of thin glass protectors designed for tablets and phones. ('Gorilla glass' - and many copies that are basically a laminate of toughened glass and probably polycarbonate with some coatings. They are also very thin usually around 0.3mm and have a convenient silicone adhesive backing.

The silicone adhesive backing allows it to stick easily to glass without any bubbles.

The glass protector is only 0.3mm think - easy to compensate in the Z offset and still allows you to use the other side if you want clean borosilicate glass for other materials.

I was concerned that the “Oleophobic” coating they add to these sheets to stop fingerprints and repel oils and almost any type of liquid would stop the thermoplastic extrusion sticking to the sheet.

I also wondered if it would bond and not release after cooling down - or worst still totally shatter when using PET/G/T.

My borosilicate glass sheet was already damaged beyond use, so I decided to try out a toughened gorilla-glass screen protector. In hindsight this was not ideal because I now have gaps under my laminated sheet, but I'm still using one of the good properties of the borosilicate glass - that it's very flat.

Some research into common size protectors and the iPad pro was too big but the Surface Pro4 was almost perfect sized for the BigBox bed (2 ) - I found them on eBay for £7.30 (under $10). From a UK seller including free postage.

You can get the exact same thing from China for under $5 with free postage.

I believe this glass laminate sheeting protector is mass-produced in China and laser-cut to size, with various cut-outs for phones and tablet computers -  so it's possible that a keen manufacturer could have some cut to exact size for use in 3D Printing...

In short it seems to be a total success - especially for PETG materials - of any size print - in fact the bigger the better.

PLA and various filled materials all seem to stick well too.

I found that I needed to increase my heated bed temperatures by +5 degrees C and Eureka! - this works.

I didn't use a brim with the PETG materials, and some were over 150mm in size with 45% rectilinear infill - (that's nasty and would take great chunks out of my bare Borosilicate glass sheet).

Some key points - 

The the “Oleophobic” coating should last a very long time, but you need to make sure not to use aggressive chemicals or alcohol based cleaners. It's designed to be easily cleaned with just a dry standard microfibre cloth and nothing else.

I also believe that when the coating does start to vanish, that will only help other materials (like PLA) stick just a little better things - time will tell on that.

You do get slightly less stick with PLA, so for very small parts (20mm or smaller) I found that a brim was all that was required to keep them stuck. It was more of an issue with transparent PLA's I didn't see any issues at all using Opaque PLA materials.

I have not yet done any testing with Nylon or woodfilled / other-filled materials.

Other PET (G/T) based materials Like Refil, nGen, _XT and T-glase all seem to work well.

Why not use a thick sheet of toughened glass? - Yes you could, but it's often not all that flat and many 3DPrinters now come as standard with a heated borosilicate glass surface. That's great for PLA, and with a 'Gorilla glass' style protector on one side, you can still use the other side for other materials if you need.

Thanks for reading,

Please leave me a comment and get in contact. If you try out a sheet of toughened glass - 'protector' do let me know how you get on with it.

Until next time.


Twitter - @Richrap3d
Google+ - RichardHorne_RichRap3D

Tuesday, 9 February 2016

Using 3D Printing for a taste of chocolate nostalgia from the 80's

Bring back the Trio!

In this post I investigate the use of 3D printing to help bring back a favourite chocolate bar - a taste of nostalgia from the 80's.

I did all this back in early December 2015 - just in time for Christmas!


In some rather great timing - literally as I was typing up this post and editing the video - United biscuits (the parent of Jacobs) Today 9th Feb 2016 - confirmed that they will actually bring back the TRIO! - I'm excited as they will be back in the supermarkets in only 1 month.

I was rather surprised and delighted by this news - but at least I got to experience my Trio! before anyone else, and I can soon compare to the real thing once again - read on to see how 3D printing can help bring back a favourite snack.


Trio!, Trio!...
I want a Trio! and I want one NOW!

Back in the 80's as a ~10 year old, one of the best things you could find in your lunch box was a chocolate bar. Maybe also some ridiculous 'Hedgehog Crisps'

Chocolate bar's of the 80's were so much more exciting than now. Many companies experimented with new ingredients and combinations, there was a big battle for the next big chocolate snack. Some were good, some were awful and were banished from the shelves very fast indeed. And don't get me started on the Marathon > 'Snickers' name change.

One Chocolate bar that was a fond favourite for me was the Jacob's Trio!

If you remember the Trio! or the memorable advert's - With a girl called Suzy that had a habit of shouting that she wanted a Trio! now! - Watch at your own risk on Youtube here

Trio's are now a thing of the past, they finally died somewhere in the mid to late 90's and unfortunately because Jacob's biscuits has been acquired by another company, they now only really focus on crackers. That's crackers in my opinion as the 30-year cycle of childhood memories is well upon us for this fantastic chocolate snack.

Another Trio! fan made a valiant attempt to bring back the bar many years back, this was a 2' long 9.5Kg heart-attack inducing monster of a 'snack'. It was not enough to bring it back or interest Jacobs, he even got into the national newspapers. You can see his glorious effort over on pimpthatsnack

Now the only option we have is to attempt to re-create the long lost bar - and that's where a little 3D printing can help.

Please read on below for details of how you can make your own, and you can watch me getting excited about the Trio! chocolate bar in the video below or over on Youtube in HD. -

Project Trio!

At the point of deciding to do this project, I was also reading a really great book called the Great British Tuck Shop by Steve Berry and Phil Norman. I can totally recommend if it you have an interest or memory for snacks of the 70's and 80's past. They mention the Trio, only briefly.

For the mold making - I'm using a great food safe silicone, you mix equal quality of Part A and B.

But first you need to design and 3D print your chocolate bar, I'm using Sketchup above to produce both negative molds and also positive chocolate bars.

After 3D Printing the end result is a set of positive molds that we can make silicone negative molds

Mix up the silicone, trying hard not to get too many bubbles, using a cutting action helps.

Pour and settle - bang it a few times down on the desktop to remove bubbles.

5-6 hours later - after removing the bars you will have a mold ready to use - after a wash and dry first.

For the ingredients, I don't have exact quantities, I was winging it for this experiment.

  • You need milk chocolate - I used Cadbury Dairy Milk. (or chocolate of your choice).
  • One Caramac bar to help make the 'toffee' layer.
  • I used 1 table spoon of salted caramel sauce - because my tastes have changed slightly since I was 10 years old :) -  (in the US that's probably about 1/24th of a 'Cup' - I'm guessing).
  • By some total accident I found that Scottish Walkers Pure Butter shortcake biscuits are exactly the dimensions we need for a Trio! - and they taste great too.

When melting the chocolate and caramac for making the toffee I recommend using a water bath (Bain-Marie)  - this helps gently melt things so you don't cook the chocolate or split the toffee.

A thermometer or other temperature measuring device is a good idea for keeping track of the melting process. If you want to also temper the chocolate before using it, then maybe practice the art of melting at a higher temperature - around 42 Degrees C and lowering the temperature to around 31 Degrees C before using in the mold.

Another directly printed mold in FilaFlex for the Caramel sections of the Trio!

Firstly you want to cast the toffee sections, melt mix and pour into the mold. They will not set hard, as you want a gooey toffee layer in the bar. I put them in the freezer to go hard so they were easy to remove from the mold.

When pouring the chocolate, just focus on getting some in the molds first.

And then give them a tap on the table, that will settle, level and release most of the bubbles.

You will see that on my Trio! bars I had a few bubbles remaining around the Trio text, so make sure you give them a really good tapping.

Add the toffee pieces and a little more chocolate, try to make sure the toffee does not move around too much.

Then you want to add the shortcake. Depending on how much chocolate you have used, you may need to reduce the height of the shortcake.

Finally cover in more chocolate and level off with a blade or spatula.

Leave to set in the fridge...

Then eat them all with a nice cup of tea.

Food / Body Safety of 3D Printed Materials - 

This is sometimes a confusing area. Some materials for 3D printing are being advertised as 'FDA approved' Food and medical 'safe'. That may or not be true in their manufacturing process to make the filament. But please remember that when used in a 3D printer, being melted in a hot-end that may have many other contaminated materials. Along with a brass nozzle that probably contains lead. It's worth keeping in mind that the FDM printing process is not a 'safe' method of producing objects that come into contact with foods or your body.

3D Printing out a cookie cutter or even a cup or spork to use once, maybe okay. But remember that 3D parts are made of layers of plastic, with lots of indentations, grooves and micro-holes where bacteria can grow or contamination can collect.

You also can't just put most 3D printed objects in the dishwasher, they usually melt, deform or discolor.

I was and still am investigating the use of TPU - thermoplastic elastomer (polyurethane rubber) for foods and mold making.

I know I had some colour leak out of a bright green FilaFlex mold when I was doing soap and candle making. (that was at higher working temperatures of around 80 Degrees C) This indicates that it may not be such a great idea to use it directly for food based products.

I did make a Filaflex Trio! bar mold (shown above on the right), in clear material, but decided to not to use it directly. And as you see above I also used white PLA printed Trio! bars to cast a food-safe mold. I feel happy with PLA as it's only in contact for a short time while the mold sets (5-6 hours) and the mold can be fully washed out with warm soapy water before and after use.

Other flexible materials are also available, Ninjaflex, Semiflex and plasticized (soft) PLA.
You can buy FilaFlex from Recreus. This material is flexible enough to allow easy removal, but also holds it's shape well. It's also ideal for temperatures of melted soap, candle-wax etc.

As ever, be aware and make up your own mind before using 3D printing for food or body based projects.

3D Model files - 

You can find all the source and model files for this project over on my Youmagine page here.

Thanks again for reading, please share and feedback on this project or anything else I'm doing.

If you would like to see more of these experiments / tutorials or examples of practical/fun uses for 3D printing - do get in contact and leave a comment below -

Until next time, I have three more Trio! bars to eat, and no you can't have one.

And remember to look out for REAL Trio! bars back in the UK shops in March 2016 - Can't wait!


Twitter - @Richrap3d
Google+ - RichardHorne_RichRap3D

Monday, 11 January 2016

Building and using the BQ Hephestos Version 2 3DPrinter

It's the start of a new year - 2016 - Sales of 3D Printers are still on the increase.

In this post I'm looking at the new BQ Hephestos V2 3D printer kit -  (Pre-production version).

This is an easy-to-assemble 3D Printer kit aimed at individuals, makers, and the education sector.

Right now if you look around at some of the big 3D printing companies, it's all bad news, redundancies and re-organisations, shift in consumer demand, lots of excuses that 3D printing is not doing what they thought it would...

But that's just not the case for many smaller and middle-sized 3D printing manufactures as demand ramps up and things start to turn from more than just a hobby.

Some companies have so much demand that even the use of '3D Print-farms', making parts to make more 3D Printers, are not enough to keep up with increasing sales orders.

BQ is one of these fortunate companies with orders outstripping manufacturing capacity. They told me recently - "something was needed to increase and grow how many 3D Printer kits could be sold to an expanding and global audience" - Sales are good for BQ.
Prusa i3 - Hephestos Version 1 - Image from

The original BQ Hephestos (Ver 1) is an Open-Source 'RepRap' made out of many 3D Printed parts and closely based on the popular Prusa i3 configuration.
BQ WitBox - Image from

BQ's previous 3D Printer (The Witbox) Is also Open-Source, aimed at a higher overall specification and bigger, enclosed build area.

BQ have merged many of the best aspects of Witbox and Hephestos V1 and incorporated them into the Hephestos Version 2 (Kit) design.

I didn't have the opportunity to try out the Hephestos V1 or the Witbox, but I am familiar with the design. This post and video is about my experience using a pre-production Hephestos V2.

BQ have offices across Europe with a remit for education, and a willingness to be Open-Sourced in all they do. If you take a look over at the BQ 'Do It With Others - DIWO) - you can start to see what drives BQ forwards and how 3D printing is playing a significant part in this educational - maker direction.

One key aspect of the Hephestos V2 is that is uses no 3D Printed parts (well, it has one 3D printed fan duct, that's all). It's still all Open-Source and runs a customised version of the popular Marlin Firmware. It also still looks a little similar to the Prusa i3, but when you look closer it's had a completed re-design of the motion system and control electronics.

The Hephestos V2 is a great machine, as long as you take into account the limits of a cold build platform (not a problem for FilaFlex).

My introduction video for the Hephestos V2 explains more about the machine and what you can do with it -

Hephestos V2 Introduction video - (November 2015).

Please read below for further details and comments. An updated summary of the build, including some further enhancements and examples of printed parts from the Hephestos V2 can be seen my following video -

Hephestos V2 Update, pre-production build and machine upgrades -  (December 2015).

Kit Packaging...

Although the packaging is almost completely cardboard, it adds a considerable weight (pre-production kit packaging - things may change for the final version).

This is an ideal flexible plastic printer - 

If the Hephestos V2 printer was designed for anything in particular, I would say it was for printing in Flexible filaments like FilaFlex, NinjaFlex and SemiFlex - TPU / TPE materials.

The excellent extruder design allows 1.75mm flexibles to be printed very fast and accurately.

Certainly not everything is quite perfect, and I'm sure some of the pre-production issues will be ironed out before the launch and production shipments.

With the Hephestos V2, I'm not convinced that such a large build platform without a heated bed makes complete sense. But I do understand the benefit of having a low powered unit, for education sectors.

I have smaller delta printers (including my own 3DR design) that have no heated bed, I really like them, my children enjoy using them and they make sense because they have a small print area.

There is also no easy way to add a heated bed. The power supply can not deliver enough power and there is no easy way to mount either a PCB heater or a flexible silicone heat pad.

The electronic's do have options for a second extruder and possibly a heated bed, but the connector is not fitted on the version I have, so it will be interesting to see if BQ have a plan for customers that want or need a heated bed option.

In the video I show some of the key steps to build up this printer. It's easy to do, and the instructions are very clear and easy to follow.

I have also been experimenting with how well the Hephestos V2 prints in FilaFlex (It's fantastic).

All the great FilaFlex colours - Printed on the Hephestos V2 3D Printer.

I have also been printing a lot of moulds for various materials, including candle wax, soap, chocolate and clay. - More on this in a future blog post - keep a look out.

Casting materials (and foods) seem to release well from FilaFlex, so it's an easy process to design, make and use a 3D Printed flexible mould from the Hephestos V2.

Take a guess what this is...

This is a shock case for a Kindle Fire 7" tablet.

Designed to be easy to print and to protect the unit.

Inductive level sensing - 

The back of the glass build platform has a thin sheet of steel bonded to it. This is so the inductive sensor can accurately check the distance of the print nozzle to the build surface.

BQ have produced their own inductive sensor, it's fitted firmly behind the extruder and hot-end. Inductive sensors are good as a distance sensor because they don't require any physical contact to the surface.

You have to calibrate the sensor by setting the nozzle just touching the build plate and storing the offset distance, this is really easy to do with guided instructions on the LCD screen. My only frustration with this process is that you can't just easily change the offset distance value. The menu options make you do the level and distance process over again if it's not close enough for your first layer. (This was on the pre-production version - things may change before the machine is shipped).

Overall it's very accurate, I have not seen any deviation in sensing distance at all, the inductive sensor is a good solution for this type of 3D printer, it is not concerned by any surface coatings you apply or the type of material used for the build plate - as long as you have a flat sheet of steel underneath for it to sense. This is bonded to the back of the glass build plate for the Hephestos V2.

I really do like this printer, the kits is very easy to build and you get fantastic print results due to the very high quality parts that have been used. It's a very fast printer and the extruder is the best I have ever used for flexible filaments and it's also great for normal PLA plastic too.

BQ have not cut any corners (apart from lacking a heated bed, but we can forgive them for that), they have made a machine for a sensible price, using high quality parts that will last you a long time and produce great printed parts.

Final thoughts (about buying a 3D printer - of any type) - 

In a market that's racing to the bottom, it's not going to be easy for any company to stand out. Alternative 'generic' 3DPrinter kits can be sourced for a lot less than most machines on the market today, but do ask yourself if they are any good before parting with even a little money.

For me we are seeing an interesting time in Desktop 3D Printing, even more companies than ever are deciding to make their own model - a slight tweak here, a different configuration or size. But to the average user that wants a 3D Printer (or should I say they want the capability of 3D printed objects at home) - it's starting to get really confusing who and what machine to buy.

We are going to see many more revisions before the 'ultimate' desktop 3D printer is with us, and that will still take quite a long time (many more years) to happen. So my advice is still very similar if you are looking to get into 3D printing -

1) - Have a 'need' for a 3D printer - remember you can use 3D printing services if you need special parts printed (gold, ceramic, very high quality details).

2) - If you just 'want' a 3D printer - then that's fine. Spend some time thinking about what you may do with it, be sure to understand how long the print process takes (many hours most of the time).

3) - If you have a need and want to operate a desktop 3D printer, investigate if  plastic (FDM) printing or SLA (resin) based machines are more appropriate for your application.

4) - Value for money - this depends on many factors, how much support you wish to get from the manufacturer. If you want a kit or a ready-built machine, and how much you want to spend.

If you spend ~£150 / $200 / 200 euro on a 3D printer - even a kit, you may be disappointed, frustrated or need to spend more to get it working well - ( Like buying a new hot-end and/or extruder for example ).

5) - As a guide, it's still sensible to be thinking about spending around £500 / 700 euro / $750 as a minimum for a machine that can provide quality prints using good components in the design.

If you plan to spend more than £1500+ / 2000 euro / $2000 For a 'desktop 3D Printer' have a very good reason and do plenty of research first.

I have many more projects, 3D printer and material reviews along with development experiments coming up for this year, stay tuned and do get in contact.

Catch me on Twitter @RichRap3D

Or over on Google+

Thanks for reading, see you next time.