Stick with it... 3D Printing - Print Bed Coating Materials. Part 1

Hello everyone,

We are looking at 3D print bed surfaces and coatings in this post.

This blog post contains an overview of the more common materials used as a 3D printing build surface. Plenty of tips and advice and we also take a look at a how well a bed coating 'skin' called 3D EeZ performs. And it's also a general call out for any other bed formula's or materials to test or experiment with. I hope you find it useful.

Most aspects of 3D Printing take time and some experimentation to refine what works best, if you already have good solution for the print bed surface, skip down to the end and take a look at how 3D EeZ worked in the test.

Will it stick?
For all the years I have been using home 3D Printers, one fundamental aspect is still causing frustration for new and experienced users alike. The issue here is getting the first layer to stick, then completing a print without unsticking, flying off mid-print or bending (warping) corners on objects.

It's now quite uncommon to have a non heated build platform, unless you choose to do so by turning it off.

One very common solution for printing in PLA, heated or not is blue 3M painters tape (multi-purpose #2090). To be honest this is fantastic for smaller machines like my 3DR design, Tantillus or the new Ultimaker 2 Go.

Have some 3M blue tape handy, it's still a really good solution for many objects and materials.

When printing small (or really tiny) and highly detailed parts, I often opt for blue tape and no heated bed. You can run cooling fans at full power and achieve highly successful and impressive prints.

Blue Painters tape - ideal for detailed tiny printing, using no heated bed.

A sheet of thick (~10mm) acrylic is another common choice when not using a heated bed, just be careful not to drive the nozzle into the surface as it will melt.

If you need to print bigger parts over ~100mm+ wide then a heated bed will help keep things flat and limit edge curl up on parts.

For PLA, PET and various filled/mixed materials (WoodFill / BrassFill etc.) you can often just use a clean sheet of glass as a suitable build platform.

Before trying to print anything, do check your nozzle is flat and vertically straight with the build platform. You also need the build surface as level as you can possibly achieve. A low cost mechanical dial shown above can really help get things set up quickly.

Top tip for using plain glass is to keep it clean, remove grease and fingerprints with alcohol or I really recommend window cleaner containing vinegar.

I use mirror glass. Handy sized 200mm x 200mm mirror tiles can be found at ikea (SÖRLI) The mirrored surface helps with checking the first layer is going down well and it heats up fast and evenly.

PLA sticks perfectly well to uncoated clean glass. If all I printed was small(ish) PLA parts,  I would always just use clean glass. The trouble comes when you want to start printing bigger (over about 170mm wide in size) or with other materials.

PVA or glue based washes are also commonly used on top of a heated glass surface. For a few years now I have most often used a 1 part PVA (elmers glue) and 8 parts water solution, it provides an excellent surface for PLA and PET printing.

PVA wash. Apply when bed is cold and leave to fully dry.

PLA being printed on a PVA coated heated glass surface.

Getting the first layer distance correctly set takes some experimentation. After a number of prints you should be able to tell when it's going down well. 

The print above is just on the borderline of being very slightly too close for the first layer. But you are usually better off being slightly compressed than not enough. Just 0.05mm more or less can make a big difference.

And this first print layer above was just slightly too far away from the build platform. 

When you print the first layer, you want to see the perimeter and infill join up, no gaps or plastic riding up in ridges. When you print the second layer, you do not want to see the hot end nozzle ploughing through the first, it should again be nicely layered on top, not digging into the first layer of material.

Gluestick, most brands seem to do a similar job. Mark Durbin kindly brought me back some genuine sticks of Elmers glue from the US. And while I remember Mark has a stunning lithophane generator up here, do check it out.

If mixing up glue and painting it onto your build platform sounds like too much work, you can also use a gluestick. Again applied when cold and leave to dry, I also recommend heating it at 60 degrees C for 30 mins before doing the first ever print.

Hairspray is another common choice. Not all brands work, you need to look out for 'extra hold' with ingredients including Vinyl, Acetate and Copolymer. No wonder it holds hair in place.

Hairspray gives a nice mottled surface to the printed object, less shiny than just sheet glass alone. I have had a few materials, like Laybrick & Laywood stick so strongly when using hairspray that the glass cracked trying to remove them, so test out your spray with small prints first.

Hairspray is easily applied and works remarkably well for PLA and PET. It does required frequent reapplication as some will stick to the surface of most printed objects. 

Nylon is a little different.

Tufnol / Garolite is often used as a heated bed surface for Nylon based print materials.

Nylon likes cellulose based fibres, so it will stick to almost anything made from wood, paper, compressed cardboard or impregnated cotton sheets (like Tufnol). The main issue is still massive warping forces as objects are printed, they will rip up or bend cardboard. I found Tufnol to be the best surface for Nylon.

Blue tape and wooden surfaces also work with various degrees of success for Nylon printing. 

A few quick notes on ABS - 

Printing with ABS is a little more tricky. You really can't print without a heated bed, and that wants to be running up at 100+ Degrees C. to get the best results. Only a fully heated chamber really provides the best way to print large ABS parts (over 200mm wide).

ABS just wants to warp and generally curl up, you really only have two choices -

Mechanical grip - 

Often a perforated board (Fibreglass FR4/FR5, shown above) can be used, this will transfer heat and also provide mechanical grip by the first layer (or raft layer) of ABS being forced into the many small holes.

Using this method a sacrificial 'raft' is often printed fist, then the object on top, when printing is finished the raft is peeled off. The raft also helps to minimise warp in the printed object, but objects printed in ABS over 150mm wide will almost always warp or have curled up edges.

The second option, and now the most common is to wipe on a solution of ABS dissolved in Acetone. This is often called ABS Juice, companies like LulzBot provide nice instructions on how to make your own 'LulzJuice', thanks guys.

I recently tested out PEI as a (heated) bed material (sample courtesy of Robox) , this is quite expensive but does work very well with almost all existing thermoplastic materials being used for home 3D Printing today. I highly expect many more 3D printer manufacturers to adopt the use of removable PEI sheets as the next generation build platform, for home and consumer machines.

PEI Sheet - expensive, but performs well.

Now to test 3D EeZ - 

Lets get on to the testing of 3D EeZ. Various materials, stickers, solutions and methods have been tried over the years. Often proving successful with one or two types of 3D print material, but usually not many or all.

Before Christmas, Tony Gaston mentioned he had a new paint-on 'skin' for 3D printer platforms.
It's not an 'open formula' and I still don't know what's actually in it, but I was intrigued by the video's and Tony is a nice chap, just trying to get people to try out his product, so decided to try it out.

Having to import the solution into the UK took some time and proved costly to get hold of, but anyone in the US should now be able to get it quite quickly to test out. I was unsure if it would work better than other home made solutions, but why not give it a test.

My existing heated bed - 

This is what my typical mirror glass build surface looks like, this particular one has a mix of PVA wash, gluestick and some hairspray for surface 'repair'. It would have printed around 100 objects to end up looking like this, and it still works fine for most materials apart from ABS and it's limited to small sized Nylon parts only. 

It's very easy to clean in a little warm water, this is an Ikea mirror cut to size. ~410mm x ~205mm and it's 3mm thick glass.

It's very thick but easy to apply, just wipe on using the sponge brush. Tony is working on a better applicators and an easier to handle container.

It's recommended to apply three layers, most recently I have dropped down to just two, but for all the testing below I had three layers of 3D EeZ - applied in opposite directions, across, up/down, and across. Each coat is applied cold and heated at 60 degrees C until it dried, about 10 mins.

It smells a little like glue, but you soon realise it's not just PVA in this mix.

My first print was just a few tiny cubes (10mm tall) in PLA with a 'brim' (skirt) to test how well it would stick. Normally after printing a PLA part I would knock them off, sometimes before the temperature of the heated bed dropped too low, so I could start another print quickly. That was not going to happen with 3D EeZ.

If you look closely at the picture above, you can see where I tried to pull off the front printed tower, I could not move it, and when I used a pair of pliers I was concerned about breaking the glass. But it did finally come off and bubbled the 3D EeZ coating. First print, first damage, not good, but it was my fault. (I think).

I decided to wait until the temperature dropped. It was a cold day and at 18 degrees C ambient the other two snapped off firmly but without any damage to the surface or part. I realised you must wait for the temperature to drop before attempting to remove the parts.

I would also later work out that it can be too cold with 3D EeZ, see later on below.

At this point I was concerned that if I printed anything too big it would simply not come off of the build surface. But don't worry that's not the case, Just don't use the brim (skirt) function for anything.

I printed various smaller parts (ok, lots of test tree frogs) in PLA, Ninjaflex, Filaflex, Bridge Nylon, Polywood and three different types of PET... And ABS too! All worked, well.

Tree frog Ninjaflex, he is quite tricky to print at the best of times in rubber, but stuck, printed and peeled off perfectly.

I didn't expect Nylon to stick, but it's actually really good.

Bridge Nylon sticks really well (almost too well), above you can see some signs of what looks like damage, but a wipe over with wet sponge and this disappears. Ignore the obvious surface damage in the middle, this was the forced 'hot removal' of the very first print tower. Again it was 'repaired' with a spot of water.

At some point printing different materials (and many more frogs) I didn't wait for the bed temperature to drop, and further damage was sustained to the surface. At this point I decided to work out what the ideal removal temperature was. It turns out that it's around 28 degrees C for most materials I tested. I believe the 3D EeZ coating changes from a semi-flexible surface to a more solid film at around ~40 Degrees C. But that's just from observation during use.

If you leave the printed part on the build platform and let it cool completely, it actually seems harder to remove than at around 28 Degrees C. I found that you can heat up a cooled bed to 40 and let it drop back to 28 for much easier removal of any stubborn parts you have let get too cold.

I now have my end of Gcode set to drop down to 28 degrees C at the end of a print making removal much easier.

It's quite easy to repair the surface, you just drip and smooth on more of the 3D EeZ solution into any damaged areas. After discovering the ideal removal temperature and repair, I managed around another 120 prints all without any further significant damage to the coating. I found a sponge wash over of water every 40 or so prints helped refresh the surface.

Getting bigger (Aria the Dragon)

And bigger - dolls furniture 

And for a really interesting test, the beautiful Astrolabe Tree Ornament by Don Foley

These rings just pop off with a normal scraper when the bed is around 28 degrees C.

Here you can see the nice matt finish you get on the underside of the print, this is particularly nice as it helps to make it look just like the other surfaces. Using just glass you get a very shiny base that can often look out of place.

The shadow rings you can see in the background will lessen next time the bed is heated, or a wipe of a wet sponge seems to help make them vanish.

I have also successfully tested ColorFabb WoodFill, BronzeFill, CopperFill and GlowFill, they all print perfectly and can be easily removed without damage to the surface coating.

I printed out bigger and more complex parts, all without the need of any form of 'brim' or skirting option.

A big Bonsai Planter (370mm wide x 190mm deep and 380mm high) by createdbygordon

I ended up printing a pot that filled the entire build platform, the only way to go any bigger is on my 3DRmega as soon as I find more time to pick that project up again.

The bigger parts were also quite interesting, as when they cooled some popped off the platform themselves, some had an edge stick but generally the bigger the part the easier it was to remove.

Does it last forever?

After around 120 prints I notice a slight yellowing of the surface.

No. For me when it starts to yellow and I can no longer revive it with a splash of water, it's time to change the coating. At this point parts were starting to stick much more strongly than before and clear ridges were visible on the surface.

It's very easy to remove, just soak in water and it will peel off. These sheets when dry feel quite different to dried PVA, they crackle and feel more crisp. They are flexible, strong and not brittle.

You can't apply it to every surface -  

That leads me on to one last bit of advice if you try out 3D EeZ - do not apply it to Kapton or PET Taped bed surfaces. You can print once, but it cracks and lifts off the surface. This makes sense really, it's just like having a PET printed part stuck to it.

I tested the material at bed temperatures from 30 Degrees C to 120 Degrees C with almost all the materials I had.

Like I said above even ABS sticks to it without a brim. I didn't print big in ABS, so that's one to do next time.

I sent over lots of feedback to Tony, he planned to tweak the formula, and try to get more manufactured as soon as possible. If he can get the volumes up and make it low cost, maybe more people can benefit.

If you are interested, Tony is also looking out for distributors in Europe/UK he is planning a launch soon.

I'm going to call this part one, as I do also intend to look at other solutions and surfaces now also becoming available. If you have any experiences of materials or coatings, want to share a great formula or have questions, do get in contact or leave a message below.

I hope that was useful and you get a chance to try out 3D EeZ at some point, it's a very interesting formulation, I wish Tony success with it.

Until next time.

Rich.

Upgrade your bed springs –> improve your print quality

Just a quick note and upgrade for you to consider - If you have a Prusa or any other machine with little springs on your build bed / heated bed especially if they are springs from biro pens think about upgrading them.

I was getting more and more issues going faster on my Prusa, most of these manifested themselves as less fine outer surfaces and less alignment on the vertical walls. So after fixing all my vertical Z issues (de-couple the Z motor mounts, make sure the rods are really straight, and adding Z rod bearing stabilisers) the only thing remaining was a little wobble on my build-bed. This wobble gets more and more the faster you print / accelerate / stop etc, so I changed my bed springs from the 'official' little ones from biro pens to ones usually used on the Extruder Idler.

On the left is a ground spring normally used on the extruder. On the right the 'official' bed spring is usually sourced from a disposable biro pen.

Change to a firm spring, under good compression.

Old Pen spring fitted, bit floppy

You need a firm spring, and make sure they are well compressed, mind you don't distort the build bed but keep them reasonably tight to stop wobble.

The change was really good and well worth doing.  I'll try to post the before and after printed objects, but I would recommend giving it a go especially if you are upping the print speeds on your machine past 80+mm/sec or having a fast machine travel speeds around ~200+mm/sec

Let me know if you change them and if it also improved your print quality.

Rich.

* Pimp my Mendel * - How to build up a LM8UU Linear Bearing Prusa

Mendel Bling is built up and running better, smoother and faster than I imagined possible.

I decided to make the Bling machine build to be a complete as possible manufacturing solution and also add a few upgrades and ease of use changes along the way, let me know what you think. It’s also a full linear bearing (LM8UU) design on all the axis.

I’m also going to use this as a *basic* overview ‘How to build a Prusa Mendel’ post, hopefully it will help someone or at least give you some ideas for your own build.

I does not cover every step, just highlights many of the main assembly points.

Frame Vertex -

My frame vertexes are a little chunkier than the normal Prusa versions, the only thing to make sure is that you have the same measurement between all frame vertex plastic faces, so if that’s 250mm like the ‘official’ Prusa or 252mm or 246mm like mine then that’s fine, just make sure they are all the same.

Axis lower frame Assembly.

Axis lower frame back with Y motor mount.

X-Carriage - 

This is Greg's X carriage fitted with 3 x LM8UU Linear bearings.

The X carriage assembly with Z ends –

Both the Prusa Z ends are a tight push-fit, I recommend wrapping just the end of the bar in Kapton tape then slightly softening the plastic before pushing it in and holding the plastic in compression against the bar until it goes cold and sets, the wrapped Kapton tape will make it impossible to remove without re-heating, so make sure you push the bars all the way into the Z end, and do one end at a time.

REMEMBER to slide on your X carriage with it’s Linear bearings before attaching into the push-fit ends of the Z axis parts.

Y stage –

If you use Greg’s LM8UU Y axis holders then assemble them in a similar way to this, there are other variants (Joem_)that have mounting lugs etc. so take your pick.

Slide them (carefully) onto the Y stage bars and space them on your frame.

So after making sure your Y axis bars are parallel, straight and spaced appropriately you can screw the Y plate onto your Linear bearing holders.

Keep going, in a short while it will start to look something like this.

The Print bed –

I slightly modified the MDF print bed by adding small wooden bars at both the Y axis ends and spray painting it with high temperature paint, you don’t need to do this, I just wanted to eliminate any flex and also have a slight lip in the Y axis direction to stop extra air movement under the bed and also over the top of the heated surface.

I added a sheet of Aluminium foil 

And then a cut sheet of silicon (kitchen baking silicon)

After this if you can get hold of it (£1 from you local pound shop) I added a sheet of PTFE cooking mesh, this stuff is great, 260 Degrees C rated. Lightweight thermal isolation and insulation for your heated bed.

The heated bed is a normal Prusa FR4 PCB; I’m going to use it upside down with the heating tracks on the top surface, so to protect the tracking and smooth it out I covered the bed in PET tape.

Top TIP – Adding the Tape (PET or Kapton) to this side of the PCB has another major advantage, any air bubbles trapped when you put the tape on make their way down the gaps in the tracks so you end up with a really nice surface, bubble free.

If you use the other side it’s not easy so apply the tape and you still end up with an impression of the PCB silk-screen printing text appearing on all your printed parts!

The stack of Alu foil, silicon sheet and PTFE mesh will give you a gap between the MDF and heating PCB, so you need to make sure you have the same spacing or a little more for the four mounting points.

Here I have used M3 stainless steel bolts, nuts and a PTFE washer to get the correct spacing.

Z Motors-

The Z drive motor connections are Nophead’s design, these work well and only need a small section of plastic tube around the motor shaft, heat it up if it’s tight to fit onto the shaft.

These should be done up tightly around the motor shaft and the M8 studding using all four bolts on each.

Z stage –

To install the Linear bearings into the Z stage you may need to heat the plastic slightly, it makes it a little easier to click them in and you can make sure they ‘set’ tightly in place and in alignment if you also use a M8 rod while you fit them in place. Secure with ty-wraps.

The motor mounts for the Z stage are both designed to have 4 bolts connecting the motors tightly down, no one actually seems to use these and it is preferable to ty-wrap the Z motors in place to allow them to ‘decouple’ mechanical wobble from Z movements.

I started by only securing with 2 bolts and using PTFE washers, but after a few prints I also ended up using ty-wraps so the motors are floating other than being held in place with ty-wraps.

The next time I take it apart I will also put a sheet of silicon in between the motors and the mount to further decouple any vibrations.

Also when you install the Z motors and the very straight M8 studding remember to add a spring inside both sides of the hexagon sections, it should be under compression between the top and bottom M8 nuts so when you do Z moves it reduces backlash to an absolute minimum.

I also added Rod stabilisers to my machines top and bottom of the Z drive rods.

I think they do help support the Z transport very well and help with vertical alignment, but you don’t really need them.

Belts and pulleys -

I have used off-the-shelf Metal pulleys this time as it’s using the 2.5mm belts and I didn’t have time to cast more of the fine pitch pulleys, I also wanted to see how tricky it was to drill and tap the metal pulleys, it’s not hard, anyone with a drill press and a simple hand tap could do it.

Top left shows how they come, you need to drill them out and add another hole for the grub screw and tap that with M3

Attach the pulleys to the motor shafts and use a good quality grub screw to keep it in place.

Greg’s X carriage has an in-built tensioner system.

Install the Y belt and if you keep the Y belts long enough so you can pull, adjust and tighten them even after the build platform is on then you should not have to disassemble anything to tighten belts in the future. 

I just taped the spare ends down to keep them tidy.

The extruder is Greg’s design along with my own Fan duct mounted on the X carriage.

The hot-end heater block is my version here - 

Electronics –

I’m not going to explain how to make up the electronics here as there is very good documentation on the Wiki and also everyone seems to have a different preference with electronics and there are so many variants.

I did have some new (at the time) RAMPS 1.3 PCB’s but I decided to use up the last 1.2 PCB I had as they are not very different in reality.

I plan to use the 1.3 Versions in a future dual extruder machine that will look quite different to this machine.

I’m housing the wiring and electronics in a simple ABS case (175mm x 110mm x 66mm).

Neat wiring and cable identification is a good thing to make it easier in the future.

RAMPS installed and some of the stepsticks for testing.

Make sure you have nice chunky wires and good crimps for your +12V power connections to the electronics.

RAMPS V1.2 does not have a separate Poly-fuse for the Heated bed, This one is 11A rated and connected separately to another set of +12V wires from the ATX PSU and then on to the heated bed, the return from the heated bed goes to the H-bed MOSFET sink (GND) on the electronics.

These are simple mechanical endstops to my own design, but you can use whatever switches you have or opto-endstops if you prefer.

I added a few LED indicators so I can see when the power to the system is on, the heated bed is active and the hot-end is being driven, I recommend doing this it tells you a lot about how your system is working and is nice to check things happen when you expect them.

When running from SD cards it’s always nice to be able to quickly remove the card to update Gcode, so I decided to extend the SD cable to the outside of the enclosure in an easy to access spot. 

To do this I simply used a cut-down Ribbon cable (an old PC Floppy drive cable) and fitted a male connector so I can fit the SD card adapter. I have not had any problems doing this, but keep this ‘extension’ as short as possible.

I’m using Stepstick motor drivers in this machine, they are almost identical to the Pololu versions, and in the photo you can see both the Pololu (top left) and Stepstick below.

Think about cable strain relief and give yourself plenty of cable to allow free movement especially on the Z axis and the extruder assembly.

Added an on/off switch for the ATX psu (this is on the low-voltage side, instead of just wiring it to ground to get the ATX to turn on), it can be classes as a panic button also as it will shut down all motors and heated things.

To mount everything and as a frame for a filament spool I use some 20mm T-slot Aluminium, you could just as easily use wood or threaded bar for this ‘extension’. The Mendel is fixed to this via 4 x 8mm printed plastic mounts at the base and two small M8 threaded rods for extra stability using the normal Mendel mounting plate at the top. The electronics box and ATX psu are all mounted to the t-slot frame.

At the very top there is space for a standard 30cm filament spool using another M8 threaded rod. This means the whole thing can be picked up and easily moved all you need is power and a USB cable that can also be removed after you set off a print using the onboard SD card.

Finally four rubber feet to minimise vibration and noise .

Software and firmware is your choice, and many more are on their way as I write this, I’m using Sprinter and Printrun (Pronterface) along with Skeinforge V41 but people are starting to use SFACT, I just had a quick go and it has much easier to understand names for the settings, and a lot less ratio's, so be sure to check it out.

First print.

It’s first print was successful and it went on to print all my multi-coloured things a few posts back.

It’s now going to be hard at work making more stuff and of course a grandchild…

Thanks for reading, I hope it was useful.

Rich.