Why You Should Add 3D Printing To Your Labs Workflow

A brief history of 3D printing:

Few technologies in recent years have been as transformative as additive manufacturing. 3D printing is an older technology that actually began its start in the 1980's as a means for rapid prototyping. The first patent published by Hideo Kodama (JP S56-144478) was filed in November 10, 1981. It described a process called stereolithography where a photo-hardening thermoset polymer is exposed to UV light in a series of slices in a model, where the thickness is controlled by 'masking' each layer to take the form of the desired shape. Little interest in this patent was shown, until it was ultimately abandoned. While this technology exists today (resin printing); we normally think of fused deposition modeling, also known as FDM when thinking of 3D printing. This process involves using extruded plastic, added layer by layer and was developed by S. Scott Crumb and commercialized by his company Stratasys in 1992.

Figure 1. Kodama's 3d printing sample using cured resin.

At its core, FDM printing is just a nozzle which melts plastic, 3 axis of movements and stepper motors to move where the plastic is melting onto. In principle, one can use a 3d pen to manually melt the plastic and shape it into the desired form. Thankfully in the 1950's some researchers at MIT servomechanisms laboratory invented the computer language known as G-Code so we wouldn't have to do it manually. Refined in 1963 by the electronics industrial alliance, G-code was first developed for computer numerical controlled machine tools (CNC) like automated lathes or mills. G-Code is simply the set of instructions that move the print head, or bed around so that your machine can place the plastic where it needs to be.

It wasn't until around 2010 when 3D printing went from being a niche research or industrial process; but one enjoyed by hobbyists. At that time, you still had to assemble your own printer using a kit. They were wholly unreliable, often caused fires due to unregulated solid state relay switches (thermal runaway) and used ABS (Acrylonitrile Butadiene Styrene) which is a semicrystalline solid which often warps if a consistent temperature is not applied to the whole piece being printed. Even so, this led to an explosion of interest in the topic.

Figure 2. An early FDM printer known as RepRap, self assembled by Adrian Bowyer (2009).

Printers have seen incredible rapid development. First we saw the advent of the commercial 'bed slingers'. This is where the heated print bed (used to keep the plastic stuck in place) generates your X axis movement. New filaments such as PLA (Polylactic Acid) that were amorphous and could produce more consistent results without a heated enclosure. Next came the 'Core XY' printers which kept the print bed fixed in place, and allowed for the use of heated enclosures. This allowed for the printing of more exotic engineering grade filaments. Even though this, it is notable that it took considerable technical prowess to not only assemble but maintain these machines of ever increasing complexity. Nowadays you can buy a printer that comes out of the box, ready and capable of producing high quality prints.

Figure 3: An example of a 'bed slinger' type printer (Anycubic Mega-S).

What can a 3d printer do for me and my team?

A 3d printer allows not only for fast prototyping; but also the fast and efficient repair or modification of your current equipment. They are cheap to setup and maintain, and in my experience in industry; an absolute vital piece of equipment in any research lab.

Where do I get started?

Here are some basic steps to follow:

Step 1: Obtain AutoCad Software. AutoCad software lets you build 3D objects with definite dimensions, so they come out the proper size and shape once printed. Fusion360 offers a free version for personal use. If it is not being used to produce commercial designs that are being sold, then it is a good option. Otherwise there are alternatives like FreeCad that offer similar features but are completely free to use. Good tutorials are numerous and can be found readily available on youtube.

Step 2: Select your printer. The current gold standard for consumer grade printers is Bambu Labs. These printers are pricey, but come ready to use out of the box. Other then cost, another drawback is the fact that Bambu currently streams all printed content on its cloud servers. If you are creating proprietary things, I would suggest looking at an alternative such as Creality or Qidi. Unfortunately these suppliers are not as good or responsive to repairs. There are some 3D shops that have supplier deals, that get good deals on parts for printers and have the knowhow to fix them. If you live in Calgary '3dSpool' is one such supplier that offers to maintain Bambu labs or Creality printers you purchase from them at the same cost you would get if you bought them online.

Step 3: Choose your filament. While PLA is an incredibly quick material that behaves easily, it has a significant number of disadvantages. Firstly is its thermal deflection, where it begins to deform above 50 - 60°C. High temperature PLA can now reach temperatures of around 150°C while keeping its form factor. However PLA is still incredibly vulnerable to chemical attack. It is not a material well suited for the laboratory environment.

Lets outline the properties you should look for when selecting a material.

Heat Deflection	Chemical Resistance	Strength	Rigidity	Printability
What temperature can it keep its form.	What chemicals is it resistant too.	How much force can be placed on it.	How tough it needs to be.	How easy it is to print.

Since this is a chemistry blog however, I am going to focus on chemical and heat resistance. The following is a chart which ranks filaments that can be printed on a consumer grade 3D printer in terms of their chemical resistance.

Filament	Chemical Resistances	Thermal Stability
Polyphenylene Sulfide (PPS)	All known solvents below 200⁰C	230°C
Polyvinylidene Fluoride (PDVF)	Strong acids/Bases, Organic Solvents and Oils	150°C
Polypropylene (PP)	Strong Acids/Bases, Most Organic Solvents and oils.	82°C
Polycyclohexylenedimethylene Terephthalate Glycol (PCTG)	Weak organic acids and bases and alcohols	88°C
Nylon (PA12)	Many Organic Solvents, Oils and greases	130°C
Thermoplastic Polyurethane (TPU)	Oils and greases	80°C

The most chemically resistant filaments are unfortunately are not just more expensive, but often more difficult to print. The crystalline nature of many of these filaments make them prone to warping. However proper precautions can be taken in order to reduce that warping. One of the more economical chemically resistant filaments is polypropylene, however it rates as one of the more difficult ones to print well on our consumer grade printers.

Figure 4: Polypropylene 3d printed part lifting off the surface of the build plate due to warping.

Figure 5: Polypropylene part printed with a raft to reduce warping

Why do prints warp, and how do I stop it?

There are excellent video's on this subject that I highly recommend. I'll do my best to summarize this topic in a short and concise manner.

Reason 1: Crystallinity

The number one reason prints will warp is due to the material being used. It is not to say that an amorphous polymer like PLA cannot warp. The reason is that as the material cools, it begins to crystallize. These crystalline zones can put internal pressure on the print in different directions, causing the material to warp and bend out of shape. Crystallinity simply refers to how often you can find a unit of the structure. Think of a polymer as long chains of a repeating unit cell. How these unit cells align affect the properties of polymers significantly. Now imagine an old soviet apartment block, where every single apartment is exactly the same. Each apartment is this polymer 'unit cell' and the orderly nature is what makes it crystalline. An amorphous compound by contrast has no regular order. Think of a homeless camp, chaotically put together.

Figure 6: A comparison of crystalline, semi crystalline and amorphous 3D printing polymers.

Fix: Swap to an amorphous polymer, or use a version that contains either fiber glass or carbon fiber reinforced filament. These tend to reduce the internal forces caused by crystallinity, as well as improve properties such as overall strength.

Reason 2: Bed Adhesion

A 3D printer bed is a heated surface, which heats the plastic to the point of where it remains tacky enough that it sticks to the surface of the printer bed. In the case of polypropylene, it does not readily stick to surfaces when heated.

Fix: Make sure that you are using the right temperature settings for the polymer you are using. Double check to make sure that the filament you are using does not require a special adhesive to work. In the case of polypropylene, you will not can any adherence unless you use a special adhesive.

Reason 3: Parts Design

Hard edges distribute the force in a way that can cause the edges of your prints to lift.

Figure 7: Illustrated difference between hard and soft edges.

Fix: Make sure to round the edges of surfaces directly on the build plate. This can diffuse those forces, and keep your part adhered to the print bed.

Reason 4: Temperature Gradient

For semicrystalline materials, they can expand and contract in differences in temperature. This is a fundamental issue with how FDM produces the part you create. When you put down a layer of plastic, it immediately begins to cool. When another layer is places on top of that, there is a significant difference between the top layer and the one below. The difference grows as you move up the layers. There is also a difference between the layers of plastic, and the surface of the build plate.

Figure 8: Exaggerated illustrated example of the uneven heating

Fix 1: Use a heated chamber. Ideally you would use a heated chamber that keeps the plastic near its glass transition point. This is before the semicrystalline nature of the filament begins to allow the polymers to align in ways that may warp your part. On consumer grade printers, we are limited to temperatures of 65°C. Also note that some of these printers rely on the heated bed to increase the overall temperature. This will cause issues if you need to have a low bed temperature with your heated chamber.

Fix 2: Shrink your heated part. The larger the part, the less even the cooling will be and the more opportunity there will be for print warping.

Reason 5: Lack of Filament Drying

Wet filament can cause an under extrusion of the filament, and uneven heating during printing. Water has a significant heat capacity, and can absorb some of the heat on the hot end meant to melt the plastic. This can cause bubbles to form, and even harm overall layer adhesion,

Fix 1: Pre-dry your filament, and store it in a dry box when not in use. Below are the recommendations for filament drying for the various chemically resistant filaments.

Filament	Drying time	Drying Temperature
Polyphenylene Sulfide (PPS)	8 - 12 hours	110 - 140°C
Polyvinylidene Fluoride (PDVF)	Not required	N/A
Polypropylene (PP)	2 - 4 hours	75 - 80°C
Polycyclohexylenedimethylene Terephthalate Glycol (PCTG)	4 - 8 Hours	60 - 70°C
Nylon (PA12)	6 - 12 Hours	70 - 90°C
Thermoplastic Polyurethane (TPU)	4 - 8 Hours	50 - 70°C

Proof of Concept:

The best way to demonstrate the power of a tool is to provide a real world example. I had the idea some time ago to make custom adaptors that could be used for thermocouples or pressure gauges that are not readily available. Here are the steps that I take to do this work.

Step 1: Create and refine your concept

Using a good set of calipers, obtain the measurements and try to draw your concept from multiple angles using grid paper.

Figure 9: Recommended tool (calipers) for obtaining measurements for AutoCad Design.

As the old adage goes, measure twice and cut once. You do not need to use expensive 2700$ Mitutoyo calipers for 3D prints. Plastic expands during printing, and tolerances that can be produces can vary as much as 0.4mm. Instead I use 35$ Kynup Digital Calipers that work perfectly for this application.

Figure 10: Demonstration of the inside measurements jaws for measuring thread pitch.

Figure 11: Hand drawn schematics of our concept, ready for AutoCad.

The glass joints that I am working with are 24/40. The larger end is 24mm, while the bottom of the taper is 20mm. To allow for a keck clip to hold the joint in place. For the inside, the thread pattern is a 1/8' inch NPT thread. Inner NPT threads cannot be modeled in Fusion 360 natively and require extra effort. Small threads are difficult for 3D printing anyhow. In this case we can print in a polymer that can tap with the appropriate thread pattern after.

Step 2: Turn your concept into a 3D model.

Now that we have our concept done, it is time to translate it into an AutoCAD software of your choice. In my case, I prefer to use Fusion 360.

i) Click on the 'Create Sketch' Icon listed in the top left of your screen. Select the 'Center Circular Diameter' tool. Click on the grid and draw your circle. Type in '24 mm' into the window. Select the finish sketch button.

Figure 12: The Sketch tool

ii) Select the extrude tool. Click on the surface of your sketch. Make the distance '-40.00 mm', and the taper angle '-2.8624 degrees'. Make sure it is on 'New Body' beside operation, and then click 'okay'.

Figure 13: Extruding to a taper

To calculate the taper angle, I used the following formula:

*Do not forget to make sure your calculator is working in degrees and not radians.

Eq 1. Taper Angle = 2 x arctan ((D - d) / 2L)

Where:

D = Major Diameter (24mm)

d = Minor Diameter (20mm)

L = Length of object (40mm)

iii) Name your object to 'Taper Body' or whatever you like. Hide the object on the window by clicking on the eye. Extrude the next object up '4 mm', set the taper angle to 0 and make sure the operation is 'New Body'. You can turn the other body's veiw back on.

Figure 14: Adding straight component to adaptor

iv) Select the lateral surface of the small cylinder. Go to the modify section and click on the 'Offset Face'. Offset by '-1.4 mm' and click okay.

Figure 15: Offsetting the lateral surface

v) Click on both objects, and select combine. Make sure that the operation is 'Join' and press okay.

Figure 16: The combine tool

vi) Select the edge between the large cylinder and the small one. Select the fillet tool under modify. Adjust the fillet to '1.5 mm'.

Figure 17: Using the fillet tool

vii) Look at the bottom of your adaptor. Click the hole tool, and drag and drop the hole to the center point. Adjust the length to '40 mm' and the diameter to '5 mm'.

Figure 18: Adding a hole through the adaptor plug

viii) Look back to the to the top of your adaptor. Left click on the surface of the top of the object and select 'Create Sketch'.

Figure 19: Creating a surface sketch

ix) Click on the 'center diameter circle' tool, and draw a circle in the center of the top face, and set the size to '10.287 mm). This is to fit the 1/8' NPT pressure transducer.

Figure 20: Adjusting the size of the surface sketch.

x) Select the sketch object on the face, and extrude down. Give it a distance of '-12.0 mm' with a taper angle of '-1.7899 deg'.

Figure 21: Creating the internal taper for the pressure transducer.

*Note before exporting

The adaptor is now ready for export! Please note that due to the shrinking or expanding of certain polymers, using the face offset becomes important to help your objects reach tolerance. For example, PLA can expand up to 0.2 mm or more so in this example you would want to have an offset in the inner diameter of (-0.2 mm). Since we are using polypropylene, I will take care of this in the slicer by adjusting the shrinkage. The shrinkage of polypropylene is 1 - 2.5%. I'll adjust the settings to 2%.

Step 3: Export the model and adjust the print settings.

Now that our model is complete, it is time to export the file. In fusion, you can right click on the object you want to export in the side menu, and export it as an 3MF or STL. If you are using some variant of Prussia slicer like OrcaSlicer, it may be better to use the STL format. This is because Prussia slicer will try and load settings from your file as a 3MF that simply don't exist, since it came straight from your AutoCad program.

Figure 22: Exporting the saved mesh for the 3D slicer.

Import it into a slicer. The slicer used often is almost always recommended by the manufacturer of your printer. In this example I am going to use polypropylene. It is important to note that polypropylene begins to distort around 80°C. In temperatures above that, it is important to switch to either PDVF or PPS.

Here are the print settings I use for polypropylene:

Figure 23: 3D Print settings for polypropylene, using the special adhesive.

Make sure to use a 'raft' in the print settings, along with a brim. The raft acts as a second surface that adds extra adherence to the print bed. It semicrystalline materials that can warp easily, it is important to do this otherwise even the adhesive might not have enough ability to keep the print stuck to the bed as pictured in figure 5.

Figure 24: The final sliced file before printing

Use a heating chamber if at all possible. In the case of my printer, the Qidi Plus 4 the heated chamber only works at print bed temperatures of 70°C or greater; and will exceed the adhesive manufacturers print bed temperature of 45°C. As a result, I was forced to abandon the use of my heated chamber.

Step 4: Prepare the print bed for printing.

Using isopropanol (rubbing alcohol) I gently clean the print bed with some paper towel. I then apply the polypropylene adhesive to the build plate. The build plate then needs to be preheated to 35°C for 5 minutes. Do a full bed leveling calibration at the correct temperature of the print bed. In this case I use 45°C.

Figure 25: Applying the polypropylene adhesive to the build plate.

Step 5: Print

This may seem straight forward, but it is important to monitor your print. Make sure that the first layers appear to be stuck firmly to the bed of your printer. If they appear to be lifting, then consider cancelling your print and starting again. Its better to waste 10 cents of filament, then 5$ worth of it.

Figure 26: Timelapse of the adaptor print

Step 6: Test, fit and iterate!

Do not worry if you struggle to get the correct measurements immediately. While 3D printers have come a long way, and even compensate somewhat for this effect.

Because we are using steel threads, we can actually use our pressure transducer to cut our threads for us. Please note that not all filaments are suitable for this. Thankfully polypropylene, PPS and PDVF all can be threaded this way. Larger threads can certainly be printed.

Figure 27: Test of adaptor, using the threads on the pressure transducer to 'Tap' the adaptor.

Figure 28: Final fit test of the pressure transducer adaptor.

And there we have it! Our completed project. If you enjoyed this article, feel free to contact me and make a request. I can also help train staff and provide instructions for prototyping. If you would rather me do the prototyping, I can do that for you as well.

If this helped at all, please share this article! I would also love to see any creations made. Photos, requests or ideas can all be sent to daniel@transitionelement.net.