Perpetual Filament Printing

Almost all the end user 3D printers use filament plastic as the raw material.
End user printers are mainly based on CNC devices which are quite mature in design.  Therefore, the most critical part of additive manufacturing systems is deposition of just the right amount of material at right location. The location precision can be guaranteed up to an acceptable level.  However material deposition cannot be controlled precisely at the moment. This results in low quality output for end user 3D printers.

For high quality printing the followings need to be improved:
- Filaments must be made of homogeneous raw material
- Filaments must have consistent diameter

The quality of raw materials mainly depends on the manufacturer. The coloring materials used in the filament also affect the quality of print.  For high quality outputs, one needs to fine tune the extrusion settings every time the filament role is changed.

Filament diameter inconsistency can be compensated by using filament compensation electronics.  The extrusion speed can be adjusted based on the filament shape.  The filament shape can be determined using hall-effect sensors.  For further detail you can check Motate128 Stepper Motor Control System. It was discussed in Filabot's forum under filament compensating electronics topic.

Another important topic on 3D printing using plastic filaments is sustainability.

For sustainability it is important to recycle the plastic. In order to obtain a nice looking print, one needs try couple of times.  The end result is a lot of discarded faulty prints.  I propose a new index for 3D printing "Discarded Plastic Ratio (DPR)".  An example of a DPR would be: “With the current technology, 100 gr of plastic used in order to print, 40 gr of usable object.  So the DPR of the printing is 60%“.  After using the printer for some time, it becomes really important to recycle the waste.  There are projects going on to resolve this problem.  Filabot is the first successful filament manufacturing device developed for the end user.  Additionally there is a perpetual plastic project.  The idea sounds good.  However, I find it quite impractical to melt different plastics into one and form a filament.  The filaments produced this way would be unpredictable in behavior.  The print quality would suffer.  Also the chemical fumes coming out of such filaments may be health hazardous.  In my opinion, the recycling of plastic is a complex process and therefore should be handled with a recycle center.  On the other hand, recycling of defective 3D prints is a more feasible solution for perpetual plastic printing.

3D Printer Raw Materials

3D printers utilize different types of raw materials. Following are my personal thoughts on two different raw material types.

Filament ABS, PLA:
- Mainly used in RepRap derivate printers.
Pros:
Relative low cost printer design. Utilizes low cost extruders. Printed objects have continuous layers, improving rigidity and smooth look.  Depending on the nozzle width, the build speed can be considerable increased in trade off lowering the detail of final object.

Cons:
Uniform and perfect circle filaments are required for high quality results.
Difficult to align layers. Difficult to control the extruded plastics thickness. Bubbling may occur.
Supporting the design is not that easy, either another extruder is required with a water soluble plastic is used or the gap can be filled with loosely build plastic.  Stringing is also another problem.
The object printed would deform while being printed due to temperature differences. The bottom platform can be heated and the extruded plastic is hot, however the plastic in between is exposed to the ambient temperature. Due to these temperature differences warping is inevitable depending on the shape of the object. Heated platform is required for larger ABS prints.


Powder Plastic (I don’t know the exact plastic composition):
- Mainly used in high end 3D printers.
Pros:
Plastic powders unlike filaments can have higher tolerances in manufacturing. No support material is needed while the powder serves as the support. No heated platform is required. Less warping occurs due to less temperature deviation. Consistent printed unit volume results in consistently high quality prints. Parallel processing can be easier to implement compared to filament based printers.

Cons:
Higher cost of printer design. Laser sintering, thermal print heads or inkjet technologies are utilized in these printers making them difficult to develop for low budget startups.
Unless multiple lasers are used printing object with laser sintering is slower compared to large nozzle filament printers. Because focused laser beam can only melt small regions compared to large size of nozzle. It is expensive to manufacture very fine plastic powder (2-3 micron).

Suggestions on Future 3D Printer Design

Current printers utilize circular extruders. These extruders have a fix width. If you would like fine details on your print you should use extruders with small nozzle size (I use the term nozzle size for the nozzle opening diameter). The downside of this is it takes a longer time to print the object while only small amount of plastic can be extruded at a given time. If you would like to shorten the print times and do not require detail on your print you can use extruders with larger nozzles.
One possible question comes to mind, “Is it possible to have an extruder with variable nozzle size?”.
A continues variable nozzle is complicated and therefore not feasible at the moment. However it is possible to have an extruder with two nozzle sizes. While printing, the printer can alter the nozzle size, either fine (0.25 mm) or bold (1.00 mm). For further detail on this subject you can read the paper   “Variable fused deposition modeling - concept design and tool path generation”.

Here are my suggestions on the extruder design:
The plastic extruder opening can be rectangular instead of being circular. One side of the rectangle can be controlled to adjust the size of the opening. With this method it is more feasible to achieve variable fused deposition system. Additionally it is possible to add channels on the nozzle to color the extruded  plastic. Special ink should be used to color the hot plastics outer surface. The ink should stay intact with the plastic even after it is cooled down and shouldn’t peel off or wash out with water. Most probably the ink used will make the surface of the plastic dull instead of shiny. However it is just a guess.

Another major change I propose, is on the plastic feeding system. The printer should use raw ABS or PLA as a granule. There will be two stage plastic feeder. First one will move raw plastic to the melting chamber. The second one will push the melted plastic from the chamber to the nozzle at a variable speed.  The trick is how to design the second stage. Most probably if it were easy it would have been implemented.
The main advantage of using granular plastic is that they are widely available and cheaper than the filament plastics used at current 3D printers. Additionally the quality of printers would be independent of the plastics shape. On the other hand filament plastics should be perfectly round and consistent in size in order to get high quality prints.

Thoughts on The Future of Personal Manufacturing Systems

Internet and smartphones are the two most popular subjects of the decade.  Web2.0 and smart phone revolutions are quite new.  Therefore people are still amazed with what they can do with them.  There is a new generation growing up with internet and smart phones.  They take these two things as granted not something new like we grownups do.  By the time this new generation grows up. They would be quite fed up with internet and smart phones. They would be looking for more…

Internet by the help of smart phones help people get whatever information they want immediately.

Internet era children will be therefore eager to get what they want.  They will not like to wait a week for internet purchase delivery. Additionally they will want to have something which is unique to them.

Next generation manufacturing systems need to keep up with the demand characteristics of the future.  Even the big industrial companies will not have the luxury of having manufacturing facilities focusing only on one product. The facilities will be small and scattered all around the world.  The supply will be there where the demand is.

Personal Manufacturing Systems and Local Manufacturing Systems will be the future of production.  My objective is to contribute on the development of these systems and I will be sharing my work on this blog.

3D Design Tricks

It is important to know how 3D printers operate before you design and print your design.  If you make your design knowing this you would get better results.

Some design recommendations:

1. Don’t forget the gravity. (The objects are printed from bottom to top. There needs to be a physical material beneath the extruded plastic. Rotate your design before printing so that there is less bulking of from the center. Some printer software can add support material bulking parts of your design.  If the printer has two extruders like Replicator, the printer can use water soluble plastic as a support material. This plastic can be removed after proper water treatment.  For the rest of the printers the support material is the same material as the object. The supporting part is printed with less fill factor making it weak. This enables easy removal of support material with breaking the part by hand.  If you have a design where the object has high detail near or above the support material, these parts may be damaged while removing the support part. Therefore it is wise to rotate the object in software such that there should be no support near the detailed parts on the final print.)

2. Avoid sharp corners and straight edges. (It is quite difficult to have results that have sharp corners and straight edges. That’s why most of the print outs shown have more rounded design.)

3. If your design has connection of multiple printed parts try to have larger tolerances on the interlocking parts. The printouts are not perfect.

The Hardware

Most 3D printers are sold unassembled.  The casing of these printers are made of laser cut Plymouth.

The assembly process takes around 8 hours of continuous work, normally two days or more is required to print your first 3D model.   Most parts are custom made for the printer. Therefore if you break one while assembling you need to order that part again and wait for the delivery.

Quality 3D printing requires precise placement of right amount of plastic to the right point. In order to achieve this following are required:

- Perfectly round and uniform filament

- The extruder speed and the temperature adjusted according to filament type and color

- X,Y and Z axis positioning reliably exact.

The printing process takes several hours if the quality is set to highest. The reason for this is more layers are required for fine details and for better positioning and better filament extrusion X,Y axis speeds are lowered. (You need to print more layers at a lower speed)

The printers do not have a well-established calibration and positioning systems. Therefore if something goes wrong even for a short duration, the system cannot correct it afterwards and the rest of the printing would be spoiled.

A feedback system should be developed for the printer to improve the quality. I will thing more about on this after I get my printer and start printing with it.

Long operating hours would loosen screws for X,Y axis motor mounts. This results in improper belt tension and therefore not exact X,Y positioning.

A rigid printer chases would be more suitable for such printers. Therefore I am preferring MakerGear’s M2 over others.

The short comings of 3D Printers:

1. Frequent calibration and maintenance requirement (Calibration: Positioning, extrusion speed; Maintenance: Oiling of gears, belt tension control)

2. Slow print speed  (Even with the fastest printer on the market it takes several hours to print a high quality print.)

3. Unpredictable print quality (The print quality depends on so many factors that, it is almost impossible to have a high quality output. The lack of a proper feedback system is the main shortcoming.  Additionally long print time increases the deformations on the plastic as a result of contraction of plastic due to heat differences between layers.)

4. Post processing requirement  (Some printers have serious stringing problem (Ultimaker), that requires careful cleaning of extra plastic coming out.  The removal of the support material.)

5. The operating noise of the printer (You can hardly stand that noise for hours. Therefore you cannot have it in your living room or your bedroom.  That’s a problem for those who do not have a garage or a study room)

3D Printing Workflow

3D Printing Workflow

Here is a summary of 3D Printing workflow written by Florian Horsch on his blog.


(Source: http://techwall.net/ultimaker-3d-printing-workflow)

As seen from the diagram the process of 3D printing requires the use of multiple programs. During the process several files with different formats are generated. The use of these applications require skill and knowledge. In order to get a quality print you need to set profiles in your slicing application. Dave Durant has a nice blog post on importance of profiles.

It would be nice if the files we get from Thingeverse would contain recommended profiles. I don't know technically, but if the profiles are standard across different printers then this could be possible. Otherwise it is impractical to include profiles for each printer on the market.
The software behind personal 3D printing needs serious improvement before it is adapted by the end users. This work can be carried out independent of 3D printing technology.

My preferences for the software are:

There should be one application handling all the required steps.

The application can be composed of modules developed by different venders, but the UI should be same and integration between the modules should be smooth.

First Thoughts on 3D Printing

I started this blog to share my experience on personal 3D printing systems.

Now I am gathering information on available 3D printers.
My initial analysis stopped me from buying a printer now.
I will wait for a couple of months more.

As a result of my analysis, I decided on buying one of these printers:
- Replicator (MakerBot)
- Ultimaker
- M2 (MakerGear)

My criteria on selecting a 3D printer are as follows:
1. Easy to assemble and use
2. Firm and Community Support
3. Reliable performance
4. Quality of print
5. Speed of printing
6. Low frequency of servicing

I was initially thinking of hacking a 3D printer. However, I decided to do it after I start using a more reliable system. Here are my thoughts on the printers based on my criteria. As seen there is no clear winner.

Criteria	Replicator	Ultimaker	M2
1. Easy to assemble and use	I can order the product fully assembled (+++)	I have to assemble to printer myself (-)	I can order the product fully assembled (+++)
2. Firm and Community Support	The firm has a well established support and the customer base is the largest among others (+++)	The firm has quick responding support. Customer base is small (+)	The firm has only IRC based support. Customer base is small (-)
3. Reliable performance	Hard to conclude from forum and blog posts however it seems a steep learning curve. (-)	Hard to conclude from forum and blog posts however it seems a steep learning curve. (-)	Metal frame is an advantage. However it is recently released therefore needs several months of user feedback (?)
4. Quality of print	Not that much impressed compared to Ultimaker (+)	If everything goes well, the results are outstanding. The biggest negative is stringing!!! (++)	No sample printouts yet (?)
5. Speed of printing	Slow compared to Ultimaker (-)	The fastest among compatitors (+++)	The speed is unknown (?)
6. Low frequency of servicing	The wooden frame may require frequent maintenance (-)	The wooden frame may require frequent maintenance (-)	The metal frame looks more reliable hence requiring less maintenance (++)