Tuesday, August 6, 2013

Why you are not going to get a desktop SLS printer*

Preface:  The first patent ever (owned by Carl R. Deckard)  issued towards Selective Laser Sintering is going to expire in 2014. Because of that, a few popular magazines wrote articles about it, saying that these (or more exactly, this) patent(s) is "holding back SLS from getting to the hobbyists".

To be honest, this is just wrong. I tried to collect the reasons why you won't get a SLS printer for your desk, and compile them into the following list. Feel free to comment on it, and discuss it with me.
  1. Cost - Divided into the different parts that make the printer that expensive
    • Laser
    • No, the laser is not going to get cheaper only because some patents, which aren't even related to it, are going to expire. Maybe the price of an appropriate laser will sink a bit, because the demand increases, but that's not going to make a huge difference. Just taking a less expensive one is not going to work because you need:
      • Accuracy
      • The right power
      • A good focus
    • Powder
    • Here it's already more likely that the price is going to drop over time. The problem is that you can't just take "some powder". You'll need a material that fulfills the following:
      • thermal conductivity
      • thermal capacity
      • viscosity
      • flowability
      • grain size distribution
      in the right amount. There is a reason, why people write their PhD about the topic of finding new usable powders.
    • Heating Chamber
    • If you want to print fast (read: efficient), the whole printing area should already be near the sintering point of the powder, which lies around 180°C (depending on the powder). Now that you have a functional and accurate (yes, 1 degree does make a difference) heating chamber, have fun isolating it.
    • Gas
    • You are going to inert nitrogen into the building area, because the powder would react with oxygen at the temperature you print at. That would influence the quality you are able to achieve heavily.
      Did I mention that you have to pay that gas? And you should also store it appropriately, because nobody likes to suffocate.
    • Leakless building chamber
    • If you are going to inert nitrogen into the building chamber, please make it leakless. Better for everyone.
  2. Size - Sounds weird, but there is a reason why the current printers are that big
    • Powder storage
    • The powder distribution system has a huge advantage (you don't need supports), but also the disadvantage that you need about double the size of the printing bed as footprint. (Why's that you ask? I'm going to answer that in the "How SLS works"-post).
    • Heating Chamber
    • It's hot in there. You should isolate it. Unfortunately, that takes even more space up.
    • Laser
    • If you get a powerful laser, you are probably going to get a whole processing-unit with it. That thing isn't too small either!
    • Cost efficiency (While this applies to the machines we have right now, you can probably ignore that for the hobbyists version of it (if it will ever exist).
    • To be cost efficient, you should print as much stuff as only possible at the same time. This is easier if you have a bigger printing area. Furthermore, you can print bigger (= more expensive) stuff on a bigger machine.
  3. Various - Even more reasons againt something I want. Damn!
    • Mess
    • After your print finished, you should get your parts out of the machine. Which is full of powder now. Welp, that's going to be a mess. In your workspace.
    • Powder
    • The particles of the powder you use are small enough to be breathed, what's pretty bad for your health. Also, they are going to fly everywhere.
    • Gas
    • Do you really want a big bottle of nitrogen in your workspace?
    • Calibration
    • You thought that the calibration of your FDM printer is annoying and difficult? I feel like it would be much more difficult if you had to calibrate:
      • laser
      • powder distribution system
      • heating chamber
      All these points are important, and influence the quality of the part you get out of your hypothetical machine.
    • Powder recycling station
    • Because the part you just created was sintered, and the space where the powder is, was hot, these particles are sticking to each other as well. This has two consequences:
      • You have to filter the used powder to get the sintered particles out of the "good" powder.
      • You need huge amounts of powder, because you are throwing big parts of it away after every print.
    • Lead time
    • Chances are that noone really started the development of a DIY-SLS printer. I guess you should expect a waiting time (if you will ever get such a printer).
    • Patents
    • Did you really believe that the ur-patent from Mr. Deckard was the only patent on SLS?
That being said, I would love it if I had a dektop-SLS-printer. I am simply trying to get some real arguments into the discussion.

*depending on your definition of "desktop", the possibility that we might get such a printer exists. It's still very unlikely.

Thursday, August 1, 2013

Fused deposit modeling

I already promised that there are going to be more detailed posts about the printing-techniques. (I mean...you waited long enough now.) So, here's the first of them:

Fdm

History

Who read my last blog post knows already that the fused deposit modeling is the oldest "printing" method. The technology was developed in the late eighties and commercialized in the early nineties by Stratasys, Inc. S. Scott Crump, the co-founder, and currently CEO, of Stratasys patented the FDM technology in 1989 with his wife and also co-founder, Lisa Crump. In 2003, the FDM-technology was the most used rapid prototyping technique.

Process

The model is - obviously - built layer for layer, either until the model is finished or there isn't any material available.
The FDM-Process
As you see on the graphic, the extruder/ the printing head or the platform where the model is built, is fixed. However, there are a few different ways of moving the extruder. The most common method to do this is the "cartesian"-system, but a different system - the so called "Delta-Printer" - is rising.

Material loading

The material used is stored on a spool respectively multiple different spools if there is more than one printhead. From the spool, the filament is led to the extruder. As you see, the material loading for the FDM process isn't too complicated. However this is very different for the other processes I am going to describe in here.

Extrusion

The extruder is the heart of any FDM-printer. It's the place where the stiff plastic filament gets molten and placed in the form of the model we want to achieve. The following graphic is pretty self-explaining, but I am going to say a few words along with it:
Material-loading and extrusion
After the filament is pulled into the extruder by the stepper motor, it is pressed inside the hot-end, where the filament gets molten. Depending on the filament (ABS, PLA, etc.) the temperature of the hot-end varies.

Advantages


  • It's cheap. As you may know, the cheapest 3d printer is the printrbot simple for only 300 Dollar. Compared to the much more expensive SLS/SLM/SLA-machines, even something like the makerbot is easily affordable.
  • The filament is available in a great range of materials, and most of it is pretty cheap.
  • Very compact build-style.
  • Easy to operate and keep working.
I know that there are a few more advantages, but I feel like these are the most important. I might even write a more detailled post later on.

Disadvantages

  • Supports are required for complicated geometries. Although there are machines that use dissoluble materials for the support is this a limiting factor for this technology.
  • If you want a clean surface, you have to go through post-processing. This is the effect of the layers.
  • Lacking accuracy in comparison to the "bigger" printing-technologies like SLS.


I hope reading this post was enjoyable and informative for you.
I promise that there won't be such a big gap between this and my next post.

Cya