In part 2 of this blog, we dive into a more technical understanding of what we call 3-axis filament.
What is a Quality 3D Printer Filament? - Part 2
In our previous blog, we got everyone up to speed on the extrusion process. We hope that background was helpful in understanding the process of 3d printer filament production. In part 2 of this blog, we dive into a more technical understanding of what we call 3-axis filament.
In filament inspection, we shoot a laser beam at the filament to get data back on how round it is, or specifically, the filaments diameter and ovality reading. Field of view becomes important here, because 1 laser is not able to capture all of the possible data on how round the filament is.
Field of view / Fault detection
In the above diagram, the red areas indicate blind zones where the laser can’t see; when the laser measures the filament, each laser only sees the outer edges perpendicular to itself.
In the above diagram, the red areas indicate blind zones
- 3-axis – 90.8%
- 2-axis – 78.5%
- 1-axis – n/a
The range of possible defects in the blind area of detection from a reliability standpoint are as follows:
- 3-axis – 7% of diameter; 0.13 mm @ 1.75 mm; 0.22 mm @ 2.85 mm
- 2-axis – 20% of diameter; 0.36 mm @1.75 mm; 0.59 mm @ 2.85 mm
- 1-axis – theoretically infinite
For 1 and 2-axis lasers, orientation of an out of round or oval shaped filament can give inaccurate readings of the mean diameter due to the restricted field of vision, as opposed to a 3 axis which will give a highly accurate mean value, regardless of orientation.
This restricted view can also give very inaccurate measures of ovality in general, with it being undetectable/falsely reading as zero depending on shape and orientation.
The actual specs of most quality laser micrometers appear to be similar in terms of resolution (on the order of 0.000001 in) and sampling rate (thousands per second). What is more important is how the user interprets the data. You could imagine that viewing this amount of data in real time would be impossible, so an average value over a certain length or time is used.
Essentially, based on however the measurement device is set up, diameter/ovality measurements can be made to look as good or as bad as the manufacturer is willing to deal with… Averaging over longer lengths of filament smooths out the graph, making monitoring much simpler, but at the expense of missing fluctuations.
To put some actual data behind this, we looked at data logs from our laser micrometer. The data in the table below shows PLA Orange running at production speeds while varying the number of seconds for each averaged data point. Each set of data is over a 1 kg spool of material. Keep in mind that each averaged data point contains thousands of actual measurements.
|Seconds per avg. meas.||0.1||1||1.5||2|
You can clearly see the trend towards a “tighter” tolerance, simply by changing how much data is averaged for a single visual data point. Again, the tolerances can be made as tight as you want.
For comparison to see where we stood, we looked at other Brands, which includes videos and literature discussing the measurement of their filament.
- Leading Brands
- Uses 2-axis measurement.
- In general, we found based on the refresh rate of the display, many manufacturers are looking at an average value over ~5 seconds. At production line speeds, this is over a fairly large length. As shown above, this takes a lot of the noise out of the chart, because it’s smoothing out many of the fluctuations. Small defects, or even large but abrupt ones would never be seen.
- Many manufacturers quote sampling rate as “several” times per second.
- Uses 3-axis measurement (advantages discussed above).
- Average value every 0.1 second, or 50X more frequent.
- Audible/visual alarms for out of tolerance measurements.
- Sampling rate ~1800/sec.
- Equates to being able to catch very small defects, over very small lengths of filament, even at faster line speeds than others.
We hope this blog has clarified what quality filament is and how we are different. Give us a try and see for yourself!