I did not want to take another topic off-topic but there was a post I wanted to comment on which I feel may promote a misconception of accuracy, precision and in the one poster’s opinion, the subject of inexpensive calipers and wasting one’s money.
Accuracy vs. Precision
Summary
Although these two topics are often used interchangeably, they are very different.
You can all do your own research but I am trying to do a synopsis of what can generally find on the web. Here is one decent one on the formlabs website
- In the first example we have three 10mm object with an advertised accuracy to within ±0.10mm
- Object #1 is 10.01mm
- Object #2 is 9.99mm
- Object $3 is 10.05mm
All three objects are considered accurate because they are within the stated tolerance of ±0.10mm
- In the case of a second set of three 10mm object which do not have a stated tolerance.
- Object #1 is 10.2mm
- Object #2 is 10.2mm
- Object #3 is 10.2mm
All three object are considered precise because they are all identical to each other.
Using the analogy of a pistol range target. Bullets that don’t hit the bullseye but are consistently in the same spot are considered precise. Bullets that hit inside the target but may have some scatter are considered accurate.
And obviously my screen clippings are neither accurate or precise, please forgive the asymmetry.
So when we are talking about printer accuracy, unless we are making parts to mate to other parts, what we really care about is precision not so much accuracy. By that I mean, I don’t care if my 10mm dowel is 10.1mm so long as my 10mm hole that the dowel will fit into is also 10.1mm. I’m simply more concerned with reproducibility. Of course it would be nice if it were both precise and accurate But accuracy is really only important if I am fitting a part to something I didn’t make but had measured. An example of this would be printing a replacement control knob for an appliance, if I wanted it to fit right, I need the measurements I took to be both accurate and reproducible in my printer.
Commercial vs industrial FDM printers.
Summary
If one does a simple Google Search on FDM PLA accuracy, one will generally find that the common tolerance is within ±0.2mm. To put that in perspective, contrast that with most filament companies who advertise ±0.02 – ±0.03mm for 1.75mm filament.
During my search, I could not find advertised tolerances for any of the consumer models but in the industrial arena and the Industrial models dance around the issue without making specific claims.
If a slicer has 3 digits of accuracy, it should be obvious that the technology generally will not deliver any more 3 digits of accuracy. This akin to what the Manufacturing folks I hear tell the Design Engineers, “Just because you can make a pretty picture in CAD, doesn’t mean it can be built in the real world”. This is something I think we all forget. The software can be programmed with 64 places to the left of the decimal point, we all know that’s math and does not generally reflect the real world. Or do we know that? It would seem that many post things contrary to this logic as was recently witnessed in the thread on STL vs STEP quality.
I’ll also point out to some of the recent posts here complaining about the lack of accuracy on an X1, well… can you really compare a $599 P1 to a $6,500 Ultimaker S5?
Cheap vs expensive calipers
Summary
It was suggested in this post that we should “ditch The uncalibrated random Amazon calipers and get get a real instrument.”
Without casting shade on the member who has a right to there opinion, do I really need an expensive set of calipers if all I am doing is making parts for myself? So long as they fit my design, shouldn’t that be what should prevail? True, if I were making precision parts for sale, it would be a different matter, but then again, I wouldn’t be using FDM technology for such a task would I?
So it got me wondering how precise and accurate are cheap calipers? Well, I took a 25mm calibration bar and used my two calipers. Admittedly this is a universe of only two samples but it is my universe. 
Here are my findings using my two cheap Amazon calipers. I measured at both the top of the jaws and the bottom of the jaws.
The cheap calipers
The even cheaper calipers
Draw your own conclusion
So if one is using a device that has a precision of ±0.2mm does one really need an expensive set of calipers? Is a set of calipers that has ±0.01mm good enough?