When we are trying to produce parts that have to interact with each other print size accuracy become one of most iportant factor. From a simple articulated figure to the most advanced mechanical processes, the precision of 3d printing becomes fundamental.
Obviously, in this article we will not deal with the adjustment of precision from scratch through firmware, but with all those chemical-mechanical aspects that contribute to not achieving the dimensional accuracy that we expect.
The lion’s share of this problem is certainly made by the first layer.
The plate not well levelled is the main cause of the lack of precision for the Z axis.
Let’s assume to print with the height of the layer of 0.2 mm and to have the nozzle at 0.04 mm from the plate, the first layer will be coated and compressed in 0.04 mm and the finished model will lack 0.16 mm.
As it’s easy to guess this inaccuracy on the first layer will also modify the next layers and we may find ourselves having the error amplified.
We recommend that you follow our guide for adjusting the nozzle-bed distance and print the calibration cube that we find at this address, to verify the actual goodness of our work.
Another aspect to keep in mind when you find a more or less serious lack of dimensional accuracy is overextrusion or underextrusion.
Material in abundance can “fatten” the model a few millimeters as a poor extrusion can make it more frail.
If you find that you have this kind of problem you can rely on this article about over and under extrusion.
Print size accuracy: Size analysis
Once we have found that we are in the case of the dimensional error and that we are not in the two previous cases, we will have to make a slightly more in-depth analysis and understand whether the difference in measurement in the size of the print is constant or proportional.
You can do a very simple test, print the calibration cube 100% (2 cm per side) and then 200% (4 cm per side). In case of oversizing the first cube will be 20.1 mm and if it is a constant error the double cube will be 40.1, in case of oversizing the proportional double cube will be 40.2.
It may seem like a subtle difference but you should deal with these problems in a totally different way.
Print size accuracy: Constant size error
The constant size error must be solved with the slicer. Some of them have a special function to compensate for horizontal dimensions.
Print size accuracy: Progressive size error
The progressive size error in most cases is due to the intrinsic properties of the filament in use. Many materials tend to shrink as they cool down, just like metal.
For some types of filament this shrinkage can even reach 5% of the size, and this can become a very serious problem.
In this case we can act cunningly and, once the shrinkage has been determined and the entity determined, we will increase the size of the model to fill the gap.
Let’s take an example, always with our cube. The 2 cm cubic cube printed after shrinking is 1.9 cm. So to print the cube with the correct size we will have to do 100/real value * design value, then 100/1.9*2 which is equal to 105.26.
Then we will set in our slicer an increase in size of the model and we will indicate a value of 105.26. Even if in the vast majority of cases the value 105 will be sufficient as a correction.