2
$\begingroup$

I realize this issue (warping) has been repeatedly addressed on this site. I've just graduated to high-temp filaments (PC in particular). I don't know much of the physics of this. I'm wondering whether the degree to which the filament contracts is proportional to the amount that it cools. If the answer is yes, then wouldn't it suggest that a lower printing temperature might reduce warping-as the temperature interval over which the filament cools is smaller? Or perhaps the difference is negligible?

Also, I see a lot of emphasis placed on good first layer adhesion. Is this still an issue if you are printing on a raft?

$\endgroup$

2 Answers 2

1
$\begingroup$

The answer already provided by @fred_dot_u at the time of writing is good, so I won't rehash what they already said, but will try to answer your questions from another angle:

I'm wondering whether the degree to which the filament contracts is proportional to the amount that it cools.

Yes. This is generally true for any solid material. This property is linked to the fundamental nature of matter in the way we understand it today.

If the answer is yes, then wouldn't it suggest that a lower printing temperature might reduce warping-as the temperature interval over which the filament cools is smaller?

Shrinkage is the root cause of 3D prints warping, however warping itself happens because of the differential in temperature between layers: when a hot layer is extruded on top of a cold one and begins to contract, it will apply a compressing force on the layer underneath, bending it.

In fact - if warping were a function of shrinkage - an enclosure would do no good: sooner or later the print would cool down to room temperature, and would warp. The reason why - contrarily - an enclosure works, is that it limits the differential in temperature between layers (which causes warping) and lets the entire print too cool uniformly and slowly afterwards.

So, would...

...a lower printing temperature reduce warping? [...] Or perhaps the difference is negligible?

Nothing beats real world data, especially when the issue has so many variables that are difficult to account for, as in 3D printing, so I would invite you to simply try to print the same model twice, only changing the printing temperature (and of course making sure the environmental temperature is the same), and see if it does.

From a theoretical standpoint, I could argue both ways.

On one end, I could argue that this is nothing different than using an enclosure set at a slightly warmer temperature than the environmental one, so it will reduce warping (even if by not much).

On the other hand, I could argue that until the filament is solid enough, it won't be able to "pull" the layer below, so it doesn't matter if the extrusion temperature (fluid state) is 230°C or 210°C, if until 190°C the filament won't begin to "pull". So warping will be identical.

Also, I see a lot of emphasis placed on good first layer adhesion. Is this still an issue if you are printing on a raft?

Yes, as also the raft has a first layer that needs to adhere to the build plate. Rafts typically have a discontinuous and over-extruded first layer over a large area, which definitively helps with adhesion, but you still have to make sure the raft sticks well. In my experience it is far easier for a raft to come off the build-plate than for the model to come off the raft. YMMV though, as the material of the build-plate, and the slicer can dramatically affect this.

$\endgroup$
3
$\begingroup$

I can't address polycarbonate specifically, but can provide a general overview of the higher temperature filament considerations.

Printing on a raft means that the adhesion temperature of the filament is accomplished. This temperature is the factor to be considered if you are thinking of dropping the printing temperature. If you drop below recommended minimums, you risk losing adhesion to the build plate and also inter-layer bonding. That alone means one should use caution when dropping printing temperatures.

Printing with a raft usually means the model's individual parts have such a small footprint that they would not remain bonded to the build plate. Rafts are also used on printers with an uncertain planar surface or irregularities in the surface. That's not applicable to this question, generally speaking.

Your question about contraction being proportional to the amount of cooling is perhaps misdirected. One could consider that the printing temperature is a manufacturer specified value and the cooled temperature would be generally considered room temperature. Room temperature would be addressed as a range, rather than a single value, but even as a range, there isn't going to be a big percentage of variation in the calculation involving the print temp/room temp.

My experience with the higher temperatures is more related to the volume of material per cross section (in all three dimensions). A printed model of substantial height with a relatively small horizontal cross section (think cylinder) is likely to have much less distortion in the x/y plane and greater distortion along the z-axis. The mass of filament cooling in the z-direction generates greater force than the smaller mass on the x/y axes.

Another factor in such thought processes is that layers are on the x/y axes and the strength of the extruded plastic is more homogeneous through the nozzle, while the z-direction creates inter-layer discontinuities, making warping and delamination easier.

I've found that I can reduce (but not eliminate) warping and delamination if I am able to maintain chamber temperature for longer periods and reduce temperature slowly. Unfortunately, I have a semi-enclosed printer and the heat loss is dependent partly on the ambient air temperature. A fully enclosed heated printer with auxiliary heating under some form of control may give you the best results.

$\endgroup$
3
  • $\begingroup$ Thanks! I actually wasn't speaking about lower ambient temp, but print temp. I think this material has a recommended range of 235-275 or something on that order. So to clarify, would cooling from 235 to room temp result in less contraction then 275 to room temp? 40C is a significant temperature variation. $\endgroup$
    – David Reed
    Feb 10, 2018 at 3:50
  • $\begingroup$ The warping you are experiencing it the cylinder example is likely due to uneven cooling and/or asymmetries in the model, but not to "more mass cooling on the z axis". In fact the height of the print subject to cooling at each given moment is probably 2-3 layers, and the extrusion of new material on top will compensate for any shrinkage of the layers below. That said: good answer, +1! :) $\endgroup$
    – mac
    Feb 10, 2018 at 6:45
  • $\begingroup$ So basically: it's a misconception to think the warping along the Z axis is due to shrinkage on the Z axis. It's still due to shrinkage along the X and Y axis, where longer portions of continuosly extruded filament are cooling, and where the tension between layers will build up at each layer. $\endgroup$
    – mac
    Feb 10, 2018 at 6:56

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .