I can imagine that the expansion of the Z-thread spindles during warmup of the chamber and the cooling down of the actual cube afterwards might be of influence on the deviations, but it seems to be almost twice the theoretical spindle expansion (approximately 0.18mm) for a length of 330mm and 45C temperature difference (20C ambient vs 65C chamber), and the shrinkage of the material afterwards would account for.
I was printing a calibration cube of 30mm in PETG (measured height 29.96mm) and the same cube in PAHT-CF (measured height 29.58mm), which is a difference of 0.38mm.
The theoretical shrinkage of a PAHT-CF cube of that size would be approximately 0.027mm.
What I have not even taken into account is the fact that the frame of the printer would eventually also expand, and should compensate for the expansion of the thread spindles.
So the only thing that comes to mind here is that the X and Y-slide might be bending due to the high chamber temperature and that possibly the bed is also still deforming a bit during the that time.
One thing I still have in mind is to heat soak the entire printer (chamber and bed) for 2 hours and run the print again to see if the deviation changes.
Does anyone have some thoughts on this observation?
Or any other ideas that might minimize this height deviation?
I planned to run the 2hour heatsoak , but electricity decided otherwise and I had a power outage about 10 minutes before I started. Long story short, when it finally came back i didnt have anymore time to do that, so please post if you do it and what the results are.
Given the whole “prosumer”“high accuracy” blah blah regarding the machine, these factors are starting to pile up in a disappointing summ…
I don’t see this immediately as a flaw, but more as a learning curve.
I have a Voron 2.4 and the big advantage of that one is that the thermal expansion of the height is much less due to the gantry that starts at the bottom.
The Trident however is also having a bed on thread spindles and possibly has the same behavior.
I am currently waiting out my 2 hours, to give it a try. Hope it won’t go in standby though.
Îťah , the context defines everything. On a random corexy? a kit? yes definetely a learning curve. On a printer marketeed as prosumer with the best accuracy possible and ready out of the box?? Flaw. Sorry.
To put it simply its all defined on what you paid for.
Once I can say I love the printer, you will see. As long as I got a prusa/ratrig/voron with a bambu label on it and I need an equal amount of tinkering to make it work as expected, you wont.
Ask me about the p1p i got , I have NOTHING bad to say about it, its perfect and everything as promised.
H2D is not. Sorry that I dont want to sugarcoat it with copium…
This would help if the height of the part is not a multiplier of the layer height.
For instance you try to print a cube of 29.9mm heigh with a layer height of 0.2mm. It would then try to adjust the last layers to have the last one exactly at 29.9. Otherwise you would end up with 29.8 or 30mm.
I do believe it has to do with the expansion/deformation of components during the first minutes/hours of the print in a heated chamber.
open to interpretation ? and other enviromental factors ?
i did read recently that some users compensate for shrinkage with a little over measurement ,cnc kitchen also has an interesting video of the subject , but its a little above my pay grade as im still learning
Well, that is based on the general expansion factor for PAHT-CF, but of course the shrinkage is also dependent on the orientation of the fibers. I personally would say that shrinkage in Z might be more due to the horizontal orientation of most of the fibers.
Well I have been doing my 2 hour heat soaking and ran my test cube again.
Same result! The height difference with PETG-HF in a cold chamber is still around 0.4mm (29.55mm).
I have been looking at the thermal expansion factor for PAHT without carbon fiber, and that should take 0.1mm of the 0.4mm deviation I have.
If anyone has some logical explanation for this, I am happy to hear it?
I am not having any ideas that could account for the remaining 0.3mm deviation, unless the firmware does something in the background that is causing this.
I have done another test with a test part of Polylite ASA, and that has a much more accurate height (29.92mm) than the PAHT-CF has.
It seems like the PAHT-CF is the reason for the deviation.
Though I still cannot think of a reason what the cause would be.
Definitely the high precision of the printer can be taken out of the equation in my opinion.
Weird thing is that the final height is determined by the last layer printed and therefore the shrinkage of the last and maybe previous layer (0.4mm) could cause a minor deviation.
This still not explains why PAHT-CF has such a large height difference.
While ASA was printed with the same chamber and bed temperature.
