I didn’t notice until maybe a half hour later, as I was busy with other things. It’s an H2D printer, two different colors of the same brand HF-PETG, a YOOPAI build plate heated to 70C before printing began and kept at that temperature during the pause. Chamber heater not activated. At the time the picture snapshot was taken (about an hour later,shown here, after the print was resumed), the print was still underway.
Once I inspected the situation and determined that there was no nozzle clumping going on, I resumed the print and it continued. The print did not lift from the bed. So, why would a pause create such an obvious print artifact?
Here’s my guess: even though the build plate remained heated, the hotend cooled during the pause, resulting in the temperature inside the chamber cooling, which in turn resulted in some amount of unwanted shrinkage, which caused the discontinuity to become visible when the print resumed. Therefore, I’m guessing that if I had been running with the chamber heater turned on, even to the minimum 40C required for it to operate at all, then maybe this would have been avoided.
As you know, layer time can be a huge influencer. Even during printing, a longer layer time due to extra infill, already shows layer lines
Normally the top most layer is not fully cooled to chamber temperature during an average layer time. The temperature of the part is transitioning to chamber temperature in the layers below.
Your 30m break basically made the entire part cool down, and probably the chamber temperature dropped even lower due to the missing heat coming from the hotend.
If you would have heated the chamber to 40C, the effect would probably be less, but I don’t think you would really get rid of it. For a visual layer line, there is not much difference required.
I don’t think that a 30min interruption in any print can be resumed without a visual artifact.
It’s off by about 0.4mm at it’s maxima, which is toward the center of the walls. By that I mean, it kinda bowed inward toward the center of the walls. So…what I now think happened is that during the print interruption, the sparse infill at the top was most directly exposed to the relatively cool air of the chamber (around 35Cish), and even though it was gyroid, it began to shrink, pulling inward on the walls of the interrupted print, and the walls would have been least able to resist toward their center.
If I take that as the etiology of what happened, then I agree with your conclusion: probably no way to avoid it. Chamber heating might have reduced it a bit, but that’s still well below print temperature. Maybe 100% solid infill might have helped some. Probably much thicker walls would have helped. It had just two 0.42mm walls (inner and outer). Reacting a lot faster would have helped.
Its basically a rule of thumb that almost any production process that is interrupted, will show a quality decay in the time period of stop to start. Especially processes that involve heat. But also processes that involve solvents or curing agents.
Challenge is to maintain a reasonable steady state to minimize the impact.
Biggest contributing factor: Cooling shrinkage.
Second? Repeated heating of colour pigments. Some pigments if reheated from different originating temperature can have different color due to physical difference.
no, if you use a thermal camera, you can clearly see that even if you have a chamber heater, the printed layer is cooling rather quickly after printed.
I can fully agree with this statement, however, the additional amount of stress between all the layers could give a serious impact on warping imho.
Preferably you would like to speed up The slow layers.
You possibly could if you would have infinite heating and cooling capacity that you could use at the correct moments of the process.
There are already AI solutions out there that dynamically adjust nozzle temperatures when the layer times are getting shorter. It also requires a thermal analysis running along with the slicer to improve speeds and temperatures.
Not that I am trying to promote Helio here, but they are trying to improve the quality and performance this way.
No. Slowing down is always better to relief stress.. Because slowing down means less heat to dissipate and that gives the molecules better chance to stabilize, re-align themselves (they’re chains) and release the stress.
At what point, if ever, does printing become too slow? By that, I don’t mean layer time, but mm/sec print speed, like we can set in our filament profile. By default it’s 10mm/sec, which seems fairly brisk.
too slow is that the previous layer has become too cold, and that would make the bonding weaker. Actually, Helio is doing a very good job at visualizing that, if strength is what you’re looking for, try use Helio to analyze it.
Slowing down also means that the temperature difference between the previous layer and the current one is bigger, causing more internal stress between each layer.
If you could print faster, with the same temperature you are indeed adding more heat/energy/time unit. But the temperature difference between the individual layers becomes less. But it requires a hotend that can keep the temperature constant while melting more plastic.
It’s a complex process and any process that involves heat is very very sensitive to fluctuations.
We just have to wait for the 3D printer that builds the part on a molecular level.
This definitely gives a good idea of the cooling process.
Not knowing what material is being printed and what the temperatures of heatbed, and chamber are, I still believe you only see half of what’s going on.
When looking at injection molding, cooling comes from the mold itself. Cycle times are mainly driven by the cooling capacity of the mold and the wall thickness. Opening an injection molding too fast, rips the part inside in 2 pieces with liquid plastic between
Here in the thermal imaging you see the outside temperature of the part. The core of the bead is still cooling down and shrinking the part further. Constant layer times are crucial for quality. The thermal behavior of the material is a big factor also.
PLA is quite forgiving, while PETG is already less stable. The higher the nozzle temperature relative to the ambient temperature, the more the shrinkage effect increases.
See how annealing deforms un-filled materials deforms parts shows you how much residual stress is in there.
Maybe we should explore 3D printing techniques that don’t use heat.
Anyone having an update on the development status of the “replicator”?
I think has do with cooling. By moving a print forward and to the right, the layer line pretty much goes away. Even because I only pause less than 30 secs, if the print is in the rear of the printer, I get a layer line.
The external part cooling fan of the H2 series can definitely be an influencer. Mine is already equipped with a deflection port.
the cool air intake at the front (assuming no heated chamber is used) can have a significant impact also. The further you come towards the front, the more likely that the cool intake air will flow over your part instead of against it.
However, with larger parts you don’t always have the flexibility to move your part around much.
I am having the same line at the same height on a similar cubic part.
that is related to layer time and the top layer of the bottom of the container / model lifting from the plate (this exclusively maybe)
I have to print a dozen parts so I hope to give you other news soon. the last cube i printed had the lesser line but the error was i did cast a wrong plate print, bed felt at 90°c from 110°c (Asa) and it lifted a bit, print ended “enough good” so i kept it. now i am printing again… and without brim because i am stupid, but this is good for getting more xp so the next print it will be with brim.
