I am about at my wits end here. I have had this happen on a couple of different printers, different filament, etc. Currently running it on a BambuLab X1c. I have also seen it on completely different models from different creators.
Filament is Eryone Silk Coextrusion, though I have had it happen with just plain PLA. Filament is dry.
Print temp is 230. I know a little high but for silk on a Bambu, it needs to be hotter.
I am using 16 top layers and have tried 2 or 3 different patterns, but mostly montonic. It is much worse if i use default top layers. Gyroid infill at 8%, though have tried several types of infill.
I have even rotated it 90 degrees on the bed to no avail.
Different speeds make no difference.
Does anyone have any earthly idea what could be causing this?
I really appreciate any advice. I normally don’t ask for help much as I usually find most of my answers doing searches and just plain trial and error, but this one has me stumped.
230°C is far to hot for PLA. It is even for most normal hotends as the PTFE-Tubing in the nozzle is evaporating toxic nerve gas.
205°C should be enough. The surface is simply remelting each time your nozzle is driving along and unlike the smaller sections on top the heat is probably stuck in the infill. Also make sure your bed is around 55°C. Depending on the surface of your print bed (e.g. glass) you can go 5°C up, because the sensor is below the bed and the print on the top is at a lower temperature. Higher bed temperatures cause soften the print on the bed and causing it to melt down.
I appreciate the reply; however, Bambu printers change things somewhat. Because of their speed, Silk PLA needs to have higher temps to melt properly and handle the speed. I can try lowering it some, but if I go too low, then the filament will not be able to keep up. This is the temp I run silk with all the time, but it is only models with flat surfaces like this that have problems on the edges. It also seems that it happens mostly on models like this where the flat surface is going to be surrounded by walls.
I also had a similar problem with a different model on my Prusa about a year ago and was using the traditional temps. It was just plain old PLA.
I will try your suggestion when my current print finishes on that printer. Not sure if I will go as low as 205 though because when I did a temp tower on this filament it did not like it that low.
Well, yeah. Print speed and print temperature are dependent to each other.
The 205°C is usually for print speeds around 60mm/s. However if the filament is rated for 190-220°C this should not be increased above these values. Even if the filament is moved very fast, during moves it is not.
While looking up this filament, I came across this Amazon review of the filament that might be of interest:
Printed with my modified Bambu Lab P1P (enclosure, aux and chamber fans, AMS). Started off with the Generic PLA Silk Profile but bumped my first layer nozzle temp to 230C and then other layers to 220C and cranked cooling a bit. Pressure Advance (K Factor in Bambu-ease) set to a modest .020.
The Bad:
Too high of a K-Factor lead to some Over-Extrusion
Ran the nozzle at 215C and had some weird extrusion that failed to adhere, literally looked like raw filament that just oozed out
Some defects that were partially too cold of a nozzle for the speed, and potentially could use some drying (Could also be a model issue since this was a Bambu Maker World Beta model, also the fact my Pressure Advance was too high along with Max Volumetric Flow being a bit high as well)
The Good:
The bad issues are easily correctable and just common 3d Print stuffs
With an outer wall speed of 50mms and the 220C nozzle temp, it has a very nice shine to it
Spool is nicely wound. No issue feeding in the AMS
The Nice To Know:
The picture in the listing doesn’t do complete justice. The colors on this, unless you shine a light directly on it, are very dark. Still shines super nicely, but its not a bright pink/purple color
Eryone has migrated to cardboard spools. I haven’t had issues with my AMS, but it is worth mentioning
Specs on the spools are pretty spot on
For folks using Bambu Lab printers, the Generic PLA Silk filament profile is pretty spot on
I haven’t seen this issue before, but I’d look into possible warping given the size of the part, trying higher infill (20%) if infill is used in the center area, or slowing down the top surface print speed.
Thanks. I have scaled the model down and been running some tests. Lowering the temp made little difference. I have been tweaking some other settings with varied results. Right now, I am trying it with the generic silk pla profile and a bit denser infill.
One other anomaly I have found is that along one side, where the angle of the wall changes, there is a strange roughness. It goes from about 45 degrees to 90 angle. It also seems to be somewhat independent of the other issue.
I might be able to upload a pic.
It only happens on this one side.
I have used this brand and type for a number of other models with no issues. Temp always at 230, and my calibrated profile. In this case, it likes to mess up, on multiple rolls of various colors.
Is that the side facing away from the large cooling fan? It seems to correspond (comes and goes) with the wavy sides at the bottom of the part as well.
As you are familiar with the filament, maybe it is the model. Just to cover all bases, I’d throw the model into Meshmixer to repair any errors that might be tripping up the slicer.
I think I am getting a handle on it. At least it is improving. Your suggestion about the infill seems to have helped, though I only increased it to 15% to save some time and material. Those diagonal lines are strange, but I can probably take care of that with ironing.
I did notice on the ends there are some oddly shaped lines, but I suspect they are seams. Since they happen on both ends identically. Strange thing was they didn’t show up on the slicer preview. Bambu/orca slicer have some limited seam settings, but I am playing with it. Here is a pic.
Yeah, I calibrate with Orca, but always have had problems with getting PA right because my eyes have trouble telling the small differences between each increment.
On the other hand, I got this good enough to go to the customer. An insert is going into it. So will hide what is left. I will continue playing with it, but at least I know I am on the right track. I may up the temp back because this filament looks shinier a bit hotter.
The problem appears to be using such a low infill per cent. I think it left some pretty big gaps that translated into the top layers. Raising it from 8-20 per cent helped tremendously.
Well, yeah. it is usually better to start from the other side when it comes to infill.
Using 20% or so to start with and then go lower with the iterations. If you design the parts yourself the opposite is feasible, as you may need to reprint anyway.
Infill is also a criteria when it comes to rigidity, even so outer perimeters (number of walls) are important, too.
Over time you get a feeling about what is required for the final part to meet all criteria.
I’m aware, but in this case, I was trying to save some time and filament as we are selling these wholesale to a local gaming store. It is one half of a box arrangement, that will have an insert, so I wasn’t too worried about rigidity. What I wasn’t used to is doing things with a flat top and how that interacts with the infill.
I’m running some PA tests right now, but that is going to be hit and miss. My eyes aren’t so good, so I may have to try a couple of times to get it right. It is just hard to tell which setting is best in the calibration tests, at least for me.
What I don’t understand is why, even with the higher infill and extra layers, the top layers are still a bit rough on the edges while the center is pretty smooth.
