Improving Bed Mesh for Better ASA and PCTG 1st Layers

I’ve been having some challenges printing ASA and PCTG. I suspect it’s because my bed is taco shaped with the lowest points being the center of the left and right edges and the highest being the front and back edges. My bed isn’t particularly bad compared to others I’ve seen but the 0.7mm range seems to be enough to cause issues at the beginning of the print cycles. I also contacted creality only to realize that their flatness tolerance is 3mm…so I don’t think they’ll be any help here.

I suspect the bed is flattening (center raising) during the print which is causing some parts to jam the extruder without enough space and other parts to have really poor 2nd layers as the nozzle digs into the 1st layers. This is also a nightmare for filament tuning.

I could probably heat soak it when printing those materials but that’s just one more thing for me to remember/pay attention to. I’ve tried to use tape but had mixed results. I suppose I should use kapton tape since I suspect the masking tape I was using may be contributing to the thermal variance. However I ordered a piece of 1/8" glass cut to size with safe edges for $20 at a local glass shop. I still may need some tape but I’m hoping the glass interface will provide a significant consistency improvement.

Has anyone else put glass on the bed and, if so, has it been an improvement? Is the stock magnet strong enough to still hold the build plate securely or should I figure out a way to get some neodymium magnets under there (either taped or glued to the underside of the frame)? I suspect the tolerances of magnet sheets like this aren’t great so I’m trying to avoid putting anything between the glass and the build plate. Any other thoughts or ideas would be appreciated.

1 Like

Borosilicate or Neoceram/Pyroceram glass is your best bet for high-temp materials like ASA. Just a heads-up - tempered glass can shatter under thermal stress, and trust me, it’s not fun to clean up after. :grimacing:

The downside?

  • Pyroceram is heavier
  • Borosilicate at K2 bed size can get a little pricey

That said, a 0.7mm deviation really isn’t bad! Totally manageable with proper prep. In fact, I just finished up a full bed leveling guide using Kapton and aluminum tape to dial in high-precision meshes - my K2 is down to a 0.056mm deviation. I’m planning to have it up on YouTube and our forum sometime this weekend.

You can check out some of the highlights here:
:point_right: Dialing in the K2 Plus to a 56 Micron Bed Deviation – The Madness & The Method

Let me know if you want help tuning yours - I’ve been living in mesh view for weeks. :sweat_smile:

That’s really helpful info. I will check with my vendor about that glass type. My goal is to have the deviation down to less than my standard layer height of .16-.2mm.

I think I still need to tune my first layer height for other materials so it seems like, beyond different filament profiles, I’ll need to develop differing printer profiles for each material. Now that I say that, it sounds obvious. But I guess that’s how most things go as we gain experience.

It would be nice if slicers could tie certain aspects of the print profile to the filament profile (eg. first layer height, bridging settings, etc.) to reduce how many print profiles need to be managed.

Also note that some users report rather significant distortion due to temperature change; i.e. a bed that is good enough at 50C can be quite different at 75C.

I have been using the glass plate and have noticed it helps to heat-soak it for a bit to help consistency. Unfortunately the magnet isn’t strong enough to pull the PEI sheet flat with the glass so I have a magnetic sheet coming for the top of the glass.

I’m looking at the K2 bed from prcise printer parts Creality
I’m assuming/hoping once they go into production, the price will come down to the same price as the others (~$100). I think they also move the sensor to the top of the bed so i’m hoping that’ll help with getting more accurate temp readings along with the improved flatness.