Bitten by bad filament again…

posted in: 3D Printing | 0

So my struggles with the questionable filament I recieved by 3DXtech a while ago has gone to a new level. I have had a theory that the first spool I recieved (since their reps even told me the batch number doesnt make sense for the material) was CF-PETG. Now CF-PETG isnt bad… by any means. But it’s not as temperature tolerant as Nylon.

So I checked out the new printer after doing a few 1hr prints of PETG (which requires a pretty hot heated bed) and I saw this:

No that’s not the perspective of the camera – Look at the line for the 2020 rail at the top, and look at the X Axis motor hanging down… Something has started to sag!

During the print the bed was at 85C which is enough to soften PETG, the motor was running ~ 40C as I was testing current ratings and acceleration/jerk settings.

The NylonX I have as the corner brackets for the Ultrabase (Which sit directly on the heated bed) are firm and not soft at all.
So my hypothesis is proving right, the reason the first spool of CarbonX printed so well, and has a smoother outer appearance than I was expecting was – it is CF PETG. I put a part on the heated bed at 85 and is was physically softer after coming to temp.

I’ve printed a few parts that are on the printer with this material… so I need to evaluate which to reprint. But the main one and the most critial one I never wanted to have to mess with after I set it up… is the Y/X gantry… To rebuild it not only requires about 10-12hrs of printing the new Y carriers, but taking ~ 1/2 the printer apart (or moving stuff around).

This suuuuuucks. But the printer is currently printing its own replacement parts. It can do it faster than my Anet AM8 and with higher accuracy, so why not. The X Axis motor on the other side that is also sagging a bit is in better shape than the one above. I turned the current settings down to keep the motor as cool as possible to not do further damage.

I also made some significant strength changes to make sure this doesn’t happen again.

The things I changed on the new design are:

  • The side brace was made 2x as thick, and removed most of the opening in it
  • the motor mount itself is .75mm thicker
  • Reduced some of the clearance cuts for the X Belt adjuster that werent needed.
  • Printing at .24 layer height instead of .2 – Nylon gets stronger the thicker the layers are.
  • Print with 50+% infill. I made a mistake with the last one doing it with 30%, but I didn’t think it would need more for NylonX as I was trying to keep this as light as possible.
  • Added a front screw/nut mount that will run between the belts from the X Motors and with some gentle tightening will brace the front of the motor with the top portion where the MGN12 rail and 2020 run through it.

Once these are done being printed the rebuild process looks like: (this is more me brainstorming this than anything):

  • Turn the printer upside down
  • slacken or remove the belts
  • Potentially remove the X and U carriages to make this easier and lighter (I am going to try not doing that.
  • remove the Y belts and disconnect the wiring to the Y carriers/X and U motors/Endstops.
  • Loosen the bolts for the MGN/2020 rail on both sides
  • Remove the 4 screws holding the Y carrier to the MGN15 blocks
  • Remove the Bad Y Carriers from the MGN and 2020
  • Remove the Motor from the Ycarrier.
  • Then do everything backwards. The calibration of the X Gantry is a trick though… took me a while to get it right before, its going to take me a while again as well. it involves not only all the work on the gantry and carriers, but I also need to loosen one of the MGN15 rails and let it center itself with the minor offsets generated with the new install.
  • insert the new carri

Overall this will be a couple days of printing and a full day of work… Fun.

Follow Joe Kelly:

Geek of many things

CEO of this Blog. It's kind of a big deal.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.