A dumb problem with bed leveling when printing nylon and a solution
So, I wanted to finally start printing with Nylon and bought a Prusa PA bed plate and some PA GF filament. I wasted a LOT of hours... LFMF
The problem was, bed leveling would often fail on some parts of the bed and always on others. I tried everything from cleaning to calibration to trying to start the print with no filament (to prevent ooze being a factor). Nothing helped.
Then I simply switched the plate to regular PEI after a failed leveling and it ran normally. Put the PA plate back in, fail again.
It seemed obvious that the PA bed plate was faulty in some way, so I started a chat with support. After waiting for an hour I had to explain everything three times, make them a video of the PA and then PEI plate leveling, and explain everything three times again. All this to learn three things:
1. Leveling fails when the probe hits something soft. Normally this would be filament ooze.
2. As soon as you print on a plate, you are on your own; forget any kind of warranty on a 100€ bed plate 🙄.
3. There is no way a plate could cause this, no sir, it's impossible.
So I got to thinking some more, and then it hit me—Prusament PA CF runs really hot, at 280°C, and the printer runs on high temp during leveling. And I had these nozzle marks on the plate after leveling. What if the PA plate material gets softer when the really hot nozzle touches it? The firmware interprets this as ooze and does a nozzle cleaning cycle. So I simply reduced the nozzle temp to 170 after a failed leveling and retried. It almost passed, so I lowered it to 90 (same as bed) and it passed immediately.
So, if you are printing nylon on a Prusa PA bed plate, you have to reduce the tool temp a lot before leveling. You can do this by inserting
M109 S90
before the
M107 ; turn off the fan ; MBL
lines in startup gcode. This will cool down the tool before leveling and make the leveling work with a Prusa PA plate.
RE:
So I dug some more and found the starting nozzle temp setting in the startup gcode under the line:
G29 G ; absorb heat
There you can set up the desired temp for nozzle cleaning and leveling, just add a separate entry for PA (it's combined with PC by default) and change the value to the one you want:
M109 T{initial_tool} S{((filament_notes[initial_tool]=~/.*MBL160.*/) ? 160 : (filament_notes[initial_tool]=~/.*HT_MBL10.*/) ? (first_layer_temperature[initial_tool] - 10) : (filament_type[initial_tool] == "PC") ? (first_layer_temperature[initial_tool] - 25) : (filament_type[initial_tool] == "PA") ? 200 : (filament_type[initial_tool] == "FLEX") ? 210 : (filament_type[initial_tool]=~/.*PET.*/) ? 175 : 170)} ; wait for temp
This would be the proper way to do it, I guess.
RE: A dumb problem with bed leveling when printing nylon and a solution
THANK YOU SO MUCH. I just wasted 2-3 hours on nozzle cleaning failed and bed failed to level. I set my temp to 90 using the M109 command above...
THIS WORKS!!!
RE: A dumb problem with bed leveling when printing nylon and a solution
I'm happy that I could help 🙂 Do note that this disables a lot or all of the nozzle cleaning effect, so you should clean it manually or you could have issues with the first layer.
RE: A dumb problem with bed leveling when printing nylon and a solution
this helped a lot. wow.
RE: A dumb problem with bed leveling when printing nylon and a solution
I´ve got to try this on Monday.
I have wasted the whole Friday morning trying to clean the nozzle and build plate.
Will write a follow up here.
RE: A dumb problem with bed leveling when printing nylon and a solution
I´ve got to try this on Monday.
I have wasted the whole Friday morning trying to clean the nozzle and build plate.
Will write a follow up here.
Follow-up:
This completely solved my issue. Of course, the nozzle needs to be clean to begin with.
In my specific case, I also had to create a new material profile on the printer, because the standard temperature for PA is 285°C and for some reason the XL heats the tool to this specific temperature before nozzle-cleaning where my filament oozes like crazy (Formfutura Styx PA6 --> prints at around 220°C).
RE: A dumb problem with bed leveling when printing nylon and a solution
Thanks for saving my sanity.
I had just upgraded the firmware on my XL before attempting to print some PA pieces. It seemed odd that it couldn't level the bed. Blamed it on the new firmware, installed what I had before, and had the same issues. Spent a few hours futzing around, and then searched the forum.
You G-code fix saved the day!
RE: A dumb problem with bed leveling when printing nylon and a solution
I have to try this with PP sheet, because I have same issue - MBL does not go through. Has anybody tested or had similar issue with PP sheet as well? Thanks
RE: A dumb problem with bed leveling when printing nylon and a solution
EDIT - Come on @TheParrotGuy, you saved my day.
M109 S90
Tthis helped. I cannot put there filament code, because I am printing lot of different filaments on PP sheet but "hard temp" on the sheet works.
RE:
By reducing the nozzle temperature, you reduce the pulses signal (PWM) that the heater receives. The PWM pulses affect the loadcell, as described in this bug:
https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/5261
Disabling the heating also works. Using this workaround, I need to compensate for nozzle thermal expansion in Slicer (+0.06mm)
Here is a video showing the problem and the solution:
min 00:00 - 00:40: if heater is on the Prusa XL fails bed leveling (and the leds flicker). The heater is heated using pulses that trigger the load cell when probing or tarring.
If I disable the heater, there are no problems with bed leveling (min 00:40 - end). But using this workaround, I need to compensate for nozzle thermal expansion in Slicer (+0.06mm).
RE: A dumb problem with bed leveling when printing nylon and a solution
I need to compensate for nozzle thermal expansion in Slicer (+0.06mm).
Could you please explain that in a little more detail?
Mods for Core One: Core One HT 450 degrees, Comfortable display , Very fast print start and Reducing noises
Mods for Prusa XL: Very fast print start
RE: A dumb problem with bed leveling when printing nylon and a solution
I need to compensate for nozzle thermal expansion in Slicer (+0.06mm).
Could you please explain that in a little more detail?
After bed leveling is done with cold nozzle, the nozzle will expand (grow) when heating, so the first layer will not be ok. The tool need to be raised 0.06mm to compensate for the longer nozzle.
RE: A dumb problem with bed leveling when printing nylon and a solution
How did you figure that out? Did you measure it? Calculate it? That's interesting.
Mods for Core One: Core One HT 450 degrees, Comfortable display , Very fast print start and Reducing noises
Mods for Prusa XL: Very fast print start
RE: A dumb problem with bed leveling when printing nylon and a solution
How did you figure that out? Did you measure it? Calculate it? That's interesting.
the first layer was not printing ok, then I live-z the print until I found the value. https://help.prusa3d.com/article/live-adjust-z_112427
RE: A dumb problem with bed leveling when printing nylon and a solution
By reducing the nozzle temperature, you reduce the pulses signal (PWM) that the heater receives. The PWM pulses affect the loadcell, as described in this bug:
https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/5261
Disabling the heating also works. Using this workaround, I need to compensate for nozzle thermal expansion in Slicer (+0.06mm)
I thought about your theory, and it does not make sense, at least with my XL:
- No other bed surface ever failed bed leveling with a clean nozzle.
- Since the nozzle leaves visible impressions in the nylon bed surface, the softness of the bed coating at high nozzle temps is clearly the cause.
- If PWM signals were interfering with the load cell, the problem would recur with any high-temp filament on any bed surface, but it doesn't, at least with my printer.
So while it is feasible that heater pulses would interfere with the load cell (as a design flaw), it sure does not look like they do, at least for me.