Nozzle gets closer to the bed every time during first layer (worse at higher temps).
I'm hoping somebody has had this issue before. I've ran through as many calibrations as I can find and just recently found the one here:
https://github.com/HelgeKeck/RatOS/blob/documentation_v2.1/site/docs/configuration/beacon_contact.md#6-first-print-and-fine-tuning
Everything works, even my beacon comes back with a 1 for the latency.
After the calibrations I'll print a large first layer - probably around 200x100mm
It'll start out PERFECT and about half way through the nozzle starts moving closer to the bed and I can see the lines start squishing out, if I adjust the z up to compensate (0.025 at a time it will adjust and keep printing good) - this is the fine tuning part further down on the page.
That's all find and dandy and the print will keep going nicely.
The issue then is the next print I do. The beginning of the first layer (which was perfect before at the beginning) is now too high because I saved the final z adjustment from before.
Today I used the "save_z_offset" command in the console and it supposedly adjusted the thermal setting, but after a couple more prints consistently back to back (just incase the bed wasn't heat soaked enough) and I still got the same consistent z height issues. Either too high at the beginning, or good at the beginning and too low at the end of the first layer.
This is also happening on all 3 of my ratrig 3.1 with beacons. I've ran non-ratos versions of klipper and the newest ratos with all the updates - doesn't seem to make a difference, different nozzles (E3D and Phaetus). Right now I'm putting an old pinda on one of the setups to see if it has the same issue ( I don't remember having this before I got my beacon sensors ).
Any ideas?? I'm at a loss and I can't babysit each of these machines through their first layer all the time - I shouldn't have to.
GitHub
RatOS/site/docs/configuration/beacon_contact.md at documentation_v2...
The preconfigured Raspberry Pi image that makes it easy to run Klipper + Moonraker + Mainsail on your printer. - HelgeKeck/RatOS
14 Replies
I believe many of us are having the same issue. I always have to increase z height by about 0.07 for it to start flattening out. The trouble is even though I don't save the z height it ends up re-meshing every print so I have to adjust again and again. There is a post here somewhere where Mikkel says to add an temp offset in the printer.cfg or something to save changes for different filaments.
Well, I'm glad I'm not the only one. I'm going to be testing out a pinda setup over the beacon and see if I have any better results. Hopefully it at least gets me printing again.
Ok, so I have an update - I'm pretty sure it's something with the beacon or the beacon programming. I installed a pinda probe on one of the printers I've been fighting, flashed a fresh copy of the newest ratos, PID calibrated the hotend, bed and did a proble_calibrate with the nozzle and bed at temp and printed my big test part and it layed down the first layer as consistent as I expected it to - the nozzle did not get closer to the bed at all. Hope this helps anybody else. I'll be running the pinda on a couple machines for the time being until we hear more from RR or Beacon.
It's good to read through this and see if this is similar: https://discord.com/channels/582187371529764864/1292556582336335904
Same issue on VC4-500
Any solution for the Beacon already?
I had a similar issue on my 3.1 500 some time ago, until I started heat soaking for at least 45 - 60 min for PLA (70° C bed temp) and > 60 min (better 90 min) for PETG (85° C bed temp). On some occasions the problem still occurs (but not as bad as before), but I figured out that this might have to do with tiny fragments of filament under the PEI sheet or maybe air bubbles when I put it on the wrong way (start at the back and then "roll" it on towards the front to get all the air out). Although I can't see why the bits of filament or other debris don't get compensated by the mesh, as the beacon should measure the height of the PEI surface (I was thinking maybe the measurement is also influenced by the distance - ideally zero - between PEI sheet and the magnetic pad). Maybe this helps
i had the exact same issue. when i take a look on ratos variables, i can clearly see my nozzle thermal expansion compensation (toolhead0 0.0628mm, toolhead1 0.08mm) and while starting a print it shows me that this expansion compensation is active. my z-offset is so automatically at 0.063mm. well, every start i needed to add about 0.05-0.07mm to get more air while first layer runs and got really pissed off.
now what to do: in slicer add "SET_GCODE_OFFSET Z_ADJUST=+0.05" (or how much you would think to give extra distance) to your start code and send your print-file to the printer. after the printer got the file and starts doing all his bed mesh calibration routine, simply pause your print! the printer will do his full routine to get started and right after that primeblob (i personally disabled it) he will then go to park position and waits for your resume. wait for about 3 to 5min and then resume your print. your nozzle will now have his full expansion and the first layer will get perfect completely. try it out! spend several days on it to check what is happening. now it works. i would wish to set a timer into start config for 300sec right before he wants to start but i'm to dumb for that by now. maybe someone can help here with that??
small update to my last post: i know found out how to set that wished timer before printing. simply added "variable_bed_heat_soak_time: 0
variable_hotend_heat_soak_time: 600" into my printer.cfg and now he waits 10min before he starts printing. and what shall i say? he does that. what else? he prints a perfect first layer now! btw: i print fiberon pet-cf17 at 280 deg celsius and wondered if he does upscale his measured nozzle expansion value between 150-250 deg celsius and yes, he does. printered measured 0.62 and because i print at 280 he goes up to 0.81. NICE work! but because i still dont get the thing with save_zoffset and all this stuff i still stay at setting my z-offset in filament startcode. i'm done, it works. hopefully you all will get it working now too.
bringing this back up a bit... was your behavior similar to this image that I just took, where you have to bump up your Z offset progressively as the print moves through the first layer, because the nozzle is "slanting" down into the build plate (poor term, but gets point across I think)?
In that image, you can see each of the bands where I had to bump up the Z because the nozzle started to dig into the print bed. If I leave it running without adjustment it just digs in completely and doesn't work at all. my skew calibration has been run a couple times, I think it's alright, but not really sure what else could be doing this.... the beacon should be offsetting for this?
not necissarily looking for a solution, just wondering if you were experiencing the same behavior or if I was mis-interpreting the post
Nope, that's the same issue that I've been having for a long time. Out of my 3 RR400 v3.1, 2 have the issue and one is fairly decent. Don't really notice the issue with lower temp plastics like PLA, but it's definitely there with ABS.
lovely. gives some credit to warping/thermal issues I suppose. there's a thread that I found that's like 2100 messages long, the jist (i think) is that there's bowing in the cross-slide or something that casues the issue but i'm not really sure how much I buy that, considering how much Z adjustments i'm having to make.
curious, does your second layer require any adjustments? when I print, I have to bump my Z height like .3mm or more which should cause issues with my second layer since the offset is different, but it's almost alwasy perfect
which leads me to think it's a software issue with how the bed mesh is applied and not a mechanical issue. idk though, not super stron on the software side of things
I'm pretty sure it's somewhat software related. I didn't experience any of these issues before I switch to the beacon and ratOS. I lost all of my old configs or I'd easily go back to plain klipper. I'm still debating building it from scratch though, seemed to be more consistent.
yeah. considering myself as well, i've got a voron 2.4 running stock klipper, pretty good experience with it so far.
wondering if it's a beacon specific issue or a beacon+ratos issue, or just a ratos issue
I was able to get much better results if I updated the start_print macro to heat up everything first then start taking measurements. I noticed that it would home first and set the Z then heat up and I'm sure that isn't as accurate as doing it once it's hot. Or at least re-homing and setting Z and stuff again after it's hot and heat soaked.
I had a similar problem when i did long first layer time prints. I have created a macro that i place in my start gcode before start print macro where i heatsoak the gantry 20 degrees higher then normal printing temp.
So what i do is home the machine, park the toolhead and then place the bed at 5mm and gantry in the middle.
For PETG I put the bed at 100 degrees from 30min to 60min. In the macro i disable all motors until heatsoak is done and then I home and z tilt again and then the start print macro is called
Right now I’m experimenting with a adaptive heatsoak that takes soms parameters of the print job and determines the temp and how long to soak
Problem is that because you have steel and aluminum in the gantry how hotter it gets the more it bends down so the nozzle comes closer to the bed