RE: Bed Leveling Failed on CoreONE upgrade
Loosen the thumb screws and lower your nozzle 1 mm or so and tighten them again. That fixed all my issues.
RE: Bed Leveling Failed on CoreONE upgrade
If you have a Core One - try this.
Gently push the rear corners with the tip of your finger. The bed flexes downwards with very little force.
Compare this to anywhere else - there is no movement whatsoever.
RE: Bed Leveling Failed on CoreONE upgrade
My cooling shroud is a good 2-3mm higher than the tip of the nozzle.
RE: Bed Leveling Failed on CoreONE upgrade
I wonder if the issue is if the nozzle is firmly up, it is pushing on the main plate and not moving the load cell as much. Except on firmer bed positions the whole extruder is giving.
Am trying this now....
Loosen the thumb screws and lower your nozzle 1 mm or so and tighten them again. That fixed all my issues.
RE: Bed Leveling Failed on CoreONE upgrade
Yeah - that's what happening... I dropped the nozzle 1mm from topping out and one probe only to level across the entire bed.
Prusa need to document this.
Thanks @discojon
RE: Bed Leveling Failed on CoreONE upgrade
If you have a Core One - try this.
Gently push the rear corners with the tip of your finger. The bed flexes downwards with very little force.
Compare this to anywhere else - there is no movement whatsoever.
On my Core One, yes, the rear corners of the bed can be bent down with finger force. But producing any perceptible flex takes significantly more force than what it takes to produce a delta of 100 or 200 counts on the load cell.
RE: Bed Leveling Failed on CoreONE upgrade
Yeah - that's what happening... I dropped the nozzle 1mm from topping out and one probe only to level across the entire bed.
Prusa need to document this.
I dunno... Prusa has documented explicitly that the hotend should be pushed all the way in. I believe the design and the assembly process are unchanged since the MK4 Nextruder. It seems unlikely that this is an issue in all units they make?
Push the hotend assembly all the way into the heatsink. There should be approximately a 2 mm (0.08 in) gap between the heatsink and the brass part of the nozzle.
RE: Bed Leveling Failed on CoreONE upgrade
Maybe your thumbscrews were tightened with too much pressure against the nozzle?
I'm happily swapping between 0.25 and 0.4 nozzles and what I do (I sadly can't use the nozzle changing tool due to my hotend cable being routed a tiny bit too short):
- Loosen thumbscrews and put one finger slightly against the nozzle, only holding it vertically in place.
- Slowly & gently take away my finger, the hotend power cable easily holds my hotend.
- Grab the hotend with the left hand and use only a 7 mm nut, no lever or anything. Make sure there's no tension on the cables!
- Screw out the nozzle
- Gently let go of the hotend
- Gently put the other nozzle through the hotend hole and guide it into the extruder
- Tighten the nozzle in the hotend with the 7 mm nut, while holding the hotend with the other hand (no tension on the cables!)
- Now the important part: put one finger on the tip of the nozzle and gently and slowly push it in, until you hit end end-stop. Apply very very little pressure, don't hold the hotend in place. The hotend will rotate into the position with the least tension on the cables.
- Tighten the thumbscrews, while the hotend can rotate and you finger very softly holds the nozzle against the end-stop.
- Done
The issues with the 1 mm gap:
- Filament can rub against the nozzle begin, spreading filament dust everywhere in that space
- Especially softer filament could bent in the gap during during retraction/deretraction, causing a clog etc.
- heartbreak might be less efficient
- print fan cooling might be less effective and rather cool the nozzle, instead of the flowing filament
In my opinion: if the 1 mm nozzle gap works, either the fan duct is hitting something or there was pre-tension on the nozzle/hotend/thumbscrews.
RE:
I'm not saying it's the right things to do as I agree with all your points. Until we get a software fix, it seems to fix all the issues we are seeing. I firmly believe the failed probing sequences is related to handling of the results in the load cell. But for now it works.
Maybe your thumbscrews were tightened with too much pressure against the nozzle?
I'm happily swapping between 0.25 and 0.4 nozzles and what I do (I sadly can't use the nozzle changing tool due to my hotend cable being routed a tiny bit too short):
- Loosen thumbscrews and put one finger slightly against the nozzle, only holding it vertically in place.
- Slowly & gently take away my finger, the hotend power cable easily holds my hotend.
- Grab the hotend with the left hand and use only a 7 mm nut, no lever or anything. Make sure there's no tension on the cables!
- Screw out the nozzle
- Gently let go of the hotend
- Gently put the other nozzle through the hotend hole and guide it into the extruder
- Tighten the nozzle in the hotend with the 7 mm nut, while holding the hotend with the other hand (no tension on the cables!)
- Now the important part: put one finger on the tip of the nozzle and gently and slowly push it in, until you hit end end-stop. Apply very very little pressure, don't hold the hotend in place. The hotend will rotate into the position with the least tension on the cables.
- Tighten the thumbscrews, while the hotend can rotate and you finger very softly holds the nozzle against the end-stop.
- Done
The issues with the 1 mm gap:
- Filament can rub against the nozzle begin, spreading filament dust everywhere in that space
- Especially softer filament could bent in the gap during during retraction/deretraction, causing a clog etc.
- heartbreak might be less efficient
- print fan cooling might be less effective and rather cool the nozzle, instead of the flowing filament
In my opinion: if the 1 mm nozzle gap works, either the fan duct is hitting something or there was pre-tension on the nozzle/hotend/thumbscrews.
RE:
If you push the nozzle all the way in, the stop you feel is the sticky out part of the main plate. This is on the other side of the load cell, so will be dampening the load cell by acting as a post/prop that bridges the load cell.
I'm 100% certain that this is the issue - proven by the fact that leaving a slight gap makes it work flawlessly.
RE: Bed Leveling Failed on CoreONE upgrade
RE: Bed Leveling Failed on CoreONE upgrade
3. Grab the hotend with the left hand and use only a 7 mm nut, no lever or anything.
That's a false friend in translation, I guess, which could cause some confusion. You are probably referring to a socket -- from a socket wrench, without using the actual wrench handle/lever? To my knowledge, "nut" in English exclusively refers to the counterpart for a screw (German Mutter), not to this socket (Stecknuss).
RE: Bed Leveling Failed on CoreONE upgrade
All the other printers with the same hotend arrangement have beds that are better supported all round. So the probing has greater resistance and something gives enough to still affect the load cell - enough.
The rear corners of the core one, whilst more than adequate for printing on, are flappier - they noticeably deflect under very little load.
I've always pushed the nozzle firmly up whilst tightening the thumbscrews (on the mk4s and XL) so I suspect the attention I have been giving to doing so is why I have been affected by this now on the Core One.
I messaged Josef Prusa and after some discussion he's pointed an engineer at this thread. Let's see where this goes. Hopefully this just requires some documentation on inserting the nozzle with some slack and no one else will suffer the same annoyances.
Yeah - that's what happening... I dropped the nozzle 1mm from topping out and one probe only to level across the entire bed.
Prusa need to document this.
I dunno... Prusa has documented explicitly that the hotend should be pushed all the way in. I believe the design and the assembly process are unchanged since the MK4 Nextruder. It seems unlikely that this is an issue in all units they make?
RE: Bed Leveling Failed on CoreONE upgrade
Oops, yes... A bit embarrassing. No idea, what my brain did there. Thanks for correcting it!
3. Grab the hotend with the left hand and use only a 7 mm nut, no lever or anything.
That's a false friend in translation, I guess, which could cause some confusion. You are probably referring to a socket -- from a socket wrench, without using the actual wrench handle/lever? To my knowledge, "nut" in English exclusively refers to the counterpart for a screw (German Mutter), not to this socket (Stecknuss).
RE: Bed Leveling Failed on CoreONE upgrade
I would like to add some information to the topic.
I have problem with TXT sheet yesterday a could not get pass the 49 points (25 ok) so I tried to find what is happening.
with support got this https://help.prusa3d.com/article/uneven-bed-31111-core-one_856294
unfortunately no difference. so I grab my old Raspberry PI with octoprint and ran bed levelling visualisation and got 2.6mm difference in bed level. that is the culprit why my printer didn't passed the test.
so this is what I found.
when I do Z calibration. the left Z lead screw keeps the bed to high. prusa support told me I have to follow the link mentioned above.
but I went different way:
I just manually lowered my left Z lead screw turning it by hand anticlockwise till it shows it is flat.
the problem is, when the bed hits the bottom, I have to do it again. Prusa support wasn't able to tell me how to fix that my left side is higher.
but I still having this Z issued when it suddenly starts printing higher.
I can print the same model, same filament and same sheet on my MK4 and it's perfect. the sheet is clean 100%. you can actually see how the filament drops from higher level
no idea who is the fix here :((
RE:
Some have printed a shim and placed it below that lead screw at the bottom so when it homes at the bottom it's correct. I imagine you could attach something to the bottom of that lead screw as well.
Others have had luck by just loosening a screw at the back of the baseplate and pulling it up and retightening. I'm not sure which one because I didn't have a C1, but that what I've read.