Luckily not during the prints, but afterward I get these long oozes coming out of each nozzle. That means it’s even oozing past the flow guard. Isn’t there a slicer setting to add extra retraction at the end of the print, or must I do it all myself using g-code? Surely everyone is having the same experience, so shouldn’t this be fixed for everyone rather than each of us coming up with our own solutions?
1 Like
Is that PETG?
Also, never seen that purple plate before, what is it?
It’s PETG in one nozzle and PLA in the other. I’m using the PLA for the supports on a PETG print, but it could just as easily be the opposite. It’s a perfect fit for the H2D’s capability.
The purple plate is a kdeavi. It’s similar to a cryogrip, but about half the price. Some people prefer it. I bought it to test, but it turns out I prefer the cryogrip for getting a perfect first layer over the entire build plate. However, for smaller print models, like this one, the kdeavi works just fine for me, and so I save my cryogrip for the truly big prints.
1 Like
I have the same problem and I’m really tired of trying to remember to clean my nozzles after every print. Often I forget and the oozed piece(s) end up on the bed during the next print.
I’d think a retract prior to cutting the filament would be the fix but I’m so out of touch with Bambu’s gcode I’m too scared to attempt it. Hopefully there’s a solution for this.
3 Likes
Exactly! You explained why the issue matters better than I did in my OP.
I’m noticing there’s an experimental feature called “Long retraction after cut”. Maybe it is the answer.
It’s meant to do a long retraction before changing filaments, but maybe it works at the end where the filament is cut prior to rewinding it? I’m fuzzy on just exactly what’s happening when, so I could be off by a mile.
Well, it’s worth a try, I’m finishing a print right now and ready to send another off. I’ll try it and see if works or not. The print is a 2 hour print, I’ll let you know how it goes afterwords.
Thanks!
1 Like
At the end of the print we know that it rolls the filament back onto the AMS. I presume that a cut must precede that action, otherwise it would be pulling all of the filament back and there’d be nothing left to ooze out of it. If it’s cut, then there’s no way after the cut occurs to retract what’s in the melt zone, because the cut occurs between the melt-zone and the extruder. So, if we are to fix this with g-code, we’d need to issue a big retraction before the cut happens. Putting in the retraction g-code in post-print g-code section would be too late. So, as a first step, we’d need to find where those cut commands are, assuming the cut commands were implemented as g-code and not baked into the firmware somewhere beyond where we can inspect.
1 Like
Well, my Bambu gcode skills are far to weak to know where to look. If we knew where the last cut is made (assuming it’s in gcode) then we could increase the retraction or add one prior to the cut. I know it needs to retract prior to the cut in order to work, just not sure where it’s at (yet).
It performs the cut by ramming the print head into the finger-like projection that pops out just beforehand. I’ll have a look tomorrow to see if I can locate where in the machine g-code section that occurs. I’m 90% confident it’s exposed g-code that we can find and edit. Once located, it should be easy to insert some kind of retraction command beforehand.
I’m just puzzled as to why no one has addressed this issue before, but I suppose there’s a first time for everything.
1 Like
Looking in the Machine G-code tab ,end G-code section. I don’t’ see the command to cut the filament, I do see where it does a retract then shuts things down including retracting the filament. Guessing it’s the M620 S65535 and or T65535 command. I think I’m going to try increasing the retract statement on my next print to see what happens. What’s the worst that can happen?!? 
;===== H2D =====================
G392 S0 ;turn off nozzle clog detect
M993 A0 B0 C0 ; nozzle cam detection not allowed.
M400 ; wait for buffer to clear
G92 E0 ; zero the extruder
G1 E-0.8 F1800 ; retract
G1 Z{max_layer_z + 0.5} F900 ; lower z a little
G90
M141 S0 ; turn off chamber heating
M140 S0 ; turn off bed
M106 S0 ; turn off fan
M106 P2 S0 ; turn off remote part cooling fan
M106 P3 S0 ; turn off chamber cooling fan
; pull back filament to AMS
M620 S65535
T65535
G150.2
M621 S65535
1 Like
Just came across this, I haven’t digested it yet but it might shed some more light.
I wonder if even just letting the fan and part cooling fan run longer before turning them off might be enough to cool the filament remaining in the melt zone (or at least the nozzle tip) enough to prevent it from oozing out. I mean, what’s pushing it out? Gravity? Thermal expansion by picking up residual heat in the hotend? I guess after the cutting happens there’s nothing left pulling it back, but why doesn’t it just sit in the meltzone like we wished it would?
Alternatively, letting it cool down more before doing the cut might achieve the intended effect.
So, several solutions seem like possibilities: 1. retraction, 2. cool more before cutting, 3. let the fans run longer, or some combination thereof. I think I’m now leaning more toward further cooling before cutting, as maybe if there’s anything left below the cutline it will just drip out anyway unless allowed to cool beforehand. There might be a risk of retracting too much if it pulls hot filament into the extruder, maybe it would jam. The whole point of doing the cut before rewinding is to ensure that doesn’t happen. Just saying without the background knowledge, we’re shooting in the dark, so maybe start with a risk mitigating strategy first.
1 Like
At least on mine it looks as though the flow guard isn’t pressing up enough against the nozzle. So, that would perhaps explain why both nozzles are oozing during the after print instead of only one.
Curiously enough, though, the left nozzle isn’t oozing during the print, even though the picture shows it isn’t completely capped off. That’s probably because I increased the filament retraction to 1mm from the default 0.4mm.
That flow guard is pretty delicate. It doesn’t take much to bend it out of alignment. What’s worse is that if it bends down too far, it can start colliding with the print.
This looks tricky in my opinion.
As you mention, a little lower and it hits the part.
Is this position this way, because it’s bent? Or is it just not nicely positioning anymore during a swap?
I read threads about the nozzle blocker not moving smoothly into place.
Eighter way, lets hope it’s not messing up your job halfway.
That nozzle blocker is bent, bend it back and problem is solved, even a new nozzle blocker need to be fitted to really do it’s intended job.
Yup. For unclear reasons, it kept dipping low and lower, until today it finally did snag and caused an unrecoverable layer shift while printing. No loose screws, so unclear why it happened,.
I followed the wiki youtube which presents its particular way of bending it flow blocker (nozzle blocker?) back into position, and it was easy.
Our H2D at work experienced a bent flow blocker in the first week we got it. It was due to a coworker sending sending a bad model that had floating parts. It hit the failed print and bent to 90 degrees. Ended up replacing it since bending it back wasn’t really going to work well. At least it was easy to do, but it made me realize how fragile and easy it is to ruin, so I purchased a bunch of spares when I ordered my h2c.