@@jes00456To my understanding some purge lines/brim things will bypass the trash bin and you’ll find them days/months later somewhere in your print room/living area.
I just realized it's possible to use the same variables for adaptive meshing in Marlin. G29 has the L, R, F, B parameters that control the scanning zone. This should be the command, I kept a 10mm margin: G29 L{first_layer_print_min[0]-10} R{first_layer_print_max[0]+10} F{first_layer_print_min[1]-10} B{first_layer_print_max[1]+10} I just tested it on my machine (Marlin 2.1.2.1), and it works.
@@SirVergil I don't have a Bambu printer, so I checked the print profile of the P1P. It looks like something similar was already implemented in the start G-code: G29 A X{first_layer_print_min[0]} Y{first_layer_print_min[1]} I{first_layer_print_size[0]} J{first_layer_print_size[1]} I'm not sure if this is correct, can someone with a Bambu machine check what I'm saying?
That's neat but adaptive bed meshing in Klipper also reduces the number of probes based on the first layer area to save time. This method wouldn't do that, but it would get a more precise mesh of the actual print area instead of the entire bed. Good job!
If it’s possible to perform integer division in your slicer’s g-code compiler, then you could set a fixed number of millimetres between each probe. Edit: on the Slic3r page for conditional g-code, it mentions that it supports functions from the C resource exprtk, which includes regular division and the modulo operator, so I think it’s possible. It would probably be easier to just use floor or ceiling or round using normal division, which it also supports.
For those with Bambu printers Click the edit button above plate type, (looks like pencil to paper)(when you hover over the button it says "click to edit preset") Click the second tab, "Machine Gcode". The top text box should be called "Machine start G-Code" At the bottom of that box, paste in: G92 E0.0; reset extruder G1 X{first_layer_print_min[0]-10} Y{first_layer_print_min[1]} Z0.8 F6000.0; position 10mm left from the lower left of the first layer G1 X{first_layer_print_min[0]-10} Y{first_layer_print_min[1]+30} E30 F360.0; extrude 30mm of filament in the Y direction G92 E0.0; reset extruder G1 E-0.5 F2100; small retraction G1 Y(first_layer_print_min[1]+40) f6000.0; move an aditional 10mm without extruding G92 E0.0; reset extruder ^ that will add the new fancy purge line talked about in the video. To remove the default purge line, remove the code (above but int he same text box) which says ;===== nozzle load line =============================== M975 S1 G90 M83 T1000 G1 X18.0 Y1.0 Z0.8 F18000;Move to start position M109 S{nozzle_temperature_initial_layer[initial_extruder]} G1 Z0.2 G0 E2 F300 G0 X240 E25 F{outer_wall_volumetric_speed/(0.3*0.5) * 60} G0 Y15 E1.166 F{outer_wall_volumetric_speed/(0.3*0.5)/ 4 * 60} G0 X239.5 G0 E0.2 G0 Y1.5 E1.166 G0 X18 E25 F{outer_wall_volumetric_speed/(0.3*0.5) * 60} M400
G-Code für ungültigen benutzerdefinierten G-Code konnte nicht erzeugt werden. machine_start_gcode Parsing error at line 247. Expecting a legacy variable expansion format G1 Y(first_layer_print_min[1]+40) f6000.0; move an aditional 10mm without extruding ^ Bitte überprüfen Sie den benutzerdefinierten G-Code oder verwenden Sie den standardmäßigen benutzerdefinierten G-Code.
i'm really liking the purge method on my xl, so easy to take off. i added a bit to reduce the height near the end of the purge to kinda cut off the extrusion before moving away to help reduce stringing from the purge line
I've always just used a skirt, but as you mentioned it gets to be a bit much, especially for larger prints, when the skirt completely surrounds the part. I'm definitely going to need to give this one a try. Thank you for this informative video.
I've just got into printing ABS and couldn't start printing the object without getting ball that would get caught in the fan duct, get pulled along and cause the entire line it was printing to be lifted. I wrote a CURA post processor specfically to read the first G0 co-ordinates (usually the move to the brim start point) and process them. It purges down one side in the normal way, across to the G0's X co-ordinate, then turns left and prints straight down the bed to the G0's Y co-ordinate. So in its initial printing it doesn't get a chance to run the duct over a stuck up tail. Works really well.
I always added a skirt when printing with my Wanhao Duplicator i3 Plus, stopped when I switched to Bambu Lab printers. The new printers are definitely an upgrade but one thing I miss about printing with my i3 is watching it print. It moved at a nice even pace and watching it was relaxing, almost hypnotic. 😌 The X1C and P1P are definitely not relaxing to watch so I’ve had settle for improved print quality and speed. 😂
Interesting! I have a Cetus 3D printer, and its dedicated slicer has used the same purge line for many years. But I never thought it could be applied to other 3D printers. Good job!
Amazing video, I just thought of one flaw… When you purge close to the model, you will either have to use your finger to remove it (adding oil to the build plate) or a sharp pit of tweezers or a knife which could scratch the bed I do like the advantage of it reducing the travel from the purge line to the print. I love to see new techniques. Thank you for the hard work you do for the community
I'll probably give it a try because why not, seems better than the traditional type of purge line, with the added benefit of not having to travel as far after
Ugh i hate not having Klipper. Don't want to bother with Klipperizing ny current printer since I'm planning to get a new one by the end of the year, but it has sucked missing out on stuff like this, adaptive temperature script, good resonance compensation, all the perks of Fluidd, etc. I really appreciate you doing this to give us Marlin users access to some of the good stuff lol. I'm definitely going to try it bc my purge lines often fail to clean the nozzle and instead just cause the oozed filament to fold or curl up onto the sides of it.
Skirt length and total extrusion depends on first layer perimeter. Might be more personal preference than actual printing issue what to use 😉 (I'm running both long traditional 2-way purge _and_ 2 skirt lines - 1 seems too easy to get loose and interfere - and don't feel like I'm wasting too much time or filament 😅)
I tweaked my purge macros on a Robox 3D printer years ago and it was always to get a no blob start print, then modded the native slicer to create. Skirt, but I can see your method of nixing KAMP is going to be more useful on my current 3D printers. Thank you for doing this.… have a beer on me.
Nice. I thought I would need to follow your video to do it on my two main printers, even though they have Klipper installed. Because it's a user-made beta version of Klipper on a machine with locked firmware. But as it turns out, I was able to install KAMP on them. Little bit of set up work and a lot of web searching little things, but I got it working.
A few questions that came to me while watching : - have you taken into account in this custol gcode to avoid having the purge line out of the print plate boundaries ? This is a real risk by printing pieces close to the border of the plate. From the video it does not look like this is checked. - how does this interract with print settings like "print by object" ? - do you think it would be possible to merge this method with the official purge lines from bambu printers, which does more that just purging 30mm of filament ? (Dynamic calibeation flow is done at the same time typically). I.e. the purge line frol A1s printers typically alrernates between 3 flows settings during this process. Would be really really good to allow just to move this purge line closer to the print onstead of doing a more basic one :)
This is great! I've definitely missed the KAMP purge feature with my Bambus. The delicate and hard to remove line along with the string and sometimes blob, have been one of largest (but still extremely minor) gripes. Can't wait to try this!
I'm still using a variation of your start code for purge from around '19. May I recommend using layers of blue tape under your flex plate to build up low areas. It's about 0.12mm thick. I have 3 layers in a couple of areas on my Ender 3. Thermal camera shows no adverse effect in heating of the flex sheet due to the tape. I haven't needed to run a mesh in months; just recall the stored mesh {55 & 70C have their own) and go.
thanks micheal i have kamp and klipper setup although didnt get adaptive purge to work correctly must of missed something thanks 🙏 will take another look 😊
For those using Marlin and Cura I use the following start code which can be set up in Cura Start code G28 ;Home G1 Z5.00 F3000 ;move table down G1 X10.0 Y10.0 Z0.25 ; Prime the nozzle G92 E0 G1 X100.0 E22 F400 G1 Z5.00 F3000 M900 K0 ;Set linear advance K factor G92 E0 extrudes 100 mm line and adjust E value to suit.
I've worked around this for years by just printing closer to the lower left corner and placing the seam also in the same position on the first layer. Its stupid but it works.
I like the KAMP purge line in Klipper but it doesn't seem to account for organic supports or wide brims. I don't know if this slicer gcode version does or not but knowing how to change the location is a good thing to know.
The purge line is the first opportunity to adjust the nozzle height by baby-stepping, so for that it can help to make the line a little longer and/or slower.
In my experience with prime lines, 30mm isn't enough, at least on my v0 where I sometimes have to increase the amount of skirt or the nozzle isn't primed enough. I've heard you mention KAMP before and had no idea what that was and I always thought I was mis-hearing you and thought you were saying CAM.
I kept meaning to do something like this. I wanted an up, right, down three sided rectangle a few mm from the bottom left corner of the print to maximize the ability to leave an nozzle drool bugger or blob in the purge line and not have it fall into the print. I want the purge line to be easy to remove and discard. Thanks!
There is a similar thing on the FlashForge slicer where you can set the length of the skirt. It means on big prints its not doing a massive skirt round the whole print
I use OrcaSlicer. There is a MINIMUM length for the skirt, if it's not at least that long it adds another loop. But I cannot find any maximum length where it stops if it gets too long. So I'm hoping that this "adaptive purge line" thing can help simulate that.
Donuts are cool too: G1 E35 F300 ; Extrude 25mm of filament at 300mm/min G4 P1000 ; Dwell for 1000ms (1 second) G1 E-0.200 Z5 F600 ; Retract 0.2mm and raise Z by 5mm at 600mm/min
This is a nice method, think i might switch from KAMP!!! What would be really useful as well would be a purge line\blob after a pause - this is when i tend to get long strings from pause location (usually 0,0) back to the printer (usually 175,175), but how to make it go Y-10, y-20, y-30 incrementally so as not to pick up the previous purge line?!
I made it slightly touching the model, just 0.3mm distance. When I release the model from the build plate, purge line is attached to the model with a tiny bit, which could be snapped off easily. I won't forget to remove purge line then.
Adaptive meshing is available in Klipper itself, you no longer need KAMP for it. I don't really find the purge line of KAMP that great. It purges quite a lot and I often had PETG from that purge sticking to the nozzle and because of the short distance to the print drawing an annoying and relatively thick string connecting it with the part. I actually went back to a purge line at the front of the bed, first extruding a blob in the air then moving the nozzle down and right laying that blob on the edge of the bed and attaching it due to some more extrusion on that bed edge, finally quickly moving away further to the riht with tiny retract and then quickly moving to the actual print.
Not sure if everyone knows this.. Prusa mk4 does actually measure the bed level on two points where it purges before the print. I was wondering how it will work in back to back prints and was relieved to see it just measures the existing purge height and adds another layer there
@10:40 Maybe this only works in OrcaSlicer, but would it be better to set the retraction line to use machine settings? something like...: G1 E-{retraction_length[0]} F{retraction_speed[0]*60}
This is a bit like the pre-extrusion setting in Flashprint slicer. It's like a skirt but you can specify the length between 50mm and 200mm and the offset distance so it doesn't do a full loop around of your print. It doesn't do a traditional purge line unless you add it to the start gcode.
I've been using this for a while... only problem I've discovered is that the purge line will intersect the brim if your using a large brim.... not sure how to fix it but its been great
The issue I had with such fat purge line very close to the print is that it can obstruct the way of the BL-touch sensor making print head jump every time probe bumps into purge line. Just be cautious about your configuration and put purge line further away from the print.
Great video, as always! - For me, your adaptive purge line is not nearly enough as I usually have decomposed material in the hotend I must purge out. I use Duet/RRF so I wrote my own purge line into gcode (similar to the way you do), that is a double purge line, the full length of one side, that starts filament and then wipes and cleans the nozzle at the very edge of the build plate moving up and down in the Z axis. This cleans the nozzle on two sides before moving to the skirt. The skirt is necessary to make sure that the Z offset distance definition is correct around four sides of the build area.
The purge line has never really bothered me, I keep my bed leveled to ~0.2 maximum offset across and that's always worked well. What bothers me more is that when it homes it always touches the bed at 0,0 and that leaves a small divot plus there's always a little blob of filament there before it starts printing if I don't push it off once it's ready to go.
Never needed purge. My printer's stock gcode had it but I removed it, instead leaving the nozzle parked at Z=0 against the bed until it heats up, and then, with nozzle still firmly against the bed, extruding a bit straight into it to prime and create some pressure.
Remember to offset yiour variable_purge_margin by at least the probe offset. But probably do 2x that just in case. Or do that + 5mm. Because the probe will scrape over the line if you don't since the purge line is much higher than your first layer.
@Teaching Tech : Michael, what about printers using Klipper firmware (with a PRINT_START macro called from the slicer).. From what I see, the *Slicer's* [Machine Start G-Code] section is limited to call PRINT_START, together with the temperature setpoints as parameters. Then all the g-code intended to initiate the print (homing, QGL, z-offset calibration, pre-heat bed and nozzle...) is located on the printer's firmware, in PRINT_START macro => In such situation, what would you recommend? A- Include the gcode in the printer's PRINT-START macro, located in Klipper's printer.cfg file ? B- Include the gcode in the slicer's Machine Start G-Code section, the line after PRINT_START is called? I've just checked, in both case, the commands are added in the gcode file. To my mind, comes: - extrusion rate for the purge line could be different for different nozzle sizes (and we have different profiles on the slicer) - but it would be cleaner / more sleek to have the purge line in the PRINT_START macro (slicer agnostic, ...) >> You toughts, best practice / sugestion?
another great video thanks Michael for sharing this info. I have added this script into the SV08 machine G-code and I get this error message when it is ready to do the purge line, I revert it back to the original and it all works fine. What am I doing wrong with this coding?
I understood the purge lines on the Bambu A1 was doing a pressure advance check? While I like the adaptive purge idea wouldn't this potentially cause other issues?
Thanks for this cool idea! What happens, if you print a skirt line? I'm using klipper integrated adaptive bed mesh and "stole" the purge routine from camp to print an adaptive purge line. But when I print a skirt, this lines won't be addressed by the min-position calculation and often the purge line overlaps with the skirt, which can lead to issues. Did you encounter this problem as well with your solution (just turn the skirt on and move it a bit away from the print to test it)?
I believe the purpose is to use this INSTEAD of a skirt. If you're printing an entire skirt, you're already priming the nozzle, so the purge line is redundant. Just pick one to use. Keep in mind that a purge line is always a set length while a skirt will grow based on the size of the print, so a large print will use a lot of excess material for no reason.
Does the code need to check if the purge coordinates are off the bed? I know with KAMP that it sends a message to the console if you need to remove the purge line before printing.
It's suitable when you have auto bed leveling, other way one time you just crush you nozzle into bed when install slighli large nozzle, when stick some new magnet for PEI and so on and completely forget about Z correction for this part. And it's luck if you Z changes greatly - it's chance that firmware detect significantly much current on stepper and stop panically. Because otherwise, when nozzle pressed into bed in, e.g., 0.2 mm, printer drag it accross and ruin bed coating.Ask me how i know. So i decided to use brim in any case. Anyway bed heating, homing and other Z_DOUDONG, and printing itself take a lot of time so brim time is neglidable, and it work out of box in any pice of slicing software and in any printer.
Ou, after some thinking i notice that's may be only in my case - i prefer tune Z offset other than mess with underbed screews (i level bed once and not come back). If you set bed level in a "traditional" way than this problem just not happen. Just had some unplesant memories pops us and force me to write upper comment without thinking.
I followed the directions for my A1 mini using orca slicer. It does, however, do the initial purge line like what happened to you in the first round. I deleted the same portion (extrusion cali line). What could be the issue? Is there another portion of the start g code I should be deleting?
What's wrong with the skirt? You have a good idea how the first layer will look, you don't see that with a purge since its a thick blob, no need to complicate things or reinvent the wheel.
What if my print is so large or positioned so close to the edge of the bed that there is no place 10 or 20mm to the side? What if I have multiple objects being sliced? Since its in the start g code it will do the purge line only once, right before starting the print of the first object, right?
Your slicer will throw up an error about a toolpath outside the print area. Yes, it will only do one purge even if you have multiple objects on the build plate. Also only one purge line when you have "complete individual objects" turned on.
I feel like there's very little actual benefits to that for most people. Tbh I'd more worried that the chunky line is more likely to fall off and intrude on printing process.
Could you maybe do a video about a PETG optimized purge sequence? I have a bambulab and the purge sequence for petg is horrible and oftel still leaves stuff hanging that eventually gets in the print.
Because skirts leave loose material near the print where they might get dragged into positions that interfere with the first layer. You need nozzle primed properly *before* it gets anywhere near the print zone.
My printer homes and purges at 350x350, but the print always starts closest to the 0x0 position, can this be changed in cura? There often is a bit of oozing across the bottom of the print that way..
Hold on. Does your x1c draw the filament calibration from right to left?! Is this another toilet bowl situation? 😂 i am in the us and my square portion of the filament calibration is on the right side of the bed. Just thought that was interesting.
Most printers I have seen have them on the left side. It doesn't really matter, though, as long as it starts the purge line closest to where the toolhead ends up after it's done doing bed calibration.
does not work with kobra, any way u can make soemthing for it.. it trys to do the line but since there is no filiment loading (kobra 3 combo) it cant do it... BUT other than that great video.
I made my purge a little lightening bolt. Takes a little more time but I like finding them all over the printer room.
😂
😂
Mind sharing?
Right on, Zeus! ⚡️
@@jes00456To my understanding some purge lines/brim things will bypass the trash bin and you’ll find them days/months later somewhere in your print room/living area.
I just realized it's possible to use the same variables for adaptive meshing in Marlin.
G29 has the L, R, F, B parameters that control the scanning zone. This should be the command, I kept a 10mm margin:
G29 L{first_layer_print_min[0]-10} R{first_layer_print_max[0]+10} F{first_layer_print_min[1]-10} B{first_layer_print_max[1]+10}
I just tested it on my machine (Marlin 2.1.2.1), and it works.
Will this work for a bambulab p1p? if so, can you make a tutorial for dumb people like me? thank you.
@@SirVergil did you not watch the vide0? For p1p just follow the steps from it
@@SirVergil I don't have a Bambu printer, so I checked the print profile of the P1P. It looks like something similar was already implemented in the start G-code: G29 A X{first_layer_print_min[0]} Y{first_layer_print_min[1]} I{first_layer_print_size[0]} J{first_layer_print_size[1]}
I'm not sure if this is correct, can someone with a Bambu machine check what I'm saying?
That's neat but adaptive bed meshing in Klipper also reduces the number of probes based on the first layer area to save time. This method wouldn't do that, but it would get a more precise mesh of the actual print area instead of the entire bed. Good job!
If it’s possible to perform integer division in your slicer’s g-code compiler, then you could set a fixed number of millimetres between each probe.
Edit: on the Slic3r page for conditional g-code, it mentions that it supports functions from the C resource exprtk, which includes regular division and the modulo operator, so I think it’s possible. It would probably be easier to just use floor or ceiling or round using normal division, which it also supports.
Thank you very much! Looking forward to seeing this easy improvement being picked up by the developer community and implemented in OrcaSlicer
Anything model-agnostic should be handled by your printer's start macro, like it is in KAMP, not your slicer
For those with Bambu printers
Click the edit button above plate type, (looks like pencil to paper)(when you hover over the button it says "click to edit preset")
Click the second tab, "Machine Gcode".
The top text box should be called "Machine start G-Code"
At the bottom of that box, paste in:
G92 E0.0; reset extruder
G1 X{first_layer_print_min[0]-10} Y{first_layer_print_min[1]} Z0.8 F6000.0; position 10mm left from the lower left of the first layer
G1 X{first_layer_print_min[0]-10} Y{first_layer_print_min[1]+30} E30 F360.0; extrude 30mm of filament in the Y direction
G92 E0.0; reset extruder
G1 E-0.5 F2100; small retraction
G1 Y(first_layer_print_min[1]+40) f6000.0; move an aditional 10mm without extruding
G92 E0.0; reset extruder
^ that will add the new fancy purge line talked about in the video. To remove the default purge line, remove the code (above but int he same text box) which says
;===== nozzle load line ===============================
M975 S1
G90
M83
T1000
G1 X18.0 Y1.0 Z0.8 F18000;Move to start position
M109 S{nozzle_temperature_initial_layer[initial_extruder]}
G1 Z0.2
G0 E2 F300
G0 X240 E25 F{outer_wall_volumetric_speed/(0.3*0.5) * 60}
G0 Y15 E1.166 F{outer_wall_volumetric_speed/(0.3*0.5)/ 4 * 60}
G0 X239.5
G0 E0.2
G0 Y1.5 E1.166
G0 X18 E25 F{outer_wall_volumetric_speed/(0.3*0.5) * 60}
M400
nice
G-Code für ungültigen benutzerdefinierten G-Code konnte nicht erzeugt werden.
machine_start_gcode Parsing error at line 247. Expecting a legacy variable expansion format
G1 Y(first_layer_print_min[1]+40) f6000.0; move an aditional 10mm without extruding
^
Bitte überprüfen Sie den benutzerdefinierten G-Code oder verwenden Sie den standardmäßigen
benutzerdefinierten G-Code.
Error occuring while slicing in Orca
@@josefno.1 Ah yes I got the same error. Let me know if you manage to work out what the correct variable is.
@@liamburgess3385 i am honest i dont know how to Coding. I thought you know how to
Ive been anoyed with having the purge line in The same place all the time.
Thank you for sharing a solution to that problem!
i'm really liking the purge method on my xl, so easy to take off. i added a bit to reduce the height near the end of the purge to kinda cut off the extrusion before moving away to help reduce stringing from the purge line
I've always just used a skirt, but as you mentioned it gets to be a bit much, especially for larger prints, when the skirt completely surrounds the part. I'm definitely going to need to give this one a try. Thank you for this informative video.
So awesome, thank you Michael.
I've just got into printing ABS and couldn't start printing the object without getting ball that would get caught in the fan duct, get pulled along and cause the entire line it was printing to be lifted. I wrote a CURA post processor specfically to read the first G0 co-ordinates (usually the move to the brim start point) and process them. It purges down one side in the normal way, across to the G0's X co-ordinate, then turns left and prints straight down the bed to the G0's Y co-ordinate. So in its initial printing it doesn't get a chance to run the duct over a stuck up tail. Works really well.
I always added a skirt when printing with my Wanhao Duplicator i3 Plus, stopped when I switched to Bambu Lab printers. The new printers are definitely an upgrade but one thing I miss about printing with my i3 is watching it print. It moved at a nice even pace and watching it was relaxing, almost hypnotic. 😌
The X1C and P1P are definitely not relaxing to watch so I’ve had settle for improved print quality and speed. 😂
Bruh 😂
Interesting! I have a Cetus 3D printer, and its dedicated slicer has used the same purge line for many years. But I never thought it could be applied to other 3D printers. Good job!
Amazing video, I just thought of one flaw…
When you purge close to the model, you will either have to use your finger to remove it (adding oil to the build plate) or a sharp pit of tweezers or a knife which could scratch the bed
I do like the advantage of it reducing the travel from the purge line to the print. I love to see new techniques. Thank you for the hard work you do for the community
I'll probably give it a try because why not, seems better than the traditional type of purge line, with the added benefit of not having to travel as far after
I'm trying this today on my Bambu printers, Thank you!!!
Ugh i hate not having Klipper. Don't want to bother with Klipperizing ny current printer since I'm planning to get a new one by the end of the year, but it has sucked missing out on stuff like this, adaptive temperature script, good resonance compensation, all the perks of Fluidd, etc. I really appreciate you doing this to give us Marlin users access to some of the good stuff lol. I'm definitely going to try it bc my purge lines often fail to clean the nozzle and instead just cause the oozed filament to fold or curl up onto the sides of it.
I must be missing something but doesn’t a skirt essentially do the same thing since is just outside the 3D Print?
I will stick with a skirt, as it's also a good way to check/dial in the first layer if you need to without affecting the actual printed object.
💯
Skirt length and total extrusion depends on first layer perimeter. Might be more personal preference than actual printing issue what to use 😉
(I'm running both long traditional 2-way purge _and_ 2 skirt lines - 1 seems too easy to get loose and interfere - and don't feel like I'm wasting too much time or filament 😅)
I tweaked my purge macros on a Robox 3D printer years ago and it was always to get a no blob start print, then modded the native slicer to create. Skirt, but I can see your method of nixing KAMP is going to be more useful on my current 3D printers. Thank you for doing this.… have a beer on me.
Nice. I thought I would need to follow your video to do it on my two main printers, even though they have Klipper installed. Because it's a user-made beta version of Klipper on a machine with locked firmware. But as it turns out, I was able to install KAMP on them. Little bit of set up work and a lot of web searching little things, but I got it working.
A few questions that came to me while watching :
- have you taken into account in this custol gcode to avoid having the purge line out of the print plate boundaries ? This is a real risk by printing pieces close to the border of the plate. From the video it does not look like this is checked.
- how does this interract with print settings like "print by object" ?
- do you think it would be possible to merge this method with the official purge lines from bambu printers, which does more that just purging 30mm of filament ? (Dynamic calibeation flow is done at the same time typically). I.e. the purge line frol A1s printers typically alrernates between 3 flows settings during this process. Would be really really good to allow just to move this purge line closer to the print onstead of doing a more basic one :)
This is great! I've definitely missed the KAMP purge feature with my Bambus. The delicate and hard to remove line along with the string and sometimes blob, have been one of largest (but still extremely minor) gripes. Can't wait to try this!
I'm still using a variation of your start code for purge from around '19.
May I recommend using layers of blue tape under your flex plate to build up low areas. It's about 0.12mm thick. I have 3 layers in a couple of areas on my Ender 3. Thermal camera shows no adverse effect in heating of the flex sheet due to the tape. I haven't needed to run a mesh in months; just recall the stored mesh {55 & 70C have their own) and go.
thanks micheal i have kamp and klipper setup although didnt get adaptive purge to work correctly must of missed something thanks 🙏 will take another look 😊
For those using Marlin and Cura I use the following start code which can be set up in Cura
Start code
G28 ;Home
G1 Z5.00 F3000 ;move table down
G1 X10.0 Y10.0 Z0.25
; Prime the nozzle
G92 E0
G1 X100.0 E22 F400
G1 Z5.00 F3000
M900 K0 ;Set linear advance K factor
G92 E0
extrudes 100 mm line and adjust E value to suit.
I've worked around this for years by just printing closer to the lower left corner and placing the seam also in the same position on the first layer. Its stupid but it works.
A brim and/or tabs help with adhesion as well as purging the filament from the nozzle.
Another great video. Just completed this and it works fantastic on my Sovol SV06!!
Thank you for the solution. I am going to try it on my creality ender 5 with biqu H2 V2S extruder.
I like the KAMP purge line in Klipper but it doesn't seem to account for organic supports or wide brims. I don't know if this slicer gcode version does or not but knowing how to change the location is a good thing to know.
It works. I tested some ridiculous brims and supports in Prusaslicer and this purge always stays outside of it.
Brilliant. Thanks for sharing Michael.
The purge line is the first opportunity to adjust the nozzle height by baby-stepping, so for that it can help to make the line a little longer and/or slower.
Klipper already supports adaptive bed meshing natively. I wonder how long until it supports adaptive purge?
I have been looking for something like this forever! Thank you!
In my experience with prime lines, 30mm isn't enough, at least on my v0 where I sometimes have to increase the amount of skirt or the nozzle isn't primed enough.
I've heard you mention KAMP before and had no idea what that was and I always thought I was mis-hearing you and thought you were saying CAM.
I kept meaning to do something like this. I wanted an up, right, down three sided rectangle a few mm from the bottom left corner of the print to maximize the ability to leave an nozzle drool bugger or blob in the purge line and not have it fall into the print. I want the purge line to be easy to remove and discard. Thanks!
This is actually REALLY nifty! Thank you!
I just made this change on my x1c. Getting ready to try my first print with it but if it even makes the line easier to remove it'll be worth it.
I believe there’s an option for max length for a skirt. That would also be a viable solution. I do something to 2:44 for my ender 3.
There is a similar thing on the FlashForge slicer where you can set the length of the skirt. It means on big prints its not doing a massive skirt round the whole print
I use OrcaSlicer. There is a MINIMUM length for the skirt, if it's not at least that long it adds another loop. But I cannot find any maximum length where it stops if it gets too long. So I'm hoping that this "adaptive purge line" thing can help simulate that.
Thanks!
Donuts are cool too:
G1 E35 F300 ; Extrude 25mm of filament at 300mm/min
G4 P1000 ; Dwell for 1000ms (1 second)
G1 E-0.200 Z5 F600 ; Retract 0.2mm and raise Z by 5mm at 600mm/min
it's so simple, yet so useful!
This is a nice method, think i might switch from KAMP!!!
What would be really useful as well would be a purge line\blob after a pause - this is when i tend to get long strings from pause location (usually 0,0) back to the printer (usually 175,175), but how to make it go Y-10, y-20, y-30 incrementally so as not to pick up the previous purge line?!
Thanks again for another valuable video. Qidi Xmax3 and Q1 Pro do this out of the box.
I made it slightly touching the model, just 0.3mm distance. When I release the model from the build plate, purge line is attached to the model with a tiny bit, which could be snapped off easily. I won't forget to remove purge line then.
Adaptive meshing is available in Klipper itself, you no longer need KAMP for it. I don't really find the purge line of KAMP that great. It purges quite a lot and I often had PETG from that purge sticking to the nozzle and because of the short distance to the print drawing an annoying and relatively thick string connecting it with the part. I actually went back to a purge line at the front of the bed, first extruding a blob in the air then moving the nozzle down and right laying that blob on the edge of the bed and attaching it due to some more extrusion on that bed edge, finally quickly moving away further to the riht with tiny retract and then quickly moving to the actual print.
Just as I wanted to do something like this myself, thanks!
Not sure if everyone knows this.. Prusa mk4 does actually measure the bed level on two points where it purges before the print. I was wondering how it will work in back to back prints and was relieved to see it just measures the existing purge height and adds another layer there
@10:40 Maybe this only works in OrcaSlicer, but would it be better to set the retraction line to use machine settings? something like...: G1 E-{retraction_length[0]} F{retraction_speed[0]*60}
This is a bit like the pre-extrusion setting in Flashprint slicer. It's like a skirt but you can specify the length between 50mm and 200mm and the offset distance so it doesn't do a full loop around of your print. It doesn't do a traditional purge line unless you add it to the start gcode.
I've been using this for a while... only problem I've discovered is that the purge line will intersect the brim if your using a large brim.... not sure how to fix it but its been great
You can adjust the purge distance from object.
@@0Gotland4Ever0 how?? I actually like the idea of removing KAMP and just using Orca
Merci beaucoup pour cette vidéo 👍
thanks! very helpful video!😄
The issue I had with such fat purge line very close to the print is that it can obstruct the way of the BL-touch sensor making print head jump every time probe bumps into purge line.
Just be cautious about your configuration and put purge line further away from the print.
Great video, as always! - For me, your adaptive purge line is not nearly enough as I usually have decomposed material in the hotend I must purge out. I use Duet/RRF so I wrote my own purge line into gcode (similar to the way you do), that is a double purge line, the full length of one side, that starts filament and then wipes and cleans the nozzle at the very edge of the build plate moving up and down in the Z axis. This cleans the nozzle on two sides before moving to the skirt. The skirt is necessary to make sure that the Z offset distance definition is correct around four sides of the build area.
Very useful, thanks 😊
Such thick purge line is taller than first layer. Fan ducts may collide with it.
There's more clearance than 0.8mm on fan ducts
The purge line has never really bothered me, I keep my bed leveled to ~0.2 maximum offset across and that's always worked well. What bothers me more is that when it homes it always touches the bed at 0,0 and that leaves a small divot plus there's always a little blob of filament there before it starts printing if I don't push it off once it's ready to go.
Ooooh, was your second ever video with a Solidoodle Printer? 🤩
Never needed purge. My printer's stock gcode had it but I removed it, instead leaving the nozzle parked at Z=0 against the bed until it heats up, and then, with nozzle still firmly against the bed, extruding a bit straight into it to prime and create some pressure.
going to use this thank you> :)
Remember to offset yiour variable_purge_margin by at least the probe offset. But probably do 2x that just in case. Or do that + 5mm.
Because the probe will scrape over the line if you don't since the purge line is much higher than your first layer.
Great video! Love your content bro!
@Teaching Tech : Michael, what about printers using Klipper firmware (with a PRINT_START macro called from the slicer)..
From what I see, the *Slicer's* [Machine Start G-Code] section is limited to call PRINT_START, together with the temperature setpoints as parameters.
Then all the g-code intended to initiate the print (homing, QGL, z-offset calibration, pre-heat bed and nozzle...) is located on the printer's firmware, in PRINT_START macro
=> In such situation, what would you recommend?
A- Include the gcode in the printer's PRINT-START macro, located in Klipper's printer.cfg file ?
B- Include the gcode in the slicer's Machine Start G-Code section, the line after PRINT_START is called?
I've just checked, in both case, the commands are added in the gcode file.
To my mind, comes:
- extrusion rate for the purge line could be different for different nozzle sizes (and we have different profiles on the slicer)
- but it would be cleaner / more sleek to have the purge line in the PRINT_START macro (slicer agnostic, ...)
>> You toughts, best practice / sugestion?
That was really cool. However if you add supports does the boundary you talked about change too to account for those?
It should. The variable used is based on the entire first layer (which should include the supports), not on the part(s).
Just checked, and yes, it does change the position based on the skirt of the supports
another great video thanks Michael for sharing this info. I have added this script into the SV08 machine G-code and I get this error message when it is ready to do the purge line, I revert it back to the original and it all works fine. What am I doing wrong with this coding?
I understood the purge lines on the Bambu A1 was doing a pressure advance check? While I like the adaptive purge idea wouldn't this potentially cause other issues?
Coming from your TTR channel and just found this, post on your other channel please!!
It's gonna be a spicy end to the season if Max lets Lando get much closer 👀
Thanks for this cool idea! What happens, if you print a skirt line? I'm using klipper integrated adaptive bed mesh and "stole" the purge routine from camp to print an adaptive purge line. But when I print a skirt, this lines won't be addressed by the min-position calculation and often the purge line overlaps with the skirt, which can lead to issues. Did you encounter this problem as well with your solution (just turn the skirt on and move it a bit away from the print to test it)?
I believe the purpose is to use this INSTEAD of a skirt. If you're printing an entire skirt, you're already priming the nozzle, so the purge line is redundant. Just pick one to use. Keep in mind that a purge line is always a set length while a skirt will grow based on the size of the print, so a large print will use a lot of excess material for no reason.
Does the code need to check if the purge coordinates are off the bed? I know with KAMP that it sends a message to the console if you need to remove the purge line before printing.
your slicer will give you an area about a toolpath outside the print area
Nice, good stuff.
It's suitable when you have auto bed leveling, other way one time you just crush you nozzle into bed when install slighli large nozzle, when stick some new magnet for PEI and so on and completely forget about Z correction for this part. And it's luck if you Z changes greatly - it's chance that firmware detect significantly much current on stepper and stop panically. Because otherwise, when nozzle pressed into bed in, e.g., 0.2 mm, printer drag it accross and ruin bed coating.Ask me how i know. So i decided to use brim in any case. Anyway bed heating, homing and other Z_DOUDONG, and printing itself take a lot of time so brim time is neglidable, and it work out of box in any pice of slicing software and in any printer.
Ou, after some thinking i notice that's may be only in my case - i prefer tune Z offset other than mess with underbed screews (i level bed once and not come back). If you set bed level in a "traditional" way than this problem just not happen. Just had some unplesant memories pops us and force me to write upper comment without thinking.
The Qidi configuration includes similar but better version - parallel lines with distance between them - it gives me information if z offset is right
It would be sick if prusa could ingrate this into there profiles for the mk3 and mini so they act similar to the mk4 and xl
I followed the directions for my A1 mini using orca slicer. It does, however, do the initial purge line like what happened to you in the first round. I deleted the same portion (extrusion cali line). What could be the issue? Is there another portion of the start g code I should be deleting?
Could you use the polygons from the exclude object functionality to make a purge line that fits the model more closely?
thx a lot
What's wrong with the skirt? You have a good idea how the first layer will look, you don't see that with a purge since its a thick blob, no need to complicate things or reinvent the wheel.
What if my print is so large or positioned so close to the edge of the bed that there is no place 10 or 20mm to the side?
What if I have multiple objects being sliced? Since its in the start g code it will do the purge line only once, right before starting the print of the first object, right?
Your slicer will throw up an error about a toolpath outside the print area. Yes, it will only do one purge even if you have multiple objects on the build plate. Also only one purge line when you have "complete individual objects" turned on.
I feel like there's very little actual benefits to that for most people. Tbh I'd more worried that the chunky line is more likely to fall off and intrude on printing process.
Could you maybe do a video about a PETG optimized purge sequence? I have a bambulab and the purge sequence for petg is horrible and oftel still leaves stuff hanging that eventually gets in the print.
Nice shirt!
Im so used to this having been a default setting on Cura and Prusaslicer that I didnt realize others wouldnt have this.
Was that a solidoodle I saw? I still have mine.
I assume you can't use this if your model is at maximum bed capacity?
Correct, although that will be a limitation of any purge line for most printers.
Answered
--I'm not being sarcastic and legitimately curious. Why not simply print a skirt and get rid of the Purge Line start g code?--
1:59
@@marc_frank Fair enough. I obviously missed that so thank you.
Because skirts leave loose material near the print where they might get dragged into positions that interfere with the first layer. You need nozzle primed properly *before* it gets anywhere near the print zone.
@@daliasprints9798 👍 TY
2:43 The XL wipes a booger onto the bed.
So Marlin still can't do bed levelling only where the part is being printed?
I just add those line through the start g code and don’t mess with my KAMP settings?
My printer homes and purges at 350x350, but the print always starts closest to the 0x0 position, can this be changed in cura? There often is a bit of oozing across the bottom of the print that way..
How would this work with multiple items in the build plate?
"I think most people shouldn't have any problems installing it via SSH"
Most people don't know what SSH is
If they have already installed klipper, they absolutely will. It can't be avoided. I also have this: teachingtechyt.github.io/troubleshooting.html#ssh
Is that a clear A1 mini? How?
I think they were given out to 3D printing influencers before the machine was released.
Hold on. Does your x1c draw the filament calibration from right to left?! Is this another toilet bowl situation? 😂 i am in the us and my square portion of the filament calibration is on the right side of the bed. Just thought that was interesting.
Most printers I have seen have them on the left side. It doesn't really matter, though, as long as it starts the purge line closest to where the toolhead ends up after it's done doing bed calibration.
Another great video but dang those UHF are annoying…
My P1S doesn't have a 'extrusion cali line' section in the start code... hmm.
What about the supports?,if the model has supports,will this cause a problem
the slicer moves the purge line based on the supports also. no problems :)
@@imonseii cool thanks
Does this work OK with idex printers?
Lol flashforge adventurer printers have it from the very beginning
does not work with kobra, any way u can make soemthing for it.. it trys to do the line but since there is no filiment loading (kobra 3 combo) it cant do it... BUT other than that great video.
And what happens if the model is the size of a print plate, like gridginity grid?
the slicer will generate the purge triange outside the print area, and then it shows you an error about printing outside the print area :D