Manual Filament Tuning on Bambu Lab Printers with Orca Slicer! Pt. 2
Вставка
- Опубліковано 6 жов 2024
- Check out the first part here: • Manual Filament Tuning...
Ellis' website: ellis3dp.com/P...
Ellis' PA tool: ellis3dp.com/P...
Start Gcode: drive.google.c...
Prorifi3D: prorifi3d.com/...
CNC Kitchen "How fast can your hotend print?": • How fast can your hote...
Perfect. I'm using Orca Slicer over Prusa Slicer few months and i'm enjoying it. Some things I learned from video!
2nd Update: I commented out the AMS section of the start gcode and its working again. Load the filament you want to run in the AMS BEFORE running the gcode. As in, use the screen, select the filament, and hit load and then run the gcode from the SD card.
3rd Update: As of 8/44/23 Ellis' PA tool was added into Orca slicer beta!
This is so freakin helpful. Thank you!
Hey there, I've been dealing with artifacts and ringing on both my P1S and A1M on 100% speeds (which I find very disappointing) on top of that, after opening a ticket I got suggested to do this VFA test together with the following response: "identify the speed range where the VFA phenomenon is most noticeable. In future prints, try to avoid using speeds within that range as much as possible." So much regarding your question @ 6:26min and their "machines".
Don't get me wrong. The machines are great and the overall ecosystem is pretty good, but not getting reliable prints on much lower of the advertised speeds is not great IMO.
PD: Yes, all machines have been properly maintained, lubricated, cleaned, readjusted, recalibrated etc.
um just going for max flow limit and then dial down the result by aproximate 10% gets you better standard printing times too. i tested this with the standard nozzle and a Cht clone that you can find everywhere. and to be honest I'm glad i did. for example, on one of my petg brands it went from standard/default 15mm3 to 29mm3 on the cht clone and on the babu hot end to 21mm3. that speed up one of my prints by over 1h. you just have to learn how to read that number: use a caliper to measure the highest it got before artefacts and if you use 0.5mm steps divide by 2 the mm you got from measuring. that number you add to your starting speed from calibration and reduce 10%. if you use step 1mm you don't need to divide by 2 but that's as a first step when you try to find a rough number.
I find it difficult to print PETG over 12mm3 on any printer. Stefan shows in his video on speed vs strength that just printing faster doesn't always keep the same strength. I've found that over 12mm3 for PETG the layer adhesion is awful and the parts are weak and brittle. I'm interested to know if you experience that or not.
@@ButterPockets i just tested one of my petg cups that i printed(i wish i could post a picture of it) and a vent that i designed for myself that has some part that are as low as 2-3mm and in terms of layers adhesion and flexibility it is flawless(the brand i am using is gray petg sunlu with a cht nozzle, and it took me a couple of days maybe even more until i was able to dial it in because all the test came out fine but when i was printing sometimes i had different issues like bed adhesion warping at the bed lvl on big prints, or wall artefacts in some regions on straight line wall, or my biggest problem that i hated the most was stringing) but i have other brands that i agree that speed made them have a bad layer adhesion and some even brittle(despite me using temperatures recommended) sunlu gray petg is the only one i am able to have such a high flow and have the petg still keep it's properties. the other brands i can't get more than 26 at best and no less than 19mm3(with a rainbow petg from a brand i never heard of or used before(but again i am using an aftermarket cht nozzle that gives you about 20-25% better flow with every filament(after you dial it in)
Vfas are not solved by input shaper, they are caused (in my theory) by the magnets and coils in the stepper motors. As you may know on some steppers you can feel the magnets and coils more than on others, in my case twotrees (usongshine steppers) steppers had very distinct pulses while on my moons steppers or flsun ones you barely didn't feel anything. The twotrees motors had really horrible vfa while the others didn't. Also at certain rpms you have more vfa or none, most steppers have really prominent vfa at 1,25 revolutions per second or at 50mm/s with a 20t pulley. This isn't something you can really tune out, maybe with more advanced stepper control and measurements with an oscilloscope. Typically you deal with vfa by avoiding problematic speeds, most commonly 50mm/s and multiples of it, although at each multiple with less intensity. Stealthchop for example also reduces vfa, but limits your max speed to 5, maybe 6, revolutions per second or 200 to 240mm/s. Here i would suggest looking at 247 printings comparison between a v0.1 ldo and formbot kit. The more expensive ldo had horrible vfa while the formbot one had ok vfa
I mean that’s a lot of what I said. Input shaper helps to reduce VFA by printing at a speed where your printer body is not resonating. I mention how tuning the driving of your motor is also one of the “real” ways to solve VFA, even though it might never completely go away. The combination of how your motors are driven, the belts and pulleys themselves, the inaccuracy of your filament and extruder, etc all contribute.
@@ButterPockets isnt input shaper affecting acceleration and deceleration? To my knowledge it counters ringing by counter ringing, similar to noise cancelling headphones.
To be fair i haven't watched till the end, had to go outside and utilise the beautiful weather we have to do some outside work
@@kilianlindlbauer8277 it does affect acceleration and speed but you acceleration and speed affect how fast your motors turn. That turning speed of your motors causes the entire printer to vibrate/resonate/ring. This in turn is one of the causes of VFA. But all of these “artifacts” are part of the larger problem.
The OP is correct, input shaper has no effect on VFA (which is different from ringing). Ringing is caused by kinematic resonance, VFA is caused by motor resonance. If your toolhead is moving at a constant velocity, there will be no ringing (kinematic resonance), but VFA could be at it's absolute worst (motor resonance).
I don't understand how to read the lines... there are more lines than numbers, so how do you know which one to pick? Love the video!
is there a right sequence to do the calibration? like temp first then max flowrate, then ... ... ... ?
I had the same questions and, to my logic, it turned out it only makes sense if you execute a test for which you already have all known variables except the one/s you're testing. The variables of interest for the case.
To me: 1- Temperature tower first.
2- Bambu tests, say PA (K value) and flowrate multiplier, in this mandatory order.
3- Max flowrate (you have to know at which temperature the material likes to run, to perform this).
- Then, dimensional calibrations for shrinkage etc if needed, with a more accurate test than a 20mm cube as CNC Kitchen pointed out what I always thought.
- TD if needed.
I wrote "if needed" because what I've learned from this clear path evryone of us seem to crave for, is that you just have to do tests for the job you're doing and for how good you want it done.
And keep in your library (drive your customer to) materials you will rebuy and you trust into. Say they will be on the shelf even the year to come.
This is not only bc, you'll find, following this iter every roll you unbox makes you waste an entire morning and close to 10% of the roll you just unpacked.
But mainly because as you swap object to print, color, nozzle or nozzle size, printer (in this order of likeliness) at least one of the values you found is unreliable or useless or doesn't fit the first one, the damn job to be printed.
It's a useful excercise, I think. I'm still following this iter, even with discretionality (according to the job to be printed).
I'm stacking the values I record for each material.
But I know (...OGs told me...) that it will be my experience that will use those value without checking the sheet, not me going to read them, for the most part.
Bambu does not recommend flow cali on a textured plate tho…?
I always do the first one, the K value or pressure advance, on a smooth plate, but for the flow x or multiplier, the second one, it prints a set of tags 2mm thick to evaluate the surface so i dont think it matters which plate u use there. @@Jacksteelmachines
But even the pressure advance has a second option for the test if u check and with that also doesn't matter which plate u use. The one with the shape that resembles army badges.@@Jacksteelmachines
Temperature, estep , flow rate , pressure advance , retraction , max volumetric speed, VFA
Good technical explanation❤👍. Wondering how the VFA tuning is performed for a delta style printer?
Great video.Then my (totally newbie) question will be.Can this method replace the lidar need on the new P1S? Looks like this is even more precise method with fillament saving even.Or am I totally wroong? I mean,many peole think about X1C because of the Lidar,but if tshi methond works well,then there is no need for the Lidar and then the P1S can be a good option for lower bugdets.
Yes!! Honestly if the P1S was out when I ordered the X1C I would have purchased that instead. This is a conventional method that you would perform on other printers. The LIDAR sounds great and will be one day, but for now I would totally spend less money up front on the printer and tune manually
Could you connect Ellis with the Softfever folks to we could get his pressure advance test into Orca? That would be amazing!
That would be incredible. I can ask, but I’m not sure if that will ever happen haha.
So I opened Orca slicer today and they added this in the latest Beta!
@@ButterPockets sweet! Thanks! Will need to do the update… questioning firmware but guess I’ll have to do that too.
@@cstategrad99 the new firmware is for the new version of Bambu studio. Which now stores PA k value on the printer and lets you swap profiles. Trying to do a video on it today, Bambu lab took a good step in the right direction.
Excellent
The sound deadening material that you people install inside inside all their car doors and on the floor to prevent vibrations would totally be perfect for putting under the printer right?
I wouldn’t think so. I think the underlying problem is that stuff isn’t really stiff or that heavy. A stone paver or marble slab would be much better.
My X1C will not adhere to ANY textured PEI sheet. I've had 2 bambu branded since November and I just received 2 knock offs. Someone is telling me I need to adjust my K-Value. It's missing from both bambu and Orca for me. When I go to the device tab, and then AMS. The calibrate for the K-value is missing on both slicers. I'm lost. Any idea what I'm missing? Awesome video btw.
It's not in the "device" tab it's in the filament tab.
It's labeled as Pressure Advance, not K value, in the failament tab.
What temps are you running your PEI at on what materials? I find that for best results I run my PEI pretty hot. 110 for ABS/ASA for instance. 65-70 for pla.
Very odd indeed. I’ve been using the Bambu textured PEI plate for weeks now with the stock presets on my X1-C with no issues.
for the max flow rate, I’m a little confused when you say Bambu have done this for us. There are not many third party filaments in Bambu Studio - eg no Sunlu at all - so unless you select generic PLA at 12mm3/s it seems like you’d be leaving performance on the table by not testing to see if it can support more?
It’s not really going to be different per filament. If the hotend can melt at a rate of 19mm3/s then that’s what it’s limited to. Which the hotend on the X1 is around 16-19
@@ButterPocketsright but the default rate for most (all?) generic filaments is much less than 19mm3.
The maximum rate at which the hotend itself can melt plastic and push it out is about 19mm3/s. Not all different kinds of filament can be printed that fast though. TPU for example would benefit from printing a lot a volumetric flow rate of around 3mm3/s, but you could also do some testing to figure how the best speeds to print at for various models rather than just “how fast can I possibly push this filament”. When I said “it’s not going to be different per filament” I really meant “per brand”. So if Bambu PLA can run at say 16 then mostly like Sunlu can run at 16. The thing is tho, setting your volumetric flow rate is for all intents and purposes the same as setting your speed settings. Selecting “generic” means that you aren’t using any kind of testing from BL and it seems obvious to me that they would intentionally set the generic setting low. What I’m really getting at is your volumetric flow rate is going to be the MAXIMUM speed at which your hot end can move and it will cap your XY movements. I guess it’s useful to know that but you probably aren’t going to be running at those speeds any way. The way BL setup their profiles tho I guess you will since their speeds are ridiculously high in their default profiles and rely on the volumetric flow rate capping them. I just don’t really print like that.
do you have a recommended step by step? for example
1. PA
2. temp
3. flow
etc
thank you for the videos!
Tune PA with 2 passes. Tune EM (flow) with 1 to 2 passes. That should really be it. If you want to you can then tune max flow rate of you need to, but this is a one time kind of test per filament type - so pretty much all PLA should run at a similar max flow.
@@ButterPockets thank you for the reply :)
@@ButterPockets I might say that “all of the same type of PLA from the same manufacturer has similar max flow rates” but definitely not “all PLA”. For instance, “galaxy” PLAs will be different than normal PLA which will be different from “silk PLAs” and good brands would be different than low-cost, no-names. Yesterday I watched a video from 247printing where he was running a beta PLA filament designed for max-speed printing and that stuff was crazy different. It looked like softened butter coming out of the nozzle.
that’s true. I should clarify more when I say that. You are correct that similar PLA types from the same manufacturer would be similar. I have high flow PLA from atomic and it definitely flows higher than their standard.
@@ButterPocketsin addition, for each calibration, which automatic test does it replace ? And how often should it be done (power cycle? Weekly? Monthly?) ?
Thanks for the video! Where did you get that sweet marble slab? That looks way better than my 16” paver.
I went to a place around me that did concrete recycling and just asked if I could buy a scrap piece of marble counter! She let me have it for like 5$. Go on Facebook market place and look for like "counter scrap", "marble slab", "granite slab". You should be able to find something near you for not too expensive.
The ellis gcode was just the stock gcode with the variables changed accordingly and some unnecessary stuff taken out.
I didn't realize how much the start gcode has changed for the bambu printers. I should probably make an updated version.
Yeah I’ve already had to change it once. I don’t think it NEEDs an update since you took out a lot of what changed.
Would the biggest artifacts not be on the side that's printed as the nozzle moves out? On the way in it's still speeding up after a sharp corner.
On the way in it’s slowing down which can cause problems as much as speeding up. Although, accelerating out of a corner can drag the filament. That’s why I turn down my xy jerk.
Is the calibration listing the tests in the order that it should be done? Temperature, Flow, then Pressure Advance or is should it be done in a different order?
Thanks
I do PA then EM per Ellis guide linked in the description.
I did the pattern, but honestly i still dont see any difference at the good area between 0.01 and 0.03, they look all the fakin' same.... :/
Any chance you can elaborate on what the G-Code you were sent vs the newer g-code "does". Is there a reason to use the older gcode over the newer gcode?
Yes, I don't know if that's worth a video, but I can write something up and post it tomorrow!
@@ButterPockets thank you! Really great video. I’ve been using the OrcaSlicer PA tests, but this one is very nice. I had wondered how to do it on an X1C. Glad I stumbled my way into your channel 😉
@@fourtwizzy ua-cam.com/users/postUgkxVb-wiQtebhfTysgjiCKKnmpffh_DeQ-q
I hope this makes sense and comment there if you want anything specific explained!
I just realized that the gcode generate with the Ellis method is causing the printer to don't end the print correctly.
Do you have a similar issue?
Can you explain what you mean? What does it do at the end?
@@ButterPockets it finish the print but after will Park the toolhead in one place and the screen will just stay with the print in progress like it is still continuing to print
Does your shelf rattle from x1c printing? Any chance you'll review the pantheon HS3 MK2? Thanks
It does move a little depending, but doesn't rattle. It's a really heavy shelf. And I don't think a company would let me near a printer that expensive haha, sorry!
@@ButterPockets would you share any info on your shelf? The heavy duty one I got from home Depot rattles like a baby's toy when printing PLA at optimized speed. I've been printing with petg lately with slower speeds, so it hasn't been shaking as bad.
I almost got some shelves from home depot but I ended up going with shelves from Uline. Pretty expensive, but nice. I believe mine are the 48x24x96 but they have a lot of sizes. www.uline.com/BL_3867/Wide-Span-Storage-Racks-Particle-Board
this does not work with ams? it tried to print without filament. :)
I actually just ran into this myself...it definitely didn't do this before. I am working to see if I can give out a fix.
@@ButterPockets Have you figured out how to make this work with the AMS yet?
@@xThree65x they added into orca slicer by default now! So it just works :) I’m filming a video on all of the updates as I type this!
where is the math for the volumetric test. u didnt explain any thing bout it since yours printed fine
There no math really and I didn’t explain it because I don’t run them (like I said). You just run it till you see it prints badly then you know that height relates to the speed(flow) at which it starts to fail.
@@ButterPockets height? What numbers do you use? I have a white Polymaker light that won’t print good at all on one of my A1’s layer lines are horrible. Sunlu prints fine. Anyways. Thank you for responding.
It shows in the tooltip for the test. It changes speed (flow) in like 1mm or 2mm increments so the height of the print relates to the volumetric flow rate. It might be extrusion multiplier