The source of the SOMETIMES AWFUL results on the Bambu X1C
ฝัง
- เผยแพร่เมื่อ 27 ม.ค. 2025
- Finally we were able to figure out the cause!
Help support this channel! / needitmakeit
This is a short list to help support this channel, thank you so much for your help!
QIDI PLUS4 - qidi3d.com/pro...
QIDI Q1 Pro - qidi3d.com/pro...
Thermal Master P2 - shrsl.com/4o6j1
Bambu X1C - shrsl.com/4j1gz
Bambu Filament Shown in Videos - shrsl.com/4ks0j
Bambu Super Tack - shrsl.com/4sjgn
BIQU Frostbite - shrsl.com/4sjgo
BIQU Glacier - shrsl.com/4sjgq
Nowhere Stair - makerworld.com...
I didn't have any major problems with my printer, but there have been so many comments saying that I should have better quality prints coming off of my X1C. I put it down to lighting mainly, but after some experimentation with the 0.2mm nozzle, it became clear that this was a real problem, and real problems should have real solutions.
After quite a lot of testing and watching what the printer was doing we were finally able to figure out what was happening, and it was pretty unexpected.
I hope you found the video helpful. I probably come off as being critical over the printer, but I really like the Printer, it is one of my go-tos, but to get the most out of it, I need to be doing something a little bit different.
#bambu
#bambulab
#3dprinting
From what I understand, the Lidar doesn't work that well when you add loads of lighting to it (widely stated fact, just Google it, there's tons of info on it). Looks to me that you're using a setup of camera lights to light the print while it's doing the calibration. This might throw off the calibration. People have reported this issue even happening from adding LED lights inside the housing.
Another argument could be that you changed many variables like hotend, tool casing, etc. I would do these tests on the stock setup without additional lighting.
I'm not going to say the lidar will perform flawlessly without the lighting and with the stock setup, but to do a proper test, at least remove these variables from the equation :)
My thought too. Every 'upgrade' I've tried on my printer (except the Obxidian) has made print quality worse. They just work out of the box, but messing with that tuning has caused me issues.
This, I noticed too that my calibration is a lot worse when I have aftermarket case light on
This 100% I was having adhesion issues on mine until I realized my Home Assistant automation for my LEDs stopped working. Fixed the automation and all is good again.
i have great results with the lidar. even tricking the system to have it auto calibrate tpu as well :P mine is an early model with one red and one blue led. i turn off all lights in the room and the only light source is the logo on the printhead and the leds on the ams´s. (waiting for an option to disable them still) i do the auto calibration from the slicer menu. for all filaments first. ive tried running them 10 times in a row and noted the result. i can confirm that 8 of 10 are the same for most filaments. so delete top and bottom value, divide by 8 and the deviation or "jitter" is well below 10%. Most filaments are about 0-3% off.
what i do now because of this is to auto first. get a ballpark value, then manual "pattern" calibration. for flow rate i dont use the fancy square chips as they take so long time just to calibrate flowrate. instead i print a hollow box in vasemode. measure the wall width and weigh the test model for that pinpoint sweetspot.
(will change with moisture level change in the filament)
@@reyalPRON I think it would be nice if the machine could automatically do ten calibrations and then take the average by itself! Actually doesn't sound like a bad idea. Why haven't they implemented this...
I think before you start tests you should return the X1C to stock as a baseline. My X1C definitely improves with calibration. I also have a Qidi Plus4 so I can do your tests too. Your X1c is not printing as good as mine (mine is stock) When using Lidar do not have lighting on, it affects it, AND do not use a textured bed calibrate on cool or high temp plate.
Or the old "Cool Plate" (I'm still upset that Bambu discontinued that plate)
The LIDAR is a huge gimmick, and not once have I gotten a better print by using it versus stock settings especially on longer functional prints where the margin of error tends to disappear compared to trinkets. I've tried all the beds out of curiosity, but the LIDAR is a huge placebo. Bambu also admits this, "The LIDAR has a 10 percent jitter" In other words it may overshoot or under shoot by 10% or a total of 20% so if you have a 95% flow rate set then the lidar might overshoot the flowrate correction to 104.5% percent and give you unintentional fuzzy skin. Honestly you are better off randomly picking a value with a 5% jitter from stock settings.
Hi, my X1C is also stock, it prints amazing with the Cool plate but it is Texture PEI plate phobic. I don't use any glue or anything fancy like that, just glass cleaner to clean the plate and the only failures I get are my own and related to oversights with some of my design prototyping.
@@Slurgical_3D_Terrain_Channeltry cleaning with water and good dish soap, air dry (paper towels can leave behind fibers and dust) and maybe try to increase bed temperature up to 60*
@ I have done that with Dawn and other brand, I went up to 75c no luck.
When I first got my X1C, the auto calibration would ruin every single print without fail. I thought my printer was broken.
After some investigation, I think I found out that it's because I was shining my flashlight into the printer during calibration, I think the bright light was messing with the LIDAR sensors.
Try playing with the lighting in your room. I have found that a darker room seems to helps the LIDAR.
agree
without reverting to the stock hardware and repeating the tests I don't think the results bear any significance. They reflect mostly on your modded printer. Going back to stock hardware might show the same results but you have to follow up in this the nake this video valuable
I'd like to see your X1C tested with the stock extruder gears and cooling duct. Maybe also without the additional LED lighting.
Yup testing a modded printer is flawed
i have an A1 and an A1 mini and both have served me well before. usually i just select the "generic " option in the slicer, press go, and not even think about it. the auto-calibration hasn't failed me once yet.
i didn't know it was so different on the x1c, but it just makes me even more convinced that i don't want one. it honestly just looks very last-gen in a lot of aspects, thanks for showing another one of those.
I've heard the lidar needs to be cleaned off fairly regularly. Could plastic-dust on lidar be a miscalibration cause, or would it warn you about it's poor readings instead?
The X1C will warn you when it detects dirt on the LIDAR (Error code HMS_0C00_0100_0001_0004: Micro Lidar Lens dirty). Never had it happen with PLA, but when I print large volumes of PETG or ABS there will be residue on the lens.
Weird, never had this error yet and I've printed quite a bit of ABS/ASA. Had the printer for two years now.
@@sirseriously I often turn off the chamber fan, that might be a factor. When the chamber fan is on the X1C exhausts through a carbon filter which catches most of the crud, when it's off all of the PETG/ABS residue deposits itself inside the chamber.
It’ll tell you specifically when you need to, really depends on the kinds of filament - some cause more dust and stringing which is what obscures the lens. When I used elegoo rapid PETG it’d tell me about once a week to clean it. When I switched to PETG HF it didn’t need to be cleaned anymore. But the carbon rods got dustier with rapid
Already been written, but you should clean the LIDAR and calibrate with dim lighting only (no studio lights/LED strips in the print chamber). And never calibrate on build plates that have a grainy surface. Haven't had a chance to try the Bambu SuperTack yet, but judging by images I've seen it does seem to be somewhat textured. 100% infill is rarely a good idea, especially when you use concentric infill. Unless your tuning is spot-on, use 96%-98%. Practically identical part strength, while giving excess filament enough space not to get in the way of outer walls and cause defects. When I print 100% infill I usually go for aligned rectilinear rather than concentric, produces significantly better results for me even when using standard filament profiles.
While I can see automatic calibration getting even better and doing a great job, I don't think I could trust it enough right now to skip manual calibration to get optimal quality. There are just too many variables at play and perfect prints take time and tuning. I am constantly impressed with my Qidi Plus4, especially the dimensional accuracy and surface quality.
Those are some of the worst prints I've ever seen come off of an X1 carbon. You definitely have something wrong with that printer, I switch out build plates all the time and I have never ran a manual calibration with perfect results almost every time. But I also clean my machine very rigorously I make sure all the screws on everything are tightened on a regular basis and also adjust the tensioners so that the belt rides in the center of the idlers on the back. I did find the bed belt on the bottom of the machine to be loose at one point so now I keep a pretty good eye on it. You might check the bed belt and tensioner on the bottom of the machine.
I look forward to you giving the QIDI Plus4 a little more love in the upcoming year. I don't own any Bambu printers. I bought the Plus4 for both functional prints as well as decorative items made in silk PLA. My wife gives the decoctive items to her consulting clients and they love them. I have nearly 400 hours on my Plus4 and it continues to turn out amazing work. I wish I understood more on how to optimize Klipper but I'm afraid to experiment too much for fear of messing up my profile. It just prints so good!
Maybe the problem with automatic calibration is all the aftermarket parts? You also cant really compare the Qidi against the x1c with aftermarket parts too. Put your machine stock before you do stuff like this! Otherwise there are too many variables.
You should really check the speed settings. Those very visible layer line are probably due to inconsistent speeds inbetween layers. This could also explain why you see differences after one of the parts finishes. Using the speeds tab on the preview section of the slicer really helps. You should try to achieve similar speeds across all layers (similar colors). Do that by reducing outer layer speeds. This shouldn't increase print times by a lot since accelarations are way more important but can have a drastic impact on quality.
It would also be interesting to compare whether its best to redue speeds or accelaration for better quality.
I have an X1C (685 hours) with a BentoBox and a 24v COB LED strip around the perimeter of my AMS riser. I just printed a calibration cube with Sunlu PLA, using auto-flow calibration on a "smooth diamond" PEO plate. I would score it a 9.9/10 (the top surface has a slight blemish where two diagonals join). I was going to do the auto-flow with and without the LED light active, but given the results, I don't think it *can* improve.
There is something wrong with your X1C.
Good video. Bit of a slow pace, would love to see a more information-dense/faster pace. Maybe something worth exploring for you?
The A1 Mini Auto Flow Calibration is straight up magic. Perfect results regardless of what filament you throw at it. Might be my favorite feature of the printer, and the main reason I use it a lot over my heavily modified frankenprinter. I would looove to get their sensors and software fitted to my other printers, would make printing so much more convenient.
Just my 2 cents... Should have put the X1C back to stock parts and did the Orca slicer full calibration. There is an amazing calibration video done by Covenant Custom that walks you through the entire process. In short it should be -> Temp tower > Flow rate pass1 > Pressure advance > Flow rate pass 2 > retraction > max flow > VFA. This might be rough if you have lots of filaments which is why I only do it on the stuff I sell or special projects.
Honestly this. Even the manual calibrations versions in this code are pretty rough.
Great channel, love the carefully thought out feedback. Subscribed!
No miracle advice, but maybe something in my experience triggers some insight. Sorry in advance for the lengthy comment, but throwing it out there and if folks disagree with something then I can learn from their experiences and thoughts as well. Mature, non-angry responses only please.
X1c hotend all stock with exception of Panda Jetpack (better air flow, reduced mass on print head decreases ringing... less mass to accelerate/de-accelerate), use DK riser (eliminates PTFE tube rubbing top glass, and has vents that can be opened/closed), use printed TPU vibration feet sitting on top of paver with vibration absorption material below paver. Jetpack gives 5-10% improvement in cooling that had been missing. For my prints, do see improvement using it. Also use a 3rd party $25 smooth/textured black pei build plate as daily driver. Can auto-calibrate using either side. Suspect due to darker pei. Single AMS sits on top of printer.
All identical X1c printers are not actually identical. OEM parts can vary somewhat between manufacturing runs, improvements, etc. For instance, maybe my stock fan runs a little slower than it should so the Jetpack helps me while others do not see a difference.
Do a combo of manual calibration and auto-calibration. Sometimes run manually only. Often let it auto-calibrate while using a "tuned", custom profile. Only use full auto-calibrate with Bambu's PC filament, and only because it gives off VOC's that I don't want to be exposed to.
General impression is that auto-calibrate can dial results in closer if already fairly close to target. Often do see a slight improvement using a combo of the 2 approaches. Suspect auto-calibrate is helping to adjust for variability in enviroment (daily changes in temperature, humidity, air pressure, slight change in filament moisture, filament color additives, etc). By itself auto-calibrate is not always close enough to get all the way there. Some stock profiles are probably by default closer to the target than others.
Keep lidar clean, door closed, lights off during calibration process. Dark glass that folks complain about is clearly an intentional design choice. Once running, door open as needed to control temperature.
Running latest printer firmware as of 2025/01/11. Quality has varied across firmware releases, but latest one appears to be in a sweet spot. For my printer at least.
Pressure advance tests using the lines are affected by other profile settings such as bottom layer flow rate. Need to dial that in before PA. Why some folks prefer to use the PA tower option since it is not affected by bottom layer flow rate. Prefer to iterate through calibrations 2-3 times in total as settings can affect each other. Seems to get closer to target with each pass. For initial calibration need to slow things down to reduce variability in the individual calibration tests. Speed, acceleration, jerk, infill (it affects flow rate test results).
Agree with 3 walls, inner/outer/inner used in combo with turning off "ensure wall thickness", and avoid crossing walls with a decent distance setting of say 100mm. For calibration, use monotonic top and monotonic line or rectilinear infill, and allow for enough top layers to get an accurate reading on flow rate (or a higher rate of infill). Reduces inaccurate reading when looking for smoothest top layer.
In v2.2.0 Orca there are definitely some glitches with Arachne. Artifacts, weird sliced lines that span to infinity, etc. Use Orca 99% of time, but aware that Bambu slicer is not currently having some of these issues with Arachne. So calibrate with Classic for time being. Generally compare the slicer preview from each before choosing for an actual print.
Having printed something similar to your printed part, found that different filaments expand differently. Poor quality example seemed to be in PLA, while Qidi example looked carbon filled. Find with some PLA's that have to reduce infill to 92-95%. With expansion and infill selection will fill out to 100%. Some will see this as a hack, but calibrations are never entirely perfect for all use cases, and this seems to work well.
Keep up the great work. Have picked up some great tips and upgrade advice from your channel. Appreciate you offering what always appears to be your honest opinion.
Honestly, the print test lines look like inconsistent extrusion speed. I'd almost want to check your extrusion wheel for wear.
There is definitely something wrong with his X1C.
Lie-dar strikes again.
You should also dry matte filaments well in my experience - it will lead to much more consistent extrusion and better surface finish. You’ll know if it’s working because your ideal EM will change after drying.
You should check the infill speed settings, when it starts to scrap off filament during print. In my case (P1S) the speed settings for .16 layer height was 450mm/s, by default profile. BambuLab changed these to 200mm/s in later profile per update. And another point, don't use crossing infill pattern in highspeed mode. And most important, dry your filament, the blobs on the outside of your hub thing are caused by moisture in the pla.
So i did 3 things to fix the under extrusion on the seams. Clean the lidar, clean and or lubricate the rods, and run some cleaning filament through the nozzle. Will probably help with your z banding as well.
From my experience with the X1C and making similar functional solid parts, I believe the issue you saw on the hub has more to do with temperature buildup within the part. This is a problem with PLA and even more of a problem with PETG. Possible fixes are slowing down, increasing cooling/ventilation, or using a glass or carbon fiber filament. You can also design internal structure to get the strength you need while lowering the thermal mass and increasing the cooling area.
Another thing to note, is that I see this same defect on parts with denser sparse infill, thick wall sections (usually 5 or more), or narrow round sections without other features in the same layers.
When it shows up in heavy-walled parts, slow down the outer wall first (this will help maintain consistent surface finish). If that isn’t enough, reduce the inner wall speed. Infill can be usually be kept fast.
Also, at the first hint of the horrid noise from the nozzle chattering across the part, switch the X1C to 50% speed (quiet mode). This may be enough to salvage the part.
A few more spinoff variations you could do:
- LIDAR calib values with stock lighting only
- LIDAR calib values with a smooth plate
You might also want to try a different filament that has a lower natural flow rate, as the stock 0.98 is used for a lot of profiles so if the natural flow-rate is near there, it doesn't highlight the flaws as much on non-calibrated because it's already close enough. Quite a few PETG's have ~0.95 .. I found Elegoo Rapid PETG tends to give a good range (~0.95 | K: 0.023).
I'm wondering if the high flow nozzle confuses the auto calibration somehow.
If you dont have a complicated file or overhangs you can print the inner/outer/inner wall or even outer first, that makes smother outer walls because it doesnt get squeezed out that much.
afaik i/o/i is identical to o/i with 2 walls, it's on 3+ walls it works differently
3 walls ioi gives much better final print quality, definitely worth it.
I have an A1 and a P1S. For the P1S I manually calibrate each new filament using the inbuilt calibrations in Orca Slicer. For the A1 I don't bother and I can't complain about the print quality.
I had similar issues. You can see with your first attempt at the hub, it starts out okay but just gets worse and worse as it goes up. As you said, there is no where for the "extra" filament to go, but there shouldn't be any extra. I assume the printer thinks it is filling the entire volume of your Layer Height x Line Width but obviously it can't do that and there will always be voids. Try turning your flow rate down by 10% and see if you can turn on some kind of flow equalization to help with the smaller lines Arachne generates. FYI, I use Cura, I'm not sure what Bamboo calls these settings.
I can definitely relate to some of your findings. The calibration process looks all precise and effective and scientific, but the stock somehow just makes more consistent prints.
Excellent video, and production quality! Did you have any growing pains when switching to the Cool SuperTack plate? Any settings to tweak? I'm having a devil of a time getting long-duration prints from not curling at the corners, despite all the typical cleaning recommendations. This is in Bambu Studio and Orca (which doesnt technically yet have a profile for this plate)
I don't see the nowhere stair with the fail rail on Makerworld.. just the stairs alone. Am I missing something?
Very helpful, I just implemented one of your connection designs on a extra large sign.
Dude you were SO close to 100K I'd basically call it a win. That silver play button is almost in the mail! Keep going I love your channel.
My A1 Mini prints really nicely with the automatic dynamic flow compensation turned on. I get much nicer results than what I see in your video. I haven't tried a 100% in fill part though.
I could never get the manual flow dynamic calibration test to work on my A1 mini the option is just grayed out, what do I do
This was really interesting, but I’m concerned about some of testing methodology that muddies the waters.
As some comments have mentioned about the LiDAR, the lighting you have on it while it was running may be interfering. I’m not sure about the X1C since I don’t have one, but I do have an Einstar 3D scanner and can definitely attest to too much brightness ruining the scan. Trying to scan small objects on a white table under a florescent light usually doesn’t go well haha. I’d like to see a control print test with the lights off so the sensor can run on its own emissions and see what happens.
Additionally, the printer was engineered and calibrated with the stock components, so using a modified X1C introduces a lot of variables that haven’t been controlled yet. It’s clear that your customized printer isn’t performing as well as it should be, but that doesn’t mean the X1C and Bambu Studio are at fault. Your manual calibration may be compensating for the problems instantiated by the customizations. But without a control group of tests, it’s not possible to say where the problem is.
All in all, this was a super interesting video and I watched it twice! I’d really like to see a part 2 follow-up before the ultimate face-off! 🎉
Very interesting tests. It looks like you have some great plans ahead for 2025 ! What upgrades are installed on your Bambu?
Looking forward to each new video - love the thouroughness of each test
I turn the chamber light off when it does the auto calibration and I get (more) consistent results - the inconsistency seems related to the lighting conditions which makes sense
A1 auto Flow dynamics calibrations are also off in my experience.
PLA k-factor default value : 0.02
Auto calibration : 0.028 to 0.030
Manual calibration : 0.022
So yes the default value is better than the auto calibration values
I went trough the work of manually calibrating adaptive pressure advance in Orca slicer. To my surprise the values vary from 0 to 0.02. At low flow rate (7.65 mm3/s) and high acceleration (20k mm/s2) 0 was the best value. At high flow rate (40 mm3/s) and low acceleration (5k mm/s2) the PA was 0.02. I was quite surprised at the quite large range of values.
The print quality improved quite a bit with the most noticeable result being the quality of top solid infill where very short lines would always look worse than longer runs.
I have found that with some filaments it is necessary to cut the speed back. Try halving most of the important speeds and try again.
I got the cool plate and I was switching back and forth between buildplates for prints and then I noticed all my prints were awful or the first layer wasn't looking great... Turns out I had misclicked the switch buildplate in this slicer. Always good to double check to keep you sane.
I wouldn't have thought the textured pei and the cool plate would drastically change the quality, but I'm guessing the lidar with the wrong buildplate selected also throws the calibration out the window.
Dont know if you did, but could you do a full calibration video for the filament ? what to look for, what you dont want, what you want when the test is done, how to save those setting for the filament, etc.
One reason the "Flow Dynamic Calibration" (checked at print time) looks nasty now, is because Bambu put the purge line on top of it. The purge line used to be in the back right corner.
About that...for a time, I was seeing a purge line in the back right, then it changed (after upgrade?) to the front of the plate.
It looks like a bad pressure advance mixed with over extrusion. If the lidar is not the issue, you could also make sure your filament is dry... My results improved so much since i put 2lbs of silica gel in my filament loader.
Go back to the .4 stock nozzle on the x1c and see if the results are better. Good topic.
no way, the e3d nozzles give WAY better print quality. its gotta be lidar and autocalibration. I have a p1s and and e3d nozzle, I do all the flow calibration myself, my print quality got better after switching from a hardened .4 stock, to a e3d diamondback.
@mrrooter601 maybe, but I would like to see him try it.
The auto-calibration on my X1C is very dependent on how recently I've cleaned the camera. Granted I mostly print ABS so the lens fogs up a lot. So I'll concur with the others that your lighting may well be causing some of the problems - remember that the printer is taking a picture of the line and comparing it to what the line "should" look like. If the lighting from the actual line and the lighting form the reference image are notably different, the printer will probably make incorrect conclusions.
I do find that cutting the speeds down from defaults makes a big difference for mechanical accuracy. The default speeds are for parts that need to work but whose aesthetic appearance doesn't matter. By cutting the speeds, accuracy goes from good to "really good." I just printed something where I kinda cocked up the CAD step and was "oh, dear, this is only 1mm thick and I need to sand it smooth to paint it." I was really worred I'd just sand through the part before I got it smooth - but with 0.12mm layers out of a 0.4mm nozzle running at about half the BambuSlicer default speeds, everythigg lined up enough that I was still able to smooth it out.
As for your blue solid part - that really looked like the filament was failing. X1Cs aren't great at printing PLA. Remember that the printer's name is actually X1-CARBON, we're all just being lazy when we type X1C - and the "CARBON" part is for Carbon-fibre nylon. X1Cs love filaments that want a heated build chambre, and struggle with ones that don't. So if that thing was PLA or PETG, maybe it was just too hot?
Personally I think that printing PLA on an X1C is using your dump truck to recycle a flat of beer cans. Sure, it can do it, but it's a waste of the machine's capabiities. I feel that if all you're planning to print is PLA, you should get a P1P or an A1 and use a printer designed to work with that material. To be sure, this is partly because I mostly print cosplay and prop-recreation stuff, so I need to sand it and join it and that is orders of magnitude easier with ABS. But the X1C easts ABS for breakfast, no issues at all, and ABS is usually cheaper anyway. I'm slowly burning off my PLA and I doubt I'll buy another spool in the future.
Enlightening. Do you find calibrating for one specific type of filament is enough? i.e one for PLA, one for PLA Matte, one for PETG, etc… or do you have to manually calibrate for every single color for best results? Assuming I'm using all Bambu brand filaments.
So one of the questions that I've tried to dial in, and prevent with extraction, line path, and everything are the hairs. I know they are common, but on something like the block, you can do a quick pass with a torch and it burns them up, and goes away no problem, but on something like the stairs, 50% of the time it seems to deform them, so what is the solution to removing those pesky hairs on delicate prints? Right now its an exacto knife and painstaking time, but on ultra fine detailed things, its just not feasible for something you sell.
Have you tried short blasts with a heat gun? While holding it with your bare hand, as a way to not get too hot.
Suggest cleaning the lidar with isopropyl alcohol and trying again.
Welcome to 2023. Everyone said that the lidar/auto-calibration functionality of the X1C didn't work and you should get two P1Ses instead of one X1C. There's no replacement for calibration by hand. The A1 automatic calibration (flux) is much better but still can't beat manual. Your blue part doesn't have enough cooling and is over-extruded. Your X1C has major layer shift due to insufficiently lubricated Z-axis screws or damaged screws. You really aren't very good at this.
The LiDAR calibration actually works fantastic if done properly. Flooding the build plate with other light sources as during the calibration sequences, as done in this video, defeats the LiDAR though because of the functional nature of LiDAR. When the receiver is expecting the reflected light to determine how to adjust pa & flow there is already a ton of external reflected light blasting into the receiver. It, of course, thinks it is over extruding and compensates by dialing back the calibration numbers to compensate.
Exactly the same principle when defeating radar and LiDAR with electronic countermeasures. Overwhelm the receiver with a similar type of radiation and it won’t be able to determine snot. It physically cannot get accurate readings when you overwhelm the receiver like that 🤷♂️
When did you last clean the micro-lidar?
What wall order do you use?
Hopefully you can include detailed tutorials in some of your upcoming videos.
Have you got a link for that cube? I'd like to try it and see what mine looks like, because yours doesn't even look like it came from a Bambu printer to me.
Its a Default Bambu Cube can be added in the slicer if i remember correctly. Been awhile since i printed one.
@@dslilws6 Found it. Maybe it's just his lighting, but he seems to be getting a lot more layer shift/variation than I get. His looks like my old Anycubic's output. Strange.
Appreciate the investigation, but I'm going to agree with what seems to be the consensus, to really determine the cause you need to "start from scratch" with your X1C at stock and don't blast the print chamber with external lighting during the calibration. You really can't sus out what factors are contributing without a "known" baseline that you and your community can validate.
The final extrusion test is supposed to look that way - i believe its used to calibrate some aspect of flow!
I have found out that using the pa tower gives you a much better control over the seam quality, you can even try orinting different towers with different gap sizes to improve the seam quality with the best quality for corners.
Mike, there is something weird going on with your Bambu printer. I just went back to watch the video where you "upgraded" the Extruder... and, yeah, the stock extruder was already printing atrociously back then - and it didn't get any better with the BIQU one. Great video, as always - but the comparison between your Bambu and the QIDI doesn't make sense until your Bambu is printing as good as any other Bambu prints.
true, something is off
Hi from Brazil. I have some problems making pieces 100% infill. To resolve i have change the infill overlap to 10%.
I love your vídeos
What’s funny is that I always used the manual calibration. I think it offers the best end results. I can understand using the auto for something you want to do quick or for something you don’t care about for quality.
I've read that you should turn off auto calibration at the menu or it will override the rate and dynamics you've entered don't know if its actually true though.
from the very few people i got to talk to that had an X1/C, before getting my p1s they just said dont use the lidar. everything else fantastic. and that still seems to be the case now. The main thing im missing on the P1S from the X1C (ive already upgraded to full hardened metal hotend) is the fact you can deselect a part of a print that has failed on a print and that an X1C can run other firmware. higher rez and fps camera is nice too. Touch screen could be nice and i would have wanted it but i very very very rearly use the screen as is on the p1s. all done by the slicer. These seem to be the only features i am really missing and only one that i really want is the failed parts selector and also better cam. the custom firmware is a bonus. I was wanting lan mode with cam but they have added that now it seems. but i dont use it coz the mobile app is useful and works without massive annoyance. cheers for the video
If you haven't already, tuning retraction and bridging might improve the staircase as well.
I use auto and never have any problems with the exception of build-plate adhesion. But there it's almost always a third-party plate with which I have challenges. Most of the time bumping up the plate temp fixes the issue, and there's always Vision Miner plate coating which never fails me when I use it.
Some ppl forget that in multi material prints, the LiDAR only auto calibrates the first filament used, if you want each material/color to be good, you should run the multi material calibration from Bambu studio, you can do up to 4 at a time from each AMS connected to store their K-values, with that done you can actually disable per print calibration but even leaving it on (though it will override the stored K value) it’ll only do it for the first filament
I have had horrible results with the Supertack on both my X1C and A1, someone suggested I changed my profile from SuperTack to Smooth PEI while using the Supertack and that worked for a couple of weeks. I made a video on my channel about it. I did a variation of temperature range following all the Bambu instructions.. But no success with them. It also seems I'm not the only one, so I am thinking that they have a bad batch run of those SuperTack plates. Other than that, those printers have been amazing and quite a game changer in my workflow. That has made me a huge fan of Bambu printers and it's not likely to change, but the 2 SuperTack plates I got are a huge disappointment.
The supertac build plate is not yet compatible with the X1C out of the box. To use it you have to turn off the auto bed sensing which could throw off other things if the software is not written well. I don't actually think this is the problem here but it annoys me greatly that I can't just use that plate on the X1C without crippling one of the features I paid extra for.
Please also compare the automatic calibration of the A1 series. It seems to use a totally different approach somehow sensing the pressure changes, so it would be nice to know whether that one is better than the lidar-based X1C one.
Damn, should have waited for the end of the video to hear you don't have one ... pity ...
Consistently good content buddy, thank you for doing this. ... first too :)
You can't honestly compare the X1 to the Qidi because the X1 isn't stock. You literally took what is an amazing printer out of the box and started modding it as if it were an Ender 3. As stated else ware, too many lights for the lidar and not even allowing the lidar to do it's job. The manual calibration that you have not done when the printer was in its stock form negates any comparisons since you don't know what the printer was doing in stock form. If you are going to do a true comparison, put the X1 back to stock. If you are still having a problem, there could be something wrong with the issue as your prints are no where near the quality of my X1 (bone stock) and all of the others that I have seen.
Love your channel. Keep it up.
also dont forget that x1c $1,350 and plus4 is $799, that's almost double the price
Unless the plus4 comes with AMS, the equivalent X1C is $1099.
on my p1p I never change the k factor from 0.02 and flow of 0.98 on pla.
To those who call the lidar a gimmick, I’m surprised no one has done a head to head test I’d like a P1S and an X1C doing 5 filaments in a row for both types of calibration. Because the X1C is hands off you just run it and it’s stored, but on the P series you have to manually type in the best looking result. It’s a QOL feature set I know but a big deal for those who use a lot of filaments and don’t have time to dilly dally
Not to sound negative here, but I think your printer issues on the X1C can also relate to the fact that you did not hit your subscriber goal yet.
You aim to be a reliable source on this topic, but you're not as knowledgeable or wise on the subject matter as you need to be yet. Hey, I'm not either, and we all can't know everything right? I'm just offering some sibjective criticism.
I think the comments here have proven some solid points about the X1C printer.
First, you need to remove all your "upgrades."
Bambu's tuning applies to Bambu's equipment. This is why rarely do they ever actually endorse a 3rd party solution. it's about controlling the process to ensure good prints.
When you have 3rd party equipment in their printer, and there's not a setting made for you to tell the printer or slicer (so it can maybe adjust the tune to accomodate your change), then you've introduced an unknown variable.
My P1S makes amazing prints and doesn't have Lidar, but it's also all Bambu. I've changed out the head and gear to the hardened steel ones required for the heavy duty filaments. But those are Bambu parts, AND there's a setting where I can tell it that I've switched those parts to their hardened steel counterparts.
Next, it seems everyone is suggesting that the Lidar does not work well with bright studio lights or LED. Definitely worth a try.
Comparing a stock QIDI machine to a Bambu that you've upgraded with parts not supported by the Bambu tune is not a fair comparison.
Your video of the X1C vs the QIDI is not a fair comparison because you've handicapped the Bambu with your mods.
If you're going to offer criticism, maybe don't start with a non sequitur. There are plenty of subpar channels with high subscriber counts, and great channels with low subscriber counts. I don't even know why you'd make that comment, other than to be deliberately negative...
I use Qidi X-Max 3 printers and always rely on manual calibration, together with a diamond nozzle for consistency. Still sometimes bad seams.
As per my experience it is about finding the sweet spot in the triangle of:
1. Nozzle temp (not discussed in the video)
2. Flow rate
3. PA (flow dynamics for Bambu users)
Other parameters like print speed, cooling, etc doesn't seem to be that crucial for fine tuning, as long as they are kept within reasonable limits.
I prefer slower print speeds, although I exclusively print technical parts.
But that is all just my personal experience I wanted to share.
Check your extruder gear?
Those prints don't look great. Might be a good idea to try stock nozzle and or fan. that 100% infill part looked pretty good done on the Qidi
For a 100% print i always assumed heat was a bit of an issue, since so much plastic takes longer to cool.
You should do the flow rate test first. The flow dynamics test is skewed if you’re not already using the correct flow rate. Doing them backwards means you’re not actually calibrated. The flow dynamics test isn’t listed first because you’re supposed to do it first. It’s just alphabetical. You are supposed to do the flow rate first.
Something’s definitely up with your unit. If you’d like to see some comparisons. I have a large number of makes on my printables “bowieinc”. But, look on the bright side, if you are having issues, that means someone else is as well and you are in a position to help:)
It obviously a filament path issue combined with overheated pla. If you want to print pla I think the other cheaper version is better. X1c carbon is designed more for abs and materials that need temperature stability, thus the full enclosure. Otherwise maybe leave the door open and remove panels if possible.
Looks like your filament is not 100% dry.
those prints are pretty bad, all have some kind of defect. is it the filament, how you store it ? any low end printer will look better
Well, you have a lot of 3rd party parts on your printer. Are you sure they are real long term improvements ? Because the print quality seems very low to my eyes. I mean we get better results on lower priced machines like our P1P and MK3s+ production machines. Despite we have a good lighting too. May I suggest you buy brand new original parts from BL and run the test again ?
If you do inner outer inner walls it will hide all over extrusions. It’s like night and day a 100% fix for over extrusion’s impact on the outer layers
Manual is the king. I've done it once for my filaments and never used lidar. It's just a useless feature for me as I don't own a farm
Dry filament?
Would love a teat between Mako, e3d, revo , swiss3d and more high flow hotends
I have an A1 and even after hours of meticuously trying to calibrate I dont get the quality and consistency i get with Dynamic Flow Calibration enabled. Ill revisit it some day, but for now im just enjoying click-to-print (coming from heavily modded OG Ender3).
On that note, it'd be awesome if you obtained an A1 to provide comparative results! Quality/settings/results wise your audience may be narrowly scoped otherwise. Great content though, sub'd!
Honestly, it's probably all of your 'upgrades'. Other than the Obxidian - everything else I've tried has lowered my print quality. Bambu tunes these machines well and they don't require tinkering with new fan ducts and the like.
Also....why glue stick? PEI or PEZ and a little temp is much easier to clean up after.
The manual filament calibration routines in Orca Slicer are much better than the abbreviated test in BS. I think Bambu limits the utility of the calibration process in the slicer in an effort to make the auto calibration look better than it is. The convenience of the auto is usually the crutch that X1 owners fall back to defend the X1, saying they can't be bothered to calibrate filaments. You get what you put in. Me, glad I went P1S and saved $500 on the marketing fluff.
Try different color filaments using the LiDAR and see if it changes. Maybe the camera doesn't pick up the color correctly or since it uses a laser certain colors don't read good because of it. I know Bambus LiDAR is a camera and a laser.
Impressive results from the Qidi. If you're doing what I think you're doing with that shaft adaptor, thermal transfer thru the shaft may become an issue for long runs; ruining that nicely toleranced fit.
Some things to consider:
1) Have you dried your filament? Even brand new filament can be too wet.
2) When was the last time you changed your nozzle? If it has been a long time that can cause bad layer lines and excess stringing.
3) When was the last time you cleaned and greased your z rods? The inconsistent layer lines could be caused by that if you haven't done it in a long time.
hehehe you're all spoiled by auto calibration j/k, if it worked really good on a modded printer, I would be all over it too!
I've only ever done manual calibration, from my first ever Ender 3 running klipper, to my newest Bambu Labs P1S, and everything in between (quite a few printers)
I really hate to say this, but until AI gets involved, nothing will ever match a manual calibration and an experienced eye... there are just too many variables IMO.
Love the channel! Looming forward to the next video, you have peaked my interest showing off that part ;)
Feels like you need the stock tool head for the X1c
I rarely write comment, but I appreciate ur vid. thank you. as far as I know the lidar flow calib doesnt save k value to the material profile. and your print look way too wet. my PLA stuff is nowhere stringy as your.. even with lower fan setting. keep up the good work!