5 things to check & tune to 3D print faster
HTML-код
- Опубликовано: 6 июл 2024
- So you want to push the speed on your 3D printer, but don’t know where to start? This video guide is for you. We will explore five limitations, and look at ways to overcome them so your 3D prints can be ready before you know it.
Thanks to Branden for requesting this video.
My free (and open source) 3D printer calibration website: teachingtechyt.github.io/cali...
0:00 Introduction
My previous speed benchy video:
Maker’s Muse slicer hacks video: • 3D Print parts TWICE a...
CHEP fast printing profile video: • Print Faster with Cust...
0:54 Firmware safety limits
How to check speed limits for each firmware: teachingtechyt.github.io/cali...
Fire video: • Flames of fire on blac...
1:42 Hot end flow rate
Simple free air max flow test: teachingtechyt.github.io/cali...
CNC Kitchen max flow rate test: • Easy Hotend Benchmark ...
Max hot end flow and speed test: teachingtechyt.github.io/cali...
4:21 Applying max flow rate in the slicer
6:19 Increasing max flow rate for free
6:47 Speed vs acceleration
Prusa calculators: blog.prusa3d.com/calculator_3...
8:48 Mechanical artefacts - acceleration test
Free acceleration test: teachingtechyt.github.io/cali...
Input shaping guide for Klipper: • Klipper guide: Input s...
Klipper input shaping: www.klipper3d.org/Config_Refe...
RRF input shopping: docs.duet3d.com/en/User_manua...
11:23 Part cooling
Stopwatch image credit: pixabay.com/vectors/stopwatch...
Buy quality and affordable filament from X3D. Buy 3, get 1 free and a free sample pack with every order: www.x3d.com.au
Get Quality Resins from 3D Printers Online. 5% off storewide for Teaching Tech subscribers [Code: tech5]
3dprintersonline.com.au/
Take a look around and if you like what you see, please subscribe.
Support me on Patreon: / teachingtech
In Cura, biggest gains for the lazy: (assuming your printer is already calibrated)
- increase inner wall, infill, skirt/brim and travel speed (don't increase Outer Wall speed).
- remove infill if the part supports it (more walls w/o infill is faster than less walls w infill),
- increase wall width to achieve same total thickness with less walls.
- enable _Filter Out Tiny Gaps,_ (assuming the model is watertight)
- set _Fill Gaps Between Walls_ to _Nowhere_
- turn on/off _Use Adaptive Layers_ as it sometimes saves time.
Agreed! Also disable infill before walls that way your infill lines won’t show through the walls when you’re printing 2 of them at 0.6mm line width.
Really appreciate the time and effort put into these videos AND the website, considering it’s all offered free with no expectations of patreon or subscriptions (although clearly he appreciates them).
Also to add, Michael makes it relatively easy for noobs to 3D printing to skip as many headaches and therefore turn them/us off of the hobby.
Keep it up! 👍
Michael, thank you very much for providing these informative videos and such innovative tools in your website! They really help to navigate around these complex topics on printers and slicers that will vary so much.
Your videos keep me modding my Ender 3 to the max. Thank you for all your research and shared knowledge!
Michel i LOVE you attention to function across every detail of 3D printing. Thank you!
I can't say enough good things about your calibration website...anyone serious about improving print quality or print speed should use it to optimize their printer profiles.
This was possibly the most useful 3D printing video I've seen. Now a subscriber.
You are a pivotal person for those that have been 3d printing for a while and want to hone in their skills to get to a professional or expert level. Thank you so much for everything you have done for our 3d printing community. 🙏
Your are the 3D printer Guru and You must spend so much time in researching.
No other person shows such good information and gives the community a website to use all the tools that you have designed & made for us to use
A big Thank you.
Thanks for giving so much to the community. Great work.
This is the exact video I am looking for to tune my printer. Thanks for creating the site.
Brilliant stuff - This man is a born educator.
Again an excellent article. While typing this I realize that positive reviews don"t help you at all to improve your skills further. And are only a confirmation that you are doing your job as a teacher very well. Chapeau!
Thank you Michael! Really amazing explanation… the best I ever seen about this topic. I will recommend for everyone that needed ;)
Thank you fot this video, much needed, I built a voron which would make a great video in itself, I had no clue where to start for faster printing, discord wasn't much help but you certainly was and placing all that on your web site was much appreciated.
You are an amazing human- great research and website!
Excellent video as always! Whenever I watch one of your videos I know I'm going to learn something :)
This is an excellent video, loads of info in here. Thanks for that!
Thank you for another informative video❤
Excellenty video man!!
Super useful video. Thank you for making this!
Perfect intro. Exactly why im here.
You are awesome, huge fan of your work sir.
This video is a wealth of knowledge
You, Filament Friday and CNC kitchen are my Goto's....
Really good video, thank you
this video is gold
Wow! Great info.
Fantastic work. Don't know where you find all the time.
Great info, great vid as always.
So now the community needs to create a database of parameters to maximize speed based on temp, material, speed, accel, jerk, hot-end type, nozzle, material and cooling.
@tradde11 ...and that is why we need to create a database for each.
Commissioning the VCore 3 500 with this at the moment. Thanks!
🥳
Great video and to add there should be something about obvious item is changing the nozzle to larger diameter to do thicker layers so less layers needed to print but sacrifice detail. I have stock .04 but now have a .06 to try out in the middle since .08 is too large for what I'm wanting
Köszönjük!
Great video! Another simple and relatively cheap flow rate upgrade is using a CHT nozzle.
I am trying out SuperSlicer because you recommend it, but out of the box I am experiencing problems, especially regarding overlap, multiple types. If you really do prefer SuperSlicer, it must be ok, maybe you could do a video on how to set it up? Thanks you so much for all your brilliant videos, have learned so much! 🙏
What an amazing video. I have a Bondtech CHT Nozzle and it's truely amazing I now need to look at more part cooling. I have used bullseye and hero me before but I added direct drive so I had to remove the bullseye. What part cooling are people using?
Some good tips and I will check out your calibration website. Have you done any testing on ringing on square/straight edges prints by comparing rotating 45° on the bed so no straight edge perimeter uses just X or y axis to print but both? Also, I've been doing some experiments with prusaslicer fuzzy skin - try depth 0.06mm, length 0.1mm - it gives a cool, subtle texture and makes parts stiffer in my opinion as the internal perimeter crosses over the deviations of the outer perimeter forming a better bond. I have tested with functional parts with a 0.3mm tolerance and they did not fuse together.
Great Video - Only thing missing IMO - especially if you go really fast - is layer adhesion. Expecially on bigger parts where the previous layer had plenty of time to cool down. The next layer needs to bond with the previous one. So If you lay down a molten noodle too fast the layer adhesion will suffer. Not critical for a "look at" model like a figurine but it can make a difference on a mechanical part. Also just checking the weight of the extruded filament IMO is not enough as at high speeds you still might get the same weight of filament but if you look at it the extruded material is shorter and thicker. This happens when the pressure inside the hotend gets too big and it compresses the molten plastic too much and then when it comes out of the nozzle it expands
If you get layer adhesion problems when the previous layer has cooled all the way, your nozzle temperature likely isn't high enough or you're underextruding from going too fast. I used to think "printing on a cooled layer" was my problem for petg, and turned fan down or even off for it, but got bad surface quality and overhangs from doing so. Just cranking the nozzle temp up allowed me to print petg with full cooling (and I have some rather extreme cooling).
I learned a lot from your video, thanks. Does the ender 3 v2 have a max speed limiter to prevent any damage to steppers or other parts from extended printing at high speeds? Thanks
On a printer without some seriously upgraded cooling, your flow test piece is likely to hit cooling failures before extrusion rate limits. You can mitigate this by blowing a desk fan at the printer but then you need to make sure your hotend is sufficiently insulated or you might just strip away so much heat from the block that the flow drops.
Another great video…
Another Important setting for maintaining quality when printing fast seens to be Jerk, corners tend do become bulges when it’s too low, even when acceleration is correct, because the speed is not translated into corner turning velocity and we have a slight stop before starting to accelerate again
linear advance / pressure advance fixes that
Step 1)
Install Klipper
Step 2) figure out input shaping
Step 3) change your max accel values
Step 4) figure out pressure advance
Step 5) yolo
step 1 will usually get you about 10x ball park
@@TommiHonkonen10x of what?
Just started using klipper recently. I actually enjoy tuning and tweaking my printer now. It's so easy to change things compared to marlin
@@icebird76 starting point
@@TommiHonkonen I still don't get what you're saying. I installed Klipper and I see maybe a 25% faster print time on average which is only 1.25x. Nothing is 10x or else a 1 hour print would take 6 minutes.
This is a nice video and the calibration website is very handy. There is still one thing that I am confused about. I can see that my firmware has limits on X and Y motion speeds. And I can see that the calibration site generates G-code with lines like "M220 S100 ; custom speed A - 20 mm/sec". My understanding is that M220 is sets a relative percentage. How does it know that 100% of /something/ is 20 mm/sec? What is that something, where can I find it, etc?
Have you been able to get the automatic died seeing to work in Superslicer?
Chep has a crazy fast cura profile with settings cloned from ultimaker settings- requires a lot of tweaks- bit worth it
Will you be doing a Cura 5 video?
I noticed that when you start to reach the limits on your hotend, the filament it pushes out becomes thicker than the nozzle size. Increasing temperature or slowing the speed will fix that. The plastic is not coming out at same temperature and speed, some of it slows down and this allows the extruded line to become thicker. Skipping is related to the wheel grip, not just hotend. I use both now, just see when the line starts to look different and that is your maximum, at those temperature. The sensor does not measure nozzle temps but block temps, it can be seriously lower inside the nozzle at high speeds. Of course, without adjusting the pressure that the extruder gears have is a way to get faster extrusion so the free air print test still applies to the current state of the machine. But it might not be just hotend that is starting to lag behind, it can be that the extruder has too weak grip.
Is the filament swelling not die swell from internal stresses?
@teaching tech im hoping you can help. i bought an used Ender 3 of FB. It was a printer that the person was still using so i know it works. The issue i have is it looks like there are a few thing that were upgraded and not sure what all is on the printer. is there a way to find out easily. im looking at upgrading the mainboard so need to know what all needs to be done when down loading the new firmware
General question, when I use the permiter speedrates you've used on your video within your teaching guide, my returns are half that for example you have 100mm/sec feeding back on the screen as 28.80mm3/sec, when I enter the same number on your website I get 9.60 mm3/sec returned in the calcuated volumetric flow?
Using volcano hotend on ender 3v2. The termistor is relocated, so now it touches nozzle's thread. I can print at 100mm/s with 1800 acceleration, 8mm/s jerk
I feel like all of the different settings and dials here would lend themselves really well to an AI application. Provided you could train with all of the parameters of a printer and feed it back real world data about how well something printed you could potentially have an AI help tune all of these settings in.
For some materials, too much cooling can lead to fragile parts because of poor layer adhesion. PETG, for example, should be printed with zero or very little cooling, except when bridging
Hi Michael, I gave it a go to your acceleration tuning test, and for some reason I can't figure out, the print head stops a second or two, will still extruding causing a blob every layer on the same spot. I use Cura. Any clue please? May be your Gcode got somehow corrupted? Ta!
at 2:54 the chart shows underextrusion at around 4mm3/s. But in your test, you were able to get ~20mm3/s. I assume 1) you have an upgraded hot end, or 2) You are willing to tolerate an amount of underextrusion so long as the part holds together. This would mean dimensional accuracy and structural integrity would be sacrificed. Am I correct?
🎯 Key Takeaways for quick navigation:
00:00 🚀 *To 3D print faster, start by identifying and managing limitations.*
01:45 🌊 *Hot end flow rate is a crucial limitation to understand when printing faster. Test your maximum volumetric flow rate to determine the fastest speeds you can use.*
04:04 🐢 *Adjust your slicer settings for maximum volumetric speed based on your hot end's capabilities to prevent extrusion issues.*
06:41 🌡️ *Increasing hot end temperature can improve flow rate, but it may lead to other printing issues, so find a compromise.*
08:32 🏎️ *Consider acceleration settings as they impact print time and quality; increasing acceleration can lead to faster prints but may require adjustments.*
10:54 🛠️ *Mechanical artifacts like ringing and layer shifts can affect print quality when printing fast. Experiment with acceleration settings and input shaping to find the right balance.*
12:06 ❄️ *Adequate part cooling is crucial for maintaining surface quality when printing fast; upgrading your part cooling system can improve results.*
12:34 🔧 *Combine strategies like adjusting feed rates for internal infill and increasing travel feed rates and acceleration to optimize printing speed while maintaining quality.*
Made with HARPA AI
Can you use your website to calibrate klipper
where do I apply max flow rate in Cura?
i have been printing 95A TPU with a 1.2mm volcano nozzle at 0.6mm layer heights.
i have theorized that i am encountering 3 types of flow rates.
the flow rates are:
The Nozzle
The extruder
the material.
In this case i found that my nozzle and my extruder could handle the high flow rates but i suspect the material cant handle the high flow.
So, by using the speeds recommended by the manufacturer i was able to determine a flow rate that is within the calculations and the prints are far superior.... althougg much slower..
For decent print speeds with TPU, Overture has a "high speed TPU" product (95A) that seems to work. I can print it at about 8-8.5 mm³/s.
@@daliasprints9798 interesting. i think mostTPUs max out at about 3mm^3/s
@@Mr_Pewpy_But-Whole Yeah. I can do a little over that, but I have a really good extruder pushing it (Flex3drive G5 which you can see on some of my videos).
Not sure if i'm doing something wrong, but I cant seem to get any of the gcode to move past the starting sequence in klipper
Visual maximum flowrate test might not be accurate - there most likely will be underextrusion before perimeter breaks (example: Bondtech CHT test model, 0.7 mm requested line width with 10ish mm3/s flow ->
Sorry off topic but is two beeps mean it did not accept the change ?
Sir,
I tried to do the volumetric test BUT the speed never changed. What an I missing?
Thanks
Kelly
Does anyone know why I set my feedrate to 500 above and my print still intact and alright?
I have set M203, M201 and M204 to exceed the limit. I have also update the firmware to mriscoc
P.S I use creality ender 3 v2
so I got my ender 5 plus with DDX and mosquito hotend using the speed and max flow. Perimeter speed up to 2000, printing abs, perimter speed 60 is defienetly the best with 80 not far from it, however above 1000 i didnt notice any difference
0.6 nozzel 0.4 layer
Its fumo friday my braddas ᗜˬᗜ
"Trust me, theres a lot of speed to be found with higher speed"
Sun Tsu or whoever-
4:51
Thank you Michael, you are our great vampire brother. I am also trying to trade PLA for blood but life is different.
I'm not entirely sure a revo with a .4mm nozzle is capable of 20mms flow rate. It's likely that like you explained, your acceleration isn't high enough during that test to ever reach max speed.
My acceleration is 5k. I made a whole video about testing the Revo at high speeds with print times included. Check it out.
@@TeachingTech The max flow rate advertised on the revo is 10mm^3/s. Is it possible you've made some mistake? I barely get 22mm^3/s on an nf crazy hf (basically volcano) and 8mm^3/s on a v6 so 20 would be surprising on the revo.
@@TeachingTech unless you're using a larger/smaller nozzle than .4
Watch the Revo video and see for yourself. 0.4mm nozzle, 0.2mm layer height is my normal. In terms of their advertising, I think they are under promising and over delivering.
Also worth mentioning I'm printing pla at 225 deg C which helps. The Revo is very good for not stringing so no problems there.
Even though your videos help me with something’s because the Enders have the same hot end that’s all they have in common so when are yougonna start using a CR10s for some upgrades,not everyone owns a Ender!!
Your flow rate calculations might give people trouble with SuperSlicer. You are calculating based on a rectangle of layer height by line width. SuperSlicer calculates the flow based on rounded edges.
You look like if graham stephan and bababooey had a baby
Okay now can you repeat all that.. but dumber. So I can understand..
You did NOT just tell people that infill underextrusion doesn't matter just because most won't see it! Are you TRYING to throw away the respect you've earned as a member of the 3D printing community!?
This was too hard to understand as a hobbyist. You lost me within 3 minutes.