Get the benefits of a 0.6mm nozzle with a 0.4?
ฝัง
- เผยแพร่เมื่อ 15 ก.ย. 2022
- Thanks to Private Internet Access for sponsoring this video! piavpn.com/Toms3D
0.6mm nozzles provide some awesome benefits, but can you get there with a 0.4 as well? Is there a downside in strength? And why did I make a skewed comparison in the first place?
Read the article to this video here: toms3d.org/2022/09/22/get-the...
Models shown:
Treefrog by MorenaP www.printables.com/model/3116...
Eastman on Artstation www.artstation.com/eastman
Relevant products
Prusa MK3 (S+) go.toms3d.org/Prusa
Prusament go.toms3d.org/Prusament
Product links are affiliate links - I may earn a commission on qualifying purchases (at no extra cost to you)
🎥 All my video gear toms3d.org/my-gear
I use Epidemic Sound, sign up for a 30-day free trial here share.epidemicsound.com/MadeWi...
🎧 Check out the Meltzone Podcast (with CNC Kitchen)! / @themeltzone
👐 Enjoying the videos? Support my work on Patreon! / toms3dp - วิทยาศาสตร์และเทคโนโลยี
Thanks to Private Internet Access for sponsoring this video! piavpn.com/Toms3D
When you have time again, maybe try a 0.6 nozzle with 0.8 settings ;)
Where's the LTTea store?
I think I have the answer to your extrusion mystery @7:43 minutes. It has to do with die swell see CNC kitchens video "My High Flow Nozzle is better than a Volcano hotend!"@ 10:30~ minutes. You could maybe turn exploring this property into a video.
Actually, Stefan had experienced the same effect a while ago; he had printed a bechy with a 0.25mm nozzle with gcode sliced for a 0.4mm nozzle. He had noticed that it printed the best... Maybe you guys could discuss this in your next podcast episode? :)
Perhaps it's the pressure build up inside of the nozzle where it has less room to slide out easily so a more restricted flow which results in more precise placement just like how a ketchup bottle is more precise with a smaller tip? I feel like long term that could cause clogging though?
sounds like room for optimization in extrusion volumes and line widths
@@ademonnamedburt1232 that can cause problems actually, the increased pressure can induce stresses into the print. The increase in pressure can also cause issues with oozing and retraction.
@@notsam498 I feel like retraction can just be increased or modified to fix that though because oozing is based apon the retraction setting?
@@ademonnamedburt1232because things often are moving very fast the retraction itself won't be quick enough. On a slow printer that's fine (kind of), on my voron it's a different story. You have to remember everything bellow the extruder can get compressed even unmelted Filament. It turns into a spring.
You’ve essentially proved why my Extra Fast 0.4 profiles (with 0.6 extrusion widths) produce better results than a standard 0.6 profile.
Your mystery around 7:50 is the filament doesn’t flow past lower layers because the center doesn’t shift as far as your sketch shows. The center of the nozzle is the same for 0.6 and 0.4. So the “over extruding” 0.4 still puts plastic on the lower layers but a bit slower (due to the 0.4 restriction). Slower typically gives smoother results so you get slightly better results in your experiment.
The center of the nozzle should not be in the same position for a 0.4 and 0.6. They should be slightly different positions (approximately 0.1mm difference)
Slightly different positions due to the slicer calculating where the larger nozzle is, not the smaller one. I upgraded my CR10s to a 0.8 nozzle, and in no way will that be in the same x/y position as 0.6 or 0.4 for that matter. I turned a 6 wall parimiter into a 2 wall parimiter and a over 7 hour print jop into 2hours and 30 minutes. These are for functional parts inside an object. So quality isn't an issue. But I found the spirale vse mode prints beautiful in their own way, and bloody strong with the 1.2mm wall thickness.
@@tylers2889 Absolutely.
@@tylers2889 Depends on the line widths, not the nozzle widths. The tool paths depend on the line widths, not the nozzle widths. At least it does in Cura 5, there's no actual nozzle width setting.
@@iskandartaib Yes, correct, but I shouldn't have to make that distinction in order to not be nitpicked. I was refuting his claim which was clearly wrong. And I didn't say nozzle width
I think you missed the correlation that the faster volumetric flow of 0.6mm GCode through the 0.4mm nozzle not only resulted in lower temps, but also that the lower temps are primarily what improved the overhangs.
That's actually a really good observation and something that I think is worth exploring!
It would not have resulted in significantly lower temps, the hotend controls the temp automatically. The temperature is effectively at a steady state and while more is flowing through it has plenty of time to reach the temp, though for some printers the printing setpoint might have to be changed if the hotend isn't keeping up. Most printers can handle significantly more volumetric flow rate than normal speeds/nozzles.
Huh? The volume of filament per time is defined by the gcode, and he used the very same gcode, so how on earth do you figure the volumetric flow is different?? The only part where there's any difference what so ever is that the flow is faster at the very tip of the smaller nozzle, but the volumetric flow is still the same, i.e., the exact same number of mm³/s of filament is being pushed through the nozzle in both cases. But why would that temporary filament speed change while squeezing through the very tip affect any temperature?
@@rileywebb4178 There's exactly the same amount of filament flowing through (not more, not less), so you're absolutely right that the temperature isn't different.
@@rileywebb4178 What makes you think the temperature is steady state? It is steady state at the thermistor, but not within the filament itself. There is already plenty of evidence that increasing flowrate can prevent the material from reaching the desired temperature even though the heat capacity from the heater is well above what is needed to melt it. The problem is that you get a thermal gradient. The more flow you put through the nozzle, the larger the gradient becomes, because the heat required goes up but the thermal conductivity remains the same.
As Thomas said, with a 0.4mm nozzle you have a smaller surface area for the heat to transfer through compared to a 0.6mm nozzle, which means the 0.4mm nozzle has lower thermal conductivity than the 0.6mm nozzle. This is also why the CHT nozzles and the other modified nozzles made on CNC Kitchen can handle higher flow rate even with the same nozzle diameter - they have larger surface areas which increases the thermal conductivity and lowers the thermal gradient.
I love how your channel actually answers questions I have when I'm configuring my slicer settings. Thanks for the video.
I've been using 0.6 settings on a 0.4 nozzle for a while now, mainly for speed, and for the functional parts that I print there has been zero difference in quality.
I got myself another printer a few months back and immediately upgraded it to a 0.6 nozzle, and have been using 0.8 settings for everything other than the external perimeter, and the quality is still great, and the speed is even faster!
What layer heights is the .4 nozzle capable of with the .6 sliced settings?
@@danielellen-barwell7332 I do 0.28 with variable later height in Prusaslicer
For me this is the real advantage of an 0.6 nozzle. I regularly print 0.8mm line widths and rarely need the detail that an 0.4 provides. The 0.6 is a good middle ground so that I'm not constantly swapping nozzles. I find that 0.6 prints 0.8 line width better than 0.8 in some cases, in the same way Thomas shows here with the 0.4 outperforming the 0.6.
So if 0.6mm line width is your goal, an 0.4 nozzle makes sense. If 0.8mm or a little larger is your goal, 0.6 seems to be the way to go.
@@d3c0deFPV what layer height do you print with 0.6 nozzle?
@@smartinsilicon It depends on the part, but I'm often printing parts for quadcopters out of TPU or Nylon which will be subjected to repeated crashes. I find that sticking with around 0.2mm layer height is best for durability. Haven't noticed much difference with up to 0.24mm or so but I saw a sharp decline after 0.28mm.
If durability is less important, 0.3mm is a good height for faster parts. Don't get me wrong, 0.3mm is still going to produce very durable parts, but when you're flying into solid objects hitting the ground at 60mph+ and trying to protect sensitive electronics, every little bit counts.
You're strength concerns were answered in an old cnc kitchen video where it was discovered with a constant wall width, increasing line/extrusion width _increased_ the part strength, with 100% infill this works up to 140% of nozzle size, with normal settings it works up to at least 200% of nozzle size. For this reason I have been using 0.5mm line widths, and have felt no need to switch nozzles. The overhang improvements and quality results are also interesting.
I for my part, I am more curious about layer _height_ than extrusion width. Wider extrusion is "easy" (to a realistic extent ofc), at least with V6 and Phaetus "shaped" nozzles, although there might be sliightly less precision control. Layer heights that 0.6 can do however, that one might be a bit more difficult to increase with a 0.4 nozzle, and still retain same or similar strength... (?)
@@pr0xZen I assume you are saying larger layer heights give more strength, but according to research, the opposite is true: Kuznetsov, V. et al (2018). Strength of PLA Components Fabricated with Fused Deposition Technology using a Desktop 3D Printer as a Function of Geometrical Parameters of the Process. (figure 5). CNC Kitchen also covered this in one of his videos.
I followed your recommendation and I switched from 0.4 to 0.6 nozzle and it was a game changer for me. Now I print at 0.35 layer height and times were reduced to a third without a mayor quality change. Thanks for this awesome tip.
The layer heights is the main point why changing to 0.6mm still can make sense. Otherwise you can stick with the 0.4mm and print in 0.6 or 0.7 line width.
Love that you don't hesitate to experiment, keep going Tom!
No this is Patrick
Haha you did a big funny wow
Considering that i hard-baked my nozzle with several years of PLA buildup and can no longer change it to a 0.6, I am very happy I found this video.😂
The .4 with .6 settings might be catching better due to increased die swell out of the nozzle. I remember from a composites course a while back that if you increase shear rate, die swell out of a nozzle will also increase. If the extrusion rate is set for a .6 but a .4mm nozzle is used, I could totally see shear rate increasing through the nozzle
Interesting. In rocket science we work on mass flow rates, constricting the flow would increase the mass flow rate as it tries to keep the amount of material the same. And no, I am a self taught rocket scientist. I find this stuff interesting.
Fascinating. Great for those who don't switch to easy-to-swap Revo's. I've got so many .4 nozzles to go through on my older machines, so this is good to know. Any new build(s) will have Revo's, so swapping to .6 will not be an issue... but... that little quality difference in the overhang is interesting enough to cause one to pause and think! Great follow-up Thomas... Good luck with your move!
You don't need to use 0.6mm settings for 0.4mm nozzle. Set perimeters to 0.65mm with 0.4mm nozzle and you will achieve the same results. Because when you use 0.4mm nozzle instead of 0.6mm, printer will push trough nozzle the same amount of material. Which for 0.4mm nozzle means just a bit overextrusion with squeezing material into wider lines.
That's also the reason why switching nozzles from smaller to bigger does not give nice results. Because in that case you are constantly underextruding for used nozzle size. 0.6mm nozzle can't print nice 0.45mm lines.
The only reason to change to 0.6mm is the layer heights. You can print 0.35mm layer heights with a 0.6mm but you can´t with a 0.4mm
You just answered my question. Thanks for putting all this effort to find out!
Good luck with the move Tom, hang in there man. I really appreciate the info you share!
This is fascinating. Please continue investigating when you have a chance!
Very interesting, good stuff to try and know. Thanks Tom
Yep that was interesting, thanks for doing the test and sharing the results. The surprise 4th result was a real treat.
I actually use a extrusion with up to 0.8mm with a 0.4mm nozzle, when printing vase Mode models. Really helps with the structural integrety of the models. You just need to up the temps a bit and slow down the Print Speere so the extruder can get enough filament to the melting point. Which is probably your issue with the matte print. As you said the filament did not have enough time to get hot enough. So upping the temps by 10 or 15 degree would probably solve this, or slow down a bit.
I've used a 0.4mm nozzle for extrusions up to 1.5 mm wide by 0.5 mm thick in vase mode with no problem. The max layer height you can do seems to depend on the extrusion width for the extrusion width seems to be limited by the diameter of the flat part of the nozzle, not the opening of the nozzle. When I tried the same experiment to see how wide I could extrude with a 0.8 nozzle, it wasn't really any different than the 0.4, because the diameter of the flat tip of the nozzle was about the same. Print speed is limited by the volume flow rate, which is a characteristic of the heater and hotend. That is, thicker extrusions only reduce print time to a point, because you eventually need to slow down print speed to stay under your volume flow limit (and cooling capacity).
Same here. You don´t even need to change speed and temp if you choose the volumetric print speed thingy for y extuder. i use a volcano nozzle in my non vulcano hot end and 15mm³ volume for it everything works great at 0,8mm.
Thanks for the follow-up
I have been using a 0.6mm (or even 0.8mm) extrusion width for at least 5 years now on my 0.4mm nozzle machines. Even back in the day when I first got simplify3d I was doing it and it was always great.
It lets you get better quality than a 0.6mm but with the same print time as a 0.6mm when you need it. And with most slicers ability to dynamically change the extrusion width it has only gotten more effective over the years.
I've done with a 0.3mm nozzle, 1.2mm widths.
Larger nozzles have very few benefits that you get, you should be maxing out your flow rate before you even think of getting a bigger nozzle.
I feel like this is the most underrated 3D printing hack ever which is only little talked about. I first started printing with 0.8mm lines and 0.4mm nozzle when I needed some boxes to only get one perimeter outline instead of two. Parts are much stronger and print way faster.
Super interesting! I'll try that. Thanks for your great videos.
It's always going to be a balance between detail and speed. .4 has more detail capacity .6 has more flow capacity. Arachne rules and I have been using 0.6 anyway for years using manually adjusted line widths (it was often annoying having to manually tweak the line width until the preview looked good) pushing the line width wider than the nozzle diameter always helps overhangs
I have found that lying to your printer about nozzle diameter can work sometimes but be super wonky other times. Instead, I just change the extrusion width and layer height values themselves, but keep the nozzle diameter to what the real nozzle is. I've been printing mechanical parts with 0.4mm nozzle using 0.6mm extrusion width and 0.5mm layer height for years now, and it works quite well. I've been designing the with dimensions that are multiples of those values, to make sure I get nice solid parts too. With vase mode prints, I keep the 0.6mm extrusion width (or sometimes bump it up to 0.8mm) and use variable layer height between 0.3mm and 0.5mm. I've found I get strong parts and height speeds! It almost seems like it shouldn't be able to get both, but here we are.
Thomas, I am so happy you are moving to Hawaii. Keep on making
I’ve currently got a 0.3mm installed so may go for a 0.4mm profile. Thanks for the tip Tom
I’ve been using this technique for a bunch of functional parts over the last week or so, and it’s really been working wonderfully!
Even though it only takes like 5 minutes to swap a nozzle, I’ve been dreading it, so just kept the 0.4 and set it to 0.6 in Cura. Really happy with results.
What settings did you change in Cura? Nozzle diameter, line width …? Thanks in advance
@@tomvs4144 Just the line widths, Cura calculates the rest. Super easy!
@@BRUXXUS great, thanks !!!
I watched this a month ago and then suddenly I needed a new nozzle. I remembered this video and gambled on the .6 to save time and honestly it took less time that you even suggested to retune the printer. Increased flow by half a percent and increased temperature by 5C. 2 setting changes from my profile apart from changing Cura to .6. No going back now. My eyes can't tell the difference at even the smallest scale I print at. Anything smaller and I would have to use SLS anyway.
Thanks for the suggestion.
I'm really glad you decided to look at this. I have essentially been doing this for years, and have even printed essentially 0.8mm widths with my 0.4mm nozzle, and it worked fine. Like you say, you get a few more holes and artifacts in infill, but if that doesn't matter in a print you can drastically decrease the time it takes to print. I still did go to a 0.6mm nozzle after your "0.4mm nozzle are obsolete" video, but that was because I could see the advantages of being able to print even wider perimeters, and I didn't really lose anything by kicking the 0.4mm nozzle off the printer. Thanks as always for the excellent analysis and informative video!
I routinely print 1.2mm widths with a 0.3 nozzle. I strongly advocate that almost everyone goes to a 0.3 nozzle as it gives them more detail *and* they aren't going to print slower because they haven't been printing at their max flow rate anyways.
@@SirSpence99Generally speaking I agree, however you can´t increase your layer heights much with a 0.3mm. You can print line 1mm wide and 0.35mm high with a 0.6mm but you can´t with a 0.3mm. I currently use a 0.25mm and 0.4mm nozzle. The 0.25mm prints 0.4mm line widths.
@@sierraecho884 If you are printing at near your max volumetric flow, increasing layer height doesn't make the print go faster.
@@SirSpence99 If so, then yes. However a 0.3mm or 0.25mm nozzle with it´s max layer higth will probably not reach max flow rate with PLA for instance. So it has it´s perks but I think I stay with my 0.25mm and 0.4mm set up.
@@sierraecho884 It has been a year since I last did the math but iirc, at 1.2mm line widths and 0.1mm heights, you get 15cubic mm/s flow at like 80mm/s speed. That would max out a v6 hotend. With all of the bells and whistles (better hotend, volcano cht) you can probably max out at ~50 cubic mm/s at that nozzle size so roughly 250mm/s travel speed.
All of those are entirely doable.
Since I was printing at 1.2mm widths, I discovered that the knockoff nozzle I had since it had a flatter bottom than the original actually printed better. Since I haven't been able to find another nozzle like it, I find the max I can do is around 0.9 mm widths.
On the overhangs: Well i'd assume that with 0.6mm gcode: The extruder is instructed to pushing 0.6mm quantities of filament through the 0.4mm nozzle, which would mean increased pressure.
And idk if i recall properly, but i do believe hearing that at high pressure the filament doesn't extrude in a clean line out of the nozzle, but instead expands a little into a blob as it leaves the nozzle because of the viscous nature of molten filament. It is the best idea i got as to why it shows different behaviour and still is able to snag onto other existing filament.
This series is such a emotional roller coaster
I’ve been using a 0.6mm setting in Cura but still have the 0.4mm nozzle. It has worked great for geometric prints and vases. Cuts time down and the finished product is crisp.
What settings did you change in Cura? Nozzle diameter, line width …? Thanks in advance
I switched to 0.6mm nozzles on my own printer and the work printer and ive just switched back. From my experience. You do lose quite a bit of detail going from 0.4-0.6mm regardless of slicer. Also with 0.6 you get a bunch of other problems like worse over hangs and cooling, and more stringing, and worse bridging.
Although the speed gain is nice, and the access to thicker layers is nice. I think it really depends on what you are likely to be printing.
Which is why im keen to get a revo or a second printer.
There's a CNCKitchen video where Stefan tested the effect of increasing line width on strength when the actual nozzle diameter overall shell thickness remained the same. I've been using 0.5-0.7 lines with 0.4 nozzle for a while and it's been pretty good
I regularly print in vase mode with a .8mm nozzle with 1.6mm line width. You get really strong prints, and I'm about to try 2mm lines from a 1mm nozzle.
Thanks Tom! I changed my 0.4 Nozzle X to a 0.6 Nozzle X on my Prusa Mini+ and the Benchy came out better at 0.2 layer height than with the 0.4.👍🏼👍🏼👍🏼
Overall my .6mm nozzle experience has been quite positive! I like the surface finishes I'm getting a little better, seemingly less stringing, faster print times, and it requires _almost_ no changes. I say almost because the one issue I did have was engraved lettering definitely wasn't as good on .6 (even with arachne). This was most apparent on engraving on top layers. Bottom layers were impacted by this too but not as much. I had some positive results changing the font, depending the engrave, increasing the size, and switching to engraving on the bottom rather than top. I was limited in how much I could do this depending on the specific model. Between all that I couldn't really get the quality I had with .4mm but overall I've been able to work around this and get acceptable quality. I can't use raised lettering due to the application though I think that would end up being better.
Everything else was the same or improved so overall it's been a really positive experience. This was all largely with MK3s'. I just setup a Voron V0 with a Revo which I may use to compare at some point. I'm holding off on Revo's for the MK3s' in hope that Prusa announces a MK4 (or MK3s+#^$ or something :P) which moves to a rigid mount Revo (like my V0 has). Given the easy swap options, I do expect I may still use .4mm for some models where I do want nicer engraving where I'm limited on how many of the above workarounds I can end up doing.
I switched to 0.6 after your video and it is great, 0.4mm layer height is just irresistibly fast since I do not mind the visible layers.
For some reason when I print at 0.4mm layer height it still takes an hour and a half plus for a regular old benchy. Is that normal?
@@zacjohnson452 No not normal, it takes about 35 minutes for me. For PrusaSlicer you need to add a new printer using config wizard with 0.6mm nozzle, then you can select the presets with different layer heights, it will be faster. For cura I think you need to set printers->machine settings->extruder->nozzle. After that you can select coarse profile.
@@zacjohnson452 not really. What is your print speed? With 0.6 nozzle, 0.2 layer height, 3 perimeters, 15% infill, and 35 print speed it takes me close to an hour and a half.
didnt expect to see you do have to move, saw it on the meltzone, hope it all works out for you
Great video Thomas
ive been doing this for years as i was lazy to change nozzle size but worked fantasticly. i found that if you use 0.6mm for everything but changed only the external walls to 0.4mm you get the sharper corners.
Same here but down to 0.38 sometimes for the external wall and down to 0.3 for support structures so they snap off easier.
Absolutely agree.
@@Jakalwarrior The support structure hack is so underrated, thank you.
Overhang explanation is simple. At overhang angles that work, the center of the nozzle orifice is above the existing material below it, but part of the orifice hangs off. With a relatively large orifice, there will be a large area hanging off with no backpressure, and material will want to shoot out there. With a smaller orifice, most of the area will be extruding against something, then squishing outwards fairly uniformly in all directions as intended by the slicer.
Oo this one is gonna be a goodie
Well, about the overhangs. I believe there are two main factors at play here. First of all, for a 45 degree overhang (with 0.2mm layer height), the nozzle will only move outwards 0.2mm per layer. So even on a 0.4mm nozzle, half the nozzle will still overlap the edge of the print. And your argument kinda goes both ways. While a 0.6mm nozzle may have more area overlapping with the edge of the print. It will also have more area that is not overlapping the edge of the print, potentially making it easier for molten plastic to droop.
The other factor is that on a 0.4mm nozzle, the nozzle will be over extruding (to make a 0.6mm line). Therefore if you move the nozzle out by 0.2 mm, the nozzle will only actually be moved out 0.1mm over the edge of the print (since the edge of the print was already 0.1mm beyond the circumference of the nozzle bore). On a 0.6 mm nozzle, this will not be the case. In fact, if you use a 0.6mm nozzle to make 0.4mm lines (like in the last video), your overhangs may even be worse.
Of course, this is just speculation on my side... So perhaps it would be worth investigating further.
My theory about why the 0.4 nozzle printing 0.6 settings is : the nozzle doesnt start over the air like your diagrams show.
(I believe Nilz is suggesting the same)
From what I have seen they always start the perimeter elsewhere then move out towards the air. So it will stick to the bottom layer first, gets squished then drag a bit out like a bridge, and i think that given the plastic usually behaves a bit sticky and all it is enough to keep that shape (especially so for your examples where they come from within and only go travel a little before going back over solid previous layers). Part cooling should also help with this new track keeping it's "shape".
Maybe the squishing also makes it more flat since 0.4nozzle printing 0.6settings uses the brass nozzle to "iron" more of the track than the other 2 scenarios.
One way to test this is to do some prints where the overhangs have longer lines parallel to the overhang edge. ie the nozzle will move out over to the air a bit, then continue moving "on air" like your diagram shows for much longer (so much wider overhang pegs on your test piece). If you do a range I think you will see the quality change as you increase the peg width, much like you will see with the bridging tests.
Edit: added part cooling effect
Hi, just one word on your last test : slicing in 0.4 and printing in 0.6mm. I did it because i made a mistake when i have replaced my old 0.4 nozzle on my QIDI X Max 3. I was surprised but the result was very good, but it was with specials conditions : the material was ABS-GF 25% and the nozzle was hardened. I think it worked because of type of material.
The overhang on the hat starts from within the part, so the extruder only makes a small loop outside the perimiter of the previous layer (print direction perpendicular to overhang edge). On more parametric parts, edges are usually relatively straight so there the nozzle will move next to the perimiter of the previous layer (print direction parallel to overhang edge). This might be the reason why the overhangs might be so good in this case. Although that still does not really explain why smaller nozzle with wider extrusion is better than both large nozzle with wide extrusion and small nozzle with narrow extrusion.
Didn't went through all comments - as several others posted: claiming to have a bigger nozzle than you actually have may result in better quality should be the same effect as telling the printer to print in higher line width. The result of claiming a 0.6 nozzle is that the printer by default picks 0.6 line width and extrudes the required amount of filament for this. Which means when doing this with a 0.4 nozzle the filament gets squeezed far more and the line is flattened resulting in better layer adhesion and stability as well as potential of achieving better overhang structures. The in-creased pressure is counterproductive but the increased overlap is helping - so there must be a sweet spot. My expe-rience is 10%-20% increased line width works best for me. So I would assume claiming you have a 0.5 nozzle while having a 0.4 should give kinda best results (estimated).
Of course, you have countless parameters and all interact somehow with each other. E.g. claiming having a bigger nozzle is also increasing the flow and to the same time the same effect as reducing the hotend temperature slightly which helps with overhangs but may have negative impact on layer adhesion. So its quiet complicated but I would still argue claiming to have a higher nozzle diameter is quiet similar to increasing the line width.
LostInTech did a great video on this a while back. Ever since watching that, I've been running my Ender3 at 6mm line width through a 4mm nozzle. I haven't noticed any real quality change, but my print times are significantly less.
I have a theory about why the overhangs print better with the 0.4 nozzle/0.6 settings. I think your sketches exaggerate the amount of offset-for a 0.4 nozzle to be entirely beside the existing 0.6-wide line, that implies a 0.5 stepover (from the center of the line to the edge is 0.3, but that means the center of the nozzle is over the edge of the line, so go another 0.2 to get the nozzle entirely in the air). 0.5 stepover with 0.2 layer height is almost a 70 degree overhang-decidedly unrealistic for everyday printing. If we reduce the stepover to 0.3, for a still-demanding ~57-degree overhang, that means the nozzle is centered on the edge of the previous-layer wall. But what this means is that the 0.6 nozzle extends 0.3mm past it, while the 0.4 nozzle extends only 0.2 past it-there's actually *less* nozzle opening hanging past the layer beneath.
But there's a possible trick here. Cura, at least, has a setting for "outer wall offset", which causes it to inset the nozzle position a certain amount when printing the outer wall, while still extruding the correct amount for the actual wall profile. If your line width is set smaller than your nozzle size and outer walls are being printed after inner walls, it automatically sets this value to -(half the difference between line width and nozzle size), so that the outer edge of the nozzle travels along the outer edge of the wall, hoping to constrain the filament flow to the correct outer dimension. For example, with a 0.6 nozzle and 0.5 line width, it automatically sets this to -0.05.
But of course you can change it manually, so here's a thought. What if you printed with a 0.6 nozzle, 0.6 line width, but set this to -0.1, so that only 0.5 of the nozzle width actually extends over the new line (with 0.1 overlapping the inner wall-which must have already been printed, but this is preferable anyway when printing overhangs), and the same 0.2 extends past the previous layer as with the 0.4 nozzle.
I hope this makes sense.
I got a nice carbide 0.4mm so this is a nice trick. Arachne + 0.6mm extrusion looks great
I spent some time considering the geometry here.
I think it depends greatly on the amount of overhang and the exact widths and ratios involved, but I think in your example you are envisioning relatively high overhang and a higher extrusion width to nozzle orifice ratio than what you have.
I think in practice, extruding a larger width from the smaller orifice leads to a higher percentage of the extrusion exiting over the previous layer, and then being pushed outward to fill the width.
e.g. A 0.6mm nozzle/extrusion that shifts 0.1mm to the side each layer will have 0.1mm of the nozzle extruding over air, while a 0.4mm nozzle and 0.6mm extrusion with the same shift will be fully on top of the previous layer.
If the nozzle does find itself mostly over the air, this seems like it would make things worse as you expected, but that seems to have not happened in this case at least.
Matt prints mean your filament is underheated (also you're risking to hit the max flow limit of your system and start getting extruder skips), and the overhangs are better with lower temperature as we all know.
The nozzle size setting in prusaslicer (dunno about cura) will limit you in how high you can set your layer height to be (max layer height=nozzle size) and the recommended layer height is generally known to be from 1/4 to 3/4 of the nozzle size. With both 0.6 and 0.4 nozzles and 0.2 layer height you are still within the limits.
Thanks for the video!
Heard about the workshop situation on the podcast. Really sorry to hear about it :( hope the next place is more stable 😥
So I've messed with this type of thing for a while before arakni was around. Ive printed with .8 set in the slicer with a .4mm nozzle. The reason your prints Look matte is related to flow rate, pressure and temperature. In short the more plastic moving through the nozzle, the more heat its taking away from the system. Just like when trying to print faster speeds, the amount of plastic flowing through the nozzle is increased, and a higher temperature may be needed to avoid jams and such. As for the matte of the parts is essentially just due to lower temperature in the extruded plastic. Its important to be aware that were not measuring the temp of the plastic, but that of the heater block.
Well, Who would've thought... Tom, I've been printing mechanical parts on 0.4 nozzle with 0.6 profile ever since I watched this video. 0.6 profile on 0.4 nozzle, 0.24 layer hight (I use 8x2 leadscrews) and a few degrees temperature bump does a great job even on complex shapes. I switched to Prusa Slicer at the same time and I;ve done my Bear upgrade all the way. Thank you for this suggestion. I works, CONFIRMED.
I was having terrible prints for a while and I couldn't figure it out. then I figured out that I forgot to switch out the 0.8mm nozzle but was still using 0.6mm settings.
I also did the reverse where I was too lazy to swap out my 0.4mm nozzle and still run a 0.6mm profile fairly well on a coat hanger hook thing.
I ended up here because Cura was ignoring cirdcular holes 3 mm when set up with my 0.6 nozzle and at line width 0.6mm. So I changeded settings to line width 0.4 still using the 0.6 nozzle and the print was just fine and the holes perfect. I am going to experiment further!
At the end of the video, I think you answered your quandary as to why the overhang on the pirate hat didn't sag as you expected. The layer height of 0.2 with 0.6 extrusion on a 0.4 nozzle is probably not calculating as much of a lateral shift as you're expecting, plus it is producing a mass at the overhang that you are not noticing as sag but it's creating the bonus of support for the next layer and the illusion of smoothness between layers. I'm also thinking the temperature of the extruded 0.4 filament is cooling just that little bit faster at the core because there is less mass for the heat to dissipate through once it leaves the nozzle.
Is there a follow up to this exploration? Feels like we need a part 3 video.
Lol, i did the 'experiment' of slicing for .4 with a nozzle worn to .6 nozzle, i had all those bad bridges and missing extrusions that i could absolutely not tune out. When i figured iout the nozzle was worn by measuring it with numbered bits i stepped my slicer up to .6 and it printed fine until a new nozzle arrived.
I like .6 printing for speed and strength but it fails on some projects needing top surface detail. I got some .2 nozzles for doing precise small lettering for clay pottery stamps where even the .4 is too wide to get clean stamps.
Every time my Prusa Mk3 or Prusa mini starts a print, it squirts out a 1.2mm wide line.
I used this as inspiration to play around with wider extrusion widths.
I found that printing with a wider width on the first layer helped with adhesion.
You can set a wider than nozzle perimeter width and a external perimeter width that is around the nozzle diameter, and get strong high quality parts that take much less time to print.
Yep! I'm flabbergasted that he is just now starting to learn about extrusion widths...
I'm excitedly waiting for the day in the next couple of months when he completely retracts his positive claims about larger nozzles.
I've found that the only things that nozzle sizes actually change are the amount of pressure that the extruder needs to apply (and thus, the limit of extrusion speed gets marginally higher with wider nozzles), as well as the minimum width of line that the printer can print with a nice finish.
I have an 0.4 nozzle currently, and I've printed line widths from 0.4mm all the way up to 3mm (yes, I meant that) and layer thicknesses from 0.1 to 1.0 (again, yes. Vase mode with 3mm wide and 1mm tall layers at a volume limit of 13mm cubed/sec looks really cool. The printer moves very slowly).
When you think about the actual cross section of the line that the printer is making, the width of the opening makes very little to no actual difference, as long as the width of the line is greater than or equal to the orifice size. Honestly I blame slicers for conflating nozzle size and desired line width, in a bid to make things simpler for users.
Yes. Although it depends a bit on the filament. The viscosity at the target temperature can be quite different between different filaments, and thus the force necessary to squeeze it through a narrower hole. And, after all, a 0.6 mm hole is more than twice the size of a 0.4 mm hole. But yes, there's practically no reason for slicers to ask for nozzle size, because it changes nothing.
@@marcus3d yup, thats the first of the two differences I've found. I've printed 3mm x 1mm in both petg and pla, with very little difference. Once you get the maximum flowrate dialed into your slicer it takes care of the rest.
"super droopy loops" definitely my new term for bad overhangs!
I think what you observed are the benefits of printing a larger line width compared to the nozzle size. For example, a .4 nozzle works best at printing .5 or even .6 wide lines instead of the default width = nozzle size that most people use. From your settings it looks like for the .4 nozzle the width is already at .45 but I am not sure. I usually print using a .4 nozzle with a .5 line width, because for mechanical parts it’s much better to have an even number of lines each mm (this way a 2 or 3 mm wall prints perfectly, they are both multiples of .5 but not of .4). With the newer Arachne slicing this is probably an outdated approach, but I still get good prints and I stick to it.
It would be cool if you compared the print strength across the 3 different methods, I bet the .4 nozzle with .6 settings will give the best results.
Cheers
What I do is keep outers at 0.4mm and inners at 0.6 along with top/bottom and infill. It combines getting details where it needs (since you'll only ever have small details that you need thin lines on the outer wall), and more material at once for more speed. As a bonus you print faster overall for the same movement speed, or same time for slower moves. Keeps the printer quieter and in my case I can actually move fast enough to cap my hotend's flow, so printing as fast as I can.
Yep! I like to run a 0.3 nozzle and print with 0.8 inner perimeters. It works great. I'm currently testing a 0.2 nozzle alongside a new rapido hotend and I am getting about 15 cubic mm/s with it. Given that at a 0.4 I should be getting 50 or more cubic mm/s, I need to do more configuring.
Nozzle diameter should be treated as a limit for only four things, *minimum* line width/extrusion size, minimum layer height (smaller nozzle = smaller minimum), maximum layer height (usually you don't want to go above about 60% of your nozzle diameter) and finally, a multiplier on your hot-ends flow rate. Testing has shown that flow rates vary by about 10% per .1mm. That is additive so if your max flow rate at .4 is 10 cubic mm/s (a v6 can do 12-15) a .5 would be 11 and a .6 would be 12.1, .7 would be 13.31 and a .8 14.64 Meanwhile, a .3 would be about 8 and a .2 would be about 5.
Here is the thing, the *vast* majority of printers are printing at *up to* half of their max flow rate. That is why people see a big speed boost with bigger nozzles, when they increase the nozzle size, they *also* increase the line widths which means that it prints more filament faster. Not because the *hardware* is faster but because they, by happenstance, have gone to a larger nozzle and changed the right setting.
Also, don't be afraid to increase extrusion width to silly numbers. Most of the time as long as you don't go above your printers maximum flow rate, you can get *up to* the diameter of the filament with minimal problems. You *can* go larger if you want.
That's a good idea about keeping line width an equal devision of 1mm. I usually design parts with 0.8, 1.2mm walls for a similar reason.
Do walls need to be in whole mm? I just choose a number of perimeters, usually 3, and don't worry too much about exactly how wide that will be. And isn't there overlap of the lines?
@@dianasharpe9484 they don't if you have some infill, but if you only have wall lines (for example a 3mm wide wall) there are only wall lines. If you set the line width to 0.5 it will be exactly 6 lines, but if it's 0.4 it will be 7 lines and 0.2 will be filled with a "bad" movement, instead of placing perfect lines.
Of course this is a problem only if the part you are trying is thinner than 2x wall width. 10mm part with 4mm wall is not a problem, as there will always be some infill to compensate. But if the part is smaller than 2x wall width, there will only be wall lines printed and it's best to have a perfect numbers of lines (line width is a divisor of the feature width you are trying to print).
Of course using 0.4 nozzle on 0.6 profiles is going to work. It's basically using 0.4 profiles but with 150-160% extrusion width. Stefan from CNC kitchen already made a video about it and it showed that the prints would look fine up until 150% extrusion width or 0.6 mm with 0.4 mm nozzle.
Arachne may also play a role a bit in ensuring better quality since your 0.6 profile showed 0.65 extrusion width, which is the same as 162.5% for 0.4 nozzle.
I switched to a 0.6mm nozzle based on your last video and it was absolutely the right choice for me. I only ever print functional parts and they now print so much faster.
Did you even try printing them at 0.6 mm with the 0.4 mm nozzle?
@@marcus3d No, I had a 0.6mm nozzle that I bought in case fibre filled filaments clogged my 0.4mm but they never did, so I'd never used it. I swapped it and tried a few prints with great results.
Have you tried 0.8mm settings with the 0.6mm nozzle
@@paranoidpanzerpenguin5262 I've done 1.2mm with a 0.3 nozzle, it works great!
@@SirSpence99 wait what exactly does that do?
My guess? The overhangs aren't printed with multiple layers directly on top, they're printed with each layer offset. With the 0.6/0.4 combo, the sliced output is doing the same, but the offsets are calculated in such a way that the next layer actually adheres a little better. I can't explain it very well with text, but if you think about offset layer deposition, I think it will make sense.
Maybe think of it as "how much of the nozzles hole is printing into air"? A 0.4 nozzle has more of its area blocked by the already printed part if it follows 0.6-gcode. So there is less room for the filament to get pushed down and the molten filament has to travel more horizontal at overhangs.
I did several experiments while waiting for a larger nozzle to come in the mail a while back. I managed to get a .4 to print up to 3mm wide extrusion at .5 layer height. Granted this was vase mode and simple vertical cylinders. The print movement had to be slowed *way* down and the temperature set to the max that PLA can take, and even with all these factors, there were sometimes print failures after a few inches of height. But the strength of the parts were unreal. I now just use a 1mm nozzle with standard 1mm settings, but it was an interesting exploration of the limits of my old nozzle.
Yep! My understanding is that the max you can get from pretty much any nozzle before you have problems is 90% the width of the filament.. I haven't seen the tests done on 3mm filament. Fun tidbit, you *can* print 1.2mm widths at .1mm just fine and it is super strong as well.
If you have your flow rate maxed, layer height does *not* affect print time except for non-printing moves.
Keep in mind that most of the time, your nozzle size sets your minimum extrusions so you are mostly stuck with pretty low detail. Another thing to be aware of is that compared to a .4 nozzle, your 1mm is probably *only* getting a 160% flow rate boost (10% per .1mm, additively)... I doubt you are printing anywhere close to the max flowrate of your printer.
@@SirSpence99 Maxing out the flow rate was one of the best things about my experiments. The printer was a giant 400x400x500 printer from Alfawise. The print quality was generally bad from the weight of the bed and the height of the gantry, but with max flow rate combined with very slow movement, the surface quality of the parts was very nice and glassy. And you are right about the 1mm settings being nowhere near max flow rate. Flow rate is up a bit, but it is nice to leave plenty of headroom to get more reliable prints.
The 1mm nozzle at .1 layer height in vase mode is also capable of insane overhang angles! If I order another nozzle it will probably be a .6mm to do more detailed parts, but for purely functional parts, everyone should have a 1mm in their toolbox.
@@Panda_Gibs I entirely agree with everything *except* your last takeaway, all of what you said can be done with a 0.4 nozzle with no problems! I don't think an ender 3 for example *could* print at the speeds required to get anywhere close to the flow rate that a 1mm would give you,, they are already pushing it to be able to reach the flow-rate of a v6 with a 0.4 nozzle!
Unless you have something like a ratrig or corexy printer, you won't have a good enough gantry to take advantage of that 1mm flow rate.
Also, keep in mind, you only get a 50% boost by doing to a 0.8 from a 0.4, so the flow rate boost isn't *that* big of a deal, unless you want to print something massive, in which case you should go with a better hotend...
@@SirSpence99 the bigger nozzle is valuable vs over extruding a small nozzle in plenty of situations. Printing overhangs, for instance. The small nozzle forcing lots of material out onto an unevenly supported layer beneath it has a much higher risk of blowing the plastic in a chaotic blob off the side of the overhang, compared to a bigger nozzle that can extrude without relying on building a pool of material beneath it. And I don't say this as a thought experiment, this is simply my observation from my tests. Plus, printing .5 layer lines is amazing both for speed, but also the visual and textural enjoyment of the final part. Some people like chunky layer lines. I love 3d printing, and I love it when it is obvious that my parts came off a printer.
@@Panda_Gibs Fair on both points, I haven't personally experienced that first one as a problem but I can see that it would be an issue. I *have* found that you can do extreme overhangs with a small nozzle if you go with a large extrusion width (1mm+)
The second one is also an excellent point... Which makes me think 0.6 and 0.8 nozzles don't really make sense, get a small nozzle and get a very large nozzle, no value in between the two.
i knew about this phenomenon for ages- i always print 0.5 on 0.4 nozzle
I believe there is a correlation between the matter prints and the better overhangs as it was pointed out the matte surface is a sign of lower than optimal print temps and on the other side bad overhangs can be a sign of higher than optimal print temps
Just tried 0.4mm nozzle with 0.6mm extrusion width and as you noted - overhangs are far more better than with 0.6mm nozzle. I previously switched to 0.6mm nozzle for faster prints with arachne, but after this test I'm rolling back to 0.4mm nozzle with another profile in slicer (0.4mm nozzle 0.6mm extrusion width).
Imagine the possibilities if slicer would be capable changing extrusion width within the print - for smaller details/holes -> 0.4mm, stright lines/infill/no details needed -> 0.6+mm together with variable layer height (not for whole layer, but again thinner where needed only). And I know you can set different extrusion width for different feature (perimeter/infill/and so), but this is for the whole model.
Something like you paint supports - paint where you need more detailed features. It'd be awesome!
Now imagine instead of multi color prints you would print multi nozzle prints. Switch between 0.25mm nozzle and 0.4mm nozzle, you would get amazing details way faster.
@@sierraecho884 Yep already thought about multi-layer thickness within the model. There is variable-layer height already but I mean change layer heights within the layer itself. Like - not detailed regions of the model 0.3mm layer height, detailed regions 0.15mm. One thick layer for coarse regions and then step down and (at the same height) make 2x0.15mm for the detail region :) and combine this with multi nozzle diameter. Mind boggling.
@@JanVokas Not even getting that complex, you can simply print the inner walls and infills with 1mm and the outer wall with 0.4mm for example. You prints would be crazy fast and crazy strong, of course if you can reach the volumetric flow.
Your diagram for the 0.4 nozzle squirting right past the layer is backwards. The bigger the extrusion width, the farther back from the edge the nozzle will be positioned to achieve the same slope. The bigger nozzle preset has a bigger extrusion width.
That was quite helpful Tom, thanks for revisting the topic. I believe a 0.4 Volcano nozzle with 0.6 settngs would be the ideal, since we have more room to melt the filament to overcome the different bore size while extruding. Any thoughts?
It should indeed achieve the same as what would happen with a normal nozzle on 0.4 vs. 0.6mm line width, but with the Volcano's about double volumetric flow.
Volumetric flow itself is one very underestimated parameter from what I see, few youtubers talk about it unless going for extreme speeds.
A normal V6 nozzle does about 10mm³/s, a volcano about 20 so you can print about twice as fast. So if your printer was on the edge of underextrusion on whatever speed you use (say, 100mm/s print moves), at the same layer height with a volcano, going from 0.4mm lines to 0.6mm ones you'll only need to move around 130mm/s, which ain't straining the machine.
By the way if you wanna calculate the flow it's quite easy. The slicer considers the line as a rectangle. So you only multiply line width per layer height and print speed, all in same unit.
@@Kalvinjj This is why I *always* look at volumetric in my slicer, I can tell if my settings are getting me anywhere near my max print speed just by that screen.
I agree with many of the comments here… this seems more about extrusion widths and whether the nozzle can keep up with the ext width x layer height volume. The acid test would be to take a 0.4mm profile, just increase the ext width and then check overhang quality. Of course if that wasn’t it, then we have a few settings to change one at a time to see what makes the difference (or the dreaded complex combination)
Thumbnail game on point
The nozzle size only gives the lower limit for the line width, not the upper limit. The upper limit is the tip size which much bigger in comparison. The nozzle size influences the maximum flow rate and the maximum layer height. If you print at a smaller layer height at slower speed you could always reduce your nozzle size. Maybe the slicers consider this in the future and start to offer variable line width. Infill usually could use the minimum line width, and print that faster and lighter while the shell and solid parts could use wide lines.
So actually, it isn't slicers considering it in the future, that actually is what arachne does.
I find it endlessly ironic that *now* is when he chose to push this (mostly false) claim that printers need a larger nozzle to print faster... Given that arachne is out, (or about to be) that, more than anything is going to be the thing that makes smaller nozzles *more* desirable, especially since most people aren't printing anywhere near their max flow-rates anyways!
@@SirSpence99 Yes, this is a bit surprising. I'm not yet into how exactly Arachne is adjusting the line width to be smaller or wider than the normal extrusion width but it seems to do both. As long as you already have lines on both sides of the nozzle I don't see any problem with narrow extrusions that fill the gap. For most lines the prusa slicer seems to go back to the default line width. In the advanced tab we can already adjust the extrusion width for different kinds of lines, so we probably should play around a bit with that instead of just loading the profile for a different nozzle size.
@@siegfriedkettlitz6529 From what I can tell, arachne adjusts line width down to your nozzle size and up to your extrusion width... Sadly the slicer I'm using doesn't have a fully functional implementation of arachne yet so I can't say much more than that.
And yes, you are completely correct. Almost all of the improvements that he was surprised by can be achieved by bumping up the external perimeter extrusion width.
I would strongly caution changing nozzle diameter away from your actual nozzle diameter, slicers have a bunch of safety/sanity features built in around the nozzle width... Namely bridges. You don't want to bridge (usually) at anything other than your nozzle width (outside of a small range from it at least) even to the point that slicers won't even let you set bridging widths except as a toggle between a few options.
For reference, I can confidently say a 0.4 nozzle can happily print a 1.2mm perimeter at .1mm layer height. You may need to drop your speeds to do so but given that the biggest quality hit on most printers is that they are limited by gantry speeds... That isn't a problem if you are pushing out filament faster. Why would you print 3 .4 perimeters when you can print just one? The quality is almost always *way* better too.
A 45° angle on a 0.2mm layer height will have a 0.4mm overlap and 0.2mm overhang with a 0.6mm nozzle/0.6mm line width.
Or a 0.2mm overlap and 0.2mm overhang with a 0.4mm nozzle/0.4mm line width.
So a 0.4mm nozzle with a 0.6mm line width on a 45° angle will be extruding 0.1mm overhang and 0.3mm overlap with the 0.3mm overlap smooshing into a 0.4mm line, and a 0.1mm overhang turning into at least a 0.1mm overhang. (maybe plus a bit from the smooshed overlap pushing in all directions).
As the gcode assumes it went 0.2mm in the overhang the next layer may only have a 0.1mm nozzle-part overlap, (assuming the above mentioned smooshing effect doesn't occur), and a 0.2mm overhang, the same as a 0.6mm nozzle.
So I would assume what would make a 0.4 mm set to 0.6mm worse than a 0.6mm nozzle is higher layer thicknesses or more extreme overhangs?
I think it is likely that the smooshing effect will help as the plastic making contact with the part will cool faster, and face higher resistance to flow from the cooling and resistance against the part - therefore some of the overlap flow will go sideways and push out the overhang distance closer to a 0.6mm nozzle - but moving sideways, not downwards. So likely looking less droopy.
Edit:corrected some errors
You have your maths wrong. At 45 degrees, if you raise by 0.2mm you have to move out by 0.2mm no matter what the line width is. Otherwise you are not printing at 45 degrees. So with 0.4mm line width you have 0.2 overlap and 0.2 overhang. With 0.6mm line width you have 0.4 overlap and 0.2 overhang.
@@ghislaincarrier correct. I shouldn't do 1am maths in my head. Thanks for identifying my error.
I think this information provides even better logic for overhangs being fine.
I use prusaslicer for a custom printer. I found to get good bottom layers above supports I have to run my support layers at 1.1mm width, on a .4 nozzle. This leaves the undersides looking like they do with Simplify3d.
I think the detriments you see with sticking to a .4mm nozzle over a .6 is limitations on layer height, layer width and probably how much work it is to squeeze the filament through the nozzle orifice. With a .6mm nozzle you could do layer heights up to ~.45mm high (to my knowledge just won't work with a .4mm nozzle). With a .6mm nozzle I think it's reasonable to get 1-1.2mm extrusion width and still have decent parts, I'd guess .4mm nozzles are limited to less than .8mm line widths. I made some parts for work and when I did the test parts that were only 1/3 of the final part tall (to test fits and mechanism) it took me over 6 hours with a .4mm nozzle and .4mm line width and .3mm layer height. When I switched to a .8mm nozzle and did 1mm line width and .6mm layer height I got the full parts (3x the height of the test part) in less than 6 hours even with having to print at a slower speed (~30mm/s rather than ~70mm/s). Of course these notes are somewhat apples and oranges because I didn't need the resolution of a .4mm nozzle and didn't need to not see the layer lines (the parts are functional, beauty is optional).
Also with .6mm orifice having more than twice the area of the .4 would theoretically mean that for the same viscosity of filament (assuming it's melted to an equally viscous state whether it's a .4 or .6mm nozzle) you should have a lot less back pressure when trying to push the filament out. Going 1.75mm filament to .4mm extrusion is a ~95% reduction in area, going 1.75mm--> .6mm is only ~88% reduction in area. Not sure if it's the right way to think about things, but I often try to gauge trends off of what the absolute extremes would be. If you didn't cut a hole in the nozzle no filament would come out. If the hole in the nozzle was 2mm+ you wouldn't have to melt the filament to get it to come out (obviously wouldn't fuse or print this way, but thinking about back pressure). With grip on filament and extruder stepper strength often being limiting factors (or at least the part that "gives up") it seems like back pressure would play a significant role in extrusion speed.
I've done this experiment and with 0.4mm nozzle have been able to print extrusions up to 1.5 mm wide by 0.5mm thick with good print quality. Usable layer thickness seems to be dependent on extrusion width, and max extrusion width seems to be based on the diameter of the flat tip of the nozzle. The resulting quality is also very dependent on your ability to cool that thick extrusion before it can sag or before the next nozzle pass comes back around and raises the material back above the temperature where it wants to flow or sag.
In short I think the extrusion size is limited by the exterior width of the nozzle tip moreso than the diameter of the hole in the nozzle. Within limits of course, I don't think a 0.1 mm nozzle open would be practical.
In the end it all depends on volumetric flow rate. Back pressure decreases the hotter the filament and the larger the bore, so for a smaller bore you need more temp to decrease pressure. As long as the pressure is not excessive I don´t see a problem with it. The pro would be you can rather fast decrease pressure by applying retractions which should result in less blobs and strings with the 0.4mm nozzle compared to the 0.6mm nozzle.
I would like to see a test with a high speed printer like a Bambu lab machine. My Idea would be to use a 0.25mm Nozzle instead. You get great quality with up to min. 0.3mm layer widths and max of sax 0.6mm layer widths for speed.
here is what i ran into. im using Cura i am also a big time "cheap charlie" i buy the cheapest filament i can find and as expected sometimes my prints were all over the place in regards to quality i run Anet A8's.
i run the Z axis up to around 50.8mm, run some filament thru making it coil up nicely on the bed then i check the dia of the filament that came out of the extruder. what i found was they vary a lot i have some that come out at 0.70mm (strange off brand brown box) down to 0.55mm (creality CR) all out of the same 0.40mm nozzle.
i figured this had to affect my overlap so i started playing with widths and layer heights to see what would happen. what i discovered (as of right now) is i check the dia coming out of the extruder then subtract 0.20mm off the layer height and 0.10mm off the width up to 0.67mm (after that my 0.20 and 0.10 fall apart). wow what a difference in my prints. what im printing doesnt have anything fine on it (sections of track for a marble run). you get a chance take a crack at trying that.
Hi Tom. I have a theory. The Bernuli Effect. Sorry my spelling sucks big time. It is like having a hosepipe, the water flows through it. If you squeeze the opening close, the water flow will now jet out [how a power washer works], but when you release the opening, the water flow returns to normal. This is how we is rocket science accelerate gasses to produce thrust to fly rockets. A nozzle is also known as a Ventury. This might explain why the 0.4mm nozzle managed to keep up with the 0.6mm flow rate.
0.4mm nozzles printing extra wide extrusions are great you just need to increase nozzle temperature and/or extrusion multiplier so you don't get under extrusion. The overhangs are better because the nozzle is further from the edge and relies on squeezing out a wider extrusion so more of the perimeter is overlapping the one below. All this while you retain ability to print thinner infill lines and save plastic where the infill is mainly just a top layer support. And you can still print fine detail.
1. Do not need to increase extrusion multiplier, the slicer calculates that automatically you need to increase temp for higher volume flow.
2. The opening of the nozzle is not closer to the edge it is at the same position for 0.6mm nozzle with 0.6mm lines and 0.4mm nozzle with 0.6mm lines.
This is not true, actually try it mate, print a 1mm extrusion width with a 0.4mm nozzle at default extrusion multiplier and temperature and see the gaps between extrusions. If you actually have done this you wouldn't be giving incorrect advice.
@@EK1H Oh I do, you seem not to understand how extrusion multiplier works.
Extrusion multiplier calculates the volume needed for give layer height and line width. So if your multiplier is at 100% for say 0.2mm layer height and 0.4mm layer width for the 0.4mm nozzle it will also be 100% for any other layer heigth and width.
The temp might need bumping up a little but only if you struggle with volumetric flow.
The nozzle just differs in it´s opening size but the amount of plastic being extruded does not differ. You basically extrude the same amount of plastic through a smaller or bigger opening that´s it. Only the pressure increases or decreases.
@@EK1H Your problem is probably underextrusion from the beginning which shows more if you increase layer with. Let me guess, you have done the "single wall" calibration of whatever it was recommended on youtube and you dialed down your EM to 95% or so instead of using XY compensation.
@@sierraecho884 you have to slightly under extruder because no filament has perfect diameter and slight under extrusion is better than over extrusion and nozzle blobs. If you go from printing 0.45mm extrusions with a 0.4mm nozzle and then bump it up to 1mm extrusion, that is almost doubling the output volume, you either have to slow down, increase temperature or increase extrusion multiplier. My method of testing is to inspect 5 perimeter commercial product I print using a 20 loupe to check xy binding between extrusions.
what makes me come here is that I did that method by mistake and came out pretty good and faster print, I was wondering if any other users using that method nozzle .4 with .6 profile, thanks for your video
This brings up an Idea:
What about printing with a 0.6mm Nozzle and sliced gcode for a 0.8mm Nozzle.
Will that eventually give even more print speed while eliminating the negative overhang effects of the 0.6mm Nozzle?
Would be worth a try i think.
I have noticed that Arachne is skipping details on the Tabletop terrain ( a house ) , even though classic renders and prints them just fine.
This with the first Alpha version and its still in the final Prusa slicer version.
So its not yet the fire and forget setting I've gotten used to with the Prusa slicer.
It took me way too long to realize the thumbnail was that butterfly meme template. Great video as always Tom. Do you think it'll work on 0.25mm nozzles too?
I've never used Prusa Slicer, but in Cura there isn't actually a nozzle diameter setting at all. There's just a Line Width setting (actually, several of them - you can set Inner/Outer walls, Infill, Top/Bottom line widths independently if you want to). So if you want a 0.6mm line width, you just set the line width to 0.6mm and it doesn't care whether you're using a 0.6mm nozzle or a 0.4mm nozzle. I regularly use a 0.4mm nozzle with a 0.8mm line width, it works very well for certain models and saves a lot of time. As you note, the Arachne engine narrows the line width when needed, and the minimum line width can be specified in Cura (the Minimum Wall Line Width and Minimum Thin Wall Line Width settings). I contend that a 0.4mm nozzle would be able to reliably print thinner minimum line widths than a 0.6mm nozzle, so if you use a 0.4mm nozzle you can set the minimum line width smaller than you could if you were using a 0.6mm nozzle. The strength issues of using a wider line width than the nozzle was addressed by Stefan (CNC Kitchen) - he found it produced stronger parts. I suppose I'd use a 0.6mm nozzle if I wanted to print 1.0 or 1.2mm lines. And yes, I think you'd have to bump temperatures up.. I've got a Volcano nozzle on my printer so I should be able to do this.
You can´t print bigger layer heights with a 0.4mm nozzle though. But in most cases one would be volume flow restricted anyways.
@@sierraecho884 True.. I almost always use 0.2mm layer heights, but line WIDTHS of up to twice the nozzle diameter work fine. Incidentally knock-off CHT type nozzles are now widely available at very reasonable prices if you find you need more flow.
@@iskandartaib Yes, so do I. I just wanted to mention that one of the little advantages of a 0.6mm nozzle is you can increase layer heights.
The e85 drag racing reference… I knew we could be friends 😎
For best performance i set 10% wider nozzle than actual for best speed/quality performance. It's better to use 0.6 nozzle with 0.66 settings, 0.8 wide on fist layer for super reliable adhesion. If you set it wider, you need to slow down print speed, you can hit pretty quickly the volumetric limit of the nozzle.
As for your overhang conundrum, do you have the setting enabled for printing internal line walls before external?
Hi @Thomas Sanladerer You're using the Beige PLA from PL, did you get that from Prusa? It was one of the FIRST spools I ever bought and I ended up odering another 3 because I really loved them. It prints very matte and unlike any other PLA I've used. Most matte PLAs look matte on the spool where as that fillament is shinny. I regocnised it right away.
That would be why your prints are quite matte with that fillament.
That's the exact filament I used! To me, it looks like regular shiny PLA with the 0.4 and 0.6, but becomes significantly more matte with the 0.4 and 0.6 gcode.
Still chuckling 2 months later to the best thumbnail on youtube... =)
Theory on the overlap, when I'm pushing the limits of extrusion speed the filament curls out of the nozzle, maybe the big gcode on the small nozzle is doing something similar with the higher pressures and giving itself something to stack on
Hi Thomas, thanks for such a great channel. I just have one question: using a 0.4 nozzle with 0.6 settings (layer height, etc), the nozzle size on the Slicer should be 0.4 or 0.6?
About the overhangs. Lets say you have a 45° overhang and 0.2mm layer height. The horizontal advance to the next layer will be 0.2mm. At 0.6mm line width, 0.4mm will connect to the previous layer, and 0.3mm of that will be directly under the orifice of a 0.4mm nozzle, with only 0.1mm sticking out over nothing (1/4 of the nozzle orifice). A 0.6mm nozzle will connect 0.4mm directly, but 0.2mm of the orifice will stick out over air (1/3rd of the nozzle orifice). 1/4 < 1/3, therefore, overhangs are better.
I also tought this is the single most important factor for the better overhangs with wider extrusions, but it works this way only for the first layer in the overhang section. The smaller nozzle causes the dimensions to be 0.1mm smaller, because the material cannot expand horizontally when extruded mid-air. Because of this, the following layer will again have 0.2mm horizontal shift from the previous one.
A look with a nozzle cam would for sure be helpful on figuring out what is happening.
@@B3M1C I think it does expand as much horizontally as it expands vertically (down). But the flow constriction is less symetric with a 0.6mm nozzle, so the line geometry will be more skewed.
I've been printing with anywhere between 0.5 and 0.6mm extrusion width on my 0.4 nozzle for mechanical parts. It works perfectly fine even on 0.3mm layers, however the printer is immediately flow rate limited on 0.3*0.6. A larger bore nozzle has larger surface area to transfer hear through so I think that's the difference. Honestly I think of you're changing nozzles, a CHT 0.4 might be the way to go with Arachne.
I'm of the view that you should almost always be printing at about 80% of either your gantries max speed or your flow rate.
Larger bore only gives you a 10% improvement per .1mm above .4, that means if you double the bore to .8, you get about a 50% max flow rate improvement. Most people aren't even getting close to their max flow rate normally anyways.
You bring up an excellent point with the CHT nozzle, another alternative is a volcano nozzle or some other "high flow" nozzle. IIRC, going to a CHT 0.4 from a regular 0.4 gives you over a 100% increase in flow rate, volcano up to 300%.
The bigger thing that I've noticed is that the nozzle diameter sets the minimum extrusion width which, *especially* with arachne is a big deal.
Thank you for this nice informative video, but does 0.6 mm settings on a 0.4 mm nozzle increase the back pressure and therefor the oozing especially with a PETG filament ?
Quick question: I'm on a Mk3s+ on prusa slicer and upgraded to 0.5mm a while ago. The problem is that I ended up with giant extrusion gaps (about half a cm) where my seams were. I tinkered around and figured out that over 5x the stock detraction speed -mostly- fixes it.
Did you have a similar problem when increasing your size in prusa slicer? If so, is there a better fix?
I know you mentioned it but I think the answer is to use a 0.4mm nozzle and double the extrusion width. No reason to tell the slicer it is a 0.6mm nozzle.
In the end, the best reason to use a 0.6mm nozzle is to reduce clogging with carbon fiber filaments.