For my (very may be clone e3d), i just pull up ''on cold'' 0.6mm nozzle fillamen without problems, for now its seems normal in 0.01+/- mitotoyo caliper
The spiral pattern is achieved by plunging a dull bit into the brass with a lot of force. They make a tool to cut out the inside profile then use it until it breaks. Probably the first 1,000 nozzle's bore is totally straight, by nozzle 2,000 though you have the spiral pattern and somewhere after that the bit breaks and they make a new one.
It's not a great application for a form drill because there is a high ratio of diameter for the tip diameter and the body diameter of the nozzle and also because the tip diameter is so small. For a 1.75mm filament bore and 0.4mm outlet dia, one would have worse than a 4x ratio between the smallest diameter and the largest diameter on the form drill. This means that you'd really like to be drilling that tiny 0.4mm diameter a much larger RPM than you can run the 1.75mm diameter at because you'll exceed cutting surface speed limits on the larger diameter. This means that you have to plod ahead at 1/4 of your best drilling rate for that tiny 0.4mm end which is going to be doing a terribly ugly drilling job for the entire length of the nozzle because it's stuck to the end of the 1.75mm body dia. It would be far faster to blitz the back end of the hole with the 1.75mm dia for nearly the whole nozzle length, then go in with a 0.4mm to peck out the last bit of length at the end. Using a second tool position to have separate 1.75mm and 0.4mm diameters will be far faster with better quality drilling and much lower breakage of tools, not to mention that standard jobber drills are way cheaper than making a form drill. I think the spiral pattern is being caused by a cheap jobber drill being pushed too far past their sharpening regimen. There's no really good excuse because clean holes in brass with jobber drills is very easy with CNC. The level of tool wear management necessary really isn't high.
I imagine that drill runout will also cause a spiral. An extremely true and concentric chuck is needed for highly accurate drilling at such small diameters.
Freaking brilliant video! Very impressed by the quality of production and the solid engineering content. The undulating helical pattern is a likely result of a worn or chipped drill bit and pushing the feedrates on the lathe too hard. Also if you don't have your drill running dead centre it will walk and cause run-out. Glad to see ours were straight and shiny as they should be, but agree that it's not really a contributor to stringing. You're dead right that the most important part of the nozzle is the orifice to tip flat transition. We've put a lot of time into ensuring that corner is 'just so'. The internal cone angle is also a surprisingly large contributor to the ooze and clean sharp extrusion, it gets you the more laminar flow you postulate and lowers pressure needs. Combine those two and you get the rather dramatic difference you saw in stringing. I suspect but cannot prove that the excessive chamfer on the clone and worn nozzles also serves as a place where material will cling to and accumulate. This little annular 'pool' of very liquid material may serve as a source of material that will be entrained into a string. Retraction is going to be completely ineffective at controlling any buildup on this part of the nozzle.
I was going to say the spiral was from a blunt drill but it has been said in detail above so I'll shut up now. Thanks very much for the great videos @cnckitchen
I just want to add ontop of your fine explaination about the helical pattern. -This is a great example why GPS tolerances (Geometric Product Specification) was invented, and specifically why it's important to have them specified when you make large orders produced out of China and the like. Not that it isn't necessary in all other countries though. :-)
Excessive feed rate, lack of pecking and worn tooling is the most likely causes, but materials like brass also have issues caused by their high thermal expansion. Materials do not heat evenly when drilling and tends to build up heat near the corners of the drill flutes during the machining process. Thus the materiel near the corners of the drill flutes is removed at a higher temperature so those areas expand more then the other areas of the cut. This leaves a smooth hole at the time of operation, but then when the materiel cools and constricts it leaves a helix or spiral shape that matches the drill feed rate. Brass and plastics it tends to become fairly prominent due to their high thermal expansion rates and high drill feed rates.
@CNC Kitchen, that's my blownout nozzle at 10:34! I got that after printing approximately 1 spool of inland glow in the dark PLA. Thank you for showcasing it in this video, it was a hard lesson learned for a new 3d printing user.
If the drill vibrates, it usually makes cones. Like an hourglass shape. The walls of hole look like melted, probably drill was overheated during the production and left that weird pattern?
25 year machinist .. that pattern happens when the drill geometry is shoddy, specifically the lead on the spiral or the point being off center slightly and your feed is way to high. Brass has a tendency to grab when using sharp tooling. A way to mitigate the grabbing is to flatten the cutting edge slightly on the perpendicular axis of the cut. This is probably the reason for the over powered feed so as to compensate for that flattened cutting edge.
@@swould333 I can confirm that, worked a lot with brass in the past, it likes short stubby cutting edges and the unevenly sharpened chinese 1.75mm drill bit is far from that. High speed vibrations + eccentric forces combined with resonance from the brass grabbing and you get forms you could not imagine can exist!
@@swould333 curious if the good nozzles are cut in three procedures and that's why they have good surface finish and geometry that clearly wasn't taken off a shotty twist drill.
The wobbly pattern of the passage through the cheap nozzle is from a warped drill bit. They used a high rotation speed (probably with inadequate cooling fluid) and overheated the bit several times and they were using a high rate f driving the bit into the brass. Its an effect I got when learning to use a home drill press.
Some more insights as to how the wobble might have occurred: The geometry of nozzles lends well to lathes (and by extension screw machines), where wobble of both the part and the drill can be more severe vs with a mill. It's also likely that the stock used to make the nozzles is at the max diameter of the part to minimize material waste, and that thinner material would also wobble more. As for feeds and speeds, the high machinability of brass allows very aggressive settings without significantly shortening tool life (and that's great for minimizing labor cost, but bad for part quality).
Also I bet the is not exactly round, but slightly elliptical due to the same reasons. I’ve learned the hard way as a home gamer that drilled holes are always bigger and never round.
Hi pal!. I just bought my first printer and I'm soaking up information with a sponge (even though I could be your grandpa because of my age... I never tire of learning!), this report opened my eyes to many doubts I had and I think a spectacular job on your part, very serious and dedicated. This is the first video of yours I've seen and I've subscribed to the channel to see the rest. A pleasure to learn from people who really care about finding the important point of the problems. My congratulations, today, December 31, 2022, I send you a big greeting from the South of Argentina and I wish you a good year!
The cross-section of the genuine E3D nozzle really sold me on their products. Such amazing quality assurance; and it really shows why it's smart to dole out the little extra cash for it.
I asked 3dpn a million times to make a video about this one topic . You really did a VERY good job putting as much thought into the tech as well as the science.
I concur with your findings, it was interesting to see the 60 degree chamfer on the inside of the e3d nozzle verses the 90 of the in expensive nozzle. Custom tooling verses standard drill bits, that is significant.
Both angles point to an custom drill bit. (Which arn't expensive when you mass produce.) The default angle on a HSS drill bit is 118°. But there can be differences for hard and soft maqterial from 110° to 130°. Carbide drills normally use 140° tip angle.
I found the section on glow in the dark filament really interesting! I had been printing a very large gift for my fiance and the nickel plated nozzle that I had been using started to show signs of damage. The filament would shoot off to the back and left no matter what filament I used. Just bought myself some steel nozzles because of it! Thank you for the great videos and keep it up!
Useful info, thanks. I have a small lathe, tried making my own nozzles, they worked ok, but no better than the bought nozzles. The hand-made nozzles took a lot of time to make, and the materials are expensive. A good compromise is to buy 0.2 or 0.3mm nozzles, and drill them out to the required size. I find 0.35mm is a good size for the kind of printing I mostly do. Carbide PCB drills are cheap, but it's easy to break them. I start by facing off a piece of scrap stock, drill and tap M6, and screw a nozzle on. I do this for each batch of nozzles to maintain concentricity. The drilling part is easy, as long as your tailstock is aligned correctly.
The chamfer on brand new (cheap) nozzles effectively increasing their diameter and hurting print quality is a very interesting discovery. Thanks Stefan, glad we have people like you in the community 👍
I think it’s worth noting though that those cheap nozzles are made by a very wide variety of manufacturers, and it may just be that the particular nozzles he tested were especially bad... I have only cheap Chinese nozzles, a big variety box of standard CR10 style nozzles, as well as one V6 style I was sent by mistake, and about 6 volcano style I bought for my Sidewinder X1. I just double checked all of the volcano and the v6 nozzle as well as a random selection of the CR10 nozzles and not a single one had that chamfer. I even double checked under a microscope in case I just couldn’t see it with my naked eye. I’m sure the internal bore of the nozzles is not as clean as genuine E3D nozzles, and there may be a difference in the internal angle, but at least it doesn’t appear that the no name brand nozzles all have that chamfer or that it’s even a common defect. He may have just received a defective batch or bought from a particularly bad manufacturer or I’m just extremely lucky... He’s 100% correct that you’re taking a gamble not buying genuine, but given that the cost of a single E3D volcano nozzle is about the same all of my Chinese volcano nozzles combined, it may be worth the risk as you won’t lose much money if they turn out bad. Though of course I do appreciate supporting the original creators. Also for example there is a massive difference in quality between the heat breaks especially on knock off E3D extruders vs the real ones that I think makes them an obviously better buy even at the in increased cost (my sidewinder for example has a PTFE tuve all the way through the heartbreak which is an issue for high temp printing, so I’m planning to upgrade to an official E3D all metal titanium head break which is actually about the same price as some of their nozzles somehow even though it’s a more complex part). Also it’s worth considering that I can buy the Chinese nozzles easily here in Colombia with local 1 day shipping, whereas to buy genuine E3D nozzles I would need to order from the US and import them, increasing the already higher cost due to reshipping and import taxes as well as significantly increasing the wait time to receive them if I need one quickly. I’m still planning to buy a few genuine nozzles to use as a baseline comparison against my cheaper nozzles to see if I get increased print quality, but I don’t think it’s fair to test a single brand of Chinese nozzle and then imply that all Chinese nozzles are manufactured that way or have similar problems. Many of them are perfectly fine from what I can tell (including from my own print quality which is quite good using only these nozzles). Anyway just my 2 cents.
Hi Stefan, I'm using probably the same ruby nozzle that you have in your ruby collection. It has a chromed copper body. I use it with a titanium heat brake (all metal) and a home-designed & milled copper heater block. The back pressure in the nozzle is greater (especially with CF-PEG) so I use a dual-drive geared extruder. With a bowden cable, in a Anycubic kossel linear plus. It took me a while to get good results again, but now it's spot on. I play around with linear-advance a bit and with normal PLA it's better not to use it (K-factor zero). Hope you're having fun with it and happy printing!
In a previous life I was a tool and cutter grinder in a mould shop. One tool I had to design and make was a D-drill for plastic injection nozzles. The interior bore ended in a 40° included angle hole. I assume this allowed for faster flow in the injection process. Materials for the nozzles were either beryllium copper or a tool steel. the nozzles also had a variety of exit holes, some on the ends and others at angles coming out on the sides of the tip.
James Crombie that's awesome! I thought that angle was really interesting. I'm guessing E3D have tried all sorts of angles to get one that works best, but I'd be interested in results from even more acute angles.
Stefan, Job well done, as always. I print mainly with flexibles and CF nylon on the same Prusa i3 MK2s. Yes, two completely different materials, each with their own challenges. I was having major extrusion issues and jams. I replaced the E3D hardened nozzle that I had run a few kilos of CF nylon through with a new E3D Nozzle X with no real change. I then looked at the heat break tube and saw some markings on the inside of it, so that got replaced. This improved my results a fair bit. I still had problems. I noticed inconsistent layer adhesion on the small parts that I print using Ninjateks Armadillo. There is a post that is 4mm diameter that when everything is right will not snap off. I print about twenty of the parts at a time and have maybe 4 - 5 that fail. I then looked at the Hobbed Pulley and noticed some wear. I didn't realize how much until I removed it and compared it with a new one. (See the photo in the attached link) There are many aspects that can cause extrusion issues. If you use the "Cause and Effect" process the best method for solving the problem is to go back the where all was good. photos.app.goo.gl/SDJUQd27NxsFrfXd9 photos.app.goo.gl/fLq8RsCXCE3vuNza7
Wow! Ones again awesome tech analytical content. As a mechanical engineer I really appreciate your channel. I will never print a cheap nozzle from now on!
There are so many interesting points in this video. It is really good to see the actual differences between the parts of different manufacturers. I've heard many TH-camrs talk about the poor quality of cheap Chinese parts, but you actually show the difference. It would be interesting to see the real world print quality difference between these, and examine why they may or may not be worth the extra cost. I would think that the quality of the print may also affect the overall structural quality of what you print as well. Basically how much does 3D-Printer part quality affect the output, and how much does it cost to mitigate the loss in quality.
In my opinion, the E3D is superior in terms of stringing because of the 60deg angle of the convergent channel (instead of 90) as it will create less flow resistance, thus lowering the pressure buildup in the melt chamber. With a lower pressure accumulated in the chamber there is less viscoelastic flow (ooze) when you stop extrusing. Also the sharp transition can help too as it will cut the strings and avoid pulling out filament. Great video as always, and excellent scientific approach to real world problems. Ps: funny enough I just noticed that Sanjay just made a similar comment.. Seems that physics is the same for all of us ;-)
Very nice tests. I have some glow-in-the-dark PLA from Velleman and it is definitely abrasive, I also noticed the nozzles being ‘sanded’ at the bottom after printing a lot of this material. The hole diameter seems to have slightly increased as well, although the only way I noticed this is that it gradually became easier to push a .4mm guitar string through the hole. A larger sanded-off nozzle bottom does have an advantage in that it makes it easier to get nice smooth top surfaces, but a problem is that it dumps more heat into nearby already printed material, which makes for worse overhangs and also it has more of a tendency to drag along PETG because it is so sticky.
Exactly. The biggest problem of abrasive filaments appears not inside(hole diameter) but outside(hole length). From my experiences, the most abrasive additive agent is "glass fiber", and not carbon fiber or metal particles. Actually it can shorten stainless steel nozzles over 0.5mm within one reel(500g). This means the nozzle diameter changes from 0.4mm to 2.0mm, or simply the nozzle has died out. If diameter of the nozzle was changed at this moment, it is not so big of a deal. I tried to use HSS_Co(used for cutting tool of stainless steel) nozzle for the glass fiber filament, but it does not work. So currently I am using stainless steel nozzle once for small parts, and flat top HSS_Co nozzle for large parts.
Klasse Video, sehr informativ. Großartig was du dir für Zeit für die ganzen Tests genommen hast! Und was mich am meisten freut ist, dass dein Kanal sehr, sehr schnell so gewachsen ist :D Sauber! Grüße, Sebastian
I think that you have definitely proven that it is worth the extra money for the quality nozzles from more reputable companies like e3d. Just the cutaway views of brand new nozzles is enough to convince me of the difference. I will be ordering some soon. Thanks for your usual in-depth research on this issue!
CNC Kitchen was mentioned on the last podcast of "3D Printing Today". They were impressed. High praise as these guys have been in the industry before it was consumer availible product.
I have actually researched this quite a bit and read about similar types of methods of testing. I think most nozzles lack precision and I have actually decided to manufacture my own and am currently building a machine to do just that and want to eliminate or minimize wear as much a possible without sacrificing precision at all. I don't believe calipers are precise enough for the tuning/calibration of a 3D printer. I have access to a fully equipped machine shop (though I understand not everyone does) and use gage blocks, micrometers, and microscopes to measure everything. I like your videos, I'll send you one of my printers when I'm finished with it.
I don't know if you realize how much some of your videos, especially ones like this one, add to the 3d printing world. In this case, we pretty much lived with "those filaments cause the nozzle to wear" and I'd bet that we all assumed that meant that the hole in the nozzle was getting bigger... instead the face is wearing off it... huh. That has totally different implications WRT to dealing with the wear. I've already been modifying the cheap Chinese nozzles for better print quality and reliability. Now you have me considering changing the procedure somewhat, I'll likely custom grind a drill bit to have a 60* tip on it (I knew about the spiral finish on the inside, it was causing my printer to jamb up once in a while so I run a drill bit through new ones to clean that up) and I'm thinking of trying polishing the points flat on them. Keep up the good work!
Sanjay Mortimer 's answer on the helical pattern seems unlikely. The consistent helical pattern is more likely to come from a drill that is mounted on a bent lead-screw. This explains the consistency much better.
I have personally seen that helical pattern when printing on my makerbot cupcake with bent lead-screws. I immediately assumed that was the cause when I saw the cross section.
Congrats for the professional video! Although 3 years has gone, this looks the right place to ask my question. There are hardened steel nozzles available today (e.g. for my Bambu Lab P1P as a spare part) these are intended for abrasive filaments, like carbon fiber. These nozzles cost about the same as their normal steel ones (their standard nozzle is steel, they have no brass nozzles) so my idea is to use the hardened steel version for all filament materials (PLA etc.) What could be the disadvantages of a hardened steel nozzle? There might be a difference between the thermal conductivity of brass vs steel (which may somehow affect print quality, maybe the max. print speed?) but I think there is no significant difference between normal vs hardened steel in terms of thermal conductivity. I can think of one disadvantage of hardened steel is that it is more brittle, but that plays no role here. What do you think?
I’m really interested in a video testing the possibility of improving the print quality using a smaller nozzle. Cool video! I always learn a lot from you. ❤
I'm looking forward seeing you machine a tungsten nozzle 😆 Great video as always! I too have a brass nozzle and the wear was primarily on one side, the filament free extruded to one side too. I used a steel tweezers to clean the nozzle regularly, which probably did some damage too. The surface finish with a new steel with zinc coated e3d nozzle is significantly better. A new brass nozzle was going to be better too of course, compared to the damaged one. I'll try to machine the nozzle too, would be interesting.
I'm glad that you tested with clone nozzles. For as much as I might like E3D parts and prefer them, if I need a pack of .4's (sometimes you just know you'll be wrecking nozzles with wood), or a couple in each diameter between .2 and .6, it's usually the difference between $10 and $50. They aren't as good, but they'll usually get the job done, and it's good to know how long they'll handle that job. So thank you.
As an engineer I always enjoy your vids. Your scientific approach and attention to detail are definitely value added. I would be curious to see if removing material from the face of the tip of the cheap nozzle, until the chamfer was removed, would improve the print quality. Also, I noticed that the surface finish of all the nozzles was bit rough. Would polishing the nozzle face help the print quality?
Brilliant. This video concludes everything you need to know about nozzles for 3d printing. I learned that brand / quality matters and that you must choose nozzle according to materials used. Thanks for the research! Great work
now i kinda want to see the new cloned Ruby (FYSETC also do a white sapphire nozzle [Sapphire is available in all colours asided frfom red, though red corundum is just a ruby so very close] that'd be fun)
Great subject to point out ! I believe that if you used an extremely strong tool (such as adjustable pliers) to tighten the nozzle it can easly deforms the inside of the nozzle and will only have a slight affect on the outside of the nozzle. I would suggest to check it after tighting (with a strong tool) a brand new nozzle.
I use MG Chemicals Super Glow ABS almost exclusively for my "White" prints and have run over 6 rolls through my machine and have only changed nozzles like once and that was just to rule out the nozzle for another issue, so I'm glad you were underwhelmed with the glow results. That matches my experience. Half the stuff in my house glows now ;)
The odd spiral pattern is most likely caused by the stock being off-center when drilled. This causes the drill to wander, due to flexing, while drilling. I have seen similar behavior when milling a circle, using a bad collet that mis-aligns the tool. It can actually create a hole that is hourglass shaped in cross-section!
Another very interesting video Stefan. I don't have a 3d printer at the moment because I can't open that can of worms at this point due to other priorities. However, I do plan to get one sometime next year. I still watch your videos because they answer many of the questions I ponder or answer questions i didn't even think about. I think when I do get a printer your videos will already have helped me better understand this process. One thing I do wonder about you touched on in this video which is nozzle diameter - because I assume it is related to wall thickness. Some of the things I am considering printing will require thin wall thicknesses. I want to make fuselage type structures as lightly as possible - with and without bracing/infill. So the questions i ask are - How small a nozzle diameter can you use? How thin can a cylinder's wall thickness be? What are the limits? I wonder if you can do a video which pushes the boundaries in that direction. Many thanks.
Hey! You skipped MY blown-out nozzle on the Reddit thread! What gives?! No, that's fine. I'm embarrassed about it anyway. Great video, thanks for such a well-done breakdown on the effects of abrasive filaments!
at 12:33 you note the point angle of the E3D is 60 degrees while the cheap knockoff uses one of 90 degrees. I originally tried an Olsson Ruby nozzle to solve my wear problems, and during a spontaneous unscheduled disassembly of said nozzle I discovered the Ruby hides a very dirty secret: it has a point angle of 180 degrees internally which is very poor in general, but particularly for feeding fiber filaments. Something to be aware of. :) I've since switched over to a tungsten carbide nozzle and am very happy with it. Much more abrasion resistant than the hardened steel nozzles I've used.
This is how the ruby nozzle looks from the inside: twitter.com/erikcederb/status/1125056521199673344 There is a point angle but then, there is a small, flat section. Interestingly the standard Ultimaker nozzles look very similar on the inside.
Mein Drucker hat für sehr lange so schlecht gedruckt, dass ich überhaupt nichts drucken könnte und ich wusste nicht, was damit passiert ist. Ich habe endlich erkannt, dass meine Düse von 0.4mm auf 1.2mm vergrößert wurde XD. Wenn es so groß wird, macht es ein riesigen unterschied. Super tolle Video wie immer!
Printing 0.4mm gcode through a 0.25mm nozzle really surprised me. You should do a test of 0.25mm gcode through 0.25mm nozzle and compare; I'm particularly interested in which one replicates super-fine details (like you'd see on a tabletop game miniature) better, and which approach reduces stringing the most. I'm also curious whether the internal angle (60° vs 90°) contributed more or less than the chamfer to the stringing; I'm constantly fighting with stringing on my bowden-tube extruder.
I wonder if the polished finish of the internal bore would assist in retraction as opposed to the spiralled bore of the cheaper nozzles. The spiral could cause more friction at the tip leading to the extraction pulling the molten filament higher in the bore than right at the tip where you want to filament to be extracted to reduce the stringiness. I feel the stringiness is caused by the molten filament still attached as the printer head moves from one area to the next. Having a cleaner retraction would reduce the amount of filament being allowed to drag from one point to the next. A larger diameter hole would increase the problem too which is shown on your test prints. Smaller hole = less filament allowed to be dragged from one point to the next. If you increased your retraction as the diameter grew bigger you may find better results?! Food for thought.
Would it be then possible to recondition slightly worn nozzles by lapping the tip flat? Additionally, you could try drilling out 0.25mm nozzles to 0.3 or 0.35mm.
This is the one 3D printing channel I keep learning valuable, practical printing information from even after a couple of years of experience. Keep up the good work!
One of your best videos ever! I wish you had shows much more before and after shots and cuts though. Like you did with them mirrored for comparison but maybe split view instead so you line them both up and show each half from their respective nozzles. Looking forward to your steel nozzle tests!
Another fine video. It was interesting to see E3D's earlier research on this topic when they introduced the hardened nozzle and had similar results but much-increased internal wear. I think they did many more hours of printing. I use a lot of XT CF20 with the hard nozzle and there is no wear detectable by eye but it does eat the silicone socks!
first video i saw of you and was realy impressed. The "auf wiedersehen" really gave me the reaction: "off course !!! german engineering!!- moment" xp realy cool research!
Hi Stephan, as to how the hole came out bulging in the middle of the shaft, I would suggest wear on the cutting bit and the drill press/milling machine being in need of balancing. As to why the chamfer on the end of the nozzle I agree, I think it was the most expedient way to finish the part. As for a future test, since you have access to a milling machine you might try cutting back a bit, say ).005 " to see if you can coax any more life out of the nozzle.
The spiral pattern is from either a slightly bent drill bit or the chuck/tool holder is off center. Or if the internals are machined on a lathe the chuck of the lathe could be off center as well.
Ich denke diese spiralförmige Oberfläche kommt von einem etwas stumpfen oder falsch geschliffenen Bohrer der dazu neigt beim Bohren zu "wackeln". Diese Teile werden vermutlich auf drehautomaten hergestellt die sehr schnell Bohren was dazu führt das eine Spirale entsteht. Bei den teureren Teilen wird vielleicht gerieben oder so. :)
Nice video. I'd change one thing and that is to show the worn out tip next the original every time (side by side) Example at 9:36 you only see the bad one with nothing to compare. This would help compare the nozzles before and after without having to skip around in the video.
This is amazing, many people like me would have bought cheap chinese 3d printer kits therefore although your test is not complete is very helpful for 3d print enthusiasts like me who are on a budget. I almost exclusively printed in white to paint my prints by hand or finish them by other means I never considered the pigment.
Great video! Could you do a test on how the 60/90 degree angle in the nozzle affects the flow of the filament? I made some filament that has two colors down the length, and I would like to know what is the best way to avoid the mixing while printing
The stringing comes from the chamfer on the inside - it is an extra space that fills up with material. Also a harder edge means it will kinda "cut" off the strings instead of pulling out the material.
I changed to cheap ebay steel nozzle a long time ago and haven't looked back. All I had to do was up the print temp by 10 degrees to 205. Good video. Keep up the good work.
Subbed for an excellent product test video, considering you are a home 3D printer guy. Yes, it looks like the E3D nozzle has been reamed after drilling the internal bore. Whereas the Chinese one has been drilled, probably very quickly to reduce time. If the E3D has also been reamed to produce the outlet hole then there's a reason for their higher prices. Their process most likely follows this sequence. Hex bar is held in an auto lathe, the threaded section is machined to 0.1mm over it's final diameter, then the thread is cut, then a special tool reduces the material behind the thread and leaves the flat flange. After that a center drill goes in followed by a drill 0.1 smaller than the final filament sized special reamer. It looks like the reamer is a custom made one that produces the bore and the 60 degree end at the same time. But what I found with all of them is that externally, both cheap Chinese and more expensive E3D nozzles showed tiny dents on the outside (under a microscope) from where many finished parts are all stored loose in a container. And without a microscopic check of every nozzle, you might receive one with a little dent right across the working end. I did when I ordered a 0.15 nozzle, but they replaced it f.o.c as soon as I informed them of the problem. Unfortunately I haven't done any printing for a year, so I can't report on anything more than that. I'm a CNC mill setter by the way.
Can you please do this same test with Bambu Labs harden nozzles. But maybe do just carbon fiber and white filament as a test subjects. I can donate 4) complete unused .04 mm P1P nozzles as I use 06 nozzles on my micro farm.
I had my og ender 3 for over a year with the same nozzle bed everything. No issues. Pla only. Upgraded to ender 3 max and did a 20 hour print with glow in the dark and turned my .4 nozzle to a 2mm
Zu deiner Frage mit der "verdrehten" Bohrung: Vermutlich wurde nicht richtig zentriert, was auf der Drehmaschine dazu führt, dass die Dreh- und Bohrachse nicht übereinstimmen, dies führt zu Vibrationen beim Bohren, da dieser exzentrisch eindringt. Der andere Punkt werden die Schnittdaten sein. Ich vermute, dass der chinesische Lohnfertiger die Bohrer bei sehr sportlichen Werten ins Material treibt und die Werkzeuge relativ lange benutzt um die Kosten tief zu halten. Dadurch hat man zwar längere Standzeiten und höhere Zeitspanvolumen, allerdings leiden die Idealgeometrie und die Oberflächengüte darunter. Zudem werden auch erhöhte Temperaturen an der Werkzeugschneide wegen Abnutzung, Aufbauschneidenbildung und schlechter Kühlung einen Einfluss haben. Super Videos, so nebenbei :D Bei weiteren Fragen zur Zerspanung darfst du gerne fragen ;D
Question: how dependent is the bore accuracy in a zozzle? I ask because I bought cheap MK8 0.4mm brass nozzles and found many in the pack are near 0.45mm which I guaged with the base of a #77 drill bit which calipers showed to be 0.45mm diameter and I can manage to get it through the tip, though snug. A 0.50 cleaning needle won't go through though.
Excellent Video! I too thought the hole in the tip would widen first, but its the tip itself that wears down due to it rubbing against the filament just laid down. You probably got less wear out of the glow in the dark filament because it was from a quality manufacturer. Cheaper versions probably have larger particles in them that wear out, and clog, the nozzle faster. Very interested in seeing your results with hardened nozzles. Tungsten looks to be the best for wear resistance v. cost. But that's just my opinion without any testing. Eager to see what you find. Thanks!
Hey ! How are you doing ? Very interesting ! I think the spiral shape is because they introduce the drill very fast and hard for production, so the drill is wobbling around, don't have enough time to leave a clean hole. This can be fixed slowing the in-out drilling or speeding up the drill rotation. Thanks for sharing !
I probably have a worn nozzle, but I'm not entirely sure at this point.... will definitely keep an eye open next time I get new nozzles to make sure they're not the cheap ones you mentioned though; kind of always wondered what the difference in price represented, glad to know now
Feel free to share and discuss the video on you Social Media plattform of choice!
Noice !
I think premiere is a touch of hype ! I like it !
How to make such a shape ? DMG Mori 5-Achs Bearbeitungszentrum
For my (very may be clone e3d), i just pull up ''on cold'' 0.6mm nozzle fillamen without problems, for now its seems normal in 0.01+/- mitotoyo caliper
The spiral pattern is caused by a dull drill with way to much feed. The spirals of the drill try to get rid of the pressure and clog on the walls.
thank you CNC Kitchen for still producing 3d printing videos that are accurate. The others have given up. Watch you soon
The spiral pattern is achieved by plunging a dull bit into the brass with a lot of force. They make a tool to cut out the inside profile then use it until it breaks. Probably the first 1,000 nozzle's bore is totally straight, by nozzle 2,000 though you have the spiral pattern and somewhere after that the bit breaks and they make a new one.
So they just jam a drill bit in there?
@@cherrify3498 sounds like it
this^^^^^^^^
It's so simple it's strange no one understands
It's not a great application for a form drill because there is a high ratio of diameter for the tip diameter and the body diameter of the nozzle and also because the tip diameter is so small.
For a 1.75mm filament bore and 0.4mm outlet dia, one would have worse than a 4x ratio between the smallest diameter and the largest diameter on the form drill. This means that you'd really like to be drilling that tiny 0.4mm diameter a much larger RPM than you can run the 1.75mm diameter at because you'll exceed cutting surface speed limits on the larger diameter.
This means that you have to plod ahead at 1/4 of your best drilling rate for that tiny 0.4mm end which is going to be doing a terribly ugly drilling job for the entire length of the nozzle because it's stuck to the end of the 1.75mm body dia. It would be far faster to blitz the back end of the hole with the 1.75mm dia for nearly the whole nozzle length, then go in with a 0.4mm to peck out the last bit of length at the end.
Using a second tool position to have separate 1.75mm and 0.4mm diameters will be far faster with better quality drilling and much lower breakage of tools, not to mention that standard jobber drills are way cheaper than making a form drill.
I think the spiral pattern is being caused by a cheap jobber drill being pushed too far past their sharpening regimen. There's no really good excuse because clean holes in brass with jobber drills is very easy with CNC. The level of tool wear management necessary really isn't high.
I imagine that drill runout will also cause a spiral. An extremely true and concentric chuck is needed for highly accurate drilling at such small diameters.
Freaking brilliant video! Very impressed by the quality of production and the solid engineering content.
The undulating helical pattern is a likely result of a worn or chipped drill bit and pushing the feedrates on the lathe too hard. Also if you don't have your drill running dead centre it will walk and cause run-out. Glad to see ours were straight and shiny as they should be, but agree that it's not really a contributor to stringing.
You're dead right that the most important part of the nozzle is the orifice to tip flat transition. We've put a lot of time into ensuring that corner is 'just so'. The internal cone angle is also a surprisingly large contributor to the ooze and clean sharp extrusion, it gets you the more laminar flow you postulate and lowers pressure needs. Combine those two and you get the rather dramatic difference you saw in stringing.
I suspect but cannot prove that the excessive chamfer on the clone and worn nozzles also serves as a place where material will cling to and accumulate. This little annular 'pool' of very liquid material may serve as a source of material that will be entrained into a string. Retraction is going to be completely ineffective at controlling any buildup on this part of the nozzle.
I was going to say the spiral was from a blunt drill but it has been said in detail above so I'll shut up now. Thanks very much for the great videos @cnckitchen
I was going to say: this guy sounds really sure about his statements, then I saw your name.
Genghisnico13 the “our” referring to the e3d nozzle was my first clue - people sound sure of themselves in YT comment sections allllll the time ;)
I just want to add ontop of your fine explaination about the helical pattern.
-This is a great example why GPS tolerances (Geometric Product Specification) was invented, and specifically why it's important to have them specified when you make large orders produced out of China and the like. Not that it isn't necessary in all other countries though. :-)
Excessive feed rate, lack of pecking and worn tooling is the most likely causes, but materials like brass also have issues caused by their high thermal expansion. Materials do not heat evenly when drilling and tends to build up heat near the corners of the drill flutes during the machining process. Thus the materiel near the corners of the drill flutes is removed at a higher temperature so those areas expand more then the other areas of the cut. This leaves a smooth hole at the time of operation, but then when the materiel cools and constricts it leaves a helix or spiral shape that matches the drill feed rate. Brass and plastics it tends to become fairly prominent due to their high thermal expansion rates and high drill feed rates.
Insane production quality, better than ever - so much time and effort and it really shows!
hello
giri nevass hi
@CNC Kitchen, that's my blownout nozzle at 10:34! I got that after printing approximately 1 spool of inland glow in the dark PLA. Thank you for showcasing it in this video, it was a hard lesson learned for a new 3d printing user.
That spiral is from the drilling procedure at the factory. The tool is flexing as it's plunging in.
If the drill vibrates, it usually makes cones. Like an hourglass shape. The walls of hole look like melted, probably drill was overheated during the production and left that weird pattern?
25 year machinist .. that pattern happens when the drill geometry is shoddy, specifically the lead on the spiral or the point being off center slightly and your feed is way to high. Brass has a tendency to grab when using sharp tooling. A way to mitigate the grabbing is to flatten the cutting edge slightly on the perpendicular axis of the cut. This is probably the reason for the over powered feed so as to compensate for that flattened cutting edge.
@@swould333 I can confirm that, worked a lot with brass in the past, it likes short stubby cutting edges and the unevenly sharpened chinese 1.75mm drill bit is far from that. High speed vibrations + eccentric forces combined with resonance from the brass grabbing and you get forms you could not imagine can exist!
@@swould333 yep.
@@swould333 curious if the good nozzles are cut in three procedures and that's why they have good surface finish and geometry that clearly wasn't taken off a shotty twist drill.
The wobbly pattern of the passage through the cheap nozzle is from a warped drill bit.
They used a high rotation speed (probably with inadequate cooling fluid) and overheated the bit several times and they were using a high rate f driving the bit into the brass.
Its an effect I got when learning to use a home drill press.
F Huber it most likely isn’t cut for brass and was trying to pull the part as well which combined with being cooked and probably dull
Some more insights as to how the wobble might have occurred: The geometry of nozzles lends well to lathes (and by extension screw machines), where wobble of both the part and the drill can be more severe vs with a mill. It's also likely that the stock used to make the nozzles is at the max diameter of the part to minimize material waste, and that thinner material would also wobble more. As for feeds and speeds, the high machinability of brass allows very aggressive settings without significantly shortening tool life (and that's great for minimizing labor cost, but bad for part quality).
@@tronique5736 most probably machined from hex bar
Also I bet the is not exactly round, but slightly elliptical due to the same reasons. I’ve learned the hard way as a home gamer that drilled holes are always bigger and never round.
Hi pal!. I just bought my first printer and I'm soaking up information with a sponge (even though I could be your grandpa because of my age... I never tire of learning!), this report opened my eyes to many doubts I had and I think a spectacular job on your part, very serious and dedicated. This is the first video of yours I've seen and I've subscribed to the channel to see the rest. A pleasure to learn from people who really care about finding the important point of the problems. My congratulations, today, December 31, 2022, I send you a big greeting from the South of Argentina and I wish you a good year!
The cross-section of the genuine E3D nozzle really sold me on their products. Such amazing quality assurance; and it really shows why it's smart to dole out the little extra cash for it.
Stefan, deine Videos sind einfach oberklasse. Qualitativ hochwertig. Informativ. Gute Geschwindigkeit. Nie langweilig. Helfen mir jede Woche neu!
This guy... This guy right here is one of the absolute best things about TH-cam.
This guy has better English than most Americans and better production quality than a lot of big TH-camrs. Great job!
I asked 3dpn a million times to make a video about this one topic . You really did a VERY good job putting as much thought into the tech as well as the science.
Happy to help.
in depth tests like this, saves us a lot of time and money before deciding. thanks so much.
I concur with your findings, it was interesting to see the 60 degree chamfer on the inside of the e3d nozzle verses the 90 of the in expensive nozzle. Custom tooling verses standard drill bits, that is significant.
Both angles point to an custom drill bit. (Which arn't expensive when you mass produce.) The default angle on a HSS drill bit is 118°. But there can be differences for hard and soft maqterial from 110° to 130°. Carbide drills normally use 140° tip angle.
Most spot drills have a 90 degree angle and are extremely rigid. Could be step one for E3D’s process...
You are correct. Machining bits can be 90 degrees.
That would make since.
With a finishing cut at 60
I found the section on glow in the dark filament really interesting! I had been printing a very large gift for my fiance and the nickel plated nozzle that I had been using started to show signs of damage. The filament would shoot off to the back and left no matter what filament I used. Just bought myself some steel nozzles because of it! Thank you for the great videos and keep it up!
The effort put into these videos is unbelievable
Useful info, thanks.
I have a small lathe, tried making my own nozzles, they worked ok, but no better than the bought nozzles. The hand-made nozzles took a lot of time to make, and the materials are expensive.
A good compromise is to buy 0.2 or 0.3mm nozzles, and drill them out to the required size. I find 0.35mm is a good size for the kind of printing I mostly do. Carbide PCB drills are cheap, but it's easy to break them. I start by facing off a piece of scrap stock, drill and tap M6, and screw a nozzle on. I do this for each batch of nozzles to maintain concentricity. The drilling part is easy, as long as your tailstock is aligned correctly.
The chamfer on brand new (cheap) nozzles effectively increasing their diameter and hurting print quality is a very interesting discovery. Thanks Stefan, glad we have people like you in the community 👍
I think it’s worth noting though that those cheap nozzles are made by a very wide variety of manufacturers, and it may just be that the particular nozzles he tested were especially bad...
I have only cheap Chinese nozzles, a big variety box of standard CR10 style nozzles, as well as one V6 style I was sent by mistake, and about 6 volcano style I bought for my Sidewinder X1. I just double checked all of the volcano and the v6 nozzle as well as a random selection of the CR10 nozzles and not a single one had that chamfer. I even double checked under a microscope in case I just couldn’t see it with my naked eye.
I’m sure the internal bore of the nozzles is not as clean as genuine E3D nozzles, and there may be a difference in the internal angle, but at least it doesn’t appear that the no name brand nozzles all have that chamfer or that it’s even a common defect. He may have just received a defective batch or bought from a particularly bad manufacturer or I’m just extremely lucky...
He’s 100% correct that you’re taking a gamble not buying genuine, but given that the cost of a single E3D volcano nozzle is about the same all of my Chinese volcano nozzles combined, it may be worth the risk as you won’t lose much money if they turn out bad. Though of course I do appreciate supporting the original creators. Also for example there is a massive difference in quality between the heat breaks especially on knock off E3D extruders vs the real ones that I think makes them an obviously better buy even at the in increased cost (my sidewinder for example has a PTFE tuve all the way through the heartbreak which is an issue for high temp printing, so I’m planning to upgrade to an official E3D all metal titanium head break which is actually about the same price as some of their nozzles somehow even though it’s a more complex part).
Also it’s worth considering that I can buy the Chinese nozzles easily here in Colombia with local 1 day shipping, whereas to buy genuine E3D nozzles I would need to order from the US and import them, increasing the already higher cost due to reshipping and import taxes as well as significantly increasing the wait time to receive them if I need one quickly.
I’m still planning to buy a few genuine nozzles to use as a baseline comparison against my cheaper nozzles to see if I get increased print quality, but I don’t think it’s fair to test a single brand of Chinese nozzle and then imply that all Chinese nozzles are manufactured that way or have similar problems. Many of them are perfectly fine from what I can tell (including from my own print quality which is quite good using only these nozzles).
Anyway just my 2 cents.
Hi Stefan, I'm using probably the same ruby nozzle that you have in your ruby collection. It has a chromed copper body. I use it with a titanium heat brake (all metal) and a home-designed & milled copper heater block. The back pressure in the nozzle is greater (especially with CF-PEG) so I use a dual-drive geared extruder. With a bowden cable, in a Anycubic kossel linear plus. It took me a while to get good results again, but now it's spot on. I play around with linear-advance a bit and with normal PLA it's better not to use it (K-factor zero). Hope you're having fun with it and happy printing!
In a previous life I was a tool and cutter grinder in a mould shop. One tool I had to design and make was a D-drill for plastic injection nozzles. The interior bore ended in a 40° included angle hole. I assume this allowed for faster flow in the injection process. Materials for the nozzles were either beryllium copper or a tool steel. the nozzles also had a variety of exit holes, some on the ends and others at angles coming out on the sides of the tip.
James Crombie that's awesome!
I thought that angle was really interesting. I'm guessing E3D have tried all sorts of angles to get one that works best, but I'd be interested in results from even more acute angles.
The spiral pattern happens when you drill a hole on a lathe but the tail stock is off center it makes the bit flex and wiggle as it rotates
Stefan,
Job well done, as always.
I print mainly with flexibles and CF nylon on the same Prusa i3 MK2s. Yes, two completely different materials, each with their own challenges.
I was having major extrusion issues and jams. I replaced the E3D hardened nozzle that I had run a few kilos of CF nylon through with a new E3D Nozzle X with no real change. I then looked at the heat break tube and saw some markings on the inside of it, so that got replaced. This improved my results a fair bit. I still had problems. I noticed inconsistent layer adhesion on the small parts that I print using Ninjateks Armadillo. There is a post that is 4mm diameter that when everything is right will not snap off. I print about twenty of the parts at a time and have maybe 4 - 5 that fail. I then looked at the Hobbed Pulley and noticed some wear. I didn't realize how much until I removed it and compared it with a new one. (See the photo in the attached link)
There are many aspects that can cause extrusion issues. If you use the "Cause and Effect" process the best method for solving the problem is to go back the where all was good.
photos.app.goo.gl/SDJUQd27NxsFrfXd9
photos.app.goo.gl/fLq8RsCXCE3vuNza7
Wow! Ones again awesome tech analytical content. As a mechanical engineer I really appreciate your channel. I will never print a cheap nozzle from now on!
There are so many interesting points in this video. It is really good to see the actual differences between the parts of different manufacturers. I've heard many TH-camrs talk about the poor quality of cheap Chinese parts, but you actually show the difference. It would be interesting to see the real world print quality difference between these, and examine why they may or may not be worth the extra cost. I would think that the quality of the print may also affect the overall structural quality of what you print as well. Basically how much does 3D-Printer part quality affect the output, and how much does it cost to mitigate the loss in quality.
In my opinion, the E3D is superior in terms of stringing because of the 60deg angle of the convergent channel (instead of 90) as it will create less flow resistance, thus lowering the pressure buildup in the melt chamber. With a lower pressure accumulated in the chamber there is less viscoelastic flow (ooze) when you stop extrusing. Also the sharp transition can help too as it will cut the strings and avoid pulling out filament. Great video as always, and excellent scientific approach to real world problems.
Ps: funny enough I just noticed that Sanjay just made a similar comment.. Seems that physics is the same for all of us ;-)
Very nice tests. I have some glow-in-the-dark PLA from Velleman and it is definitely abrasive, I also noticed the nozzles being ‘sanded’ at the bottom after printing a lot of this material. The hole diameter seems to have slightly increased as well, although the only way I noticed this is that it gradually became easier to push a .4mm guitar string through the hole.
A larger sanded-off nozzle bottom does have an advantage in that it makes it easier to get nice smooth top surfaces, but a problem is that it dumps more heat into nearby already printed material, which makes for worse overhangs and also it has more of a tendency to drag along PETG because it is so sticky.
Exactly.
The biggest problem of abrasive filaments appears not inside(hole diameter) but outside(hole length).
From my experiences, the most abrasive additive agent is "glass fiber", and not carbon fiber or metal particles.
Actually it can shorten stainless steel nozzles over 0.5mm within one reel(500g).
This means the nozzle diameter changes from 0.4mm to 2.0mm, or simply the nozzle has died out.
If diameter of the nozzle was changed at this moment, it is not so big of a deal.
I tried to use HSS_Co(used for cutting tool of stainless steel) nozzle for the glass fiber filament, but it does not work.
So currently I am using stainless steel nozzle once for small parts,
and flat top HSS_Co nozzle for large parts.
Klasse Video, sehr informativ. Großartig was du dir für Zeit für die ganzen Tests genommen hast!
Und was mich am meisten freut ist, dass dein Kanal sehr, sehr schnell so gewachsen ist :D Sauber!
Grüße, Sebastian
Vielen Dank!
I have my voron in production mode, and have about 2000 hours on it so far. I used a ruby nozzle just so I never needed to worry about nozzle wear.
I think that you have definitely proven that it is worth the extra money for the quality nozzles from more reputable companies like e3d. Just the cutaway views of brand new nozzles is enough to convince me of the difference. I will be ordering some soon. Thanks for your usual in-depth research on this issue!
Impressive quality of this video: How you shot it, that you cut the nozzles in half, content-wise. Good job.
Great as always. I'm also happy to see I'm not the only one who cut away the bottom of the silicone sock.
These are actually the new silicone socks, the closed ones are now called "pro" socks.
@@CNCKitchen Funny, I don't know what's pro about that. I had to cut away a mm or two of the bottom of every sock so far.
CNC Kitchen was mentioned on the last podcast of "3D Printing Today". They were impressed. High praise as these guys have been in the industry before it was consumer availible product.
Oh cool, gotta listen to it!
great video...suggestion....take a worn nozzle and CNC a new flat extruder surface and see the difference in the benchy....
That's what I thought. And another option (which I'm going to try after fixing my printer) is to… Do the same with sandpaper.
@@JayOhm i have tried the sand paper but cant get it flat enough. Looks ok to the naked eye, but under a microscope you can see the rounded edges
I have actually researched this quite a bit and read about similar types of methods of testing. I think most nozzles lack precision and I have actually decided to manufacture my own and am currently building a machine to do just that and want to eliminate or minimize wear as much a possible without sacrificing precision at all. I don't believe calipers are precise enough for the tuning/calibration of a 3D printer. I have access to a fully equipped machine shop (though I understand not everyone does) and use gage blocks, micrometers, and microscopes to measure everything. I like your videos, I'll send you one of my printers when I'm finished with it.
I don't know if you realize how much some of your videos, especially ones like this one, add to the 3d printing world. In this case, we pretty much lived with "those filaments cause the nozzle to wear" and I'd bet that we all assumed that meant that the hole in the nozzle was getting bigger... instead the face is wearing off it... huh. That has totally different implications WRT to dealing with the wear.
I've already been modifying the cheap Chinese nozzles for better print quality and reliability. Now you have me considering changing the procedure somewhat, I'll likely custom grind a drill bit to have a 60* tip on it (I knew about the spiral finish on the inside, it was causing my printer to jamb up once in a while so I run a drill bit through new ones to clean that up) and I'm thinking of trying polishing the points flat on them.
Keep up the good work!
Sanjay Mortimer
's answer on the helical pattern seems unlikely.
The consistent helical pattern is more likely to come from a drill that is mounted on a bent lead-screw. This explains the consistency much better.
I have personally seen that helical pattern when printing on my makerbot cupcake with bent lead-screws. I immediately assumed that was the cause when I saw the cross section.
Congrats for the professional video! Although 3 years has gone, this looks the right place to ask my question.
There are hardened steel nozzles available today (e.g. for my Bambu Lab P1P as a spare part) these are intended for abrasive filaments, like carbon fiber. These nozzles cost about the same as their normal steel ones (their standard nozzle is steel, they have no brass nozzles) so my idea is to use the hardened steel version for all filament materials (PLA etc.)
What could be the disadvantages of a hardened steel nozzle? There might be a difference between the thermal conductivity of brass vs steel (which may somehow affect print quality, maybe the max. print speed?) but I think there is no significant difference between normal vs hardened steel in terms of thermal conductivity. I can think of one disadvantage of hardened steel is that it is more brittle, but that plays no role here. What do you think?
I’m really interested in a video testing the possibility of improving the print quality using a smaller nozzle.
Cool video! I always learn a lot from you. ❤
Thank you for the information. You helped solve why my prints were degrading and why I've been having an increase in stringing in my prints. Cheers.
I enjoy your attention to detail and you methodical approach to testing..
Excellent video as usual from the best 3DP channel on TH-cam. There's always good information to digest from the CNC Kitchen :-)
I love it when we get the actual science, and in a form where we can see it for ourselves!
I'm looking forward seeing you machine a tungsten nozzle 😆 Great video as always! I too have a brass nozzle and the wear was primarily on one side, the filament free extruded to one side too. I used a steel tweezers to clean the nozzle regularly, which probably did some damage too. The surface finish with a new steel with zinc coated e3d nozzle is significantly better. A new brass nozzle was going to be better too of course, compared to the damaged one. I'll try to machine the nozzle too, would be interesting.
I'm glad that you tested with clone nozzles. For as much as I might like E3D parts and prefer them, if I need a pack of .4's (sometimes you just know you'll be wrecking nozzles with wood), or a couple in each diameter between .2 and .6, it's usually the difference between $10 and $50. They aren't as good, but they'll usually get the job done, and it's good to know how long they'll handle that job. So thank you.
Is it possible that the tip of the nozzle sands down by touching the prints, so it sands down faster at lower layer heights?
yes and if over extruding even a little bit the nozzle will drag more.
That means one should beware of excessive ironing, right?
As an engineer I always enjoy your vids. Your scientific approach and attention to detail are definitely value added. I would be curious to see if removing material from the face of the tip of the cheap nozzle, until the chamfer was removed, would improve the print quality. Also, I noticed that the surface finish of all the nozzles was bit rough. Would polishing the nozzle face help the print quality?
Brilliant. This video concludes everything you need to know about nozzles for 3d printing. I learned that brand / quality matters and that you must choose nozzle according to materials used. Thanks for the research! Great work
now i kinda want to see the new cloned Ruby (FYSETC also do a white sapphire nozzle [Sapphire is available in all colours asided frfom red, though red corundum is just a ruby so very close] that'd be fun)
Great video as always! Don’t forget to try a slice engineering vanadium nozzle in the abrasive resistant shoot out.
Great subject to point out !
I believe that if you used an extremely strong tool (such as adjustable pliers) to tighten the nozzle it can easly deforms the inside of the nozzle and will only have a slight affect on the outside of the nozzle.
I would suggest to check it after tighting (with a strong tool) a brand new nozzle.
I use MG Chemicals Super Glow ABS almost exclusively for my "White" prints and have run over 6 rolls through my machine and have only changed nozzles like once and that was just to rule out the nozzle for another issue, so I'm glad you were underwhelmed with the glow results. That matches my experience. Half the stuff in my house glows now ;)
The odd spiral pattern is most likely caused by the stock being off-center when drilled. This causes the drill to wander, due to flexing, while drilling. I have seen similar behavior when milling a circle, using a bad collet that mis-aligns the tool. It can actually create a hole that is hourglass shaped in cross-section!
Also, unevenly sharpened drill bit does the job. Possibly they didn't even use center drill.
yeah no center drill would do it, i wonder boring it on our own would do any improvement
I don't yet have a 3D printer, but I really enjoy how much thought goes into these videos.
I wish there was a way to like the same video 1000 times. This guy definitely makes amazing videos.
Another very interesting video Stefan.
I don't have a 3d printer at the moment because I can't open that can of worms at this point due to other priorities. However, I do plan to get one sometime next year. I still watch your videos because they answer many of the questions I ponder or answer questions i didn't even think about. I think when I do get a printer your videos will already have helped me better understand this process.
One thing I do wonder about you touched on in this video which is nozzle diameter - because I assume it is related to wall thickness. Some of the things I am considering printing will require thin wall thicknesses. I want to make fuselage type structures as lightly as possible - with and without bracing/infill.
So the questions i ask are - How small a nozzle diameter can you use? How thin can a cylinder's wall thickness be? What are the limits? I wonder if you can do a video which pushes the boundaries in that direction.
Many thanks.
Hey! You skipped MY blown-out nozzle on the Reddit thread! What gives?!
No, that's fine. I'm embarrassed about it anyway. Great video, thanks for such a well-done breakdown on the effects of abrasive filaments!
Your tests match mine. I found that nozzles wore in both hole diameter and height, but wore in height more signficantly.
Honestly you are making the best 3d printing videos. Sau stark. Weiter so.
at 12:33 you note the point angle of the E3D is 60 degrees while the cheap knockoff uses one of 90 degrees. I originally tried an Olsson Ruby nozzle to solve my wear problems, and during a spontaneous unscheduled disassembly of said nozzle I discovered the Ruby hides a very dirty secret: it has a point angle of 180 degrees internally which is very poor in general, but particularly for feeding fiber filaments. Something to be aware of. :)
I've since switched over to a tungsten carbide nozzle and am very happy with it. Much more abrasion resistant than the hardened steel nozzles I've used.
This is how the ruby nozzle looks from the inside: twitter.com/erikcederb/status/1125056521199673344
There is a point angle but then, there is a small, flat section. Interestingly the standard Ultimaker nozzles look very similar on the inside.
I’d be curious how Anders Olsson’s ruby nozzle compares!
Stefan@CNC Kitchen, and Stefan Gotteswinter. My favourite TH-cam subscriptions. So logical. So scientific.
Looking forward to your Ruby nozzle review.
Mein Drucker hat für sehr lange so schlecht gedruckt, dass ich überhaupt nichts drucken könnte und ich wusste nicht, was damit passiert ist. Ich habe endlich erkannt, dass meine Düse von 0.4mm auf 1.2mm vergrößert wurde XD. Wenn es so groß wird, macht es ein riesigen unterschied. Super tolle Video wie immer!
Printing 0.4mm gcode through a 0.25mm nozzle really surprised me. You should do a test of 0.25mm gcode through 0.25mm nozzle and compare; I'm particularly interested in which one replicates super-fine details (like you'd see on a tabletop game miniature) better, and which approach reduces stringing the most.
I'm also curious whether the internal angle (60° vs 90°) contributed more or less than the chamfer to the stringing; I'm constantly fighting with stringing on my bowden-tube extruder.
I can't get rid of stringing with eSun PLA+. Not sure if it's the filament though.
I wonder if the polished finish of the internal bore would assist in retraction as opposed to the spiralled bore of the cheaper nozzles. The spiral could cause more friction at the tip leading to the extraction pulling the molten filament higher in the bore than right at the tip where you want to filament to be extracted to reduce the stringiness. I feel the stringiness is caused by the molten filament still attached as the printer head moves from one area to the next. Having a cleaner retraction would reduce the amount of filament being allowed to drag from one point to the next. A larger diameter hole would increase the problem too which is shown on your test prints. Smaller hole = less filament allowed to be dragged from one point to the next. If you increased your retraction as the diameter grew bigger you may find better results?! Food for thought.
Would it be then possible to recondition slightly worn nozzles by lapping the tip flat?
Additionally, you could try drilling out 0.25mm nozzles to 0.3 or 0.35mm.
Should work but probably not worth the effort if a good, genuine new one is just $5.
This is the one 3D printing channel I keep learning valuable, practical printing information from even after a couple of years of experience. Keep up the good work!
The flow of filament is the most important, you can print 0.6 with a 0.4 nozzle .Filament will expand outside because of pressure difference.
One of your best videos ever!
I wish you had shows much more before and after shots and cuts though. Like you did with them mirrored for comparison but maybe split view instead so you line them both up and show each half from their respective nozzles.
Looking forward to your steel nozzle tests!
Excellent video Stefan. I found this video very helpful. You put so much work into your high quality videos.
Super interesting findings. I really appreciate these types of investigatory videos that you do.
Thanks for the test. Could you also test ruby nozzle from some chinese manufacturers (ie trianglelab have one and it's for half the price of ollson).
Another fine video. It was interesting to see E3D's earlier research on this topic when they introduced the hardened nozzle and had similar results but much-increased internal wear. I think they did many more hours of printing. I use a lot of XT CF20 with the hard nozzle and there is no wear detectable by eye but it does eat the silicone socks!
first video i saw of you and was realy impressed. The "auf wiedersehen" really gave me the reaction: "off course !!! german engineering!!- moment" xp realy cool research!
Now i finally know why i get strings...cheap nosel + lots of Glow in the Dark! Thank you for testing these things!
Hi Stephan, as to how the hole came out bulging in the middle of the shaft, I would suggest wear on the cutting bit and the drill press/milling machine being in need of balancing. As to why the chamfer on the end of the nozzle I agree, I think it was the most expedient way to finish the part. As for a future test, since you have access to a milling machine you might try cutting back a bit, say ).005 " to see if you can coax any more life out of the nozzle.
The spiral pattern is from either a slightly bent drill bit or the chuck/tool holder is off center. Or if the internals are machined on a lathe the chuck of the lathe could be off center as well.
Ich denke diese spiralförmige Oberfläche kommt von einem etwas stumpfen oder falsch geschliffenen Bohrer der dazu neigt beim Bohren zu "wackeln". Diese Teile werden vermutlich auf drehautomaten hergestellt die sehr schnell Bohren was dazu führt das eine Spirale entsteht. Bei den teureren Teilen wird vielleicht gerieben oder so. :)
Nice video. I'd change one thing and that is to show the worn out tip next the original every time (side by side) Example at 9:36 you only see the bad one with nothing to compare.
This would help compare the nozzles before and after without having to skip around in the video.
I love these videos, but I'm now off to look for that CNC mill thing, wow!!
Chapeau bas. The amount of work in this video is overwhelming.
This is amazing, many people like me would have bought cheap chinese 3d printer kits therefore although your test is not complete is very helpful for 3d print enthusiasts like me who are on a budget. I almost exclusively printed in white to paint my prints by hand or finish them by other means I never considered the pigment.
Great video! Could you do a test on how the 60/90 degree angle in the nozzle affects the flow of the filament? I made some filament that has two colors down the length, and I would like to know what is the best way to avoid the mixing while printing
The stringing comes from the chamfer on the inside - it is an extra space that fills up with material. Also a harder edge means it will kinda "cut" off the strings instead of pulling out the material.
I changed to cheap ebay steel nozzle a long time ago and haven't looked back.
All I had to do was up the print temp by 10 degrees to 205.
Good video. Keep up the good work.
A fun test would be milling off the very tip of the cheep nozzle (one or two thou) and see if it improves the print quality.
Subbed for an excellent product test video, considering you are a home 3D printer guy.
Yes, it looks like the E3D nozzle has been reamed after drilling the internal bore. Whereas the Chinese one has been drilled, probably very quickly to reduce time. If the E3D has also been reamed to produce the outlet hole then there's a reason for their higher prices.
Their process most likely follows this sequence.
Hex bar is held in an auto lathe, the threaded section is machined to 0.1mm over it's final diameter, then the thread is cut, then a special tool reduces the material behind the thread and leaves the flat flange. After that a center drill goes in followed by a drill 0.1 smaller than the final filament sized special reamer. It looks like the reamer is a custom made one that produces the bore and the 60 degree end at the same time.
But what I found with all of them is that externally, both cheap Chinese and more expensive E3D nozzles showed tiny dents on the outside (under a microscope) from where many finished parts are all stored loose in a container. And without a microscopic check of every nozzle, you might receive one with a little dent right across the working end. I did when I ordered a 0.15 nozzle, but they replaced it f.o.c as soon as I informed them of the problem. Unfortunately I haven't done any printing for a year, so I can't report on anything more than that.
I'm a CNC mill setter by the way.
Can you please do this same test with Bambu Labs harden nozzles. But maybe do just carbon fiber and white filament as a test subjects. I can donate 4) complete unused .04 mm P1P nozzles as I use 06 nozzles on my micro farm.
Stefan, great job, always look forward to your good engineering and scientific method. Danke.
I had my og ender 3 for over a year with the same nozzle bed everything. No issues. Pla only. Upgraded to ender 3 max and did a 20 hour print with glow in the dark and turned my .4 nozzle to a 2mm
Zu deiner Frage mit der "verdrehten" Bohrung:
Vermutlich wurde nicht richtig zentriert, was auf der Drehmaschine dazu führt, dass die Dreh- und Bohrachse nicht übereinstimmen, dies führt zu Vibrationen beim Bohren, da dieser exzentrisch eindringt. Der andere Punkt werden die Schnittdaten sein. Ich vermute, dass der chinesische Lohnfertiger die Bohrer bei sehr sportlichen Werten ins Material treibt und die Werkzeuge relativ lange benutzt um die Kosten tief zu halten. Dadurch hat man zwar längere Standzeiten und höhere Zeitspanvolumen, allerdings leiden die Idealgeometrie und die Oberflächengüte darunter.
Zudem werden auch erhöhte Temperaturen an der Werkzeugschneide wegen Abnutzung, Aufbauschneidenbildung und schlechter Kühlung einen Einfluss haben.
Super Videos, so nebenbei :D
Bei weiteren Fragen zur Zerspanung darfst du gerne fragen ;D
Question: how dependent is the bore accuracy in a zozzle?
I ask because I bought cheap MK8 0.4mm brass nozzles and found many in the pack are near 0.45mm which I guaged with the base of a #77 drill bit which calipers showed to be 0.45mm diameter and I can manage to get it through the tip, though snug.
A 0.50 cleaning needle won't go through though.
Excellent Video! I too thought the hole in the tip would widen first, but its the tip itself that wears down due to it rubbing against the filament just laid down. You probably got less wear out of the glow in the dark filament because it was from a quality manufacturer. Cheaper versions probably have larger particles in them that wear out, and clog, the nozzle faster.
Very interested in seeing your results with hardened nozzles. Tungsten looks to be the best for wear resistance v. cost. But that's just my opinion without any testing. Eager to see what you find.
Thanks!
I’m trying to learn about nozzle wear and then a WHOLE LTT episode over intel gold cpus pops up as an ad lmao. And it’s 1 of 2!!!!!!
Hey ! How are you doing ? Very interesting ! I think the spiral shape is because they introduce the drill very fast and hard for production, so the drill is wobbling around, don't have enough time to leave a clean hole. This can be fixed slowing the in-out drilling or speeding up the drill rotation. Thanks for sharing !
I probably have a worn nozzle, but I'm not entirely sure at this point.... will definitely keep an eye open next time I get new nozzles to make sure they're not the cheap ones you mentioned though; kind of always wondered what the difference in price represented, glad to know now
awesome vid! put 500g of glow in dark through, now trying to solve stringing and was worried i messed up my nozzle.