13 Changes that might Improve Prusa i3 mk3 Print Quality
A list of change could will improve The prusa i3 mk3 print quality due to extrusion issue 602. A lot of people watched my original video on the inconsistent extrusion issue and so it seemed only right that I also provided a suitable update. There are a number of issues that make up the inconsistencies of issue 602 so here are a list of fixes that you should look into if you think you have inconsistent extrusion or anything that looks like issue 602.
Affiliate Purchase Link:
Buy Prusa Mk3s+: shop.prusa3d.com/en/51-origin...
0:00 Intro
03:15 loose extruder screws
03:40 bearing quality
04:20 loose z-axis rods
04:55 z-axis nuts
05:44 lose hotend and extruder assembly
06:51 ptfe tube
07:31 incorrect meshing on the Bondtech gears
08:33 filament biting variation
09:40 x ends being too loose
10:33 belt rubbing
11:02 linearity correction
11:22 stock rubber feet
12:06 belt quality
13:11 hotend cooling issue
Prusa Blog Post: www.prusaprinters.org/prusa-r...
Github Post: github.com/prusa3d/Prusa-Firm...
Fixes Wiki Page: prusaowners.com/wiki/index.php...
More in depth discussion at the discord channel: / discord
Gear: www.vector3d.co.uk/my-gear
Blog: www.vector3d.co.uk/blog
Website: vector3d.co.uk/
03:15 loose extruder screws 03:40 bearing quality 04:20 loose z-axis rods 04:55 z-axis nuts 05:44 lose hotend and extruder assembly 06:51 ptfe tube 07:31 incorrect meshing on the Bondtech gears 08:33 filament biting variation 09:40 x ends being too loose 10:33 belt rubbing 11:02 linearity correction 11:22 stock rubber feet 12:06 belt quality 13:11 hotend cooling issue
90% of "changes" is just maintenance of your printer...
Sure are. and I'm happy to have the list of things to check
Thanks for the tip on the heater block being loose, that fixed a clogging issue I have been having for a while!
Very good and informative video! I replaced my bearings with Igus Drylin bearings. They're like the PLA bearings shown, but smoother, different (self-lubricating) plastic. Very nice. One thing to check that wasn't mentioned is the "squirrel cage" fan. Mine was fine up until about 60% RPM's, then I would get decent enough vibration from it to give me a little salmon skin. Fans go bad sometimes. Replaced it with a Delta fan and all is well.
I received my MK3 kit in June and haven't had the inconsistent extrusion that is seen with issue 602 so it is odd that it is hit or miss with each printer. Also, about the heatsink not staying cool enough, Prusa released a new hotend design about a week ago that adds airflow as well as a better cooling fan mount/angle. I printed it in PETG and it definitely does keep the hotend cooler, as well as increases part cooling especially for bridging and overhangs.
i do like the explanation (half verbal/ half non verbal) of the relativley warming of the cold end on 14:30, thanks a lot for that. C.
Have you done any of these modifications on your printer? If so, what were the results? Kind of dissapoitned that there are no before and after shots to see if all of these modifications really make a difference. Especially since one of your recommendations was to replace the bearings with Misumi bearings which run $10 per bearing; $100+ to replace the entire set.
well thx it's old video by today but not speaking about your knowledge and explanation which are from my point of view excellent , but personnaly i like to have lot of pics and details on the pics during talking . We can undersand that you really know what you talking about and did some test and your analyse is right...now to follow you especially as foreigner is not always easy and trying or doing the change or correction that you mentioned is hard only to follow what you say..hope i am clear and it can be a positiv critic..thx a lot and happy to follow u
I have been running Bondtech gears since they were an alpha product. The only way you can cause issues with them is to over tighten them forcing the idler gear too far into the root of the driven gear. I think recommending people grind off the teeth of the idler gear is unnecessary and a bit extreme. The filament acts as the couple between the gears as well as the gear meshing, so a light gear mesh is all that's required. But thats just my 2cents after 4+ years of running Bondtech gears.
12.30 That is not high tensile strength, it is high Young's Modulus. Materials can have a high tensile strength and stretch a lot.
Thank you. Very good report.
you talk about the screws coming loose. Can you use something like a loctite threadlocker to help the screws stay tight?
This was nice. I am in work to rebuild my mk3.
Thank you! Best summary!
im not happy with my prusa i3 mk3 , i ll try what u say . thanks
I dont know but 75% of the tipps is to replace parts, I don't see why this is a suitable solution. Disclaimer: I also own the mk3s and I am happy with it. If I would have to replace everything in order to get it working properly, maybe it is not the right printer in the first place.
Thanks for the video! You mentioned replacing belts. What do you think about steel reinforced belts?
@@Vector3DP they are, indeed, much stiffer and pretty hard to bend... that's what got me worried. I am waiting on my MK3 kit to arrive, and will try the steel reinforced belts, since that's what I have at hand.
What about that Layer issue now??? Is there any fix already?
It is not just loose carriage, it may be too tight. That was my problem. A test I propose is to loosen the x-carriage slightly. If the x-belt tightness goes up (in settings) , i.e. looser then you are probably going in the right direction. Now retighten x-carriage with a consistent and slighter lower tension than before. This solved the calibration problem for me. Albert
It’s an old video but maybe somebody can make use of this info: One thing that helped me is to check the position of the axis of the free spinning bondtech gear which you tighten with the screws. The axis is to short so you have to make sure it’s placed centered so the bondtech gear is fixed to both mounting holes. If this is not the case the gear will wobble and not be able to properly grab the filament with can cause inconsistent extrusion. In my case the benchy printed fine every time even though the axis was out of place but my own designs failed. This might be because the settings in the GCode for the benchy are put very conservatively? IDK I’m new to this... However, it only started to get better for me after i properly centered the axis so the bondtech gear was firmly in place and could lead the filament into the hotend.
Comprehensive and nicely done, with lovely humorous bits as usual. But did you follow your own advice and fix your own MK3?
I'm not sure most MK3 owners even have the problem, at least not the same problem you had whereby the print quality got significantly worse over time. Degradation over time is the aspect of the problem that almost everyone discussing it has overlooked. I was hoping you could figure out WTF happened to your own printer and fix it. Theoretically, this shouldn't require anything more than adjusting, tightening, or reprinting parts or replacing them with new identical ones. Changes to the design and/or the use of non-Prusa parts should not be required. Then, if you are so inclined, you can mod your MK3 to your heart's content. If you do, it would be most helpful if you could demonstrate that the mods are actual and not just theoretical improvements. That's a lot to ask, of course, but you seem to have the expertise to do it.
Hi ,Thanks for the video. It was great and helpful, Which is the best in terms of accuracy and quality prusa i3 Mk3 or CY-10 mini? Thnx
Hi I have tested Prusa i3 Mk3s, Cr-10 mini and the ender 3 pro.. Ender 3 pro (or even Ender 3) is the best in terms of quality IMHO. Way cheaper and easier to use too.
@@ashwinmohan4503 lmao........what rubbish
Prusa posted a detailed explanation of the problem a few days ago in their update. The biggest contributing factor to the 602 issue is improper assembly of the kit, and I fully agree. Yes, you may have gotten all the parts in the right places but that doesn't ensure success. I found the balls in the linear bearings need to be lubricated with light machine oil, and the bond tech gear roller bearings greased liberally before installation. The linear bearings are VERY sensitive to compression and will start to make noise even from the pressure of your fingers squeezing them. So when installing you need to make sure the clamping parts are not too tight--especially the U-bolts on the Y axis. When properly assembled, there should be very little noise and only light pressure required to slide the table and X-axis. On mine, the X-axis bound up after 1 week of use. I disassembled the x-axis carriage and oiled the balls inside the linear bearings. When reassembling I stopped tightening the X axis carriage screws as soon as I noticed the linear bearings starting to drag. I do not have the 602 problem.
What do you expect? particularly if you're new to 3D printing. A customer is going to follow the assembly manual and hopefully do the best job possible. It's good that you know this but unless it's in the instructions, how are we supposed to know?
Lol. My 3 other 3D printers do not exhibit the same results. Pretty sure I assembled my kit right. There is a number of issues still!
@Tyler. Exactely - I have also 6 printer kits but all work proper. My MK3 worked perfect but you simple see the low quality when it comes to the Prusa 3d printers after a month of using. Really for that money you should get much better stuff, it is just a shame. Got same time my Tevo Tornado and compared to the MK3 it is just working and not needed me multiple hours of fixing some shit.
Unfortunately I think i have the same issues @Sicha Dutrottel. Out of the box firmware worked awesome! After a couple months I decided to update the firmware because WHY NOT? Normally it should benefit the machine. Not In this case. With the other issues like crap bearings, faulty filament sensor that never worked, and inconsistent extrusions, etc, etc, My 380$ "Analog" Wanhao is more reliable. ;P
Prusa's blog writer claims the problem is NOT related to the printer's mechanics and that it is present in ALL FDM printers. Then he said, they were working on the problem... These are Prusa's claims, not mine.
Love your work! It’s like watching The Goodies talk about 3D Printing, slightly. Informative and funny.
Just wondering, could adding a few lock-washers help with the loosening screws?
I know that you've probably long since solved the problem, but Loctite 425 is supposed to be safe for metal-plastic fastener mates. This should prevent loosening due to vibration and allow for future readjustment if needed. Superglue can also be a possible option.
Thanks for sharing :-)
if you replace the stock ldo motors with moon motors, this also gets rid of artifacts and makes your prints look a lot better.
Very informative....I really wanted one of these printers and all the "reviews" raved (and drooled all over Josef) about how awesome they were. Now I keep seeing issue after issue with them. Hopefully this is resolved
I fell for how awesome they look, I got two weeks of excellence from a kit and now it's next to useless. I can't even print a successful first layer or calibration square. At the end of the day it's how they manage their customers that matters, that and all of the unresolved issues on their forum.
I think the heatwave has caused quite a few problems for people too.
I haven't really seen Prusa want to talk about these issues in other videos he is in. It certainly turned me off from wanting to buy one.
Try looking at this for example: shop.prusa3d.com/forum/general-discussion-announcements-and-releases-f61/wire-fatigue-from-spiral-wrap-burnt-cables-t18155.html I observed the beginnings of that myself.
Nice video I have mk3 and I am having issue with Abs print, " heating done massage" it stop the print and filament eject message is displayed than I have loads the filament again, it works and stops again, the same file in pla setting works perfect, any suggestion
I'd contact prusa support, they're fantastic.
What system do you recommend for a technically minded tinkerer to get in the UK that might want to mod the printer? Prusa seem way too expensive. I'm thinking a clone. Can you give a specific link please as I see massive variations even between exactly the same clone models
John Michael Stock If you watch other 3d printing KZhead creators, there were very substantial complaints about Chinese manufacturers changing the product without calling it something else, effectively beta testing on their customers. So be aware of those.
One thing I have noticed on my MK3 is the extruder stepper motor runs hotter than my old MK2. The heat is transferred to the Bondtech gears, so I have to wonder if that is pre heating the filament as well. Seems like they should reduce the current to the stepper. Non of the other steppers get hot or even warm only the extruder stepper.
Robothut they actually have reduced current. Any lower and the extruder stepper starts skipping. Kinda bad design imo.
Yea that's what I thought too. But they run hotter so go figure. I see some guys changing the design so the stepper on the extruder drives the filament gears with a belt, the belt keeps the heat of the stepper from heating up the gears. Kind of crazy. But it all works for the most part, so it is still amazing. I guess the stepper current can be set in the firmware so I think they change it with the updates we get. They may have jacked it up in the last updates.
they do but no idea why. it could just be more friction/teeth with the bondtech or the heatbreak since it's not 100% the same as the e3d. it's the mmu version.
Robothut j
I had a similar Problem with my turnigy fabrikator. Adding a cooling fan to the extruder motor has greatly improved print quality
hi, thanks a lot for the video. At 03:56 you mentioned about swapping the stock bearing with with high quality ones as something from ? Sorry, I couldn't get the name correctly. Could you please give a link to the bearings or the name? Cheers.
He said “Misumi” If you got your mk3 recently (mk3s+ and maybe mk3s) then your printer already has misumi bearings
@@stc_ Thanks for clearing my doubt. I got mine around Sep 2020 so perhaps it might have Misumi as you mentioned. Thanks once again.
@@rw4833 if you flip your printer over you will see “misumi” written on the bearings
@@stc_ Fantastic, thanks for the tip. I will check that out.
i found my hotend wasn't loose per-se but it did "twist" really easily. so even with like 10 zipties on the wires the movement of the printer could still twist it. bad design. needed to by snug. regarding the fan, they should of used the 20mm, bit more static pressure. unfortunately it was probably too big for their "design".
I think you'll find. 1. You were unlucky with the twistyness, mine had practically 0. 2. If you had a larger fan, too much air would probably be pushed onto the hotend which you definitely don't want. Also, I'm pretty sure 99.99% of people don't have inconsistent extrusion issues due to lack of proper cold end cooling.
Was there ever a real resolution to issue 602? As helpful as the siggestions in the video are, they do not fix the issue.
I have never ever had a z problem , a lot of people seem to but i dont really get it and often it is not a z issue but a under or over extrusion. The z azis moves so little, so slow that it is very seldom the z and somethkng else instead masking itself as a z issue.
Awesome video. For those wondering, the ceramic tape mentioned is just the usual yellow and white heatblock insulation you see many printers shipping with, usually attached with kapton tape. You can either wrap the entire block, or do a small piece between the block and the heatsink (the latter method being preferred by those using socks, as they don't cover the top of the block). Not a lot of testing went into if this helps, but we figure it can't hurt anything. R3 parts and better fan is definitely the ticket to the cooling problem. A few other mentions that may help: 1. The extruder head is cantilevered off the rods, meaning they are usually in a state of being flexed a little and letting the head sag downward. It's very easy to lift the head a touch from this position, and that's exactly what the filament does when feeding from the spool or bearing guide. A reverse bowden prevents this entirely, and will prevent some fat/narrow layers being caused by the nozzle being lifted and let back down repeatedly. There's a couple designs for this on Thingiverse, and once you've used it you wonder why all machines don't have reverse bowdens. 2. Bad filament can cause many issues. This was a common source of confusion during the 602 ticket, with many people who were experiencing filament diameter artifacts thinking it was 602. Spend a few extra bucks trying out filament with higher tolerances, you may be surprised at just how much it helps. 3. For the more adventurous of you, upgrade your machine to JLTX's wonderful Skelestruder ( prusaowners.com/wiki/index.php?title=Skelestruder_by_jltx ) or some other geared system. The gearing helps to mask some of the stepper moire problems while giving numerous other benefits. The biggest thing to be gleaned here is that this isn't a single issue, it's many, many issues combining to make absolutely shitty looking prints.
is there a place we could find A/B comparisons of what the Skelestruder does to mitigate this issue? I am curious if is simply a matter of having more torque available so that the extrusion distance is more rigid and correct compared to the command. I think it would be really tough to compare because by the time a person converts the machine there are so many changes it's difficult to sort out a single variable.
Don't quote me on this, but I think Jltx (the Skelestruder designer) has stated that Skele wasn't built to correct 602 and doesn't want to imply that it is a fix or that he's put any 602 specific work into the extruder (so I wouldn't expect comparison shots from that source, maybe from a user who upgraded). I think he wants that beast advertised as an all around upgrade anyone would want, vs being "the fix" for 602. :) I agree with you that there are too many variables changing in that situation to really get anywhere towards solving 602. If that was your end goal, I would say swap to a geared stepper on the stock assembly somehow. That seems to be the biggest change going to Skele. Also, it's been proven many times over by imrija in the Discord that this isn't simply one issue or malfunction. This is many artifacts from many sources at once, and they have to be knocked out one at a time. You may swap a part and discover you've knocked out moire, but still have bulging layers. You may get rid of layer bulges but still have horrible Y ringing due to rubber feet. The list goes on and on and on...
Really interesting & informative video. Completely puts me off buying this printer partly because of the inclusion STILL of 3D printed "Rep-rap" stylee printed parts. I mean seriously? Key axis guide rods sat in printed parts? Are you kidding? Sheez.. for the price point get rid of those quickly printed 3D maker farm shite and replace it with something more dimensionally accurate and more accurately reproducable. Crucial nuts in 3D printed parts? the list just goes on. Other mechanical stuff secured with tie wraps?? What? This thing, as is, is a hotch potch. Prusa. Drop the daft cursory nod to the origins of "rep rap" Why bother with those roots? It's the past and the original rep-rap was truly rubbish. Oh! sold to a hungry media as a machine that "could replicate itself" which they eagerly took up and run with, and which of course, as we all know, is a pile of bullocks.
Cool vid, thanks for sharing. Do you have any suggestions for a fan-upgrade?
5:22 - This is still about printers, right?
lol
I was having under-extrusion issues that were, in-part, caused by the idle gear axle moving out of alignment. I modified the axle supports in the Idler Door part to fix the issue. It eliminates any possibility of the axle shifting during printing by ensuring the axle is always in the optimal "non-clicking" position. Most importantly it allows for even extrusion to provide good prints. This is the remixed part: www.thingiverse.com/thing:3504682
Also, it looks like one of the reasons that slowing a print down and turning up the temperature can also reduce or eliminate clicking/under-extrusion issues is that the “custom” PRUSA e3D heatbreak changes bore size from 2.2 to 2 close to the heat zone. Proto-Pasta did a bunch of testing that showed that this was probably the major cause of the click (stepper slips) and under-extrusion sometimes leading to jams. kzhead.info/sun/l7eJgNaaZ3xudpE/bejne.html If this is your problem, the solution is to by the non PRUSA custom e3D heatbreak www.printedsolid.com/products/e3d-v6-heatbreak-single
the reason why the soft rubber feet create more artifacts is because it keeps all the printers vibrations locked to the frame which means less mass resonance has to move and also a larger range of resonance that moves up into the higher frequencies as well as the normal lower frequencies. When you add hard feet you essentially increase the mass of the printer to the printer plus the mass of the table/other it is connected to divided by the transfer efficiency of the connection from the printer to the table/other. The best way to reduce noise and reduce vibrations is to attach the printer to something designed to absorb vibrations/sound or something that works well to do that in general. MDF and particle board is a good example of this. Then underneath that you can add softer feet if you like. The more layers you have the more the vibrations into a resonating object can be reduced. For my printer i have homemade dense silicon flat strips that are in between a particle board and my printer. Then below the particle board is another set of silicon feet. I get almost zero vibration through to my table which is partially metal and before doing this had terrible resonance. I hear a cement slab also works quite well.
Are you attaching / solid mounting the printer to the silicon, mdf, particle board, or only setting the printer on the surface? Thanks
Yeah that's all go out and get a concrete slab to make our overpriced printer make quality prints sounds like an apple idea.
I have my Anet A8 firmly fastened to a stabilized platform. No vibration issues.
It disappears in vase mode I think it's a extrusion problem
Yes the cooling fan. The fan shouldn't be mounted on the side. When going left it's pushing the air in, but if you left it will suck the air out. Hence you have a problem.
Zee axis?
For some reason, I never see this issue on my mk3 and I don't have the slightest idea why. I wish I knew so maybe it would help with the solution
Jonathan, it may be that you have not printed the kind of parts or with the kind of filament that shows the artifact. This truly is quite a minor thing for many prints and is a non-issue for those that I print. If you are using a filament that can show it and printing flat faces and if you shine a flashlight across the face at a low angle likely you will see it. If not, please pop by the forum and look at the test prints that are being made. Have you run the linearity correction recommended by Prusa? Did you see it there? I feel your printer probably will act the same as the others but honestly - we don't know til root cause is found. : )
I'm assuming since it's been 2 years since and theirs the mk3s+ model out they've probably fixed a majority of these minor issues? Can anyone confirm?
Most of the issues are fixed by now IIRC. The 45 degree fan I think also helped a lot
which one will you suggested, Prusa I3 MK3 or LULZBOT MINI 2
Get a Tevo Tornado and add an autolevel sensor... I own 6 printers incl. the MK3 and all kits from China are better when it comes to quality than the MK3. It will just cost you useless hours of fixing cheap things while all others print exactely like it should be.
Absolutely excelent video. Using it for my MK2 and its helping a lot.
I noticed one of the forum members on Prusa's site called this artifact "basket-weave" and I think it is a good description. I have the MK3 and I can find this on some prints when I look for it. It doesn't bother me much and for my purposes has no affect at all on overall print quality (my prints are used as mechanical parts, not show pieces). I think it is a very interesting problem to study and when we do figure it out maybe it will advance the industry a little bit.
There are many other machines out there that do not have this problem and are up to three times cheaper than this machine I think it's more of an apple mentality. People think that the owner is some kind of Genius when in reality this printer is nothing special and should have already had all of the bugs worked out before they were shipped.
When money is no issue. Wich bearings for the head would you recommend?
Sounds to me like money is already an issue you paid 3 times as much for this machine it should have better quality Parts built into it.
Get to it!
iFixit FTW!
has anyone really looked at the threaded z axis rods? they wobble quite a bit because the top end is virtually unsupported.
Welp, so much for my theory! :D
What bearings did he recommend?
Matt-MST watch the video
Nice update. I am still new to 3D printing and bought a cheap China printer to see how I would like 3D printing. I found I do enjoy it but would have issues that were a hassle to deal with. I was ready to get the MK3 from all the reviews I read when the 602 issue came up. After reading many articles, forums and blog posts it seems there are 2 camps. One camp says this really is not an issue and you can tweak the printer and get good if not better prints than other more expensive printers. Then there is the other camp of people that say it is worse than a $200 printer, this is a real issue and needs addressed, Prusa is ignoring it as it is not an issue on the MK2. This makes it very confusing as to what to do. Should I get an MK2 for now and wait to see how this all works out? Are the ones that say it is a real issue and provide prints showing the $200 printer did a better job than the MK3 just have a bad MK3 or some issue in the build of it? Is it really that the MK3 was not built correctly by the person complaining about it? Has anyone else had these thoughts and what direction did you go?
this particular thing (602) is one of the most minor things that could go wrong with a printer. I definitely would not be using it as a criteria. I like my MK3 as it is a nice improvement over the Printrbot that I had before this. The issue can be what your expectations are. I use my printer almost daily and often make prototypes or test parts for work such as enclosures for electronics. I print in PLA, PETG and recently have done some PC-max. The best thing about this printer is the heated bed, magnetic plate and PEI sticker. The system is killer. some caveats - 1) I have never used the filament sensor, I left it unplugged from the start. 2) I run straight from the sd card always. 3) I have not placed my printer inside a heated enclosure. my point is that I have avoided some problems that others have seen by avoiding the use case that has led to some bugs. If you can stand waiting a few months or longer, sit tight and look for the 'next' thing. The market should move toward using either all 32 bit high performance micros or even possibly Klipper or similar to offload motion calculations. The current boards and software are reaching their limits imo.
Very good points. I did think about waiting on MK3s if that is on their roadmap. But then you have more teething of new tech. So maybe an MK3 now and in a few years watch the new tech get refined.
Thanks for reading my comment and replying. That positive feedback helps.
My first printer was worse than Chinese quality. I would recommend the MK3, but buy it pre-built for the warranty and ability to ship the printer back to Prusa. Being honest, there're several design flaws such as wire fatigue and flexible braid sagging that I personally saw developing. Two parts of the printer have just been revised but not put into production. It might help to wait a couple of months.
same here - a total disaster when it comes to QC. Get a chinese kit and you get probably way better parts.
Holy shit, Prusa is the new Tarantula.
Fully assembled the prusa mk3 costs 1000€ and the kit costs 750€ and i am expected to invest more to get a working printer? You´d be probably be better off by buying a cheap ender 3 and the duet wifi board. A friend of mine has both and his opinion is that the ender 3 with all the upgrades is better than the prusa mk3 and cheaper
No you're not expected to invest anything else in the printer and neither do you need to. Thats not at all the point of the video. Ender 3 is ok, but its really not as good as a mk3.
@@Vector3DP That´s bullshit. I even have a Zaribo for 1600€ (upgraded Prusa) and the same bullshit. My Creality CR-10 prints much much better results. Prusa or Zaribo. Bullshit stuff for much money! Or let´s talk about the MMU2S ... what a buggy piece of shit! I hate it! All the time issues issues issues .... 1600€ vs ~ 400€ incl upgrades for my CR-10 = CR-10 is the winner! It prints and prints without any issues!
Yes yes yes! Bearings and rods? Hmm not so sure, loose parts definitely, idler gear no teeth definitely, rubber feet also. Actually I think its an engineering/customer engagement issue. Thomas hasn't realised that poor quality doesnt bode well for company growth. I manufacture products too. Our goal is to have a totally satisfied customer. Sometimes we make zero profit to achieve this but thats cheaper than cost to brand. There are full metal x axis mk2 assemblies. I am going to try this 1st. It solves almost all of the design issues mentioned here. Also think Prusa could do better with a customer advocacy engagement model. They have 1000s of customers who would offer advice input and experiential feedback which is free R&D and is reliable and field tested. Grow slower and protect your brand sales are critical but not at the expense of customer experience. Out of 10 I rate that a 4/10 build quality is a 6/10 overall experience is 6.5 it should be 8 or 9 or even 10 and this is achieved by outcome FOR THE CUSTOMER.
I'm very happy with mine. Great prints, convenient, great customer service in my experience (DOA filament sensor, replacement shipped to .au and received in 4 days). I also notice from their recent "spring update" email that they appear to be spending $$ to find a solution for those with problems, and are working on general improvements like the improved extruder and cooling as Adam says.
@@KieranShort it seems to me that since you pay like 3 times as much for one of these compared to similar models they should have work the bugs out before then mail them it's not like it's a new company.
I would even be more critical of them since you pay a lot more for one of these. a lot of the problems you have with these machines have been around for a long time they just don't seem to get fixed.
I'm curious to see/hear if anyone has tried swapping the x carriage between a mk2 and mk3. That would narrow things down to whether it's a hardware or firmware effect. Linear advance on the firmware side or something to do with the extruder (not enough torque from weaker drivers, not quite as good cooling, etc.) seem like the most likely culprits to me. I know I keep saying this like a broken record but to me it looks like the spool is tugging on the extruder as it moves, I've had similar looking results on one of my printers because of this issue and it lines up with not enough torque on the extruder. That being said, most of the things mentioned seem to be general print quality things, like slop/flex/stretch and don't in my mind line up with what the issue specifically looks like.
That was really a nice explanation on how to improve your print quality. I can also add to what you have said two points that are difficult to solve because it is inherent to the design of the all metal hot end and the properties of PLA. To have the most consistent extrusion it is important that all dimensions are tight tolerances, and that includes: filament diameter and the cylinder path of the filament while going through the hotend. If we theoretically had a quality filament then the only problem left is the tightness of the metal throat, which usually is not as good compared to a regular Teflon type hot end. I always experienced better consistency when using teflon hot end like the : J-head, especially with PLA compared to all metal hotend. The other problem is because of the nature of the molten PLA itself. Molten PLA fluid behave similarly to a Non-Newtonian fluid. A good example of non-Newtonian fluid is : water and starch mix. Non-Newtonian fluids viscosity depends on the force applied to the liquid . When molten PLA is suddenly retracted and pushed with force back into the nozzle it forms a semi-solid state that counter the gears pushing force and slowly let go and eventually the solid state turns back into regular liquid. The trick to avoid this inconsistency in the fluid behaviour is: to find the sweet spot of the retraction speed that is not enough to produce the solid state behaviour, also, to have a throat and the nozzle both plated with the least friction material like: Nickle.
interesting. If this phenomenon is leading to the basket-weave inconsistent extrusion artifact then doing spiral vase mode should eliminate it. I assume in spiral vase mode there will not be any retraction after the first few layers.
My observation might have some weight to it especially based on the feed back from the community the MK2 which has a less force extruder (Greg's Wade style) did not exhibit the same problem. And yes, using the vase mode or the least number of retractions has significantly improved the quality of the surface. Also, using less viscose filaments that has a higher Glass Transition temperature (Tg) like: ABS & PETG would give better results because of the lower viscosity at higher temperature unlike PLA. My solution for the problem was to replace the current Extruder with the Titan from E3D. The Titan might have less torque than the Bondtech extruder but it has a much refined control on push force.
gotta boo then gotta boo =D haha made my day ...very good vid. thx
15:55 "By upgrading to a fan that's a little more ... "air-pushy" "... LOL
He meant one with a higher static-pressure spec.
The best solution I think is to upgrade to machine that cost about 3 times less than this 1 to get a better machine.
What appears to be a shocking number of quality issues and general poor design in certain areas on a 3D printer that reports to be the best in the market for the price. As a retired senior mechanical/electrical engineer, I never like to hear the words poor quality bearings/rods and parts not being strong enough.....Emm perhaps the mk4 will address the issues????
I wonder if anyone has considered a peltier based phase change cooling/heating system for a hot end assembly. I've run one of these on a highly overclocked PC system in the past, and one side of the plate will be so cold you have to worry about condensation, and the other side is nuclear hot. Seems like with a little tuning, this could be utilized to create a hard transition point for the melt zone. If going that far is too difficult, perhaps a better heat sync design like the Noctua processor coolers that use proper heat pipes and cooling fins, rather than just some milled fins.
Glad I passed on this machine !
Meh its a pretty good printer if set up right. I think its a winner
Great video. The Prusa i3 is a premium priced machine with non premium parts :)
they were part of RepRap which was trying to make a printer with mostly 3d printable parts. Not premium, but definitely efficient
7:41 Hahahahaha!!!
When you guys are so humble these you probably should use some Loctite
but...but I thought the Brown colour from Noctua added +5 airflow!
I bought 4 of them just for that! damn it!
It was never about flow amount. The Noctua should have all the flow it needs, it's 20% better than the 30mm fan that E3D supplies with their products. E3D doesn't say anything about static pressure but Idon't see where it's worse than the 30mm. The Mk3 extruder body might have more constriction on the flow though.
E3D's own 30mm fan also has to push air through the same constricting fins so I'm not seeing where that's really the problem.
The new ones come with a black fan, does that add to the bonus?
JeffDM No it isn't. It's got a third to a half the static pressure of the stock one, but slightly more airflow. Loud fans aren't loud for no reason. The E3D one is a XYJ12B3010H. And the Noctua one is a NF-A4x10, which I think is the 5v model.
Just got my Mk3 kit and when I built it I put removable thread lock on all the screws. After building to many crap printers, that's always an issue that annoys me.
Okay - the ptfe tube fix is a good call out...
Okay so, this may be completely stupid, and I don't pretend to know more than the dozens of people who have researched this, but you guys have tried disabling the extrusion sensor right? I'm just thinking if it's not calibrated properly, or it's getting misreads, it might cause this issue by speeding up and slowing down the extruder to compensate for filament size. I'm sure someone has tested it, I'm mostly just curious. I've only heard of this issue showing up on the Mk3, and it has a part specifically to compensate for this sort of thing, so it just seems like it might be malfunctioning. I could also see it trying to compensate for differences in the filament size at the location of the sensor, and not compensating for the distance between the sensor and the hotend, so it's got thin filament in the hotend but thick filament in the cold end. But again, this is me who doesn't own a Mk3, spitballing ideas.
it's a good thought. However...I have never used the extrusion sensor even from day 1. My printer works great but when I look for it, I definitely have this basket-weave artifact on some of the prints. I've done some experiments looking for root cause and so far have turned up empty. I know it is not caused by my (unplugged) filament sensor : )
Zed-axis* 😉
Sorry, but these are not 13 changes but 13 checking steps.
damn, i think an video "improve Anet A8 print quality" would deliver mostly the same suggestions. But its way cheaper :(
So to summarize: The solution is to actually inspect and maintain your equipment more periodically.......
I agree, maintenance is important. One thing I would like to add is that part of creating a quality product is designing it to be low-maintenance. The less maintenance required to keep the product functioning like new, the better.
Maybe if they built the quality into the overpriced machine, maintenance wouldn't be such an issue
umm, you are aware this is one of the cheapest printers on the market?
Hugo Ortega more periodically??
@@hugoortega195 never heard of an ender 3? lol prusa is 5x the price here in aus.
wow! the bestprinter in the world has so much poblems? for only 900 dollars?
At the point you recoment selfprinted Linear bearings i stop the Video to write this. I got agressive when i heard PLA "SELFMADE BEARINGS" this is the worse thing you can do to your printer! Even IGUS Bearings wich are way more precise as selfprinted ones- DO NOT WORK! They need at least 45Nm to be tighten and this is too much for your steppermotors. I do Not have any of this issues even They're acceptable for a low Price printer. You can improove even the best System. So i dont think youre talking about something significant New improovements
FFS - got to 4:22 before the waffling on made me stop - get to the point
It is really funny a 1000 USD printer is delivering similar prints as a 200 one and people are still trashing the cheaper one... I was properly geared up to getting the prusa but i have yet not find one selling point that is worth the difference of 800 USD...
I agree. Get the ender 3 or 5 and buy the duet 3d wifi board for 180$ and its as good as the prusa. You can add all the functions to the ender 3,5 and you will still be paying less than for the prusa kit
Don't like the video, it is just stating a whole list of fixes posted by other people and to me they don't all seem correct/right. And there is just no comparison or proof actions fix it. I'm really curious about the rubber feet for example. I think it also fixes some artifacts. But also statements about the bearings, or the fan. I find you state everything in a way that it is all true and all is better then the origenal.The video might be a bit better to watch if you state these all are suggestions from other people that might help.
Would it be too much to ask to see objective, side-by-side prints from different printers using the same settings with the same filament in the same light? Every such comparison shows the Mk3 looking "different" but not necessarily "worse". The KZheadrs with a stable of printers could do us all a favor and print off a 1 perimeter 20mm hollow cube on each with silver filament, stack them, and film them in the same light from various angles. There is definitely an issue, and Prusa is working on it, but prints are still good.
Maybe you should do your own testing instead of asking smart people who paid a lot less money for their machines.
Meant it in the general sense. If you're going to compare things, actually compare them. A video that doesn't show before and after comparison images is less than helpful. If one's better, by all means, let's see it. This is feedback for somebody asking for subscribes and likes. He did ask for comments.
cooltek fans- cheap as silent as the noctua but produces a better airflow!
7.45 😂
Boobies!
IgnoreMyChan go back to youtube kids
@@AquaticSCP go back to KZhead adults bruh smh
Xedric playz you mean normal youtube I’m already there (I know it’s a joke I just ignore terrible ones)
@@AquaticSCP No I mean KZhead adults because you're so smart
The machine sometimes crash and continues printing in wrong position
i dont even own a prusa mk3
do u have fake hair? at the top
Haha, no lol. what a strange question.
So buy a £700 printer and replace 1/3 of it... Ok
This more of a video not by one lol..
lol remove the printer and talk about the filament shooting out. sounds like an Ad for a constipation pill.
Buy a 1000€ printer and change all the parts. ok. Will just not buy that printer. Thanks for the video though.
Still the best printer in this price range my a long way.
@@Vector3DP Vector 3D Thanks for the answer. Im looking for a printer and was interested about that Prusa for long time. But recently I started to read the reviews, and they are a lot (a lot) of bad reviews. For the failures of some parts (out of the box) that Prusa doesn't want to take in warranty, for the bad reliability when printing very long projet (50% of failure for some people), and for the bad team support. I checked many videos, and the quality does not seem to be better than a 300€ printer (?) What about the low quality of the orange parts of the printer? Some people will tell me that its just visual aspect, but it doesn't make me feel comfortable... Same with the LCD screen that look 20 years old. I will not speak about the wifi missing. Why not to go with a Creality Ender 3 pro, add 50€ of component to have quiet printer, auto leveling, and you will have the same as a Prusa? Even the quality of the printer, LCD seems better with creality. I never saw a Prusa, neither a Creality, so my opinion is certainly not accurate. But still, we are speaking about a price 3 times bigger than a Creality? It should be out of the box without any problem! I am really struggling, because I would be ready to pay 1000€, even more, and I though this one was a good choice. But the more review I watch, the more complain I see about the quality of the printer, the quality of the printed parts, and the quality of Prusa support. That makes me feel not very confident... :) Edit: as i understood, the 2,5 version was perfect, then it started to downgrad with the 3, and now the 3s. Will have to dig a bit more on KZhead to find out if it worth to buy it, or if I will stick with an Ender 3
These machines have so many problems especially when compared to machines costing so much less.
Solution don't buy a MK3, any of the clones can be tweaked to produce the same quality of prints or better than a Prusa that is the sad truth, in fact many of them perform better out of the box than the MK3 even after all of the "fixes" it had. A $900 "kit" should contain machined and reinforced injection molded parts only, paying this much for flimsy 3D printed parts is laughable, seriously I understand that the original Prusa was designed to be assembled from nearly anything but paying $900 for something that still uses zip ties to fix it's axis rails should be a big nope to anyone. The Prusa is now all about reliability and precision but the build quality of it is anything but that, they could've made the Prusa MK3 fully extruded easily at this price range and even good extrusions like those for examples used by Cetus which are 100 micron dimensional accuracy extrusions aren't expensive these days. They also could've upgraded it with a much better interface and built-in connectivity, the rotary scroll wheel Marlin interface is showing it's limits a new touch interface should be used and wifi/ethernet should be a standard for a printer at this price.
So how much did you pay for one of these machines seems to me you since you paid 3 times more then what other companies charge for their similar machines that they should have been worked out before they shipped. It's not like it's a new company. I see it's a fault other people believing in the owner thinking he some genius kind of sounds like an Apple product to me.
why would you need to do the changes prusha is infallable he can do no wrong
best change? buy something cheaper, it hurts less when you throw it through a window 😁😁
To receive a printer with so many defects, I pay a lot less and I get a geeetech prusa i3 clone
this video was uploaded 4 years ago
so basically everything is shit on my new printer
pla bushings are rubbish, they wear out fast!