[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/diy/ - Do It Yourself


View post   

File: 423 KB, 1800x1800, collage.jpg [View same] [iqdb] [saucenao] [google]
1732568 No.1732568 [Reply] [Original]

I have no manifold and I must scream edition


Old Thread >>1726943

All the info you need about 3D-printing: https://pastebin.com/7Sb4TVdy

>Need help with prints? Go to:
https://www.simplify3d.com/support/print-quality-troubleshooting/

If that doesn't help you solve your print problems, please post:
>A picture of the failed part
>Printer make & model
>Filament type/brand
>Bed & extruder temperature
>Print speed

>What printer should I buy? [Last updated 7-9-2019]
Under 200 USD: Creality Ender 3
Under 500 USD: Creality CR-10
Under 1000 USD: Prusa i3 (Mk2 or Mk3)
Over 1000 USD: Lulzbot or Ultimaker
Buyer beware: some chinkshit clones are garbage. Some can be genuinely good, though.
Instead of buying a new printer, you could consider building your own: https://reprap.org/wiki/

>Where can I get free things to print?
https://www.thingiverse.com/
https://grabcad.com/
https://google.com/

>What CAD software should I use?
Solidworks, Inventor, AutoCAD etc. all work, but Blender and Fusion 360 are free:
https://www.blender.org/
https://www.autodesk.com/products/fusion-360/

>> No.1732581

Has anyone tried modding the ender 5 for independent extruder?

>> No.1732589

https://drive.google.com/folderview?id=1-178_r7mWNaMLkU9F8PgMiWPv_1l91gs

German and Englisch

>> No.1732596

>>1732581
Independent?

Direct drive you mean?

>> No.1732605

>>1732568
Blessed is he who puts /3DPG/ in the title.

Also surprised to see my print in the montage.

>> No.1732614

Any suggestions? Just started a new job and we're dealing with 3D printers one of my tasks was researching best techniques for best results and I know there's lots of parameters besides temperature. I'm very new to this here's some information I've typed up so far with sourcesSimplify3d.com
https://www.simplify3d.com/support/materials-guide/abs/
Abs requires temperture control not only on bed and extruder but the atmosphere within the printer. Ventilation is required. Ideal bed temp is 110°. Extruder temp 220 - 250°c. It is recomended to heat the extruder 10-20° higher for the first few layers to prevent seperation. Larger pieces may require an enclosure to maintain heat and prevent drafts. Consider ABS slurry (abs mixed with acetone) to improve bed adhesion as well as Kapton tape. Also if possible disable fan cooling, but keep in mind overheating is possible.
3dprinting-blog.com
https://3dprinting-blog.com/tag/3d-printing-parameters/
Infill: 20-30% should be enough since there is an internal lattice. This will save on materials, but reduce durability.
all3dp.com
https://all3dp.com/2/3d-print-stringing-easy-ways-to-prevent-it/
Enable Retraction to prevent stringing
airwolf3d.com
https://airwolf3d.com/comparing-bowden-vs-direct-drive-3d-printer/
Bowden vs Direct Drive
Bowden has less moving mass and will cause less imperfections from motion. Direct drive is more responsive, and requires less push from the extruder, less chance of filiment slipping, and a smaller stepper motor. Thus, direct drive is prefered unless speed is required.

>> No.1732615

>>1732605
I'm jealous, my penis+tentacle got snubbed. Damn blue boards.

>> No.1732620

>>1732614
There are usually two fans on the hotend, a part cooling fan, and the hotend cooling fan
in general, for printing ABS you want to turn off the part cooling fan but leave the hotend fan running since the hotends are usually designed to require a cooling fan running at all times

And as for direct vs bowden, basically the difference is that there is little to no "springiness" in a direct drive compared to a bowden system and what that means is basically that you get a better control of exactly how much filament you are feeding the nozzle and when.

As for slicer settings, if you have a quality machine you probably wont need to change the factory slicer presets at all, i've got a Prusa MK2 and it produces really nice results with the default settings on Prusaslicer, the only thing i tweak from time to time is the infill ratio and wall thickness.

>> No.1732621

>>1732596
Independent dual extruder is when you have two separate hotend on the x axis

>> No.1732624

Though i do prefer Fusion 360 Autodesk hands out education licenses to their other software freely, so if ya want a professional cad, you can get Inventor for free pretty much legally straight from Autocad

>> No.1732625

>>1732620
Thanks

>> No.1732653

is there suppose to be a new model of the elegoo mars coming out next year?
been looking for a resin printer mainly for my D&D stuff and for small statues. im not in a huge rush to buy one but i dont want to get one only for a new version to come out soon after.

alternatively, how is the photon resin printer? seems its listed at a slightly higher price point but i know that usually higher price doesnt necessary mean a better product.

>> No.1732656

>>1732653

The higher price of the photon likely has to do with there being more sheet metal in its construction. The mars is a newer, more cost-optimized design that uses a top cover instead of a door.

>> No.1732666

>>1732653
The newer photon S is supposed to be bad, all plastic and flimsy, the old one is roughly comparable to the mars. There is a mars pro coming out in a couple months with some quality of life improvements, but nothing that will actually improve print quality besides the linear rails. If you don't mind waiting for the mars pro it will be a better printer, but not drastically so. Elegoo is a better company than anycubic, so I'd pick a mars over a photon given a choice, but they produce almost identical prints in the end.

>> No.1732668

>>1732666
gotcha, thanks. maybe it was the photon that i heard getting the new version and not the mars.
looks like as of now, im gonna stick with the mars.

>> No.1732672

I love the Mars, and having a 40w light source over the 25W light in the original Photon makes prints significantly faster.

>> No.1732677

Holy shit. Brand new filament, an aftermarket nozzle and better calibration and I almost have a test print.

32% complete. When should I expect it to fail?

>> No.1732678

>>1732653
Yes, the Mars pro or whatever will basically be similar in concept to the photon S's upgrades. Matrix parallel lights and better rail. If you want a good one like that right now have a look at the epax x1dj. Solid construction with the parallel lights and it's backed by an american store that will help you out. It's expensive though, but also preleveled out of the box so it's as close to a "ready to go" chinese oem resin printer as you'll get

>> No.1732682

>>1732677
yur mum

>> No.1732683

Are there better connectors for 3d printers than the xt30?

>> No.1732685

>>1732677
And failure at 34%, started air printing again.

>> No.1732695

>>1732683
I like the deans T connector

>> No.1732696

>>1732683
soldered

>> No.1732700

>>1732696
I need connectors and the xt30s are soldered anyway

>>1732685
Was looking for something that's crimped and also has a 3 terminal option

>> No.1732714

just read the op, 3d printing doesnt seem too hard to get into. looks like the Creality Ender 3 is the best option to go with as a beginner. any tips before i make the plunge? also i dont think i saw a website for purchasing filament, so as long as it isnt chinkshit i should be good right.

>> No.1732717
File: 241 KB, 1133x554, Capture.png [View same] [iqdb] [saucenao] [google]
1732717

>>1732714
oh and theres a sale for the pro for the same amount as the regular version, this is a no brainer right

>> No.1732718

>>1732714
are you half way confident with CAD modelling parts or at least willing to learn?
because if not 3d printing is a useless gimmick
you can only print what you can design

>> No.1732721

>>1732717
You could probably get the non-Pro for 175$ from a third party reseller, but yeah, the Ender is really good considering how cheap it is. And since it's such a popular printer, the issues it does have all have well-documented solutions. If I were getting my first printer today I'd pick the Ender for what it's worth.

>> No.1732725

>>1732718
>are you half way confident with CAD modelling parts or at least willing to learn?
no, have never touched it. however it looks like a fun hobby that can actually be useful compared to other hobbies. so yes, im interested in learning
>>1732721
if i went with the non pro, what features would i miss out on? though i imagine it doesnt matter to me cause its not like id know the difference anyways.

>> No.1732729

>>1732725
>if i went with the non pro, what features would i miss out on? though i imagine it doesnt matter to me cause its not like id know the difference anyways.

The pro is quieter (don't actually know if this is still true or if they use the same stepper drivers in the standard model now) and the bed is slightly more stable, due to the increased width of the extrusion it rides on. I actually have both, and, really, the pro isn't worth it, unless you plan on running it while you sleep. There just isn't any real difference in performance between them.

Plan on cutting some glass to use as a bed, though. On both of mine, the stock beds are uneven as hell, to the point of effectively halving the print area if you actually expect your model to stick. Others have gotten good ones, but I wouldn't bank on it, myself. Plus, glass is basically unbeatable for PLA, I find. I haven't tried steel though, so take that for what it's worth.

>> No.1732732

>>1732729
watched a video on the ender 3 and the guy was also talking about a glass bed. whats the difference between a glass bed and the stock one? ill probably end up going with the non pro then, i prefer noise to drown out my tinnitus anyway

>> No.1732742

>>1732732
The stock bed is aluminium with a BuildTak clone, it's got really good adhesion but like the other guy says some of the beds are warped. The Pro has a magnetic version, you can pull the material and the print off and then just flex the thing to pop it off, it's more convenient than using a scraper directly on the aluminium. Not worth the extra money just for this feature though, and if you want it you can buy a magnetic PEI sheet for the same cost, which is way better.
Glass is guaranteed flat, and well-tuned and with some hairspray you'll have good adhesion. Most prints come loose on their own once the bed cools. You can't print PETG and similar plastics though, since they bond directly with the glass and can rip bits of it out when you remove the print.
You can get glass cut in five minutes at basically any glasscutters (look for signs like "car window repair" when you drive near a walmart, you'll find dozens), so if I were you I'd start with just ordering the Ender. If your bed does turn out to be warped it's a quick fix to go buy a 220x220 glass pane and four binder clips (used to mount it to the bed), but if you get a good bed the stock buildtak is honestly better. Glass needs to be clean and sprayed regularly, the Ender buildtak you only need to wash off under water if it gets really dusty, otherwise it's basically foolproof.

>> No.1732804
File: 1.59 MB, 2592x1944, DSC03310.jpg [View same] [iqdb] [saucenao] [google]
1732804

Finally figured it out. The extruder tension wasn't tight enough.

>> No.1732918

why do 3d printed things have those horizontal lines?

>> No.1732920

>>1732918
Layers, nearly all prints are sliced into planes and built up one by one
Non-planar slicing is an experimental thing but not at all common yet
>tfw no 5-axis printing

>> No.1732994

>>1732804
This looks like a 3D print from 6 years ago.

>> No.1733080
File: 491 KB, 2025x1215, tete.png [View same] [iqdb] [saucenao] [google]
1733080

Almost done.

>> No.1733111

>>1732568
>want to order https://www.thingiverse.com/thing:2756468 at a local 3d-print shop
>get the reply that its too small and finicky to print
they are using FMD and offer 0.10mm ABS printing, 100% infill, just like the thingiverse states
could i print this with a ender3?

>> No.1733120

>>1733111
Yeah, just make sure your settings are right.

>> No.1733124
File: 19 KB, 480x360, hqdefault.jpg [View same] [iqdb] [saucenao] [google]
1733124

Has anyone tried to build a printer with a robotic arm ? I kinda want to make one for potential 5-axis printing and with a modular head for various functions other than printing.
I'm aware speed wise it will be far behind regular systems but i'm aiming for about 100-200mm/s and a precision of about 0.1-0.05mm on all axis, and a payload of max 25 kg on the end of the arm.

>> No.1733128

>>1733124
I wish you the best of luck with that.

>> No.1733130
File: 2.09 MB, 3008x2000, 1436072163920.jpg [View same] [iqdb] [saucenao] [google]
1733130

>>1733124
>various functions other than printing

>> No.1733132

>>1733130
Every thread you have to push your faggot bullshit onto unrelated fucking topics. Fuck off.

>> No.1733176

>>1733132
Can't blame a fag for being faggy. That's what they're like, have you never seen gay parades?

>> No.1733206

>>1733080
>hundred dollar extruder
>chink aluminum extrusions and rollers
lmaoo

>> No.1733219

>>1733206
extruder is most likely a chink clone

>> No.1733220

>>1733219
oh alright, that makes sense

>> No.1733222

>>1732920
You've piqued my interest. I'll get back with results it might be a while. I may know a genius or two.

>> No.1733230

>>1732994
Yeah, it's my first (complete) print ever, so I know its not great but I'm just happy my printer finished a job. I didn't set it for fine detail or anything, I just wanted the printer to not fail at 58% like my previous 'best' print did. This is all new to me.

>> No.1733234

>>1733124
Here's the major design hurdle I see: the end of your extruder is going to be bouncing around like it's attached to a spring because it's not braced like a normal design. Not to dissuade you from building an arm (could really be useful for automating a print farm for example), but I wouldn't expect precise prints out of one (definitety not at those print speeds).

>> No.1733256

I want to take a black and white image and make it into a 3d print. Is there a software that will just allow me to pull the parts of the pic up to do this?

>> No.1733258
File: 358 KB, 820x1664, IMG_20191212.jpg [View same] [iqdb] [saucenao] [google]
1733258

>>1733206
>chink aluminum extrusions and rollers

not anymore. And nobody is retarded enough to buy an original bondtech (I hope).

>> No.1733262

>>1733234
>going to be bouncing around like it's attached to a spring because it's not braced like a normal design
On a shitty homemade arm, absolutely. Maybe he's planning to use an old industrial arm or something though, those are very rigid and surprisingly fast. I expect you'd still get more ringing than even on shitty printers, but it should work.

>> No.1733265

>>1733256
Yes, the same program/website used to make those lithophanes.
https://all3dp.com/2/how-to-3d-print-your-own-lithophane/

>> No.1733268

>>1733265
Thank you.

>> No.1733282

>>1733258
>linear rails but a chink bondtech
lmao dude

>> No.1733316

>>1733282
Do you even know what you're talking about?

>> No.1733380

>>1733282
Chinese make linear rails too. American manufactured hiwins cost like $200 per half meter new, cheaper surplus but still really expensive. Chinkshit knockoffs are like 50 for a set large enough to build a printer.

>chink bondtech
>implying
Bondtech might as well be cheap chinkshit, most of the diy/reprap serving companies are.

>> No.1733471

>>1733128
Thanks man

>>1733234
>>1733262
Yep, that's one of the things I'm afraid of, while it will be a homemade arm, I still plan to make it rigid with an aluminium core and composite skin. Around 3-4k budget for an arm that can extend to about 1.5m. As I said, payload of about 25kg for the heaviest kind of head ( so for a ~1kg extruder head I think it should be ok, precision and vibration/inertia wise) And about 50Kg for the arm itself, maybe 100kg if needed. It would be close to an industrial one but still done at home, with articulations machined with a CNC.

>> No.1733481

Does the air filter on the Anycubic Photon S work well?

>> No.1733511

Is it safe to use bondo with PETG? I'm working on some moulds and have no fucking idea what I'm doing. Just want to smooth out the surface.

>> No.1733531

>>1733511
People use automotive filler primer before sanding to smooth parts out all the time. I would imagine that bondo would work alright. It takes a lot of time and sanding to get anything nice looking in my experience.

>> No.1733586

>>1733511
Works pretty well, no idea if what you're putting into said molds will stick to the bondo though.

>> No.1733594

>>1733176
All that's required is the smallest modicum of restraint. This is a blue board, he can post his filth fucking elsewhere.

>> No.1733605

>>1733471
Good luck then. Just bear in mind that rigidity is #1, much more so on an arm than on a regular printer. There can be absolutely no play in the mechanism. On my printer I can wobble the head about half a millimetre by pushing on it when the steppers are locked. Your arm must be more rigid than that, because any play in any joint is going to amplify at the head. If you were using, say, an ABB IRB1100 I would have some confidence, but homemade stuff, not really.

>> No.1733606

>>1733594
I'm telling you, these people are physically incapable of exhibiting restraint. They're attention whores, it's literally what they do. No matter how tolerant you are of them they're going to demand more and become more outrageous.

>> No.1733607

>>1733111
Get it SLA printed instead, it'll probably last longer and will definitely work better

>> No.1733641

>>1733471
>As I said, payload of about 25kg for the heaviest kind of head ( so for a ~1kg extruder head I think it should be ok, precision and vibration/inertia wise) And about 50Kg for the arm itself, maybe 100kg if needed
that payload is unrealistic an industrial arm of that capacity will weigh at least 250kg
2.5kg should be doable for a semi professional diy project
btw how do you plan to control it are there any open source or at least cheap controllers available?
because writing your own software would be a far bigger project than building an arm of any size

>> No.1733650

>>1733111
It would print, with maybe a little tuning. An msla printer would do a much better job for not much more than an ender though. You should still be able to find someone willing to print that for you somewhere though.

>> No.1733664

>>1733641

>that payload is unrealistic an industrial arm of that capacity will weigh at least 250kg
>2.5kg should be doable for a semi professional diy project
I see, thanks for the heads up, I haven't made a robotic arm before and I didn't run the math yet. I don't have many references outside of 1ton arms in factories.
Is the budget of 3-4k I set realistic (for a 2.5kg payload)? Considering that I can machine myself some of the parts.

>btw how do you plan to control it are there any open source or at least cheap controllers available?
I didn't start thinking about it yet, if there are existing kits then great, if not I could code a basic G-Code interpreter on a microcontroller. Its not hard, just a bit long since there are a fuckload of commands. The G-Code itself would generated with existing software like printer slicers or Cambam, or some other stuff.

>> No.1733668

>>1733664
>I could code a basic G-Code interpreter on a microcontroller. Its not hard,
Uh-huh.
Good luck mate.

>> No.1733670

>>1733664
>Is the budget of 3-4k I set realistic (for a 2.5kg payload)? Considering that I can machine myself some of the parts.
if you can machine all structural and specialized parts like gears and joints yes than that's achievable
assuming you hunt ebay for used servos and controllers

>I could code a basic G-Code interpreter on a microcontroller. Its not hard
yeah anon, it'll be way over your head the kinematics for a robotic arm are orders of magnitude more complex than a xyz gantry you find on 3d printers
no way in hell you'll be able to do that with this attitude
hopefully there's some existing project you can use

have you looked into used industrial arms? for 3-4k you might be lucky and find one
alternatively there's https://www.kickstarter.com/projects/1383636492/the-smallest-servomotor-robotic-arm
it's a lot smaller than what you are aiming for but also way cheaper and probably a great tool to learn about robotics before you commit to a big project

>> No.1733677

>>1733670
if you can machine all structural and specialized parts like gears and joints yes than that's achievable
>assuming you hunt ebay for used servos and controllers
Structural and joints yes, gears no. For motors I was thinking of a couple of Nema 34-42 with sensor feedback.

>no way in hell you'll be able to do that with this attitude
Sorry if I did pass as a condescending ass, wasn't my intention. But I do have a background in transport engineering (senior undergraduate, 5 year cursus). And I have already done robotic arm tier kinematics during my studies. Plus I can use Matlab/Simulink for simulating the thing. The real risk imho is getting stuck with writhing code for a hundred time consuming different commands.

>that kickstarter
Interesting, I'll look into it, but ultimately I want to DIY my own thing.

>> No.1733690 [DELETED] 
File: 78 KB, 1466x751, 5-axis-3dprinting.jpg [View same] [iqdb] [saucenao] [google]
1733690

Sorry for the robotic arm thread hijacking.
Going back at the initial subject, did anyone here had the chance to use 4-5-6 axis printer?

>> No.1733696
File: 78 KB, 1466x751, 5-axis-3dprinting.jpg [View same] [iqdb] [saucenao] [google]
1733696

Btw did someone had the chance to use a 4-5-6 axis printer?

>> No.1733746

Changed my Ender 3's aluminium extruder and tube to capricorns and now it seems to be printing much thinner lines that ruin prints.
Any clue what the issue should be?
Clogged nozzle, overly tight extruder spring, overly tight extruder wheel, broke couplers etc?

>> No.1733752

>>1733746
you ought to get a Prusa

>> No.1733754
File: 362 KB, 1247x1080, pic1.jpg [View same] [iqdb] [saucenao] [google]
1733754

Any tips for a newby on CR 10. I have lots of stringing problems. Also any info of toxicity or PLA. A fucking sibling is making problems because of the smells.

>> No.1733756

>>1733754
>toxicity or PLA
Pretty low but you don't want to be inhaling it 24/7.

>> No.1733757

>>1733746
Underextrusion may be caused by several factors that can easily be solved by reading the sticky. I'd bet a clogged nozzle (fix with atomic pull method), or some new extrusion ratio on your new extruder.

>>1733754
Stringing problems can be solved by reading the fucking sticky. Since it's a bowden system, you probably don't have a high enough retraction set, or your temperature is too high. Start with stringing tests and adjusting retraction, if that doesn't help adjust temperature and try again.
PLA isn't toxic, but it just smells funny. ABS and nylon are much more pungent, and much more dangerous.

>> No.1733759

>>1733756
Cheers. I read on most of the stuff but my sibling is making problems over micro particles which are literally everywhere and according to CDC the dissipate withing few inches. She is a retard who can't listen. Anyway. Anyone want the .stl to try it out?

>> No.1733761

>>1733757
>PLA isn't toxic, but it just smells funny. ABS and nylon are much more pungent, and much more dangerous.
Cheers. Do you have any sources so I can get her off my back?

>> No.1733763

>>1733761
https://lmgtfy.com/?q=pla+toxicity

>> No.1733766

>>1733763
Yeah. All the answers would just give her munition and she already read the Monsanto paid clickbait shit.

>> No.1733776

burnt pla is probably a carcinogen (as much as glyphosate or bacon, IARC > eu classification) but melted PLA is safer than your teflon tubes and machine oils on your printer.

Monsanto didnt deserve to be sued for 300mil/2bil btw, you cant prove something is cancerous in a lawsuit, especially when its an obvious devaluing move when bayer wanted to buy them out. the 100mill on appeal is at least appropriate for doing something you still cant prove today is wrong.

>> No.1733779

>>1733754
it's a bowden system with a really long tube, what did you expect?

>> No.1733784

>>1733776
I put it at 200c and plate at 50c. Should I take the nozzle temp down?

>> No.1733799

for a bowden?
you dont need the hot bed unless you use pei, set first layer to 210-205, increase your retraction length, extruder acell and speed to the max before grinding (mine is like 70mm/s with 60 return speed and 8mm length, depends on if its all metal and if its geared or not), add a z hop of 2x layer height, no coasting, no firmware retraction, wiping can help (cura's small feature setting, once its tuned, can help with fingers stringing together), combing might help if your extruder is shit. cura tree support works as a partial ooze shield and I have become dependent on that combined with manual support columns.

>> No.1733806
File: 128 KB, 628x472, featured_preview_Girl_1.png [View same] [iqdb] [saucenao] [google]
1733806

>>1733799
Thank you very much. I will try it next time. I uploaded the model on thingiverse but since I am a newb I have to wait 24hrs. Here is a preview.

>> No.1733835

>>1733746
Capricorn tube is slightly narrower than normal tube, check if it's secured firmly at both the extruder and the hotend. If it moves back and forth a mm or two on retractions, it'll cause loads of stringing.

>> No.1733851

>>1733746
>>1733757
>>1733835
Found the problem. Creality's official metal extruder adds a little rivet nut that the third party ones (and thus upgrade videos) don't have to keep it very tight, without that inbetween the spring it'll be loose and it'll cause extreme underextrusion like less than half as thick lines as they ought to be. I couldn't tell straight away because the loose spring still seemed to be about as tight as the old ones.

Still good advice on the tubing, if I start seeing more stringing than I used to after everything's calibrated to perfection i'll be sure to check them.

>> No.1733861

>>1733851
As long as the capricorn is fitted securely you should see a lot less stringing. Because it's a narrower tube there'll be less play in the filament inside it, and thus less slack eating up your retraction range. Did you make sure to cut the tube as short as possible? The shorter the tube, the less bad your stringing will be. It's why direct drive is so popular, on that the comparative tube is something like one twentieth as long as in a bowden.

>> No.1733865

>>1733861
Yeah I cut it about as short as I thought I could get away with and still have it extend the full axis dimensions without issue. I have more of it I can cut if it turns out I messed up anyway but good to know that can be an issue.

>> No.1733882

So I have a friend who is super interested in 3d printing and I always tell him about my experiences with crazy new experiments (most recent was non planar printing) but he doesn't own a printer. What should I print him as a Christmas gift that will get him super excited?

>> No.1733919

>>1733882
If y'all are outdoors folk or into a city scene or something geographic, go to jtatch.com/Terrain2STL/ and make an stl of whatever geographic region might be meaningful and print it. I've found that increasing vertical scaling by .2 makes things look great.

>> No.1733925

>>1733919
Broken link

>> No.1733926

>>1733925
Missed an h. My bad
http://jthatch.com/Terrain2STL/

>> No.1733991

>>1733531
>>1733586
Thanks, bought some spot putty and will head out tomorrow for some filler

>> No.1734101
File: 105 KB, 983x551, 1573434610343.png [View same] [iqdb] [saucenao] [google]
1734101

>>1733991
this or just the filler primer brand (more colors). I know these two solvents wont destroy (but will adhere to) PLA but will melt ABS very quickly if you dont use thin coats
if you use a brush, you might wanna get a surfacer primer as well, if you plan on sanding, just to clean up the sanding and brush strokes.

>> No.1734199

>>1733919
Great idea! How big of a scale do you usually print? 10km? 50km? I'm not entirely sure what would be the coolest

>> No.1734213
File: 838 KB, 3096x4128, fjgoz2o1g3x31.jpg [View same] [iqdb] [saucenao] [google]
1734213

What is this bullshit and how do I stop it?

Every layer with a hole is like .01mm narrower than sold layers. Seems to happen on every thing when there's odd geometry changes. IDK if its my Ender 3, Cura being stupid, me using stupid settings.

>> No.1734257

>>1734213
Do you have variable layer height turned on? And if not are you using magic numbers for your set layer height?

>> No.1734299

>>1734199
If you play with vertical scaling, you can make anything look cool. I've only done a few, and unfortunately didn't take pictures. You can play with the box settings so it's rectangular and stretched too, like I did a stretch of river my buddy and i like to raft. I did the entire small farm town valley and surrounding mountains that my grandpa grew up in and gave to him, I've done mountains we've backpacked. The valley town had the biggest coverage, and I increased vertical scaling to like 1.4 (default was 1) and it made it a little more interesting and easy to tell where things were without looking too exaggerated.

Also, the link won't do buildings, but there are tutorials for ripping buildings from maps.

>> No.1734433

>>1733806
Where is the link?

>> No.1734442

>>1733806
Is that a converted daz model?

>> No.1734460

>>1734213
no variable layer height, yes magic numbers

>> No.1734484

>>1734433
Finally. Here it is https://www.thingiverse.com/thing:4046260
>>1734442
Nope. Scratchbuilt. I have craploads of these that I did over the years. Now I will do something that actually makes me focus.

>> No.1734485

>>1734433
Please post results. If anyone has a resin printer I'd love to see how it comes out.

>> No.1734566

>>1733641
Not him but robotica is (iirc) an inverse kinematics solution for matlab.

All he needs is the D-H parameters of his arm and it will compute the necessary joint angles for any required motion of the tip.

Could I do it? No, but the hardest part of the job has been done for him.

>> No.1734571

>>1733806
Why do you have to wait?

>> No.1734581

>>1734571
They seem to have a period of 24hrs before they publish anything done by new creators. Anyway, its up - https://www.thingiverse.com/thing:4046260

>> No.1734596
File: 263 KB, 784x693, Capture.png [View same] [iqdb] [saucenao] [google]
1734596

>>1734484
>>1734485

I"M DYING.....lol

He modeled the cervix

>> No.1734627

>>1734596
how's that the cervix?

>> No.1734630
File: 10 KB, 246x205, FULLY MODELED.jpg [View same] [iqdb] [saucenao] [google]
1734630

>>1734627
Well if it isn't, then it is a very deep anus. Regardless it was kind of a reference to a video game with similar modeling.

>> No.1734646

>>1734596
That is not the cervix. I made the cervix disappear with dynamesh.

>> No.1734864
File: 61 KB, 500x500, source.jpg [View same] [iqdb] [saucenao] [google]
1734864

>>1734630
>>1734596
>FULLY MODELED VAGINA
As expected of /3DPG/

>> No.1734885
File: 120 KB, 1280x720, photo6028205656609239846.jpg [View same] [iqdb] [saucenao] [google]
1734885

Printer: Anet A6 Reprap i3
Material: PLA
Bed & extruder temperature: 50 C° & 200 C°
print speed: 30 mm/s or 20, not certain.

Original project
https://www.thingiverse.com/thing:1763518

I know is made for ABS, but there are people in the comments who said it worked for them in PLA.

Did not work for me, the parts don't flex.

>> No.1734897
File: 163 KB, 720x1280, photo6028205656609239845.jpg [View same] [iqdb] [saucenao] [google]
1734897

>>1734885

>> No.1734901

>>1734897
Eek. Your first layer looks a little high and sparse.

>> No.1734904
File: 3.46 MB, 4032x3024, 20191215_212401 - Copy.jpg [View same] [iqdb] [saucenao] [google]
1734904

Printing a watch and phone stand for KK gifts

>> No.1734913
File: 550 KB, 1620x2160, IMG_20191215_122010.jpg [View same] [iqdb] [saucenao] [google]
1734913

Any clue what the problem here might be? The print consists of a tall and pretty thin frame section in the background, this prints perfectly (two walls thick, which is why the infill is visible through the side), and a tree support in the foreground. The actual part itself prints perfectly, and so does 90% of the support, but this one side of the support has these huge holes. I don't think it's extrusion multiplier, because the part prints perfectly, and I don't think it's the speed I'm printing at since the entire support prints at the same speed and only this one face has any issues. The opposite side has no issues. You can even see that the issue happens even on the skirt at the first layer, the edge there looks a little frayed even though it's perfect everywhere else.

Generic PETG printed at 235 degrees, stock Ender 3 fan running at 30%.

>> No.1735007

>>1734897
>>1734885
looks horrible dude
should have bought a prusa

>> No.1735009

>>1735007
>>1734913

>> No.1735040

>>1733124
>100-200mm/s
maybe
> precision of about 0.1-0.05mm
somewhat realistic
>payload of max 25 kg
not happening. dildo sexrobot (from >>1733130
)weighs 250kg and has a payload of 6kg. and modern robots of similar size have only double that

>>1733132
lol

>> No.1735048

The RepRap looks fun to build, has anyone tried it? What did the cost come out to be?

>> No.1735066

>>1735048
>The RepRap
which one?

>> No.1735074

>>1735048
I remixed an entire motion system for my device. if you have acess to a printer here's a few things for making your own printer:

pla bushings work for years, going on 3 and no signs of wear on both the shaft and the plastic, use silicone lube.
follow an established reprap design, like the hypercube, at least for the mounting setup, so you can modify our use other remixed setups people around the world already have done for ducts, multi hot end mounts and the like
buy your own bowden extruder, don't print one, its a waste of time and metal ones are cheap
pla works, even for motor mounts if you have tmc drivers, my first hot end mount was pla and that worked until the fan broke
design for the parts you have, if you dont want to buy new bearings, modify the mount from whatever kit you're remixing (or use PLA bushings) I use a fusebox motion system with a hypercube mount just because I own the hypercube fans and already had the correct length rods for the fusebox motion system

the cost was negatable, I had 500$, spent 300 on a kit, broke the acrilic parts after a year and with what I had, other than rods, bearings, motors, and hot end I already had, I spent 50 bucks to fix, with another 200 on a brand new duet wifi (on sale at the time) because at the time smoothieboards just came out and where shit but I needed a good corexy board.
Expect spending the price of an equivalent kit unless you find the parts on sale or already own them, plus the price of a roll of plastic prototyping.
Controller-wise, buy a 50$ 32bit board with tmc drivers, avoid klipper unless you already have a pi and an 8 bit board.

>>1735066
idk what he ment by that, maybe the reprap project, self replicating manufacturing. something that died when we moved from screw rods, drills and bandsaws to aluminum extrusion and cnc'd hotends.

>> No.1735091

>>1735007
Are you sure the mistake is in the machine then?

>> No.1735095

>>1735091
Hard to say, back when i had a chinkshit machine these kinds of issues were an everyday occurrence, never really got properly rid of them

>> No.1735098

>>1735091
Flash Marlin to it if you haven't already. Big improvement on my chinesium machine.

>> No.1735104

>>1735066
Scrolling the page I realized after there are many different varieties to build so I still have to choose.

>>1735074
I have an Anet A8, I've been making it work for about a year and I'm pretty tired of all the little complications it has but I've pulled it apart and fixed it enough to have (I think) a good understanding of how to put one together. It functions well enough to print the parts needed.

$300 is what I bought my Anet for and probably have put about $200 into it to make it function better. I'm assuming this isn't including filament on the prints.

I'm most interesting in a learning experience.

>self replicating manufacturing. something that died when we moved from screw rods, drills and bandsaws to aluminum extrusion and cnc'd hotends.
Threaded rod isn't used anymore? What do you mean by aluminum extrusion and cnc'd hotends? Do you mean the brass nozzle? Isn't that always cnc'd? How does this lead to the end of self replicating machines?

>> No.1735108

>>1735095
>>1735098
I can attest to both of these with my A8. Get Marlin but still, there will always be something a little wrong unless you spend all the time you could've spent printing things you want to build printing things to improve the machine.

>> No.1735111
File: 479 KB, 520x350, 1560086623009.png [View same] [iqdb] [saucenao] [google]
1735111

>>1735104

>> No.1735112

>>1735091
That guy's been going around commenting the same thing on pretty much every sub-par print, just ignore him. I don't have an A6 so I don't know what exactly your problem is, but I know it's solvable.

>> No.1735122

>>1735104
Dude just download the prusa STL pack, buy a couple of extrusions, cut the frame out of plywood and you're set

>> No.1735123

>>1735122
Don't I still have all the wiring/board and motor stuff to do, that's where I was thinking cost would rack up. Unless I just decide to gut my A8 parts in the end.

>> No.1735124

Has anyone tried metal filament? I've been having some good luck getting used to using Hatchbox's wood PLA and using the metal PLAs from reviews I've read seems similar. No bed heating, more abrasive to nozzle, larger size nozzle, watch for clogging because it heats and cools faster. I do construction work and I have some ideas for prints at work. Nothing that will have crazy exposure to damage but drill bit cases or even hole saws for drywall would be amazing to have in a pinch when I need something specific.

>> No.1735125

>>1735123
Yeah sure gut up your A8, or go on aliexpress and order a few motors, a board, and a power supply

>> No.1735128

>>1735112
you can't polish a turd

>> No.1735166

>>1735128
this, the a8 modding community seems to be the neon of the scene

>> No.1735176
File: 311 KB, 1918x1787, back.jpg [View same] [iqdb] [saucenao] [google]
1735176

>>1735048
Built an i3 rework around 8mmx500mm rods and a wood frame.
$350 from various ebay sellers.(had good records until a hard drive failed)
Actually spent about $450 from buying stuff in bulk because I plan to eventually build something else.
Nice knowing that nothing could break that I couldn't diagx and fix with off-the-shelf parts
You can even get the printed parts for about $25 from ebay

>> No.1735207

>>1735176
Wow, that brings me back to 2014. Good effort dude! Got any prints to show us? Does this monster beat the Ender?

>> No.1735226
File: 250 KB, 1300x704, brace.png [View same] [iqdb] [saucenao] [google]
1735226

Hi 3DPG
anyone know how i can measure a hole inclinasion from a stl part with solidworks ?
im trying to adapt this rigidity addon for my wanhao i3 , the hole have a certain angle on both XY axis, i need to recreate this feature in a custom part

>> No.1735234

>>1732568
Whats the downside of resin printing? thinking about getting an elegoo mars

>> No.1735265

>>1735234
Much smaller build volume, weaker parts, and it's messy to deal with and potentially hazardous to your health if you're careless. That said I love my mars and would reccomend it

>> No.1735308

>>1735265
I agree with this, post processing is a bit of work too and I think a lot of material is lost in supports but I could be wrong. I have an Anycubic and from what I hear from Mars owners makes me wish I had one instead.

>> No.1735316

>>1735176
I like the cable management, looks neat. A8 wiring lengths makes it look like an absolute eye sore. Your method seems pretty reasonable, especially buying in bulk. I wish I had more access to wood working equipment. Thanks for the advice and inspiration

>> No.1735319

>>1735308
>From what I hear from Mars owners
It's the same process, how could there be a radical difference? Post processing and support removal is still a thing

>> No.1735346

is the tevo tarantula pro a good alternative to the ender 3?

>> No.1735391

>>1735226
Take a side view and make an extruded cut that goes up to about halfway of that hole. Then, open a sketch on the new side surface, and then use convert entities to give you the planar aspect, and then measure that using smart dimensions.

>>1735346
Minor differences between the two, I'd consider the Tevo a good alternative. There are various factors that add a bit of cost to the Tarantula over the Ender 3, those also make it more moddable - which may or may not be a good thing, depending on your skill level.

>> No.1735399

>>1735319

There isn't a difference, it's just that Mars owners are more vocal about their printers, like Ender3 owners. The hardware is nearly identical apart from the LED which is supposed to have a couple of extra watts, but that's it.

>> No.1735417
File: 70 KB, 1000x1000, 61OlQbV986L._SL1000_.jpg [View same] [iqdb] [saucenao] [google]
1735417

So my father just bought for 470€ a ANYCUBIC Chiron FDM.
Good choice?

>> No.1735420

>>1735417
Pretty good, I got mine for cheaper but it does rather well. Throw away the autoleveling probe ASAP, good manual leveling is much better. You might want to consider a glass bed, but you definitely want a 0.8mm nozzle to fill that build volume at any appreciable pace. Join the Facebook group, they've got some good support. Stock spoolholder is garbage too, print a seperate one.

>> No.1735421

>>1735226
>>1735391
AWSOME, thanks a lot

>> No.1735439

>>1735420
this, I dont even trust bl touch auto levelling, unless you have a competent inductive sensor for mesh leveling, use glass.

>> No.1735572

>>1735265
The newer resins are about on par with thermoplastics except shock/impact resistance. So they're about on par with the better cast resins.

>> No.1735656

>>1735124
Aren't those just metal particles suspended in pla? You still have to sinter the resulting green body.

>> No.1735661

>>1734885
I've printed this before, snapped in PLA multiple times, didn't work until I printed it in PETG (and it was overly flexible, wasn't very springy)

>> No.1735667

>>1735399
25W Photon VS 40W Mars is a significant advantage when it comes to print time reduction

>> No.1735688

>>1735656
I guess you would but is it necessary? Could you just print it and leave it brittle? I don't 100% understand sintering, just heating the metal particles so they fuse right? This is why I'm asking

>> No.1735690
File: 1.08 MB, 3264x2448, IMG_3187.jpg [View same] [iqdb] [saucenao] [google]
1735690

Does anyone have a good primer on direct metal laser sintering? The technology looks amazing.

>> No.1735695
File: 1.03 MB, 1196x973, file.png [View same] [iqdb] [saucenao] [google]
1735695

Just got my first 3D printer today (Anycube Mega-S) and my first print came out great

>> No.1735734

>>1735439
That's autistic. A lot of people have consistently good prints with auto bed leveling. If it's not repeating consistently just slow down the probing

>> No.1735735
File: 277 KB, 918x1632, 2019-12-16 22.23.37.jpg [View same] [iqdb] [saucenao] [google]
1735735

How do you stop swiss cheesing in only one part of a print? In this case only one side of the rock and the shaft of scythe suffer from underextrusion.

>> No.1735742
File: 632 KB, 868x473, this does not go in the oven.png [View same] [iqdb] [saucenao] [google]
1735742

anons, dont attempt to sinter aluminum PLA, unless you ant a puddle of nothing in the end
sintering is nothing as you guys describe, its a very simple process
dmls is directly melting little beads of metal with a powerful laser into each other (thats the LS of sls or dmls) https://www.youtube.com/watch?v=bwFspzVGUF4
the metal has no crystal structure, so its more like a ceramic right now (metal ceramic is interesting, but the cool ones are made from aluminum oxide and zirconia at very high temps), many companies have their own methods for fully sintering (moving from green to brown) these things, typically by baking the metal.
there's also metal powder and binder, that iro3d process where you melt a weaker metal into the powder, or casting (from pla or sla casting resins), all which have their own design limitations and considerations (this is why most sls/dmsl metal prints still have support material for sindtering/resinderting)
https://www.youtube.com/watch?v=fzBRYsiyxjI

BASF Ultrafuse and Virtual Foundry can be printed on a desktop 3d printer, but they use a 90%+ fill and since they were put in a thermoplastic in order to print, they need to be sintered in a process more like recasting by melting off all the pla into sand or some material, similar to how crucible materials absorb all the shit left in metal refining, so they wont turn into a pile of mush.
you could also cast some bismuth materials in plastic molds, which I find interesting https://www.youtube.com/watch?v=770cSWbmBRg
this exists too, but RFID gas https://www.youtube.com/watch?v=W_jcYga95aY
there's also welding, like whatever musk or nasa did with the robot arms and melting metal in its free form. I dont trust it, all the info is patented and it seems like the caveman process with big seams ready to burst when its used, I cant even find out if they sinter those things. https://www.youtube.com/watch?v=DKkcBoSeUOg https://www.youtube.com/watch?v=iLndYWw5_y8

>> No.1735747
File: 84 KB, 969x481, 1549064309186.png [View same] [iqdb] [saucenao] [google]
1735747

mesh printing really only needs ~IR range accuracy, actual z probing for multi z axis homing needs the consistency of ~microswitch (no lever) or by hand like duet does for their homing methods (2-5 point homing when z steppers are initiated, loading of different heated beds depending on temp or bed material measured with a removable probe), or you will have first layer issues, but as long as its within .05mm every time, you shouldn't be too concerned. this is true going down to a 0.1mm first layer, by that point you should use a raft.

>> No.1735749

>>1735417
>>1735420
I just got my Chiron and its been driving me nuts. The Fan for the blower will not turn on unless I power cycle the whole unit and then manually send the M106 command, and removing the M107 from the Start Up script.

As soon as the print starts, or one has finished, the fan refuses to turn back on until power cycle.

I set both in Cura and Simplify3d to start full power at layer 1, layer 2, doesn't matter, will not do it.

Did you have this issue at all? I'm waiting on the chinese support to reply back but they are...slow, regardless of time zone =/

>> No.1735768

>>1735735
Check the pressure on your extruder gear, fixed it for me when I had that issue

>> No.1735773
File: 1.67 MB, 4032x2268, IMG_20191216_221804365[1].jpg [View same] [iqdb] [saucenao] [google]
1735773

Also about the Chiron is this shit...

My poor belt twists...

>> No.1735831
File: 60 KB, 540x720, 1568629346188.jpg [View same] [iqdb] [saucenao] [google]
1735831

Anyone here using the creality borosilicate bed with petg successfully? Thinking about switching but I heard petg is eating glass beds like crazy.

>> No.1735840

>>1735831
Any glass bed with PETG should use a thin film of hairspray as a release agent, this stops the PETG from killing your glass bed.

>> No.1735849

Is linear advance snake oil? I went through the hassle of converting my ender3 to direct drive and configured LA but now I have ringing on the Y axis AND on the X axis due to the weight of the extruder.

>> No.1735851
File: 3.92 MB, 2340x4160, IMG_20191216_223926.jpg [View same] [iqdb] [saucenao] [google]
1735851

>>1735849
Forgot pic

>> No.1735901

>>1735851
Linear advance totally works. On Marlin 2.0 and my Anet A8 housefire it makes a huge difference.

>> No.1735911

>>1735901
>my Anet A8 housefire
hats off to you, fine sir

>> No.1735930
File: 1.04 MB, 2560x1440, Z seam.jpg [View same] [iqdb] [saucenao] [google]
1735930

How do I get rid of these z-axis artifacts? They show up in all my prints. Ender 3, cura, pla at 200 (and 205), black, red, and green filament.
My z axis is calibrated perfectly. I haven't calibrated my flow but there are no other flow-related artifacts. I have tried seam hiding, wipe, coast, and a myriad of other settings. Other than that, my prints come out beautifully - No stringing, slant, weird layers, etc.
It's the flow, isn't it?

>> No.1735931

>>1735930
Try reducing layer height first, see if that makes a difference. If it does not, yes, it's probably flow.

>> No.1735933

>>1735930
please make a circle arund what you refer to

>> No.1735936
File: 1.70 MB, 1440x2560, 20191217_084834.jpg [View same] [iqdb] [saucenao] [google]
1735936

>>1735931
My bad, I forgot - I have tried .12, .2, and .24mm layer height. This one is .2mm.

Print speed has been tested between 60-75mm/s. Jerk control enabled and disabled.

>>1735933
My print that bad? Added circles.

>> No.1735940

>>1735930
>>1735936
They're called z seams and you can't truly get rid of them. They're left over from when you z axis moves up to shift layers. If your model has corners you can tell the slicer to align the seams to them and they won't be visible, but if your print is cylindrical the best you can do is align them so that they're all in one spot and easier to sand off. I think strata sys has a technique to reduce these, but its patented so we'll get to use it in 20 years.

>> No.1735954

>>1735690
What are you interested in?

>> No.1735958

>>1735936
Technically linear advance should help because of the reduction of nozzle pressure at the end of a layer. But dunno how/if it works on a bowden printer.

>> No.1735971

>>1735958
it works, but sharpest corner and retract on layer change are more important for those end blobs, just dont coast with a bowden. the linear advance settings, when done on bowden, only really helps with ringing and tolerances for stuff like gears or lithophanes, and works better for stiffer materials. (thats really all the linear advance tuning does, changing the extrusion values dynamicly in order to make a consistent line when printing at different speeds)

>> No.1735985
File: 150 KB, 1108x976, old print, but shows barely any blobs from layer change.jpg [View same] [iqdb] [saucenao] [google]
1735985

>>1735936
ok, first use software retraction, then try to reduce temps, enable the sharpest corner or hide seam settings other than random, disable/enable different combing methods for what the printer will do when it moves to the next island, adjust the layer width, maybe try lowering the layer height, maybe coasting will help, (coasting usually doesn't with small details and changes quality if you increase/decrease temps or even fan speed), linear advance tuning could help, but even my longest, super detail slicing settings still get stringing (between small parts) and these blobs (or marks where speed is reduced and gets shiner where the blobs would be).

>> No.1735995

>>1735954
Key fundamentals beyond the basic intro, where it's at and where it's going. I was reading the fall issue of metal AM magazine, and they seem to think its a relatively "mature" technology, but seems expensive

>> No.1736040

>>1735985
>>1735958
>>1735940
Thanks for the info and help. I'll apply it to my next print. Don't have anything circular planned for the near future but I'll update when I do.

>> No.1736057

>>1735688
It's necessary otherwise all you have is just a regular plastic print with metal powder in it.

>> No.1736118
File: 189 KB, 800x469, napoleonic_diorama_2.jpg [View same] [iqdb] [saucenao] [google]
1736118

Fantasy and Sci-fi minis are overly easy to find, but what about historical minis? Sometimes I manage to find a model or two around, but nothing more. Is there any good repository of good and diverse historical minis, either free, purchaseable, or [spiler]pirated[spoiler]?

>> No.1736165
File: 62 KB, 940x1024, 1572265938148.jpg [View same] [iqdb] [saucenao] [google]
1736165

Ey frens, noob here. In the store where I am going to buy my first printer it only has 4 colors in PLA, what color should I choose for a noob like me? They have purple, transparent red, light grey and black. I like purple but maybe a light color is better to see mistakes? like gray

>> No.1736166

>>1736165
I personally use white because you can't see sanding color disfiguration.

>> No.1736200

>>1736165
I like safety vest orange. Seems to hide printing imperfections best of all the colors.

>> No.1736206

>>1736165
If one colour looks more matte than any other, go for that. If not, emergency option is white. Glossy materials will show imperfections the worst, black in particular will _always_ have visible layer lines when held up to the light. Matte materials however will pretty much always look great, it's why I love printing carbon fibre.

>> No.1736289

>>1736057
From what I read it's 80/20 metal to plastic. So what is it just going to give a metallic look or will it be more durable than a PLA print?

>> No.1736293

>>1736118
https://cults3d.com/en/tags/historical?page=2

Best I could find.

>> No.1736333 [DELETED] 
File: 1.22 MB, 1920x1080, Liberator_Test_Fire.webm [View same] [iqdb] [saucenao] [google]
1736333

https://archive.org/details/gunplans2019

https://gun-plans-2019.jimdosite.com

>> No.1736337 [DELETED] 

>>1736333
Thanks Mr.FBI. Seriously though great shit thanks anon.

>> No.1736344
File: 13 KB, 243x243, 9B9BE591-B942-4829-9B98-A38FF68B143F.jpg [View same] [iqdb] [saucenao] [google]
1736344

Overnight printing. Do you guys let your prints run overnight?

>> No.1736345

>>1736040
>>1735936
That looks like Z-seams, Cura has an option to randomise those at the cost of increased print time. Makes them much less visible.

>> No.1736346

>>1736344
Don't really have another option if your print is 24 hours or more. Make sure thermal runaway is enabled, make sure there's no flammables close by, and pray to the gods it'll work.

>> No.1736347

>>1736346
or just buy a prusa and print without worrying

>> No.1736349

>>1736347
You can fill the print volume of a Prusa in less than 24 hours, those things are pathetic in terms of what you can build with them. Overpriced, too, and the amount of goddamn shills for them is too damn high. Those people should really off themselves, trying to sell inferior printers to newbies that can't tell the difference between a modern, large volume printer and the antiquated Prusa design.

>> No.1736357

>>1736349
what the heck are you talking about? like literally every printer ever has a print volume of 200x200x200
and that is for a good reason, ordinary fdm 3d printing is best for parts that fit into that volume, you go bigger and you really start to see the limitations of the tech

>> No.1736358

>>1736357
CoreXY format knows not of such things
>>1736349
Agree, but they exist to bring the tech to home gamers, and to enrich joseph prusa

>> No.1736359

>>1736358
Yes it does.
Cartesian is way better than CoreXY anyhow.

>> No.1736360

>>1736359
Removing the artifacts of carrying an additional stepper motor will increase the cost of production, or lower the quality.

Corexy, all you need is good belts, but all of this is better than i3

>> No.1736362

>>1736360
nothing stopping you from doing bowden in a cartesian
other than bowden being shit
but it is equally shit with CoreXY

>> No.1736363

>>1736362
Y gantry requires a stepper motor to move the belt

CoreXY was designed to provide an alternative for this

>> No.1736366

>>1736357
>ordinary fdm 3d printing is best for parts that fit into that volume
Nigga what. I have a Chiron at home that can print it's 400^3mm build volume just fine, at work we have GRR500 which most definitely needs the build volume for the tasks we're given it, and since FPF is a subform of FDM (fuck Stratasys patents), you can go much, much larger than just 200^3mm.

>> No.1736369

thingiverse down?

>> No.1736371
File: 1.63 MB, 2140x2386, IMG_20191218_120345__01.jpg [View same] [iqdb] [saucenao] [google]
1736371

90€ for the supervolcano? No thanks I'll make one myself

>> No.1736404
File: 2.63 MB, 3264x2448, 20191120_190231[1].jpg [View same] [iqdb] [saucenao] [google]
1736404

>>1736344
All of my printing is unattended after the first 10 minutes.

>> No.1736405
File: 2.93 MB, 4032x3024, 20191031_095936[1].jpg [View same] [iqdb] [saucenao] [google]
1736405

>>1736357
>like literally every printer ever has a print volume of 200x200x200
Everybody look at this post and laugh

>> No.1736410

>>1736405
considering that's basically two printers in one
he's not that far off

>> No.1736421

>>1736410
>muh goal posts
Only if you set it in duplicate mode. Otherwise you can print a single part that is 400mm wide.

Machine size limits are due to moving mass, and linear guide accuracy. FDM printers that scale beyond 400mm tend to be pellet-feed since the available print volume can easily go beyond what is convenient to produce from even 3kg spools.

>> No.1736446

>>1736289
It's actually less durable than a pure PLA print, because plastic bonds less well to metal dust than to other plastic. For an analogy, imagine that you cast something out of brass, but also mix the brass with sand. Less durable.
It does look pretty convincingly like metal though. It'll feel slightly cold to the touch and be a bit heavier.

>> No.1736523 [DELETED] 
File: 2.69 MB, 720x1280, shuty.webm [View same] [iqdb] [saucenao] [google]
1736523

https://archive.org/details/gunplans2019

https://gun-plans-2019.jimdosite.com

>> No.1736533

Is there a single up to date guide on upgrading the marlin firmware with the BLTouch on an ender 3 from stock? (have to do it since the v3 bltouch apparently doesn’t work on old firmware)
I’ve got the silent board with a bootloader but most videos on the subject seem to be for old fw stuff that doesn’t apply anymore and a lot of the configuration info for marlin is just hidden around those videos and i’m not sure all of it really applies anymore either. There’s also claims of the arduino ide integrating the sanguino and u8glib stuff on newer versions but I dunno if that’s bs or not.
Pretty much the only upgrade on the ender 3 that’s incredibly annoying to find up to date information on so far.

>> No.1736540

Anyone here have any idea about the costs of injection molding? Interested in making an outdoors product but I'm afraid any 3D printed materials wont last in the long run or am I completely wrong? Think gardening usage.

>> No.1736541

>>1736523
Isn’t there a huge issue with heat distribution for printed guns?
I guess you could cast them in metal though.

>> No.1736545

>>1736540
Printed ABS should hold up fine.

Cost of injection molding directly relates to geometry and volume (size) of the part in question.
Parts that are 1 cubic inch or lower can be done on benchtop equipment below 5 tons. And the tooling for those is piss-easy to make.
Parts beyond 2 cubic inches are going to cost you pretty pennies in just the die cassette machining costs.

>> No.1736547

>>1736540
Kind of interested as well. Injection molding plastics with a LNS 150A is like $1800 for a very small build volume even for me as a miniatures guy, at which point i’d rather just go with multiple resin printers and fill up the build plates.

>> No.1736551

>>1736541
You've got to shoot quite a many mags if you want your barrel so hot it melts ABS

>> No.1736558
File: 2.63 MB, 1812x964, help.png [View same] [iqdb] [saucenao] [google]
1736558

I just got a 3d printer and tried my first print and failed. I don't know what this is called, but this is what happened when I tried to print. The actual model is to the right. Somehow, it doesn't translate properly when slicing.
I'm using the Anycubic Photon S with a resin material.

>> No.1736560

>>1736558
Could be a model with holes. Check it with photon validator and fix it in netfabb/meshmixer.
Could also be the supports, throw it into chitubox for those. (Angle your model 45 degrees or so to let the supports go on the actual model, I know some people also like to cut off the base and print it separately)

>> No.1736562

>>1736558
theres a lot of problems with your approach
first of all, that model probably isn't solid and needs at least a meshmixer inflate pass, I don't think typical mesh repair would help in making that close to printable. even if you printed that in a sls printer, the paws are too thin to be cleaned up by even dusting
2nd, you bought a msla printer as your first, you need to use the proper supports (thats what its called, anon) for that, all I personally know about that process is there are multiple programs that will generate compatible supports, the only one I use is meshmixer and I don't even use the tree support option for fff on there, chitubox is what everyone uses it seems.
3rd, since there is a base being printed, a big slab of cured resin gets stuck to your bed or film, of what I know from msla, printing like this is awful for the bed/film.
look here for getting all those details printed https://www.youtube.com/watch?v=b3wFEdN7TqQ
the more unsupported details, the more work you have to do to clean up

Personally, I would cut up the model into little parts (for that model, like 10 and a base) and increase the model size by ~200-500% to get all those little details printed and able to survive print cleanup.

>> No.1736565

>>1736562
also I want this, anyone with msla printers comment on the lifetime of resin in those vats? https://www.youtube.com/watch?v=VTBqOeTdQPs
a big wide vat seems like all the solvents would dry out quickly

>> No.1736567

>>1736558
furries...

>> No.1736573
File: 131 KB, 843x645, 1576433999577.png [View same] [iqdb] [saucenao] [google]
1736573

>>1736567
he attempted to print this thing...
https://www.thingiverse.com/thing:3954949
I wanted to see the 3d hubs price with a 50mm scale ver of this and I think I broke the site's compatibility checker

>> No.1736574

>>1736567
Not everything that has to do with animals is furry related.
>>1736560
>>1736562
Thanks for the help. I'll look in to these suggestions.
>>1736573
But I think the actual print is fucked so I'll probably just print a metal gear or something in its place.

>> No.1736576
File: 447 KB, 1360x724, 1555432944516.png [View same] [iqdb] [saucenao] [google]
1736576

>>1736574
at over 600 shells and 20,000 intersecting faces, this is not printable in the slightest
do the test print, a benchie or a Gayer-Anderson cat print first before trying to go all out with the supports, do you even have ipa for cleaning the resin before cureing?
https://www.myminifactory.com/object/3d-print-festive-gayer-anderson-cat-52539

>> No.1736577

>>1736576
Most of that could probably be fixed automatically with a pirated copy of netfabb t b h.

>> No.1736579

>>1736577
no I own nettfab, I know what's possible
Ill get a copy up and running in a couple hours to prove windows 10 auto repair or web netfabb is a better choice than installing a useless tool if you want.
this model seems like something to show a simple write up comparing these things

>> No.1736580

>>1736579
Use the the most extensive auto repair option & post results. Actually kind of curious what it’ll do with a model like that.

>> No.1736586

>>1736580
ill pirate netfabb 2020 to get best results compared to:
-meshmixer "autorepair all" and "solidify" (solidify with the highest settings, prob will still fail, solidify isnt really meant for repair and is a secondary step in my make printable process)
-https://service.netfabb.com/login.php (typically the best free choice I use)
-https://tools3d.azurewebsites.net/ (prob the same as web nettfab^)
-3d builder app auto repair (prob the same as the online tool^)
-https://makeprintable.com/
-blender 3d print toolbox repair
-MeshFix
-Mesh lab (2019 github)
and anything else I can get to without having to talk to someone to demo their software for a fake company I'm linked to.

>> No.1736600
File: 22 KB, 765x209, 1567954560108.png [View same] [iqdb] [saucenao] [google]
1736600

also for mods, netfabb breaks the MIT license, so I'm just enabling features that are freedom free.

>> No.1736601

>>1732568
>>>b

>> No.1736675
File: 2.56 MB, 4640x3480, IMG_20191218_211713.jpg [View same] [iqdb] [saucenao] [google]
1736675

Help me, I just started resin printing. the skirt and sword on this failed to print. should I just add more supports? and Are resin support supposed to be this damaging?

Epax X1
Elegoo Mars blue
45 bottom time
6s exposure

>> No.1736676
File: 316 KB, 550x527, 1549874648926.png [View same] [iqdb] [saucenao] [google]
1736676

>>1732568
>when 3 of your files are on the OP

>> No.1736683

>>1736675
>should I just add more supports
can't tell if you don't show the supports in the first place.
>Are resin support supposed to be this damaging
I don't know, since your camera focused on the foam and not the figure.

>> No.1736685

>>1736676
my image tho

>> No.1736697
File: 1.93 MB, 4640x3480, IMG_20191218_204846.jpg [View same] [iqdb] [saucenao] [google]
1736697

>>1736683
wrong picture, supports are gone though

new pic coming up

>> No.1736709
File: 2.48 MB, 4640x3480, IMG_20191218_224655.jpg [View same] [iqdb] [saucenao] [google]
1736709

skirt didn't print so well but the final product looked great from the front. what do you guys do to clean up resin supports?

>> No.1736712
File: 2.42 MB, 3480x4640, IMG_20191218_230301.jpg [View same] [iqdb] [saucenao] [google]
1736712

>>1736709

>> No.1736717
File: 236 KB, 1228x636, 1566248291144.png [View same] [iqdb] [saucenao] [google]
1736717

so far results have been interesting, gotta find a consistent non-manifold xray tool like the 3dhubs viewer
the blender render should be cool with the colored difference textures, a few of the tools are broken tho, like makeprintable changing payment models and the azure server queue taking forever

>> No.1736719

>>1736712
>>1736709
>>1736697
The angle's too flat, the whole point of angling it is to minimize the contact area of the print with the print surface while balancing it with print time. 45deg or so should be better. If you're using a Photon, the default program it comes with is garbage at generating supports, so install PrusaSlicer and use those to make them instead. You can export the whole thing as an .stl and then slice like you usually do. Don't always rely on the autogenerated supports of any program, check it manually in case there's anomalies.

Use a VERY sharp blade to trim the supports close to the model, and flush cutters to get the larger pieces out of your way. A pack of 120 blades is $10 and will last you ages (https://www.amazon.com/gp/product/B07CR7QBRB).). Get the supports off BEFORE post-curing, otherwise they become extremely brittle and difficult to remove without shattering/marring the part.

>> No.1736722

So no god hands? fug

>> No.1736741

>>1736446
That's interesting, thank you. There's some 250g reels I can get, I'd still like to play around with it to test it, I imagine I could make it pretty durable with a high infill

>> No.1736745

>>1736712
I use Chitubox for slcing. With supports usually anything that's hanging over will droop down. Try to keep supports in the back away from what you'll see when you display it just in case there are imperfections. I'm not sure why the supports are shooting out the side like that, looks kind annoying for post processing.

>>1736719
Since you have experience with Resins I have a few questions. what do you use to post cure? I use IPA and water but find that my prints are a little sticky still after and bits from post processing sanding/trimming get stuck pretty good.

>> No.1736746

>>1736404
How much have you into this set up?
So many prints I've decided not to do because I don't want to micro manage a bunch of 20-30min prints for tons of small parts.

>> No.1736751

>>1736745
Post-curing involves some sort of UV light, not just rinsing the extra resin off. You can either buy a special setup for it (somewhat expensive), or use a UV nail gel curing machine for an hour or so (cheaper), or just stick it on a windowsill for a day or two (cheapest). If sticking it in the nail gel thing or in a windowsill, flip it around halfway through so both sides get even exposure.

>> No.1736763

>>1735911

After printing like, all the shit to make it more rigid, getting a printbed sticker and bltouch, it prints like a charm. All the cool shit in marlin 2.0 works, still have to dial in that sick junction deviation to get less ringing.

Upside of having such trash printer is, you learn a lot. I broke mine like 3 times since i got it and fixing it and upgrading made me understand whole process much better.

Anet A8 is great for the price if you have time and get a bed sensor.

>> No.1736769

>>1736540
Injection molding gets slow and complicated when you start trying to save costs and do things manually. But you really just need a way to make a mold and a press that can pushing molten plastic into the mold. Problem is air pockets are an issue and so is the rapid cooling as the plastic enters a metal mold, so you either need tons of pressure like true injection molding machines, or you need to keep the mold hot enough to keep plastic flowing until it fills up everything. There are a lot of simple enough designs out there you can look at but it's basically just a hot pipe with a nozzle filled with plastic that you press on to have the nozzle shit out plastic into a mold.

>> No.1736783
File: 2.23 MB, 4640x3480, IMG_20191219_033126.jpg [View same] [iqdb] [saucenao] [google]
1736783

Resin Slayer

>> No.1736799
File: 1.06 MB, 1100x1664, 1572418473264.png [View same] [iqdb] [saucenao] [google]
1736799

ok, after blender render is out a thingiverse comparison and opinions should be uploaded
https://send.firefox.com/download/7dc55886bf74c8db/#KayNh6HoKQkwUvR3v88SSg
none of these are printable to scale, netfabb preformed the best, but materialize was a nice set of software.

>> No.1736800
File: 540 KB, 1289x827, 1545381000698.png [View same] [iqdb] [saucenao] [google]
1736800

>>1736799
Forgot this part, the worst ones (and a descent one) are here

>> No.1736817

APB
Do not buy the kelant s400. It doesn't have even the PRETENSION of being a reliable, working product.

>> No.1736828

>>1736746
>How much have you into this set up?
In the past 2-1/2 years I have worn out
4 monoprice select minis (new)
1 lulzbot mini v1.0 (used)
1 lulzbot mini v1.04 (new)
3 lulzbot minis v1.04s (refurbs)

Pictured and still working are
1 lulzbot mini v1.04 (new)
2 lulzbot minis v1.04s (refurbs)
4 Qidi X-one2s

The Qidis are pretty great for farming, but their one downside is that you can't control them with octoprint servers because the USB port is too slow to feed gcode through.

So to date I have spent $10,200 on printers.

>> No.1736831
File: 2.95 MB, 4000x3000, IMG_20191218_180519.jpg [View same] [iqdb] [saucenao] [google]
1736831

It's a little bigger

>> No.1736921

>>1736799
>>1736800
Wow anon, thats some nice comparison work. Some of those conversions look pretty impressive. I should really learn to use a non CAD 3d tool one of these days.

>> No.1736962
File: 37 KB, 400x562, For what purpose.jpg [View same] [iqdb] [saucenao] [google]
1736962

>>1736831

>> No.1736968

>>1736962
More area for transferring heat into the filament. Allows for higher print speeds.

>> No.1736988

>>1736962
Printing at ridiculous speeds with 1.2 mm nozzle

>> No.1737078

>>1736828
Damn man, what are you printing, what's your endgame with all this? It looks like your doing big projects,especially if you've worn out printers. How much filament are you burning through on average? How costly has it gotten?

>> No.1737123

>>1736968
>>1736988
Not that guy I'm still curious. The reason is because more filament is heated at any given time, therefore, more can printed quickly?

>> No.1737134
File: 350 KB, 1108x1080, 1568836038613.jpg [View same] [iqdb] [saucenao] [google]
1737134

I just bought the CR-10 V2, how fucked am I?

>> No.1737152

>>1737134
buy a 32 bit board, a LOT of people are going to tell you to get an arduino to unlock, a full size pi for controlling, and an under volted or inaccurate z probe; just buy a wifi enabled (or a 10-5$ pi0w enabled, whatever's cheaper) 32 bit board and an inductive probe if you wanna be hands off.

>> No.1737154
File: 19 KB, 255x258, whatdoesthemuttsay.jpg [View same] [iqdb] [saucenao] [google]
1737154

>>1737152
Im quite new to 3D printing, do I need all that extra stuff like a raspberry pi for controlling, etc?
Doesnt it work "good enough" out of the box?

>> No.1737168

>>1737152
piw0 is barely fast enough to feed a printer at full speed. If you want to run a webcam on it or the full chatty web interface, it struggles hard. And it'll take a good 5-10 minutes just to boot.

>> No.1737169

>>1737154
it does if you wanna be a bitch and not automate
all I'm saying is dont fall for the traps when you do
>arduino
gay and only is needed because creality doesn't want with people destroying their printers by trying to circumvent the locked down marlin. not wanted and not needed with 17 fucking dollar prime delivered drop-in replacements on the market
>full size pi
this is for klipper fans and tards that think you can slice models on the printer instead of just using a dedicated thrift store pc. unneeded extra tool to break that does the job of a esp32
>bl touch
the most expensive probe is literally worse than a switch taped to the extruder

the printer will work, you'll tune and fix issues with the shitty filament runout sensor or change the bed type, but those fixes are cheap and only cost time and a few cents of plastic. the worst expenses are like glass and pei sheets. the BS involved with hardware upgrades or wifi enabling is where i'm concerned. you dont need any of this shit but people will tell you that you'll need to spend 75$ and learn both the arduino IDE and rasbien just so you dont have to move an sd card.
before octoprint, people used to shill for micro to full size sd cards so you wouldn't have to keep your pc connected; the problem was those devices fucking broke the printer if you tilted them at all, meaning you had to spend more money to buy a new board and reprogram it or go back to the "stone ages" of keeping a pc connected to your printer.
its all the same shit, this community's reddit shills find some solution to their lazyness and anyone who wants the machine to work a bit more efficiently gets fucked over into spending more cash when the best solution is always the simplest.
if you want a faster printer that's not locked, the bottleneck is the chip itself, and luckily today thats one the cheapest pieces of hardware to replace.

>> No.1737171

>>1737168
I'm meaning just feeding files into a storage media, octoprint can suck a big one, 32 bit boards dont need a(n?) usb tether to tell it what to do.

>> No.1737174

>>1737171
no, neither do 8 bit boards. But the wifi and monitoring is nice to have separate from the processor controlling the hardware. Having twice the features in one place seems to cause ten times the bugs.

>> No.1737176

There seems to be nothing in the sticky on firmware upgrades? Is there a renowned forum or wiki or something? I copped some shit a while back for still using skynet

>> No.1737178

>>1737176
firmware upgrades for what?
Each ecosystem is going to have its own hacking groups. Probably not much overlap.

>> No.1737180

>>1737178
For an Anet A6. So you suggest just digging around Anet forums? Thanks :)

>> No.1737181
File: 179 KB, 1640x1035, 1575186401733.png [View same] [iqdb] [saucenao] [google]
1737181

>>1737174
>wifi and monitoring is nice to have separate from the processor controlling the hardware.
even the duet uses an esp32 to do exactly that
you dont need a rasberry pi to do this, a pi zero can do this VERY well, even when running thru octoprint you can do file operations to save processor time, run the server, AND control the printer over basic ass usb.

>> No.1737183

>>1737180
No idea for Anet. I'm a RepRap/Prusa hacker, so.....

>> No.1737185

>>1737181
As someone currently moving his print farm from Pi0W to PI3A+, the PI0Ws struggle just to keep up with WiFi. I mean, they work, but they fall behind really easy.

>> No.1737192

>>1737185
thats a problem with the bloat then, or the shitty wifi chip; the 2$ esp8266 on smoothie and duet have had none of these issues, something with magnitudes more power should be able to run a little web server. are you sure its not all that wasted time trying to keep an 8 bit board usable is not the issue.
https://www.youtube.com/watch?v=tpxHwl4AUBY looks like even in 2016 there was exploration for this on marlin

>> No.1737194

>>1737192
It's specifically a problem with the PiZeroW. The way it implemented USB and WiFi kills the processor. But seriously, it's got no power even leaving that aside.

>> No.1737195

>>1737194
ok, all one would need is 2$ chip instead of a 7$ computer because somehow you can fuck up two gpio pins and a wifi connection. a pi is entirely useless in my eyes now.

>> No.1737208

>>1737195
An interesting conclusion. Well, have fun.

>> No.1737226

>>1737208
let's go through this one more time:
a printer controller needs to: control at least 4 steppers, calculate the position of the head and extruder from gcode and load that gcode.
for very old machines that meant a direct connection to a computer that sent the motor commands over a serial connection, something an 8 bit board is perfect for. Eventually we were able to only have to send the gcode commands to the software, and for simple motion systems that works just fine, but we have limmited our max speed to around 200mm/sec of detailed motion on the printer due to processing of 8 bit boards.
the problem comes with storage, wifi connectivity, and complex motion system calculations like corexy and mesh bed leveling.

the solutions to the limited speed of 8 bit processes are: offsetting storage and network to a separate processor(octoprint), offsetting the entire motion system(klipper), or replaceing the slow processor for a fast one(32 but boards)
>octoprint
+sbc can keep commands going to printer at a high enough speed to reach 200mm/s max, wifi and ethernet options included on most sbcs, compatible out of the box, cameras
-reporting temps, filament runout checks, and baudrate can slow the 8bit; shitty optimization or boards mean a full computer is needed, no pi0ws
>klipper+octo
+in addition to octo, can calc additional features and run fast on corexy machines
-complexity inc and compatibility decreaced; only powerful sbcs can run; more data is moving over usb
>32bit boards
+can exceed 200mm/s speed, can run special modes on steppers, dead simple environment, can be dropped in with prefab configs, can easily run corexy, demonstrably better surface quality, ipcameras
-some dont include a wifi chip like the smoothie

the costs:
>octo
30$+ sbc and an sd card
>klipper
40$+ sbc, sd card, Arduino or jtag programming tool for bootloader flashing
>32 bit board
17$ board + 2$ wifi option

what feature am I missing that makes the 3A+ better than the 32bit board?

>> No.1737228

>>1737208
and as for the web interface I already explained an esp8266 can run a fast web client that'll hold up to large files. even if you love octo, which is fine (the outernet cloud connection option is nice and I did forget to mention that), the featureset of a 32 bit board should be a higher priority than a 30$ wifi upgrade.

>> No.1737249

>>1732568
Ender-3 is on sale for $174.99 (USD) btw
Not sure how long the sale will go for though

>> No.1737265

>>1737123
Yes. I use the normal V6 (0.6mm nozzle) and if I print to quickly with petg my stepper motor can't handle it and skips steps. I have switched to the volcano but I want to print even bigger and faster

>> No.1737273
File: 208 KB, 1280x958, 6C14F6A7-A8CB-4778-9048-1763A81CE8A8.jpg [View same] [iqdb] [saucenao] [google]
1737273

>>1736675
Way too little exposure. Print at 45 degrees and have 8-10s exposure after initial layer

>> No.1737277

Trianglelab inductive sensor y/n?
Never used any sort of bed leveling but I like the inductive one because it's slim, not a bulky white plastic suppository like the blt
I'm using an ender 3 with skr board btw.

>> No.1737289

>>1737152
>>1737154
Just try the cr-10 v2 out of the box.
All this massive efforts in modding it are often not worth the time and hassle, especially when you're not knowing what you are doing.
A pi cam etc. is not necessary at all for successful good prints.
Modding became a sport on its own.

>> No.1737324

>>1737226
>let's go through this one more time:
no, that's OK. I see the same realities you do, my priorities are different. Putting the front end processor onto the board vs. having it as an add-on (like integrated support to plug in any PI as a piggyback) isn't really any difference technologically. Having one processor do it all tends to make a difference in terms of code management.
As far as $20 vs $2? Dude, whatever. I just spent $100k for another fiber line into the office. because we might need it in a few months. Your reality and mine are simply different.
>what feature am I missing that makes the 3A+ better than the 32bit board?
Off the shelf component with long term company and widespread community support. *shrug*.

>> No.1737328
File: 29 KB, 500x500, 41YLD+WeFPL.jpg [View same] [iqdb] [saucenao] [google]
1737328

>>1732568
I don't even know how these work but would it be worth paying some anon to print a bunch of sponge filters?

>> No.1737337

>>1737328
...no? Sponge costs, like, nothing.

>> No.1737338

>>1737277
you mean 3d touch? works okay, just make sure you buy one with plastic probe.

>> No.1737342

>>1737338
I mean this one.
https://www.aliexpress.com/item/33006152423.html

>> No.1737350
File: 257 KB, 2018x1009, caliburn.jpg [View same] [iqdb] [saucenao] [google]
1737350

>>1737078
>Damn man, what are you printing, what's your endgame with all this?
I sell Nerfs
>How much filament are you burning through on average?
10kg per week
>How costly has it gotten?
45% expenses, but very little of that has been the printing itself. Each lulzbot mini proved very reliable for the first 8,000 print hours. But eventually the Z-axis motors wear out. And when they do they tend to take the controller board with them. Replacing those three parts proved to be expensive and not worth the effort so I'm transitioning over to these cheaper printers.
The monoprice select minis each only lasted around 800 print hours.

>> No.1737354

I walk into a thread trying to talk about all the bs involved in upgrading marlin on a cr10 and get jumped by some guy that has to defend his interface
>>1737324
>Off the shelf component with long term company and widespread community support
this is retarded, I want you to just glace at what's on /mcg/ and I want you to know that callig a pi a "component" is retarded.
> PI as a piggyback
there's a reason why prusa moved to 32bit with integrated networking.

run an astroprint farm for all I care, dont say a pi is a better choice than a board replacement, buy a melzi board and see it get smoked by the cheapest SKR money can buy. a 32bit board would be great for people who already have octo.
the reality is you've had your heads shoved up your ass so far talking about the interface "cant be run" from a 0w that you forgot I'm talking about the mainboard and how the locked down bootloader problem can be avoided by swapping the printer's brains, instead of all this reddit bullshit that people try to use in order to polish a turd. thats my reality.

>> No.1737355

>>1737350
>But eventually the Z-axis motors wear out.
! That's unexpected. There shouldn't be anything that wears that quickly in a stepper.... LDO or some other make?

>> No.1737356

>>1737354
>and get jumped by some guy that has to defend his interface
projecting.
>there's a reason why prusa moved to 32bit with integrated networking.
Yeah, just like there was a reason they stuck with the Einsy for so long. There are no absolutes here.
>the reality is you've had your heads shoved up your ass so far talking about the interface "cant be run" from a 0w
It can be. I've done it for about a year. And it's had performance issues over that year, so I'm going for a literal plug-in-a-new-component replacement that solves it. I don't even need to reflash anything, just swap the SD boot card.
> locked down bootloader problem can be avoided by swapping the printer's brains
What locked down bootloader? I can flash anything I want.
Hardware is cheap. Labor is expensive. Software is ephemeral. Interfaces last decades.

>> No.1737369

>>1737355
Because the z-axis has two motors and two lead screws I think they drift out of parallel over time and it causes binding. That bindingd slowly gets worse and the chip and power circuitry on the controller isn't specced out high enough for that eventual load. I've tried replacing just the controller or just the motors to fix the issue and it doesn't resolve the problem. They seem to go bad in sequence or at the same time.
Replacement stepper motors with matching specs to OEM range in price from $12 to $45 depending on brand/source.

https://smile.amazon.com/gp/product/B07GWRGFGY/

The controller board however can't be found any cheaper than $90. And replacing all three parts hasn't proven to be worth the hassle. The other lifecycle issues the lulzbot mini has are the cable carrier over time will wear out one or more of the smaller gauge wires inside of it. That part of the wiring harness will develop an issue and need to be replaced every 6,000 print hours. Sometimes it's just the limit switch not working. Sometimes it's the thermistor dropping out, which causes the print job to freeze up and error.

>> No.1737370

>>1737356
I mostly ran my shit from an sd card display on my printer until I bought a duet. my octoprint experience was fine, I could boot and automatically set step configs on boot and monitor from another room when I used octo to load sd card prints and I partially moved to a very old version of klipper when I finally unlocked the bootloader.
I do not care about the pi or how you move files to the printer, a 2$ chip does the same shit in my eyes, I'm fine with disagreeing on how to connect your printer to wifi, I want people to know the most worthwhile mod for their printers is 32bit.
when someone wants to add a zprobe, run their printer at 60mm/s without losing quality due to the printer processor not being able to catch up, add a probe on a locked down cr10 board, add tmc stepper drivers or use savable mesh leveling, it costs 17 bucks.
>I can flash anything I want.
creality can't, and when people spend 200$ on a printer, component price is a factor.

>> No.1737374

>>1737370
>>I can flash anything I want.
>creality can't
boo hoo. cheap is as cheap does.
> when people spend 200$ on a printer, component price is a factor.
Definitely. When you're so cheap you squeak... Sure, go for a PiZero and live with it. You recommended the anon start there, all I said was "you'll have performance issues if you do".

>> No.1737396

any way to use the Duet maestro dual stepper expansion on a normal duet 2 wifi board?
I accidentally ordered the wrong board but i would prefer to use the wifi board instead of Ethernet on the maestro

>> No.1737408

>>1737396
https://duet3d.dozuki.com/Wiki/Duet_Expansion_Breakout_Board
you need to wire this up according to the diagrams

>> No.1737414
File: 1.07 MB, 1283x1200, 1545496812644.png [View same] [iqdb] [saucenao] [google]
1737414

>>1737396
https://duet3d.dozuki.com/Wiki/Dual_Stepper_Driver_Expansion_Module
you could wire it up directly actually, you would just need to break it out yourself

>> No.1737415

>>1737414
>>1737408
wow thanks!!!!!
i will try that

>> No.1737426

>>1737289
>>1737169
thanks anons!

>> No.1737438
File: 104 KB, 1108x513, c7dd52b9d3aa78edaa150e3f709776f6.png [View same] [iqdb] [saucenao] [google]
1737438

>>1737350
What would be the few most important things you would say to someone designing a 3d printer?

>> No.1737493

>>1737438
Well, don't use t-slot aluminum as linear rails for one thing. Make changing nozzles and PTFE as easy as possible. Provide a quick thumb-release of some kind on the extruder idler. Avoid having the filament interacting with any printed parts because many novelty filament types will eat through them. Every female thread in your design will need to be metal, via inserts or nested hardware. And as much as possible make all of the major components removable on their own without having to tear down or remove whole other assemblies in order to get to them.

Automated leveling is over-rated based on my experience. It's only useful for overcoming warped/skewed buildplates, especially as print area/volume scales upwards. Which is something that could be overcome by using a much thicker buildplate, or having borosilicate glass on-top of it instead.

>> No.1737509

>>1737328
Cheaper than buying "aquarium sponge"? Sure. Cheaper than buying any other kind of sponge and cutting it to shape? Hell no.

It's probably just that you need to make sure the sponge doesn't contain chemicals that are dangerous to fish. So basically any kitchen sponge should be fine, since the only special thing fish can't take that I can think of is soap and copper. Unlikely to find any copper in a sponge, and soap is as simple as buying a sponge that doesn't come pre-soaped, which is like 99% of them.

>> No.1737510

>>1737493
Thanks a lot, i can see that you're heavily focused on the longevity/maintainability aspect, as one would expect from you, i will keep your advice in mind when designing things.

>don't use t-slot aluminum as linear rails for one thing
i don't think i have too many other affordable options, the gantry axis will be over a metre long, steel shafts are too heavy and/or not rigid enough, i suppose rollers with the extrusion but i've heard good things about PLA linear bearings

>> No.1737511

>>1737493
>It's only useful for overcoming warped/skewed buildplates
All buildplates warp as they heat up. Not only because of inherent differences inside the metal, but also because the heat is never 100% even. The edges for example get cooled more than the centre, because they're more exposed to cold air. The centre also tends to be where 90% of prints are located, which insulates that part even more.
Autoleveling still isn't _necessary_, but it can be very convenient, especially for large buildplates (300^2+)

>> No.1737514

>>1737511
Not a problem if you use TheBest™ method of leveling a bed, ie. first eyeballing it and then seeing how the first layer turns out and adjusting based on that.

>> No.1737516

>>1737514
Pretty much. I wrote a custom gcode that just homes without the autolevel and then prints .2mm thick and goes from corner to corner until I cancel it. I adjust the bed manually until it prints perfectly, then I cancel. Autolevel means that even as if drifts out of calibration it'll still print well, and then once every other month or so I can just redo it. Takes two minutes.

>> No.1737518

>>1737516
If you've got autolevel why don't you just set the proper offset and have it never drift out of calibration?

>> No.1737519
File: 321 KB, 500x500, 1525653507608.png [View same] [iqdb] [saucenao] [google]
1737519

Does heat affect induction bed sensor detection distance? I think i've had varying first layer heights with different bed temperatures

>> No.1737522

>>1737518
Because the printer has little wheels on the bottom of the buildplate and they probably turn very very slightly whenever I peel away a print. I don't know exactly why but without the autolevel the bed does drift out of level after a handful of prints.

>> No.1737525

>>1737522
Oh i see, i had somehow completely forgotten about those adjusting screws having only printed with my prusa for a year, they really ought to put locking mechanisms on the nuts.

>> No.1737537

oh we're talking about autolevelling methods, my gcode, after homing the z axis motors via a three point tilt (not compensated, just a macro that trams the bed on startup), loads up either a cold or heated bed mesh (AL bed, if I want to add a glass bed, which is smaller than my new bed, I disable and check manually) and kinda follows prusa's start gcode where it prints a line across before each print for the ideal z gap.

>>1737519
I have heard of this, but its good practice to have separate cold and hot values, so they're all relative to whichever you are using at the time, although warping of the bed for 300mm+ printers seems to be more of a reason to do so than fuzzier values. the duet can also be set to probe an n number amount of times per point to get an average.
I have no issues, but I run my probe at its rated 12 volts for accuracy and a farther detection distance, and I guess if it were to get hotter while it probes it would skew the values. for 5 volt probes I'd recommend maybe letting it heat up or probe twice and save the data that prints clean via the console and load that up each time.

>> No.1737538
File: 598 KB, 1920x1536, _SAW5999.jpg [View same] [iqdb] [saucenao] [google]
1737538

Been playing with the Prusa Mini for the last couple of days at work.

So far,
Print Quality & Speed: Excellent, no different than the I3 on PLA and PETG. I haven't tried flexible filament yet.

Interface: Crazy responsive, I can't wait till the eventual I3 MK4 upgrade happens. Octoprint is not compatible with the new board, and the prusa web interface isn't complete yet, so USB stick loading only for now.

Noise: In an office with HVAC noise, effectively silent. I could barely make out a noise from the geared extruder on retractions, but it might have something to do with that filament sensor dangling off the side.


Gripes: The nozzle crashed into the PEI sheet on initial calibration, putting a nick in the finish. (there's no first step to cal with the sheet off like the I3) No issues after moving the PINDA probe down.

The nozzle clogged during a print with Hatchbox Wood PLA. The extruder was grinding the filament down, so it may have just needed more tension, because the clog cleared easily by forcing more filament in by hand. I haven't had a chance to try again.

>> No.1737548

>>1737510
>i don't think i have too many other affordable options
Accurate linear guide carriages for t-slot are pretty expensive (upwards of $90 for those double units). PLA linear bearings would be so-so, but at least the end user could reprint them if you provide the file as open source. So if that is the approach then I suggest making the linear bearing portion a separate print from the rest of the extruder assembly. So that reprinting it and then swapping it out (or the parts that comprise it) doesn't involve disassembly of anything else. I would also recommend keeping the whole extruder unit as an modular sub-assembly. That has saved me a ton of hassle over the past year as I have spare extruders on hand for all of my printers should I run into a jam I can't clear while it's on the printer.
>the gantry axis will be over a metre long
And over that distance t-slot extrusions will have some flex/twist distortion, so it's going to matter how accurate you want your positioning to be. The double profile you have shown is less of a problem provided you don't rely on rollers.
>steel shafts are too heavy and/or not rigid enough
Rigidity depends on diameter and whether they are supported or not.

>> No.1737567

>>1737511
>Autoleveling still isn't _necessary_, but it can be very convenient, especially for large buildplates (300^2+)
Once you try mesh autoleveling, you will *never* go back. It removes 90% of the fiddling/failed prints in the first few layers and you never have to think about it.

>> No.1737568

>>1737567
that only applies to brainlets

>> No.1737570

>>1737519
>Does heat affect induction bed sensor detection distance?
It affects some sensors. Don't use those.>>1737568
>that only applies to brainlets
that applies to anyone who values their time over saving $5.

>> No.1737579

New Thread >>1737578

>> No.1737755

>>1737265
Nice tip, I didn't know I could reduce printing times that much.

>> No.1737780

>>1737538
How assembled does it come? I would think that if it was a full kit, the nozzle crashing into the PEI would be user error during assembly, but sounds like it's not the case?

>> No.1737869

>>1733080
What program is this

>> No.1737873

>>1734897
Add skirt. Make passes smaller. And thicker in cura. It will add time but it wont look like dog shit.

>> No.1739382
File: 2.40 MB, 4640x3480, IMG_20191221_040029.jpg [View same] [iqdb] [saucenao] [google]
1739382

How do I improve my print quality?

EPAX X1
elegoo grey resin
0.05mm layer height
60s bottom exposure
6 bottom layers
8s exposure per layer
180mm retraction