help-with-3dprinting

[faint sky]

You would want something beefy.

brads are for very lightweight things. for a book shelf you'll want something much thicker and something that can be used with much larger drywall screws.

brads are like .5mm or 1mm they're tiny like staples.

[iron remnant]

Oh, neat. I printed a whole bunch of LED mounts that clips to the shelves in my geekroom, except I was going for plain COB strips so there's a bit of material there to act to diffuse the light.

[misty gyro]

Chiming in here (again), you need to be careful when mounting things to a wall. In particular sheetrock isn't all that strong. If you're mounting anything heavy you really want to use a stud-finder and put a decent sized nail or screw into a stud.

[iron remnant]

The dad joke of pointing the stud-finder at yourself is, of course, entirely optional.

[coarse coral]

Yeh I was thinking on the lines of angle brackets as a triangle to reach out to the distance but I doubt it would support. I'll maybe use the 3d printer as a solution for this but likely it may end up within a cabinet since a cabinet will act as an enclosure at the same time

[coarse coral]

Yeh it needs a good amount of attention to think it over

[coarse coral]

I'm just going to go with @iron remnant s advice

[iron remnant]

So, here's what I ended up doing

It's all 2020 extrusion that I got pre-cut from Misumi USA. So the goal was to hold a printer (I sized it so it'll fit my current printer or the larger printer I'm probably going to upgrade it to) and all of the accessory bits like filament, plus my networking gear and linux box, plus whatever else happens to fit.

For me, at least, it's a lot easier to switch to a fully enclosed printer instead of making the printer space enclosed. The doors would smack into the adjacent shelving (it fits in a corner) and also part of the advantage of an enclosure is that the print is stronger and warps less because the enclosure holds in the heat.

And then the nice thing about 2020 extrusions is that they are open to a lot of modifications. So I've got a Honeycomb Storage Wall 2020 mount that I need to post at some point, plus a lot of cable management and DIN rail mounting stuff.

[acoustic quiver]

Hi there, I have an Elegoo Neptune 4 Pro, and am currently trying to do PID tuning on the heat bed and at the end of the process it comes up with an error saying !! pid_calibrate interrupted and Can not update MCU 'mcu' config as it is shutdown

Has anyone else experienced this issue? or have any pointers on how to resolve it?

I am running Firmware Version V1.1.2.03 and UI Version 1.2.4

[coarse coral]

This is a well thought out stand for a single printer I like it a lot, and also could be good because u changed my thinking that the printer would be on display in my lounge since it's pretty quiet but having this type of solution means there's space for filament it can be enclosed and further to that I have a wall cupboard that it would sit really well within which is external vented. This mean I can line up my print and forget about it with all those benefits plus the benefit of not looking at the print after first layer until it's finished (developed this habit with printers I had trouble with).

Neat! Ideas!

[coarse coral]

Sorry I have a Neptune 3 plus but haven't did pid tuning yet

[faint sky]

Just happened to me. Over time the nozzle worked its way loose and I found it inside the spaghetti mess. Both the nozzle and sock came off. I know for a fact I tightened the heck out of the nozzle when I installed it, and it was on a brand new extruder that never had filament through it before. This will likely be a recurring issue as it's happened to both my sprite extruders now.

This is likely another culprint for when people get giant molten blobs around their hotends. It requires manual labor to clean up (while the hotend is hot).

I think it's due to the way filament is twisted or when the nozzle moves over a part and meets rotational friction, it slowly loosens over time.

I had my printer dialed in extremely well so when I had to raise the z-offset by .5mm and noticed molten filament coming out the side of the silicon sock, should have been my first clue. 5 minutes later the nozzle and sock and were on the bed surface along with some spaghetti when I came to check on it.

Now I'm thinking that nozzles need some way of keeping them in place like a cotter pin or locking mechanism because this is the 2nd brand new extruder this has happened with.

Yes I have 2 sprite extruders. 1 is for .4 nozzle and the other for .8. It's much easier to swap out the entire extruder assembly than just the nozzle.

[misty oasis]

Ok hello again,
I had a friend come over who’s experienced and he couldn’t figure it out either. Something’s wrong with the balancing of the heat bed he said. Unsure what that means, and now unsure what to do next.

[faint sky]

Tighent the vrollers on the y-axis (front to back) bottom? With too much free play on the vrollers it might make the carriage tilt side to side.

[vocal flax]

I'm having problem with bed leveling left of the bed always low compared to right side. Its i3 style bed slinger. I've tried leveling gantry (both manually with a caliper and g29 (individually driven z). Trammed bed both manually with a paper on 4 corners and with tramming assistant in marlin. Tried glass bed and pp bed, (glass unfortunately shattered can't show it right now). I am out of ideas.

Any ideas on what might be causing this

[faint sky]

What printer specifically? Left side definitely a hair lower than right. Does it have automatic leveling and/or adjusting screws?

It's not off by much but noticeable enough to cause quality issues later in a print. Off by about .15mm

Does it have a pinda style probe for autoleveling and bed mesh?

[vocal flax]

I have automatic leveling and adjusting screws

Ive already said it's i3 clone. Prusa style bedslinger

It has a touch probe with opto interrupter

[faint sky]

Do manual probing with a sheet of paper starting in the middle first. Then do the outside corners, repeat that a couple times until you've got everything trammed like you want. Then do an auto bed probe. The more level you can make the entire bed prior to bed probing the better.

[vocal flax]

M48 gives good enough results

As I said, tried both manual tramming and assisted tramming

[faint sky]

You mentioned gantry though, are you concerned it's not square somehow?

ok then AFTER the bed mesh is created then you can still tweak it with the adjusting screws. bring that left side up a little?

[vocal flax]

I've tried to level gantry to bed to y rods and to motors as reference points all same result

[vocal flax]

Then next print the left side is low again.

Using auto bed level every print

[faint sky]

does it probe all corners prior to every print?

ahhh that's what's doing it.

[vocal flax]

It probes 4x4 matrix every print

[faint sky]

in which case yes i'd be concerned with the gantry being a hair higher on the left.

[vocal flax]

I've tried 3 different method of doing that no luck

[faint sky]

fixing a .15 or so deviation with the gantry isn't trivial... but the bed probe should accomodate for that, so your bed probing isn't working as it should either.

[vocal flax]

The error is repeatable

And m48 gives good enough results

[faint sky]

i would do a bed probe, make a mesh, then disable it from doing it every print. that way you can make the screw adjustment once and then never touch it again.

i have no experience with i3 only enders, unsure how that will translate.

[vocal flax]

That's a workaround and probably a good workaround but it doesn't solve the problem

[faint sky]

it does if you never touch it again and only use bed probing when something changes. that's how i use my S1 pro and works quite well.

[vocal flax]

I3 is the same thing but it has real linear rods than v grove sigma

[faint sky]

my bed probe only comes down in the center once per print just to verify bed height, and uses the same existing mesh every time.

there are others here with rod experience that hopefully with jump in. i've never owned one with rods, well except the Anet A8 but that was a decade ago.

[vocal flax]

It's not a solution, it's a workaround and credits to you it's a good one. Unfortunately, I have 2 different bed surfaces, glass and pp sheet. And I sometimes use heated sometimes not heated. I have 3 combinations of bed.

[faint sky]

yeah sorry that's all i have for you 😦

[vocal flax]

It doesn't work for me. Also, this printer is a hobby itself, so would like to solve it

[vocal flax]

No worries thanks

[faint sky]

maybe try Prusa support discord? there's gotta be some specialists out there.

[vocal flax]

My thing is very close to anet A8 tbh

[vocal flax]

I tried 608 wasteland, I guess I can try my luck there

[faint sky]

problem i had with the A8 was the rods warped and rusted over time. rods only work when they're perfectly straight and cheap rods just destroyed the entire thing.

not to mention the entire thing was a complete fire hazard waiting to happen

[vocal flax]

Oh I don't have that problem, mine are alloy and has thick plating or something. Could be outside induction hardened I don't remember.

Haven't seen any wear or rust on rods for 10+ years

[faint sky]

well i used it in a garage, in florida, high temperature, high humidity, turned everything into rust within a year. it was cheap cheap horrible quality A8.

[vocal flax]

The marlin config is my own no fire hazards here 👌

[faint sky]

the fire hazard was the A8 PSU and wiring

[vocal flax]

Custom built, bought piece by piece

[faint sky]

very nice! hope you figure it out. if you do please come back and let me know what fixed it. all i can think of is tricking the bed probe.

[vocal flax]

My prime suspect is rigid couplers, which has no basis, I just am out of options and change everything I can

If that doesn't solve Ill turn table 180 degrees to put double bearing side on the right

Again has no evidence these could be it

Actually I might do 2nd first because it makes sense to have something wrong down than up

[misty ibex]

@faint sky did you finish your new infinity mirror?

[coarse coral]

Left just looks off. I don't know if you have build plate tighteners but why don't you just lower right a touch and bring up left?

Seems quite straight forward I know.

[faint sky]

No it's in the pile of unfinished projects. Had to take a vacation, then work on halloween stuff, then get back to it. Month went by and forgot to get back to it.

[misty ibex]

I hate it when that happens...

[balmy pulsar]

never happens to me.....

runs

[torpid jasper]

“Pile of unfinished projects” implies a “pile of finished projects” and frankly I’m not sure such a thing exists

[balmy pulsar]

no such thing as a finished project

[vocal flax]

As I explained again and again the error is repeatable, if I adjust left gets low again

[coarse coral]

Maybe your springs are weakened

[vocal flax]

Springs are tight. I've tried very compressed less compressed different ways of tramming

Result is same always

[shy kelp]

Something like tpu or silicone, but cheaper?
Latex is cheaper, but it's a pain to work with, if you need some thickness

[tardy juniper]

Hey, so I have a CR10 Smart but for some reason, it is autoleveling but then ignoring the auto level data when it goes to print

I cant figure out whats causing this

Also when it prints I have to set my z offset rediculously high which I think is also due to the same issue

[faint sky]

You must add gcode to your slicer machine start gcode settings (in Cura) to use the bed mesh otherwise it will ignore it.

Each sliced model gcode file is embedded with the code to use the bed mesh or not.

For my S1 Pro it never used the bed mesh until I added M420 S1 Z10

You have to look up your slicer software settings for your printer and figure out how to add the gcode to use bed mesh. M420 S1 Z10 only applies to Marlin. If you're using a Bambu or Klipper you have to figure out how to do that on a software basis.

[tardy juniper]

Thanks thats perfect

I think thats exactly what I was missing. No idea why I couldn't find it anywhere online

[faint sky]

Use G28 prior to that to home all axes first too

[tardy juniper]

Yeah

So G28 and then G29 and then M420 S1 Z10

[faint sky]

Remove G29 in favor of G28 and M420 S1 Z10 both.

[tardy juniper]

Oh so no g29?

M420 S1 Z10 will auto level?

I thought that only told it to use the saved auto level mesh

[faint sky]

G29 does a bed probe prior to each print which is unnecessary if you do M420 instead

[tardy juniper]

Will M420 auto level?

[faint sky]

So do an autoleveling with your control pad. Add G28 and M420 lines and it will use your last autoleveling bed mesh.

[tardy juniper]

Ok

[faint sky]

G28 will home all axes and do 1 single probe in the center to confirm bed height. Then it will go to the closest left corner, use the M420 bed mesh, and start on purge lines when nozzle temp and bed temp are reached.

This makes it so you only ever have to use autoleveling when something changes. It will never do a full 16 point (or how ever many probe points you have) again until you manually do it again. It saves so much time prior to each print.

When you do it for the first time have your hand on that z-offset compensation to move it up/down on your first print. As it's printing the purge lines that's when you do your final z-offset compensation.

It's mostly to help avoid bed scraping but also will ensure your first layer goes down at just the right z-offset you want.

From then on both your bed mesh and z-offset will remain the same every print... and you can print forever like that.

M420 itself does not do autoleveling no, it only uses the mesh from your last autolevel probing.

As long as you have that line in your printer settings it's exported with every gcode file you create.

Therefore your printer interprets that as always use last saved bed mesh.

You can redo autoleveling at any time and your gcode file will use that new bed mesh instead. Autoleveling creates the bed mesh.

The first thing I do is a manual bed level, get everything as perfect as you can, then do autoleveling for the bed mesh.

Your manual leveling cannot be more than 1mm off on any corner. Autoleveling cannot compensate more than 1mm deviation.

[tardy juniper]

Ok so I left in the g29 just to get it to level before printing (using m420), but it seems to not compensate still

Like the z axis screw is just stationary until it goes to the next layer

[tardy juniper]

Thats how my prints used to look 😭

Is it possible that since it g29'd in the same print as when it tried to load the mesh from eeprom that there wasn't a mesh?

Also is there a separate command for saving the mesh to eeprom or does g29 do that by default

[faint sky]

unlikely but with this method you can redo autoleveling as many times as you want to create new bed meshes and the cgode will use the new one.

[tardy juniper]

I see

Ok I'll give that a try

[faint sky]

https://marlinfw.org/docs/gcode/M420.html

[tardy juniper]

Also not sure if this helps, but my z offset is currently at the max value

And if I decrease it at all the nozzle is too close

[faint sky]

do you have increment set to 10mm?

[tardy juniper]

For the z offset?

No

[faint sky]

use .1 or .5 increments

[tardy juniper]

I think its value is 1.5mm rn

The value is called "z-axis compensation"

[faint sky]

https://www.youtube.com/watch?v=q31XVAvAbOI

Do not use this video for guidance, it's complete garbage and extremely deceptive.

[tardy juniper]

🤣

I haven't seen that one honestly

But I'm not suprised, crealitys products are kinda trash lol

[faint sky]

Your touchpad is setup a lot like my S1 Pro.

[tardy juniper]

It might be that

One sec

[faint sky]

Not sure what you mean by max value... I've never maxed out z-offset compensation.

[tardy juniper]

Oh wait

Its a S1 pro

Not a cr10 smart

[faint sky]

🤦‍♂️

[tardy juniper]

😅

This one

Thats what I have lol

[faint sky]

https://www.youtube.com/watch?v=GmhBYOEb-ro

[tardy juniper]

Oh wait no sorry

Yeah its a cr10 smart

[faint sky]

That video by Ricky Impey is really good and should be close enough to adapt for your uses since your touch pad is very similar.

[tardy juniper]

Ok

Thanks

[faint sky]

and for people who have odd leveling issues due to gantry alignment issues there's this one to pay attention to.

https://www.youtube.com/watch?v=9eQ_rBzfm5s

It's easier for me to link videos than to write paragraphs on this stuff over and over. If you have a printer issue the first thing you should be doing is scouring youtube for videos on the subject. Most everything I've learned about 3D printing has been from Youtube.

[tardy juniper]

Oh yeah for sure

My issue was just finding anyone who was experiencing the same issue 💀

But I do think the links you sent will give me the right direction, so thx

[faint sky]

I hope so, good luck. If you can't figure it out I'm always here.

[tardy juniper]

For sure, thanks!

[faint sky]

Finished installing the printed u-channels for the side-lit strips. Took about 10 hours of printing. Has a PIR sensor for motion activation.

[vocal flax]

@tardy juniper if you go devons' route make sure you use store settings option. Default marlin option is to scrap any bed leveling data prior to homing and m420 brings values from eeprom (or equivalent) which is normally zeroed out during flashing

[faint sky]

G28 doesn't delete a bed mesh it only disables it. If the very next command is either G29 or M420 it is reenabled. So the combination of G28 and M420 is homing, disable bed mesh, enable bed mesh. If you use G29 it will take a new bed mesh prior to each print. If you use M420 as long as you don't physically change anything the same bed mesh will always be valid. This is why I do not touch the bed surface AT ALL, EVER, and I certainly don't remove and flex the bed surface to pop off prints. If it ain't broke don't touch it. M420 will work forever.

However as I've recently discovered you do have to account for possible nozzle unscrewing itself. I didn't believe that was a real thing until it happened to me. When it started unscrewing it lowered the z-offset the more it came unscrewed and yeah it gouged my bed a tiny bit in one spot. So even if you do everything perfectly correct, things can still go wrong.

[coarse coral]

I opted to disassemble my printer and rebox for moving and I only had one job but managed to break one of the small wires in the process

[vocal flax]

If you haven't store it on non volatile memory, m420 brings values from non volatile memory (which is erased when freshly flashed) then you lost the bed leveling information. I didn't said g29 erases it.

[faint sky]

How often do you flash your firmware?

[vocal flax]

If he never leveled the bed and stored the settings (write to non volatile memory) it's safe to assume no data (all 0es) pulled from memory

[faint sky]

Some of the newer enders don't require manually storing to eeprom. It's done automatically during autoleveling.

Which is why you want to manually level as perfectly as you can before running autoleveling. After that is the final adjustment of z-offset while the purge lines are going down on your first print.

He said he did level the bed manually and with autolevel but his prints weren't using the bed mesh because he was missing M420.

The reason I'm against using G29 every time is the probe will rarely measure the same data every single probe, there's variability. You remove the variability with M420 and can adjust to that manually if needed. I know it sounds weird but it works. G29 can have you chasing your own tail sometimes when the prints come out slightly different every time and that's because it's taking a new bed mesh every time prior to printing.

In an ideal world G29 results will be identical every time. In reality, in my experience, far from it.

[coarse coral]

https://youtube.com/shorts/E3FjnnP8z1M?si=5iiF9jDszz6HUC0W

His first print is so clean

[vocal flax]

G29 is not supposed to be identical every time because thermal expansion (of bed and many other things on printer) is random (enough).

Not leveling every time will be good enough for 3d printing

Also you can test your probe to be good enough for the application with m48

[faint sky]

That's true if you're experimenting with different filaments. Good point. I've been using the same stuff for so long I don't have to touch anything.

If I were to go back to printing with PETG yup I'd have to recalibrate again.

Different nozzle temp with different filaments = different viscosity which would require z-offset changes at the very least.

[oak cairn]

gex

[faint sky]

Modified a Pi 5 model to include their active cooler.

Not perfect but should be close enough for an enclosure design.

[balmy pulsar]

radus on the fan hub too big. also doesnt have the ejector pin mark. fail

😛

looks nice

[balmy pulsar]

haha

[faint sky]

Yes I completely guessed on everything, didn't measure a thing with calipers. 😛

[balmy pulsar]

hehe

yeah. im used to being given a couple bad photos of a car and expected to make something that seamlessly transitions from live action...

these days we have scanners, so much nicer

[faint sky]

I'm too lazy for that. Only needed a rough estimation. The enclosure is going to be about 3x the size of the pi because of an external SSD.

I have a stupid idea of making a kind of windmill from the exhaust fan.

[balmy pulsar]

yeah. its nice having models for making stuff. so glad so may suppliers give them out

windmill haha

[balmy pulsar]

i want a wind mill. my area gets frequent 100+kph winds.

ooh, a decorative one

[faint sky]

kind of something like that where the air gets directed into a cone that blows a rotating thingy.

[balmy pulsar]

why a cone? to accelerate the air?

[faint sky]

a whimsical enclosure

yes because the fan isn't that powerful so the heat will help act like a geothermal vent along with directed airflow... that's the idea anyway.

[balmy pulsar]

hm

[faint sky]

but it won't be like a real windmill because there aren't enough blades

[balmy pulsar]

and theres no mill

[faint sky]

will be more like a paddlewheel

[balmy pulsar]

now im lost.

haha

[balmy pulsar]

the wind turns the fan blades. then mechanically spins the paddle wheel, whuck paddles waht?

a creek?

[faint sky]

yes it will be stupid but it will be my kind of stupid

[balmy pulsar]

ok

i thought maybe it would go in the bathtub, haha

bird water fountain

[faint sky]

there will be no water involved, pi's are hard enough to get as it is.

[balmy pulsar]

hahahaha

[faint sky]

hoping the cone will increase velocity like a turbine

[balmy pulsar]

indoors?

im still lost. a fan (from a pc?) is gound to blow air on another fan in hopes of spinning it?

*brain broken, does not compute, NAN

i am trying to launch my keycaps later today. got a few more things to do, take a bunch of pictures etc

[elder oxide]

The packaging is really nice

[balmy pulsar]

thanks

im pleased with it mostly. lots of little things to improve next go around. but for this week, im done haha

my cnc machine (one servo) died while making a vacuum fixture to cut these out. so i ran out and bought a cricut.

actually not that bad a machine haha

[faint sky]

General idea, it's impractical but I like the idea because it's impractical.

It's a Pi Wheel case...

[balmy pulsar]

oooh

one problem

the fan on the pi blows DOWN

[faint sky]

shut up no

[balmy pulsar]

haha

[faint sky]

😢

[balmy pulsar]

oops

sorry

maybe you can duct it

[faint sky]

I still don't have a way to power it on so didn't know that lol

[balmy pulsar]

yeah, fans blow into the heat sink then the air exhauts out the sides

[faint sky]

glad i didn't go any farther with the design you saved me a lot of effort. was going to make a little cottage around it and everything. 😦

[balmy pulsar]

eek

i bet you still can. just need to harness the air differently

that type of paddle wheel is for water as well. youd need something that will atch air better

[faint sky]

Can always add an external 5V blowing up but that would defeat the purpose. Maybe can do it on top of the SSD I'm sure that'll get warm.

yes was just a rough design to show the concept

[balmy pulsar]

ah

pic of the real pi fan?

im betting it also blows air on the heat sink beside it

[faint sky]

i posted one, scroll up, already installed.

[balmy pulsar]

cant see the side

[faint sky]

There is no real holes on the side which is why I thought it blew up.

it's just a square tin can all the way around

[balmy pulsar]

...

[balmy pulsar]

it must all come out the bottom

[faint sky]

eh tried powering it via usb c from the pc but didn't like that. spun the fan for a second. it spins clockwise.

[balmy pulsar]

no im right, its on the side with the fins.

[faint sky]

aww crap

[balmy pulsar]

thats really kinda a terrible deisgn haha

weird

[faint sky]

right, I would fully expect it to blow up but they're the ones that have put a ton of R&D into it so they'd know best.

too used to PC exhaust fans i guess

there's no slots to blow it toward the heat sink either. it's just to put airflow across the entire metal plate.

🤷

[balmy pulsar]

no, would never blow upward. that creates negative pressure under the fan and would not cool anything.

same reason your pc case fans blow in

with the rear exhaust being the optional one

but the fins on the heat sink are sideways. obstructing the fan exhaust

just makes it look like an afterthought

[faint sky]

if there was enough vent space to draw heat up below the can that would work. i guess they wanted as low profile a solution as possible.

[elder oxide]

It’s easier to push air through than pull it through

[balmy pulsar]

no. fans always blow into the heat sink

[elder oxide]

Physics are weird lol

[balmy pulsar]

never the other way

[faint sky]

if it's in an enclosure then it's just going to recirculate the warm air inside back into the fan.... i dunno.

[elder oxide]

It all has to do with the pressure system created by the fan and it’s contact area with the heat sink

[balmy pulsar]

your enclosure needs to guide the air in and out

[faint sky]

knowing this will change my case design a little to ensure it can suck outside air in from above the fan.

[balmy pulsar]

otherwise the fan will serve no purpose.. liek on a steam deck :x

[faint sky]

making an enclosure for this just got a little more complicated :/

[balmy pulsar]

you an also make your printed case shroud the fins, guiding the are to an outlet

[faint sky]

i have some mesh grill for pc fans, can just make something to incorporate some mesh.

[balmy pulsar]

real enclosures are serious business. they may loock like a tin box, but they take into account a lot of physics

i think most pi enclosures are just boxes though, haha

[faint sky]

yeah most enclosures i've seen are purely aesthetic.

[balmy pulsar]

yup

i remember the first ibm server we got. was the first case id ever seen where the maker clearly took airflow seriously. several injection moulded ducts that channeled air

as opposed to the typical row of 40mm fans that did very little

[faint sky]

the problem with ducting in 3d printing is removing the support material

[balmy pulsar]

open side ducts. channels. not tubes

the pcb is the 4th side of the duct

in the context of a pi case, it would be like ribs

that are in highly specific locations

[iron remnant]

I had a bunch of aspirations of doing a more elaborate DIN mount Pi case with fans and ducting and stuff but instead I just printed a simple one and never got around to doing the fancy version.

I did do, long ago, a mockup for what a modern chassis designed to look and be shaped like the Mac IIci/cx/Quadro 700 case from the days of yore and at least one guy nagged me because he really really wanted me to make it real.

And, like, sure there's plenty of mac-esque pi cases but the scale is all wrong.

[faint sky]

Finding a square display that exactly matches the Mac would probably be the hardest thing. Unless you did some kind of CRT bodge but then the Pi would be inside close to the CRT and get a ton of interference.

I think it would be cool to make a modern version with a widescreen with HDMI. A modern take on a classic. Nothing wrong with that in my opinion.

Just don't post it to Mac enthusiast forums, they'd get frothy.

[iron remnant]

Well, I was mostly going for the afforementioned IIci/IIcx/Quadro design.

If I wanted to make the Mac enthusiasts frothy, I'd make a replica IBM PS/2 Model 25 but with Apple logo and colors.

[faint sky]

or just build an m-ATX mid case PC and slap an apple logo on it with a Windows 11 desktop.

I think my build volume might be enough to print an entire Mac classic. Problem is I really don't feel like squinting anymore and prefer the really big monitors of today.

[iron remnant]

Well, the notion I had was to get a MicroATX motherboard, a SFX power supply, and space for the usual accessories and a fully faithful implementation of the design guidelines.

Unfortunately it's still 400mm on one axis.

[faint sky]

print in half, solder melt it together, bondo, sand, paint?

[iron remnant]

Actually, no checking the file, it's 320x400mm so it's gotta be drawn and quartered still no matter what.

Assuming anything smaller than a 350mm x 350mm bed.

[faint sky]

Most ideas sound way better in the mind, sometimes just have to let ideas go. I have a desk full of half completed projects from always getting sidetracked with new ideas.

[iron remnant]

Yeah, like, I designed it as a learning project.

And I posted photos in case it eventually would cause someone else to go through the trouble of launching a kickstarter for the best retro Apple case design.

And, really, I don't have room for a random extra PC solely because I wanted something cute like a Mac IIcx.

[coarse coral]

Will putting my 3d printer in a closet which isnt much more in size than the width and depth of the printer have similar benefits to an enclosure?

Does vertical space play a part in this, it's basically a floor to ceiling closet usually for hanging coats.

I'm thinking of just making a shelf for the printer, nothing much really.

[coarse coral]

Sorry but this design although it may be total genius still seems hilarious

[iron remnant]

So, check this out: https://www.youtube.com/watch?v=WrQiUL0jAzc

A closet is going to give you some potential benefit from microplastic and VOC pollution

It's probably not going to maintain the heat like an enclosed printer would.

But also small aerosol particles and gasses are not necessarily easy to wrap your brain around the transport of, as evidenced by COVID and some of the less effective measures of protection.

[coarse coral]

Thanks @iron remnant looks like anything where ambient temps are close to printer temps delamination is less and strength increases. Looks like the printer will be in the closet for 3 reasons now. It doesn't take up floor space and as above.

[faint sky]

I wouldn't put it in a closet that has your clothes as a fine particles will coat everything eventually. An empty unused closest would be ok.

[faint sky]

A Ring Doorbell Enclosure that acts more like a wireless security camera by hiding the doorbell light.

[coarse coral]

Yep this will be a single purpose closet (it's actually a walk closet rather than IKEA) and it looks like this could be the ideal purpose.

[faint sky]

and I made it backwards. It's supposed to be angled in the other direction. Noticed after the 7 hour print finished. 🤦‍♂️

[coarse coral]

The plus is i can't imagine it used many grams of plastic?

[arctic dragon]

It’s still 7 hours of print time.

[faint sky]

It's actually quite thick. It used a lot of plastic. Finally nailed it on my 3rd print.

The black angle bracket that comes with it has a weird curve to it. It's not a perfect radius, it's a little squared off. Eventually dialed it in but it took more revisions and plastic than I wanted. Spent about half a spool getting it right. That's the way it goes sometimes.

and the .8 nozzle doesn't do curves very well 😛

[radiant bridge]

Can anyone provide the physical measurements for https://www.adafruit.com/product/420 ? I'm looking to design and print an eclosure for this and the Matrix Portal S3. The data sheet does not have it included

[vestal thicket]

The S3 Matrix Portal 3D CAD is here: https://github.com/adafruit/Adafruit_CAD_Parts/tree/main/5778 Matrix Portal S3 but I don't see the matrix panel you linked to

[radiant bridge]

Thank you, I am looking for the display as well

[faint sky]

Dimensions: 192mm x 96mm x 12mm (7.6" x 3.8" x 0.5")

a link to the product datasheet is at the bottom of the product page too. 😉

I know there are P4 & P5 brackets but I don't think anyone's done P6 brackets yet if that's where you're headed with it.

and there's no actual 3D models of matrix panels yet to my knowledge.

[radiant bridge]

The product data sheet does not show the dimensions, even the one on digikey shows the gerber files for the matrix portal and not the led matrix.

[coarse coral]

True

I don't print much at all so my lack of patience with 3d printing has returned to newbie levels again

[faint sky]

The dimensions are different at the bottom of the page depending on which pitch you choose. So those dimensions should be accurate for the P6 panel.

[radiant bridge]

was hoping to find something like this that I could use accurately create a case around https://learn.adafruit.com/assets/117190

for the display, will break out a ruler and calipers and test print some I guess.

[coarse coral]

I actually came up a solution to just screw a ring doorbell down using metal hinge pieces this was like 2 years ago. I sprayed all the brass black and the total cost was under £4

Somewhere I have a photo

[faint sky]

The goal of my project was to cover the doorbell button and light. Yeah a metal case could do but I have a 3D printer and it's printing season. 🙂

[coarse coral]

Keep on smashing it 💪

[faint sky]

I'm confused, you want to try to use the ISFL31 driver to drive a matrix panel? I don't think that's going to work. You want a Matrix Portal S3, it'll make your life much easier.

[radiant bridge]

nope, was just an example for the size/image/measurements I was looking for

[radiant bridge]

I will do some simple modeling and post a link

[hollow blade]

best 3d printer for value or are 3d printing shops better?

[toxic kite]

value depends on how much fiddling you want to deal with I suppose

[coarse coral]

If you want large build size and lower price I think Neptune 3 or 4 plus is a good purchase. Faster printing speeds on the 4 now also

Personally I am done with creality products

[arctic dragon]

Also worth taking into account is the amount of printing you anticipate, and the timeframes you would need your parts in. There’s unfortunately too many variables to have a single right answer.

[faint sky]

There's also resin printers if you want higher resolution detail and don't mind a smaller build volume.

I'll agree with the Neptune 4. It's a good value for the money and comes with Klipper for higher speed printing.

The 3D printer marketplace is so saturated that most often you get what you pay for.

If you buy anything except a Prusa be prepared to have no manufacturer tech support.

[coarse coral]

I'm impressed at Neptune's pretty solid build quality and overall aesthetic. It's a nice printer for the price. My Neptune 3 just sorta works and the first levelling calibration I was very impressed with it.

But yeh if I could swap for a Neptune 4 I totally would.

[vivid helm]

... and sticky, stinky resin.
... and hundreds of dollars in consumables.
... and reduced versatility.

All uncured SLA resins are toxic.

[shy kelp]

... And waste that can't be just throwed
... And it's a bio-hazard
... And in the stinkless resin, you don't know what actually is
... And it's a pain to cure it right
... And it's a pain to generally handle it

[balmy pulsar]

resin haters

[nova minnow]

henloa

does anyone know why my print is showing that it'll be full of holes

when the fusion file is completely solid

[faint sky]

Might be z-seam points?

[vivid helm]

Yeah those are just seams.

[nova minnow]

ahhh

will it leave a hole or no?

[faint sky]

it can produce an artifiact which is why the slicer is letting you know where they'll be. depends on how dialed in your printer is.

[nova minnow]

aslong as it doesn't leave a hole the whole way through it'll be fine

[faint sky]

typically it's the opposite and will leave a little bump

[nova minnow]

alright good

i can just fix that after

[faint sky]

I just found out about card scrapers and burnishing tool to scrape artifacts away instead of sanding. Just got my card scrapers in. They're not even sharpened and make a huge difference.

[nova minnow]

well its just a prototype for something im working on

i brought a massive spool of recycled filament from old plastic bottles

[faint sky]

https://www.youtube.com/watch?v=a4e9JJkinvk

[nova minnow]

so im making a straw that you can slot into itself

properly

perfect size for any drink container

[faint sky]

I make big stuff and these are really handy.

[nova minnow]

but i just need to see if my design works

looks pretty good so far

[faint sky]

interesting

[nova minnow]

yes i noticed a design flaw

that hole you see on the bottom right

wont be there soon

imma just extrude it cause i dont need clearance there

[faint sky]

getting the infill out of that channel will be challenging.

i mean support material not infill

[nova minnow]

first thing i do is try printing it without supports

if it breaks then i add them

it seems like it should be fine

in fairness

to make it easier to print i sloped it kinda

so its not like a solid overhang

it kinda builds itself up as it goes

[shy kelp]

You can even put it in the oven with grainy stuff like salt
I wonder if you can get decent effect with a brush and a power tool
A lot of microplastics...

[faint sky]

Sometimes you can print on air but I wouldn't count on it. If you want to ensure you can do the channel without supports then make the walls 45 degrees between the parts.

[balmy pulsar]

usually you will see a small over extrusion bump from transitions. no bid deal.

[nova minnow]

i'll probably just bend some sheet metal if i have to

put that in there

break the support out

it'll be fine

[shy kelp]

There are also arc overhangs

[nova minnow]

i hope it prints without supports because the whole idea is to use as little plastic as possible

[faint sky]

I have a 90 degree pick tool that would be good at something like that.

[nova minnow]

it's supposed to be a recycled thing

[shy kelp]

You may use the "lighting" infill with a thick shell then

[nova minnow]

i might add custom supports

in my cad

[shy kelp]

Or just slice it in two

[nova minnow]

just so they're easy

to remove

[balmy pulsar]

the salt thing is basically a recast. You are effectively making a cast part using the original part as the pattern and salt as the mold.

[nova minnow]

cause super slicer makes a royal mess of anything support related

[faint sky]

Pick tool set can come in handy for getting hard to reach support material out.

[nova minnow]

oh i have them already

[balmy pulsar]

teeeny long nose and curved pliers too

[nova minnow]

i meanit's about there rn

the bridge over the little slot came out fine

[faint sky]

i meant the channel on the first part you showed

[nova minnow]

im doing it at 0.08mm cause its supposed to be water tight

weather it will be or not is a separate question

[balmy pulsar]

youll never get water tight on a print. ive tried

well, not "never" but i wouldnt rely on it

[faint sky]

that's a very fine setting. keep an eye on it for excess buildup around the nozzle.

[nova minnow]

tyty

[nova minnow]

dw it's less of a water tight test rn more of a does the design work

test

[balmy pulsar]

what is the nozzle?

[nova minnow]

i'll worry about the water tight part when i get to it

[nova minnow]

0.2

[balmy pulsar]

oof

good luck

hehe

[faint sky]

yeah might be possible with flow rate adjustment, i've never gone below 0.12

[balmy pulsar]

ive done a lot of 0.15mm layers with a 0.3mm nozzle. it clogs a LOT

[faint sky]

yup that's another concern, very likely to clog

[balmy pulsar]

it does work, but id say every 2nd small print is a fail

[faint sky]

there are smaller nozzle sizes for finer detail but i don't have an personal experience with them.

building a 0.08 profile sounds like a challenge

[balmy pulsar]

i have some 0.2mm in a box. ive gone full on the other way though, i like 0.8mm nozzles now.

for building larger structural parts fast

[faint sky]

same

for industrial boxes, enclosures, etc.. I'll take function over form. The .8 nozzle puts down some thick lines.

[nova minnow]

i have a superstition about my printer

if i dont listen to money for nothing whilst it prints

it always fails

every time

when i put it on

it comes out perfect

every single time

[faint sky]

lol they all have their own personalities for sure. keep it happy and it will keep you happy.

[nova minnow]

yes

my printer just loves Dire Straits

i mean the start of money for nothing is 100% what makes it print right

just the best intro ever

after them asking for their MTV

yk the bit

[balmy pulsar]

i feed mine cookies. and pizza.

[nova minnow]

wooooooooo

[shy kelp]

I am using my k1 as shoe drier
It never ends to clog
Literally my 10 year old ctc i3 wood gives me less problems

As they say when you go cheap you end up spending more huh

[balmy pulsar]

im all about the cheap

haha

or even better

ha

[acoustic quiver]

Hi there, I can create a thread for this issue (as it may get a little extensive lol) but, I am using an ELEGOO Neptune 4 Pro, trying to do some Pressure Advance calibration using the Pressure Advance Tower test built-in to OrcaSlicer and running into some issues, I can specify further but I just want to gauge first if anyone has experience with that or may be able to help, thanks in advance! 🙂

I am dealing with underextrusion in the corners of the test print. A theory is that the firmware is not respecting Pressure Advance changes via G-Code

Sorry if the quality isn't great, but you can see that in the back corners light is clearly passing through, and in the front corners there is definite underextrusion and stringyness

I am probably going to run a test following the Klipper Documentation rather than the test built in to OrcaSlicer, and see if maybe I have some settings wrong, or missing a step or something. I will update with how that goes, but any input or recommendations would still be greatly appreciated!

[acoustic quiver]

I am currently running the test with the provided STL from the Klipper documentation. The corners look much sharper so here's to hoping this test works. I am wondering if maybe I am supposed to issue the TUNING_TOWER_COMMAND stuff or whatever, but am unsure because I'm pretty sure I ran this same test a couple weeks ago with the built-in OrcaSlicer test without issuing any commands and the corners came out fine back then.

I am unsure as to what the issue could be there, and can only hope I didn't mess up a setting somewhere and don't realize lol. As far as I know the only things I have changed are printing temp, Flow Rate, Rotation Distance, and that should be it

[inland galleon]

any 3d modelers around? trying to figure out how best to 'fold' the highlighted face down into the flat part of the back

this is in f360. i can do basic squarish things ok but this is by far the fanciest thing ive ever tried to model, and its been stalled for like 2 years ^^;

this is loft (with some fillets, and im not in love with it

[balmy pulsar]

loft needs a gap for best performance. split the object and leave some space between and loft the gap. use the guide tools to make sure is is straight. Also, dont mate round to sharp corners with a loft. Add a small fillet to the top edges of the second part.

[inland galleon]

ill def try the fillet, but im not sure what you mean by the gap

you mean split the obect at the part im trying to merge (vertically) and put a small gap between them?

[balmy pulsar]

nope

split on the face selected in the first image

then make a gap between thr 2 parts

[split raven]

For people using CAD for 3d printing, do you try to use dimensions that are multiples of layer height / extrusion width? Does it matter? I assume that the slicer will round up / down values so if you want more control / accuracy you might as well take those into account when designing?

[vocal flax]

I think there are 2 reasons for this. The design was robust, and repeatable. And back then availability of low quality components were low. I'm sure they were present but I assume the price cut wasn't worth the supply chain complications.

[vocal flax]

This is probably best resolved using surface tools. as far as I understand you want to "extend the surface then stitch with other surfaces to create solid".

[vocal flax]

Layer height/extrusion width yes. Depends but most likely not. Slicer might do some stuff, depending on settings. Arachne setting in prusa slicer may vary line width to match dimension or slicers do some filling movements for small gaps similar to ironing.

[shy kelp]

No? I just use some random precision table online by material, and scale by that value. Or print some dummies to have a real world value

[faint sky]

I don't scale at all with my models. I attempt to get them accurately represented to the thousandth of a mm. I do not take into account what the layer height or any other slicer settings will be. Doing so is pointless unless you want to make a model for every nozzle size and setting. The models are what they are, and the slicer has to compensate for that not the other way around. Are there advantages to doing it the way you mention? Well yes but will take up so much more of your time that's its not worth the effort.

[balmy pulsar]

you will need to scale if you intend to match any pre existing object. the material shrinks unevenly. for pla i often have x and y set to 105%. z tends not to shrink significantly due to the nature of the way it prints.

[faint sky]

I suppose if you're going to make a model of something that won't fit on a print bed like a full sized car then yes scaling will be required. For shrinkage I do those adjustments in the slicer.

[split raven]

Interesting responses, thanks. I didn't think about compensation for shrinkage. That's the kind of thing I would expect the slicer to do

[balmy pulsar]

i mean scaling in the slicer....

you dont tamper with the cad model of course

[split raven]

Do you do scaling in the slicer in addition to adding some clearances in the CAD design?

[inland galleon]

thanks. looks like i have a new tool to learn

[balmy pulsar]

clearances you would do in cad if needed. scaling in the slicer is just to get the output to "match" the cad as close as possible (its never perfect)

[inland galleon]

i only worry about layer size if im making something thin. if its going to be like 1.5mm+ i figure im fine with the error margin

[shy kelp]

Has someone ever tried to print some cage nuts?
Didn't found them in the hardware store, I've got standard nuts which work fine, I don't have time to do some testing so I was hoping to find some already made model
It's easy, but I don't trust too much the plastic wings

[faint sky]

Haven't tried it but seems about the right shape. https://www.thingiverse.com/thing:5248527

[shy kelp]

Not really the case, it's for a rack, I ended up designing some crap without knowing the exact holemount (although it has some tolerance)
Let's see how it goes tomorrow with 4/5 hours of sleep yuk

[leaden slate]

I am printing one of these
https://www.printables.com/model/41018-microphone-voucher-box-mikrofon-gutscheinbox
the head is 2 pieces and im going to print it out of conductive pla so it can be a capacitive touch object. Do I need to do anything special when i put the 2 halves together? The wire will be connected to the bottom half, I am concerned that if I use some super glue that the top half wont work
https://media.printables.com/media/prints/41018/images/406494_58751cba-ec3f-4572-b2c3-b669781c9d98/thumbs/inside/1600x1200/jpg/head.webp

[faint sky]

could redesign it so it's snap fit

or put large metal scissor springs that make contact with the inner shell once put together.

[leaden slate]

that makes sense. my skills are limited to openscad. I added a couple holes to the stl for the clip for the bottom half, maybe I could add more holes and a peg to hold them together, the only downside is it took a long time to render

[leaden slate]

There's also nothing stopping me from putting a screw into it

[faint sky]

coat the inside with aluminum foil then make a big ball of foil to fit in there. it's bound to make contact that way too.

i mean there are about 100 different ways to skin that cat

[ornate imp]

is there a reasonable and sane way to split a mesh for a print in half or am I going to have to keep fighting blender

[vivid helm]

PrusaSlicer has a tool for that.

[ornate imp]

seems if I had use mirror modified on a mesh, applied it, did other work, and then try to bisect it fails to bisect properly on one of the sides

[ornate imp]

can it export back to stl? I normally use cura, not sure how programs etc differ

still kind of an amateur with 3d printing and 3d modeling in general makes me irate

[vivid helm]

Yeah, if you right click you can export.

[ornate imp]

I'll give that a try, thanks. Blender refuses to slice this properly and I can't figure it out

and this is all just so I can make sure this helmet frame I'm creating is the right size, but it's slightly too big for my ender's bed

[vivid helm]

It could be that your model is screwy.

Some models are like that.

[ornate imp]

considering I'm trying to print a wireframe of my original model that's also possible, but it's incorrectly splitting on exactly one half of where I had it mirrored

really annoying

[faint sky]

I don't use Blender, what I'd do in Fusion 360 is make a flat plane to cut the two sections in half. Then delete the plane.

Technically I would extrude the plane negatively so there is no plane but that's the basic principle. No idea if that'll help in Blender.

[ornate imp]

Bisect is a mesh splitting tool that does the process you're explaining, but the actual mesh separation afterward is breaking due to the mirrored mesh state or something, even after applying the modifier so t hat it's just a raw mesh before bisecting and it still does this

Every time I use blender I have to re-learn how to do the most basic things and anything that's not just doing an extrusion or something ultra-simple is outdated info or breaks in weird ways, I've never enjoyed do any 3d modeling at all so I just hate it every time

20 years of repeatedly going back to do any basic 3d modeling and it drives me insane every single time

Generally I believe people are capable of doing anything if they apply themselves enough but 3d modeling is one thing I cannot do adequately, ever.

[faint sky]

If it's mirrored and you're deleting the original side of the mirror then it should change the other half. In theory. Your idea is sound, no idea why that's not working.

[ornate imp]

The weird part is those attached vertices on the center-to-right, are the original non-mirrored portion that I had before it was mirrored

which makes it even weirder

[faint sky]

oh it tedious for sure and there's always a learning curve.

[ornate imp]

yikes

okay, I have no idea how I'm going to adequately test the proportion of this without burning through entire spools

I thought going wireframe would reduce the filament but then I need even more supports, so just one half is 600g

[faint sky]

there isn't a measuring tool of some kind? so you can scale it the correct dimensions?

[ornate imp]

I can 1000% guarantee you I cannot properly measure the size of my head for this model

that's why I want to do a basic 'shape' print to test the proportions

[faint sky]

lol, youtube videos. there's a video out there somewhere.

looks like bisect mode is what you want

[ornate imp]

well I split in half in prusa but the filament cost is too much for me right now

I'll figure out something eventually, just a lot of things for my project are oriented around framing this helmet correctly

[faint sky]

https://blender.stackexchange.com/questions/5320/cutting-a-mesh-in-half

[ornate imp]

yes, I literally looked at that page specifically

and my result is breaking

[faint sky]

😦

[ornate imp]

[faint sky]

https://www.youtube.com/watch?v=N3H5GQ6wVkM

[ornate imp]

just going to print the original model not this wireframe one, still have to use prusa to slice though since blender is being dumb

haven't done a print longer than 8 hours before so this'll be an experience

thank you for looking though, I think it's just an actual bug in blender or some kind of extra situational knowledge we're not aware of

my very much WIP project is being shaped by this

[worldly comet]

Hi. I like my Ender 3, but now I want auto bed leveling and two colors at a time. Suggestions?

[faint sky]

Sovol SV04 is an IDEX (independent dual extruder) setup. Sovol is a Creality clone that uses mostly Creality components. That would be the cheapest entry level for an IDEX. Then there is of course the Prusa MK4 with MMU3 addon or a quite pricey Bambu with AMS.

There are also some mods available for the Ender 3 to make it an IDEX if you're not wary of tinkering with your setup.

[worldly comet]

Thanks!

[ornate imp]

Not surprised but the first half of my helmet 3d print cracked at layer lines when removing it from the baseplate; I semi-melted those portions back using a heatgun, though kind of overheated and damaged it, but I'm mostly testing size/shape as a very rough print so I'm not upset about that.

I printed with standard PLA (not PLA+), 'low quality' 0.28mm layer thickness, 20% infill and the reason I did that is to speed up the printing time while reducing filament usage - although apparently using a thinner layer line with more infill would've used a lot less supports and thus less filament, but I saved 10 hours off half of this entire model in print time.

As I continue working on iterations of it toward the final goal I'm probably going to use 100% infill, thinner layer lines, and might look into PLA+ or another filament type.

Curious if anyone has some suggestions on how I can ensure it can be kept sturdy in case something bumps into it that it doesn't split along the filament line, including potential post-processing (which I may end up doing on final result with sanding/painting/etc).

[faint sky]

I wouldn't waste money on PLA+. Either use PLA for indoor projects or PETG for outdoor projects. PLA+ is an attempt at a hybrid and falls short of its intended use.

If you're going to print hotter with PLA+ might as well print with PETG anyway.

[ornate imp]

So far I've only used PLA filament, and this probably wouldn't be used outdoors much at all if ever

I'm just concerned about strength of the print's resistence to impacts/falls/etc so it doesn't split open, but again it's probably infill and layer thickness that should solve that

[shy kelp]

Indeed, or another way may make a more or less void shell (use lighting infill), and then infill the inside with some other material, like any cheap construction potty, but it won't be light

You may also want to use some infill combination to save time while not having bad resisted
In the bambuslicer it doesn't seem to he there, but in some others it was... You could replicate in cad anyway

Or maybe make an outside coat of resin

Or then use the 3d print thing, as a base for sturdier materials

[ornate imp]

given it's a helmet for an XR wearable project I'm working on, the lighter I can make it the better, because I've also got electronics, wiring, and batteries to fit in as well.

I'll also have to ensure it is breathable so it'll probably have some vents here and there which will weaken the structure but also make it harder to just 'cover it' it with another material, but this type of design work is new to me

I'm probably going to be dealing with 5-10 iterations of this before it's even close to what I want. this first print was just a size test print for the initial structure so I can get an idea IRL if it fits on my head, what parts I need to give more room for, etc befoer I work on the actual model more

it's 2.7mm thick so it's not all that big, I'm just concerned about it splitting too easily even when I use 100% infill or once I make more structurally weak sections

and I don't want to spend hundreds on filament trying to figure this out if I can since a single print attempt costs me $10+ in material

[shy kelp]

Make a aeries of smaller copies then

Or just one and call it a day
Or use it as a base for other materials like paper and vinavil

[ornate imp]

if I had 500x more experience as a 3d modeler (because I have effectively none) I'd consider 'just one' as a possibility, but that's not possible heh

[shy kelp]

It needs to fit your face? You could easily replicate with a couble of photos
Or maybe 3d scan your face through photogrammetry and use that, but likely you'll have an heavy mesh that you need to retopo (I am bad with it, I can't help much more)

[ornate imp]

Fitting my face is the reason I'm doing this initial print to get a size gauge, and I'm almost done printing

I already tried using a premade high quality model and trying to edit it was extremely futile. ignoring retopology I couldn't seem to edit anything without severely breaking the mesh

my concern is making sure I'm not spending dozens/hundreds of hours trying to make a model that inevitable just won't even work if it's never going to be structurally strong enough to do this project

I'm newer to 3d printing and absolutely new to making wearable structures for a project, but this isn't just a cosplay thing to wear once or twice, I'm making a custom XR headset that is going to be used quite a lot

[shy kelp]

Yet another option is going full "analogic" get a negative of your face through life casting using like silicone (not sure, google that, I haven't done anything since I was in art highschool)

[ornate imp]

Again, that's not my current concern

but I appreciate all feedback regardless

[shy kelp]

I mean you should try what works or you could cold consider to pay someone else to do the job for you
But unless you're getting someone from a lower income country, it won't be cheap either way

[ornate imp]

Yeah I'm not paying someone else to do this, I'll just have to brute force this on my own.

Because it's a low-poly model I can very easily measure offsets needed from this first print off the flat quad sides, and then make adjustments from there in my mesh.

This is the cracks it had, and then it looks funny because I tried using heatgun to re-attach it lol

that's just the back half, and so far I can tell it fits my head well. front half is finishing tonight

once this finishes and I get a feel for the sizing while figuring out offsets needed, places to put my electronics etc, then I can do a more thorough quality print which hopefully won't crack like that

currently takes 500-600g of filament for one iteration to print and I don't expect that to change more than +/- 50g

[faint sky]

Use a soldering iron to remelt pieces together. Theres a knife attachment for most soldering irons or plastic welders that is specifically for that kind of thing.

[ornate imp]

Y'know, when I got my soldering iron + heatgun station it came with about a dozen tips and I've really only used one so far.

[faint sky]

Cracking apart means the shell is too thin. But you also dont want to waste a ton of filament for thicker prototypes.

[ornate imp]

I wondered if the knife-tip was just some kind of weird multi-pin desoldering tool instead

[faint sky]

It is but also works great for plastic welding.

[ornate imp]

The shell is too thin in general (2.7mm)? Or would having thinner layer lines be better? Or both, heh

[faint sky]

Just ensure to clean the tip well after use so you can solder with it later. Wet sponge does well.

[ornate imp]

Yeah I have sponge and brass wool

brass wool for the solder ofc

I wouldn't want melted plastic in that

[faint sky]

Yup i use my soldering iron on both electronics and 3D prints. As long as you clean it well its a dual purpose tool.

[ornate imp]

speaking of soldering iron and 3d prints, I should print a case for my pinecil I just got

[faint sky]

Shell is just too thin if walls are 3mm. I make beefy 5-10mm thick parts. They’re more prone to warping than cracking. You’d need a baseball bat to crack it.

[ornate imp]

Any process for the final design to be uh... cooked? Obviously a bad idea to put it in your home oven, but something to that effect that could cure the layers a bit more

[faint sky]

But i wouldn’t want to do prototypes thicker than 5mm. Thats a lot of filament for a helmet sized project.

[ornate imp]

Or maybe some kind of primer sprayed on

Yeah, I just did a basic modifier to the model to give it volume and printed that to see what would happen. I physically measured the print afterward, heh

[faint sky]

If you more strength for a helmet just brush on some clear epoxy resin.

You can add fiberglass to the outside, even kevlar, and epoxy that.

Of course sanding and painting too.

I built an entire 16 ft boat using that method. If you want a strong helmet that’ll do it.

[ornate imp]

The worst case scenario I can imagine is it falling to the ground in terms of impact

I probably don't need it to survive a baseball bat, but that would be a neat bonus

[balmy pulsar]

new toy. (well, old toy, but being heavily modified)

this is what it used to look like.

basically i am making it smaller.

[faint sky]

looks beautiful and heavy

would need a good motor for that lead screw

[balmy pulsar]

the motor is on the top there. 500w ac servo.

86mm square.

and yeah, its heavy, 500lbs fully built.

haha

[ornate imp]

@faint sky second half cracked when I removed it despite trying to be gentle, but the knife soldering tip trick helped seal it 'enough', ty

[ornate imp]

I ended up welding them together using it too

[worldly comet]

Hi. I’m setting up a new SV04. I turn it on and do a 'home' and the right extruder moves to the right and tries to keep going after it hits the limit. Could it be because the metal piece is bent? I really don’t want to disassemble and pack this thing and send it back.

[faint sky]

Does it have a limit switch on that side? Malfunctioning limit switches are a common issue for a printer not homing correctly. Also possible if there is a limit switch that it wasn't installed 100% correctly. No the metal bend shouldn't be a problem that I can see.

[worldly comet]

I used a wrench to straighten the bent blue metal piece, and that allows the part to slip into the limit switch. So that problem is solved.

[faint sky]

I stand corrected. It didn't look that bad but if it required a slot to fit the limit switch trigger yes that would be an issue.

[misty ibex]

So close to getting my new office setup and getting my 3d printer settings tweaked. I'll be asking for a lot of help here....

[shy kelp]

On the bambu slicer, do I have no way to use a svg or the kind to introduce patterns on the outside shell?

[hollow blade]

hey guys I found something to 3d print at a shop but have a problem

are the dimensions already predefined

https://www.printables.com/model/143745-raspberry-pi-pico-case

[faint sky]

with an stl yes the dimensions are predefined, no scaling should be necessary

[pseudo isle]

does anyone have CNC machine that might want a little side work?

[balmy pulsar]

depends what the work is

I am in xmas gift mode https://www.youtube.com/watch?v=pSjNn7hZf3g

next up, wood magnetic knife blocks

[pseudo isle]

what i need is acrylic cut into a circle

[balmy pulsar]

ah, no. acrylic is better done with a laser.

milling it has a tendency to make cracks

[bleak flare]

@pseudo isle there are quite a few cut to order shops on Etsy that will allow you to upload a shape and then they laser cut it

[toxic plover]

My ender 3 pro has a tendency to not extrude properly, its been happening ever since I added this sharpie mount. I removed it since, but still doesn't. I replaced the tip, tried leveling, and cleaning the hotend but no luck

[vocal flax]

Have you reseated/changed PTFE tube. I've heard PTFE tubes can cause under extrusion.

Cleaning extruder gears is also an easy option. Can be checked off easy if not done in a while

[faint sky]

define not extruding properly?

under extruding, over extruding, inconsistent extruding with gaps?

PTFE tubes on a bowden setup are considered consumables. They will eventually degrade near the end, get crusty, and particles can break off causing clogs and inconsistencies. Partial clogs can be a pain to track down because it usually involves tearing down your hotend assembly up to the extruder gear to find the issue even on a direct drive system.

[vivid helm]

I just had to do that recently

[faint sky]

With direct drives the most common cause for clogs is too close of a z-offset and too fine of a layer height. Backpressure will cause clogs faster than a physical defect.

yeah it's not fun but necessary maintenance occassionally. extruder gears can wear down so have to check those for proper operation and wear too.

calibrating e-steps is part of that. whenever you have to disassemble the print head assembly then you have to recalibrate everything again. just goes with the territory of being a 3d printer owner.

and there's the possibility of your nozzle literally unscrewing itself due to friction on layer lines. that has been a continual issue on my Ender 3 S1 Pro. i've figured out that's a major reason people wake up to their entire print head assembly being covered in a gigantic blob. if it unscrews even a little filament will start oozing out the size of the nozzle where it screws in instead of the from the tip.

if you start getting excess buildup on the sides above the sock good chance the nozzle has unscrewed a little.

and the main culprit for that really is over extrusion and incorrect flow rate or esteps

[toxic plover]

That's the next thing I'll try

[toxic plover]

Very thin lines and occasional blobs

[faint sky]

yup partial clog or too much back pressure

[toxic plover]

Ight I'll see if I can fix it

[hardy inlet]

Could any one enlighten me about why my print turned out like this,

Is it a clog or not dry filament or something else?

Any help would be appreciated

[balmy pulsar]

clog or slipping extruser

[toxic plover]

Yea basically mine if they even get that far

[hardy inlet]

Thanks for the help, I heated it up a little and pulled out a long thin filament and it seems good now

[hardy inlet]

That's unfortunate, I hope you resolve the issue

[balmy pulsar]

clogs are annoying. ive had some filaments like matte PLA that are really prone to clogging

i guess cause it literally has sand in it

[ocean orbit]

yes that looks less than ideal

[coarse coral]

Terrible. Impossible to recrimp

I could solder it at risk of melting the plastic

I feel solder is the best thing

Or do I just snip the block and try to crimp

There was absolutely no reason for me to disassemble this thing just to move house

[coarse coral]

I need a new filament holder this one maxes out at the ceiling (not something you hear every day) and also need to model a holder so that the LCD can rotate 90

Not sure what I feel about this. Yes it's a sort of ready made enclosure but it just seems somewhat restricted. Lowering the shelf is probably the smartest move

[coarse coral]

Yeh this has been cancelled

[balmy pulsar]

O-o

did it make an inapropriate twitter post?

[coarse coral]

Yeh the printer was taking a q and a and got most of the answers factually accurate. The host decided that wasn't good enough, so yes. Cancelled.

[vocal flax]

Why not reverse bowden and tush type of spool holder

[coarse coral]

It's too high either way, I will struggle to remove the build plate at the only orientation it fits. I'll probably come up with a shelf in some other location of which none are apparent to me yet

Was considering up high there but that space is too valuable for media and consoles

[faint sky]

Don't place it so high that you have issue reaching to swap out filament rolls.

Some type of square stand would go well there. Maybe something with cabinets to store spools and tools in. The one thing most people don't account for is where all those boxes of spools will go.

[shy kelp]

Did I f my tip? The back stuff is melted petg

[vocal flax]

Nah if you use brass (or something soft and heat resistant) to clean it off when it's not it should be fine

[balmy pulsar]

petg will always build up on the nozzle. you need to just keep cleaning it off. sometimes several times per print

main thing is to keep it from hitting the print if it builds up a lot

[faint sky]

While it's hot rotate a wet sponge around it and that helps a lot. Then the brass sponge helps take off the rest and will get it back to a nice clean tip you can add solder to.

[balmy pulsar]

most printers come with long tweezer sets. this is what they are for

[faint sky]

Once it cools down you can scrape most of the filament off with a fingernail too. Doesn't all have to be cleaned up while it's hot.

like if you intenionally leave a chunk on there when it cools it'll pop right off.

[balmy pulsar]

i find when it is cold petg is bonded on. not gonne ever come off.

[coarse coral]

Thanks @faint sky Yeh I have 3 wall storage wardrobes from floor to ceiling. I really want one of those to be freed up for all project work.

One currently is storage for golf clubs and decorating things for my place. The other is overflow storage which could be re organised a little bit and the other acts as a closet. I think the second will need to be overflow and closet and free up the last.

The last wall closet should ultimately be the 3d printer at a height that's practical with storage under for filament and all my electronics tools and project thingies.

I considered getting rid of my desktop pc and going to a laptop recently but I doubt thats going to happen. I don't want to feel like I'm not in the pc master race any more

I'll probably just buy some off cuts of wood and make a shelf. Then add storage to that over time.

But then again I just measured and for correct orientation I need 76cm deep and sadly all I have is 60. Closets are not an option now and where the previous photo shows that's not an option either as it leaks into the walking area.

The only other obvious place is table mounted next to a radiator which is bad for many reasons. There's actually no place that I can see 🙈

It's probably going to be going in horizontal on a shelf to wall closet number 2.

When Ur 3d printer is bigger than Ur house

[faint sky]

If your 3D printer is bigger than your house then you just print new additions to the house. Wouldn't that be nice.

[coarse coral]

Clever idea. I'm on it

[shy kelp]

Not going to be legal though
You know you should first pay tons of licensing first

[coarse coral]

Print closet version 1. Codename: Creally Squashy C1

I measured this to a height that was optimal, and also identified places of the walls to drill into. Ultimately I ended up putting off cuts atop offcuts in order to get the right structural integrity. This is engineering (not) at it's finest

[coarse coral]

Definitely going to find a more solid piece of wood for the shelf itself or maybe some aluminium. That's so janky

[sudden lion]

Looks like I wrecked this end stop switch, ordered a new one

While I'm at it I should probably clean up the dust

[faint sky]

Yup that's what it looks like when the silver trigger tab breaks. :/ I got a pack of cheap ones from Amazon but only about 1 in 10 actually worked correctly. Test the switch continuity to make sure it works before going through the trouble of installing it.

[vocal flax]

If it still presses its better to not use the little metal part. It improves repeatability and it doesn't catch random stuff.

Overall more robust

[faint sky]

if you can get it to reliably press in the same spot... that one looks pretty borked.

[sudden lion]

Yes it's not able to press it because it's broken out of place

[reef sapphire]

same happened to me but if you have the metal bracket just put ut back itll work

[sudden lion]

It somehow got bent/moved to where it's no longer laying on it's side, I am not sure the proper way to move it back or if it's fixable. I'm not very mechanically/electronicy inclined. I'm afraid I might break something. I'm just a hobbyist but I know someone who is an expert on this kind of stuff

It looks like it's lifted off the PCB

[faint sky]

The limit switch was directly soldered to a PCB? Usually they're screwed down into a mounting location. That does not look easy to access. What are the chances you can contact the manufacturer for advice or warranty?

[vocal flax]

That looks like MakerBot limit switch clone those aren't screwed in neither has screw holes exposed (PCB covers)

It's just the metal legs bent if you fold it over how it was or to a place where itd press (assuming it doesn't spring back) it could work.

Additionally if you're handy you can just scrap the whole board and make switch holder and manually wire it

[balmy pulsar]

just pay $3 and get some omron limit switches 😛

[hardy inlet]

I was 3d printing today and my first layer had a skip in this section. It looks like the nozzle still went through the entire section but the filament did not come out. By the time I saw it, it was already done with the first layer and the rest of the print went fine.

Do I need a new stepper motor for this or is this a different problem?

Any help is appreciated!

[ocean orbit]

i would keep an eye out for a repeat of the issue

[coarse coral]

Yeh keep an eye, it might be just your nozzle or tube had a bit of a clog. I'd replace nozzle and tube (or at least snip) if it continues to happen. Also I've found that Bowden's are naturally lubricated with the material they are made of and some times it becomes somewhat abrasive leading to chatter in the flow. In the past I've replaced the tube and the chatter goes away

[shy kelp]

No luck to find some kind of spring constant/topology sheet between different materials, or do I necessarily need to waste material into some testing?

[ocean orbit]

a 3d printed objects properties will depend on how its printed so you would probably need to do testing anyway, depends what you are doing though

[balmy pulsar]

yeah, a solid pla or abs spec sheet is kinda meaningless

[shy kelp]

Ehm outside of pcbway, there ain't any cheap metal cnc manufacturer?

[balmy pulsar]

jlcpcb has it too

i think

https://jlc3dp.com/cnc-machining-quote

more limited options. just aluminium and plastic and copper

i think jlc does in in house though and pcbway aggregates verndors

[faint sky]

Send Cut Send comes to mind, if you want sheet metal laser or water jet cut and formed. Depends on your design needs. Actual CNC from a PCB manufacturer doesn't sound like a bad way to go if you want a metal casting mold or something like that.

[balmy pulsar]

cnc is not one button. thats the main issue with the services. you need to be super clear what you want and to what tolerance. every decimal place in tolerance can add 2 zeros to the price

[ocean orbit]

considering hte amount of times i mess up my 3d prints i would be terrified of ordering lol

[balmy pulsar]

i have a friend for big stuff or precision stuff i cant do. i trust him to make sense of what i am asking.

maybe i could save 20% with pcbway but... not really worth my effort

we have a colour code system. blue is "inimportant", yeloow within 0.005", green 0.001", and red is "as good as you can possibly do"

so i just colour all the surfaces in fusion

[faint sky]

I think the point was for people who don't have a friend with a CNC and/or don't want to pay an arm & leg for a small personal project for a one-off. The whole point of 3D printing is being able to design & produce small parts for hobby use.

[balmy pulsar]

well no, the point of 3d printing is to make things that are otherwise difficult or impossible to make any other way and to make functional models quickly. it has adapted for hubby use very nicely because it is becoming more and more of a one button solution. CNC is still complicated and somewhat expensive whether you do it yourself or send it out.

anyhow. cnc service form pcbway and jlc seem quite organised, but it will all hinge on communication.

ive used JLC for metal prints, and they communicate issues very well

[vivid helm]

Safety is another major consideration. Also power.

[hardy inlet]

On the right is the first layer of the print and on the left is the rest of the print, there seems to be some extrusion issues with the infill. Does anyone know the solution to this problem? I just cleaned out my nozzle and Bowden tube. I'm starting to think it is my extruder stepper motors fault.

[shy kelp]

It's also incredibly loud, I can't have that equipment at home even if I had the money
I kinda regret refusing a job into a factory that did manufacturer machineries to print on metal, but I guess that from now on I am banned from working there since I refused the job like a week prior the start date lmao

[shy kelp]

I may be wrong, but to me, it seems that you have some leveling issues or filament that bubbles

[hardy inlet]

Thanks for the advice, I was dumb when cleaning and messed up my Bowden tube so I trimmed it and when reinstalling I didn't have it touching the nozzle which caused the issue and a lot of clogging.

It's printing fine now but I only have half my bed to use since the Bowden tube is shorter now, just waiting on the replacement to come in

[shy kelp]

Any cheap 3d scanner (like 200? Or more, but the less the better), that makes up usable meshes?
Photogrammetry is a pain and expensive
Random things that I find on the net, are like limited in object size, having a turntable table

[balmy pulsar]

200... 000? yes.

:x

i think creality has one for $600. thats one of the cheapest i know of

[ocean orbit]

Im going to go with probably not for anywhere close to that price

Theres the openscan but its certainly more than 200 and very limited in object size, what kind of thing did you want to scan? Someone might have an idea

[verbal sail]

What are you looking to have done? I built a small lathe and may be able to help

[balmy pulsar]

i just sold my lathe :/

poor thing

[ocean orbit]

I would love a lathe and other tools, sadly tiny apartment 😦

[balmy pulsar]

my lathe was destroyed in a flood, but i cleaned it up a little and sold it to the university of guelph race car team. should be interesting ha

[verbal sail]

Build one. Some extrusions, brackets, and a grinder/ hobby saw and you have a basic lathe that turns aluminum well and fits on a desk

[ocean orbit]

oh for sure but i have no space sadly, nevermind though!

[balmy pulsar]

grinder i think is way too fast. but the principle is sound. especially since the "cheap" mini lathes are now almost $2000

[verbal sail]

This was an early version of what I have. I made some alterations since then to make it more stable. It has variable speed as well. Fun part is, I used the grinder to build the lathe

[balmy pulsar]

ooooh, that kind of grinder

thats really cool

[ocean orbit]

That is cool, amazing what you can do with a bit of ingenuity

[verbal sail]

I'm going to be redoing the instructions for the build if yall want it

[ocean orbit]

I couldnt use it but im sure some will be able to

[balmy pulsar]

my sad lathe

it still worked more or less fine. but its quiter rusty

[verbal sail]

I'm gonna be adding a mill section to it as soon as I figure out how and can afford it

[balmy pulsar]

mill is harder, because it needs so much rigidity and precision.

[verbal sail]

Yeah. I'm thinking I may have to add an extrusion cube with cross beams. Take some inspiration from architecture and construction

[balmy pulsar]

i am trying to put this one together economically. it is not really turning out to be worth while for me though. i can buy a better/bigger one for less time and money. :/

"real" used cncs are so cheap now too. if you have any space (i dont).

[verbal sail]

Me neither

[balmy pulsar]

$800 for that

but $2000 to ship to me

haha

sigh

and its still much too big for most people

[ocean orbit]

Is that a crt?

[balmy pulsar]

yeah. its from 1992 ish.

crts were all the rage

😛

[ocean orbit]

Amazing

[balmy pulsar]

this one is also $800. 1999 ish.

[ocean orbit]

When a machine weighs 3000lbs the additional weight of a crt isn't a big deal

[balmy pulsar]

i actually want that one, but the shipping makes it not viable

[ocean orbit]

Cnc Mill vs manual is always a question

[balmy pulsar]

id not bother with manual at this point

[ocean orbit]

See thats kind of amusing because I likely wouldn't bother with cnc

[balmy pulsar]

what i do with these brother machines is rip out the 90s controller and put linux on

[ocean orbit]

I'd get a manual mill if I could but yet again, space

Eventually maybe

[balmy pulsar]

https://youtu.be/sgKMCBDJcTk

i used to have 2 of those. i sold them when i moved. but now i want another one

anyhow... haha

[faint sky]

You converted a benchtop grinder into a PWM lathe? I didn't even know that was possible. That's rad!

[vocal flax]

More like VFD probably

[faint sky]

I've never seen that done before. It's a really neat idea.

[verbal sail]

I dunno about Pam. Its just variable spead through a pot. Simple wiring inside. But yeah. Otgere tgan yhe pwm i did turn a bench grinder into a lathe, with optional table saw, stone cutter, and rotary tool. Technically its for wood, but it can handle soft metals like aluminum

[vivid helm]

Is it a pot or an autotransformer?

[shy kelp]

In this use case I wanted to import something made out of clay in a digital copy. Although it would also be nice to import other objects (usually cad-like) so I can avoid redrawing them

[faint sky]

I think PT actually had a short article on using your phone as a 3D scanner. It what he used to make a 3D mesh of the first Bell Telephone hacker van. I'm sure it's in the Adafruit blog somewhere.

If you can make a 3D mesh of an entire van then small objects should be much easier in comparison? I've never tried it but it's been on my mind.

Ah it's called Photogrammetry. https://blog.adafruit.com/2017/03/02/photogrammetry-3d-scanning-for-beginners-6-must-haves-recommended-by-sketchfab/

and there's an Adafruit Learn Guide https://learn.adafruit.com/use-3d-scanning-to-make-low-poly-3d-models/creating-a-3d-scan

Making them low poly is optional. It starts with a high poly model and low poly is more like an optional artistic choice.

[balmy pulsar]

yeah, thats what he is trying to get away form 🙂 there IS a scanner add on for the iphone, but ive tried it and it's not great. Its a tricky field, and even many of the higher end expensive scanners are pretty terrible.

Scanning is one of my main jobs, so ive gone through most available options

[faint sky]

Ah here it is. PT scanned an entire van. https://www.youtube.com/watch?v=JokGUXH3u9I

ah ok. yeah i would take ihavenofish advice on this subject. they have... professional experience in this field.

[balmy pulsar]

yes. my advice is to avoid the field like the plague hahahah

[faint sky]

lol

[balmy pulsar]

it is mostly geared to construction and engineering, and they really do not understance "why you would want a nice mesh"

[faint sky]

it's likely a rabbit hole of hardware devices that come up short in delivering on their marketing hype?

[balmy pulsar]

actually, a lot of times the hardware is solid, but the software is abysmal

like our leica scanner. the machine is amazing. the software is borderline useless (and costs $7400 a year)

[faint sky]

there have always been 3D digitizers on a small scale. i think what makes the phone option attractive is the ability to scan huge objects.

[balmy pulsar]

well, photogrammetry can use any camera. in some cases you can get godo results. but it is mostly useless for anything that needs to be acurate, or is small

[shy kelp]

(Or something usable)
Photogrammetry makes up heavy blobs

[balmy pulsar]

blobs are software. were you using agisoft?

when i use agisoft i then bring it into cloudcompare which can select polygons by density and delete them

blobs are usually low density patches

[faint sky]

so there's a lot of points in the mesh you end up having to correct manually anyway. have at least a quick mesh is better than starting from scratch?

[balmy pulsar]

i use cloudcompare to mesh lidar usually

[shy kelp]

Tbh, it's a while that I haven't done any photogrammetry (camera shots+random software), I don't really remember
But I needed to retopo or do some subprocess to make it usable

[balmy pulsar]

well, you always need to retopo. that is a given

[ocean orbit]

ive seen some really good results on the miniature scale, like, small objects

[faint sky]

ah the scanning apps require a phone with lidar like newer iphones?

[balmy pulsar]

lidar is "real" scanning. but lidar is not all created equal. the iphone was pretty useless

my leica is lidar. that is acurate to a mm over 50 metres

but thats also $125000

:x

[shy kelp]

Bruh

[faint sky]

i'm still using an iphone 5 because the lidar feature on new phones just creeps me out.

[balmy pulsar]

most lidar fails on anything mildly reflective as well

or black

[ocean orbit]

ive sen some excellent results from the open scan mini, but thats for like, warhammer model sized things

[balmy pulsar]

is openscan using the pi cam?

i saw some good results for micro scale scanning with a pi based one

[ocean orbit]

pi cam or the upgraded one, i think you need the higher res one for it to get decent results

[balmy pulsar]

k

[ocean orbit]

its a few hundred bucks, like 3 or 400 i think all told? but i was surprised

[balmy pulsar]

the other main this we use for tv is a booth of camera. its a mix of photogrametry and structured light patterns

thats mostly for scanning people

[ocean orbit]

is that like the kinect that uses the grid?

[balmy pulsar]

no, this is like 80 canon SLR's

in a dome

[ocean orbit]

sick

[balmy pulsar]

we dont own one of those. they cost too much

[ocean orbit]

https://www.youtube.com/watch?v=tp5p2nDnON8

this is the video i watched

surprisingly good results on a small scale

[balmy pulsar]

yeah.

when you have a fixed sensor and structured light source, you can do some amazing things

[ocean orbit]

ah, structured light IS what the kinect uses

[balmy pulsar]

basic photogrametry "fails" because it has to deal with arbitrary positions and lenses, and often shallow depth of field

[ocean orbit]

i thought it might be, though obviously a simple version

[balmy pulsar]

kinect yes. artec is structured light as well.

(higher end hand held scanner)

[ocean orbit]

anyway, 3d scannign is hard and expensive 😛

[balmy pulsar]

then you have scanners like creaform that use markers

[ocean orbit]

i saw a reddit post that basically put it pretty well i think, if its just a one off or something its probably cheaper just to pay a guy to copy whatever it is manually

[balmy pulsar]

and the last type is the good old arm. faro or romer or nikon. laser on the end, but the arm is calibrated to some microns, so it always knows where the laser is. these are usually the most precise options for engineering

[ocean orbit]

ive seena few requests to 3d scan a part i could build in a couple of hours and im a massive amature

[balmy pulsar]

on set, our main purposes of scanning is to get accurate dimensions.

for tracking and modellsing extensions

[shy kelp]

So in a nutshell there's nothing decent and cheap yet?

[balmy pulsar]

but i also scan props accurately so they can esplose etc

[ocean orbit]

this makes sense, it also makes sense if you are replicating a tonnnn of stuff

[balmy pulsar]

not your kind of cheap

[ocean orbit]

i mean, it depends on your definition of cheap and your definition of decent and the size of the thing you want to scan

but 200 is gonna be ROUGH in any definition

[balmy pulsar]

yeah

[shy kelp]

Lowest amount? Just to have an idea, and see if it may be worth for me

[balmy pulsar]

the creality is $500. thats your entry into something "useful"

but i would stop short of calling it good

[ocean orbit]

how big of an object are you planning to scan?

[balmy pulsar]

my neighbour made a scanner. it was really hated by customers. i forget their name

[ocean orbit]

you can do up to 18cm squared wirth the open scan?

[balmy pulsar]

oh, thats not too bad

i thought it was more tiny

[ocean orbit]

the mini, thats used in the video, is only 9 cubed

idk which is more accurate

[balmy pulsar]

ah

the rovepoint seems to be quite hated.

it looks a lot like the creality

[ocean orbit]

i just think right now 3d scanners at the hobby level are going to be a pain either way

[balmy pulsar]

yes

[ocean orbit]

probably less of a headache to just pay someonme on fiver to do it for you

[balmy pulsar]

hehe

well, as an example, we charge lets say as a random number $2000 to show up and scan a set. We then charge $5000 or more to turn that scan into something useful. The "scanning" is the easy/cheap part. People in the hobby world want 1 button instant gratification, and scanning is not there (yet)

[ocean orbit]

whereas you cna probably pay someone on fiver like, $30 to copy a simple part 😛

So yeah, id maybe look into the open scan https://en.openscan.eu/
if you are only doing small parts but even then you will likely need to do cleanup

[balmy pulsar]

ha. i wouyldnt go that far. but yeah

[ocean orbit]

and its gonna be like 3 or 400

unironically i have considered looking into doing some modelling jobs when im more experienced but it would absolutely violate the fusion360 license so oof

[balmy pulsar]

100k is the fusion lic

[ocean orbit]

i thought they had swapped it to any proft, ill check

[balmy pulsar]

if you get to 100k, you can pay the 450

oh. i think officially you need to move to the registerd start up version

i forget now

they have changed it a few times

[ocean orbit]

they sure have

ok so free version is up to 1000 USD a year

[balmy pulsar]

i sell fusion models of guitar. i made $1250 this year. ha.

which is not bad for not really doing any work

[ocean orbit]

anything more than that and you can do the startup

oh that sounds interesting

[balmy pulsar]

selling models is a completely viable "job" if you pick the right niche

[ocean orbit]

can i peek what you are selling?

i was more thinking of going on a commision thing personally but i am interested

[balmy pulsar]

this is one

they are files for cnc maching guitars

[ocean orbit]

ooooh

yes i get you, an interesting niche!

looks at his neglected guitar

[balmy pulsar]

this will be the next big one to launch.

kid/travel size

but it is quite complex, so i need to make it "ingestible" to people.

[ocean orbit]

yeah, plus making it useable, makign something that L:OOKS like a guitar is i imagine the easy part of the situation

[balmy pulsar]

i want my cad sales to get up to over $1000 a month this year. we'll see

[ocean orbit]

I will cross my fingers for you, my goal this year is to expand my skill

been doing a bunch of modelling for various little projects and i suddenly realised "wow i should have done this at university"

[balmy pulsar]

ha

miniatures do super well. things that are printable.

[ocean orbit]

i am not artistic in the slightest so i would be screwed doing miniatures

[balmy pulsar]

people sell minature cad models for $5-$20 each. model one a week. after a year you have something pretty self sustaining

ahh

yeah im not great at organic

[ocean orbit]

sameee

[balmy pulsar]

better at spadce ships

😛

i am prepping to print this

needs some work

i cant sell that of course (technically i guess i did "sell" it to microsoft hah)

[ocean orbit]

hah even then im kind of rubbish, creatively i struggle, if you want a practical part or bracket or mechanism or something along those lines i can probably do it, but making something creative and aesthetic

not so much

[balmy pulsar]

also need to make a printable 6 foot tall rocinante one day

but they never alocate time for it

[ocean orbit]

thatw ould certainly take a while

I have the basic idea for a clock kit in my head at the moment, mostly 3d printable with an electronics BoM

[balmy pulsar]

well the rocinante model is a complete dumpster fire. it is meant to render, not print of course. it will take an intern like 3 weeks to make it "manifold"

[ocean orbit]

ouch

I need to grab a couple of solenoids and see what i can get away with...

[balmy pulsar]

ok, so fusion "start up" is $150 for 3 years now. so not free, but still pretty acceptable

[ocean orbit]

Thats definitely doable if you are making over 1k a year

[shy kelp]

There's also rhino
1k Perpetual licensing
Or about 200 as a "student"

[balmy pulsar]

rhino is not a similar software

rhino is freeform surfacing (its more than that, but that is the core)

where fusion is parametric solid modelling

there is some overlap, but they have fundametally different purposes

[shy kelp]

Mah I don't see them being so much different
As much rhino is not parametric by default it can be with grasshopper, which now is well integrated

Bruh this year I'm 24

[balmy pulsar]

ive been doing 3d since before you were born. 😛

ha

waitn, no, i mean im 29

yeah thats it

[ocean orbit]

if i was to swap from fusion i would likely move to something free rather than another paid product

[shy kelp]

What's there freecad? Pretty bad imo

[ocean orbit]

though the free options are from what i have heard, variable in their utility

[balmy pulsar]

i need the machining side of things, and there isnt much free that is worth using

[ocean orbit]

yeah

watchign a 5 minute intro to freecad