/diy/ - Do It Yourself

The apple light is pretty rad.

How's it made, just 3d-printed diffusers in front of LEDs? And what material are the diffusers made out of?

>>1272345
Lol the fleshlight made it into the pic

>>1272350
>How's it made, just 3d-printed diffusers in front of LEDs? And what material are the diffusers made out of?

Correct. Natural (clear) PLA placed in front of LEDs that act as diffusers. Hopefully by next week I should have a guide posted on instructibles. I finished writing the guide and programming, now I have to edit some pictures and post.

>>1272360
With diffuser on.

>>1272360
Do you sand it down to a nice matte?

>>1272383
No sanding required. I experimented with sanding and all it really does is make the diffusers look somewhat dirty.

The diffusion seems to be a function of diffuser thickness and distance away from the LED. Everything else is negligible.

>>1272360
Awesome, thanks anon. I got my printer working last week so I'll probably try this as soon as I get some decent-colored filament.

>>1272392
No worries. Note though that those diffusers are not colored filament, but completely clear/opaque. Typically I would recommend 3D Solutech, but their good quality filament is actually a detriment in this situation. There is a shitty Chinese manufacturer called Inland that makes a more opaque natural filament.

>>1272355
Thats a jellyfish

Yay my stopgap x-carriage made it in the pic

Currently trying to fix Z now, has anybody here used integrated leadscrew motors? They're not very cheap, but i've ordered a pair in the hopes of fixing Z-wobble and eliminating the need for a coupler.

>>1272781
Did you consider some ball bearings on the top to try to cut down on wobbling?
Ps, if you try that, skateboard bearings won’t fit, I tried.

>>1272810

I actually printed a pair of long bushings out of PLA that fit the dimensions of the leadscrew exactly, leaving it without any play. While it did decrease some of the wobble, the leadscrew had a slight kink which made it move off-center by ~0.2mm, introducing wobble through the flexible beam coupler at the bottom. This is with the Z-axis supposedly decoupled mind you - the nut was rocking on its X axis in addition to moving around on XY, twisting the x-axis ever so slightly.

Anyway, since the kink is at the very top of the screw, i figured i could use a rigid coupling and just leave the top unconstrained. Big mistake! Never use rigid couplings on a 3d printer - due to the way you use a set screw or clamp it down, the act of tightening it always misaligns it very slightly. Even a 0.5 degree tilt on that end translates to a few millimeters offset at the top end. It shows how Prusa Reseach has definitely done their research, because they use the same type of integrated leadscrew steppers, just with an 8mm pitch instead of the 2mm ones i ordered.

>>1272781
Lead screws won't fix your wobble unless you have wobble in one direction, in which case three leadscrews in a triangle could fix it.

I would try using more or better (tighter fit) linear guides to fix the wobble.

>>1272355

And so did my christmas tree ornament design.

The anon in the other thread was right. I should have sanded it down more :(

>>1272816

The idea is that the leadscrew would be as centered as possible since it's part of the rotor. A more rigid Z-axis is on the todo, as i'm already using 10mm rod there, just need to space the linear bearings further.

>>1272345
Apple sign instructions are up for the anon who was looking into how to make one of these.

https://www.instructables.com/id/Apple-Neon-Sign/

>>1272846
I have a Z-bed balanced on twoleadscrew integrated motors. Like I said, if it's wobbling just in one direction a third leadscrew can help, in which case you should go with a single motor in a closed belt loop driving 3 independent screws. If it's just generally wobble you want better linear guides, it's what they're for. The drive mechanism isn't for stabilizing the path of the hardware, the guides are. How many rods and guides do you have? I'd recommend one in each corner or at the center of each side of the bed and the long type of linear guides.

>>1272936

Ah i see, you must have an Ultimaker-style printer with a "cantilever bed" configuration. In my case it's a Mendelmax 1.5 printer where the X-axis is suppored on both sides by the Z-leadscrews and linear rods. Adding any more leadscrews isn't an option there, and most of the wobble really comes through as a rocking motion in the direction shown in red in my picture. I have a good idea on how to remedy that though.

Since you're using integrated leadscrew motors - how much do they wobble? They're supposed to be perfectly centered from the factory, but sometimes get bent during shipping. There's even this video describing how to replace them if that's the case:

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

What happened to reprap?
They had so much talk about printer being able to make another printer, now they seemed to basically give up.

>>1273070
More or less because the quality of true in the spirit of reprap printers are kinda meh compared to open source + open hardware.

>>1272949
Taking apart motors sounds like a bad idea considering they're magnetized at the factory when assembled. Also, I see what you mean now, for some reason I thought you were talking about the bed.

Are you sure it's Z-wobble? As in each layer is slightly offset from the other? That's typically more of a XY problem, or if it is a Z-axis problem it's probably more likely to do with bad linear guides, bad settings (like bad motor Vref) or could be the X-axis carriage being bad.

I've never had a problem with the Z-axis screws, even ones with cheap couplers like on the wanhao i3.

What could be behind these small gaps on the bottom layer?

-0.8 nozzle with 0.96 extrusion width, 0.4 layer hight
-200 C pla
-100% first layer heigh/width (Could this be it?)
-3 top/bottom solid layers
-35% outline overlap
-35mm/s

>>1273234
Ok im pretty sure its thin wall behavior. How should I adjust the settings?

>>1273234
nozzle too wide, extrusion width too big, design dimensions not optimized for extrusion width, ie all the problems you must accept with a fuckhueg nozzle

>>1273239
Yea well. Might just widen the edges then. The part is already extremely rigid

>>1273150

They talk a lot about that in the comments, and apparently the torque you lose by taking a stepper apart is negligible. One guy even made a video comparing the holding torque of a stepper that has been taken apart to one that hasn't, and they were virtually identical. It especially doesn't matter for Z anyway, because leadscrews can't be backdriven.

I'm pretty sure it's wobble since i've observed how the X-axis twists when doing a fast move. The easiest way to spot this stuff is to see if the wavy pattern matches the screw's thread pitch (it does). It is typically an XY problem, but i mentioned previously that i've decoupled the XY motion, meaning that the nut is free to move around on XY, but not Z.

>>1273070

The idea behind the RepRap project was never that realistic - you obviously can't print stuff like motors, stepper drivers and control electronics to an acceptable quality level. What made it gather momentum initially was its open hardware aspect, with designs that let any DIY-er build their own printer using cheap hardware store parts. As focus shifted towards just buying kits or premade printers from online stores instead of sourcing things yourself, so did the designs. You get to see more non-reprapable parts with tighter tolerances like lasercut sheet material, aluminium extrusions, custom machined hotends etc. Most people care about print quality, not reprapability, so things just naturally went in that direction.

Post some models to print please.

Got my Rostock v2 calibrated and level and sticking. Now Im working on the perfect first layer. Problem is I'm getting this. I've got the extrusion multiplayer at 1.4 and it's just not connecting. Is it speed? Temperature? Any idea anyone?

>>1273306

Calibrate your Esteps/mm. If you're getting that kind of underextrusion at x1.4 it must be off by a ton.

>>1273311
Damn, didn't think of that. Thanks anon.

So what is a "normal" level of temperature change during a print?

I'm getting a bit of temp wobble whilst printing, would PID tuning help or is this to be expected?

>>1273311
>>1273306
Also make sure in your slicer you have your filament set to 1.75mm or the actual measurement of the filament diameter

>>1273335
Yeah I did do that. Calipers came up with 1.72 so that's what I've been using.

>>1273352
Currently running with matter controller and matterslice for my slicer.

>>1273286
boat.stl

>>1273330
That looks okay to me, im getting +-3C usually

>>1273387

Oh... Sometimes I have weak layers in my prints. Doesn't seem to be a clog, and doesn't happen at the same z-height (even using the same gcode).

Thought it might be variable temperature issues. Back to the drawing board on that one then.

Tuning PID can't hurt though

>>1273388
You can try to print test pieces at different set temperatures.

Say you are printing at 200 and your variation in temp is +-4C, you can print one sample at 204 and one at 196 to see if it really weakens or strengthens your part.

Tuning wont hurt yes

I have the Anet A8 and its fun. Then i got a CR-10, which is even better. Now i lust for a RAMPS board with TMC2100 drivers in silentstep mode for that extrasilent printing. Am i ok? Will i survive winter? When will it stop?

>>1273436
>Am i ok?
Nope!
>Will i survive winter?
Let's not get hasty. Work on surviving November.
>When will it stop?
It really doesn't. Maybe you light all your stuff on fire, but really, that's only a temporary setback.

Anyone have experience with building a filament extruder?

>>1273545
Yes. It's difficult.

Hey I recently bought a shitty aldi printer (he170703) and am having problems slicing models, I'm using cura with recommended settings from the manual. But when I print only a tiny amount of filament is extruded and it it too high above the bed and doesn't stick in the slightest. I have tried changing the layer size with no effect. I'm sure it's something I am doing wrong but I have no idea what.

>>1273689

Have you levelled your bed? Thats the first thing to check if its printing too high

>>1273690
Yes I have leveled it about 20 times so far, I played with the height a bit making it higher and lower with no difference, so basically what it makes is a tangle of string each time.

>>1273690
Also the default models that come with the printer work fine, it's just the ones I slice.

>>1273685
Care to elaborate?

>>1273255
>The idea behind the RepRap project was never that realistic

This makes me sad.

>>1272361
Looks very nice. I have some clear pla lying around. How thick is the surface just over the leds? Thanks

>>1273695
>getting plastic pellets is difficult, no one wants to sell you less than 500 kilos and the few that do will overcharge so much it's not even worth making an extruder
>even if you get pellets somehow, you won't really know the specific characteristics and it may or may not behave differently based on additives
>for the machine itself you need a strong motor with a big reduction, 12V wiper motors are a meme, won't extrude for shit
>matching the auger and pipe ID is difficult, depends on exactly what plastic you want to extrude and the viscosity of it, more viscous plastic needs a gap between the pipe and the auger, while something like PET needs a very close match
>sourcing the right pipe is difficult
>the pipe will bend on you because of the internal tension when you cut the opening for the pellets and ruin the pipe
And when you eventually get to extruding plastic you still get blobs and bubbles in it because you didn't use an industrial drier prior.

>>1273702
The diffuser surface? Typically between .2-.26 thick. Distance from the LEDs should be about .5-.7 inches.

Fuck you all mptherfucking nigger faggots. You were no help at all. I was just a curious youngling wanting to get into 3d printers. Asked advice on good cheap printers, etc etc. Nothing. Faggot cunts. Well o built a machine. Got it running. Levelled it. Etc. Etc. Printer out this test box and it all worked great first try. I even did a burn test and strength test which is why you see it beat up and broke in the back. I can do everything wothout your help.

Fuck you again useless niggers

>>1273733

I want to make the cheapest excuse for a 3D printer possible, for the memes. I figure $5 hot glue gun from china, some tiny DC electric motors, and some markings printed on a sheet of acetate + a few LEDs and photodiodes to make a bootleg optical position indicator, and a frame made from pine and drawer slides. But how would I send it 3D printer files (G-code?)? Could I just send them to a microcontroller, or would I have to use a raspi or some specialised hardware?

>>1273739

You could do the "old CDROM parts" memeprinter with one of those 3d printing pens as the extruder:

https://www.youtube.com/watch?v=anIy6eb1fW0

>>1273747
I thought of using those 3D pens, but they're kinda expensive for a meme project. I figured with a hot-glue gun with a modified thin nozzle and possibly hotter temperature could work, and I might even be able to use PLA or something with it. The CD drive parts is a good idea, though they would limit it to a fairly small size.

Also pallet wood + hot glue gun will be really good for a youtube-tier meme project.

>>1273756
Hot glue guns are not cheaper than a regular 3D printer nozzle, both are like $3-4

>>1273747

pretty cool video.

Got me a Titan extruder. I'm still waiting on the aero heatsink for it. The thing is so flimsy, does it actually work?

Also, is there any store in Aliexpress I can buy cheap brass gears to make a geared extruder for bowden (i.e: kinda like some ultimaker 2 extruders)?

>>1273739
>and a frame made from pine and drawer slides

That's actually how one of my customers built his first 3d printer, and it worked well enough for him to print the parts for his second one, for which he built a frame out of aluminum extrusions

>>1272781

You eliminate z-wobble by using 10 or 12mm linear rods (8mm are flexible enough to be bent by the leadscrew/whatever) and a proper straight leadscrew (rather than a threaded rod)

>>1273764
>The thing is so flimsy, does it actually work?
that's what all your mums say about my dick but it still delivers, even without heatsink

>>1273761
All I can find are these nozzles, and longer, threaded things, neither of which appear to have a heating element within. I guess I could wrap something like that in nichrome, which could be pretty cheap. If there's something even more cobbled-together looking that I could heat as a nozzle that could be good, all it really needs to be is a metal cone. Maybe I could bend some thin computer-case sheet metal into place and wrap it with nichrome? All it would take is some plumber's tape to insulate it.

>>1273776
https://www.aliexpress.com/item/3D-printer-V6-remote-extrusion-print-head-extruder-with-Thermistors-Cartridge-Heater-J-HEAD-hotend-long/32810836239.html

>>1273777
Ah, thanks for that, I'll definitely pick one of those over a hot glue gun. Looks like a 3D pen would actually be pretty cheap to make if you just got one of those and a housing.

Still wondering if I do make a shitty diy one, what material to use. I've seen teflon parts, does PLA tend to stick to metal?

>>1273766

I'm already using 10mm rods (for all axes actually). The leadscrew is a proper 10mm trapezoidal one too, there's still wobble. Not extreme wobble mid you, the banding is barely visible unless you inspect the part up close.

>>1273764

I got a Titan Aero now and it never struck me as flimsy. The hob on your picture looks really weird, and the plastic is way thinner than it is on mine... Is ithis some sort of bootleg?

>>1273794
It's a Tevo clone, not the original

>>1273794
>I'm already using 10mm rods (for all axes actually). The leadscrew is a proper 10mm trapezoidal one too, there's still wobble. Not extreme wobble mid you, the banding is barely visible unless you inspect the part up close.

How are you attaching your z axis assembly to the rest of the frame?

>>1273798

For Z axis assembly and wobble direction see >>1272949

Being a Tevo clone would explain the flimsy feel - original is pretty solid, and so is the TriangleLab clone i believe. It also explains the weird hob - main issue with clones is that they don't machine the tooth profile as precisely as E3D's because it increases the overall manufacturing price. Tevo must've decided to go for the sort of gear profile you see in direct drive extruders instead to remedy the situation somewhat while keeping price down.

>>1273818
Show your prints. Just because you can move the axes a certain way by hand doesn't mean they're moving the same way during prints. A cube will shoe z wobble and tell us whats wrong

>>1273818

Triangle labs one was a bit more expensive. I ordered the heatsink from them, I wonder if it will even fit the Tevo clone. Guess I'll have to order the TL extruder as well, minus the motor. I can keep the Tevo as a bowden extruder, anyway.


If you don't want your gantry to twist like that, make an X carriage that is as balanced as possible. If you have your motor all the way back then the carriage will twist the beams as it gets closer to the middle since the anchor points aren't infinitely rigid. That happens on my machine since I use aluminum rods and they flex easily. A hiwin-like linear rail should mostly eliminate this "issue"

>>1273715
I have access to pellets, grinder, drier, and an industrial extruder, but i cant change the tooling so i can only extrude 2 mm strings with shit tolerances. This is why i want a smaller more accurate extruder.

I have a drawn pipe that would work, and a gearbox to get enough torque.

Thanks for the heads up tho.

>>1273691
Bed too high plus not enough stick. Try blue tape.

>>1273691
Did you try babystepping the z axis when the print is starting? I can't reliably set my z-endstop so I always have to raise the bed 0.28mm while the printer is making the skirt so I'll get a nice first layer

>>1272360
>>1272361

Looks really great. I wanna try that.

>>1272361
fucking nice

>>1273733
I told you and countless others to buy the monoprice one.

>>1273950
>>1273960

Unfortunately the arcade sign was a little too thin. The reds and some green shows through. I shelved that project for now, but if you keep a diffuser thickness of at least .22 inches you can't go wrong.

So I've been printing for a few months with my Flashforge Creator Pro and I've just been using the ReplicatorG splicer that came with it, but I'm not starting to notice some of the bugs and issues with it. What's a good free/cheap Splicer to use?


Printed and painted this for a friend who's a fan of Madoka. May as well contribute with it.

>>1273689
I have the same printer. Cocoon create yet?
Go to officeworks/bunnings whereever and get some very thing double sided tape. Works like a dream. I couldn't get anything to stick without it, it's not your bed leveling.

>>1274075
I think I might be completely fucked in the head this morning.

Get some very THIN* double sided tape. Don't mind all the other mistakes I have been up for a long time. Just buy the tape and have zero problems.

>>1274012
Cura

>>1274012
Cura is great.

Just made this useless lampshade in fusion 360 and sliced it with cura.

17h and i got a brand new lampshade i dont really need.

>>1273234
Bump the heat up to 220 and it should come out nicer. Also you might want to lvl the bed a tiny bit more.

perhaps lower get the nozzle a tiny bit closer to the bed aswell.

>>1274211

does that mean it took 17 hours to print that? I'm thinking about getting a 3d printer. Is that normal for most of the affordable printers, or was that abnormally slow for some reason?

>>1274214
I got a decent 3d printer. Prusa mk2.

Its a rather large build so it takes a while. It will be done a little bit sooner. The time estimate is off by a few houres.

But yeah 3d printing takes a while. Unless you got a lazer printer.

>>1274216
>Unless you got a lazer printer.
What is this meme

>>1274217
>dont tell >>1274214, wait untill he asks where to put the filament

>>1273908
Take a look at the precious plastic extruder, I was following that design with mine.

What stuff do you guys use to make a smoother finish and fill in gaps of your 3d printed stuff?

>>1272345
Why I would want to learn about 3D printing?

I could make money knowing this?

Serious question

>>1274240
For fun, and no you can't really make money off of it. Maybe cents per print unless you make original content, in which case you're better off with a real job anyway

>>1274012
sli3r

>>1274264
>no you can't really make money off of it
Really?
Buying a 3D printer is useless for business purposes then?

What a waste of time, and... why you learn about it then?

>>1274226

For ABS - acetone smoothing
For PLA - ethyl acetate smoothing (which isnt as effective as above for me)

For any wargaming stuff I make, I just fix it up with greenstuff if needed. Greenstuff is great.

>>1274277

Depends what business youre in.

Do you work in manufacturing and need quick prototypes? Then they are worth the investment.

Do you work in fast food? Pointless to buy.

And why learn anything if you cant make money from it? I suggest you look up "hobbies"

>>1274286
I see. what about filler primer or wood filler? I hear that resin epoxy is also good, any thoughts on these methods?

>>1274288

I've also _heard_ good things about various resins, but not used them myself so didn't recommend.

I also spoke to a dude that uses a 3D printing pen to fix things mid-print if it goes wrong. Tenpted to get one myself to see if that can help

>>1272345
Is there a cheap chink kit recommendation list somewhere? I used an makerbot at my former employer so I have experience with 3D printing. And now I want a cheap starter kit to do some prototyping at home.

I have looked at these 2, but chink 3D printers are a jungle.

https://www.aliexpress.com/item/2016-Newest-TEVO-Tarantula-I3Aluminium-Extrusion-3D-Printer-kit-printer-3d-printing-2-Rolls-Filament-8GB/32596996503.html

https://www.aliexpress.com/item/2017-Newest-Large-Printing-Area-260-260-350mm-Open-Build-Aluminium-Frame-3D-Printer-kit-printer/32797885083.html

>>1274012
I just use makerbot software with my creator pro. it works perfectly well, but i have been thinking about trying out some other slicers myself

>>1274322
Look for a prusa mk2 i3 clone. Some good videos on them and about the same price range.

>>1274277
I can design my own stuff, for personal use only though.

Btw, has anyone ever used an aquarium pump to cool prints? I'm trying to figure out an effective way to cool prints on the e3d chimera. Fan shrouds are usually too bulky for both ends

>>1274335
I have thought the same, i dont think its enough.

So what speeds does everyone here print at?

Anyone got a resource of warhammer 40k .stl files? There's a bunch on thingiverse but not all that I know of

>>1274287
>"hobbies"
you must be a milliionare

>>1274371

>chinkshit i3 clone printer:- $150
>PLA filament:- $20 a kilo
>Second hand laptop:- $40
>Linux, slic3r, FreeCAD & Blender:- $0

Whatever m80, this hobby is cheaper than vidya

>>1274240
You could sell models for tabletop RPG, though perhaps having an easy way to make the models would help since it's doubtful that many of the players who want models made will be proficient in blender or solidworks or whatever. There might be 3D character generation software, at least for 5e.

>>1274340
Why not?

Somewhat unrelates. Are there any plans for diy ink printers? I ask because i want something that can print in white without costing $20,000

Little update on the TPE jellyfish. One side I've printed on 50mm/s for the most part and ~75mm/s for finishing it off because I had to leave, and that turned out fine in my black filament. Throwing a flat 75mm/s on white filament for another side resulted in these lines, which made it really weak, much weaker than it's opposing side. Rather frustrating. I swapped because I was running low on filament, but I guess I'll find out about it soon when I print it again in black. Attempting 70mm/s on black and seeing if it does the same pattern again.

Beginning to have bad thoughts on my clamping design. Looks like I'll be slapping teflon tape at its seams.

If it all fucks up again I'll suck it up, get more filament and eat the 16 hour print times on 50mm/s.

Frustrated, annoyed. Aiming to get results by Wednesday.

>>1274428
Not possible or feasible in any way. Even those $20,000 printers can't print white without somewhat of a hassle. The nozzle and pump system for the white channel needs to be used frequently and/or flushed periodically. Major pain in the ass and expensive. 220cc of Roland white costs $320.

To get to 300dpi, the printer needs to be hell of a lot more accurate than what 3D printers can do. The ink dispensary system, drying stage, feeding mechanism, and encoding system all need to be more finely controlled than what's DIYable. It's miraculous that simple ink printers cost less than $50 nowadays.

>>1274491
Ah fuck. 70mm/s resulted in wall delamination in a small test run, and striations on layers. Quite bad! Have revised design for better proportions.

Looks like I'll be sticking to 50mm/s.

>>1274538
>It's miraculous that simple ink printers cost less than $50 nowadays.

The trick is they use linear encoders rather than shitty stepper motors

>>1274555
Couldn't you use rotary encoders instead? They'd take up less room.

>>1274546
>70mm/s resulted in wall delamination in a small test run, and striations on layers.
might help to bump up the extruder temp a bit. but you would probably be better off just taking your time and printing slower

>>1274571
The linear encoder is actually tracking the position of the print head, the linear track runs across the printer and the encoder is on the head. This is better than a rotary encoder because the rotation will need to transfer into linear movement which can slip and will have slop in it.

>>1274555
Good point, it's always confused me as to why nobody uses inkjets to build 3D printers. Are there limitations that stop this from being possible? The only one I've seen is a delta made from dot matrix printers.

>>1274582
You need to heat the plastic still as it's falling from the head but you cant blow hot air because that will fuck up it's path.

Maybe shine a laser and heat up where it's supposed to land?

Particle size and propellant is a problem too. It's really interesting, I'd love to do it, but I don't have the spare cash to invest in a printhead like that.

It seemed obvious to me, so I kind of assume there must be something stopping people from using it.

>>1274585
Same here, I'd totally do it if some parts at the thrift store caught my eye. I imagine it being a regular fdm design, maybe stealing from ultimaker and copying those xy rails. I could see the z axis as being a problem, but other than that it seems like a great way to build a cheap printer.

Gregg rulez ok

>>1274538
The ALPS printers could do it for under $2000, not sure why they went out of production. They're fine printers based on the decals I got

>>1274582
https://m.youtube.com/watch?v=t650YXt-_jQ
It's like you people don't know how to google. Nobody does it because there's no real advantages to it.

>>1274580
Ah, I guess it's fairly easy since they've got room along the flat rails. For a non-position recording linear encoder, I think you only need two binary optical channels, one offset from the other by the spacial period /4. I was thinking I'd have to use 8 channels for my shitty timber printer, but with limit switches and a manual zeroing controller I don't think I'd need that at all.

>>1274571

You actually can, some printers use servos rather than regular steppers (i.e: Zmorph), but you can't even tell the difference from print quality alone

>>1274641
Polyjet is a kind of inkjet printer. I can't wait until that shit comes to DIY 3d printing, as I don't think SLS will ever take off and DLP is pretty messy


https://www.youtube.com/watch?v=Som3CddHfZE

>>1274641
>>1274930
I understood he meant like this. Not using it as just the sensor, but literally using a head built around ink jet technology to lay down your plastic.

>>1274953

How about combining the 2 for a multicolour printer?

For example, using a clear/white filament, that passes through a small inkjet spraying chamber with the usual CMYK inks to colour the filament before it goes through to the hotend?

I've not really thought this idea out, just whilst writing this post, so I'm sure it's massively flawed in fundamental ways

/diy/, i'm thinking of printing a H0 wagon as a flat assembly kit instead of single part.
I think it might increase print speed(can be printed multiple at once), quality (less layers visible), stability and cost. What do you think about this, considering i haven't even decided on a printer yet?

>>1275010
Ink wont color plastic like that. Hell even most dyes have trouble. That said, HP colorjet or whatever works similarly to that, but iirc it's polyjet and liquid resin is easier to dye than only partially molten plastic

>>1275021
Print both out and tell us how it really was.

I am trying to find 3d models or dimensional schematics that are to scale to a S&W Model 642 revolver or of similar size. Looked around and it is difficult to find good models to print from. I'm printing it for a buddy who does leather working for holsters and needs a cheap form to mold the leather around. Anyone have experience to share about going about this?

>>1275021
Use miters instead of butt joints.

>>1274924
Is that a motor with a built in encoder? Pretty smart.

>>1275231
Yes, which makes it a servo

There's a kickstarted servo called uStepper which can be used as a drop-in replacement for 3D Printer motors. Supposedly it eliminates missed steps.

https://www.youtube.com/watch?v=GzB2n2bIjzY

>>1275246
>that's what a servo is
ooooohhh

>>1272345
Is just mine or repetierhost is so buggy

>on rotate if I press backspace removes the object
>if I press winkey repetierhost crashes.

Damn it, ran out at the very end. Looks like I'll have to get more filament. Actually unusable since the taller cut off section is part of the inside of the cast, can't risk it being all shot to shit.

Otherwise, >>1274573 the walls turned out a lot better with 200C on PLA, much more consistent of a finish.

Live and learn.

>>1275246

why do morons think it helps to make a video of them jumping around from one menu to another? WRITE A FUCKING ARTICLE.

>>1274491
>>1275461
Can't wait for you to share your designs, senpai. Keep up the good work!

>>1275473

But you can't monetize an article with the excellent YouTube advertising platform from the benevolent leader Google!

>>1272345
So uh, what is this supposed to be?

>>1275638
clearly, it's an exercise in producing an inverted archimedes screw

a mild success, i must say

>>1275638
It's a sex toy.
Don't tell admins desu~

>>1275638
Is the thing above it and in the middle an anal probe of some sort?

>>1273306
I have the same machine, so I might be able to help if you answer some questions. What material? Temp? First layer speed and height?
1.4 is way too high of an extrusion multiplier if you're using the stock extrude. Also turn off your damn flash or take the picture at a sharper angle.

>>1275638
>>1275659

It's a jelly fish

>>1275661

that's clearly a minimalist christmas tree statue


you guys need to get your minds out of the gutter

>>1275680
Thats my boy!

>>1275638
It's a jellyfish obviously.

>>1275661
Unicorn horn.

>>1274609
looks under extruded as fuck

Try fiddling with the filament size setting in your slicer. Either use calipers to get the actual filament size, or experiment by decreasing the size setting till all the lines are filled in.

What's the best way to make a print air tight? I want something capable of holding air pressure for a while.

>>1275739
Acetone vapor for ABS
Epoxy coating for PLA
If you need it quick, use superglue coating, dry it quick in water vapor.

I used the later successfully for vacuum piping. Only coated the outside because i didnt have space, but inside coating would be the best i guess.

>>1275638
it's a jellyfish. this was explained quite clearly by the anon who posted it

Good lord, is this a beautiful machine or what?

https://www.youtube.com/watch?v=fZTYfo1o7Lg

>>1275849
Nice design, but I can see the chinese knock off linear guides in the price.

>>1275849
God damn how many more cetus shills are we going to have?

>>1275849
Are there seriously 4 screw holes in the build plate that is also quite bumpy?

>>1275993
They're vietnamese

>>1273733
except spell.

/thread

>Prints coming out crappy no matter what I do (inconsistent extrusion)
>Suddenly think to check odometer (reset when I do a nozzle change)
>4km left (exclusively PLA), 2km right (mostly PETG, some luminous PLA)

The black shit from the PLA wasn't surprising but the PETG nozzle, for some reason, has lost all of its internal nickel coating right up the inside (even where it was touching the PTFE) and the tip of the PTFE tube has somehow become crimped.

>>1272345
I...is that 3d printed fleshlight I see?

> watch videos printing all these great things with ease
> read this thread and see all this technical jargon
> afraid I'll waste a few hundred because I'd never figure it out

>>1276148
Literally after setting up your first print, you'll understand 95% of 3D printing "jargon".

>>1276128

That's a *jellyfish*, can't you read?

>>1276084
replace the teflon tubes, and use a drill bit and needle to clean and unclog the insides of the nozzle

>>1276190
Thanks, I just replaced the tubes and the nozzles since they're coated.

Real bad clog today fellas, and my nozzle and throat were seized in my heat block...

Took ages to loosen and replace, and I dont know how it happened as I've printed like 5km of filament with no issues, and changed no settings.

Life is pain. Always keep spare parts on hand kids.

>>1276275
Sometimes, there's just crud in the filament that's introduced during manufacturing. It could be anything from a clump of dye to an unfortunate insect that got extruded.

>>1276152
Thanks anon. I have a few projects to make. Hopefully I can get my rocket finished and post it here.

>>1276148
>>1276380
I highly recommend the Marvin keychain for keying in your settings. It should take less than a metre of filament and lets you test everything from post-extrusion cooling (the top of his 'face') to inter-layer adhesion (how strongly the loop of the keychain stays on).

>>1276391
Much appreciated m8. You've been very helpful.

So I printed this weird hotend mount/z-probe combo

The idea is that the hotend sits on a hinged platform, and there are electrical contacts on the other side. When the bed pushes against the hotend, the contacts open and the z-min is triggered.

It works, but it turns out my X-axis rods are so flexible they bend even before the hinge moves. The endstop will be triggered, but I'll have to lower the bed a few tenths of milimeter until the x-rod isn't flexing anymore.

Is there anywhere in marlin where I can add those tenths of milimeter as some kind of offset for the z-probe/z-min?

I'm awful at fusion 360 and need to make an octagonal shape for a print. I want the four holes shown in my model to have their four corners cut off to make them resemble the buttons on the controller on the left. I've figure out how to make an octagon with equally shaped sides, but I'm not sure how to draw the real type of shape I need. Thanks if you can help me.

>>1276513
Chamfer.

First timer here. Do all my models require wall thickness or are they okay as long as I have them water tight?

>>1276541
Okay as long as they're water tight

Could I design print out a custom attachment for my shoes that has holes underneath for wheels + some system of locking the wheels in place retracted/extended (say a sliding mechanism on the sides of the attachment?
It would simply be held in place with straps or something over the toe of my shoe

sortof like the amazing rollerskates rikka has but without having to scour for shoes that fit and that are durable

wheels are pretty cheap to buy separately but I have no idea if 3D-printed stuff is strong enough for stuff like this

>>1276721

Well, it was possible with injection moulding when I was a kid

>>1276723
I suppose using a mould would work too
but I have no mould so I would have to print one I suppose?

>>1276721
>>1276723

...or something similar atleast.

Just combine the 2 examples. It would be like wearing platforms though, so you _will_ be accused of being a manlet when the wheels are retracted

>>1276725
the issue is finding shoes that fit though since these are usually sold from online stores and feet are tricky because two shoes of the same size might not necessarily fit the same

that flip-mechanism is pretty clever though rather than a raise+lower one

>>1276721
>wheels are pretty cheap to buy separately but I have no idea if 3D-printed stuff is strong enough for stuff like this

It can be made strong enough, by either using materials like taulman nylon or making it thick enough

>>1276084
I think luminous filament is very abrasive if you don't have a hardened nozzle.

>>1276527
Exactly what I needed, thank you

It just occurred to me that it should be possible to 3d print injection molds IF you use an extremely low melt temp thermoplastic like Instamorph. (InstaMorph-66°C PLA-180°C) In this case you'd 3d print the hollow mold in PLA of what you want then melt and pour/inject instamorph into the hollow mold. I havn't tested this, and I think there may be problems such as sticking or warping if the temp is too hot.

WTF should I do with this new knowledge?
Mass produce dildos?

>>1276835
dildos or get some color dyes, maybe just use toner idk, use your ideas to make cheap -ish chess pieces, sell on etsy, make some cash while you're fucking about, maybe it'll pay for your filament after awhile

>>1276835
You can 3d print injection molds if you use some high temperature abs-like resin

>>1276841
unfortunately instamorph costs more than filament per KG. The one benefit is it's 100% reusable. It can be melted and re molded an infinite amount of times. I guess it'd be good for parts that frequently break through mechanical stress.

>>1276844
If you search for Polycaprolactone you might get it cheaper without a retarded brand name.

>>1276183
Delphin snout.

>>1276835
Not sure if this Instamorph has any use besides making toys or repair things quick. Your idea might be feasible but the material what you can mold makes it irrelevant.
More problems with the tooling:
Surface finish because of the fdm process.
Wear due to the high speed mould.
Accuracy due to the warping of the tooling under high pressure.
Delamination of weak layers.

>>1276843
I personally know the guy who developed a method for producing tooling with polymer jetting technology to get parts with decent accuracy. You can produce 100-1000 part with a single mould, pretty useful!

Can anyone suggest me a good program that i can use to collapse triangles or merge vertices in an stl file?

I am too lazy to write my own.

>>1275217
try grabcad dot com

>>1275021
have a look for Depronised on thingiverse

>>1277088
Blender -> Decimate mesh

>>1277132
Can it do only 3 selected triangles?

What does /3dpg/ think of garbage designs like this? Do you think people design them because people in biological siences don't know better and have access to funding?

https://www.youtube.com/watch?v=jbOlXAAzB4g

>>1277148
yes

the only design is shit and the idea of fdm for stem cell + scaffold system has major problems.
https://vimeo.com/155880305

pumps at that level of accuracy are either this http://www.mycorsolutions.com/uploads/4/3/2/6/43268381/pneuwave_brochure.pdf
or syringe pumps, used in the video.

if they focused primarily on just a delivery mechanism of the stem cells using tech like that constant pump or a peristaltic pump system, they could make some patent money.

>>1277183
>>1277148
I dont think they give a shit. If it works dont fix it.

I'd love to play around with different length intake trumpets. Also to make other components as projects progress.

>>1273436
>Anet A8
I'm just starting out and wanted to know your experience with the A8.


Also, meseeks any good YouTube vids/channels. Thanks

>>1277183

It's like they paid an artist to come up with a model and neither the artist nor the guy who come up with the idea have any clue how anything works

What is that big ass nema23 doing?

>>1277183

We could come up with a fancy design with a pair of syringes (call it bio-nano-colloid-dispenser) and earn hundreds of thousands of free kickstarter bucks

Should I go with the regular cr-10, or the cr10 mini? the mini is a good 50 bucks cheaper for me(canada), and the smaller size and improved quality would be a plus.
I just don't know how many parts they changed between it and the base model, so I don't know if I'll be able to replace shit when it breaks. like, if the hot-end is different that might not be a simple drop-in-fix, or if the motherboard is different I might not be able to swap that at all if it fries.

Any insights?

>>1277221
Wait for sale or coupons for a CR-10. Cant beat that huge build volume.

>>1277240
they're already both on pretty heavy sales. The mini is about 90 bucks off on banggood, and the fullsize is 60 bucks off on gearbest, which I want to avoid using if possible.
I doubt the prices are going to drop that much. The build volume is neat, but the mini really isn't THAT much smaller. I'll mostly be building small parts for hobby electronics. The biggest shit i'll need is enclosures, where z-depth will never matter, and an extra 8cm really isn't worth the extra 50 bucks and having to deal with fucking shitbest.

>>1277247
I think you just answered your question. Beware of the glass bed come with the cr-10 line since mine bows right out of the box. Otherwise its not a bad chinkshit printer

/bant/ brought me here, is there any way to make kapton tape usable?
I have a pei sheet on (actually level) glass and it seems to work similarly, but not as good with pla, it keeps sliding off kapton.
I have glue, but I want something that can work on a permanent bed for a few prints without touch up.

blue tape is my default,but sticks too well with certain brands of plastic and leaves a bad (but tolerable) finish.

>>1276450
marlin has a probe/ probe offset setting you should turn on, you can change offset in the lcd or through the commands, just make it positive rather than the normal negative.

>>1277267
photo related

>>1275662
Material is pla, the stuff seemed sells, temp has been all over the place but 205 seems to be the best so far. First layer height is .2 and I'd have to check the speed.
I am still running the stock extruder but the hotend is the upgraded he280. Sorry about the shitty pic, my office is pretty dark.

What CAD programs do you guys use?

Buddy gave me this pile yesterday. Anyone recognize the brand? And do you think it's salvageable?

>>1277359
Honestly i just use AutoCAD. For 3d I use inventor, creo and Autodesk fusion.

>>1277359

FreeCAD. I keep dabbling with blender for more "organic" modelling, but it's a clusterfuck.

Even compared to FreeCAD!

>>1277370
>>1277361
I would be doing mechanical stuff rather than say characters. Thank you guys.

>>1277360
Get some 2020 t-slot alu, build a new frame, reuse the PSU and the steppers, rest is probably shit

>>1277360
>brand
That looks like a frankenprinter, but if the components are still functioning theres plenty to salvage there

>>1277383
This

>>1277383
Yeah, sounds about right. Probably going to need an actual extruder as well.
>>1277386
He said it was some chink kit he picked up. The whole things made of laser cut press board. Everything works though so here's a chance to build my own small set up.

>>1274328
Not that anon but are those the defacto standard for starting out?

>>1277546
I wouldn't say there's any standard exactly, but Prusa style cartesian printers are definitely pretty easy to debug, and if you get a kit, it's likely to have a community built up around it that will be able to help you with that kit's quirks.

>>1272345
Is that an onahole in the OP?

>>1277555
We've been over this. It's a flexible fitting for a water pump and definitely NOT some sort of masturbation device.

>>1274353
>9000

>>1274397
>5e
Not that anon,
I need to make my cleric. What software did you have in mind?

>>1275021
Wow, how did I not think of this? I have n scale and this could do a ton of things for a model rr.
>>1277120
Thanks

>>1275739
Depends on what you're making. A groove for a rubber o-ring could be one solution.

>>1277220
Don't forget to compensate yourself well before you "fail" due to "technical challenges"
>>1277200
>Also, meseeks any good YouTube vids/channels. Thanks

Bumping for this.

Recommendations for a measuring caliper?
I bought a cheap one from Amazon and I hate it. The thumb thing that moves the jaws doesn't roll smoothly.
Also, should I go digital?

>>1273739
Just pull apart old printers. Take the steppers and/or DC Motors plus Encoders from them.

Buy a cheap extruder for a few quid.

Then run it all on a shitty old computer with a parallel port using Linux CNC, or buy a chinese mega clone and ramps board.

>>1277673
How accurate do you need to go? This is the one I have:
https://www.amazon.com/dp/B017KUC6XQ/ref=cm_sw_r_cp_apa_7YHcAbSEC2QX0

Not that I've compared it to a gague block or anything, but it's accurate enough for my hobby porpoises.

>>1277679
I got 3 h bridges in corporate samples, look into those, you could also use linux and a wired controller for a cheaper arduino uno clone controller,

pc power supply or a 12 volt laptop one you could find or steal from work (they wont notice)

dvd drives work too for steppers ( two per drive, easy polar printer right there)

how do you get encoders from a closed-source printer system?, found a nema 17, a wierd 80's metric, and a cheapo stepper from an old type writer. I thought an easydriver would be fast, but if it is already there, I could use it like a floppy drive controller,

>>1277673

I got a digital one from AliExpress for cheap. Only goes to 100th of a mm, but thats accurate enough for me

>>1277673
Mitutoyo.

>>1277708
If he's asking on 4chan, he don't need no $150 measury pliers.

>>1277717
If I buy 15 Harbor Freight calipers and take the average of 15 measurements, will they be as accurate as one Mitutoyo?

who here

/maximuminfill/

>>1277703
>how do you get encoders from a closed-source printer system?

http://homofaciens.de/technics-base-circuits-transmissive-optical-sensors_en.htm#F4210

http://reprap.org/wiki/Optical_encoders_01

then

https://github.com/misan/dcservo

You need to tune the PID but it works.

Generally you can find ground by hooking it up to the printers circuit board and running a continuity test. The other side of the LED is then 5 volts and the remaining two wires are channel A and B.

>>1277708
>>1277724
>Memetoyo

The reason Mitutoyos are so expensive has more to do with overall build quality than accuracy. Even the cheapest chinese calipers are accurate enough. You may need to zero a chinese one more often to prevent it from drifting off too much and it may feel crappy but thats about it. Personally i use an old-fashioned vernier caliper for most measurements down to 0.05mm, and a micrometer when extreme accuracy is required. My micrometer is like pic related btw, before anyone starts accusing me of being a moron who never tried quality instruments.

>>1277717
This. I know the brand and would love one but I'm a general hobbyist using this for model trains and rockets.

>>1277563
Well I just suggested 5e because it has much less variation in character designs than pathfinder or 3.5. But I have heard of some IRL friends of friends doing it, I even saw one once, and I don't think they have the time on their hands to model them from scratch. Of course they could have picked a generic one off thingiverse.

There is Heroforge, which might be what my pals were referring to:
https://www.heroforge.com
But it probably doesn't let you download the 3D file, just order it.

>drawing my cetus clone
>looking at cetus 3d pictures
>notice the linear guide carriages are connected to the belts with a tiny part
>they're closed belts
>there's no way to put something on the other side of them to clamp onto the belts

cont.

>>1277679
>>1277703
NOT SHITTY ENOUGH
Diy linear optical encoders are damn shitty looking, and so will be whatever perfboard abomination I decide to make. I've got a RPi 2B lying around that I could use, though If I can use an arduino with minimal peripherals that will be even cheaper.

>>1277852
>cut a small bit off a spare belt
>punch a hole through it
>put a button-top m3 screw through the hole
>add a 3d printed spacer on the other side
>clamp the piece of belt to the printed part with a thumb screw

Holy dog shit they're pretty fucking sneaky if they actually did this.

I tried to break the belt by pulling it apart with two pliers. Nothing. The hole stretched a tiny bit, but as soon as I stop pulling the belt apart it returned to normal. Turns out fiberglass reinforced timing belt is strong as fuck, unlike those white steel-wire-reinforced PU belts.

>>1277856
Looks like this when the belt is in the aluminum extrusion channel.

>>1277855
try to use a set of digisparks with sprinter and the pi

>>1277865
wait, I call dibs on that, need something to test the new drivers, its the new drivers with stealthChop2, a silent mode actually designed for 3d printers in the datasheet

also if anyone knows whats the difference between the trinamic TMC2208 and TMC2224?

208 already works with the step sticks, so now that they've shipped, I want to test them out before I buy an adapter.

>>1277200

I wish I had a bowden extruder when i started printing because I fuck the z level calibration every time I change filament.

>>1277865
Both a Pi and some ATtinys? Are the ATtinys needed for expanded I/O ports or something? Or is it required for the sprinter firmware?

Is the duet wifi the best option for a controller right now?

>>1277900
>bowden extruder
Thanks, I'm looking it up now.

>>1277924
sorry not sprinter, that's used with dumb people

klipper, not sprinter https://github.com/KevinOConnor/klipper
needs updates for a attiny but just a linux cnc could work

its a setup that instead of gpio steppers from a pi to drive the driver pins, it uses individual micro controllers to drive the steppers, just to add a bit more layers of shit to interface

you could only use one or two to interface as well https://www.youtube.com/watch?v=Z0vQIaUZGu4

>>1277941
Well for the existing software I could write a bit of code for a microcontroller to turn normal stepper-motor signals into those that the encoder+motor combination will take, but the feedback delays might put things off. How hard would it be to write something to turn G-code into motor movements from the ground up? I'm thinking no more than 200 lines, assuming G-code is the "sequential co-ordinates and feed/no-feed commands" haven of accessible data I think it is. And that could probably run entirely off an ATmega328p + MOSFET board.

>>1277949
you should read up on how klipper works, with my limited knowledge, it offsets the computation to a python script on the pi and is sent to the microcontrooler via usb or gpio for multiple (unique) controllers using different drivers/ controllers for each axis defined in configuration,
https://github.com/KevinOConnor/klipper/blob/v0.5.0/docs/Code_Overview.md

Ever since I was a kid I wanted to design a car. This looks a hell of a lot easier than to work with clay.
Anyone here ever make their own car? Any /o/fags in general itt?

Oh wow, I just found out you can cast metal. This is amazing.

>>1278018
you can cast lots of things...

>>1277984

I think other threads on /diy/ would be better suited for making your own car.

Come back here when you need window handles or a unicorn horn gear shifter

>>1278023
Elaborate please. I'm very compelled to buy a 3d printer. Hopefully some will go on sale during black Friday/cyber Monday

>>1278024
Wasn't aware, I'll check the catalog, thanks. It just popped into my head as I was reading the thread.

>>1275662
Speed for infill is 40mm per sec, outside perimeter is 30.

>>1277855
>NOT SHITTY ENOUGH
Diy linear optical encoders are damn shitty looking

lol. So you're building one these shitty Zebra Crossing Piano Key type Linear encoders.

>>1278060
That looks perfect, but I'll have two channels like pic related so I can tell which way it's going more easily.

>>1278060
>>1278075
You can use the sensor from an old optical mouse and a piece of printed paper with tiny zebra lines on it

Source
https://youtu.be/CIRKRzw54Zs

>print few models that have holes
>holes always undersize
>print circle and square test
>all the sizing good

Hmm.. Well I'm new to this thing hope I'll solve it soon.
Thinking of making an fpv tricopter when I'm done with my printer upgrade. Still waiting for that filament shipment

>>1278108
You should always design parts with oversized holes, I usually do 3.4mm for M3 screws which are in reality about 2.8mm

>>1278108
>waiting for that filament shipment

Where is a good place to buy filament?

>>1277900
My flsun kossel delta came with a bowden, it works pretty great yeah.

>>1278110
Guess I should learn editing STL files or designing in general.

>>1278116
Depends from where you come from. I'm from SEA so I ordered mine from Hobbyking, 22usd per generic chinese filament including shipping.
I don't know any other good place and I was in a hurry. I think US/EU have better selection.

>>1278116
I get my ABS filament from 3dplastx, they ship from Lithuania in the EU, so shipping is cheap. 0.92 kg for 12-13 euros.

>>1278126
Enlarging holes and stuff like that is easy in TinkerCAD, and they are easy to get started with also. Made some brackets for my bicycle a few weeks back with it, works great if you don't need to make anything complicated.

>>1278116
>Where is a good place to buy filament?
Amazon

I just ordered some from HobbyKing and after shipping costs they seem to be averaging $13 per kg for whatever they have in US inventory.

That's only slightly cheaper than your average kg of filament from Amazon, assuming of course you have Prime and get "free shipping" that way.

>>1278025
Lightning bolts etc..

>>1278142
>>1278110
Isnt there a setting in slic3r to print holes to proper size?

>>1278240
In all honesty it really isn't the slicers job to factor in margins for holes.

Are there any consumer printers that can actually achieve <0.1mm tolerances? FDM or not. I was looking at the form 2 but a forum post I saw says it's as bad as 0.2mm to 0.5mm

>>1278317

But it _is_ the slicers fault if the size of parts don't match the dimensions of the STL file, when properly calibrated

>>1278338

Wanhao D7?

>>1278086

How about printer motors that come with the attached rotary encoder inside them?

>>1278343
>>1278240

It is not necessarily slicer's fault. I also use Slic3r and if I make a hole diameter 3.05mm it prints a hole that makes a tight fit with 2.85mm filament. But only if I print it at 0.2mm layer height. If I print at 0.3mm, the hole is too small.
I don't think it is a problem in slicer, I think that shrinkage of printed plastic is the problem. If plastic isn't cooled fast enough it shrinks. And it mostly affects small features.
- Try printing with more cooling or different fan shroud.
- Try lowering Small perimeters speed in Speed settings
And see if there is any change. Also your printer might not be properly tuned. If you have slight over- extrusion and a bit lower steps/mm in x and y axis, You wouldn't notice it on outer dimensions of parts, but you will get much smaller holes.
And elephant foot and seam can also be a problem.
There is also XY size compensation setting in Advanced tab. It says it is for fine-tuning hole size, but I never tried it, so I don't know how it works.

>>1272345
is there a way to skip/ignore fine details with cura 2.6.1?
one particular model wastes time printing dots of filament on the top layer

>>1278369
>It is not necessarily slicer's fault

Change the word "necessarily" for "entirely" and I agree with the rest of your points

When you change layer height/cooling settings in your slicer, and if you've properly calibrated steps/mm settings, then the slicer's gcode _should_ make a part that matches the model.

Output should match input details, in a properly calibrated setup.

Yes, there are lots of variables, but thats why there are so many settings for properly calibrating your setup

>>1278386
>2.6.1 cura

>>1272345
whats the consensus on slicer choices?
I got god-licenced s3d 4.0 for most stuff, sli3r just to see if I can just move over s3d settings to it, can never get anything good right now.
have the new cura 3 release and have long prints but could get good quality once I can get the cooling down and octoprint to stay working.

I also have marlin 2 but used to use the Italian marlin clone for a long time, that Italian version is good if you want sd only prints.

>>1278369
last time I did a hole test in slic3r it failed all the way to .4 when I can get .2 for square and .3 for round holes in s3d, so any help with those settings would be great

>>1278396
>2.6.1 cura
im lazy to update and afraid i'll loose all my settings

>>1278398
lat time I used cura I had a wanhao i3 back in july (not the included one, that is still there) and updated it two weeks ago, all the settings are still there. you could also export them.

>>1278338

It's gonna be very hard to get that kind of accuracy out of a consumer printer. *Perhaps* with a DLP printer like the Wanhao D7 as >>1278348 mentioned, but there are no guarantees it can actually produce parts like that. If you already own an FDM printer you may be able to get ~0.2-0.5mm detail out of it using a 0.2mm airbrush nozzle attachment, but it will be tricky.

>>1278400
>0.2mm airbrush nozzle attachment
Why? Nozzles already come in that size or smaller

>>1278414

Airbrush nozzles are pretty good for very small diameters because at that scale you need a pointy profile that will radiate the least heat possible over the part. While you can usually find nozzles with a 0.2mm diameter hole, you won't fine one with that kind of profile.

>>1278086
The Logitech G303 does have a pretty damn high resolution, but in either case I'd rather do it with new cheap parts than having to scour through the garbage.

>>1278350
Same goes for printer servos.

I might experiment with both if I happen to come across any, but I know the optical mouse method will be damn hard to code around.

is there any clones of the Deltaprintr Mini Hotend?
those custom parts look expensive, but the fast cool down and heat up speeds would be good for a dual extruder.

>>1278420
Where can I get one for a e3d chimera? Or two rather

>>1278433
never mind, found it, not that hard to find so I wont link spam.

>>1277221
I got a Cr-10 in September, I've been printing pretty consistently. Its a great starter but also has a ton of customization for people with more skill. Stay away from Gearbest, Banggood is good they have a representative on reddit that will help you a lot if you need it. There is a facebook group for the Cr-10 and they will provide a lot of help with questions regarding the printer and have some print profiles, stls, etc. There is a huge online community and the build volume and quality you get out of the cr-10 is worth the extra 50$ imo. People make kits for upgrading and you can print a lot of upgrades for it. First thing I recommend is leveling knobs so your fingers dont get raw and then the fan.

>>1277360
Take steppers and any other parts and sell them and buy a kit imo

>>1278439

Normally they're sold here, but they're out of stock:

https://well-engineered.net/index.php/en/shop/airbrush-nozzle-adapter

The only other case where i've seen airbrush nozzles is on the Merlin hotend, but that's an old design with a PEEK cold end.

>>1278350

Yeah, I got one of these out of a Brother Printer, it operated some pump mechanism for the Ink Jet.

https://www.youtube.com/watch?v=PU6tgQ84CEA

Not too shabby.

>>1278659

Now I know what to do with all my printer servos

What surface heat bed plate do you guys use? I’m not sure if I want the sticker thing, heard it left a big mess after.

I just started using glass sheet, it’s good as fuck but first layer not sticking well. Had to use masking tape with glue stick and now it’s sticking too fucking well.

>>1278722

Have you tried hairspray? It works for me for ABS, PLA and PETG. Also using a glass plate.

>>1278398
Back it up then, jeez.
Or install the new version separately.

I don't think this warrants a new thread, and this seems to be the most fitting one since a lot of the hardware is (vaguely) similar; I want to get a laser cutter soon. I'm not looking for something that can handle the thickest materials and is geared for mass production, but at the same time, I'm not looking for Babby's First Laser Cutter. I don't really know what to look for in these things, and unfortunately they seem to fall into these two categories - either I find $150 aluminum extrusion and 3D printed things that handle pieces that are 6x6", or I find massive machines that are >$5k which have a huge operating area. Is there anything in between, or am I SOL? Are there any reliable but not too expensive brands I can look into?

>>1275021
Recalculate your fucking normals, bucko.

Say I want to keep my printer in my garage. Is rockwool a feasible insulation/enclosure material to keep it toasty within the print zone? Is there anywhere i can get pieces cut to size instead of spending big bucks on a huge block of the stuff?

>>1278876
Make your own

>>1278876
Your categories aren't so much "small and cheap || big and expensive" but "co2 || not co2". You can /diy/ a not-co2 laser cutter of any size on a budget, something like 100x100cm might cost like $300 to make, but it will be slow as fuck because the best regular laser you can get is around 4-5W. To get an idea how fast they go, look it up on youtube. They're basically for cutting thin wood.
CO2 lasers can have a much larger wattage, but they're not garden-variety shit, you actually need to think about protecting everything else from the laser, and the laser is not cheap. You might need more precision in your cutter to run the beam to where it needs to be.
I'd recommend considering the option of building a CNC router instead, because while it will cost more and need a way more rigid construction, it gives you quite a bit more flexibility.

>>1278722

I'll likely be testing Printbite in a few weeks, will report here. Other than that, for glass i just put gluestick on, works universally enough. You can even use it with a cold bed.

>>1278659

Are encoders even supported in firmware like marlin? Would be nice if one could just stick a pair of encorders on X and Y for a more reliable closed-loop system.

is plastic 3d printing good for anything other than ornamental/trinket stuff?

>>1278923
Yes, you can use it in about the same areas you see injection molded plastic used. Cases for things, some mechanical parts, fittings.

>>1278929
when would you print a custom fit plastic mold part instead of just buying a new appliance? I don't see how 3d printing is anything more than a meme.

>>1278931
And I fail to see how your post is anything but bait.

>>1278931
When you custom design a device and there is nothing commercially available. If you come here with a foregone conclusion, you might as well just selectively read existing posts and leave the thread having confirmed your bias instead of shitposting it up.

>>1278931
Because no company makes this custom part I need. If you don't have mechanical needs and lack CAD abilities, you're stuck with a plastic garbage printer.

>>1278933
>>1278935
>>1278936
has anyone in this thread used a 3d printer to make something that is not just ornamental?

>>1278937

I have used mine to make useful things. Repaired my cat-flap, shower curtain bracket, cable tidying systems, phone holder for in my car, butt plug, childproofing hooks for kitchen cupboards

I use mine mostly for custom household shit. Most things I could buy, but it's cheaper and more fun to design and print them myself

>>1278937
Off the top of my head, I've printed a filament puller with gear reduction, a lawn mower, parts for quadcopter, a camera mount for plugging into people's asses, a custom LED worklamp's casing, some CNC parts, cabinet mounts for a LED strip, all the pulleys for a large belt grinder and of course various 3D printer parts which I imagine wouldn't count.

Hello i am interested in this stuff but i have one question.

How sturdy are the made components compared to "normal" factory production like molding and stufin general.

Als would it be possible ti leave empty spaces where you could insert metal parts for extra strenght.

Imagine an Z shaped product wich is 5mm thick and has interbally a left space lige zigzag to put in sheet zigzag metal to strenghten it even further.


Sorry for bad explanaiton but my pc is broken so i cant really draw in paint.

>>1278949
They're about 80-90% as strong as injection molding from the same plastic iirc. I don't quite get what you have in mind, but leaving spaces is possible to an extent, you might need printed supports depending on where it exactly is.

>>1278937
I intend to build a force feedback flight yoke with 3D printed parts, I also want to design and print custom flight sim panels, also print war gaming scenery and I've just had/stolen an idea for a mini holder to make painting easier. I'd also like to prototype a firearm action. Now that I think about it I had an idea yesterday to create a custom handle with built in hot keys for my graphics tablet so that I can hold it like a palette.

Use your imagination anon.

>>1278937
New collar for kitchen sink, panels to conceal edges of cutout in shelf, plug for sewage pipe, paintbrush holder, SD card holder

This is one of those things where you're literally only limited by your imagination.This is like complaining that you can't paint and your brushes are useless due to your lack of ability.

>>1278937

I'm using mine to print parts for a pen plotter CNC machine. I'm printing as I design them.

I also used it to print archery triggers, parts for another 3D printer, parts for itself, etc.

>>1272345
>>1272345
Just noticed this.
How fucked am I?

>>1278904
CNC router is also on the list, but I've really got my heart set on a laser right now - something that can cut metal would be grand.

>>1279003
your nozzle is not tight or your throat might be broken if its an anet a8

>>1279004
Then you're probably looking at a CO2 laser. Since the only thing on the moving rails are the mirrors and lens, they can be pretty lightweight and expandable, if a bit of a health and safety nightmare. The major cost is the laser tube itself, but if you want to trust the Chinese with them they're only $150 or so including shipping for a bottom-end 40W.

>>1278876
40w desktop co2 cutter

>>1279004
>something that can cut metal would be grand

Cheapest thing that can do that would be a plasma cutter.

Laser machines that can cut through metal are on their own category


https://www.youtube.com/watch?v=w92SSzBQGpE

>>1279041
That one flat-out states it's not for metal.

>>1279043
I don't mean things like 1/2cm thick steel, more like the mm-thick stuff you can find at Home Depot or craft stores.

>>1279044

A 150w co2 tube + PSU for a laser machine that can cut that kind of metal is like 2000$

https://www.youtube.com/watch?v=OCro_Q8SBdU

get a laser for your laser

Delayed due to weekend, print came out fantastic under 30mm/s. Casting sometime soon, can't tell when but I really underdesigned the clamping system. I could have printed a set of rings(I have one, at the top) instead of tabs to clamp it down on, and the locating rods at the bottom were supposed to be for screws, but I don't have any in this current size so I made them into locating holes.

I hope this'll stand up.

>>1278949
>>1278954
The z-axis is always weak, like as much as a small fraction the strength. If you print a rod straight up and try to hang something from it it'll snap very easily in comparison to a injection molded part.

In the other axes it's plenty strong since each layer is a near continuous line of plastic. You can definitely put a metal insert into the prints and just the z-axis alone is usually enough.

>>1279180
Than you two.
With Spaces left open i mean if you google an modern pvc window frame in cut secrion, you se inside the frame there metal parts biult in for additional streanght. Something like that.

Icwill think about it and if i start i will post here. Can polycarbobate also be 3d printed?

>>1279180
>z-axis is always weak
That's only if your print is shit, my parts from PETG never seem to have any affinity towards breaking along Z. It is inevitably somewhat weaker than other axes, but it shouldn't be a big difference.
>>1279205
Look at what kind of stuff others have 3D printed, you should be able to see if there's structure like what you want. Most likely it's possible.
>PC
Yeah, but you need some special parts for it, most notably a steel nozzle and an enclosed printer.

>>1279206
I mean like pi related where you can insert extruded aluminium for further strenght.

Can be bparts made a way that they can be clipped tofether so you can construct bigger stuff wich wont fit your printer directly?

>>1279218
Yes, although it's a hassle, but it's possible. You'd want to print parts with holes along Z.

>>1279219
That should be doable. Its basically like anny hollow object.

If we would have this tread on /b how long would it take for the first guys to want to print guns....

>>1279221
Printed guns are shit

>>1279223
I know. But you know /b

I would love to print stuff. But i dont know what i yould. Well an fan holder for.my gpu would be noice

Generic PLA is a NIGHTMARE!

So me and my roomates want to buy a 3D printer for shenanigans, figures to paint and so. We'd like something with quality but we're aware our budget can't afford a really good printer.

For now we're looking at a Prusa i3. How hard is it to build and use? We all have build our PCs and have above average tech knowledge, but we're not tech gurus or something. We also know there's some vapors method to smooth printings.
So how does the Prusa i3 sound for dedicated first timers? we have a budget of tops 600€ for everything, Printer and any accesory or extra hardware we need to buy.

>>1279147
You design these yourself?
What program do you use?


>Would you share the files...?

>>1279223

Not if they're from an industrial SLS machine

>>1279269

It's not hard at all, but it's nothing like building a PC, which will work just the same if the parts are crooked.

Vapor smoothing is typically used for ABS, which I don't think you'll want to print on an open printer like the Prusa i3.

Prusa i3 is fine, but if you want to spend even less I suggest some clone from Aliexpress, like the Anet A6. Once it's properly set up it will print with a very high quality, and you can replace the acrylic frame with one made from aluminum extrusions by printing some parts and ordering the extrusions off Misume or wherever. It's about 200$, and you may spend some money to replace all the rods and bearings it comes with.

>>1279322
Isn't Vapor smoothing something you do after printing it outside the pritner?

Also we're really not that cocnernced about the price and going cheap, we want the best quality for that price.

>>1279326

Yes it is, but you need an enclosed 3d printer to print ABS (which is the easiest plastic to sand down/vapor smooth with acetone) or solvents like chloroform to smooth PLA

Even really cheap printers can print with a good quality after a few inexpensive upgrades

>>1272345

How is the Tevo Delta? I only see shill reviews but it seems like one hell of a deal for $760 still.

>>1279329
What kind of upgrades? how hard is to apply them?

>>1279326
>best quality from the price
Look into Josef Prusa's i3 (prusa3d.com) - he's the guy who developed this line and is constantly upgrading it. There's 3 different versions now, but you can only get the later 2. Cheapest (kit) is 600, and it's pretty easy to put together.

>>1279331
>upgrades
Heated bed, rigid frame, enclosure
Bed's not too hard, frame requires almost a complete rebuild, enclosure can literally (and I mean LITERALLY) be a cardboard box

>>1279348
>Prusa i3 MK2kit for 600bucks

Jesus. Does it have ALL I need for a good 3D pritner or I still should buy some accesories?

>>1279351

It's 600 bucks because it's the one that works right after you assemble it

>>1279355
Is that not the case with other printers?

>>1279351
>all you need for a good printer
The only thing I could say it's "missing" is a steel nozzle, which you only need if you're planning to print more exotic materials and is cheap anyway. It works like a charm. Pic related, minimal adjustment.

>>1279359
>is that not the case
Absolutely not. A lot of the kits you'll get could have missing hardware, warped frames, dead steppers, cheap electronics that burn out, etc. etc.

I'm not trying to scare you but that's just how it is. If you're willing to put forth a lot of effort to fix/improve it, you'll get a lot of use out of the $200 printers but be prepared for a LOT of tinkering/replacing components.

>>1279370
How big is the difference in quality between a 200€ pritner and the Prusa i3 MK2s of 600€ (700 with shipping).

700 is too high for us.

>>1279359

It's a "your mileage may vary" type situation depending on how chinky and cheap of a kit you get. With cheap asia kits you'll notice deviations from the original i3 design like frames made of laser-cut acrylic instead of aluminium, crappy bootleg E3D hotends that jam, fire-hazard electronics... Stuff that a more passionate DIY-er would fix and upgrade over time, but you mentioned you wanted a "good printer".

>>1279382
I definetly want a good printer and I'm open to go a bit higher than our budget for it, but the model we want is too expensive (700) and the models you guys sais are around 200. It's not that I don't trust you because you definetly know way mroe than I do but such a gap in cost makes me doubt. Isn't there any model for like 300-500, kit or not, with good wuality? Or a 200 printer with some upgrades is really enough?

>>1279389
The prusa costs as much as it does not because it has that much more printer in it, but because it's a time-proven, famous kit and they provide support with it. The actual cost of a prusa is around $300. The $200 kits will usually have some shortcomings depending on the specific kit, ie. acrylic frame bends out of shape over time, control board terminals catch fire, but these are generally well known problems so once you decide on a specific model you can read up on it and know what to expect.

>>1279389

You may find the Creality CR-10 to be within your budget. It's not "guaranteed smooth sailing" like the i3 MK2 but it's likely to be something you can live with.

>>1279372
they speak truth, owner of a flsun cube kit,
starting from the power supply,
no ground wire,
no fuse,
no enclosure for supply,
no external mosfet for large headed bed, all power has to go though board (this can happen in some premade ones too)
cheap e3dv5 clone with an extra thermistor for "reasons"
only 4 stepper drivers included
uses a fucking stepper separator for dual z, somehow works fine fo you get the correct wires
2/5 of the stepper wires are inverted (and unlabeled), correct in software
ALL ACRYLIC PARTS - broke in a month
had to wait for new bearings to send in before I could use it, took 3 weeks to fucking send
extruder mount is so shitty it cant print petg until you add a flex adapter
no part cooling fan
bed leveling is shit, unless you upgrade
don't get the touchscreen upgrade, software hell
the community cant update their own software, flsun hasn't

buying a cheep kit will always bring problems, but wither get a entry, low feature premade because it will have a better warranty and safety standards, monoprice and the cetus3D are good, monoprice you need to make sure it has the mosfet though.

fuck creality, I got a printer with the same size (after upgrade) with auto leveling, faster printing, less z and y ringing, cube style footprint and enclosure compatibility all for 100$ less (150$ less from amazon)

I would recommend the cetus, great printer but it depends on what features you want.

>>1279393
>>1279398
>>1279402

Those problems you mention are fixed by simply buying a replacement for around 100 bucks, install it and that's it?
Also I imagine that even if they aren't that time-proof as the i3 it would take months or several dozens of prints to go shit.
I mostly want it to print decent sized figures (that's why the prusa interests me, 20x20x20 working space) for then be smothered and painted. Nothing too complex. Dunno about my roomated but hey I'm the one looking for all the info so fuck them.

Thanks for all the replies, btw. I really appreciate all this help and I'm writting down all of it.

>>1279404
you want figures, get a cetus 2 extended (brand new), 18x18x23 size, has a warranty, (unfortunately probably cant get a dual upgrade anytime soon)

creality is bigger and has a heated bed, but has a shit melzi with very old marlin that you need an arduino to unlock and update. but if your plan is turn on and print, both have flying colors and the ability to return if needed unlike others.

>>1279404
It depends on the problem. Some fixes are $1, others more. Ultimately the best thing is to build your own printer because you learn the ins and outs of it and can troubleshoot a lot better later on, and it's the cheapest option as well, but obviously the most time-consuming.

>>1279312
Yes.
Solidworks.

No, they're stupidly simple to make

>>1279450
Any reason you printed a mold for your *fuel vortex generator*? Wouldn't a big hunk of PVC be easier to cast into?

What's the most forgiving plastic for beginners?

>>1279614
anything cheap that's not ABS or has less than 3 stars on amazon

>>1279615
Thanks. I figured ABS was something I had to work up to. Can't wait to learn it so I can make some car parts.

>>1279614
PLA, I use HatchBox but they've had stock issues - I heard eSun is also good.

>>1279645
Thanks. What printer do you have?

>>1279645
esun petg is great if you have a dehydrator or use it only when its dry, their pla+ type stuff is good too, never had their normal pla tho, got a huge stockpile of hatchbox when it was on sale.

>>1279666
Prusa i3 MK1 (no longer sold, and I feel I don't need the upgrades [yet] )

Are delta builds a meme that can't match the quality of a prusa i3 clone?

>>1279686
yes, there is only 3 styles, mendel, ultimaker gantry and coreXY gantry. the mendel with one arm still uses the same mech, just needs less weight.

all others are shit.
prove me wrong.

delta, polar, darwin and 3drag should only be used when absolutely needed, such as large hotends for 3drag, belt/idler issues or nema 23 on a mill for darwin, and moving head weight or size constraints for polar and delta systems.

>>1279562
No way to non-destructively release the mold, and it would possibly increase surface area to the point of being unable to remove the positive. A three part mold with two of the parts being a shell will help me release the cast item. This way I am also reducing the number of materials I have to handle and have the material come into contact with, and seeing PLA effects on casting such matterials. Will be checking for deformities and creep.

>>1279802
try using the ski handle mold on thingiverse, also should put an air vent so insert wont have the suction,

I also think a flight style case would be the best style mold shape for handhold and possibility of a case with suction control

>>1279813
>air vent
A good suggestion. In this case, release pins may be beneficial, but the elasticity of the material(mind you, not silicone so will be much less rigid) may damage the material and will require a large enough pin. As for air vents, I may look into this.

>FL designs
Yeah, nah.

>handhold
A plebiean method, but to each his own. The end product is closed ended, and suction is entirely reliant on internal design - a ski handle is terrible and is overly bland.

>>1279691
Polar shit, darwin is basically an Hbot it works but meh and the twisting forces can cause issues, 3drag your asking to increase ghosting affects 10 fold by moving the bed around on the y and the x axis

delta had its hayday and its past.

Only kinematics that matter Cartesian, coreXY, UM style and HaqXY for FDM

>>1279686
Its shit, only good for impressing people.

Are there any printers out there that are particularly fast and precise for $1000 or less?

I'm in the rapid prototyping business and would rather not wait 3 days for a 12" tall model.

>>1279919
You can build one on that budget, buy ready-made, probably not.

>>1279921

I'm new to 3d printing, what format should I build for that budget?

>>1279923
CoreXY or delta. I personally prefer square printers so I'd go and have gone with CoreXY. But if you're new, best idea would be to look up plans for an existing printer and copy those rather than trying to design something from scratch. What you want to look for are things like
>lightweight gantry, potentially with a flex shaft extruder
>volcano nozzle and heater block
>wide GT2 belt, 9mm+
>either VSlot rails or linear guide rails, something that isn't shitty steel rods with LM8 linear bearings
>big motors, I designed mine for NEMA23 but most people stick to NEMA17
>maybe ballscrews for Z, this is not that important
>Z driven by a single motor
>preferably AC heatbed, especially if you want a larger build platform
>good cooling, I haven't gotten this far with my build so I can't say what works at high speed, but you need well designed delivery of air
>32bit controller, Arduino doesn't work well for high speed
Size doesn't make a huge difference in the design, most are pretty easy to scale either way. Except for rigidity of the frame, you need to put more effort into that on larger (300+) printers.

3d printed CNC or Chinese CNC?

>>1279932
Both of them are gonna be shit so just go with the cheaper

>>1279691
>>1279823
not that anon, but aren't delta printers (with bowdens) able to move faster than XYs/cartesians/etc?

>>1279003
If it's big print you'll end up with loads of little blobs that look like unpopped popcorn on the bed, just take it off and screw it back in when its hot

>>1279940
The speed increases aren't going to take a 12 hour print to being a 1 hour print, the increases in print speed, acceleration and jerk settings are only going to have a small increase compared to cartesians.

That's where coreXY can pick up the coverage.

Also worth noting that the hotend + extruder carriages aren't nearly the weight that they used to be.

Should I go ahead and buy a cr10s now for $490 shipped or wait until black friday?

>>1279928
>three ball screws
>nema23 motors
>two 20x20 beams for the x axis
>each of them with its own linear guide

What the fucking hell are you doing?

https://www.youtube.com/watch?v=6EtJfaiQ9H0&t=35s

>>1279919
Ultimaker2 clones from Aliexpress

>>1280097
Are you trying to imply that any of these cost more than custom CNC machined aluminum parts making up the entire body of the printer?
The ball screws are cheap, the linear guides are relatively cheap, and since I'd want to have at least one direct drive on the extruder I figured I'd make it more robust. I can always rebuild it to single-beam if I get a flexdrive, but I'm not convinced it's worth blowing $100 on.

>>1280101

No, I'm just seriously interested in what you're doing. Why not just one 20x20 beam for the x axis, for example?

>>1280101

If you're using two extrusions for the X because of the weight of the extruder, it doesn't really matter because a 500g or so load can't really flex 2020

>>1280103
As I said, I want to put a stepper on there, and a single beam would have the center of gravity in a weird position, which at high acceleration might deflect the printhead hanging off. I don't know if it would, and it depends largely on specifically how you mount the parts, but I figured I might as well go for the more robust design, these beams aren't that heavy anyway.
>>1280104
I know it can't flex the 2020, but rather I did it to avoid rotation around the beam.

>>1280108

That makes sense

Any thoughts on the flashforge creator pro?I'm not sure i can beat the price building my own assuming the print quality is good.

>>1280098
Cant find this one anywhere on the site for some reason

>>1280131
Search for CL260

>>1280132

No luck finding one for sale, just some Youtube videos showing how unreliable they are.

I just finished putting my chinkshit kit together. My heat bed and hot end aren't working. No power going to any of the blue 12v ports in the pic. The weird thing is that the heat bed worked the first time i tested it but not since.

Any ideas?

>>1280243
forgot pic

>>1280243
Fan and lamp on?
Thought E0 is hot end connection?

>>1280348
I've tried E0 and E1, neither are powering on (checked with multi meter). All the motors and fans are working.

>>1280243

Perhaps you have a new record on how fast the 12V mosfets on a chinkshit printer became out of order? Check if you're getting any voltage with a multimeter.

can someone explain marlin ubl?
I just took out my geometry ti83 triangle solver and made a 15deg triangle from each side based on my abl settings...

compiling now, on marlin 2 debug btw

I want a decent printer, but I would like to spend the least possible amount of money and at the same time I don't want to end up with a DIY plywood ghetto printer.

Any suggestions besides killing myself?