|
The globe ones she just buys from the dollar store or in bulk at Walmart on the cheap for craft projects. While she can probably save some money printing them with a resin printer, she probably wouldn't want to go through the hassle of the safety precautions and washing and curing the prints. I don't think time is as much of an issue with her, usually when she wanted me to print stuff it was a "Hey can you do a print for me overnight" type of situation.
|
#
?
Nov 23, 2021 16:23
|
|
|
|
| # ? May 30, 2024 13:39 |
|
|
SEKCobra posted:Any Longer LK1 owners done calibrating? It's driving me mad right now because none of the Marlin gcode works like I would expect it. M503 reports nothing, PID Tuning just seems to kill the machine etc. Anyone know what might be up with that? They don't even seem to have my firmware version on their website. I'm pretty sure there are only two owners of that particular machine ITT, tops (if you aren't literally the only one).
|
|
#
?
Nov 23, 2021 16:28
|
|
|
biracial bear for uncut posted:Also this. You could also do some pretty cool stuff with transparent resins and some little LEDs.
|
|
#
?
Nov 23, 2021 16:32
|
|
|
biracial bear for uncut posted:I'm pretty sure there are only two owners of that particular machine ITT, tops (if you aren't literally the only one). I bought one but haven’t taken it outta the box yet so I can only offer moral support.
|
|
#
?
Nov 23, 2021 16:37
|
|
|
biracial bear for uncut posted:Because even the fastest filament printers on the market take hours to print things of any appreciable size, and the linear print movement speed dick waving people do about their machines mean very if you're printing anything that isn't a box. Klipper wants a word with you. until my ender 3v2 broke from the speed until i beef up some other areas. But until then it was printing 2x as fast as I was for good quality
|
|
#
?
Nov 23, 2021 16:38
|
|
|
biracial bear for uncut posted:I'm pretty sure there are only two owners of that particular machine ITT, tops (if you aren't literally the only one). I'm pretty sure we were at least 3.
|
|
#
?
Nov 23, 2021 16:40
|
|
|
Roundboy posted:Klipper wants a word with you. Yeah? People like to say this but never post a "This was printed at X linear speed setting and took Y total print time" vs. "This was printed with Klipper at X linear speed setting and took Y total print time" in this thread to prove they actually gain anything where it matters. Oh, a print that normally takes 12 hours now takes 10 hours? Wow, good job putting excessive wear and tear on your machine by pushing it beyond the parameters it was designed to handle and reducing it's total lifespan to print one thing faster.
|
|
#
?
Nov 23, 2021 16:45
|
|
|
biracial bear for uncut posted:Also this. I haven't done any ornament printing, but I'd actually think you'd better off doing ornaments in two pieces on a resin printer. The globe with a big opening and one end and a separate "cap" for the hook that you can insert and either glue or weld in with resin. A nice big opening like that plus a relatively thin shell would mean you wouldn't even really have to worry about resin getting trapped inside. On that note, the discussion here last week or whenever about drain holes and hollow prints got me to do some experiments. I'm beginning to wonder if drain holes actually make things worse under certain conditions. I tried printing a bunch of tiny tanks and tiny spaceships (pretty much my only personal use case for hollow printing) both with and without drain holes. Did my usual cleaning routine and then left them both to sit for a few days before cutting them open. The interiors of the vehicles with drain holes clearly had some poorly/incompletely cured resin. This is almost certainly 100% my fault and something that could be addressed with better and more thorough cleaning because I suspect it's due to goopy IPA-diluted resin that didn't dry out enough. That said, the models without drain holes were fully cured inside, or at least cured enough that I can't imagine any kind of failures occurring. What seems to be happening is that basic gray resin is translucent enough for resin on the inside of the print to cure as long as it's just a thin layer and not big blobs. The interior resin had clearly cured fairly well since it was brittle enough to snap bits off. Just to be clear, I'm not suggesting that it's a good idea to print hollow models without drain holes. I think this works under the very specific conditions of printing models that are basically big boxes with relatively thin shells. I also have no idea if there's still any kind of issue with fumes from curing causing pressure problems that could crack/burst a model. I don't really have time to play around with this more at the moment, but I think it's something I'm going to try putting more effort into in the future.
|
|
#
?
Nov 23, 2021 17:03
|
|
|
biracial bear for uncut posted:Yeah? People like to say this but never post a "This was printed at X linear speed setting and took Y total print time" vs. "This was printed with Klipper at X linear speed setting and took Y total print time" in this thread to prove they actually gain anything where it matters. Fair point. i did post a benchy with as good or better quality and a time saver. But yes, my printer is down with the added stress of the retractions as I didnt tune it properly. I will say that klipper and linear acceleration / pressure advance / G2 arcs? does look nicer then the default cura stuff. Especially with blobs on curved surfaces that can't be hidden. Plus the remote send aspect and monitoring (not a klipper exclusive, but octoprint) as been better overall
|
|
#
?
Nov 23, 2021 17:06
|
|
|
Any slicer developer that figures out a way to have G2/G3 supported natively in the output that goes to the printer is going to find some radically increased print quality just by eliminating all of the direct linear transitions that come off of STL output from whatever 3d modelling software (this is usually an "arc fit tolerance" in your more traditional CAM software that generates output for CNC mills/etc.) I know there are scripts out there that do this but they have limitations as far as how good a job they do of this vs. most of the arc fit tolerance scripts used by the traditional subtractive machining CAM packages out there.
|
|
#
?
Nov 23, 2021 17:37
|
|
|
Paradoxish posted:I haven't done any ornament printing, but I'd actually think you'd better off doing ornaments in two pieces on a resin printer. The globe with a big opening and one end and a separate "cap" for the hook that you can insert and either glue or weld in with resin. A nice big opening like that plus a relatively thin shell would mean you wouldn't even really have to worry about resin getting trapped inside. Also, the one thing to really watch out for is the suction forces on the model if there isn't a "drain hole" (that name is pretty misleading). Models will rip off supports very easily if there isn't at least one hole to reduce suction forces -- this happens a lot where models do have holes but they don't occur until well into the model.
|
|
#
?
Nov 23, 2021 17:37
|
|
|
biracial bear for uncut posted:Any slicer developer that figures out a way to have G2/G3 supported natively in the output that goes to the printer is going to find some radically increased print quality just by eliminating all of the direct linear transitions that come off of STL output from whatever 3d modelling software (this is usually an "arc fit tolerance" in your more traditional CAM software that generates output for CNC mills/etc.) isnt the problem the control board of the printer? I have read that I can install arc welder and convert all my gcode to g2/g3 movements in Cura all day, but when it gets to the ender (3v2 board revision 4.2.2) it simply can't do those movements so its internally translated to small x/y movements anyway. This is bypassed with klipper since that (via PiZero 2) is becoming the new control board. I think i read I can magically get this in my current board if I cut a trace / remove a resistor ,etc but ill just buy a new board before then ..... for my Voron build
|
|
#
?
Nov 23, 2021 17:42
|
|
|
InternetJunky posted:This is a cool test and I look forward to your future experiments about it. That said, I'd be nervous if there is even partially-cured resin at all left in the model. It really does seem to disintegrate the cured resin. Yeah, I have a lot of concerns about this in general, so I'm mostly just trying things out of curiosity. That said, I was surprised at how well cured the inside of the parts were. I'd also be worried about the fact that you can't really know if it's working without cutting the piece open or waiting to see if you have a goopy mess a year in the future. Might be useful for saving some resin if you're printing a ton of identical models and don't mind wasting one to check the interior first, though.
|
|
#
?
Nov 23, 2021 18:09
|
|
|
What's the maximum thickness I can print something in resin and be sure that a curing station will fully cure it?
|
|
#
?
Nov 23, 2021 18:13
|
|
|
Roundboy posted:but when it gets to the ender (3v2 board revision 4.2.2) it simply can't do those movements so its internally translated to small x/y movements anyway. ... all movements will eventually get translated to small x/y movements on a cartesian machine.
|
|
#
?
Nov 23, 2021 18:20
|
|
|
Bucnasti posted:now I look and feel like a proper mad scientist while printing. As one should.
|
|
#
?
Nov 23, 2021 18:40
|
|
|
mewse posted:... all movements will eventually get translated to small x/y movements on a cartesian machine. Right, but any controller worth using can do the math to produce a much smoother arc than most slicers will output without producing ridiculously long strings of code output. CNC controllers have been able to do this since the 1940s. We aren't stepping that far back on processing power with 3d printers. EDIT: The CNC shop at Delta here in Atlanta still uses a machine originally built back in the 1960s to machine parts (edit: I used to work for one of their subsidiaries and a tour of that shop was a neat field trip to go on when I worked there). Some Pinko Commie fucked around with this message at 19:05 on Nov 23, 2021 |
|
#
?
Nov 23, 2021 19:02
|
|
|
biracial bear for uncut posted:Yeah? People like to say this but never post a "This was printed at X linear speed setting and took Y total print time" vs. "This was printed with Klipper at X linear speed setting and took Y total print time" in this thread to prove they actually gain anything where it matters. Citation needed on reducing machine lifespan, because I'm pretty sure you're again talking out your rear end -- or you'll cherry-pick one obscure blog post from 2013 to support it.
|
|
#
?
Nov 23, 2021 19:22
|
|
|
insta posted:Citation needed on reducing machine lifespan, because I'm pretty sure you're again talking out your rear end -- or you'll cherry-pick one obscure blog post from 2013 to support it. Literally on this same page. Roundboy posted:until my ender 3v2 broke from the speed until i beef up some other areas. Also just in case you forgot that 3d printers are CNC machines: What happens when you push stepper motors to run faster for longer periods of time? Related question: How much does the lifespan of a stepper motor decrease every second it spends operating at temperatures that are higher than it's duty rating? Some Pinko Commie fucked around with this message at 19:32 on Nov 23, 2021 |
|
#
?
Nov 23, 2021 19:28
|
|
|
biracial bear for uncut posted:Right, but any controller worth using can do the math to produce a much smoother arc than most slicers will output without producing ridiculously long strings of code output. I'm just taking issue with the "klipper is doing magic" thing when a 32-bit control board not running klipper can interpolate an arc just fine
|
|
#
?
Nov 23, 2021 19:34
|
|
|
mewse posted:I'm just taking issue with the "klipper is doing magic" thing when a 32-bit control board not running klipper can interpolate an arc just fine I know, it's maddening that people think Klipper does anything that a slicer can't do if the developers adapted the same kind of scripting you can find elsewhere to run on each "slice" of a 3d print. Also it can be run on an 8-bit controller as well. We literally have a CNC machine at work that has one as the math processor to handle arc interpolation (and that's a machine from the early 1990s). Like, you can replace literally thousands of lines of linear movement code with one G2 or G3 line of code. It's stupid that slicers haven't integrated this functionality just because "oh well, STL files don't have real curves so neither should the GCODE output have G2/G3 commands".
|
|
#
?
Nov 23, 2021 19:36
|
|
|
An excellent point R the "Why the F are we still using STL" pile.
|
|
#
?
Nov 23, 2021 19:53
|
|
|
because nobody wants to implement a CAD kernel just to slice models
|
|
#
?
Nov 23, 2021 20:09
|
|
|
To be fair the break was likely related to retraction settings being appropriate for a normal bowden tube setup, but insane high when using pressure advance. My filament broke right outside the extruder in the tube. Its an easy, but annoying fix, and my latest issues were related to lack of adhesion causing stringing, and normally that is not an issue, but with the acceleration speeds being so much higher, it wrapped itself around some fun areas and snapped my printed shroud as it cooled. Honestly the most annoying part is that I had to replace the thermistor wires a couple times, which sucks when its covered in hardened plastic. Knipper really isn't doing anything crazy that modern boards should also be able to do, but they just dont implement them. I think its the simplify 3d vs Cura type thing. The 'standard' was set and Cura comes along and starts throwing in features that become commonplace.
|
|
#
?
Nov 23, 2021 20:39
|
|
|
Roundboy posted:To be fair the break was likely related to retraction settings being appropriate for a normal bowden tube setup, but insane high when using pressure advance. My filament broke right outside the extruder in the tube. Oh, so the Klipper-caused broken printer was ... the filament snapping? Followed by nozzle blobbing, which is also 100% caused entirely by Klipper and doesn't happen on Marlin period? GUESS BIRACIAL IS RIGHT, KLIPPER IS TRASH.
|
|
#
?
Nov 23, 2021 20:45
|
|
|
I know this is in jest, but i believe if my acceleration was normal marlin levels my fan shroud would have survived. Then again that part has been a weak point in the past causing me to reprint it a few times due to settings and infill and temp, etc . I switched to proper PLA because of silk pla screwing me on that.
|
|
#
?
Nov 23, 2021 20:55
|
|
|
BMan posted:because nobody wants to implement a CAD kernel just to slice models I remember hearing once that e.g. STL is ridiculously easy to work with and integrate into a software project, like you can be up and running in an hour easy. By contrast one of the open/free CAD formats like STEP was basically "Ugh I give up" territory. (I don't remember who it was specifically I heard that from but I do remember they were been-there-done-that for slicing software so I figured they probably knew what they were talking about.)
|
|
#
?
Nov 23, 2021 21:00
|
|
|
Zach Weinersmith knows what's up http://smbc-comics.com/comic/halt 
|
|
#
?
Nov 23, 2021 21:22
|
|
|
BMan posted:because nobody wants to implement a CAD kernel just to slice models I do. Also at least one person is working on it. https://github.com/karl-nilsson/sse/blob/master/README.md Also scripts like this exist: https://github.com/DeepSOIC/SplineTravel But should really be part of a given Slicer's output from pressing the Slice button. Some Pinko Commie fucked around with this message at 21:49 on Nov 23, 2021 |
|
#
?
Nov 23, 2021 21:41
|
|
|
canyoneer posted:Next do a fractal vise i’m actually planning on it / working on early part modifications to be castable; the fractal vise design is inherently very accommodating to my awkward and limiting 3d printed mold:cast metal workflow constraints, where any individual component requires it’s own mold to be designed and printed from like $90 USD/litre Sculpt Ultra resin, and where the molds must fit on my tiny Mars Pro print bed. Making a single large vise component, say, is largely impractical because I probably can’t produce the tooling without a lot of compromises to make it printable on a bed just~ 2.5” wide, and the large tool print will cost a lot without any guarantee it’ll print successfully or produce good castings. The fractal vise only requires several individually-smaller molds per size of jaw element, and then i just cast multiples of various parts and assemble. The vise body/foot itself is gonna be the biggest component, but it doesn’t have to be cast at all if I tie everything into burly aluminum extrusions or sth like that. Much less costly to experiment or take a gamble on any given mold design decision or tool geometry choice. None of the existing fractal vise designs ive seen so far are suitable for being cast, not without a lot of modification or needlessly-complex molds. Most of them don’t seem like very good vise designs in the first place, frankly, the design seems extremely vulnerable to upthrust and deflection away from the vise bed because of all the extra jaw articulation points, and I haven’t seen a good approach for preventing that deflection (or they exist but are fragile and inherently-uncastable). That said, they look cool as poo poo and people like them, and being able to quickly and economically manufacture usable metal fractal vises based on expired patents could turn me a pretty penny down the road, so i’m picking away at my own revised version to make tooling for. No way in time for Christmas, but down the road it should come together. That and a hundred other projects… Ambrose Burnside fucked around with this message at 22:11 on Nov 23, 2021 |
|
#
?
Nov 23, 2021 22:06
|
|
|
I first saw the fractal vise from Teaching tech who apparently designed one for 3d printing. Is this the same vein as what you are talking about or is this one inherently different ? https://www.thingiverse.com/thing:4904044
|
|
#
?
Nov 23, 2021 22:14
|
|
|
There is also this totally legitimate looking site if you want to buy a metal one.
|
|
#
?
Nov 23, 2021 22:16
|
|
|
Randalor posted:What's the maximum thickness I can print something in resin and be sure that a curing station will fully cure it? I’ve never hit an upper limit for this, and i’ve printed things that had features that were functionally inch+ square blocks of solid resin. My impression is that the initial LED layer cure is the level of cure the ‘deep interior’ of any resin part ends up with, and that’s fine because it’s sealed off from environmental UV and oxygen. post-curing seems primarily a surface treatment, at least for opaque resins; my impression is that it provides a harder-wearing face over a tougher, less brittle core. I believe it functions much like case-hardening in metalworking, at least in the mechanical property sense. A full ‘deep’ cure could make a part more brittle than you’d want; thinner parts clearly cure brittle all the way through in a way thicker parts don't, which maps on to my own experience with case-hardening steel. cured resins are actually fairly complex composite structures of multiple materials, assembled in very particular structures/patterns instead of the homogenous mass we see with the unaided eye, so I’d think that ‘differential tempering’ plays a significant role in print mechanical properties just like layer bonding from pull-off /suction etc forces play their own roles. The other important thing postcuring does is react any unreacted photo initiators in the surface resin. Sometimes prints “never cure fully” and remain tacky on their surfaces forever, no matter what you do; this is because the photoinitiators can undesirably scavenge oxygen from the atmosphere, depleting them before they can do their work on the unlinked polymer component of acrylate resins. Post-curing does a really thorough job of using up all those surface photoinitiators to guarantee an inert, fully-cured outside surface. Ambrose Burnside fucked around with this message at 23:20 on Nov 23, 2021 |
|
#
?
Nov 23, 2021 22:30
|
|
|
Has anyone used Siemens Solid Edge? I want to put threaded holes in my STLs without paying for Solidworks.
|
|
#
?
Nov 23, 2021 22:53
|
|
|
Roundboy posted:I first saw the fractal vise from Teaching tech who apparently designed one for 3d printing. Is this the same vein as what you are talking about or is this one inherently different ? Same basic thing, yeah. There's at least 3 or 4 iterations with design differences, they all popped up within a week or two of an original antique fractal vise from the late 19th/early 20th century getting a lot of play on reddit/imgur, iirc. The features that link all the jaw members and constrains how they can move are the involved parts that complicate molds, I can't just reproduce any existing design due to the undercuts and lack of drafts, the cast parts would lock in the molds and you'd have to destroy the tooling after a single use to free up the cast part. biracial bear for uncut posted:There is also this totally legitimate looking site if you want to buy a metal one. I saw that. I enjoy all the PATENT PENDING bluster, given how prominently this all started linked to a century+ old defunct patent. also extreme lol @ taking $150 preorders for a product that has no projected launch date and no projected final price honestly a fully-machined 'proper' steel fractal vise just seems to lack a sensible use-case; it's going to be a $$$$$ labour-intensive specialist fixturing tool, which means machinists and other serious fabricators are the target market, but they already spend lots of time and money on their workholding, they don't need a one-stop-shop solution, especially if it doesn't perform as well as the Kurt or toolmakers' vises they already own. The alloys I'd be using can sustain 12 tons of force per square inch before creep or deformation becomes an issue, which is just as good as cast iron or steel for almost all visey applications, and could be sold for a fraction of the cost of a steel fractal vise by entirely avoiding machining anything. getting ahead of myself as usual, let's see where i'm at once i've got these cast involute wrenches locked down Ambrose Burnside fucked around with this message at 23:20 on Nov 23, 2021 |
|
#
?
Nov 23, 2021 23:17
|
|
|
https://www.youtube.com/watch?v=XHndF7PU5Yo&t=189s I am looking for the top feeder print to move my stepper/feeder to the top of my machine, aiming to take the strain off the filament traveling down the side of the machine and into the feeder then the extruder (Using brittle glow in the dark). Anyone able to point me in the right direction? Also anyone able to point me in the right direction for a better fan shroud, just ordered replacement fans but I feel like the shroud could use an upgrade also. I have 2x the stock fan sizes coming so it can use more of them if need be, but am looking to improve it also while I am working on my printer again. Thanks in advance
|
|
#
?
Nov 24, 2021 00:13
|
|
|
The Eyes Have It posted:I remember hearing once that e.g. STL is ridiculously easy to work with and integrate into a software project, like you can be up and running in an hour easy. By contrast one of the open/free CAD formats like STEP was basically "Ugh I give up" territory. That's the difference between polygon meshes and NURBS surfaces, yes.
|
|
#
?
Nov 24, 2021 00:28
|
|
|
GotDonuts posted:https://www.youtube.com/watch?v=XHndF7PU5Yo&t=189s Assuming it's PLA, the filament shouldn't be brittle... that most likely means it needs to be dried. Longer feed paths are generally not a good thing. Having the most consistent feed path from the extruder to the head is ideal. Having it on the Z is a good place for it. There are lots of filament guides, or you can go with a side-saddle mount. Chep publishes a good one. I have been able to run absoltuely everything through my ender using the standard spool mount, mounted backwards (so it's on the front of the machine) through this filament guide: https://www.thingiverse.com/thing:3950799 The Satsansa duct is very good, and supremely easy to print. https://www.thingiverse.com/thing:4369859
|
|
#
?
Nov 24, 2021 01:34
|
|
|
Sagebrush posted:That's the difference between polygon meshes and NURBS surfaces, yes. Also my takeaway was that the formats are not just worlds apart data-wise, but also existing tool-wise as well. It sounded like to implement STEP he was looking at having to roll his own importer from the ground up. Sure the format is open, but it's also complex and there was really nothing out there to make life any easier integration-wise. It turned out to be way more work than could be justified for a feature that was solidly in the "nice to have if feasible, but can't justify spending too much time on it" column.
|
|
#
?
Nov 24, 2021 01:35
|
|
|
|
| # ? May 30, 2024 13:39 |
|
|
I have an Octopus Pro board. It's hooked up only to power from the PSU. I have TMC2209 drivers connected into MOTOR0, 1, 2, 3, and 4. I have a Fysetc 12864 Mini panel plugged into EXP1. There are no other cables plugged in. I haven't changed any jumpers - they are as they were when I unsealed the board from the packaging. I compiled a Marlin firmware update, saved it to an SD card, inserted it, and powered on. The Power LED comes on and stays solid, the Status LED blinks a few times in sequence and repeats the same sequence, and there's ticking sounds from the 12864 display. If I eject the SD card and reboot, the same issue persists. If I unplug the 12864, the ticking noise doesn't come back, but the Status LED just blinks on and off rapidly. Anyone know if I missed something or need to do something to get it to a basic power-on point?
|
|
#
?
Nov 24, 2021 03:08
|
|
