MEMBER | PRINTERS | WILL PRINT FOR YOU? |
epicepee | kessel mini | |
vvp | rostock delta (from kit) | no |
damorgue | REDACTED | NO TITANIUM FOR YOU |
CPTBadass | REDACTED | NO DMLS, ABS, PLA, NYLON, STEEL, ALUMINUM, TITANIUM, ETC. ETC. ETC. ETC. :| |
__red__ | unknown | NO (idea) |
interesting interesting. i hadn't seen the delta systems, but there's very much a mill vs lathe thing going on here.
i don't know what your plans are regarding your printer, but we should do some same-model assembly tests and comparos to see how the two methods stack up. on the one hand, you have much bigger bounding box. otoh, your servos have to move things around that have variable mass over time and could potentially be pretty heavy.
needs: programmers to help me make a faster skeinforge (translate it into a nicer typed language, basically)
modelers to make stuff
models that are well suited to 3d printing
people who are willing to go bananas
Why 3D print a crappy-ass gun when you can go to Cabela's and buy an heirloom piece for a few hundred bucks more?More?
I counter your argument with a "Why not?"Because it could bring the Justice Department to your door. Because it could blow up in your face the first time you shoot it. Hobby 3d printers aren't always known for the strength of the finished item. Particularly if you don't get the settings right, your plastic got exposed to moisture, or you just got cheap filament.
hope you're planning on putting SA profile caps on those :))
yah, it should be able to pump those out though.
that was the first thing boost asked for
that was the first thing boost asked for
(emails mkawa 3d models for handcuff keys)
Whoa whoa whoa, why do we want plastic forms of me Tym? :eek:
Whoa whoa whoa, why do we want plastic forms of me Tym? :eek:
I mean of your avatars 0.0 Slow down there! :P
needs: programmers to help me make a faster skeinforge (translate it into a nicer typed language, basically)
modelers to make stuff
models that are well suited to 3d printing
people who are willing to go bananas
as part of my actual work work, i will very likely be working on a gcode tolerance estimator and validator. a good benchmark suite and a clean gcode generator will be imperative.
otherwise, it's going to be a relatively high quality 3d printer whose services will be available to the community at much lower than market margin (which is anywhere between 5-10x and 20x). proceeds will be reinvested in the community (3d scanner, other tooling, etc.)
a couple possibilities: 1) rewrite skeinforge in a highly disciplined dialect of C. use Clang for enlightened compilation and analyze the llvm. 2) rewrite skeinforge in a Java 4. Use wala for enlightened compilation and analyze the bytecode. 3) rewrite skeinforge in haskell or ocaml (ie, some ML dialect). hack up ghc or ocamlc (I have extensive experience in the latter, and access to enough experience to get going in the former) to do enlightened compilation. analyze the AST (afaik the lowest IL in both compilers is just an AST).
advantages and disadvantages to all options. will have to think on this one. If you're a phd applicant, I encourage you to apply to either UCI or UCLA's (latter is iffy) PL program to participate in this. If you do though, you need to let me know first because I will have to write the grant to fund you ;). Otherwise, you're welcome to pop in on the github anytime as we work. :)
Why 3D print a crappy-ass gun when you can go to Cabela's and buy an heirloom piece for a few hundred bucks more?More?
Considering the price of a 3d printer, the real gun is cheaper, mine is up to $700 in parts (almost $100 in nuts and bolts alone!), and the one Mkawa is getting is about $3000. You still need about $30 in plastic as well.
I don't know typical gun prices, but I'm sure $700 would get you one (used maybe?), and $3000 certainly would.
If you plan on having someone make it for you... Forget it, you need a firearms manufacturers license to do that.I counter your argument with a "Why not?"Because it could bring the Justice Department to your door. Because it could blow up in your face the first time you shoot it. Hobby 3d printers aren't always known for the strength of the finished item. Particularly if you don't get the settings right, your plastic got exposed to moisture, or you just got cheap filament.
Mostly, because it was meant as a technological demonstrator. It wasn't meant to really be used, it was meant to show off 3d printing as well as scare gun manufacturers and anti-gun nuts alike. You have to admit, it's impressive when you can anger both gun lobbyists and anti-gun lobbyists equally, and at the same time.hope you're planning on putting SA profile caps on those :))
yah, it should be able to pump those out though.
You would probably be disappointing with 3d printed keycaps. Not only are they not smooth, but the last one I saw who tried, the stem broke.
As for pumping them out... a Keycap with a decent finish, just a guess, but at least 5 minutes each. You also need another 5-15 minutes to heat up the bed and nozzle before you can even start and similar cooldown before removing it. A full keyset, even doing several at once, could easily take all day.
needs: programmers to help me make a faster skeinforge (translate it into a nicer typed language, basically)
modelers to make stuff
models that are well suited to 3d printing
people who are willing to go bananas
as part of my actual work work, i will very likely be working on a gcode tolerance estimator and validator. a good benchmark suite and a clean gcode generator will be imperative.
otherwise, it's going to be a relatively high quality 3d printer whose services will be available to the community at much lower than market margin (which is anywhere between 5-10x and 20x). proceeds will be reinvested in the community (3d scanner, other tooling, etc.)
you can make a plate full of adapters/keycaps in a single run of the printerIf they fit on the build surface, but you are still looking at 10 hours or more, especially with decent quality.
As Damorgue said, purpose plays a large part in which is best.
The Replicator 2 is a nice printer, it's certainly one of the more professional looking and the easiest to get up and running. You can certainly do much worse. Just beware the build area, I doubt it will handle much more than a GH60 case. I went with a Delta style system for that reason (the last of my parts arrive Thursday).
The biggest problem I see you face now is expectations.
Those who don't understand 3d printing tend to underestimate them, but those with some understanding of them, tend to greatly over estimate them. The claimed 100micron, due to the system, has to be done EXTREMELY slow and printing in general is slow to begin with.
these should be printed undersized on a coarse setting on their side in PLA. to avoid rafts or other supports, the cherry cruciform will be highly tapered orthogonally to the way that MrInt has tapered his, and the alps end should taper and have dimples chunked out as it hits the bottom of the switch socket. this way, the adapter has all its sheer strength vertical to switch, and focused at the joint between the two cruciforms.
once the r2x comes in, we can reprint a some of your more egregious parts. i would also consider redoing the plywood eventually with heavier material. the light weight of the outer frames of the diy printers are a classic "bad thing". the error introduced by the movement of a head is inversely proportion (to some reciprocal function) of the weight of the frame. this is why bridgeport mills weigh several tons.The parts all fit now, so no worries there, just took some elbow grease. The wood isn't nearly as much of an issue on a Rostock Delta as it is on Cartesians as it's not really applying much force to it. Newer Deltas are eschewing wood, but not because of wood being problematic, but that they created simpler/cheaper designs that no longer require it. I could have started there, but I wanted the extra build volume of this style.
oh man, i think you've infected me leslie. a delta max design could be incredibly helpful for certain kinds of projects here. for example, the really long build area could end up working extremely well for keyboard cases (which otherwise have to be cut up and then fastened together).:)
oh man, i think you've infected me leslie. a delta max design could be incredibly helpful for certain kinds of projects here. for example, the really long build area could end up working extremely well for keyboard cases (which otherwise have to be cut up and then fastened together).:)
I love watching the Deltas work, it's just amazing to see them function, it's almost organic.
Like I said earlier, that length is why I went with this, space to do a gh60 (and TKL), of course it has to be done vertical but it should work with some scaffolding, and eventually two at a time to reduce that, if I make them for others. I don't plan on keeping the Rostock as my only printer for long, I plan some upgrades for it, but I also want to make a Cerberus derivative, which has a 400mm high build area, large enough for a TK.
I don't have a problem with the Cartesian/Reprap/Makerbot design, I just like the Delta design better. It uses less desk space, has more room, and just looks so futuristic. I kind of look at it as the second generation of home based 3d printing. However, time will tell if it becomes the more dominant design. There are a lot of people invested in the style you have but many are also switching over as the design becomes more stable. The Rostock itself is only about a year old and it was extremely revolutionary. Many doubted it would even work and even claimed it was fake.
If you do decide on another, especially a Delta, I would recommend looking at either a Kossel or a Cerberus (both are still in development by the designers). Both are as cheap or cheaper than a Rostock. The Kossel has a similar build area as your Makerbot, but makes for a real nice, and cheap Delta that performs extremely well (and MUCH quieter!!!), I passed on it only due to build area size. It's about $100 less to build than a Rostock, and is the second design from the same guy who designed the Rostock. Cerberus is the work of a Lotus kit car designer who redesigned the Kossel and Rostock to form his own printer. It has an even larger build area than the Rostock for a similar or slightly less price.
Other options are Mini Rostock, which can be done for pretty cheap as well and would make for an excellent second printer for smaller items. If I start making things that sell, I plan on one of these. Parts are easy and cheap to get/make, including pre-cut Lexan panels instead of plywood (Ebay has these). The biggest drawback is the size (of course) and the noise... The linear bearings on Rostocks are VERY NOISY! Many recommend printing ABS bearings to replace them. Which isn't a bad idea anyhow as it prolongs the life of the rather expensive rods (quality rods for a full size can run from $90-$180!). The rods are why the Rostock Max, Cerberus and Kossel were designed the way they are now.
All of this is open source (including most of your Makerbot), so you can swap over parts, use yours as reference, and of course, build another entirely. Like I said, your second can be a lot cheaper if you D.I.Y. it.
Oh.
I forgot about another printer earlier, turns out there is a pre-build Delta and of decent/non-wood quality. It's called Spiderbot (http://www.spiderbot.eu/). They are still getting going, and their build area is smaller, and you would have to import it, but it's a nice, pre-built all metal Delta for a decent price. If I remember right, their base kit is $1200 after exchange rate and the top end was $1800. I like it better than the Rostock Max, but if it means saving $1000, I'd rather just build one myself.
If it wasn't so expensive to operate, I would build a stereo-lithographic printer as my next. These are the ones that use a laser to cure resin, they have incredible resolution. The printers themselves aren't any more expensive (you can use a Blue-Ray laser and you need no extruder). The problems start once you get finished building it and find that the resin costs 25 times more than ABS. :eek:
So instead of costing $0.001 per key, it will cost $0.0025? :eek:
HMMMM some intriguing stuff here XDYeah, but how does it change when heated... does it warp or expand?
http://www.mcmaster.com/#glass-stock/=mue6y7
Can either of your printers print with soft materials?negative. shrinkage is crazy with these materials, and their melt points are way too high for FFM printers.
Something along the lines of 10A Rubber?
Or can they only print hard ABS plastic?
MCM may be good, but wanting $30 per linear rod (I needed 6) and $20 per bearing (I needed 12) was a bit on the expensive side.what McM is good at is high quality very precise materials. this makes pricing look pretty bad at low MOQs but it's actually pretty good because the guys on ebay and amazon that are selling rods are not selling rods that are as precisely cylindrical as the McM parts.
Even nuts and bolts were 3 times the price.
I haven't looked at glass-mica, but for even heating. most guys are putting two thermisters at opposite ends, and if your heater covers the entire surface, you should get relatively even heating. With your aluminum bed, that shouldn't be an issue.
Can either of your printers print with soft materials?
Something along the lines of 10A Rubber?
Or can they only print hard ABS plastic?
what McM is good at is high quality very precise materials. this makes pricing look pretty bad at low MOQs but it's actually pretty good because the guys on ebay and amazon that are selling rods are not selling rods that are as precisely cylindrical as the McM parts.I used VXB Bearing, they have quality rods and bearings for about half MCM's price. The rods I got were good, the bearings, meh. I plan on printing replacement bearings anyhow.
Seems plenty stiff to me. To stiff actually. But I figure its better for me than plastic.Can either of your printers print with soft materials?I'm not sure how stiff 10A rubber is,
Something along the lines of 10A Rubber?
Or can they only print hard ABS plastic?
but there is filament that replicates silicone.Yes I saw that mentioned on some random website. Of course there are a lot of different types of silicone at all different hardness levels. When I hear "silicone" the first thing I think of is "boobs" and at first I was thinking that would be too squishy to print keycaps out of... but I did some more research and at least some silicone should be strong enuff to hold up.
This is from a Rostock Max (wood framed) with an E3d head/.40 nozzle/no cooling/ABS/heated glass bed/magnetic rod ends.Good enough? Gosh, it's almost perfect as far as I can see, geee ...
Seems good enough for me...
Do you need a scanner as well?I think we missed this one, sorry.
Can you explain a little about how a dual extruder head helps print/extrude silicone? Does it spread the heat out between two heads?
Do you need a scanner as well?
Can you explain a little about how a dual extruder head helps print/extrude silicone? Does it spread the heat out between two heads?
You don't need dual head to do silicone (it's actually a plastic that feels like silicone), however if you were to make keys from it, it would be far too soft and floppy.Soft and floppy is what I want. Of course they need to support their own weight and the stem is kinda small. And given how small keycaps are I figured we might have to make them with an extra thick stem and maybe an extra thick top so they would have the required strength and rigidity.
My thinking was with dual head, you could build a regular key underneath and make a coating of silicone over the top, similar to rubberized keys.Aha! ur so devious! ;D
I was thinking it could be done with a single head by just doing it in a separate operation, but odds of pulling it off are just too slim, time consuming, and just an insane way to do it.I doubt it would ever be accurate enuff trying to use 1 head in 2 totally different construction passes.
A second head isn't that difficult or expensive to add anyhow (about $75 for me). Once I get mine running how I want, I will likely do it.Gummi Bears support their own weight. That's what I need. Of course Gummi Bears are cheating and have 10x the thickness of the wall of a keycap stem.
Here is one (http://www.youtube.com/watch?v=MlMTFr9qu9Q) silicone like plastic, it's a bit stiffer than the other I saw. The other I saw is soft like Gummi Bears.
Soft and floppy is what I want. Of course they need to support their own weight and the stem is kinda small. And given how small keycaps are I figured we might have to make them with an extra thick stem and maybe an extra thick top so they would have the required strength and rigidityYou cannot thicken the stem much at all as it must go into the top of the switch housing.
I doubt it would ever be accurate enuff trying to use 1 head in 2 totally different construction passes.Actually a few companies are working on quick, auto change systems right now. One is close to production. It works on CNC machines, no reason it can't here, the problem is making it cheap enough.
Gummi Bears support their own weight. That's what I need. Of course Gummi Bears are cheating and have 10x the thickness of the wall of a keycap stem.The only way is to layer it over something stronger.
If there was such a thing as a material that was strong in one dimension while being soft and squishy in the other dimension that could work perfectly. Horizonatlly rigid while being squishy vertically would theoretically absorb all the shock of bottoming out.That is one of the wonders of 3d printing, like fiberglass, how you layer things and how solid you make it, can make things flexible in one way and stiff in another. They are making mini plastic leaf springs that work, it's just a matter of figuring it out.
pics!Show Image(https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcQCqzI3Br-9Lcyag-WujmSHLI4_EortT5kBxQAeNV-aVaRffjz42A)
210mm x 210mm x 325mm build area
It's probably going to make it's (horrible) first print shortly. Then it's on to fine tuning.
The latest method though is using an Xbox Kinect, they are having great results from that, and the new HD version has people really excited because the added detail will make scans far better.I visited a startup company not long ago: Volumental (http://www.volumental.com/). You use Kinect, a browser plugin (Windows and Mac only) and their servers do the computation. When it was demonstrated to me, uploading and downloading took most of the waiting time. The result is a mesh and a texture map.
Tell us how noisy it is. :)Actually, since I got the better bearings and rods, it's not that bad. It's similar to the square tube railed ones on Youtube, such as Cerberus and Kossel.
Probably a lot till linear ball bearings are not replaced with bushings.Yeah, it's a bit smaller especially since my heated bed is smaller than that as well. A more realistic number right now is 150mm-180.
Leslieann, do you have diagonal rods longer than the common 25cm? Because if not than you cannot do 21cm x 21cm square, it is more a circle with about 24cm diameter. The one I can get to is not finished yet and it has 25cm diagonal rods.
First print.
A 10mm crooked cube. LOL
taking potatocam to a whole new level lol
First print.
A 10mm crooked cube. LOL
Oh God.. WHAT HAVE YOU DONE?!
taking potatocam to a whole new level lolThe lens was blurry, it's a tiny object, and there was movement in frame, not a good situation for any potato cam.
can you illuminate (lol) how you did your heated plate? materials etc? i want to (but may not get around to) playing around with rotating heated mica/silicate beds. the canonical example of an object that will print well in a polar system and not a cartesian is the simple cylinder (whereas the cube, or worse, the high ratio rectangle) is the vv example.I got a PCB based heat bed (which is smaller than my build area) along with belts/pulleys, Arduino, sd adapter, and endstops in a smoking combo deal on Ebay.
i will definitely start by playing with mica-silicate non-rotating beds though, so definitely, parts, bom, measurements, etc. (reminder: pick up a type-k probe and fluke adapter or dedicated unit for precise contact measurement -- IR is going to pick up and average too much of the bed, exactly what you don't want when you're measuring uniformity).
I got a PCB based heat bed (which is smaller than my build area) along with belts/pulleys, Arduino, sd adapter, and endstops in a smoking combo deal on Ebay.using a pcb as a uniform filament... BRILLIANT!!
I covered it with green glass and Aqua Net hairspray (make sure it's cold or it will look like Michael Jackson's hair). Also you want to cover anything such as mechanicals to avoid the spray as it can gum things up. I sprayed some on a paper towel and wiped it onto the surface 5 prints so far without having to reapply. I use 8 clamp type paper clips to hold it around the perimeter and ensure good contact as the pcb's are never flat.buy a granite surface plate at your local machine shop shop, knife shop, or carpentry shop. lap the glass on the plate. the glass is much heavier than the pcb and as long as it's thick enough (it's thick, right? if not, buy a thicker one and run more current through the pcb) will not be deformed by the pcb.
PCB beds heat faster, cost more and use more power, mine plus glass and clips takes about 7 minutes to heat up, but I think I need a stronger PSU, as the heat system cycles I can see my LED build area lights flicker a touch. My bed alone is sucking up about 11 amps, plus the 5 for the rest of the system really taxes the PSU the little PSU I'm using (I have 1 or 2 spare amps on an old, used psu). I also had to add a fan to the Arduino as the heater was overloading the circuits.HMMMM... interesting point about not-needing uniformity... as much as the right temp at the center (where you're mostly printing)... food for thought.
No matter how round you try and make something, you have to remember that you are working in digital, at some point you hit the pixel level and it will simply not get anymore round. Although, I'm seeing it, despite being digital, if you slow down the jerk movement and run a few degrees cooler/more fan, you can actually round the corners. At least on a Delta. While making the cubes, a few corners on a few levels ended up rounding due to the speed of the arm jerk versus the speed of filament extrusion. I ended up with a really nice rounded corner.
As for temps, I fired a few temp probes on my bed, one I crushed two were damaged while adding heat shrink (thermistors are damaged by an open flame, but soldering irons are fine...) so I am using a 10k resistor from an old pc fan I had laying around. It's quite inaccurate, however I'm having no issues with the bed temps, I just kinds guessed on a how off it might be and it works fine, it's off by about 2-3 degrees at room temp, so I set it 6-8 degrees below target temp, How did I get that, well, for each degree after that, it gets increasingly longer to heat up, and I got tired of waiting and just decided to print and see what happens. LOL It worked.
Point is, don't worry about temp accuracy, find a temp that works. Nozzle size, filament type and even filament brand will change what temps work best. On a .5 nozzle, I get flow around 220c, but I print at 235c. On a .35 nozzle I don't even see flow until 240 and print at 245. When you go too high on either plastic, you can smell it and when it's too low, it won't do anything. My experience so far has been that it should slowly drool when it gets near the right temp and a full nozzle.
To get started, worry more about getting things running right, instead of trying to have it all figured out and measurement-wise perfect. Nothing on these is perfect, and you will probably just over complicate it. You are going to probably spend a long time just figuring out the best temps, nozzle, and THEN you get to play with extrusion settings. Precise temps are the least of your worries and are something easily figured out through trial and error, and are in fact best figured out in that manner. I have heard of ABS varying from 215 to 260 for best melting point. Like I said before, forget the precise number, find what works for you and your combination.
I had the same problem with my first saltwater aquarium, I went all out with all sorts of expensive equipment and I just over complicated it. I downsized and went to a simple all in one, with nothing special and had a great working tank. Same with the printers, get what you REQUIRE and figure out where to go from there.
the makerbot alum plates cost 75$. the idea is that you use them for a little while until the heating and cooling cycles warp the metal or the teflon coating nicks or worse. then you throw it out.WTH!?
i have power supplies. christ i have a lot of power supplies.I do too, however finding one that can push 18+ amps, wasn't in the scrap pile. :(
:O it's in colour! Thought 3d printing can't include colours in it
holy crapDamn that came out great, I'm a bit envious at how easily you got up and running.
Even the cheap DIY 3d printers can do colors. The question is why would one bother with it?
Everyone can join in on this fun too now!
http://www.kickstarter.com/projects/pirate3d/the-buccaneer-the-3d-printer-that-everyone-can-use
Noooo, kawa, post more stuff here. I love this thread!
What differentiate the $2000+ printers from the $300+ ones?the same thing that differentiates all expensive from cheap tools: precision and accuracy.
great list kmiller. this is now officially the living 3d printing thread.
don't forget the upcoming photocuring printers. i can't remember the name of any of them at the moment but the basic idea is that you take a resin that cures under UV, you point a projector downward at a tub of the stuff, and you slowly raise/lower whatever it is you need to raise or lower to form a solid object.
people 3d printers owned will print for you?
-------------------------------------------------------------------------------------
leslieann rostock delta (from kit) no
mkawa makerbot rep2x (from box) not yet! debugging!
kmiller8 still deciding duh, no
What differentiate the $2000+ printers from the $300+ ones?the same thing that differentiates all expensive from cheap tools: precision and accuracy.
don't forget the upcoming photocuring printers. i can't remember the name of any of them at the moment but the basic idea is that you take a resin that cures under UV, you point a projector downward at a tub of the stuff, and you slowly raise/lower whatever it is you need to raise or lower to form a solid object.Resin is 15x the cost of ABS, I looked at building one.
basically the practical method behind additive manufacturing is to take a material that transitions between phases of matter (solid, liquid, gas, plasma), and bring from a non-solid to a solid in a controlled, compounding matter. this differs significantly from subtractive manufacturing in that you usually take a big old brick of solid stuff and hack away at it until it looks like you want it to. one was invented by cavemen. THE OTHER BY MEN AND WOMEN OF THE MODERN SPACE AGE
anyway, this is all a roundabout way of suggesting that you consider investing in more tooling than product, leslieann. at the moment, you're so focused on making a print that you're not really sure what, where, or how everything is working.
What differentiate the $2000+ printers from the $300+ ones?Replicator 2x is more of a high end home printer. It's no more accurate than any other as it's all similar systems. It's just a Prusa or Mendel in a fancy box with better documentation and all the bugs worked out.
completely different error profile, scale and purpose. resin photocuring has the potential to model additive photolithography. you're unlikely to want to make a safety cone with it (and you would have some definite trouble doing so anyway since you're relying on a chemical reaction that may or may not be malleable in terms of color), but you might want to make a the exoskeleton of a tiny robotic swarming bee.don't forget the upcoming photocuring printers. i can't remember the name of any of them at the moment but the basic idea is that you take a resin that cures under UV, you point a projector downward at a tub of the stuff, and you slowly raise/lower whatever it is you need to raise or lower to form a solid object.Resin is 15x the cost of ABS, I looked at building one.
basically the practical method behind additive manufacturing is to take a material that transitions between phases of matter (solid, liquid, gas, plasma), and bring from a non-solid to a solid in a controlled, compounding matter. this differs significantly from subtractive manufacturing in that you usually take a big old brick of solid stuff and hack away at it until it looks like you want it to. one was invented by cavemen. THE OTHER BY MEN AND WOMEN OF THE MODERN SPACE AGE
The printers costs similar, but the resin costs will kill you.
It's no more accurate than any other as it's all similar systems. It's just a Prusa or Mendel in a fancy box with better documentation and all the bugs worked out.only in my wildest dreams. a few bugs are worked out such that you can print a traffic cone when you unwrap the box (or at least you may be able to). anyway, as i also have said incessantly, error analysis and correction is all incremental. a bridgeport is just a sherline in a fancy 3-ton box (this is not an exaggeration). but the weight of a bridgeport is essential to reduce specific kinds of error and produce the kinds of extremely high tolerance parts that power the increasingly efficient internal combustion engines and composite molding tooling that the modern world is made out of.
Cheaper ones have smaller print sizes, and depending on model, more or less tinkering required to make quality prints. Some also require special filament, which is normal filament, just wrapped in a fancy overpriced package.what is filament? it's wire made out polymer? what's the composition of that polymer? what is the typical shape is the extrusion? what is the tolerance of the bounding diameter of the filament? the word fancy hides the details that drive the incremental improvements that need to be made to achieve that incrementally higher accuracy and precision that may be boring to some, but is extremely important in empirical science.
What is filament? it's wire made out polymer? what's the composition of that polymer? what is the typical shape is the extrusion? what is the tolerance of the bounding diameter of the filament? the word fancy hides the details that drive the incremental improvements that need to be made to achieve that incrementally higher accuracy and precision that may be boring to some, but is extremely important in empirical science.You missed what I was after there.
... all the diy printers we're discussing are all hobbyist grade. there is a high degree of experimentation required to print useful output with all these devices, and the only two materials that can be printed thus far are PLA and ABS (and then only filament)
PLA is a nightmare to get right, it turns out I broke both 120psi pneumatic fittings that hold the bowden (feed) tube. Seriously, the fittings hold 120 psi, and I snapped the hell out of both of them. :eek:This is interesting. Lets assume 120psi is the typical rating for pneumatic fittings used in DIY 3d printers.
What differentiate the $2000+ printers from the $300+ ones?I would like to know too. The point is that companies charge whatever the market will bear. What they charge is unrelated to how much effort is behind the product and what is the product value to users. There is only a lower limit on price. That is the cost of producing the product. If the 3d Printer market is commoditized then you can expect that the price corresponds to the value. Is this true for the current 3d Printer market?
very cool. this is only a little higher than the temp that the MBI filaments are locked at, but that's a software limitation. i suspect there's some fudge that would allow 240C and maybe a tiny bit more without melting anything important.... all the diy printers we're discussing are all hobbyist grade. there is a high degree of experimentation required to print useful output with all these devices, and the only two materials that can be printed thus far are PLA and ABS (and then only filament)
And nylon and maybe more. But yes PLA and ABS are the common ones.
Glad to see some good come out of my posts.PLA is a nightmare to get right, it turns out I broke both 120psi pneumatic fittings that hold the bowden (feed) tube. Seriously, the fittings hold 120 psi, and I snapped the hell out of both of them. :eek:This is interesting. Lets assume 120psi is the typical rating for pneumatic fittings used in DIY 3d printers.That means the force rating of the fitting is about 10 N. Typical extruders (for 1.75mm filament) push with force in the range of 10 - 15 N. So we are overloading the fitting a bit. But still looks like your fitting had a defect since the safety factor should be at least 2.
- 120psi (i.e. about 827kPa) pressure rating
- a bowden for 1.75 filament has about 2mm radius
Or you pushed the filament with significantly higher force.
Thanks that you document the problems you experienced. It is an interesting reading.
and leslie, the MBI machines aren't even sherlines; have you seen the machining on one? they are very well made, and subtractive printing is something that is pretty well understood at this point. i just went into detail on how half my "turnkey" machine is totally bogus and i have to put hundreds into measurement, tooling and elbow greay to improve my heated bed ON TOP of the thousands that MBI spent to produce the heated beds..I consider what you have to be more like a Cadillac with all the bells and whistles. The fact that you already found a bunch of issues (many of which go back to the design it's based on) just reinforces that idea.
the posters in this thread have many years of engineering experience combined, leslieann. you can choose to ignore their suggestions, but it sounds like what you're doing isn't really working for you, so we have offered them.
fwiw, the first thing i would do about your extrusion head is buy or build a multi-input temperature readout device (fluke is overkill, extech is about right, no-name is not worth it. you will be using this a lot. velleman offers reasonably priced kits, although i think they still use PICs -- yuck. sparkfun or adafruit might have a more modern kit using an atmel MCU...), a number of contact (type K is imo the de facto standard) thermocouples with nice high thermal limits and reasonable accuracy (+/-10C is probably fine), and a roll of kapton tape, then measure the temperatures up and down your head device from the nozzle up to and including the feed tube coupler.
i'd also buy a very high quality and small DC clampmeter that can measure the amperage feeding all the power devices in your chain non-contact. these things generally push too much current for general purpose ammeters. this will cost serious money, as you'll want to be able to measure relatively small (10s of amps) currents with it, and the cheap ones are basically useless until you're pushing car motor starter current (3-500A). keep in mind that these all use hall effect (basically they're measuring the B field perp to the current), so you want a small, sensitive clamp. i haven't had the greatest luck with these so i can't offer more specific suggestions.
the only thing i can really tell you is that i generally don't buy constructed things if i can construct something comparable myself with reasonable effort, as i have far more invested in tools than i do in _stuff_. that said, i do admire your tenacity and persistence. you'll get there, but i will insist that it might take a bit of a change in perspective if you want to get there sooner rather than later.
also, i find it odd that only mechengs really know and understand controls and tolerances and yet when they inevitably get turned into digital computations, it's completely up to a CS or worse, an EE to implement, and we know nothing about controls. it's quite silly actually, and points to a huge gap in both curriculums.
Several professors were asked to solve the following problem: "Prove that all odd integers are prime."
Mathematician: 3 is a prime, 5 is a prime, 7 is a prime, 9 is not a prime - claim is false.
Physicist: 3 is a prime, 5 is a prime, 7 is a prime, 9 is an experimental error, 11 is a prime ...
Engineer: 3 is a prime, 5 is a prime, 7 is a prime, 9 is a prime, 11 is a prime ...
...and the first engineering joke i was ever told :PQuoteSeveral professors were asked to solve the following problem: "Prove that all odd integers are prime."
Mathematician: 3 is a prime, 5 is a prime, 7 is a prime, 9 is not a prime - claim is false.
Physicist: 3 is a prime, 5 is a prime, 7 is a prime, 9 is an experimental error, 11 is a prime ...
Engineer: 3 is a prime, 5 is a prime, 7 is a prime, 9 is a prime, 11 is a prime ...
Mathematician: 3 is a prime, 5 is a prime, 7 is a prime, therefore, by induction all odd numbers are prime.
Physicist: 3 is a prime, 5 is a prime, 7 is a prime, 9 is an experimental error, 11 is a prime ...
Engineer: 3 is a prime, 5 is a prime, 7 is a prime, 9 is a prime, 11 is a prime ...
the only open source gcode generator that i know of right now is skeinforge. i'm planning on rewriting skeinforge for the purpose of rigorous analysis using formal methods. i'm also hoping to look at some of the open source CNC mill and lathe compilers as well. if you know of any other FOSS manufacturing compilers or can NDA me in on anything, i'm happy to explain to you what the goals and methods of this research are.http://slic3r.org/
it's often told in a way that's less kind to the mathematician:He should have used COQ or some other theorem prover ...QuoteMathematician: 3 is a prime, 5 is a prime, 7 is a prime, therefore, by induction all odd numbers are prime.to explicate the joke: the mathematician uses formal reasoning to make a dumb mistake.
...
As for the head, I found some the issue with the print head. The teflon liner shrank, probably from heat, apparently, it's a common but little discussed issue with this head. The other problem was too much retraction, I was sucking up a blob of filament into the teflon and getting stuck. I used a bit of my bowden tube to replace the teflon insert, and cut way back on retraction and I'm back to getting okay enough prints that I can make my J-Head mount and possibly even the magnetic arm parts I need.What was the too big retraction? What retraction do you use now?
What was the too big retraction? What retraction do you use now?J-Heads have about 20 different versions, some longer than others. Some use cartridge heaters, others have a different Peek end. They aren't just revisions, sometimes they are just options. Seems a bit much to me.
Edit: Hmm, J-head seems to be quite short compared to what I saw at the printer I can get to. I do not know how it's head is called.
this is brilliant! thank you!!!the only open source gcode generator that i know of right now is skeinforge. i'm planning on rewriting skeinforge for the purpose of rigorous analysis using formal methods. i'm also hoping to look at some of the open source CNC mill and lathe compilers as well. if you know of any other FOSS manufacturing compilers or can NDA me in on anything, i'm happy to explain to you what the goals and methods of this research are.http://slic3r.org/
AGPLv3 license
AFAIK, it is only for 3D printing.
I use Slic3r (mainly because it's simple and I learned on it), but you will want to use something to verify the code. It's known for not always doing things right and I've had several things I fed into it and come out withing entirely different. Like a horseshoe shape coming out as three cylinders...that is so perfect for my project i literally want to cry with joy right now.
The Rostock here does not have any problems with direct drive extruder. It uses 0.45Nm stepper with drive wheel radius of about 4.5mm. That represents force of about 100N. Should be plenty for 1.75mm filament and 0.5mm nozzle. It does not show any problems when printing at 6cm/s.
Here is an interesting info about forces required from an extruder cold end.
http://airtripper.com/1338/airtripper-extruder-filament-force-sensor-introduction/
Unfortunately he does not specify his nozzle and the filament.
I use Slic3r (mainly because it's simple and I learned on it), but you will want to use something to verify the code. It's known for not always doing things right and I've had several things I fed into it and come out withing entirely different. Like a horseshoe shape coming out as three cylinders...that is so perfect for my project i literally want to cry with joy right now.
sounds good leslieann. slowing down is NOT a bad thing in any way, both the head and your methodology. increasing motor speed introduces a ****load of physical error, not to mention makes it much harder to deal with temperature issues etc. etc. etc. etc. etc. 235C is about the right temp for the filament. what i haven't been understanding is why you have so much heat and pressure in your extrusion system in general. if you've been trying to get it REALLY hot and then trying to cool it down REALLY quickly so that you can print uber fast, that would explain quite a bit.
I wasn't trying to heat it fast and then cool it, I was trying to maintain a stable temp through the print. I've determined that the problem is heat soak on the hot end. The Peek and internal teflon heats up, increases pressure on the filament and things come to a halt. If I lower temps and speeds, it takes longer to heat soak but still does. If I increase temp it heat soaks faster, either way, I don't get any further with a print job. Fans do nothing, insulation on the head did little. The head just heat soaks the same no matter what.wait wait wait. slow down here turbo. i'm not sure what you mean by heatsoak, because i've only heard it used for what i imagine is a different meaning (heatsoaking in automotive terms is the process of thermal migration. with big old cast motor blocks it takes a long damn time for heat to spread). my guess on what you're saying is that you have filaments that you're pushing current through, and you have a thermistor somewhere (where?), but you have very little control over the temperatures in the head for some reason? the way this typically works is that you have a big fat filament, you have a simple controller (my meche friends keep telling me variants of 'i wasn't even aware there was an alternative to PIDs?!' sigh), and you feet the thermistor and the filament voltage into the pid with the filament voltage as the variable and the thermistor as the control target, yah? does this describe what you're doing? if not, please describe what it is that you're doing.
also the latest news on my end is that i have everything back together and i'm printing a bunch of traffic cones to use to push through the carpet under my machine + random pieces of acrylic that the machine sits on. it turned out that the machine was totally not level at first because it was missing a little rubber foot from the factory (HAH), and on top of that, all i really have to mount it on is 40lbs of random acrylic and a berber carpeted floor. so, traffic cones.You should be able to get head to plate down to about .0025mm, I'm down to about that. I'm also getting similar accuracy in x and y, so long as my belts are tensioned right. Still getting the hang of how often and how tight they need to be. This is still with sloppy arms.
few reasons for this: my head to plate tolerances right now are around +-0.1mm
wait wait wait. slow down here turbo. i'm not sure what you mean by heatsoak, because i've only heard it used for what i imagine is a different meaning (heatsoaking in automotive terms is the process of thermal migration. with big old cast motor blocks it takes a long damn time for heat to spread).
I wasn't trying to heat it fast and then cool it, I was trying to maintain a stable temp through the print. I've determined that the problem is heat soak on the hot end. The Peek and internal teflon heats up, increases pressure on the filament and things come to a halt. If I lower temps and speeds, it takes longer to heat soak but still does. If I increase temp it heat soaks faster, either way, I don't get any further with a print job. Fans do nothing, insulation on the head did little. The head just heat soaks the same no matter what.wait wait wait. slow down here turbo. i'm not sure what you mean by heatsoak, because i've only heard it used for what i imagine is a different meaning (heatsoaking in automotive terms is the process of thermal migration. with big old cast motor blocks it takes a long damn time for heat to spread). my guess on what you're saying is that you have filaments that you're pushing current through, and you have a thermistor somewhere (where?), but you have very little control over the temperatures in the head for some reason? the way this typically works is that you have a big fat filament, you have a simple controller (my meche friends keep telling me variants of 'i wasn't even aware there was an alternative to PIDs?!' sigh), and you feet the thermistor and the filament voltage into the pid with the filament voltage as the variable and the thermistor as the control target, yah? does this describe what you're doing? if not, please describe what it is that you're doing.
hence, what is happening is either that the teflon is getting too hot (your theory, leslieann) and causing the abs to hit transition too early OR the abs is not moving quickly enough at some/any point, and the filament at the heater is getting hot enough to melt the filament all the way up the teflon tube. now the tricky bit is that it seems to me that once either of these things happens, both will happen, and then the whole thing will blow up.
which seems to be what's going on.
is that blackened bottom part where the teflon slides into the peak? or does it actually slide into the heater (because that is bad.)That black strip is actually where the bowden tube clamped it, before I used it as a new liner. You can ignore that.
The teflon does butt up against the brass nozzle, but they all do (otherwise it creates issues). The difference is that this end uses a huge metal end/head which encapsulates both teflon sleeves. There is no way to even begin cooling the teflon until it reaches the Peek material. This makes for a very large transition area for filament.
here is the problem i have with this. teflon is a thermal insulator (among plastics it's certainly not the best, but it also has a very very high melt point and is slippery, so i think it's a good choice of materials here). that means that it's hard to heat up, but that also means that it's hard to cool down. it should not be getting hot, period. correct me if i'm wrong, but i believe what you just said is that the teflon does extend into the heated metal portion (unless the brass nozzle extends all the way up into the peek -- it is not clear from the diagram. imo this is a design error to me (but a fixable one). ideally, you want an instant transition from solid to very not solid when you hit the heater. that means no thermal insulator ever between heated portion of head and not-heated portion of head and then infinite amounts of current into the heated portion such that it is a theoretically constant 230C or so (ie, instaneous thermal recovery).
If you were to maintain a sharp temperature gradient, then the heat input could be increased, thus allowing a more rapid build speed and decrease the risk of clogging before it reaches the nozzle. The way this is usually achieved is by a better insulator (mostly delays the point where the gradient reaches through, which isn't suitable for long builds) or you apply a barrier which removes heat, ie a cooling block. The cooling block would need to be compensated by increased heat input, but it alloows for the temperature gradient to be sharper.Correct
This is what I gather from your descriptions.
darmogue breakin it down. basically you either need to keep your cool end cooler while keeping the hot end constant or v.v.
You will need to remove heat from the PEEK by any means possible. I was trying to think of a good way to divert heat away from there, but the problem is that it moves around and any extra size and weight like heatsinks or connections in the form of hoses to the moving entity will carry disadvantages.Correct, it's a Rostock Delta.
Just to confirm, with that model it is the head which moves around and not the build, correct?
i would rank "exploding head" as a bigger problem than "adding some inertial error" and start adding heatsinks. cheap extrusions are plentiful at any electronics surplus storeYes, but water cooling is so geek chic! :cool:
Here is an interesting info about forces required from an extruder cold end.
http://airtripper.com/1338/airtripper-extruder-filament-force-sensor-introduction/
Unfortunately he does not specify his nozzle and the filament.
Interesting, I skimmed it a bit but will go back and read more of it.
He did mention the head, under a picture, it's a J-Head V9 clone (J-head is an open source design and tons of people are making them, some good, some not so good and in many wild variations).
His filament is 1.75mm based on the pictures (scale of teflon vs other parts of the extruder). I have an Airtripper extruder, works well, but it's much more complicated than some of the newer designs.
few reasons for this: my head to plate tolerances right now are around +-0.1mm and my head to head (remember, two heads) tolerances are about 0.05mm. with abs and a 0.4mm head, i should actually be able to do about 0.1mm slice heightsI read somewhere (cannot find the web page now) that bed leveling should be precise to 1/5 of the layer height used.
He uses a modified Airteripper too and somewhere else he mentions that it is not good for 3mm filament. So even from that we can assume 1.75 filament. No idea what material was used, since temperatures he is trying seem too hot for PLA and cold for ABS. Does J-Head V9 come with different nozzle sizes? Or is it only 0.5mm? .... I'll try to ask these things on his web directly.His temps could be a thermister not reading quite accurate, but also some ABS melts higher, some melts lower.
what was the opening diameter of your old nozzle, leslieann?
my tiny family of traffic cones to push through the silly carpet. :DThat's hilarious, it's a bummer you didn't print them with a smiley face. :))
anyway, i'm noticing that i'm getting the most inertial error on small y movements. the makerbot slicer's gcode generator doesn't slow down for these high frequency small distance back-and-forth y-movements and it's definitely causing some error -- you can see the whole 80lbs shaking when it does this. interestingly, the y-axis is driven by two belts due to some odd motor placement. thinking..
that looks like your cold end is too cold or your platform is not heated..I doubt it is that. I think the temperatures are already stable when I start printing skirt. My guess would be that when printer is not running for a longer time them some gas bubbles get into extruder (somehow) and till they are expelled I get the blobs. Always after a while the extruded filament is nice and even.
I do not know. I used only ABS and the filament is in the rage 1.68-1.70 all the time but mostly 1.70mm.
The only thing which is not OK with it is that when starting a print the extrusion is "bloby". See change in the extruded material amount when printing a skirt. It has 4 strings and the outer one (the top) has clearly visible blobs about every 4 mm. This typically goes away when a skirt is finished, sometimes it lasts even during a part of the first layer. Sometimes the filament connecting the blobs is extremely thin even to the extend that it disappears and I only get a sequence of about 1mm long blobs :rolleyes:
Maybe it is not a filament error.
that looks like your cold end is too cold or your platform is not heated..I doubt it is that. I think the temperatures are already stable when I start printing skirt. My guess would be that when printer is not running for a longer time them some gas bubbles get into extruder (somehow) and till they are expelled I get the blobs. Always after a while the extruded filament is nice and even.
Btw. I thought for a long time that Rostock design is the most easy to bed-level, but with the right software support. Looks like the support is arriving now. I have even written down the equations for automatic bed leveling myself but I did not get to the implementation of the HW and the testing. Anyway I do not need to do it now since it is already done:
At least incorrect tower positions, diagonal rod length, incorrect top endstops, and z-axis length should be easy to compensate for (so one would not need laser cut top and bottom plate). No idea yet what kind of errors are compensated in the video.
and leslieann, re: the tolerances on your filament, not to say i told you so but... ;)One of the bigest complaints I have heard with cheap filament is size., so I did check it when I got it and before buying the PLA. The first 10% of the ABS was decent, I wouldn't have said it was great, but it served it's purpose, it was only as I got further into the roll that things went south. Had I seen it this bad, I wouldn't have ordered the the PLA. It too seems fine many meters in, but it's not like you can easily unroll a pound of filament and check.
we should identify the _good_ filament manufacturers here. frankly, the MBI stuff is obviously being made in china now, so we can isolate the good extruders at factories in the same way that we isolate the good keycap molder factories over there... after all, a filament extruder is basically the front end of an injection molder with a tool at the end that extrudes long strands of filament in a very controlled way (rather than a big old mold under pressure..)
I'm going to contact the supplier tonight about the filament and see what they say. The seller claims an oval shape is normal due to how they roll it, however, mine is round, it's just HUGE. When half the roll is 1.98 and 1.99 round, of course it has issues in a 2mm ID bowden tube. The seller and manufacturer also claims .1mm accuracy.. riiight. I know it's cheap filament, but I would like to at least get most than a quarter pound out of the 4 I have. Luckily, I'll be near the seller Friday and may just stop in and show them.When a filament is not round or when diameter changes over the length of the filament then the filament should be thrown to a garbage bin and the supplier should not be used again.
hey guys, so does anyone have Cherry keycap models? I really dig the ALPS adapters, geniushttp://www.thingiverse.com/thing:79673
Well I got my kit yesterday... Got everything assembled except the extruder because I'm missing a few bearings. Hopefully by the end of the week I'll be printing.
topre keychains!! :-X
people 3d printers owned will print for you?
-------------------------------------------------------------------------------------
leslieann rostock delta (from kit) no
When a filament is not round or when diameter changes over the length of the filament then the filament should be thrown to a garbage bin and the supplier should not be used again.I agree, but he also needs to know he is selling utter garbage.
Even 0.1 mm accuracy for 1.75 mm filament can mean that the volume pushed through nozzle is almost 12% more (or almost 12% less) than the slicer expects. The error seems high to me.
people 3d printers owned will print for you?
-------------------------------------------------------------------------------------
leslieann rostock delta (from kit) no
This information is incorrect.
Mine is not a kit, mine is built from scratch.
Also, I have no problem printing for others, once I have everything worked out.
oh, good point. could you put one dense post together that summarizes your build? plans, BOM, some brief description of anything that needs to be fabbed, any specific issues you had early on that we haven't been beating to death for the last n pages... ;)Yeah, I can do that, and yes, the Rostock is exactly like that... I get things like this:
going on plans and sourcing your own parts is something that should be discussed more. a lot of the reprap plans are more like loose recipes (ie, get _some_ kind of stepper motor with at least these specs...) that provide room for a lot of budgeting etc... if the rostock is like that, a more detailed post on your build would be very education.
I agree and disagree with you about the .1 accuracy. On one hand, yeah, it's a hassle, on the other hand, how much are you willing to pay to get higher accuracy? If it's not effecting the print, it's not worth a massive price increase.You are right, there is no reason to pay more when it is not needed.
The rostock here is the old one as defined in the wiki: http://reprap.org/wiki/RostockThat's the same Rostock I have, just done in parts by me, no kit. Yeah, the instructions stink. LOL My next will be a Cerberus or oversize Kossel.
I do not own it. It was from a kit. The kit had few bad parts: too loose filament guide on airtripper (fixed by inserting a piece of bowden in it), defect on the heatbed copper traces (fixed by soldering), universal joint holes should have been smaller, obsolete and unfinished build instructions.
It prints ok at low speeds (at most 50 mm/s), 0.3 mm layer height, external perimeters at about 10-20 mm/s.
A smaller change you could make: The larger the wheels, the less influence you will get from flex in the drive belts. You sort of gear it up, then add the tolerance and then gear it back down again, also gearing down and lowering the influence from the added tolerances in the middle of this transmission.Larger pulleys, while they do have less belt influence, make for less positioning accuracy. Though 200 steps per mm is probably a bit on the excessive side. :))
when you get your novel up i'll link to it leslieann :))First version is up (3500 words :-X)
finally got the bed as level as it will get with an indicator gauge, but i forgot to check head to nozzle clearance (or offset that appropriately) FACEPALM.
the result: (Attachment Link)
oops.
finally got the bed as level as it will get with an indicator gauge, but i forgot to check head to nozzle clearance (or offset that appropriately) FACEPALM.
the result: (Attachment Link)
oops.
finally got the bed as level as it will get with an indicator gauge, but i forgot to check head to nozzle clearance (or offset that appropriately) FACEPALM.I assume that's (only) the Kapton tape... Ouch.
the result:
oops.
Could you guys 3D print another 3D printer?
Nice, and my filament went from good stable 1.7 mm to 1.61 mm, I guess at the end of the spool it will be like a sewing thread :) It is probably a good idea to check filament diameter each day before printing.I usually only go to about 95c, with hairspray on green window glass.
Also I tried the latest Marlin firmware from https://github.com/jcrocholl/Marlin.git (2013-06-11 22:33:58) and it does not work well. Movement is jerky around the center of the heatbed.
Uff, and I found that it is hard to make the first layer stick if its layer height it lower than 0.2 mm. I read many times things like: "Make first layer thin to make it stick." It looks like it may be true but there is a lower limit. This was ABS @ 230°C on kapton tape on glass @ 110°C.
Did any of you try sandblasted glass?
Ridiculously thin (and also slow) is 0.025 :) http://richrap.blogspot.sk/2012/01/slic3r-is-nicer-part-3-how-low-can-you.html
have been avoiding glass (yes even pyrex) for its horrible heat conductivity.So long as I don't have a fan blowing on it, my bed is up to 95c in about 7or8 minutes.
at 230C i'm getting blobs of PLA stick in my head. my _theory_ is that these blocks have gotten way too hot and depolymerized, because they get wildly stuck in the head -- they stop flowing no matter how much power the heatercore pumps into them.230c for PLA is way too hot and will clog your head for sure, that was part of the problem I was having.
heat stress of doing that would destroy my entire head quite quickly. makerbot r2x nozzles are not replaceable.at 230C i'm getting blobs of PLA stick in my head. my _theory_ is that these blocks have gotten way too hot and depolymerized, because they get wildly stuck in the head -- they stop flowing no matter how much power the heatercore pumps into them.230c for PLA is way too hot and will clog your head for sure, that was part of the problem I was having.
One thing I later found was let the head cool to room temp, then push some ABS through (at ABS temps) and it would clear it out
Also, solvent will eventually clean things out, but leaves residue, meaning they will clog again soon enough probably.
This (http://www.youtube.com/watch?v=6bTfl35zlHE)is my preferred method for cleaning them out. So far, I have yet to need to clean my J-head, but I cleaned my old head all the time due to issues. Beware the alcohol fumes, they can get you drunk or sick quite fast. It can also catch fire.
heat stress of doing that would destroy my entire head quite quickly. makerbot r2x nozzles are not replaceable.
also the r2x firmware happily told me to extrude PLA at 230C :facepalm:
i only have a couple spools of PLA an honestly the r2x handles abs so beautifully that i just stuck them in a bag until i get further into design of the Geekhack True Polar (tm). up early so i'm going to build up my arduino uno motor shield and rotate some leadscrews :D
Ahhh yes, I forgot about that, which is why so many replace it with a J-head.
Mine does ABS quite well as well, other than fumes, I have no issues with it.
ahahahaa this cracked me up so hard i couldn't read the rest of the postheat stress of doing that would destroy my entire head quite quickly. makerbot r2x nozzles are not replaceable.
also the r2x firmware happily told me to extrude PLA at 230C :facepalm:
i only have a couple spools of PLA an honestly the r2x handles abs so beautifully that i just stuck them in a bag until i get further into design of the Geekhack True Polar (tm). up early so i'm going to build up my arduino uno motor shield and rotate some leadscrews :D
Ahhh yes, I forgot about that, which is why so many replace it with a J-head.
Mine does ABS quite well as well, other than fumes, I have no issues with it.
Update on my printer:
Switching to a geared extruder was the best thing I could have done. Good filament or bad, it will push it. It really made life easier... Except during filament changes. Oh, and I will never buy loose 1lb filament again, what a hassle to deal with.
My biggest problem now, is the perfectionist in me. I redesigned the carriers for mine, and I cannot count how many revisions I've done, but I have over a dozen prints of the part (at an hour each) sitting in the scrap bin. Granted it's only the third part I have designed, but between finding what works and what doesn't, accounting for shrinkage, and just changing how I want certain things has lead to many, many changes. Last night when i finally got what I think was finally it, Slic3r decided it no longer wanted to properly slice it. Cura, Kisslicer and Skeinforge want to run my extruder in reverse,
flip it in software not hardwareIf I could find where, I would.
It seems most slicer programs run the extruder opposite to Slic3r, the only thing I can figure is that the Repeteir firmware may be confusing them, I'm not sure.It cannot depend on firmware. Slicer can generate G-code without printer even connected. It cannot query firmware. It must be because of slicer configuration. If you use absolute coordinates for extruder then you can easily check the G-code. Look whether the E Values are increasing when progressing further in the G-code file.
It cannot depend on firmware. Slicer can generate G-code without printer even connected. It cannot query firmware. It must be because of slicer configuration. If you use absolute coordinates for extruder then you can easily check the G-code. Look whether the E Values are increasing when progressing further in the G-code file.Actually, you hit the mail on the head, or darn close.
quite weird. this doesn't even seem like it should exist. what does an absolute extruder coordinate represent? position on the great strand of life?
quite weird. this doesn't even seem like it should exist. what does an absolute extruder coordinate represent? position on the great strand of life?I read that at the beginning they did not know what is better. Absolute or relative. So they implemented both options in firmware. Slicers added an option for this to support firmware. Allegedly, almost everybody uses absolute coordinates nowadays and support for relative ones is obsolete.
actually i just had to swap the filament spool on the fly. i was just letting it run and realized it was down to the last few feet of filament on the spool. good thing, because i'm pretty sure it just air prints if you run outYes, it will most certainly air print. More than a few people have made filament sensors to tell them when it runs out.
I read that at the beginning they did not know what is better. Absolute or relative. So they implemented both options in firmware. Slicers added an option for this to support firmware. Allegedly, almost everybody uses absolute coordinates nowadays and support for relative ones is obsolete.This runs counter to what I just ran into and what I read in the reprap.org wiki (http://reprap.org/wiki/G-code#M82:_set_extruder_to_absolute_mode).
Question (Serious): What do you guys use your 3D printers for? Because whenever I see people showing off theirs, they just generally print random crap they don't need. Just for the fact they say they have a 3D printer.I have seen similar, however, I know at least two who use them to generate income. One is planning on leasing out his latest to local businesses.
LOL, you made me to search for my source: http://www.renosis.net/Skeinforge-41-guide.htmlI read that at the beginning they did not know what is better. Absolute or relative. So they implemented both options in firmware. Slicers added an option for this to support firmware. Allegedly, almost everybody uses absolute coordinates nowadays and support for relative ones is obsolete.This runs counter to what I just ran into and what I read in the reprap.org wiki (http://reprap.org/wiki/G-code#M82:_set_extruder_to_absolute_mode).
Cura and Skeinforge both use relative by default, and the page I read made a specific note that Repetier/Slic3r was odd is using absolute by default. I had to switch to relative in order to get those others working.
I am told by Tonokip (a firmware author) that the reason the absolute and relative options are there, is, at the time of implementation, they were weighing the advantages and disadvantages of each mode. So both are options are there, however, pretty much everyone uses Absolute mode now.I guess that means it is not really settled what is going to be used more.
is it possible to use it for printing keycaps?
I'm just going to float this out there. Today I learned that Plastic Splinters Suck. I was getting trained on the 3D printer and while handling the models, I managed to get two plastic splinters.
first aid protip: before you go digging into your skin with a knife to get a splinter out (or just to cut yourself -- hey, it's not my business) get a HOT lighter, butane preferably, and light the blade up. it will sterilize it. the lighter will leave some carbon on the blade edge. just leave that there. trying to get it off will contaminate the blade.
first aid protip: before you go digging into your skin with a knife to get a splinter out (or just to cut yourself -- hey, it's not my business) get a HOT lighter, butane preferably, and light the blade up. it will sterilize it. the lighter will leave some carbon on the blade edge. just leave that there. trying to get it off will contaminate the blade.
I hope hand sanitizer and hand soap work just as good as fire 'cause that's what I had.they don't, you're screwed.
Possible: Yes. Suitable: No
The FDM printers of today will have troubles with getting a good enough stem mount as well as create a nice surface. If you treat the parts after printing, for instance with acetone or sand them manually, then you can get a nice outer surface. I recon the stem fitting will remain a bit troublesome though, but probably possible. SLS and SLA are far better suited for printing parts with small details such as key caps.
I am looking forward to seeing some interesting cases and all kinds of projects come to fruiting because of these though.
For a "living" thread this is pretty dead.
cheggit this converter I made to make MX ALPS compatible
It works ok, I had to print it at like 90% the model size and SUPER SUPER SLOW, but it kinda works after a little post processing.Show Image(http://i.imgur.com/d8WEe3E.png)
For a "living" thread this is pretty dead.Only diamonds are for ever ... and not even those ... I guess :)
For a "living" thread this is pretty dead.
cheggit this converter I made to make MX ALPS compatible
It works ok, I had to print it at like 90% the model size and SUPER SUPER SLOW, but it kinda works after a little post processing.Show Image(http://i.imgur.com/d8WEe3E.png)
I'm sure you've seen the thread over at DT but in case not similar project, although not currently active
http://deskthority.net/workshop-f7/cherry-to-alps-adapters-t4934.html
Quite smart indeed, perhaps try Mx to topre, I believe there would be an elevation issue between a design like that? (Keycaps raised a bit higher)?
I made some converters along with mrinterface, but they were found to be too tall. It remains an option if we ever make a custom alps board, where we could make a deeper case which would enable us to use converters and MX caps. I am away from home, so can't really send it atm though, but it looks very similar to yours kmiller8.
Hi folks,
Relatively new rep2x owner here. I received my r2x in June-ish, and have been racking up the build hours since then. I swear I have had more fun with this in the last month than I had in the first 3 years of grad school (wait, those were terrible by any stretch.. sorry, I'll try to think of a better analogy as this review 'evolves').
So far, this review is unfortunately universally negative. Bottleworks delivered quite quickly considering these are parts machined, presumably, in the US, and I'm well aware of how expensive machine time in the US is. However, what I received were a) very roughly machined aluminum parts. In particular, the build plate base machining was clearly optimized for speed above all else. the arms were similar. there was no final fine finishing cut on any part and many parts have larger nicks and irregularities that show a general lack of respect for detail. I understand that the price of these parts is very low, but unfinished aluminum with very rough cuts is kind of the worst of all worlds. aluminum oxidizes in weird ways if you don't finish it, and with the HBP in particular, one needs to be quite careful about the flatness tolerances, as the use of an aluminum base plate means that you will have to deal with thermal expension, even if the very top layer of the build plate is borosilicate. for about the same price (when they have stock, which is another matter entirely), MBI's gravity cast alu plates are at least flattened to 0.3mm, and their injection molded ABS arms are precisely sized, even if they warp out into nowhere land after about 50 hours of build time. (they are cheap, fwiw!)
3 more points that i feel are necessary to warn people about before i end this bit of the review (but i will be installing at least some of the parts, so there will be updates! hooray!)
1) the bearings that are included at 10mm ID, 19mm OD LM10UU linear ball bearings with printed retaining rings. this would be all well and good if the r2x (and possibly the r2? i have no idea) used steel Z-stage support rods. however, they actually use 10mm ceramic coated alum aka feather shafts. everything i've read about feather shafts recommends using frelon sleeve bearings and not ball bearings. i have seen this many many times from every vendor of these rods. note that MBI uses sintered bronze bearings. while not frelon, they are definitely softer than steel balls. my understanding is that the deal here is that while the ceramic coating lowered coefficient of friction, it is not quite as hard as a steel rod (which are typically rockwell C60 or so), and if the 10-20-micron RMS ceramic coating wears off, all you have left is some pretty soft aluminum. in short, you're going to have a bad day. now, because he specifically cut the arms for LM10UUs, you can't use the MBI bearings, unlike carl's alucarrier. the MBI bearings are 15mm OD, which is a more common bearing size for 10mm IDs. to be fair, bottleworks' product description specifically states this, but it still took me by surprise when I took everything apart and started taking a closer look at all the MBI and bottleworks parts.
that said, these arms and retaining rings should fit a number of other vendors' frelon sleeves quite well. i've ordered what's in stock, but may just wait on lead time if i find a particular nice set that needs to actually be manufactured. fortunately, it seems that MOQs are pretty low in the bearing industry. in particular, SPD and QBC both have highly compatible parts in their catalogs.
2) i pretty firmly believe that the heater wattage on this kit should be upped. i tossed just the aluminum base with heater and unidentified insulation (corners are sealed with what looks like RTV silicone, mostly a black fiberglass looking weave). just measuring surface temperature on the alu base with an unrestricted 130w into the unit, it had a significantly hump at about 67C, and it was clearly struggling to get hotter than that. stick a borosilicate plate on top with a thermal conductivity about 2 orders of magnitude less than aluminum and you're going to have a bad day. also, despite the heater basically covering the bottom surface, i got an odd 5-ish C gradient from the center of the unit to the corners. mcmaster carries a silicone backed 270w 24v 6x9 unit that i may end up swapping in for this
3) THIS IS PROBABLY THE MOST IMPORTANT POINT:
i have gotten zero post-sales support from bottleworks. worse, my questions were at first answered with highly defensive appeals to volume ("i have hundreds of kits out there <snip several paragraphs in which questions are not answered>|") then when i repeated the questions i got ("i'm not going to argue about this" when i didn't recall arguing about anything at all). finally, his ultimatum was that i must send the kit back to him and if they were sent back in the state they were sent, he would refund me because i was a 'hypochrondriac' of some kind due to my HBP measurements. i simply stated the conditions that the HBP base were tested under, and i got back, verbatim "ok kid". in short, not only should you expect no post-sales support, but if you do contact him, prepare to be berated and insulted at frighteningly short cycle times. in fact, i'm still receiving emails from him. his last email could easily be seen as a threat of some kind, as it involves him mentioning personal information of mine out of context and with no other content in the email. thankfully, i am an internationally published researcher, run geekhack.org, a 23k user webforum, and generally have zero fear of doxing, but there you go. if you order, don't email him after you get your parts. you're going to have a bad day.
so that's it for now. i will be using what parts i can salvage from the kit, hand finishing parts, etc. i had to disassemble my bot anyway to replace both heads (i cut one of my initial plate levelings too close early on scraping them pretty badly, and the heads were growing increasingly out of tolerance as i printed the crap out of them :D), and my initial horizontal Z gantry plate, which now says "makerbot" on the front (it's really really warped), so except for the vague threats from this vendor, i'm actually a pretty happy clam. hell, even his insults are better than grad school. also, to be fair, i leveled one shot at B about me being an actual engineer. it was in the heat of the moment, and i apologize to the fellow.
that is a review of "bottleworks makerbot accessories", which is a company i won't bother linking here, NOT makerbot, who has been absolutely wonderful.Ahh, okay, I wondered why things weren't adding up.
i like this part of that /. post best: "The emission rates were similar to those measured in previous studies of several other devices and indoor activities, including cooking on a gas or electric stove, burning scented candles, operating laser printers, or even burning a cigarette."
lol.
anyway, if i'm reading this right, there's not much we can really do about the UFPs but fully ventilate the area, or use a fairly effective fume hood that ejects to atmosphere. hepas don't go down to nm scale, and my cheap charcoal is basically useless unless the stuff is much larger and non-inert. your dryer vent idea is probably the best, leslieann. i'll probably just open the window more often during long prints. i have nowhere obvious to route air to without opening a window, so i might as well just open the window more often and point a box fan at it or something
alas, this is also morbidly funny because i was just arguing with someone about the ABS smell, and my position was that it couldn't possibly be vaporizing. d'oh!
hmmm... i didn't try the ptfe lube with the PLA because i suspected the problem was that it was depolymerizing (which of course it would if it did get stuck to the heatercore). i will have to try this. if you still have mineral oil in your heads though, you wash them out with solvent (iso alcohol is a pretty safe one), as i think mineral oil doesn't react nicely with the brass nozzles. a little ptfe lube should be fine though, and the worst effect it can have is slightly lower the temperatures in-head when printing ABS, but ABS tends to flow extremely well, so i suspect that won't be an issue.
if you read elsewhere, the PTFE oil we like best here (no, in fact, WE WORSHIP IT) is dupont's krytox basic lubes, 102, 103, etc. the base is a polymer oil, which separates it from many ptfe dry lubes (which are very light petro oils with ptfe nanoscale particles). and it can be combined with the krytox line of greases to get like literally any viscosity. also the friction coefficient, it is so low. SO LOW
edit: i'm mentioning this because NOW, when my machine is down, i'm getting a bunch of printing requests. bug me again next week!
the MBI engineers basically knew they couldn't build a flat, stable (ie, non-moving) z-stage with their budget. hence, they had two goals: first, when you can't make something solid, you try to control the oscillations, so they went with a suspension system and tried to dampen its frequency with every parts and material choice. second, they obviously thought hard about the temperature cycling, and wanted to protect this suspension system as much as possible from exposure to heat. hence, gravity cast aluminum for the HBP; it's cheap, and easy to make flat since it's so soft. personally, i've lapped mine down (the last time i was less dumb and did it near op temp), 2-3 times. the heater for thin aluminum is also pretty cheap and can be low wattage, even covering a large build plate. then, they tossed two layers of a fiberglass/aerogel like substance (but again, a cheap one) under the heater and threw a thin alum sheet in with the leveling nuts. on the bottom side of that, they tossed some pretty hard springs (note: measure K). and nuts for leveling the plate. some brilliant dude even made their business card a feeler gauge for optimal nozzle to plate clearance. HAH!
now, z gantry. in the center is a long buttressed plate with some reasonably hard resin. i'm guessing it's about gf20 and PPO or PO. this has a high temp resistance, plastic transition way above the heater temp, and high tensile strength due to the glass fiber. basically, its young's modulus is high, and it shouldn't yield. unfortunately, it seems to over time. my gantry plate with several hundred hours of build time (at least?) is very much convex. so, final question: why the abs arms? i think it's actually because they were trying to lower the oscillation frequency with the arms and the feather shafts. the feather shafts have a lower modulus of elasticity than steel shafts do, as do abs arms vs PO + gf20 arms. basically, they wanted the gantry plate to stay as good a leveling reference as possible (ie, they wanted it to deform as little as possible) so they made the arms and the shafts the path of least resistance for oscillation so that the kinetic energy of the plate bouncing around is dissipated as far away from the plate as possible. if you think about it, the alternatives are a) the frame. this is basically all bad. b) the gantry plate. this is exactly what you don't want.
what i think i am going to try to do with my bot is actually to transfer that oscillation to the frame and then to dampen is as much as possible in the frame. there are two ways to do this. first is to make it heavy as sh*t. i've done a bit of that already, in that my bot sits on 40 pounds of acrylic and under another 5lbs. the second thing to do is to plate the sides with material that adds weight but also dampens oscillation. one upside to this is that the one can also improve on the heat insulation goals of the side plates as well. i have some stuff i'm going to play around with for this, since steel is cheap, and heavy, and if it doesn't work out i can always make other crap with it.
as for the HBP flatness issues. it's a bit of a myth that glass is naturally flat (i can't tell you how many times i've seen this on the various printing sites and it bugs me to no end...), but i'll lap down some borosilicate -- might as well start with the bottleworks stuff, and see how well it works out.
for small parts, i'm just going to use high alumina ceramic. it's not that expensive in small pieces, it's ridiculously hard (it's basically sapphire powder in binder, compressed then fired), and it has metal-like thermal conductivity properties if porosity is controlled while having basically 0 thermal expansion.
for large parts, i have no idea. lots of power applied directly to thick surface ground silicate maybe? silicate is at least cheap... it's hard to really say what's best from a hobbyist perspective, which means there's plenty of room for experimentation :D
i expanded on the suspension system comment on the makerbot users group. i think this is actually what's going on there:
ah, that i didn't pick up on. care to guess what alloy? something like t316 24ga would make a lot of sense. it's quite inexpensive to produce and has extraordinarily high yield strength without much weight.
this will let me (just barely) build a GH60 case flat on the bottom, and a full size TK case on end.
this will let me (just barely) build a GH60 case flat on the bottom, and a full size TK case on end.
Could you explain how you will print the tkl vertically? Just curious how your machine manages sudden horizontal surfaces without anything underneath.
Question: I have had alot of stuff printed with nylon, but a new company im trying to work with is offering ABS is that the same, different? Stronger etc?Nylon is far more flexible and less brittle than ABS. ABS comes in more colors. Your keyboard case is ABS as are most doubleshot keycaps. Nylon is most commonly used in rope.
Ive honestly no idea, thank y'all
ptfe melts at 260 (that is, MELTS, not enter glass transition), which is why you don't regularly want to be getting PTFE that hot and nylon is NOT for printing in these printers.E3d and Trinity heads can handle it, there are a few other all metal ends as well, they were designed to be capable of it.
you're probably talking about the lower temp taulman with the process temp of 245. there is another taulman nylon with a process temp of 260. please please please don't try to print that stuff with a head that has ptfe in it!Yes, the 240-245c... (depends who gives the number)
you automatic belt tensioners now? or are they manual? automatic belt tensioners rock. so. hard.Not automatic, but I have been thinking about it.
where are these MX to alps adapters that i am supposed to be churning out?
it's just too big for my bounding box. i've done some pieces like this at say a 45 degree angle, but the printer ends up spending 8 hours building scaffolding and then 7.5 hours in it turns out that one small detail piece (like the fastener holes on his design) breaks off due to a small error that would have to be designed out and the whole print is a bust.How big are they?
if i had the source files with splines i could cut the design into printable pieces, but only the steps and stls are provided on the ergodox page. i've pinged dox to see if he has time to cut up the design himself (it's very much conducive to it) and beef up certain structural parts (since the price per cm3 of my prints are at least an order of magnitude less than shapeways), so we'll see.
it's just too big for my bounding box. i've done some pieces like this at say a 45 degree angle, but the printer ends up spending 8 hours building scaffolding and then 7.5 hours in it turns out that one small detail piece (like the fastener holes on his design) breaks off due to a small error that would have to be designed out and the whole print is a bust.How big are they?
If they can fit into a 400mm diameter, I should have you covered shortly. :thumb:
thinking of picking up the makerbot scanner so that people can sculpt and send then have molds printed. thoughts?
where are these MX to alps adapters that i am supposed to be churning out?
you scan a part using their camera + dual laser system and it provides you with an stl. you can extrapolate from that STL to make a solid or you can print it directly using their slicer. if you were extrapolate out the STL into a solid with splines you could do an inversion to make a mold in CAD software.
the adapters are just not that hard to model. i say someone take the hour that's needed and just draws them off of the datasheets in solidworks, annotating the model with the datasheet tolerances.
I could otherwise recommend sculpting in software directly. Zbrush and Blender both have some tools for it. They have made the process quite similar to sculpting irl.
I could otherwise recommend sculpting in software directly. Zbrush and Blender both have some tools for it. They have made the process quite similar to sculpting irl.
I was actually just thinking of this the other day. Is there a "known good" model of a basic MX keycap that one could use as a starting point for such modeling? I don't want to have to worry about tight/loose fitting stems, etc.
Thanks in advance!
i'm very hesitant until we can come up with a really good use case for it now. we need a small mill way more than we need a scanner
shapeways scales at the size of eg a single cap, but for larger objects like complex chassis shapes, you're using way more material than they seem to want you to use.Shapeways warns not to go under 3mm for wall thickness in structural parts, even in metal.
that is really tight.That is just a rough quick layout.
are you sure it wouldn't be better to print it vertically with some supports?
Leslieann, do you have different auto-levelling than the one created by Johann?i think it's a useful advance. sure, it's not perfect, but that's the whole point of mentioning that larger beds give you a higher probability of hitting flatness tolerance issues.
Because Johann's auto-levelling does not correct for the curved print-bed as Damorgue proposed.
Johann's code curves everything exactly the same way as the bed is curved. Actually this is an important feature of Johann's code because it allows to correct for slightly incorrect tower positions and diagonal rod length. This is a great feature since one does not need to have exact rods and top/bottom plates and prints will still work well enough and there will not be issues with first layer not sticking. But incorrect tower positions and rod length specification in firmware lead also to second order errors in layer flatness. That means printer does not know what is flat i.e. you cannot assume head is moving in a flat level. This second order error is only linearly approximated at 7 points in each axe direction. So you do not want tower positions / rod length too wrong so that the approximation works well enough (and so that the probe deploy/retract can position head well enough).
I know only 3 approaches to delta bed levelling:
1) Johann's - this is what most people know about.
2) Math based one. It is currently not completely automated but there is some proof of its feasibility. Here is more information about it: http://forums.reprap.org/read.php?178,237655,237789#msg-237789
3) Manual one i.e. the tedious one. Most delta calibration descriptions do not bother with incorrect tower positions but the thread above has some hints here: http://forums.reprap.org/read.php?178,237655,237789#msg-237789
Mkawa will not like Johann's approach but I think it is good enough. Thick mirror is flat enough if you mount it only in 3 points (so that you do not expose it to forces which can bent it). Really if your keyboard case is only as bent as a mirror is then it is a already a win. Especially considering how easy to use is Johann's bed levelling. And mirror does not get curved more after few hundreds of uses. My experience is that once delta gets levelled it stays that way (if you are not replacing printer parts).
I still plan on setting rod lengths and such to achieve flat, to me, that is part of the process of tuning a delta.Good thinking.
oh crap, fell asleep during this post too. anyway, it's all complicated.
Hi all 3d printer. I just saw https://www.massdrop.com/buy/magic-plastic-pellets on massdrop (it's some plastic pellets you can melt and shape to make anything).this is also called instamorph and has a bunch of other trade names. i have a big bottle that i used to make dong-shaped objects (just kidding, that was just a shout out to the d-squad and kirkle!). it's fun stuff. you just dump a bunch into boiling water, pull it out and treat it like clay until it cools down.
Anyone know if they are good? Maybe it could be used to make keycap stuff like Binge handmade marvels?
Waiting on Kawa's explanation of all of this plastic properties :rolleyes:
Hi all 3d printer. I just saw https://www.massdrop.com/buy/magic-plastic-pellets on massdrop (it's some plastic pellets you can melt and shape to make anything).this is also called instamorph and has a bunch of other trade names. i have a big bottle that i used to make dong-shaped objects (just kidding, that was just a shout out to the d-squad and kirkle!). it's fun stuff. you just dump a bunch into boiling water, pull it out and treat it like clay until it cools down.
Anyone know if they are good? Maybe it could be used to make keycap stuff like Binge handmade marvels?
Waiting on Kawa's explanation of all of this plastic properties :rolleyes:
see look! i don't always go into atomic detail! :D
also just buy it off amazon as instamorph. no need to gb it. it's so cheap!
i managed to brick a mightyboard. oops
Leslieann, did you use Berry tripper at the end?Berry Tripper is a Wades gear reduction combined with Airtripper, and yes, I'm using it. Other than sometimes having to "aim" the filament into the output hole, it works fabulous.
I'll need to change extruder too. So I'm looking for some bastard child of Airtripper and Wades/Greg.
Something which uses gears and hobbed bolt but has filament guide and clamp as airtripper.
My computation that the direct drive should be enough was wrong. I didn't know that when a stepper is using micro stepping then it's actual torque is only about 70% of the nominal rating.
i managed to brick a mightyboard. oopsDon't do that. ;D
Does anyone happen to have a 3ds max file for a cherry esc key?
Does anyone happen to have a 3ds max file for a cherry esc key?
Please?
Does anyone happen to have a 3ds max file for a cherry esc key?
Please?
Go geared, you won't regret it.
Does anyone happen to have a 3ds max file for a cherry esc key?
Please?
I am inclined to not release mine until I have finished all the rows and 1u;1.25u;1.5u and so forth and is absolutely certain that they are correct. What are you going to use it for?
Yes, I am planning to. Maybe I could push airtripper more but I do not want to add fans on the stepper driver and the motor. Berry tripper uses M5 bolt and a hobbed pulley as airtripper has. I have only a hobbed bolt at hand. I'll probably design something which will fit nicely in the corner of the traditional Rostock. I do not want to unnecessarily limit the built volume at the extruder location.I put the extruder outside the wall, and then later replaced the wall with a tube. I used printed brackets to hold a tube at each back corner, and clamps to hold the extruder and spool mount. It not only resulted in a much more open look, but the frame was more stable as well.
I put the extruder outside the wall, and then later replaced the wall with a tube. I used printed brackets to hold a tube at each back corner, and clamps to hold the extruder and spool mount. It not only resulted in a much more open look, but the frame was more stable as well.
that said, i don't think MBI's pricing is all that bad, and their filament is good and they have a good color selection.$42 for a pound of extruded plastic? No.
kg ma'am, not a pound.I did per pound, I cut the price in half (1Kg= about 2.2 pounds)
plastic is a complicated beast. we think we know what it is because it's so malleable and easy to touch, but the more i learn, the more i realize how much of a miracle it is.
You're welcome to process and extrude these ABS pellets you're finding on the spot market at those prices. I think you'll be surprised at what happens when you attempt to re-extrude the filament you produce (assuming you manage to produce usable filament at all :)) )I'll find out soon, I know someone with a Lyeman extruder for this very purpose.
i like your stinginess and gumption, but prepare for a fair amount of frustration :)I won't have to be first, I'll let the other guy figure it out first. LOL
that said, i don't think MBI's pricing is all that bad, and their filament is good and they have a good color selection.$42 for a pound of extruded plastic? No.
It doesn't cost that much, sorry, no way in hell, not going to #$#%^#$ do it, that is Lego level pricing for extruded plastic. Almost all of it's coming from just a few suppliers in China.
i can literally show you a dual extruded thing where the two colors have wildly different shrinkage, adhesion and every other mechanical property. ABS is not ABS is not ABS. there are as many formulations of ABS as there are particles in the universe (recall, this is the greek letter 'c'). they range from the truly ****, recycled-but-no-one-will-tell-you-that, to the direct from henkels or dupont or other certified master batch maker with extensive documentation on what it is made out of to measurement error to how it will behave when extruded (ONCE) within tolerances.I'm still not paying MBI prices.
cool new autodesk tool for manipulating stls and step files directly: http://http://blog.123dapp.com/2013/09/introducing-autodesk-meshmixer-awesome-3d-remixes-at-your-fingertips
STLs are output files. meshmixer is a postprocessor like makerware or repg. you should not be working with solid meshes. you want splines to retain precision. sure, solidworks can import stls, but you literally are not allowed to manipulate them. same with rhino, even maya. no design package speaks voxel meshes. not even game programmers work with polygonal or voxel meshes at the design phase anymore. (in fact, gaming engines do dynamic rendering for LOD purposes but that's neither here nor there..)STL's may be an output format, but it's standardized format that other 3d printer programs accept. It's a massive oversight in many peoples opinions.
you're free to pay any price you can find for filament, but be aware that a statement like "ABS costs X$ in y shape" is an inherently silly and wildly imprecise statement.
use <MCAD/involute_gears.scad>
gearAxeDist = 25;
tinyTeethCnt = 16;
bigTeethCnt = 32;
boreD = 5;
height = 8;
roughness = 0.1; //plastic surface roughness
gearPitch = (gearAxeDist-2*roughness)*360 / (tinyTeethCnt+bigTeethCnt);
echo("Gear Pitch [mm]: ", gearPitch/180*3.14159);
// tiny gear
translate([gearAxeDist+10,0,0])
gear(
number_of_teeth = tinyTeethCnt,
circular_pitch = gearPitch,
gear_thickness = height,
hub_thickness = height,
bore_diameter = boreD,
backslash = 2*roughness );
// big gear
gear(
number_of_teeth = bigTeethCnt,
circular_pitch = gearPitch,
gear_thickness = height,
hub_thickness = height,
bore_diameter = boreD,
backslash = 2*roughness );
The easiest and quickest is just swapping the gears in the extruder you use (Berry Tripper?).I actually use herringbone gears and found a generator for it, I just need to make them.
The idea is something like this:
I actually use herringbone gears and found a generator for it, I just need to make them.Right, that is a normal gear with non-zero twist argument + add its other half using mirror transformation. But no need for that if you already have a generator :)
It's not the movement that's an issue, the herringbones run quieter.Ok, I did not know that. It is a good reason to use them.
It's not just herringbone gears, it's why we use angular cut gears on everything we can.It's not the movement that's an issue, the herringbones run quieter.Ok, I did not know that. It is a good reason to use them.
straight cut gears are more durable. more surface area per unit diameterActually, angular cut gears have more surface area and tend to run more smooth as they are always engaged, leaving them less prone to shock damage and giving them a smoother operation. In general, they tend to be considered stronger overall. The two main selling points on straight cut gears are that they're easier to produce, and they don't produce side loads. Another benefit is that when shaped properly, straight cut will have lower friction, however, this tends to equate to more noise, particularly as speed increases. Friction and ease of production is why race cars use them, while street cars use angular whenever possible for noise reasons.
but my spatial reasoning is frankly ****. is there a really simple way for me to grock why arc cuts give you more surface area?Very over-simplified answer...
I'll stay with straight cut gears. I like the fact I can put a gear on without moving gear axe away first.I'll probably go to them at some point, it would certainly make assembly easier.
yep, that was my immediate thought, but it's a packing problem, and straight lines are easier to pack, no?but my spatial reasoning is frankly ****. is there a really simple way for me to grock why arc cuts give you more surface area?Very over-simplified answer...
The straightest distance between two points is a straight line, and a straight cut gear, is a straight line.I'll stay with straight cut gears. I like the fact I can put a gear on without moving gear axe away first.I'll probably go to them at some point, it would certainly make assembly easier.
I have RAMPS 1.3. It's wiki page claims 11A for heatbed (which is about right) and 5A for steppers and hotend (which is too small). Most motors (driven by the most common (green) pololus) can take about 1.3A without thermal shutdowns. So the minimum rating for motors and hotend should be 4*1.3 + 12/6.8 ≅ 7A.
... maybe it's possible to use a 24volt bed and leave the rest on 12volts.Yes, it is. Though, you need a bed which will not pull more than 11A at 24V if you increase the voltage. If you have 12V only PCB heatbed then you need to switch it. I have seen a dual voltage (12V/24V) PCB heatbed. I think (not sure) it is still pulling about 10A at 24V option. If so then you would get a heatbed which heats up considerably faster. I personally do not care. When I heat it up I cover it with a piece of cloth and it is quick enough for me.
like those buttons i printed for you. you're going to get a lot less elasticity out of the sls version of those...
the problem with long tall skinny parts, and really really small parts is collision. if a head collides with a printed bit of extrusion, bad stuff happens depending on how much support that part has w/rt stiction to the build plate (which is ultimately where all the forces into the part go while it's being printed).On tall thin stuff, I usually still a pad off to the side with a thin upright (1x or 2x nozzle diameter), this lends support and can be broken away relatively clean after printing compared to putting a pad and support on the thin wall.
Leslieann:Yeah, I have been reading up on it a bit more and saw that.
I just noticed that the 11A fuse for heatbed on my RAMPS 1.3 is rated for only 16V (the 5A fuse for steppers and hotend is rated 30V). Your RAMPS 1.4 is probably the same so if you would want to switch heatbed to a 24V version then you should change the fuse too.
with your bowden tubes.. have you guys tried using sacrificial walls and towers as a wiper mechanism?I have my retraction down pretty well, to where I don't have a need really.
you don't have issues with creep out of the hot end at all with the bowdens? naturally we have gravity bringing a bit of extra melt down out of the nozzle and need to wipe every once in a while.It may be needed with dual extruder when one head is heated and does not print anything for a longer time. I did not try to use it but I think the wipe column is almost useless for single extruder rostock. Moreover it can be harmfull since it adds retracts which (if frequent) deforms filament and results in wrong extrusion amounts. The other point is that non-printing moves are quick and head is not pressurized during them. Hardly anything gets out of the nozzle because of gravity in such a short time (typically below 0.5s). The bigger problem is retract and retract reverse times. Depending how quick your extruder stepper is, these can take about 0.2s. But the important difference is that the nozzle pressure is rising/falling during these times. The filament is pushed to the hotend with forces around 20N (for speeds about 2.5 cm/s, 185°C PLA). This value is about 2e5 times higher than the gravitational force for a bit of melted plastics in the nozzle. OK, 2e5 is the maximum at the time of retract start and retract reverse end only but still it is a huge number.
The consequence is that tiny blobs of material are deposited at the retract and retract reverse locations. Slicer can handle this easily if there is enough internal volume to hide them in. For very tiny parts there is not enough internal volume and it causes visible/measurable seem on e.g. thin walls. Thin pillars not only do not have the internal volume but also bent which is much worse problem actually. And if they are alone there is also problem with their cooling.
That is the reason I'm trying to get not only stronger extruder but also a very quick one. This also made me to go to 24V for the motors. It increased the usable extruder speed by about 60%. I actually expected speedup by 100% but it was not true. I do not know why I achieved only 60% :confused:
Anyway, I believe the problem with tiny blobs during retract and retract reverse are mostly fixable with better slicers. The slicer should have some model to estimate how flow will change with the falling/raising pressure and move the head along the planned extrusion path acordingly. If the bowden friction is predictable enough (I hope it is) and the fluid dynamics in the hotend is predictable enough (I think it is) then this should be doable.
Now you have me curious...Probably not. Depends on the speed you need for normal extrusion when printing. My guess is that your typical printing speed is 100 mm/s. With layer height of 0.3 mm and extrusion width of 0.6mm we get filament feed rate of (100*0.3*0.6) / (π*(1.75/2)˛) ≅ 7.5 mm/s. Typical filament driving pulleys have diameter 9 mm, and lets assume 200 steps per revolution, then we get stepping speed of 7.5/(π*9)*200 ≅ 53 steps per second. That is a too small number for driving voltage requirements. I have here a voltage/torque characteristics of a stepper (not a NEMA17). There is only about 3% difference between torques at 12V and 24V driving voltage up to the speed of about 70 steps per second.
I was interested in 24volts for my bed mostly, but now, I'm wondering if maybe 24v on the motor side would allow me to go back to a direct drive extruder. The EZ Struder is a fantastic extruder, it juts lacks the power for any nozzle below .5mm.
Cool. I'm thinking about building a wider (and shorter) delta printer. And stiffer. But I'm not in a hurry. Maybe in a year or so :)
Put a bit more clamping pressure or lower the amperage so the motor skips first. I think extruders are the biggest pain the neck on printers.Good point. I should decrease the maximum stepper current. Skipping is better than filament grinding. The clamp is already as tight as it can be.
My monster is up and running, it's leveled to about .02mm across a 500mm (20inch) build surface. Still chasing bugs, but it's pretty neat. A bit quieter than then Rostock, however, it's shocking how much noise comes from the print head.0.02 mm is very good. Did you use laser cut top and bottom plates?
Good point. I should decrease the maximum stepper current. Skipping is better than filament grinding. The clamp is already as tight as it can be.
Or replace the filament driving pulley. Are you aware of anything with a big diameter (at least 15 mm) and bigger teeth spacing (so that the filament grit does not stick between teeth)?
0.02 mm is very good. Did you use laser cut top and bottom plates?Depends on the print job and things, but on some printers you can hear the head rubbing ridges of the part you are printing. In my case, cheap pulleys were to blame, which I've since replaced. That significantly lowered the noise and made my layers much more uniform.
What do you mean by noise from "print head"?
Smaller pulleys are better for torque.My idea is to get bigger pulley so that there is more contact area with the filament. This should lead to grinding at higher forces. I can adjust torque easily by changing the gear ratio.
However, the extra long wires on the large printer makes my thermister WAY, WAY off. By the time I see 175c at the nozzle with the infrared thermometer, Repetier is showing 250c. Only off a little, LOL.This does not sound right. If you use 100kΩ thermistor then resistance difference between 175°C and 250°C should be about 700Ω. Even long wires should have resistance of about 2Ω at most. That is too far from 700Ω. My guess is you have a bad connection somewhere (broken wire, bad solder joint) or you do not use 100kΩ thermistor, or something else is broken or improperly measured.
This does not sound right. If you use 100kΩ thermistor then resistance difference between 175°C and 250°C should be about 700Ω. Even long wires should have resistance of about 2Ω at most. That is too far from 700Ω. My guess is you have a bad connection somewhere (broken wire, bad solder joint) or you do not use 100kΩ thermistor, or something else is broken or improperly measured.
IR thermometers measure total radiance emitted over a volumetric cone type thing and aren't useful for trying to measure point temps, but that's just one issue in that whole stack. there is also the thermistor. simple point thermistors are just not very accurate. RTD-style resistance thermometers are accurate, but expensive. for point measurement, you really should be using a thermocouple.
kmiller8: Did you figure out why it did go wrong in the upper part?
Thanks Leslieann. I looked at Cura generated gcode and it does not look like it does anything special with fan for the first layers on support. I'm tempting to think I could achieve even more with an option to make the print speed for the first layer on support really slow. It should behave like bridges. But I do not see such an option in Cura, nor KISSlicer. Slic3r has a bridge speed option but I do not know whether it would be used for the first layer on a support material.The lack of fan tweaks and such is probably at least some of the reason Cura slices so fast.
RFC about fan mounting.A duct shouldn't matter, the heater should be able to heat up plenty regardless.
Is it worth to add a fan duct to direct air more below the print head?
Or is fan without any air duct better?
Upgraded my Z-rods (smooth)my god, where did you get the rod on the left from? that's a bit of a disaster right thereShow Image(http://i.imgur.com/1trvJNg.jpg)
My friend printed Yoda the other day for me, the detail is amazing. :pwow, that's fantastic. do you have any info on what printed that?
I don't know a thing about 3D printing but I guess this belongs here.
(Attachment Link)
A duct shouldn't matter, the heater should be able to heat up plenty regardless.If I would had known I would not start designing some duct.
Many deltas just stick a small fan blowing across everything and have no issues, this makes for a lighter moving assembly. I tend to go back and forth between effector mounted and just a loose fan, however I have yet to find a duct worth bothering with.
My friend told me its a cheap Chinese copy of Makerbot Replicator, it is printed with ABS plastic with 0.01 mm layer.My friend printed Yoda the other day for me, the detail is amazing. :pwow, that's fantastic. do you have any info on what printed that?
I don't know a thing about 3D printing but I guess this belongs here.
(Attachment Link)
yes, i've seen these. they're sold as 'flashforge' and they've managed to second-shift the makerbot/ssys tooling to the point where their machines actually say makerbot all over them :P. they actually cost quite a bit considering, and use very well-made mbi revD controllers (from thingiverse) and all the other open source parts that MBI put up before they closed much of their design (and wasted tons of engineering time on posting images of your print to facebook :facepalm:).My friend told me its a cheap Chinese copy of Makerbot Replicator, it is printed with ABS plastic with 0.01 mm layer.My friend printed Yoda the other day for me, the detail is amazing. :pwow, that's fantastic. do you have any info on what printed that?
I don't know a thing about 3D printing but I guess this belongs here.
(Attachment Link)
I don't even know what a mbi revd controller is and none of this 3D printing stuff. Just wanted to share this little Yoda I got :pyes, i've seen these. they're sold as 'flashforge' and they've managed to second-shift the makerbot/ssys tooling to the point where their machines actually say makerbot all over them :P. they actually cost quite a bit considering, and use very well-made mbi revD controllers (from thingiverse) and all the other open source parts that MBI put up before they closed much of their design (and wasted tons of engineering time on posting images of your print to facebook :facepalm:).My friend told me its a cheap Chinese copy of Makerbot Replicator, it is printed with ABS plastic with 0.01 mm layer.My friend printed Yoda the other day for me, the detail is amazing. :pwow, that's fantastic. do you have any info on what printed that?
I don't know a thing about 3D printing but I guess this belongs here.
(Attachment Link)
i feel like one of two things is going to happen as the hobbyist machines progress. either we'll have to start programming our gcode by hand like all the big boys do, or we're going to have to get a lot smarter about gauge checks and JIT with parallel simulation. one of the reasons the silly expo was so depressing was that everyone is trying to build a cheaper and/or flashier printer. fail. the hobbyist folks have yet to produce a printer that actually, dependably, works. that is the problem that needs to be attacked..You've eclipsed what I would consider hobbyist at that point.
mkava: The part is already in two pieces. The one shown lately is the top part only. But I could divide the top part to the duct section and fan holder section as you proposed for even better result. Good point. I believe this would help much more than thicker walls.yep, a huge part. and it gets worse when, like the makerbots, your idler pulleys can move.
I have very good experience with connecting ABS parts with acetone too.
I also can confirm that part designer should try to avoid sharp corners. Especially then they should happen in the section of part which has low slopes. Sharp corners in the low slope area are unprintable with current slicers. Well if you re not willing to do everything very very slowly. I guess it could be printed if the corner itself is printed really very slowly.
I attached a picture of the failed bottom part of the duct which did have a sharp corner in the low slope area. Notice that the corner is collapsed. You can see the same part redesigned without the sharp corners in my old posts. After filleting the corners it turned out pretty well.
I think to get better results mathematical model of the FDM printer should be part of the slicers (for open loop) or the firmware (for closed loop). It is not like it is a new physics. It is just not a simple physics.
Btw I was asking somewhere in far history about belt young modulus. Here are some approximate data:That indicates that for machines with long belts (e.g. deltas) the belt stretching can be significant part of corner errors at high printing speeds. Depends on the stepper motor rotor inertia in comparison to the inertia of the rest of the printer moving parts.
- 6mm wide T2.5 belt with steel core - will stretch about 0.02 mm per 1m of length per 1N of force (from a rough measurement of a belt user)
- 6mm wide GT2 belt with glass core - will stretch about 0.06 mm per 1m of length per 1N of force (from a provider data sheet http://www.bbman.com/assets/files/pdf-library/Engineering/Timing%20Belts/BeltTensileProperties.pdf ); notice the data sheet has incorrectly specified unit for their belt modulus; they specified it as lb/in˛ but it should be only lb
That indicates that for machines with long belts (e.g. deltas) the belt stretching can be significant part of corner errors at high printing speeds. Depends on the stepper motor rotor inertia in comparison to the inertia of the rest of the printer moving parts.I see this on my large printer, I can't run the belts as tight, and with their length (2 meters each), and those long diagonals, it really can create issues. I have to dial down travel speeds, acceleration and deceleration quite a bit to keep it from causing problems, about 1/2 to 2/3rds the Rostock with it's 1.4m belts.
[re: leslieannI'm not saying the problem can't be solved, just not for the money most people are willing to spend on a printer. Particularly one with limited material.
3d printing is just motion control with a filament extruder. motion control is a mature problem and the practical solution that people have come up with over the last 30 years is hand programming. sad but true. i personally think that's garbage and that all motion control can be done better faster etc. with smarter compilers; like vvp is saying, the physical parameters of the printer need to be given to the runtime and/or compiler. further, i would insist that a _lot_ more open loop feedback needs to happen.
as far as whether this technology can is beyond the scope of hobbyists, i really don't think so. hobbyist affinity really depends on whether technology is packaged up and black-boxable. it is less about total complexity and more about exposed complexity
that's my 0.02USD anyway.
I see this on my large printer, I can't run the belts as tight, and with their length (2 meters each), and those long diagonals, it really can create issues. I have to dial down travel speeds, acceleration and deceleration quite a bit to keep it from causing problems, about 1/2 to 2/3rds the Rostock with it's 1.4m belts.Is it using GT2 belts you mentioned in the past? Do they have glass fibre core?
Regarding the belts: I think the Z axis accuracy is far more important when it comes to reliability and failed builds. X and Y will cause less exact parts, but inaccuracy in Z will cause delamination, pressing the head into the part and general failures to a larger degree imo. This is why I suggest that FDM printers should strive for a large build surface but limited Z axis. This of course also means that the rods have to be able to withstand the weight and not drop too much over its length but strengthening them is fairly easy.
Is it using GT2 belts you mentioned in the past? Do they have glass fibre core?
It looks like most GT2 belts sold use glass fibre core and most T2.5 belts use steel core.
GT2 belt producer claims GT2 should has more precise belt-pulley meshing compared to T2.5. And also GT2 (with glass core) should last longer too. I can agree with them about these too. They recommend it for precise positioning. That maybe be OK but only for short belts (if we are talking bout GT2 with glass core).
Glass core GT2 is not so good for deltas. From this point of view the reprap wiki page on belts is rather misleading by pushing GT2 heavily without even mentioning the problem of low modulus of glass core GT2.
Wolfram alpha claims these young's modulus values:
glass - 69GPa
kevlar - from 70 to 179 GPa (depends on type)
steel - 205 GPa
Looks like steel core is the way to go, maybe kevlar if the belts use the best one (DuPont Kevlar 149).
It would also help to use wider belts than the common 6 mm width.
Regarding the belts: I think the Z axis accuracy is far more important when it comes to reliability and failed builds. X and Y will cause less exact parts, but inaccuracy in Z will cause delamination, pressing the head into the part and general failures to a larger degree imo. This is why I suggest that FDM printers should strive for a large build surface but limited Z axis. This of course also means that the rods have to be able to withstand the weight and not drop too much over its length but strengthening them is fairly easy.yes, a huge number of the printer designs i've seen just use rods that are too small everywhere. this includes the makerbot. and then they don't spec them to high straightness tolerance or high strength material. and then rod diameter is generally difficult to change without redesigning everything, so i've had to go to exotic materials for pretty much everything in the bot. everything is either 440 or case hardened steel at this point, with straightness to < 1 mil and surface hardness at C65 or better. further, bearings obviously play a huge part in error components along the linear actuation, so i've had to go exotic on those too. everything is either extremely high precision (mostly misumi) ball bearing or rulon lined sleeves when there is more surface area and less duty cycle.
a huge number of the printer designs i've seen just use rods that are too small everywhere. this includes the makerbot.And the original Rostock. It uses Φ 8 mm rods and linear bearings. I'm surprised Johann did not go for Φ 12 mm rods. Based on a local supplier here the rods would be 35% more expensive and linear bearings would be 10% more expensive. Rod error specified as at most +0μm -8μm. Bearings specified as at most +8μm -0μm. It is the same for both 8 and 12 mm versions. 12 mm versions look worth it. Especially when some people put linear guides on their Kossels which are much more expensive than rods and linear bearings.
LOL, looks like our el-cheapo 3dPrinters are designed so they just barely work. And if one wants higher quality either he needs to print really slowly or just replace almost all the parts :)i can't believe you actually sat down and calculated to determine this. it's pretty obvious if you just look at it. a) well, it seems to print something b) what the hell? how could this possibly work?
Makes sense.
And the original Rostock. It uses Φ 8 mm rods and linear bearings. I'm surprised Johann did not go for Φ 12 mm rods. Based on a local supplier here the rods would be 35% more expensive and linear bearings would be 10% more expensive. Rod error specified as at most +0μm -8μm. Bearings specified as at most +8μm -0μm. It is the same for both 8 and 12 mm versions. 12 mm versions look worth it. Especially when some people put linear guides on their Kossels which are much more expensive than rods and linear bearings.
i can't believe you actually sat down and calculated to determine thisYes, the conclusion is fairly obvious :)
This was all part of why I did my own. As Mkawa found, trying to re-engineer something often just creates more and more problems. I tried scaling other designs and they just didn't work well, each part change required other parts to change._everything_ causes problems, as we all know. we have all had our share of incredible frustration. now that i have a more global view of the FDM printer market, i can laugh about it (even 60 grand printers are incredibly frustrating), but my response to this is to sit down and engineer stuff. after working in research for so long, i don't really expect anything to work, ever. ;)
I do not think that the model I used for rod bending is much good though. If I put higher forces in it then the results do not conform to experimental measurements.you cannot rely on material specs on the web for steel. if the part is using chinese steel formulations that goes doubly so. even if you have completely traceable material sourcing, the way that the part is formed and machined is just so stochastic. even the "expensive" rods i source are not aerospace or space part quality at all. and all space parts, aerospace parts and structural parts are tested per batch for compliance by the engineering firm to ensure that QC was performed and that the material will generally have the specified characteristics. we would be hard pressed to actually accurately determine young's modulus for anything we have on hand, so at some point we just have to put the parts in and see what happens.
OK, back to the belt stretching of the delta here (classical rostock with 6mm wide T2.5 belts with steel core). It is still true to the original except the diagonal rods were changed to carbon rods with MP-Jet ball joints.While I haven't looked for steel core, most Rostocks and Kossels are using fiberglass core GT2 belts (not steel or T2.5). I found a chart showing stretch, but I would have to dig it up. I looked at Kevlar, but it's expensive. T2.5, especially steel would be MUCH stronger than common GT2.
{cut}
Btw, the maximum usable acceleration of a carriage is about 8g. At 15g, the stepper would definitely skip steps instead of moving a carriage :-) At 8g, the side force on the smooth rod is about 15N and that can bent it about 0.8 mm. Just to get you some idea what the very worst situation can be.
_everything_ causes problems, as we all know. we have all had our share of incredible frustration. now that i have a more global view of the FDM printer market, i can laugh about it (even 60 grand printers are incredibly frustrating), but my response to this is to sit down and engineer stuff. after working in research for so long, i don't really expect anything to work, ever. ;)Tell me about it. LOL
so kids, that's why you give your nozzles a serious acetone bath every now and again, and then _get the resin that's sitting around clogging things_ out. chances are that you're going to have moments where that's a bit too much backpressure, or you're below or above process temp and this stuff will build up over time no matter what you do.
It's pretty obvious when your printer takes a second and a half to cross a 170mm print bed that you aren't hitting 500mm per second. Even if you account for some acceleration, it shouldn't take that long.500 mm/s is hight speed. If they use the common 1g acceleration then (even with 0 jerk) crossing 170 mm bed should take 0.34 s. That is really quick. Although 1g (10000 mm/s˛) or 0.9g (9000 mm/s˛) is the default acceleration in Marlin. It is rather high. E.g. it takes only 12.5 mm (and 0.05 s) to get to 500 mm/s with 1g.
Don't even get me started on those claiming 300mm per second on a Kossel Mini with magnetic ball joints. The ball bearings add a LOT of weight to the moving assembly, probably doubling it. I barely trusted them at 150mm per second on my Rostock.They probably do not really know what speed they are actually running. I'm very satisfied with my ball joints. No way I would switch them for magnetic joints (especially after what you reported about them). One ball joint weights only 5g (even with the associated screw/nut to platform/carriage and the screw to the carbon rod). One finished rod with two ball joints and the nuts to connect them to platform/carriages weights only 15 g. It is stiff, strong and does not have any play I can notice. No way I could achieve this with magnetic joints.
your stepper moment depends on the driver and motor, as the driver does not pass off vcc to the stepper. the driver is buck charging down to a lower voltage and high current. if you buy matched drivers and steppers, the driver datasheet will give you the holding torques, but the vcc voltage of the driver doesn't really matter, since it's buck charging whatever it gets down to the stepper voltage. the advantage of giving a driver higher voltage is generally that the power supply will run more efficiently, as it doesn't have to push out nearly as much current.OK, I'm not that good with stepper drivers. Do at least some of them even multiply voltage internally to ramp up current quickly? Because if they do not, then it takes some time to raise the current (because of stepper winding inductance). And if this time is significant part of the whole time the coil is driven (which is shorter with higher rotor speeds) then it lowers the available torque.
I'm sorry because I did not take the time to read the 600+ messages of this thread and I'm afraid my question may be off topic...for a porous surface i'd start with a layer of white artist's gesso. it's a very white acrylic primer that can be bought from any art store for almost nothing and loves porous surfaces. just brush it on.
A friend did print my Ergodox case using SLS 3D printing with white polyamide powder.
The case is really nice. The only point is the granularity/porosity of the surface which will make it get dirty quite quickly and impossible to clean up.
I got told to spray the case with Sikkens Spotprimer then spray with any color I want.
I built the Ergodox because:
- I found the project really nice
- I'm not 100% satisfied by my TEK
- I'm a geek (at least this is what the other say)
- I'm willing to reduce some elbow/shoulder pain
I would like to avoid exchanging elbow RSI with some finger skin allergy due to the "any color I want" I would have sprayed on my everyday keyboard.
Any suggestion about which color or kind/type of color I should select for that usage ?
Thanks.
Well, it looks like DOMO cannot accept a Cherry MX switch into that nice middle hole. If so what it is good for?
This is what I would like to print ... eventually. It can accept 40 switches. Not only one, not only after the teeth are filed away :D
(Attachment Link)
On the other side, DOMO has a decisive advantage. It can be printed now. My part is far from ready to be loaded to a slicer.
for a porous surface i'd start with a layer of white artist's gesso. it's a very white acrylic primer that can be bought from any art store for almost nothing and loves porous surfaces. just brush it on.
once the gesso is on, you can go crazy with any kind of paint you want. seal with a layer of acrylic clearcoat. i use a golden glossy UVLS clearcoat (uv protection layer) because it's water based and yet very hydrophobic once it dries. it is also made to allow even oil paints to dry underneath it. however, once it's on, it's on. you can't paint over this clearcoat.
the domo keychains can in fact accept MX keyswitches in their mouths.Nice, I'm p0wned.
the prusa i3 has an 8x8" bed, which isn't really large enough for vvp's design.The plan is to print one half of the keyboard in 4 pieces. Two pieces will be acetone-glued to the top part and the other two to the bottom part. So it comfortably fits to a 20x20 cm heat bed. I do not yet know how well the gluing will work but I believe it can be done well enough. So far my experience with acetone (and ABS-glue) is good. We shell see how it turns out.
So it's better to use a mirror or alu plate?The rostock here uses a mirror. There was no problem with flatness. Actually there was one. But it was not because of glass. The problem was that too much insulation was below the PCB heater and that caused the PCB to bent. When the mirror was clipped to the PCB it did bent accordingly. Putting smaller amount of insulation under PCB heater fixed the problem.
Is it really so bad to switch out the plate if it breaks?
vvp: from just looking at my ergodox it seems like the keywell itself most certainly does not require 8", have you tried to split the keywell from the rest of the model and print the wristrest etc as a separate piece?The standard ergodox probably does not fit 20cm x 20cm heat bed. My contoured version easily fits since:
What has your experience been with regards to cherry mx switch mounting in the printed material?
ANYBODY EVER USED 3-D PRINTING TO MAKE KEY CAPS YET?
ANYBODY EVER USED 3-D PRINTING TO MAKE KEY CAPS YET?Yeah but the stem was the wrong size. Quality off of a FD bed though was pretty damn good when sliced with a decent infill
i do think it's possible to make much better tweener printers, and it's very sad that people aren't really trying to do so.I'm trying. I stiffened the frame with a metal beam :)
The common dirt cheap FDM printers will not produce a keycap of good enough quality by themselves. If I would want to do a keycap I would use take some generic low profile Cherry MX keycap and I would cut sides of it away (leaving only the stem and the flat top of the keycap). I would 3dPrint the keycap shape I want without the stem and then I would glue it with the stem.
Even with an well calibrated FDM printer, you can get quite nice detailed prints with small nozzle (e.g. 0.3 mm), small layer height (e.g. 0.1 mm) and really low print speed (e.g. 3 cm/s). The only thing I would be afraid about is the strength of the stem (if you would 3dPrint it too).
See e.g. this as an example how small prints can look like: http://www.tridimake.com/2013/05/3d-printing-with-smaller-nozzle-diameter.html
If you would add some acetone treatment to it then it would look even better (if you do not mind shiny).
Otherwise some SLA (stereolithography) printer can probably do it strong enough even with stem. But that is also much more expensive.
Hi guys, what is the best 3D printer for making a keycap? I'm interested to get one but I need to learn more about 3D printer. I have seen a potato keycap from massdrop. As link provided, the spec is 3D Printed Beige. So If any 3D printer can do the same thing as that keycap, I will be so excited to get one! :eek: :p
https://www.massdrop.com/buy/potato-keycap?s=keycap
Some of them look like metal. So that would be SLS. Are the plastic ones done with SLS too?
I do not know whether there is any SLS printer which can be home made. There are some SLA home made printers.
Do you have any idea how is SLS expensive compared to SLA? As for as home made FDM, that is dirt cheap. Mostly the price of filament, which is somewhere around 25€ per 1kg (for a good quality). Leslieann mentioned that resin for SLA is about 10 times more than filament for FDM.
Hi guys, what is the best 3D printer for making a keycap? I'm interested to get one but I need to learn more about 3D printer. I have seen a potato keycap from massdrop. As link provided, the spec is 3D Printed Beige. So If any 3D printer can do the same thing as that keycap, I will be so excited to get one! :eek: :p
https://www.massdrop.com/buy/potato-keycap?s=keycap
As long as you make sure to adapt your cap a bit to the technology there are plenty of printers capable of making decent key caps. None of them are very cheap though, and as mentioned, FDM is probably a bad idea for this. Examples:Show Image(http://i.imgur.com/PJZLkfn.jpg)Show Image(http://i.imgur.com/GYPvxGH.jpg)Show Image(http://i.imgur.com/e0CMtrY.jpg)Show Image(http://i.imgur.com/MyKdfWZ.jpg)
The last image shows some compromises which I recommend you to do to make the stem far easier to print.
Hi guys, what is the best 3D printer for making a keycap? I'm interested to get one but I need to learn more about 3D printer. I have seen a potato keycap from massdrop. As link provided, the spec is 3D Printed Beige. So If any 3D printer can do the same thing as that keycap, I will be so excited to get one! :eek: :p
https://www.massdrop.com/buy/potato-keycap?s=keycap
As long as you make sure to adapt your cap a bit to the technology there are plenty of printers capable of making decent key caps. None of them are very cheap though, and as mentioned, FDM is probably a bad idea for this. Examples:Show Image(http://i.imgur.com/PJZLkfn.jpg)Show Image(http://i.imgur.com/GYPvxGH.jpg)Show Image(http://i.imgur.com/e0CMtrY.jpg)Show Image(http://i.imgur.com/MyKdfWZ.jpg)
The last image shows some compromises which I recommend you to do to make the stem far easier to print.
Those look nice! I want to make my own keycap but I don't know which printers suit me well. Thanks though
Hi guys, what is the best 3D printer for making a keycap? I'm interested to get one but I need to learn more about 3D printer. I have seen a potato keycap from massdrop. As link provided, the spec is 3D Printed Beige. So If any 3D printer can do the same thing as that keycap, I will be so excited to get one! :eek: :p
https://www.massdrop.com/buy/potato-keycap?s=keycap
As long as you make sure to adapt your cap a bit to the technology there are plenty of printers capable of making decent key caps. None of them are very cheap though, and as mentioned, FDM is probably a bad idea for this. Examples:Show Image(http://i.imgur.com/PJZLkfn.jpg)Show Image(http://i.imgur.com/GYPvxGH.jpg)Show Image(http://i.imgur.com/e0CMtrY.jpg)Show Image(http://i.imgur.com/MyKdfWZ.jpg)
The last image shows some compromises which I recommend you to do to make the stem far easier to print.
Those look nice! I want to make my own keycap but I don't know which printers suit me well. Thanks though
For tiny plastic objects, I'd say SLA is the way to go.
SLA is a type of additive manufacturing, not a brand. Here are a few machines which sort of target the home market:For tiny plastic objects, I'd say SLA is the way to go.Where could I get it? any links would be appreciated! I think it's not available in Thailand. For making keycaps by using 3D printer, any basic knowledge of programming (design) needed? Thanks. :thumb:
SLA is a type of additive manufacturing, not a brand. Here are a few machines which sort of target the home market:For tiny plastic objects, I'd say SLA is the way to go.Where could I get it? any links would be appreciated! I think it's not available in Thailand. For making keycaps by using 3D printer, any basic knowledge of programming (design) needed? Thanks. :thumb:
Form 1 (https://www.kickstarter.com/projects/formlabs/form-1-an-affordable-professional-3d-printer)
Titan 1 (https://www.kickstarter.com/projects/kudo3d/titan-1-fastest-tallest-print-high-res-sla-3d-prin)
Pegasus (https://www.kickstarter.com/projects/fsl/pegasus-touch-laser-sla-3d-printer-low-cost-high-q)
system, which has positive tolerance towards the left on the left side and toward the right on the right sideThis is the part which I do not see how it could apply to corexy.
let's forget about bias and polarity of forces on the carriage here for a second. assuming iid error on the A and B steppers and respective belts, dY = c(dA - dB). error cancellation.system, which has positive tolerance towards the left on the left side and toward the right on the right sideThis is the part which I do not see how it could apply to corexy.
It is not like the two corexy steppers work like the opposite sides of an anti-backslash nut. The steppers do not somehow apply a pretension on the belt forcing the belt to be always aligned to one side of the pulley teeth.
The two steppers are independent. Really (from the point of view of carriage movement and belt alignment to pulley errors) corexy looks only like a mechanism which transforms the coordinate system. In corexy it is rotated by 45° compared to a traditional cartesian. If you would not mind that the build area is diamond shaped (and not square shaped) then you can drive corexy just like a traditional cartesian. In corexy, driving only one stepper just moves carriage along a diagonal and changing direction of this one stepper (the second one still standing) will experience backslash just like in a traditional cartesian. The only difference is that the backslash will be along diagonal.
If corexy would somehow eliminate pulley to belt alignment error (the backslash) then you probably can also claim that traditional cartesian can eliminate this error too when it moves only diagonally (drives both steppers at the same time).
assuming iid error on the A and B steppers and respective belts, dY = c(dA - dB). error cancellation.
If anyone want to have a go on my just designed arcade buttons that use cherry MX (how come nobody thought of that before?!) here you go:
print one of these http://www.thingiverse.com/thing:421598
and one of these http://www.thingiverse.com/thing:417509
just pay attention to the 2 paragraph instructions on each "thing" and you should be fine.
I don't have a printer yet. And i'm dying to know if the plastic "plate" will survive one match of street fighter :D
If anyone want to have a go on my just designed arcade buttons that use cherry MX (how come nobody thought of that before?!) here you go:
print one of these http://www.thingiverse.com/thing:421598
and one of these http://www.thingiverse.com/thing:417509
just pay attention to the 2 paragraph instructions on each "thing" and you should be fine.
I don't have a printer yet. And i'm dying to know if the plastic "plate" will survive one match of street fighter :D
What printer are you planing to get? As you designed it, it will be very hard to print with an FDM printer.
The problems to fix if you want to use FDM:
- Remove the top rim of the housing so that the cylinder wall ends at the top suface of the Cherry MX mounting plate.
- Align the tops of the clips to the top of the mounting plate.
- You may need to make clips thicker so that they are stronger.
- Turn the whole thing upside down.
- If the inner overhang is wider than nozzle diameter then make it smaller.
- Print with support touching build plate enabled in slicer so that the outer overhang does not sag.
The button itself should be printable as you have it but the stem may not be strong enough.
what you're saying is true regardless of which direction the steppers move. however, look at the next pulley. in each system. we're looking at the error _at the carriage head_. in y-motion, the force of the two belts oppose each other. that's where the error would otherwise be coming from. you have to calculate ultimate positioning error due to red and blue forces.assuming iid error on the A and B steppers and respective belts, dY = c(dA - dB). error cancellation.
I thought about it but the only result is that that I more believe that corexy does not have any built-in error cancellation for stepper pulley / belt meshing errors. That equation is just a mechanical coordinate transformation thing.
Here is the picture:
(Attachment Link)
Lets assume the carriage is moving up first. Both steppers are moving the red/blue arrow directions, i.e. red (left) stepper counter clockwise, blue (right) stepper clockwise.There you go. There is full play even in the Y direction. No cancellation.
- The bottom teeth of the red stepper pulley will be aligned to the left side of the belt teeth.
- The bottom teeth of the blue stepper pulley will be aligned to the right side of the belt teeth.
- Now we change the direction of both motors so that the carriage is moving down.
- First both steppers move a little bit without moving the belts/carriage.
- Then finally the red stepper pulley will engage the right side of the belt teeth and move the red belt.
- At about the same time the blue stepper pulley will engage the left side of the belt teeth and move the blue belt.
CoreXY is cool. If I would built a cartesian bot it would be corexy. But I do not think it has any error cancellation features related to the play between pulleys and belts.
what you're saying is true regardless of which direction the steppers move. however, look at the next pulley. in each system. we're looking at the error _at the carriage head_. in y-motion, the force of the two belts oppose each other. that's where the error would otherwise be coming from. you have to calculate ultimate positioning error due to red and blue forces.If you mean there are no moments of force on the big carriage (and therefore it should not twist a bit while it is changing direction) then I agree. That is the advantage against hbot.
There may be something wrong with Makerbot Gen 5 machines:
http://www.hacknorway.com/project.php?id=28
There may be something wrong with Makerbot Gen 5 machines:
http://www.hacknorway.com/project.php?id=28
Can't help but be a little happy that Stratasys seems to fail at everything they try.
yes, i'm talking about the former, specifically twist and ringing. obviously during a steady state feed, the thing is just moving and that's about that.what you're saying is true regardless of which direction the steppers move. however, look at the next pulley. in each system. we're looking at the error _at the carriage head_. in y-motion, the force of the two belts oppose each other. that's where the error would otherwise be coming from. you have to calculate ultimate positioning error due to red and blue forces.If you mean there are no moments of force on the big carriage (and therefore it should not twist a bit while it is changing direction) then I agree. That is the advantage against hbot.
If you mean that the forces will not move the small carriage because they compensate each other then they better should. If they would not then it would be worse than a simple cartesian because what was intended only a Y motion would cause a small X motions too.
If you mean something else than I do not understand.
Do you guys know a good reference for designing for FDM? i can't find anything.if you're doing true FDM and not filament fusing, you can use the shapeways references. they're pretty good for the majority of machines. they have specific wall tolerances to their machines, but generally you want to stay much larger than those anyway.
If you live in or will be visiting the Northwest December 12, consider this your invitation to come see us at our big Open House at our USCutter location in Redmond, WA. There will be food and drinks galore. We’re giving tours of our warehouse and showrooms. And we’ll be showing off the latest new Mimaki, our full line of 3D printers from Leapfrog, Phoenix, and other equipment that can help make your 2015 more prosperous. We’ll also be giving away a Mimaki 28” CG-SRIII and a Phoenix 3D Printer as door prizes and other goodies as well.
The fun starts at 12 Noon and we’ll be open until 7PM. Mark your calendars now and plan to attend. We'd love to see you there!
You do not need to print it upside down. FFF printers are good enough at bridging to print a keycap. Just design the bottom side of keycap top to be horizontal and it will print just fine without support. See here:
https://geekhack.org/index.php?topic=43362.msg1507680#msg1507680
Don't mean to bump a dead thread, but didn't wanna make my own if it's already there. I've got access to one of those Cube 3D printers and a MakerBot Replicator 2. I've tried printing a cherry R2 keycap and it turned out alright, needs a bit of filing, but I recently found out I can change the filament to be much thinner. I've also tried printing a topre keycap, but had no support and the edges turned into spaghetti. I don't have much time to tinker around with 3D files at the moment, but once I do I would want to make MX compatible sliders next. Anyone have any experience with these?
1 spool per project would drive up the costs, then yet again, running out halfway could be worse. Show us some pics of your 3D print once it's done :)
Anyone else get a new spool for every new project? :)Not me. But do you know that you can pause a print, replace the spool/filament and continue later? The filament will not be perfectly primed so you will be able to notice it on the printed part but it mostly does not matter.
Anyone else get a new spool for every new project? :)Not me. But do you know that you can pause a print, replace the spool/filament and continue later? The filament will not be perfectly primed so you will be able to notice it on the printed part but it mostly does not matter.
Edit: The best place to do it is during infill.
1 spool per project would drive up the costs, then yet again, running out halfway could be worse. Show us some pics of your 3D print once it's done :)
Here is a prototype bronze plate:
(Attachment Link)
After clogging my initial hotend with bronzefill, then dislodging the replacement E3D with XT-CF20, sticking to PLA seems like the better choice, tho I have a hunch the E3D should handle bronzefill well, but can't risk it
Anyone else get a new spool for every new project? :)Not me. But do you know that you can pause a print, replace the spool/filament and continue later? The filament will not be perfectly primed so you will be able to notice it on the printed part but it mostly does not matter.
Edit: The best place to do it is during infill.
yeah :D
used to do it a lot, not for replacement but mainly for fun, mixing different colors
last time I paused a print tho, during a power outage, things went to hell, for some reason the print didn't continue as expected, it turns out if I waited 2-3 minutes the power would come back and the ups would last - after that event I didn't risk another pause yet, let alone a spool change :)
In hindsight, I'm guessing it was a Simplify 3D bug
The size of the printed part is limited not only by the build plate but also by warping (especially when you do not have heat chamber).
My problem with laser cutting is that I cannot laser cut a contoured plate like this: https://deskthority.net/post247068.html#p247068
The top part of the case is printed in 2 pieces (keywell and palm part). Then they are glued using the technique I described. Works pretty well.
Off topic, those keycaps are just "mmm"Nothing special. The black ones are OEM DCS salvaged from an old keyboard.
Another challenge with multi-part printing is the sub-1mm imperfections, when you put 2 pieces side by side, the 3d printing imperfection is pretty obvious even if it's 0.1mmYes that is the problem. That is also the reason to leave 0.3 mm gap at the places where parts are going to be glued together. 0.1 mm from each side for surface imperfection and 0.1 mm in the middle for the glue.
Leave 0.3 mm gap between the pieces.
My previous assumption was the print to be stateless, so the printer could execute the g-code at hand, and after the resume, continue executing the remaining g-code simplify3d sends
[...]
Even if it turns out perfect, 3d printed parts are awful to touch :D
After building a couple of kits I decided to take the plunge and 3d print as much of as Ergodox as possible.
So far the case is done and I'm working through printing all the keys - I've included some shots of my WIP.
Because I'm using a FDM machine and want to avoid supports as much as possible I've based the case around Lister's acrylic case.
For my printer I had to make the switch plate 1.3mm thick for the clasps on the switches to properly engage, but that was the hardest part.
The keys themselves can be tricky on a FDM style printer, but I've found something workable @ 60 microns and 30 mm/s
The stems were particularly tricky to get "just right" so they don't grind anywhere.
FWIW material costs are insanely low - the case cost perhaps $10 to print (including failed attempts) and the keys are 5-10 cents a pop. :thumb:
The a circuit board, switches and teensy from MK set me back another $100 or so.
For my next attempt I'm going to see if I can use ninjaflex to mask the traces on some copper clad board and etch it myself :))
Isn't that stuff carcinogenous?
You can paint PLA over with acetone/ABS slurry, it bonds to PLA VERY well.
You can add a bit of a powder dye to it as well, to make the colors better.
Actually, ever since I got into that 3d printing thing, ABS/acetone slurry started to cover most of my gluing needs, replacing cyanoacrylates and epoxies in most cases. Stuff's fast drying, sturdy, has great elasticity, and shrinks when drying, clinging mechanically to elements it can't make a chemical bond with. And you can easily dilute or concentrate it.
Wiki claims THD is about as poisonous as acetone but it is suspected of causing cancer.
Styrenes (released when printing with ABS or sublimating from polystyrene) may be linked to cancer too.
Wording about PLA was "probably not causing cancer". So it is not completely safe either.
I guess only few substances nowadays are not suspected of being linked to cancer.