The Living 3D Printing Thread

[mkawa]

on that note, i got two material samples in today:

1) borosilicate 0.125" thick sheet
2) alumina ceramic 0.025" thick sheet

the borosilicate is clearly going to be much easier to machine. it should basically lap flat just like glass. which is good, because this sheet is _not_ flat.

the ceramic requires diamond tooling iirc to machine, so what you get is what you get flatness wise. however, the flatness of this sample is reasonable, and more importantly, it conforms quite easily to my surface plate and apparently requires a huge amount of bending force to crack or otherwise break. thermal conductivity is fantastic. i could see a sandwich where the top layer is a very thin layer of alumina ceramic, the middle layer is a filament carrying a ton of current, and the bottom layer is something very flat, but not thermally conductive. not only would this heat quickly and uniformly, but it would be almost impossible to damage.

[Leslieann]

Yes, Ceramic will need diamond tooling.

You are going to spend a ton of time making a super flat bed, then heat it up to 80c (which can cause things to shift), inside a heated box (again, things shift), and then you plan to lay down plastic, which warps like CRAZY as it cools.

Even if you could print a keyboard case flat, ABS would prevent you from doing so. Any more than a few inches and by the time you get a few layers, as they start to cool, they will shrink, pulling the ends up. Experts actually only recomend a 10cm length tops because of the warpage.

Take a look here:

This guy is awesome.

[mkawa]

first of all, i actually think that we should just do large objects in PLA. unless there's some material reason why we need to print something in abs we should just print it in PLA, period.

second, i think we need to recall high school physics: there are two kinds of effects that we refer to as errors in any process; there is stochastic error, this is the isometric shrinkage that causes surface abnormalities and roughage in FFM printing, and there is systematic error, which causes distinct geometrical errors in the shapes that are printed via FFM. how is this relevant? flatness of the surface plate is an attempt to minimize systematic error, and heating the surface plate is meant to reduce stochastic error. however, it's important to note that the error factors introduced by these two flaws in the manufacturing process are not independent!

in particular, as i think i ranted about earlier today, every damn thing is stochastic! some things are just lucky enough that their random error variables have tiny short tails and low variance. i will admit that i didn't watch that dude's entire youtube rant, but my feeling is that he's not analyzing his sources of error; just pointing out that you can never remove the error completely is not a productive use of one's time. so here's my feeling. we can do a lot better than what's contained in the reprap wiki and/or what's commercially available. yes, i'm going to spend a ton of time experimenting and hopefully build a flat plate that stays flat when you heat it to 100C or so. then, i'm going to lay down PLA on it in layers, and hopefully the thing that comes out is going to be incrementally closer to my CAD than the thing that was produced before i sat down and thought really hard about the sources of error in the original design

if not, i'm sure as hell going to understand the issues better, and nothing's going to stop me from trying again )

ps, can you tell i'm going stir crazy from having my printer another month out? )

[Leslieann]

Reducing error is good, don't get me wrong, but I just think you are going a bit extreme on a system that is only really capable of .1mm (.003in) accuracy under the best of circumstances.

As for PLA vs ABS, PLA isn't as strong, nor are the colors as vibrant, however, I hear it's great to work with.


Looks like I finally got mine dialed in other than scale.
I may need to add an extra cooling fan to the head to sharpen up my corners a tad, but otherwise my prints are looking quite good.

[mkawa]

PLA is only incrementally less strong than ABS, and remember that an object that's printed FFM is going to be way way less dense than an object that's molded under high pressure. as everyone is constantly pointing out, the defects in the bulk material are HUGE, much bigger than the 0.1mm that _the stepper motors only_ are capable of.

instead of thinking "well, this is a low accuracy manufacturing method, screw it", think "ok, how do i reduce my other error sources to get to the fundamental accuracy (computational and electromechanical) that this manufacturing method is capable of".

if we were talking about that cool but weird heated filament pen i might be like "yah, screw it, we're limited by the human hand", but stepper motors and computational units can be incredibly accurate. the only thing holding FFM back is the introduction of bulk defects from physical sources; otherwise, it's actually a pretty damned accurate way of making things..

[Leslieann]

PLA is actually quite a bit more brittle. Number-wise they may seem similar, but in actual use, it's quite a bit different. I watched a Youtube video of a guy messing with and breaking both, ABS fared much better. PLA is also fickle in terms of moisture and we have high humidity here. Installing and removing filament after each job is cumbersome, you can't just pull it out of the extruder when finished.


As for the accuracy,
Of all the things in your printer effecting quality, you are focused on the part that is already the most accurate part in the entire machine and where you will see the absolute least (if any) gains. You can't even level the bed to the level of accuracy you are after. Accuracy is good, but focus it on something that will visibly make your prints better. I can't tell you where that is on a cartesian, probably belt tension or something, but it's certainly not the bed accuracy.

[mkawa]

abs is actually way more fickle in terms of humidity. i think i mentioned earlier, but dry box.

bed accuracy is huge because it's systematic, and shrinkage is one of the biggest issues, especially with abs.

ps: steppers and threaded rods

[Leslieann]

Everything I have read says exactly the opposite about humidty. PLA is soy and corn based and just soaks it right up. ABS is more heat fickle and while humidity bothers it, it's nothing like PLA. Strength and humidity were why I started with ABS.

Bed accuracy can be huge, but again, it's already the most accurate item in the printer. You are targeting the most accurate item hoping to help make up for the shortcomings of everything else, which are known to be quite inaccurate.

Until you deal with the other inaccuracies, you can put the most accurate bed known to man, but it won't make a lick of difference in your prints. This goes triple since you have yet to even try the printer and see just how many things are going to cause trouble before you even start putting down plastic.

[mkawa]

neither material is hydrophilic, and bed accuracy (flatness and cooling issues) is the least accurate part of a threaded rod-based printer FOR PRINTS FRIENDLY TO ITS COORDINATE SYSTEM

what actually happens with polymers is that water gets between the polymer chains, and becomes a defect in the bulk filament. between PLA and ABS, ABS is the stickiest water-wise, and hence has the most problem with this, in addition to having the worst shrinkage and thermal behavior. PLA was championed earliest for FDM and FFM printing because a) lower temperature resistance and property change points means it's most stable while cooling and b) less sticky in terms of water. further, even if you do get water in the filament, you don't need to heat it as much to get the phase change and hence the water produces fewer defects in the printed material and is less problematic when the filament hits the heater. that said, any bulk polymer filament should be stored in a dry box.

the head position on a cartesian printer is by far the most accurate part of the printer. it's completely analogous to a CNC mill. as long as your stepper motors are accurate, your rods are accurate, and the head itself is fixtured with enough force, basic controls can position the head to near the 0.1mm the threading and steppers are capable of. (note that at this point all i'm talking about is moving to the point and then sitting there)

your printer has a completely different set of error sources; your head positioning depends on tie rods and bearings, and apparently the tie rods are attached to belts at some point? completely different, and a source of much more error for certain types of operations.

for example, consider the g-code execution of printing a cylinder on your machine vs a makerbot. the makerbot is going to have positioning issues due to computation accuracy, nasty inertial forces all over the place, etc. your machine has fewer fewer movements to make, and can take code in polar coords (the point of that design). further, the inertial forces are going to look completely different. in this case, your bed accuracy is going to play a larger part and my bed accuracy is going to play a smaller part in the printing error.

now, consider the g-code execution of printing a cube on both machines. the makerbot is going to be happy as a ****ing clam. only one motor's going to have to move at once, and now the bed accuracy is a BIG DEAL. your machine is going to be pretty unhappy.

make more sense now?

[Leslieann]

I still don't buy it, I don't think it will hurt, but I still doubt it will do much.
The plate you are are getting is already considered one of the better ones in the industry. I'm not saying it can't be better, but have serious doubts that making it better is going to do much. It;s not like your printer comes with a plywood bed.

[mkawa]

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. even so, it's a plate that's too thin to even bother machining flat, even if it had uniform heating (the corners have a tendency to "wing out"). it is one of the better in the cottage maker industry, but the cottage maker industry just doesn't have a lot of resources at the moment (although we're here right now working to change that ). the makerbot solution is nice because it's "good enough" for a "long enough" lifetime, and then it's easily replaceable. and note that even then, you have a tilting table under the bed that you have to use to try to level your printing space for each job, or else, as i was told by an MBI rep "yah, your prints will probably warp otherwise. we recommend a [tolerance] height gauge and adjusting the table every few prints".
u
let me put it this way. a bridgeport 3-ish axis cnc mill costs 70 grand without even the gcode compiler, and its weight is measured in portions of tons. when you want to make a circular hole in something with a bridgeport, if at all possible, you take a machine bit the diameter of the hole you want to make, and you drill out that hole directly, because it's not quite accurate enough to make a circle anywhere near as accurate with a smaller tool.

[mkawa]

christ, i'm totally rambling at this point.

let's see your new prints!!

[Leslieann]

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!?
Okay, I get it now, that's insane.  I wouldn't play that game either.

I'll post some fresh prints in a bit, I am ordering magnetic arms this weekend, which will help, but at the moment I had to recalibrate things as I redid my upper and lower plates because the old ones were done in a rush (and were quite off).

Other issues:
The cheap aluminum pulleys I got were complete junk and stripped out. I found while replacing one, that it was rocking about 1/8th turn on the shaft. It's amazing how good my prints were, all things considered.

I'm still working on getting correct scale and final calibrations on things. I got a new Extruder, put in a cooling fan and duct and just getting things squared away, but the prints are getting better. I did a hollow cube earlier and it came out fantastic (needed more fan), but before getting a picture, I broke it apart to see how it looked inside. My bridging inside was good, as was the strength, so I know my retraction and cooling is pretty much dead on.

I'm going to get some PLA next week as well and see how that goes. I also plan on getting another head soon. I bought this one with the intention of using it only to get a handle on things as the one I wanted was out of stock. I need to adjust my P.I.D. tonight too (no idea what it is), so I get better heating on my bed and head. As it is, I'm pushing the little power supply quite hard by the time I run lights, heaters and fans.

This is what happens when you build from scratch though, and it's interesting.


I will admit, if there is one thing I dislike about deltas, it's the calibration. Getting it to print flat, at the right scale, and the right height... It's like groundhog day.  Make a firmware adjustment, home all the towers, measure and adjust, make another adjustment to firmware, home all the towers, measure and adjust, wash rinse, repeat until the towers all match the bed center. Any time you change one thing, every measurement change. You can't just do it once and be done, so you spend hours, going up, going down, upload, measure, up, down....

And it gets worse, these measurements are supposed to done while hot. So while you are sliding a bit of paper under the hot head, you are trying not to bump it with your knuckles, while watching the screen, and clicking the mouse to move the axis. During all of this,  your fingers are insulated from the 80C bed, only by a thin bit of paper. NOT FUN.

[Leslieann]

Potato cam time...

TL:DR
Left is latest print @ 2x speed, on the right is my very first print.



I know some of you are probably wondering why I'm still tuning this thing. Well, first off, as mentioned I made some changes, had some problems... This is a home built machine built from (rather loose) open source plans, and I can be a bit obsessive at times.

Anyhow...
On the right, is the first thing/cube I printed. A 10mm solid cube (why solid?) printed with a .5mm nozzle, it was 20% over sized (.0785in) and, well, just look at it. It leans, no uniformity... This was a result of bad frame alignment, loose belts, and what I later found was a pulley that was holding on for dear life and several rods weren't tightened up all the way. But hey, it printed so I was happy at the time, I finally had something, regardless of how horrible it was,  to show for all the money and effort.

On the left is the latest, yes, I know the top infill isn't great (I was experimenting)... However, it was printed at twice the speed as the one on the right with a .35mm nozzle, and is completely hollow, the walls are only 1mm thick and it won't fall apart like the one on the right has done, it takes a pair of pliers to crush or damage it.  It's also within one thousandth of an inch of what it is supposed to be, I think I can cut that in half and possibly triple the speed with some changes. Maybe your super bed isn't so crazy after all! It wasn't until I got my calipers and I converted my numbers to inches that I realized. Now I'm kind of shocked to be honest, I really didn't expect that sort of accuracy. Maybe it's time to stop fooling with it and print something for real.

The one corner looks bad (as does one on back corner) for a couple reasons. The corner you are seeing happens to be where the head was stepping up. On something this small, the loops are small and so the previous layer hasn't cooled quite enough to compensate for the added time of the head being there while moving up and adding more hot plastic, so you get a slight melted look. The back side suffers because my fan doesn't reach it as well, I just installed it today and may add a second of a deflector to redirect the air back there. I had two fans before but with no ducting and was getting worse results with two, larger fans than I am now. Some of this can be fixed by adding a pause between layers, but that can leave the head sitting in a bad spot, lowering temps, more air flow and more... That's part of the problem though, every problem has 10 solutions, each with different results that may impact something else, so there is a lot of trial and error.

Anyhow, there is very clear progression there, in fact, if you ignore the heat issues (due to size) and my poor infill (again, I was experimenting), my prints are better than the parts on the printer itself at this point and only getting better.


Edit: I just noticed the picture makes the new one look like it tapers in at the bottom, so I checked it with my calipers, it's showing a difference of .0005in between the two. While I have doubts that my El Cheapo caliper is even that accurate, I can live with it.

[mkawa]

you're getting there. that's not bad for abs

PID algorithm: http://en.wikipedia.org/wiki/PID_controller

it's the simplest control algorithm

[Leslieann]

Thanks.
I'm pretty happy with it, Just a little more and I think I will consider it done.

Then I'll start working on my next one. LOL
I have a sneaking suspicion this one may be sold by the end of the month. I have some people interesting in getting a printer, and I'm pretty sure at a meeting I have later this month that I could probably sell it to finance the next one pretty easily. If so, I'll be building a Cerberus or Kossel next (I will whether I sell it or not!).


There are a LOT of pitfalls doing it from plans.
If anyone is considering it, research, research, research. It's easy to get lost or over spend.

[mkawa]

IT SHIPPED!!!! IT SHIPPED!!#@*)(#)*(@!)(*#!(*)@#()* ao ewxzcirwewsklj ao wzxirwskjlhakj aso excited i can"t even type!?!?!?!?!#!#()_#!)(_#!

[Leslieann]

Awesome!

I have some PLA coming, magnetic arms (better precision) and a stronger power supply. The latter is something I hear yours actually can benefit from as well.

I have 17 amps (old PSU), the board uses 5amps and the bed uses 11... supposedly. However, when I flip on my LED lights I can see them flicker depending on system load, so I have a 20 amp coming to replace it. Plus, the bed struggles to warm up when I have even the slightest breeze. In fact I was getting some airflow from my Arduino cooling fan and that was hurting heating times even. I have never gotten it over 80C. I built a guard to redirect that air and it helped, but more amperage is needed. The current PSU is a noisy little bast*rd anyhow. I have seen three of these power supplies and not a single one has a good fan in it.

I thought I blew up a motor or driver the other night as well, the driver modules are just fickle. Too much power and they overheat or cause the motor to click, too little and the motor won't move. I think it was fine, but I ramped up the power to stop it clicking, which only made it worse. I think the real problem is I may have a clogged nozze. Weee...

[mkawa]

i have power supplies. christ i have a lot of power supplies. i also live 10 minutes away from allelectronics

power supplies will not be an issue.

clogged nozz's suck!! datavac to the rescue?

[Leslieann]

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.

I got a nice low profile one with 20 amps for $20, it's overkill, but I would rather be over powered than under.


I got the bed up to 80 finally. I swapped back to my .5mm nozzle, and still having some feed issues and retraction problems. Just not quite set right I guess. One of those instances where a pre-built would be a lot easier to deal with, but, meh.  I did this as an experiment, albeit an expensive one, but you have to pay to play.

Also, if you haven't tried it, try Tinkercad. It's not a fantastic cad program, but you can knock out simple stuff extremely fast.

[mkawa]

forcing myself to learn solidworks. yuck but what're you gonna do

[kmiller8]

Hey, I just bought a 3D printer too! can I join the party OuO

[Acetrak]

mkawa I can give you free lessons np

[Leslieann]

Here is a tip, when trying to trouble shoot, keep a log of changes you made...

My extrusion issues have been solved, I found a small setting I had changed waaaaay back trying to solve a different problem. I fixed the other problem, but this was haunting me now. I'm thinking of calling this printer "Butterfly effect", because any minor change you make, can have HUGE results elsewhere.

The air nozzle I'm printing looks FANTASTIC, bridging is good, no lift, no overhangs, no odd noises, no droops, no melting, it's just soooo smooth. This is with a .5 nozzle, .35 layer height, 60mm per sec print speed (average speed for a cartesian).


Yes, I'm swooning, it's the best print I have made so far, and it's with my large nozzle and a higher speed than I have been using. This is what I have been aiming for.

[Leslieann]

I like PLA... for about 5 minutes, then my print head shaft gets too hot and clogs the whole works. This stuff is REALLY, REALLY temperature fickle.

I have reconfigured my fans 4 times now trying to get it to work with no luck. Best I have managed is about 10 minutes of printing with it.  All this time I thought my extruder was underpowered, it's not, in fact I snapped off more than half the metal teeth in my Bowden tube press fit connector.  So apparently, I've been battling this head from the start.

I get fantastic prints at slow speeds, then as the teflon liner warms up, it gets worse, and worse and worse, until it just clogs entirely. I had it printing at 200mm a second and after 10 minutes it couldn't even manage 1mm per 30 seconds. By the way, watching a delta at 200mm per second is awesome(!) but wow do things to to shaking. LOL

Oh, I don't mean the printer, don't get me wrong, it was swaying and shaking, but it was printing great, it was more than capable. No, it was shaking the desk, which in turn was reverberating into the floor. You would have thought someone was having some extremely hardcore sex, during an earthquake, while on PCP. Needless to say, I didn't let it run like that for long. Something would give under that kind of punishment.

Sooo, next paycheck, I'm ordering a new head. This was only meant to be temporary, but E3d heads are still a 3week wait (been that way for weeks). Trinity has a good one as well, but theirs is meant for 3mm, not the 1.75 I use. The only other all metal one I found, doesn't have a built in bhowden attachment. I'm starting to consider just making my own, they aren't that complex really. In the meantime I may order up a J-Head and see how that does, unfortunately, those too lack a bowden attachment.

[mkawa]

holy crap

[linziyi]

:O it's in colour! Thought 3d printing can't include colours in it

[vvp]

:O it's in colour! Thought 3d printing can't include colours in it

Even the cheap DIY 3d printers can do colors. The question is why would one bother with it?

[Leslieann]

holy crap

Damn that came out great, I'm a bit envious at how easily you got up and running.

Watch out for shrink on ABS, it's like dunking a hot guy in freezing water if your bed temp isn't above 90C. 



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. 

When they broke they not only jammed my nozzles, one piece also jammed the entrance to the print head. What stinks is I kept redoing my fan layout, thinking the PLA was clogging. It was, however it wasn't a temp issue, it was metal in the feed system causing excess pressure, which allowed the PLA to overheat.

Now that everything is cleaned out, I'm printing ABS like a champ (even that had begun to clog after 5 minutes of printing). I have two new fittings on the way, and I'm reworking my fans a little more to hopefully get PLA flowing good. I'm finally able to start making things other than small simple stuff.



I found out a few other things.
I upgraded the PSU from a tired 17amp unit to a new 22amp, added a larger heatsink (with heat sink compound) to the Ramps board, and increased the awg size on my bed from 18awg to 16awg. WOW what a difference. I immediately saw a huge change in how the bed heats up, and no more light dimming when the lights are on. However, this unveiled a new issue... The new PSU showed I was still undersize on wiring and the wires were heating up bit enough to burn, but quite warm, had I still been using the 18awg, I'm sure I would have melted something. I put 14awg wires on the bed and not only are the wires cold, but the bed heats twice as fast as it used to.

I did some checking, I was on 18guage wires and according to some charts, for a short run, 18awg will handle 11 amps. However, I had issues even with 16awg. This tells me that either some of the wire size charts are wrong, or the bed is pulling more than 11 amps. Before, it would take 15 minutes or more to reach 75c, and that was with no fans (sometimes it struggled even with that), now it heats up to 90c in about 5 minutes, even with a small fan blowing across the glass . 

I also insulated my head (currently in the garage drying), and I have the parts for my magnetic arms, which will clean up slop, but I need to buy new JB weld as the stuff I have I think is bad, it didn't hold very well at all. Once I get PLA flowing I'll finish the magnetic arms, and then I will be off and running with good accuracy.  I still want   better head, but now that it's not as much of a hurry now that I know the head wasn't the main problem. I can always use the old one for another color. I just need another Nema motor and bowden tube.

Oh, and I did finally get the PID done, but I have to redo it again now that I changed the wires and insulated the head.


I hope no one has the wrong idea, I built mine from scratch from a newer design. It's bound to take some tuning and tweaking. While Mkawa is printing great, he spent a lot more to do so. I wanted the newer style and the experience of working from scratch, mine was as much about the journey as it was about the destination. Both methods work and eventually (with luck) arrive at great working printers, we just took different routes. Some of my issues were my fault, others were luck of the draw, some are due to lack of documentation. I'm sure Mkawa is happy with his, as am I with mine.

[Leslieann]

Even the cheap DIY 3d printers can do colors. The question is why would one bother with it?

Two colors allow for lettering or designs to be placed on objects.

[mkawa]

rule of thumb is to run twice as many wires as the theoretical max. the theoretical max assumes no defects in the bulk wire (whether stranded or solid). this is _never_ the case. defects raise resistance, which causes dissipation in the wire (resistive losses = power expended in wire = hot wire). also pay close attention to the specs you're reading as theoretical maximums. they may be ratings at something like 60C, which is way too hot for practical runs of wire.

it's a bit like speccing power supplies. switching supplies are typically happiest at 50% max steady-state load @ 40-50C. raise the temperature or load and you're likely making some semiconductor or dieletric very unhappy for short, or worse, long periods of time. you're also typically increasing noise and ripple, which can cause problems through the circuit that the rail supplies if you're say, analog or mixed signal and psrr is not impeccable.

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. it seems to me (and feel free to correct me if i'm wrong) that you're attempting a print, guessing a fix, attempting another print, guessing another fix, etc. i may be too far in the other direction, but i don't like to start the machine until i have concrete properties and measurements in my head and can validate what _should_ happen. then, i can stop things and perform a control step the second i see something i don't expect. this has nothing to do with starting from a kit vs a built machine; although i am also surprised the built machine was able to pump a few small things out so easily, i can guarantee you that it's not all wine and roses.

that said, the first thing i'm doing right now is actually building a mount for a dial indicator so that i can do a regular topographical map of my build plate(s). believe it or not, the carriages and driver for such are pretty damn good, even with a few belts in the mix (doh! i was wrong about that one, only the Z axis is a threaded rod), and the build platform is definitely the limiting factor for me.

based on the first map i'm going to have to decide whether to try lapping, different materials, holders, etc. interestingly, the rep2x plate approach is to raw _cast_ (i'm guessing it's a gravity cast from the look of it) an aluminum plate. this is interesting because ideally you would actually want a forged plate to maximize uniformity of expansion, and then you would want to machine the finish to get verifiable flatness at room temperature. of course, as i said before, 75$ for a new plate, so a raw gravity cast it is . hilariously, the factory even misaligned the kapton tape (they use full on kapton instead of just teflon, ironically; the film probably costs more than the plate), so i have a gigantic seam running across the tape where one roll ended and the other began. you can actually see this in the traffic cone, although i'm not sure if i got it in the picture.

anyway, one of the goals of this project is to develop a new machine that, as i said before, uses truly polar gcode and construction. one of the nice side-effects of the traffic cone test print (included on an sd card with the machine!) is that you can see the computational approximation in the "circular" cone. this was pleasantly surprising, as going into this, i was sure that physical error would dominate computational error everywhere.

ok, novel over. fun awaits! feel free to ping outside the thread if you want to powow on really gritty specifics leslieann!

[mkawa]

hah! i just remembered why i bought tack-spray the other day. i have a sheet of teflon from mcmaster i was going to play around with (after cutting new mouse feet out of it ). yesss!

[tsangan]

Everyone can join in on this fun too now!

http://www.kickstarter.com/projects/pirate3d/the-buccaneer-the-3d-printer-that-everyone-can-use

[CPTBadAss]

Noooo, kawa, post more stuff here. I love this thread!

[kmiller8]

Everyone can join in on this fun too now!

http://www.kickstarter.com/projects/pirate3d/the-buccaneer-the-3d-printer-that-everyone-can-use

I had one of the $250 pre-orders on that, but the build area was too small, and the delivery date is too far out for me.

Some other cheap options.

makibox starting at $200, shipping sometime within a few months. (I think it's fully assembled)
printrbot starting at $300, shipping sometime this month at best. (Kit/fully-assembled for more)
solidoodle starting at $500, ships in a few weeks (fully-assembled)

Those are all the "cheap" ones I've found and would consider buying.

* kmiller8 has looked a lot into the options out there before making his decision.

[mkawa]

Noooo, kawa, post more stuff here. I love this thread!

dude, i have a real metals project that i need your help on (design, FEA and machine time). email a bro.

[mkawa]

great list kmiller. this is now officially the living 3d printing thread.

[The_Beast]

The soliddoodle is a decent printer. It took a while for my friend to set his up and get all the settings right, but now that it is, printing is really easy

[PointyFox]

We have a Solidoodle at work and the thing looks like garbage.  It's held together by parts made from 3D printers and various kinds of tape.  Also whatever that gold stuff is on the platform is also chipping off.  It could be the fault of whomever built or is running the thing. 

[linziyi]

What differentiate the $2000+ printers from the $300+ ones?

[mkawa]

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.

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

[mkawa]

What differentiate the $2000+ printers from the $300+ ones?

the same thing that differentiates all expensive from cheap tools: precision and accuracy.

[kmiller8]

great list kmiller. this is now officially the living 3d printing thread.

here's some more

Makerfarm RepRaps has kits ranging from $520 to $585, but doesn't come with powersupply/glass, so it's a bit more than that
Rostock Max is a delta-type printer of $1000
Mendel Max $1600 but one of the most solid reprap designs and a sexy looking printer
MakerBot $2200 "standard" for FDM-type printers (honestly a lot of cheaper ones can print similar quality, but after a lot of calibrating)

Obviously these are just a few of the options available, but these (to me) are some of the best kits and 3D printers.

[kmiller8]

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.

The popular one is the Form1 ($3300)
But there are DIY methods that would cost around $700

The big problem with these types of printers, is that the resin is very expensive, however the print quality is quite a bit better than FDM printers can achieve.

[kmiller8]

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

kmiller8 got a makerfarm reprap 8"

[mkawa]

the ultimaker is the current reprap produced design (reprap seems to be a loose-knit developer community at this point, and "the reprap" seems to refer more to the designs they product than any particular machine). that design looks quite nice.

power supply-wise, my recommendation is to search for HUGE used tdk lambda, meanwell, elpac, B&K, et&c supplies on ebay, then go to your local electronics surplus store and get a passive power supply tester. in my case, that's an 8 ohm 18" long piece of ceramic with wire embedded in it (next time i go i expect to find a 1ohm. if not, there will be blood). it will apparently sink 300W without catching on fire, but i'm somewhat doubtful, and regardless if i need to seriously test something past 150W i'll probably just put it on top of 12" long aluminum extrusion from same store and aim a fan at the combo. throw your voltmeter across it and eyeball that the voltage stays reasonably constant. remember, it's a voltage source, so if it starts to fail, the voltage will fluctuate (then it might explode, so put some glasses on or something).

now, just making a used power supply source a huge amount of current is NOT a full functional test. ideally you want to measure ripple under across the load range, etc. etc. etc. etc. well tough. if you really needed that kind of precision, you would buy new and it would cost you thousands of dollars instead of about a hundred bucks. hence, my recommendation is to buy a supply rated for 3x or so of the current you expect to actually use in steady state, and to not exceed 60-70% of rated peak current.

that said, modern switchmode supplies are incredibly efficient, reliable, and being pumped out by the buttload AS LONG AS you're ok with 12vdc. in particular, the makerbots run on 24vdc, and stringing switchmode supplies serially across a common AC ground naively is NOT OK. it's possible, but the results will not be pretty, and switching power supplies are cheap. in fact, they exist only because they're so cheap, so just find one that provides the power source you want as cleanly as you need.

[linziyi]

What differentiate the $2000+ printers from the $300+ ones?

the same thing that differentiates all expensive from cheap tools: precision and accuracy.

I heard that replicator x2 is "professional grade" 3d printer, so it must have insane accuracy... But for the cheaper ones, are they still reliable for easy tools like screws, nuts and blots?

***I know maybe this is not a very good question since most people only have dealt with one 3d printer***

[mkawa]

"professional grade" 3d printing is currently limited to 10-60k$ machines. formiga, stratasys, and so on make high accuracy high volume SLS and FDM printers that can print in multiple materials at much higher temperatures. the makerbot, cubex and 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)

[Leslieann]

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.

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

Resin is 15x the cost of ABS, I looked at building one.
The printers costs similar, but the resin costs will kill you.

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.

It's just what you are seeing based on what I write. I'm kind of trying to document some of the things I ran into trouble with.

I actually researched a lot and do test, measure and retest. Part of the issue, is as I mentioned, documentation. One place will say one thing, and another will say something else. Being open source a lot things are kind of like comparing Red Hat to Ubuntu. You get info from one that doesn't work with the other. As such you get a lot of conflicting info or even out dated info.

Your kit was built the same way, however they already went through all of the process of working out the kinks.
This is also why working machines fetch a decent amount of money.

[Leslieann]

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.

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. Speed can also be a factor in price.

The biggest thing though, is how much do you want to tinker. As one company says, "A sense of adventure required".