3D Printing: The Return and Reintroduction of FoxtrotZero

Status
Not open for further replies.

FoxtrotZero

Jr Member
I thought a post like this might be better suited for the New Recruits forum, but apparently I'm no longer able to create new threads there. The actual content of this post pertains, primarily, to 3D printing, so I figured this is a good secondary choice.

If you don't know me, don't worry, there's no real reason you should. I showed up a few months back and was active for a while before my studies at University consumed my life once more (I've been buried under 19 units this semester, plus a schedule that has me waking up at 6:30 most days to catch the bus). If you're one of the few people that does recognize me, I'm "back" - I can't guarantee I'll maintain a strong presence, but I'm going to try to maintain a constant one, not least of which because seeing the great work around here is my largest motivator (that, and the guilt trip imposed by my workbench).

If you want to know more about me, I wrote one hell of a text wall when I first showed up on this site, which you can find here. I can't honestly recommend reading all of that unless you're a bit of a masochist. Most of it pertains to my proposed build for a set of ODST armour, which I've since refined a bit, but not actually gotten far on.

Since posting that, I acquired some materials and started on a pepakura build. The first thing I did was the low-poly tester helmet, which I'm glad I did, because it did a lot to help me understand sizing (read: was a good several cm too small). That's still sitting, somewhat ravaged, on the shelf over my workbench. Then I got into it proper with one of the shin pieces. "It'll be good experience", I thought. "It's got a little bit of everything", I thought. "I can **** it up a bit and nobody will notice", I thought. God how I was naieve.

See, the shin pieces have a few things going for them that make them absolute hell. First, they're probably the largest or second largest piece of the set (the other being the thigh piece). Second, the entire front of them is a mess of small ridges and indentures. Third, every piece that isn't some tiny intricate bit involves these large, natural, calf-like curves, which are terrible to do in pepakura. And fourth, it's a closed, tubular structure (as opposed to a flat chest piece), which makes final assembly and test fitting a bitch. I still haven't finished that shin piece.

For the impatient or time-pressed among you, this is where I start discussing 3D printing

I'm terrible at segues so I'm going to jump right into it: It would be a whole **** of a lot simpler if I could just 3D print these pieces. It would be easier, probably faster (as far as man-hours is concerned), and probably more lightweight (since the primary material becomes plastic, not fiberglass). But I don't own a 3D printer, or have the files. Plus, some of these pieces would need to be broken up, and I don't know if the plastic (ABS, right?) is sturdy enough for what I want to do. Let us address these problems in order:

Access to a 3D printer: This one turns out not to be a problem. I have access to a fair few facilities through my university. As a freshman, they're not yet willing to let me into the Engineering department's fabrication center, but I happened to discover that the library has a 3D printer. They allow you to upload files and have them inserted into the queue for printing. The upside is that, best I can tell, they offer this service completely free. The downside is that, because they've only got a single Makerbot Replicator, it can take up to two weeks to have your part processed. Aside from the restrictions on size, the part has to print within 3 hours, and if it's over 40MB, you have to turn it in physically, with a flash drive. I really don't know how long it takes to print any size/amount of item, and thus, I don't know if this 3 hour limit would be a significant restriction.

Access to 3D files: I don't really expect this one to be a problem. Obviously 3D files of ODST armour exist, and I can imagine it's only so hard to turn these into 3D printer-acceptable formats. It also might not be a walk in the park. If anyone knows where I can get these resources, I'd appreciate the knowledge. To get a feel for the 3D printer, I've been thinking about having them print up the mask of an NCR Veteran Ranger from Fallout: New Vegas. The caveat is that I've never found a file for it, but since I have three different autoCAD programs on my computer, I think it might be worthwhile to exercise those skills. It is, after all, the printer I'm testing, not my own abilities.

Breaking up the pieces: Obviously this isn't inherently difficult, and might even work to my advantage with pieces such as the shinguards and vambraces (which I probably can't snake my way through, but will instead have to fasten together somehow), but there are two snags. The first is breaking them up in a clean manner, and the second is joining them back together. I'm not certain what sorts of glues or other harsh chemicals it takes to join two pieces of plastic like that, so that's knowledge I'd appreciate recieveing.

Structural stability of parts: I mean, I'm not expecting them to fall apart. But I expect they're going to get dinged around a bit, and it would be nice if they wouldn't crack. I assume the outside only needs sanding smooth, not a whole lot of bondo like a pepakura build needs. I don't know if bondo or fiberglass will adhere to the plastic (or, more importantly, if things like fiberglass resin will cause unpleasant reactions and/or gaseous products), but if one will do so (agreeably), then perhaps putting a thin layer of that through the inside would help with give it some sturdiness. If this is an option, it might actually be important, because thinner pieces would make it easier to avoid running up against the 3 hour time limit.

So, I don't know a whole lot about 3D printing other than the theory, but it's definitely something I'd like to learn about, and this is a great opportunity. If anyone can tell me whether this is a feasible project and, if so, help me answer some of these questions or at least point me in the right direction, I'd really appreciate it.
 
The 3 hour print time will definitely screw you over. Printing out all the parts for one of my helmet files that I'm working on right now (split into 4 parts after being hollowed because of build area restrictions) would take something like 36 hours... divide that by 4 and you're still looking at upwards of 9 hours for one part... however I would bet that if you came up with a good plan of action, research, etc. and presented it to the powers that be at your engineering department they would be willing to let you use their printers, possibly in exchange for you representing them with your sweet new armor ;) lol

The 3d files part is probably going to be more of a problem than you expect. Cleaning up game or pep files to be acceptably 3d printable is no small task - doing it for the Destiny helmet I'm currently working on probably took me over 20 hours so far and I've been working with 3d stuff for several years now.

As far as re-adhering parts and structural stability I'm not too much of a help there.
 
The 3 hour print time will definitely screw you over. Printing out all the parts for one of my helmet files that I'm working on right now (split into 4 parts after being hollowed because of build area restrictions) would take something like 36 hours... divide that by 4 and you're still looking at upwards of 9 hours for one part...
Yeah, I was thinking about that a bit. I spent this morning in the library and somehow, for the first time, saw the room off in the corner of the area with the big Build IT sign. I walked over and saw the printer on the desk, which read something like "100%, 01:32:56" or something in that realm, and it wasn't very big. I did, however, just look at their website again, and they specifically say "three hours per week". Which A) makes a bit more sense, as there's no way you could make much more than small trinkets otherwise, and B) means I might still be able to do this, albeit while taking even more time than desired. Maybe I'll send them an email and see what they have to say on the topic.

however I would bet that if you came up with a good plan of action, research, etc. and presented it to the powers that be at your engineering department they would be willing to let you use their printers, possibly in exchange for you representing them with your sweet new armor ;) lol
I actually went down and spoke to the fellow in charge of the Fabrication center a couple weeks ago. He says that, as I'm an engineer, I'll have an opportunity to be introduced to all the machinery soon - and that if I'm not familiar with it all by the time my senior project rolls around, I'm going to regret it. It's just that if he had a bunch of freshmen running around the place, he'd have even more problems than he does now. The subtext, thus, was "just wait".

The 3d files part is probably going to be more of a problem than you expect. Cleaning up game or pep files to be acceptably 3d printable is no small task - doing it for the Destiny helmet I'm currently working on probably took me over 20 hours so far and I've been working with 3d stuff for several years now. As far as re-adhering parts and structural stability I'm not too much of a help there.
Well I was hoping I wouldn't have to do a whole lot from scratch, especially since the programs I'm familiar with (Autodesk Inventor, Creo, and SolidWorks) are autoCAD programs, not 3D modeling programs, so they work a good deal different, and I don't think they're capable of importing existing models that aren't in their format (which has to do with the way models in AutoCAD aren't just a stored shape, but a stored sequence of operations such as extrusions, chamfers, and lofts, with specified parameters that lets them do things like automatically pull dimensions into the engineering drawings).

Obviously people have taken raw game assets and turned them into fairly clean models that are digestable by Pepakura. If I could get my hands on a model right before it had its polycount reduced for Pepakura (I assume that's something that happens before importing it into designer, lest you get an incomprehensibly complicated fold pattern), that would be perfect. Honestly, if I can verify that something like bondo will work on the ABS plastic, then I'd even be willing to use an existing pepakura model and shape the exterior on my own.
 
Obviously people have taken raw game assets and turned them into fairly clean models that are digestable by Pepakura. If I could get my hands on a model right before it had its polycount reduced for Pepakura (I assume that's something that happens before importing it into designer, lest you get an incomprehensibly complicated fold pattern), that would be perfect. Honestly, if I can verify that something like bondo will work on the ABS plastic, then I'd even be willing to use an existing pepakura model and shape the exterior on my own.
Actually, the game files are very low poly count, normally the same as the pep ones (technically sometimes double but only because they are all tris whereas many times the tris are converted to quads before pulling them into pep). The detail is applied with textures and normal maps in the game because it's less intensive than using actual high poly models. So, nobody has access to the high poly originals.
 
Actually, the game files are very low poly count, normally the same as the pep ones (technically sometimes double but only because they are all tris whereas many times the tris are converted to quads before pulling them into pep). The detail is applied with textures and normal maps in the game because it's less intensive than using actual high poly models. So, nobody has access to the high poly originals.

Well ****. Like I said, I don't mind having to base my work off of a pepakura original, if that's what I have to do. This assumes that it's possible for me to turn one of these (or, more likely, one of its precursors) into a 3D printable file, and to add to it and shape it with a material like bondo. Which, admittedly, makes things a bit more sketchy right now. I guess we'll see what happens.
 
If this were me, I'd pep and resin the basic form and use this to get the sizing right. Maybe a layer of fibreglass too. Then for all the horrible details, I'd just use foam. It's incredibly easy to detail foam, it sands reasonably well, you can carve a line and then it'll open up into a seam when you heat seal, you can sharpen a piece of PVC pipe and twist it through, etc. Bondo is best for compound curves that have to be freehanded with a file. Point being, there's no reason to use one single method here. Different materials are more suited to different purposes (like I need to tell an engineer that, but you get my point). Use each material where it excels.

What tools do you have access to in general? Can you find a laser cutter on campus? Is there an industrial or product design center with their own tooling? It's hard to gauge what direction to push toward when the setup appears to be "3D printer or bust".
 
HA! small world eh .Foxtrot I started on a massive goal this past Monday . To print a entire armor set .
I see you are on the same goal too! . Ive opted to do a total re draw in CAD and some other software I have.
( congrats on going in to Eng for college by the way! . I am a EE and I love it! .)
So far I have the helmet half done from when I started on monday..( STLs from thingaverse)

If you want I am gonna free open source my suits armor's STLs as I make them .
Some good programs to use for modeling besides CAD is Bryce by DAZ 3d .
its dirt cheap and It plays nice with OBJs and its toolset is a good way to do things that normal Eng CAD is bad at . ( also can let you add textures to a object ! good for that battle worn look . )
I have a flashforge Dreamer with a 9x6x5.9 ish bed . close in size to the makerbot you have access to.

As of now I have half of the helmet done via STLs from thingaverse . 20 hours in to it.
I est 150 or more hours of print time to do a entire armor set .

If you use ABS filament a quick way to get a worn look is ABS pipe glue . It adds a nice texture and make the print lines almost vanish.
for reinforcement I have planed to fiberglass the inside of the prints to stick them together and make them crack resistant ..
Peter D
 
entire 3d printed suit you say.. done that IMG_0987(2).JPG
 
If this were me, I'd pep and resin the basic form and use this to get the sizing right. Maybe a layer of fibreglass too. Then for all the horrible details, I'd just use foam. It's incredibly easy to detail foam, it sands reasonably well, you can carve a line and then it'll open up into a seam when you heat seal, you can sharpen a piece of PVC pipe and twist it through, etc. Bondo is best for compound curves that have to be freehanded with a file. Point being, there's no reason to use one single method here. Different materials are more suited to different purposes (like I need to tell an engineer that, but you get my point). Use each material where it excels.

While you're entirely right, and as time goes on, I'm seeing how much less labour intensive a foam build is, I originally settled on pepakura and fiberglass to get that rigid armour plate feel. I can't get completely away from foam, of course, and will eventually be using some of it for the undersuit.

What tools do you have access to in general? Can you find a laser cutter on campus? Is there an industrial or product design center with their own tooling? It's hard to gauge what direction to push toward when the setup appears to be "3D printer or bust".

At home, I don't have much more than a work desk and hand tools, as well as some old power tools like drill presses and a belt sander (there's a bandsaw in there, but it's currently in a fairly disassembled state and I think it might have some form of mechanical failure). At the University, I don't *currently* have anything. I should have access as early as this summer (I don't know if they'll let me in over summer, or if the place is even staffed) and as late as some time fall semester of this year. I know they've got a few forms of mills (CNC and the like, I presume) and I'm pretty sure there's at least one 3D printer in there (PLA or ABS, I couldn't tell you). I wouldn't be surprised if there was some form of laser cutter in there, but I really can't say with any certainty.

The problem I'm trying to avert is that the process of folding with pepakura is very time consuming and, I feel, kinda easy to **** up on detailed pieces. When I have the time I'm going to try to push through with it some more (this spring break I hope to finish the leg piece up, as well as repair my tester helmet and start trying resin and fiberglass on that). *If* there were a practical way for me to get access to a 3D printer, I'd be willing to use the printer even just to replace the pepakura process, filling all the details in with bondo as I would normally. Admittedly, I don't really know what I'm getting into with the likely high amount of sanding I'd have to do, but this just reinforces my point that the more man-hours I can streamline out, the better off I am.

HA! small world eh .Foxtrot I started on a massive goal this past Monday . To print a entire armor set .
I see you are on the same goal too! . Ive opted to do a total re draw in CAD and some other software I have.
( congrats on going in to Eng for college by the way! . I am a EE and I love it! .)
So far I have the helmet half done from when I started on monday..( STLs from thingaverse)

Yeah, as soon as I open my eyes to it I'm seeing more and more people who have, or are attempting to, 3D print a set of armour. And thanks, congratulations to you as well. EE's are a rare breed at my university - in the Eng department, we're mostly Mechanical, with some Aero sprinkled in.

If you want I am gonna free open source my suits armor's STLs as I make them .
Some good programs to use for modeling besides CAD is Bryce by DAZ 3d .
its dirt cheap and It plays nice with OBJs and its toolset is a good way to do things that normal Eng CAD is bad at . ( also can let you add textures to a object ! good for that battle worn look . )
I have a flashforge Dreamer with a 9x6x5.9 ish bed . close in size to the makerbot you have access to.

You definitely should do that. I can't guarantee I'd use them myself, but it's always good to have more resources floating about. I am definitely learning the shortcomings of AutoCAD, as compares to a regular 3D modeling program. I've worked with Autodesk Inventor, Creo Parametric, and SolidWorks, and I'm noticing there's some things they just can't do too well (as an experiment, this weekend I started attempting to model an NCR Veteran Ranger mask in Creo, and it's a real bitch, an absolute mess of reference planes and lofts, and I've barely gotten anything done). I might try and pick up Blender or something, since it's open source (if I'm not mistaken) and people tell me it's relatively user friendly.

As for the makerbot, I went down to the place Friday afternoon (the Makerbot belongs to a group based out of the library, called Build IT, not to be confused with the Engineering department's fabrication place, which I don't yet have access to and don't know what kind of heat they're packing) and it's definitely cool, but they currently have an absolutely hard limit on three hours per item. That might change in the future, but I don't think it will change before a better method appears.

As of now I have half of the helmet done via STLs from thingaverse . 20 hours in to it.
I est 150 or more hours of print time to do a entire armor set .

If you use ABS filament a quick way to get a worn look is ABS pipe glue . It adds a nice texture and make the print lines almost vanish.
for reinforcement I have planed to fiberglass the inside of the prints to stick them together and make them crack resistant ..
Peter D

See, this is the sort of thing I wasn't sure of. Things take a hell of a lot longer to print than I expected, which is fine, but does require me to reconsider my plan of attack. I do wish you luck in your endeavour, though.

I'm not sure if I'll be using ABS or PLA (Build IT uses PLA, because they don't have the ventilation to deal with ABS, but I don't know about the fabrication lab), but I'll definitely keep that in mind. I was mostly planning on adhering the various slices together while held together with clamps, and then just sanding the whole thing smooth, should that be required. That's assuming I have a photorealistic model, which I don't think I will. More likely it'll be akin to the pepakura I'm replacing and I'll still have to make use of the bondo.

And nintendstroid, I'm impressed. I wasn't aware there had been any fully-printed ODST suits yet. Kudos, looks good.
 
Last edited by a moderator:
yeah its all printed out only have to finish it. but commisions and some other great projects are making it take a backseat.
the cast helmets you see are from my 3d printed model odst so technically its also 3d printed.
the master is just way to heavy to wear.
Abs is not a great medium for printing large pieces. it wil crack while printing.
most of mine are all in one piece i have a 600mm printer.
i strongly advice you not to use blender but start using autodesk max. blender is free and lightyears behind. you can get a 3 year student licence for autodesk

current 3d printed project is this and the halo 5 ar plus a secret project i cant reveal yet10847425_445506155607990_1624449195067449407_o.jpg
 

Attachments

  • IMG_1204.jpg
    IMG_1204.jpg
    1.9 MB · Views: 263
If you can... get some PLAmax from octave systems . it's some of the best PLA out there .
I use ABS and I dont get cracks very often . I did have some lifting issues this week ! . a I had 2 10hour prints fail back to back . . so I have to raft stuff.
Here is what I did the first day/night ...
16257260284_19472d6caa_z.jpg




16673538350_f548330cfb.jpg

From STLs http://www.thingiverse.com/thing:158120
 

Attachments

  • 16257260284_19472d6caa_z.jpg
    16257260284_19472d6caa_z.jpg
    57.7 KB · Views: 212
Parametric modelling isn't really intended to work for large compound pieces like an entire helmet. It's designed to produce a series of parts that interlock in such a way as to produce a final product, as you would expect in engineering. So you would need to break down, say, a helmet, into its component parts, and model them all separately. All those seam lines aren't just for decoration! (Well, they are in this case, but you know what I mean.)

While you're entirely right, and as time goes on, I'm seeing how much less labour intensive a foam build is, I originally settled on pepakura and fiberglass to get that rigid armour plate feel. I can't get completely away from foam, of course, and will eventually be using some of it for the undersuit.

If it's good enough for Weta... http://instagram.com/p/0gZuk1Iq00/ :p You can also use Smooth-On Epsilon to apply a rigid coating over the foam, so that it's merely a substructure and the outside is hard resin like fibreglass. That seems like a best-of-both-worlds scenario I may look into in future, even for masters given how easy foam is to use.

At home, I don't have much more than a work desk and hand tools, as well as some old power tools like drill presses and a belt sander (there's a bandsaw in there, but it's currently in a fairly disassembled state and I think it might have some form of mechanical failure). At the University, I don't *currently* have anything. I should have access as early as this summer (I don't know if they'll let me in over summer, or if the place is even staffed) and as late as some time fall semester of this year. I know they've got a few forms of mills (CNC and the like, I presume) and I'm pretty sure there's at least one 3D printer in there (PLA or ABS, I couldn't tell you). I wouldn't be surprised if there was some form of laser cutter in there, but I really can't say with any certainty.

The problem I'm trying to avert is that the process of folding with pepakura is very time consuming and, I feel, kinda easy to **** up on detailed pieces. When I have the time I'm going to try to push through with it some more (this spring break I hope to finish the leg piece up, as well as repair my tester helmet and start trying resin and fiberglass on that). *If* there were a practical way for me to get access to a 3D printer, I'd be willing to use the printer even just to replace the pepakura process, filling all the details in with bondo as I would normally. Admittedly, I don't really know what I'm getting into with the likely high amount of sanding I'd have to do, but this just reinforces my point that the more man-hours I can streamline out, the better off I am.

Hmm, seems like for the next few months, pep is realistically your only option. Whether you want to do that though, I dunno, it seems like it might be a lot of effort for little to no payoff in the end. That said, an engineer should understand materials and how things go together. My practical skills come full circle and inform my 3D modelling and digital painting. Theory's all well and good, but good engineers and machinists understand the materials they're working with on a more visceral, fundamental level. In other words, it's worth getting your hands dirty even if you're not very good and in the end it's for naught, because it'll open up engineer-brain pathways you didn't know existed. 3D printing the basic structure is an interesting idea though, might have some merit. However...

Given that the engineering department is probably going to have both a 5-axis mill and a vacuformer, it makes no sense to waste lots of time and effort at this point 3D printing something fragile that needs a lot of finishing work, when in 4 months or so you can model and mill a wooden buck, draw a polycarbonate or ABS pull of that, and have your perfect, tough, lightweight set of armor that didn't take several months to model and print.

I think the allure of 3D printing is that it creates "something from nothing", but realistically it's only useful for "sketching" products and rapid prototyping, or finishing and moulding to replicate in a material more suitable to the task (ie. urethane). The impulse to use 3D printing for everything is strong when you have access to one, but every process has its place and FDM isn't replacing anything any of the others time soon. Stereolithography may have more potential given its higher resolution and increased material homogeneity (I'm assuming), but I don't know if the UV-cure resin is useful for much besides tabletop models.

Blender vs 3Ds Max... You can get the latter for free while you're a student, but I don't know how long that license holds. Getting screwed after you graduate would suck. They're both equally capable at modelling, but each workflow is more natural to different people. I have no issues doing fairly complex modelling with Blender after a few tutorial videos and it's capable of both parametric and polygonal modelling, though I haven't done much with NURBS yet personally.
 
Aw crap, absolute textwall status incoming.

Parametric modelling isn't really intended to work for large compound pieces like an entire helmet. It's designed to produce a series of parts that interlock in such a way as to produce a final product, as you would expect in engineering. So you would need to break down, say, a helmet, into its component parts, and model them all separately. All those seam lines aren't just for decoration! (Well, they are in this case, but you know what I mean.)

This I am definitely aware of. I've been trying to flesh out an NCR Veteran Ranger mask (I mentioned that somewhere, can't remember where) in Creo (I picked Creo because I technically have the most instruction in it), and it's been an absolute ****show. I think I might have decided to scrap that, though. More on that further down. On a related note, I'm replying to you over breakfast, and when I finish my breakfast, I've got to get back to fininshing the Arbor Press assignment in SolidWorks that's due tomorrow(!).

If it's good enough for Weta... http://instagram.com/p/0gZuk1Iq00/ :p You can also use Smooth-On Epsilon to apply a rigid coating over the foam, so that it's merely a substructure and the outside is hard resin like fibreglass. That seems like a best-of-both-worlds scenario I may look into in future, even for masters given how easy foam is to use.

I mean, at this point, the thing really tying me to a fiberglass build is the material I've already bought (though when you consider it, the only material I actually bought was the bondo, as the resin and matting were already laying around my garage due to a tub patch job not too long ago). That said, foam is usually more freehand work, and I don't think I'm comfortable working with it yet. I do intend to, at some point, resurrect my NCR Ranger vest build (that died in a fire due to the impracticality of moving in a bulky half-vest of 1/2in EVA foam, so I'd probably do a double layer of 3/16 or something similar, so it'll flex as needed) and I think at some point in the future I'd like to do a custom Spartan (think Jorge with a custom flair) and that's definitely something that's bulky and large enough that I'd be better off (and more comfortable) attempting out of foam.

Hmm, seems like for the next few months, pep is realistically your only option. Whether you want to do that though, I dunno, it seems like it might be a lot of effort for little to no payoff in the end. That said, an engineer should understand materials and how things go together. My practical skills come full circle and inform my 3D modelling and digital painting. Theory's all well and good, but good engineers and machinists understand the materials they're working with on a more visceral, fundamental level. In other words, it's worth getting your hands dirty even if you're not very good and in the end it's for naught, because it'll open up engineer-brain pathways you didn't know existed. 3D printing the basic structure is an interesting idea though, might have some merit. However...

Yeah, you're really right here. And I am going to get back on the pep. A large part of my avoidance has been the travesty that was attempting to make the shin piece pretty much right off the bat. It's damned intricate, and actually really large. I might skip that altogether for now. Instead I think I'll finish up the tester helmet (it's suffered some damage, and was missing some pieces) and continue to use that as my testbed, and then maybe work on the bucket proper. That, or the big chestplate. Either of these are items that, if I move forward with, I can really start to see the fruit of my labour and get some morale going, which I'm going to want over the summer if I'm going to get anything done.

As for the idea of 3D printing the basic structure and filling it out, that's why I decided to stop the work on the NCR mask. Unless I get around to modeling it in another program in higher detail, that's probably how I'll go about making the mask (and, as you say, it's good experience, as that's a methodology I'd now like to see the results of).

Given that the engineering department is probably going to have both a 5-axis mill and a vacuformer, it makes no sense to waste lots of time and effort at this point 3D printing something fragile that needs a lot of finishing work, when in 4 months or so you can model and mill a wooden buck, draw a polycarbonate or ABS pull of that, and have your perfect, tough, lightweight set of armor that didn't take several months to model and print.

I think the allure of 3D printing is that it creates "something from nothing", but realistically it's only useful for "sketching" products and rapid prototyping, or finishing and moulding to replicate in a material more suitable to the task (ie. urethane). The impulse to use 3D printing for everything is strong when you have access to one, but every process has its place and FDM isn't replacing anything any of the others time soon. Stereolithography may have more potential given its higher resolution and increased material homogeneity (I'm assuming), but I don't know if the UV-cure resin is useful for much besides tabletop models.

You're right in that once I actually get into the lab, I'll have a lot more potential solutions available to me, and I'll definitely have to explore those, as there's no way I'll pep out this whole damn thing before I get into the lab. And I am going to trudge on with the pep, since it's my only option right now.

You're also right in that is definitely the appeal of 3D printing, but it's definitely a question of what you're making. Making small batches of anything plastic can be very expensive, because moulds are really expensive, whereas 3D printing allows for the creation of even a single object. I met a guy the other day who had a 3D printer that was mostly comprised of 3D printed parts, brackets and such. About half of them were printed by that very same printer as he continued to improve upon it. I think the technology is also going to see a lot more maturity as we get better at various metal deposition/laser sintering techniques. Currently, any 3d printed metal parts have to go through a bit of an intensive heat treating process, but I know companies like SpaceX have been using them for stuff like rocket nozzle cones (something about having a guarantee that the inside of the part is metalurgically homogenous, so you don't have to make some just to cut them open and assume the rest are good).

Blender vs 3Ds Max... You can get the latter for free while you're a student, but I don't know how long that license holds. Getting screwed after you graduate would suck. They're both equally capable at modelling, but each workflow is more natural to different people. I have no issues doing fairly complex modelling with Blender after a few tutorial videos and it's capable of both parametric and polygonal modelling, though I haven't done much with NURBS yet personally.

I'll probably go with blender just because that open source community is going to be better at offering tutorials to completely new modelers. Any time I try to look up something simple about Inventor or Creo or something, it's impossible, because the entire community is a bunch of trained professionals with reference books getting dusty on their shelves. It's nice to have access to programs that normally cost thousands of dollars to license, but it's not nice to rely on them - as you said, especially once the student liscenses run out. I think I have a copy of Maya on my computer somewhere, but that had to do with trying to convert Autodesk .ipt files to .obj or something for Space Engineers (a process I never saw through because I have too many competing interests).
 
Aw crap, absolute textwall status incoming.

I like it when people have considered a topic deeply enough that they're able to have a lengthy, detailed conversation on it.

This I am definitely aware of. I've been trying to flesh out an NCR Veteran Ranger mask (I mentioned that somewhere, can't remember where) in Creo (I picked Creo because I technically have the most instruction in it), and it's been an absolute ****show. I think I might have decided to scrap that, though. More on that further down. On a related note, I'm replying to you over breakfast, and when I finish my breakfast, I've got to get back to fininshing the Arbor Press assignment in SolidWorks that's due tomorrow(!).

I bet, I can imagine that's not the easiest model to start with, for sure. Not that I'm one to talk about over-ambitious starting projects.

I mean, at this point, the thing really tying me to a fiberglass build is the material I've already bought (though when you consider it, the only material I actually bought was the bondo, as the resin and matting were already laying around my garage due to a tub patch job not too long ago). That said, foam is usually more freehand work, and I don't think I'm comfortable working with it yet. I do intend to, at some point, resurrect my NCR Ranger vest build (that died in a fire due to the impracticality of moving in a bulky half-vest of 1/2in EVA foam, so I'd probably do a double layer of 3/16 or something similar, so it'll flex as needed) and I think at some point in the future I'd like to do a custom Spartan (think Jorge with a custom flair) and that's definitely something that's bulky and large enough that I'd be better off (and more comfortable) attempting out of foam.

Foam doesn't have to be freehanded in the slightest if you don't want it to be; you can use guides and templates for the entire thing, if you want. I think I'm going to do a Halo 3 Marine top half (don't need the hips down for what I'm doing) by converting pep files to foam. Especially for the helmet since compound curves are hard, but foam can be encouraged with heat a little too. Might not need a starting template for the other pieces, I'm not sure yet.

Yeah, you're really right here. And I am going to get back on the pep. A large part of my avoidance has been the travesty that was attempting to make the shin piece pretty much right off the bat. It's damned intricate, and actually really large. I might skip that altogether for now. Instead I think I'll finish up the tester helmet (it's suffered some damage, and was missing some pieces) and continue to use that as my testbed, and then maybe work on the bucket proper. That, or the big chestplate. Either of these are items that, if I move forward with, I can really start to see the fruit of my labour and get some morale going, which I'm going to want over the summer if I'm going to get anything done.

Sounds like a plan to me. Try to get some Bondo practice in too. That stuff's awesome, and so easy to work with.

You're right in that once I actually get into the lab, I'll have a lot more potential solutions available to me, and I'll definitely have to explore those, as there's no way I'll pep out this whole damn thing before I get into the lab. And I am going to trudge on with the pep, since it's my only option right now.

The half-finished pep should provide a good testbed for various techniques though, like you say. You can practice everything from resin to paint on some realistic geometry.

You're also right in that is definitely the appeal of 3D printing, but it's definitely a question of what you're making. Making small batches of anything plastic can be very expensive, because moulds are really expensive, whereas 3D printing allows for the creation of even a single object. I met a guy the other day who had a 3D printer that was mostly comprised of 3D printed parts, brackets and such. About half of them were printed by that very same printer as he continued to improve upon it. I think the technology is also going to see a lot more maturity as we get better at various metal deposition/laser sintering techniques. Currently, any 3d printed metal parts have to go through a bit of an intensive heat treating process, but I know companies like SpaceX have been using them for stuff like rocket nozzle cones (something about having a guarantee that the inside of the part is metalurgically homogenous, so you don't have to make some just to cut them open and assume the rest are good).

True, for certain specific objects of a particular size, complexity of geometry and length of production run, there's an advantage to 3D printing. And yeah, the aims of RepRap are admirable and getting increasingly attainable as the abilities of the technology progress. However, I'm inclined to say there's a fairly vast gulf between a MakerBot and the type of DMLS machine I assume SpaceX would have. The abilities of one can't justify the use of another, despite their relationship to one another, but judged on its own merits... And right now, standard FDM is still wanting for many higher-intensity practicalities. Carrying a repair kit for it might be tricky too. :p

Open hardware, motion control and home manufacturing are one of my "things", so I'm not down on the tech. But it's important to be realistic. Issues that are irrelevant at small scale (like, say, torsional rigidity) can become much bigger problems at full size. That's not to say you couldn't, say, model in some channels for a bit of alu bar or something, and besides, nintendstroid already demonstrated that it's achievable. But it's one thing when you buy a $1200 3D printer that you have 24-hour access to at home and another when you have access to a million-dollar manufacturing facility as part of your studies, y'know? Since you're going to have relatively limited time set aside from your studies and whatever social stuff they like to foist onto freshmen there, it's probably important to have clear direction on where you're spending that time. That said, you do have 4 years ahead of you, so it's not like there's a huge hurry.

I'll probably go with blender just because that open source community is going to be better at offering tutorials to completely new modelers. Any time I try to look up something simple about Inventor or Creo or something, it's impossible, because the entire community is a bunch of trained professionals with reference books getting dusty on their shelves. It's nice to have access to programs that normally cost thousands of dollars to license, but it's not nice to rely on them - as you said, especially once the student liscenses run out. I think I have a copy of Maya on my computer somewhere, but that had to do with trying to convert Autodesk .ipt files to .obj or something for Space Engineers (a process I never saw through because I have too many competing interests).

The Blender community is awesome. Anything you want to find out is easily available. There's a BlenderGuru on building a spaceship corridor interior or something that you might be interested in looking up, as far as an introduction to modelling goes. It goes over most of the important stuff. I hear you on the competing interests!
 
Status
Not open for further replies.
Back
Top