Question about resolution

Hi all, I'm new and wanted to preface my question by exclaiming how awesome reprap and its community is. I am really excited about the possibilities and implications of personal fabrication.

I've never used a reprap before or seen an object made by one. They seem from some pictures to have lines and a certain texture to them. I have seen a lot of development threads about exciting topics like making more parts reprappable, recycling common plastic, turning reprap into a 3d scanner etc. but I haven't seen anyone discussing increasing the potential resolution of reprap. One reason I asked is I wanted to create a creative commons licensed boardgame and sell pieces and thought it would be cool to provide the .stl files for people to print their own if they had the desire but I think the pieces may contain too much detail.

Am I wrong in my assumptions as to the textural quality of reprapped objects? What is the resolution like now? how small a detail can you achieve? Could you for instance print something like this:



about 3 inches tall with a one inch base? Is molding always gonna be a better solution?

I just can't shake the idea of how cool it would be to have wargames like warhammer with reprapped pieces so you could actually play such games without having to fork over hundreds of dollars to games workshop et al.

It's not so much a question of resolution, but rather build method. you can get higher resolution by using a smaller orifice and higher gearing. anything hanging in air, like the scythe, is hard with thermoplastic extrusion. it would require lots of support structure.

A powder bed machine would be much better at building these types of models.

You will also fork hundreds of dollars over to your reprap machine resource supplier.

If you want an army of zombies to do your bidding, I'd perhaps think about making a mold in silicon from one non-copyrighted design.

I hate to be a discouraging word, but we're not really at the cheaper than mass produced point yet.

I think the UP! 3DP is nearly capable of that level of print quality. Most RepRaps haven't gotten there yet, but might well soon.

Molding will probably be better, though I do like the idea of being able to print figurines at home to replace missing or broken ones.

One thing you could do is print rough blanks. They would be in the general shape of the figure, but you (or the user) would have to carve out the details. the plastic is easy to cut; if you are a good whittler it would be easy. (actually, it would be a lot of work, but at least the user has the option.)

Another idea: Print moulds (front and back) in plastic and then get the user to fill them with clay. Each clay figurine would need attention (eg: cleaning up, details, etc). The user would need to fire them to get make them hard, but the plastic doesn't need to be involved in that process. Once fired, they can be painted however you like.

BTW: The idea of having all the figurines in STL (or at least in a 3D format transferable to STL) makes sense anyway. The technology is still evolving, and what you design now could be directly printable in a few years. It also might be possible to use the same original 3D models to produce a 'virtual' board, with the ability to share that board with other users (eg: GM and players), in an interactive/online way.

I see mendel supporting support material soon. ish.

http://ifeelbeta.de/index.php/support

http://reprap.org/wiki/Mendel_Multiple_Extruder

Texture-wise, imagine you're making the object from slightly bendy 0.3 mm or 0.5 mm pencil lead. Powder printer will be better, but that's ~1-2 year vaporware right now.

I am a massive moldmaking geek, but casting-moldmaking is its own discipline. It is not something one can learn overnight, and we can print stuff which is non-trivial to mold. Like a pair of hands, fingers loosely interlaced. Or a quadruped.

One reason I asked is I wanted to create a creative commons licensed boardgame and sell pieces and thought it would be cool to provide the .stl files for people to print their own if they had the desire but I think the pieces may contain too much detail.

Sounds fun. As a lark, or a serious entrepreneurial venture? There may be the corner case where the game takes off but everyone prints the parts and the boards rather than buying from you.

In which case you'd support it via "merch" like t-shirts.

Also, do you have any stl files you'd like to stuff into the wiki?

---

-Sebastien, RepRap.org library gnome.

---

Remember, you're all RepRap developers (once you've joined the super-secret developer mailing list), and the wiki, RepRap.org, [reprap.org] is for everyone and everything!

Well, I don't know if you guys have played Wargames much but the following warhammer miniatures cost $72 + shipping. Call the tank 5, and the rest of the figurines still cost more than $2.50 each, not counting shipping, and that's on sale.



These 5 legs cost 10 bucks online:



They charge a lot for these miniatures and while I obviously don't have much experience 3d printing, seeing videos of people printing new reprap parts I cant imagine each of those widgets costs 10-15 bucks in plastic. Correct me if I'm wrong but I think that the base plastic is cheaper than those figurines. Again, I could be wrong but if the raw plastic is that expensive then it seems it will be a continuing problem in general.

Models could be made such that they would be easier to print in 3d, so issues like the scythe could be avoided.

In my searches, I found this article which I thought was relevant to the conversation:

[fabbaloo.com]

It reminds me of Multiresolution modifiers and Ptex Texture Mapping in Blender. The basic idea being that you could apply a technique like that in the same way, using a larger stream of plastic and printing quicker for areas with less detail and using less/ printing more slowly for fine details, so areas without detail wouldn't need to be printed slowly. A bright coder could probably even set it up such that the software could gauge how much detail a certain area had given the number of vertices in that area of the model. Again I'm just talking out of my ass, but it seems it takes a lot of talking about stuff that is stupid and wont work before anyone stumbles upon a good idea.

The raw plastic granules are very cheap - an injection molding machine and creating the proper molds for the parts are not!

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

well I was referring to a comment made earlier that even ignoring resolution, 3d printing miniatures would be too costly. Their comment was actually in favor of molding over 3d printing at this point in time.

If you just want to replicate those figures, you probably better make a silicone mould from them and fill these moulds with polyurethane or epoxy. The printing resolution of RepRap machines isn't that great for fine details.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

What about commercial 3d printers like Stratasys? Do they have the resolution to print this?

What about a 5 axis mini milling machine? Like this [www.sherline.com] , it cost 1225 $. Has anyone any experience with milling?

As for the cost, the plastic filament costs 45$ for 5 lbs, natural, coloured a little more [store.makerbot.com]

That's about 20$ / kg

What's the weight of the miniature? 20 grams (a random guess)? that's 40 cents to print plus some energy, a minor cost.

I would checkout Shapeway's see [www.shapeways.com]

Not as cheap as if you have a RepRap but you could at least try a figure or two without investing a large amount
of time and money

Keith Sloan

Autarkyboy Wrote:
-------------------------------------------------------
> What about commercial 3d printers like Stratasys?
> Do they have the resolution to print this?
>
> What about a 5 axis mini milling machine? Like
> this [www.sherline.com] , it cost
> 1225 $. Has anyone any experience with milling?

... with proper separating the figurines you can go away with a 3 axis mill - i did some moulding with chocolate ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Autarkyboy Wrote:
-------------------------------------------------------
> What about a 5 axis mini milling machine? Like
> this [www.sherline.com] , it cost
> 1225 $. Has anyone any experience with milling?

I have Sherline's 8-direction mill. It's only a 3 axis machine. The other directions are for complicated setups; they can't move while it's machining. It also can't handle fine detail as-is; there's too much backlash since it was originally designed to be a manual machine. The best upgrade I've seen to address backlash replaces all threaded rods, nuts, and sliding surfaces.

Autarkyboy Wrote:
-------------------------------------------------------
> What about commercial 3d printers like Stratasys?
> Do they have the resolution to print this?
>
> What about a 5 axis mini milling machine? Like
> this [www.sherline.com] , it cost
> 1225 $. Has anyone any experience with milling?

The five axis mill is only part of the puzzle. Even if the listed machine was 5 axis (it is 3) it does little good if the machine is not fitted with CNC controls. Lets set that aside, and figure you could cobble together a little 5 axis machine by converting a Sherline 3 axis (I'd use the 5400 series as many have) and adding a 2 axis trunion table. If you know what you are doing a system that could cut machinable wax or plastics would get into maybe 3-4k, including the computer to run it, more if you don't know what you are doing .
Now you need to have CAM software, which is not cheap, especially for simultaneous 5 axis. And it needs a post processor matched to your machine, or you can write it yourself, which can be "dicey". Then you need to learn to use 5 Axis CAM. Despite many illusions that there is a "load solid model into the program, 2 clicks and you have finished parts" CAM programming takes a lot. 5 Axis guys can make upwards of 100k, and for good reason.
The very allure of a printer like the RepRap is that you DON'T have to do all the intermediate steps to get from a design to the finished product. Solid modeling is easy compared to CAM programming.

Autarkyboy Wrote:
-------------------------------------------------------
> As for the cost, the plastic filament costs 45$
> for 5 lbs, natural, coloured a little more
> [store.makerbot.com]
>
> That's about 20$ / kg
>
> What's the weight of the miniature? 20 grams (a
> random guess)? that's 40 cents to print plus some
> energy, a minor cost.



Yea, that was my hunch. If they could actually be printed it would be rather simple to make 3d printed board and wargames.

I guess it's just out of repraps capabilities at the time. I was noticing the stark difference in quality between these two prints, and was wondering if someone with more expertise could explain why the UP! printer has prints that are so much cleaner and is this a concern for reprap developers? To improve the overall quality of the prints.

Reprap:



UP! (I'm guessing because it was done by the pp3dp folks)




I just look at it like I look at everything open source, it takes a little longer but eventually it matches or surpasses proprietary rivals. I really am curious to know what it is that makes the prints so much cleaner. Granted it's loads more expensive, but besides the extruder and the motor I can't really see what could be improving the quality.

The top picture has strings on it because the extruder is not reversing properly at the end of a filament run. Judging by the background it was printed on a Makerbot which used a DC motor for the extruder. These can't reverse quickly and accurately as steppers can. My Mendel does not leave strings.

The towers are wobbly at the top because they got too hot. A fan or some delays would fix that.

The UP printer prints in a higher resolution which requires a smaller nozzle. That is made easier because they use smaller filament like commercial printers. We use 3mm welding rod, which is more readily available and cheaper. It wouldn't be hard to make a Mendel extruder that worked with the UP plastic, just make the barrel smaller bore.

---

[www.hydraraptor.blogspot.com]

As well, the UP print has wobbly towers too (although less so). The wobbling problem, as Nophead says, is because the prints are too hot. Those peak layers are small so they're finished very quickly, before they can go down. I'm pretty sure the newer firmware waits for the layers to cool before it continues.

very cool to know. I can't wait to get my hands dirty with this stuff.

I have seen people making their own glass nozzles, because apparently there's a problem with what was being used before. Would this problem (sorry I dont remember the exact issue but I'm sure you know what I'm referring to, something about heat loss) make it impossible to do what I mentioned some ways above? Namely extrude the normal amount of plastic for areas with less detail and extrude less plastic for areas with more detail, somewhat like a multiresolution modifier in a 3d program? So you could use the thicker filament, and still make parts and blockier less detailed objects more quickly, but flexibly use a higher resolution for areas where it's called for?

I'm one of those people tinkering with glass nozzles, but I don't think there's any problem with what was being used before. The state-of-the-art nozzles right now are brass or stainless steel with teflon lining, and these work great - they can also do what you're talking about (that's more of a software than a hardware challenge). I'm experimenting with glass because I see it having some potential advantages over metal, like being able to inspect conditions inside the melt chamber, having a sharper transition zone, and being able to be drawn to a fine smooth point instead of drilling.

As i understand it, if you run the extruder faster than nomal, the filament will swell after exiting the nozzle orifice. So, a .5 mm filament can swell to .7 mm.

it's not a lot of room to play with, but taking advantage of the swell for infill should give a couple percentage points decrease in build time.