RepRap 3.0

ok, when it comes to mendel and multiple print heads its probably going to start with paralel heads with bowden mechanisms. and a good support material has been found in using pla as the support and abs as the build material. nophead and erik i believe did some great experimental work with this just before christmas. right now im doing work with my makerbot to install dual heads but having issues with the interface and geting the makerbot firmware to recognize multiple tools. which is probably more my fault than anything. as for a powder printer, ive been visualizing a darwin modified to be a powder printer by altering the print bed to be a two part object with a powder resevuar taking up much of the half of the current bed that the print head is not able to reach by design. then either using a print cartrige (aka hp style) or a modified paste extruder to deposit the liquid binder into the powder.

to be honest its something im working on but only theoretical at this time.
as for metal sinstering. ive been working on ebfff hardware some also and i believe that it would be posible to make a home made eb gun safely that MIGHT be usable in a metal powder sinstering aplication.

but then again someone else may get to each of these projects before i do which would be fine with me.

ive looked over the requirements that this award proposes and I do not see any of them being either imposible or even unlikely in reality. One of the greatest gifts that the makerbot has given the reprap comunity has been a solution to the chicken and the egg problem. In order to make repraps you need a reprap, so how do you get more repraps out there? make more repraps. but you need repraps to reprap to reprap etc.... and while a makerbot is not optimal for printing out a mendel it is capable of doing it, and if people are able to take the mini mendel idea and actualy create a functional printer then it will make a makerbot even more able to solve the problem since makerbots biggest problem is some of the mendel parts are right on the edge of their build zone.

If we want to get serious about this than a dedicated wiki and perhaps blog are in order...then its time to start talking rig design...this non-trivial for laser sintering for sure as laying the material for each layer at a uniform thickness will be very challenging...along with the other issues you mentioned.

We haven't enabled blogging yet, but you are very welcome to use the wiki for your stuff, including research notes, files, images, and so on. It helps cross-integrate with everything else if it's all in one 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!

goinreverse Wrote:
-------------------------------------------------------
> The problem for me is installing chrome/chrome
> frame on tons of different computers and OS's,
> just not something I have time/energy to deal with
> right now. I have used wave, seems like a lot of
> people say the opposite, that etherpad has some
> things they are waiting to be added to wave...
> YMMV.

I would like to talk to the guy who thinks etherpad is anywhere near wave...i would put etherpad in the category of reprap software

maitri982 Wrote:

> I would like to talk to the guy who thinks
> etherpad is anywhere near wave...i would put
> etherpad in the category of reprap software

It is so bad in fact that I can't see myself using it. I have started a google wave for this. If you want to be invited to the wave just send me your email kcramer at sapiosciences _dot_ com

... for lasers: - i think i have found a really cheap source for many pigtailed IR-diodelasers with 2.5Watts output power, but have to dispute the terms.

What's needed more is the power-source (i use parallel LM317 in current driving mode), one or two lenses and some tubes for adjusting and fixing the lenses.

But the most serious task is the security - i use laser-goggles when working with my lasers, but they'll cost around 300 USD, so not really common ...

How can i guarantee when selling the laser-kits that nobody will be harmed when using the laser or don't make nonsense with?

Viktor

I have compiled a list of plastics an there properties including a bar chart of melting points...see attached PDF.

I have not seen mention here of PVC, but it appears to have the lowest melting point of the common plastics.

Actually PVC has a high melting point, what you have is 80C, which is the glass transition.

---

[www.hydraraptor.blogspot.com]

thanks naphead...

FYI...my source was here:

[www.dynalabcorp.com]

but here shows quite a broad range of melting points from 100-260C:

[books.google.com]

They also say the fusion temp of unplasterized PVC is 160-170C, which is perhaps the numbers we would be most interested in.

also posted on google wave...repeated here:

regarding the structure of our machine, why can't it just use something like this:

[www.onlinemetals.com]

This should be plenty strong and it is cheap. It can serve as the frame and also as the guide rails for the x/y head movement. I may go to Lowes here and just see what they have in stock that I might be able to use.

Quite interesting article is here: [www-rcf.usc.edu]

It is quite clear from it how to layer material. Also selective inhibition of sintering is interesting with XY head - you can only trace outline, no need to scan entire volume.

I don't think that material will have the finish and tolerances needed. Also as mild steel it will rust, distort and is generally not easy to make into a square cage. For high accuracy printing we are talking about getting to tolerances under hundredths or even into thousandths of an inch, that means the planes and gantry need to move predictably on those tolerances. Something like the mcwire has a tolerance maybe to a couple tenths of an inch, mendel properly tuned I think can be accurate to within perhaps 5 hundredths, I would really like to see 1 hundredth as a goal for this machine. I am not even sure if the extruded aluminum channel I am waiting for from mcmaster will be straight and finished enough.

The major costs end up being how to do you move the planes along the axes, you can do 2 paralell shafts as in the makerbot but you end up having to use relatively expensive high precision shafts and then build a carriage with linear bearings/bushings to traverse along it, it makes construction more complex because you really need to laser cut the parts to have the shaft alignment be accurate enough. The potential advantage of the extruded t-slot channel (or perhaps strut channel) is that there are off the shelf bearing guides that are relatively cheap and because the channel is straight within x tolerances that carries through to the bearing guides.

I have a sketchup I am working on, and will post to etherpad, Will take a couple more days to complete.

Prior to looking up these rods i looked up t channel...the only thing that turned me off to it was the pricing...its rather salty...but probabyl requires less other connectors so maybe its a wash with rod based structures...

i read the paper...I actually like that method *a lot*.

it does remove many of the issued of dealing with lasers. Using printer heads is also straightforward I would think.

if any others here want to join this discussion on google wave...please send your email to me...

Not sure if I have seen bothacker in the forums but his design here is the same as the t-slot version I have as gantry-only (rather than cube):

[bothacker.com]

8020 looks like a good source for aluminum rail in addition to mcmaster. Some parts mcmaster has 8020 does not like the 2 and 3 way corners, some things 8020 like the raw bearing pads.

nice drawing and design...much more straightforward.

I note that he is not using t-slot for the guide rails for the gantry, but a rod of some sort. I assume you are using t-slot + roller bearings per your prior comments.

Also, this design will work for FDM, but not for laser sintering or another sintering process.

After looking at the 8020 system (8020.net) in detail I don't have any interest in a mendel or mcwire and pretty happy to just be done with that, For 70% the cost of a makerbot I can put together a 2'x2'x2' device. That is without a doubt the best system I can possibly conceive of for doing a wide range of things including this. I am going to be focusing my energies on building an FDM with that and with the plan to explore SL once that is working in a rock solid way. Best of luck to all the reprappers out there. I will keep posting in the forum on extruder updates and electronics from time to time, But I think I definitely need to fork a new project, no more printed parts!

i looked at their catalog here: [8020.net]

they have very nice stuff indeed. Even a whole linear motion catalog with special parts for heavy duty cycles like I guess you would need for FDM.

Best of luck!

But the most serious task is the security - i use laser-goggles when working with my lasers, but they'll cost around 300 USD, so not really common ...

How can i guarantee when selling the laser-kits that nobody will be harmed when using the laser or don't make nonsense with?

Dig up something like this, tack on warning text, print labels with your laser printer.

Label the box and the plastic bag the laser kit comes in.
[www.myconfinedspace.com]

Also, 'upsell' the glasses as an add-on or suggested extra with the laser kit in your online store.

Cut and paste the image and warning text into a few of the wiki pages that have the build instructions for the laser system. Once you've done that once, printing more labels, and pasting text into webpages is low-effort.

---

But I think I definitely need to fork a new project, no more printed parts!

goinreverse, maitri982 would you be interested in mirroring your working notes or document your build on the wiki? I think it would help a lot of people. Or would you prefer to do it all in another space and then cut and paste it all in when it's finished?

We've got gigabytes and gigabytes of space in the wiki for guys like you. Also, let me know if you need a dedicated forum for Extruded Aluminum-based RepRaps.

---

-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!

Sebastian,

I am still working out what media and formats I am going to be documenting with but I have no problem whatsoever with it appearing on the reprap site and will most likely license it under creative commons SA which I think is compatible with repraps documentation license.

There should be a pretty foolproof 8020 repstrap coming from me forthwidth. I am talking to them about making it into a kit (they offer that as a service for pretty nominal fees) that you can just direct order through any of their distributors and it comes flat-packed, pre-cut, etc in 2-3 days flat. Just add water, err electronics and motors.

As a side note they are outstanding folks over there at 8020, one of the easiest industrial companies I have ever had the pleasure of working with. I should say I work for a business that is likely to do some large volume production with them on other things but my guess is that they would be just as nice to anyone off the street. If you are in the West US I would also strongly recommend copper state nut & bolt (an 8020 distributor).

I am still working out what media and formats I am going to be documenting with but I have no problem whatsoever with it appearing on the reprap site

Excellent!

You may want to start mirroring early or immediately, or it becomes a real chore to duplicate onto the RepRap wiki, plus you start generating mirroring and versioning issues.

and will most likely license it under creative commons SA which I think is compatible with repraps documentation license.

Hmmm ... That is probably so. You may want to release your work under 1) CC-SA, and 2) GPL. The GPLness helps make it more cross compatible with all the RepRap stuff. I tend to go pure-GPL for that reason.

License-discussion quickly turns into license-talking-past-one-another and then license-flamewar which quickly gets moved by me into the Library forum where I can derail it into 'Let's make the RepRap website better and take over the world!

I'm glad to hear that 8020 (and the other extruded-aluminum "ExAlRap?" folk, presumably) are so organized. This means that once folk have developed and documented, it will be super-easy for you and other folk to do up and sell a 'ExAlRap' vitamin kit.

Personally, I need to design, do, and document Eiffel and RBS before I can get into 8020.

Hi Sebastien,

... as i already stated in the German RUG the diodes are mostly 5Watts or connected in pairs for 10 Watts output at a 100micron-glassfiber, so a really dangerous 'toy'

When i have the first kits functional, i'll try to start local and check the security and liability issues with selling class-IV-lasers as OEM-modules.

I think the liability can be delegated with warnings and disclaimers, but i have to secure too no underages will get them, or i'll be in trouble either

Viktor

I'm glad you're looking into it. Try these guys:
[www.eff.org]
or use google to find up a local German electronic hacker rights or information must be free! group that's got a lawyer bolted on.

Happily no one has make a RepRapped handgun or RepRapped self-guided RC plane with "interesting" payload yet, so I don't have to think about these issues. (I'll bury my head back in the sand like an ostrich now rather than thinking about it.)

... When you do find something out, please paste in the response or make a note in the library forum to the german language mailing list discussion. (In german, of course.) RepRap is a global project so we have to think about localization everywhere. You may want to tack on "Deutsch:" in the title if you do post in Library forum in german, but that's only if it is convenient. If the subject is
DonaudampfschiffahrtsRepRapLaserLegalQuestiongesellschaftskapitän I'll probably figure out that the text will in German regardless of a note in the title.

---

Viktor, it may make sense to just write some stuff up in German language but in a dedicated RepRap laser forum. Otherwise it will start to get really difficult .001 to 12 months from now when we've got multiple different languages all doing RepRap laser stuff.

Just tack on "Deutsch:" in the title somewhere.

... Or let me know if you need a dedicated Deutsch-laser forum.

All this are random just-thought-up ideas. Write it up wherever it makes the most sense to you. I'm serious about Deutsch-laser, Deutsch-Mendel, Deutsch-Whatever, Deutsch-VDXBot#15, as discussion outgrows the current Deutsch forum.

... i'll post some infos at the wiki when the lasers are working and i have some related images.

I think the start could be in English ... in our German CNC-forum we have some serious disputes regarding lasers and security, but most real work is done beside the forum in P2P-networks and small 'closed' groups with 2 or 3 persons, so this actually isn't a point to worry about legal stuff.

But this will change rapidly, when the kits are in the wild ...

Viktor

Viktor,

Just post the parts list and specifications in detail. That way individuals can acquire the parts completely independently of you and you won't incur any liability. Mouser, Digikey, Alltronics, Mcmaster all sell heaps of dangerous parts that when put together can become even more dangerous, there isn't a big issue because they sell it as parts (also for example makerbot sells unassembled parts). Just search instructables and you can find about 10,000 things more deadly than what we are talking about here where the primary risk is blindness or fire (though the extruder is a major fire hazard itself, probably more so).

Wickedlasers here in the US has fair pricing for reasonably high quality laser eye protection.

That fact that we all already play with suspect electronics connected to 900F nichrome wire, I don't think I really see a laser as being any more dangerous than that. I think the main danger is that people think if they can't see it it won't hurt there eyes which isn't true and as long as that info gets out there accidents can be avoided.

Just post the parts list and specifications in detail. That way individuals can acquire the parts completely independently of you and you won't incur any liability.

I think he's planning to.

But if he can make a profit selling kits as well, that that's even better for him and for us, because it will fund his research, while at the same time, the blog, wiki, and forum provide publicity, documentation, and support for him, because that's the blog, wiki, and forum's job.

... This means it's the forum's job to figure out the laser safety stuff for the wiki so developers like Viktor can concentrate on developing and documenting.

Here's a stub page for people who want to work on this.
[objects.reprap.org]

Sebastion Et Al,

I think a separate section of the wiki is a great idea. I am enthused about some of the directions things are taking.

I had thought of an 80/20 prepacked as well, but perhaps now I don't have to worry about it thanks to goinreverse. The nice thing is there will still be room for people to innovate within whatever design we start with using extruded T's.

As I previously stated, lasers provide great promise to solve several key FDM problems, so I am glad to see progress on that front as well. The sintering inhibitor process is also interesting as it may achieve similar results with no needed laser.

If some of these designs can be multi-tooling...even if its just:

* High accuracy FDM
* CNC
* Pick and Place

Then you may have a device that can have a real impact on the world in a positive way. if we get lasers working too then that would be a real coup.

Of course each step presents difficult challenges, but getting a good solid structural design of T-channel with solid electronics should accelerate needed improvements as this is very flexible framework to work within.

Best,
Kevin

... some time ago i saw a report where airplane-wings were made from carbon-fabric: - first cut the exact contours, then sewed the parts together, inflated, impregnated with 2K-epoxy, cured and last trimmed/milled for fitting.

Has anyone thought a similar strategy for building a solid frame?

Idea: cut the parts of the base, frame and supports from watertight fabric, sew or glue the borders together and fill it with some hardening material - e.g. 2K-epoxy, concrete or a sand-epoxy-mixture ... for better stability/fitting you can insert some rods or other static parts prior to infilling ...

You can inserting some stiff plates and fixatures for mounting the additive components or mill the fitting areals after curing ...

Viktor

Kevin,

You may enjoy having a look here and entering your own wishlist / research pathway.

[objects.reprap.org]
[objects.reprap.org]

---

... some time ago i saw a report where airplane-wings were made from carbon-fabric: - first cut the exact contours, then sewed the parts together, inflated, impregnated with 2K-epoxy, cured and last trimmed/milled for fitting.

Has anyone thought a similar strategy for building a solid frame?

I've been thinking about an Eiffel / RBS PourStrap with large [[RBS/Beams]] of steel or sheets of acrylic to make the mold, and using food-type plastic wrap as a release element.

Or laser cut sheet for molds. Or CNC router made molds. I think about molds a lot. And self-replication a lot. That's why I like the idea of using [[RBS/Beam]] for as a mold to make [[RBS/Beam]]. Or using [[RBS/Beam]] and large sheets of acrylic to make a [[PourStrap]]. Or using a [[Laser Cutter]] to make a [[FlatPack PourStrap]].

You can inserting some stiff plates and fixatures for mounting the additive components or mill the fitting areas after curing ...
I'd definitely go with with embedded fixtures or RBS-holes. Milling is a bootstrapping problem. (The RepRap wiki doesn't understand milling yet - [[Spindle]] and [[Dremel Holder]] are currently empty wiki pages.)

Edited 2 time(s). Last edit at 02/12/2010 01:44PM by SebastienBailard.

Hi Sebastien,

... not moulds ... i think you know the air-inflated 'jump-castles' where children can crawl in and jump over the elastic/springy basis and against the walls?

A similar concept but redesigned for an inflatable rerpap-frame with some support-structures ... all made from sheets of fabric or plastic, glued or sewed together to represent the 3D-structure when pumped up or filled with a fluid but hardening material ...

So you're working only with flat sheets, cutting contourlines, combining the right pieces along the outlines ... and when inflated - voilá! ... a 3D-structure

Viktor

Edited 1 time(s). Last edit at 02/12/2010 03:07PM by VDX.

I begin to understand. A "SandBag PourStrap" or "Old pair of trousers
PourStrap", where you fill the bag or pants* (clothing) with (polymer)
concrete. (A PourStrap is a RepRap made with a mold rather than a
thermoplastic extruder. This is true even if you RepRap the molds first.)

It is a beautiful concept, including fun things like papercraft or
laser cut sheet-molds.

*Please remember to take off your jeans before filling them with cement.

---

Technical issues (this is all guesswork on my part, I'm no expert):

Hydrostatic pressure

Epoxy and concrete has a high hydrostatic pressure and will burst
through where you have sewn if you have tall column of concrete. This
may be true unless you are using generic firehose-canvas used by
firefighters.

The seams are not as much of a problem as the holes where your rods
and static parts (fixtures, through holes, embeded nuts) are. Those
holes in the fabric will be difficult to seal properly, and material
will ooze out. You can seal them from the inside with modeling clay.
... You can frantically patch leaks from the outside with modeling clay.

Your second mold will have many less leaks than your first mold.
If you are lucky, you will not end up with a 50 kg pair of pants
epoxied to your workbench or floor as a permanent desk ornament.

Geometry and mechanical degrees of freedom

If you are making a monolithic PourStrap, you have a number of
fixtures and rods, and you need to align them with 1-2-3 blocks, lego
bricks, masonite or steel pegboard, etc. (In which case you might
make the mold from pegboard and food-type plastic wrap rather than
fabric.)

If you are making a Lots-of-Parts PourStrap, like a Mendel clone,
Darwin clone, Eiffel Clone, Delta-clone, you need each poured subunit
to have 2 geometric ... "constraints"? 2 through-holes, 2 flats, 2
fixtures, 2 embedded nuts, 1 embedded rail + 1 throughhole, etc. Call
it 2 geometric "features".

Because if we describe the Lots-of-Parts PourStrap as a 3D mesh or
graph located in space, even if it is made from toothpicks and
marshmellows, each marshmellow has 2 toothpicks and is connected to 2
marshmellows. (Ignoring 2 end marshmellows).

Locating and controlling the orientation of those 2 features requires
pegboard, 1-2-3 blocks, or RepRapped thermoplastic molds or fixtures
(rod holders).

You get 1 free flat area at the top of your cast, where gravity makes the top of
the epoxy "perfectly" flat. But you need jigs and fixtures to locate
and hold a rod at a specific angle and location with resepet to
to this flat area.

One thing you need for a pourstrap is a plastic drafting triangle,
T-square, and lots of shim stock. Maybe a straight edge. Or a full
machinist toolchest, if you have 100-500 dollars of measurement
technology on hand.

---

An octopus or hermit crab uses optical control and feedback (aka
looking and correcting movement) to specify where the toolpoint is
moving, and doesn't need 2 geometric "features".

Since most RepStraps are dumber than hermit crabs, I think it is
necessary to build in precise geometry.

I apologize for not really understanding the real technical issues or
mechanical engineering to describe the geometry requirements better.

---

I think a monolithic or lots-of-parts Eiffel is the best way to begin
with PourStraps. Right now the best design is lots-of-parts and
lots-of-metal-rod. I think that is a little silly, but it is the best
working design we have, and ships well via the postal service.

... I want to make a plaster PourStrap.