Formlabs Form1 SLA 3D Printer for about the same price as Replicator2 / X2

[happyman]

I think it is very exciting that SLA 3D Printer begin to follow the trend of FDM 3D Printer. At the present, there aren't many SLA 3D Printers below US$3,000 to choose from (the most prominent brand that target home 3D Printer that I am aware of, is B9Creator) and most of them seem to be an indiviual start up. Formlabs seems to be backed by some deep pocket investors so its future appears to be brighter than the rest. It has a nice software too with following features

 

• - Automatic generation of smart support structures
• - Autolayout optimizes part location prior to printing
• - Manual tools for rotation, scaling, and duplication
• - Analyze individual print layers with 3D stage viewer

It also comes with finishing tools as well.

 

It has already hit the target of US$100,000 in Kickstarter.

 

Hopefully, this will help to accelerate the overall 3D Market technologies to the next level.

 

Will Makerbot consider to expand its line of 3D printer offering with other technologies like SLA or SLS? Hopefully, we will know soon with the increasing pressures from the competitors.

 

 

 

 

[Aaron Double]

Something of note between the formlabs and b9 creator.

They are similar but different, the formlabs one uses a laser and is a traditional SLA printer except they are building from the bottom up, the B9 creator is a DLP printer. Uses a video projector to expose an entire layer at one time instead of drawing the layer with a laser.

Not a positive or negative for either one, just a difference.

Aaron

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/iALRwHc1BdEJ.
To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.

[Matt Smollinger]

It looks like a great option. However FDM can be 90% of what SLA produces, at 1/10th the cost. $150 bucks for resin, and thats cheap? Yuck, and oh yeah, don't get it on your hands cuz its toxic and stains worse than anything else.

I'll stick with my PLA FDM for now. 

--Matt--

[Andrew Russell]

Really beautiful design. I *almost* pulled the trigger on buying one.
Then I remembered how much fun being an early adopter* with my
Replicator has been, and I'm going to pass. If they are still around
in 2 years, maybe I'll buy one. :)
I sincerely hope they do well. Any word about their prep software
being OS, or even the hardware?

I've read (on this forum?) that UV reactive polymers are *always* UV
sensitive, and the prints get brittle over time. Lost wax casting
could be a way around this, but I don't have the time to experiment
with it.
For what I want to make/print, I'll stick with ABS or PLA.

I'm really happy the consumer/prosumer focused 3d printing market is
exploding right now.
Now where is the kickstarter for an OS 3d modeling package that isn't
Blender and is focused on the needs of 3d printing? (better booleans,
flawless shrink-wrapping of arbitrary geom, open GL accelerated previs
of gcode, etc.)

-Andy

* kudos to the Cupcake and ToM owners. Thanks for being the *early*
early adopters and getting us to where we are today.

> https://groups.google.com/d/msg/makerbot/-/ZLdCLD8V_YIJ.

[g. wygonik]

I met the formlabs guys over a year ago through work and they had working prototype machines way back then, and the pieces they printed were awesome. I can only figure that a year later they're now even better and consumer-ready. I backed it, so we'll see in February--when mine ships--if my hunch is correct ;-)

I'm not so keen on the cost of resin, but hopefully I can stretch that cost out by using my ToM for "first run" prototypes, and then just print the final version on the Form 1. Or, sell time on the Form 1 to cover the material cost :-)

g.

[Dan Newman]

On 26 Sep 2012 , at 7:34 PM, g. wygonik wrote:

> I met the formlabs guys over a year ago through work and they had working
> prototype machines way back then, and the pieces they printed were awesome.
> I can only figure that a year later they're now even better and
> consumer-ready. I backed it, so we'll see in February--when mine ships--if
> my hunch is correct ;-)
>
> I'm not so keen on the cost of resin, but hopefully I can stretch that cost
> out by using my ToM for "first run" prototypes, and then just print the
> final version on the Form 1. Or, sell time on the Form 1 to cover the
> material cost :-)

Well, the target price for 0.5 kg of the liquid resin for the B9Creator was $59.
(That's about 500 mL judging from my 1.0 kg bottle which is 1 L but I've not
opened it to see how close it is to the fill line and, understandably, the
bottle is opaque. Guess the density is close to that of water.)

We'll see if that price changes over time. The B9 uses a resin which cures with
visible light and can handle layer heights of 0.05mm (maybe even smaller). Part of the
formulation is the opacity which then impacts cure depth (along with intensity
of the light and exposure time).

Dan

[g. wygonik]

Indeed - hopefully time (and competition) will be our friend and drive prices down :-)

g.

[Zip Zap]

    The videos show just a small tray of resin being used.  Also, Walmart and hardware stores already sell fiberglass resin as either an auto body repair or for fiberglass molding projects in general.  That stuff is nasty.  However, I've gotten messy with it even when wearing gloves.  Yet,  I'm still alive.  Also, one gallon of fiberglass resin cost $36 at Home Depot. I'm not saying fiberglass resin is the same type that is being used for SLA printing but it should be similar in principle.  Fiberglass resin hardens after adding a chemical hardener wheareas whatever SLA uses hardens due to light.  In other words, heat.  Anyway, I believe SLA is the future standard once the cost and kinks are worked out.  There is less mechanical axis movements involved and doesn't even look like printing.  Is as if your reaching into a bucket of plastic slime and pulling out whatever object you desire.  

---

From: Matt Smollinger <m.smol...@gmail.com>
To: make...@googlegroups.com
Sent: Wednesday, September 26, 2012 11:44 AM
Subject: Re: [MakerBot] Formlabs Form1 SLA 3D Printer for about the same price as Replicator2 / X2

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/ZLdCLD8V_YIJ.

[Kai Backman]

On Wed, Sep 26, 2012 at 11:24 PM, Andrew Russell <ganst...@gmail.com> wrote:
> Now where is the kickstarter for an OS 3d modeling package that isn't
> Blender and is focused on the needs of 3d printing? (better booleans,
> flawless shrink-wrapping of arbitrary geom, open GL accelerated previs
> of gcode, etc.)

There has actually been a number over the past few years but they have
all tended to fizzle out pretty quickly. The only one I know that is
still standing is Shapesmith.net and it uses OpenCASCADE and doesn't
roll it's own geometry kernel. OpenCASCADE is good but has a lot of
limitations when importing random STL files for example.

Having actually implemented a full CAD stack in Tinkercad, starting
from the geometry kernel upwards, I have a few hunches why this hasn't
happened. For starters "better booleans" or more specifically the
geometry kernel problem is non-trivial. The team at Tinkercad that put
together our Gen6 kernel has an average experience of 20+ years of
working with these problems. The "6" in Gen6 literally means we wrote
six production geometry kernels, ran all of them against live users
for months and rewrote each new generation from scratch. For Gen6 the
lead developer (literally) holed up in a bunker under the city, where
the local university library keeps its less accessed books, and didn't
emerge for several months.

Having contributed to a dozen or so open source projects over the past
few decade that type of focus is hard to muster without some external
form of funding. A PhD student could probably do it but they would be
lacking the experience of running five previous versions live. As a
reference, when Dasasult systems bought the ACIS kernel from Spatial
they paid $25M in cash. To make a really good CAD tool focused at 3D
printing you need a kernel that's more advanced than ACIS. Neither of
the two big commercial kernels available today, Parasolid and ACIS,
are good at handling triangle imports like STL. Either they are really
slow or just plain break.

Anyway, it's a long and technical answer to your simple question. :-)

Out of interest, what do you mean by "flawless shrink-wrapping of
arbitrary geom"?

Take care,

Kai

--
Kai Backman, CEO
http://tinkercad.com - Mind to design in minutes

[sjlane7160]

Stephen Lane wrote:
     I can strongly recommend reading up on the subject of photosetting polymers they don't set due to heat (which I agree is "infrared light") as suggested below the light at specific frequencies ie UV in some cases causes the polymer chains to cross link changing their physical properties in this case from liquid to solid, the tech has been in use for many years in fact it is the basis of photoresists, silk screen emulsions & many other tricky things when you go searching. But I do agree as they  become more popular and wide spread the cost will come down.

  Regards
Stephen

[Aaron Double]

http://bucktownpolymers.com/polymers.html

Has UV curable polymers for $144 a gallon. Wonder if it is compatible.

Aaron

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/ipnfRF7spDUJ.

[Martin Galese]

I'm curious about that too.

The Form1 people have said, not surprisingly, that they're not going to be supporting 3rd party resin, but it seems like the worst case scenario is that you throw away some expensive resin.  

It looks like they're using a 405 nm laser and at least some of Bucktown's stuff should polymerize at 420 nm and below.  Also, at least some people have had luck using the UV (385 nm) curable resins with a 405 nm laser.  See, e.g., https://plus.google.com/113952010793377772767/posts/VomPVvhrgJt

Martin Galese

[Blizzard]

I'm quite interested in getting a Formlabs printer instead of a Replicator 2... but the decision isn't easy... What would be the advantages of the Replicator 2 instead of just cheaper material and bigger objects possible? It seems to beat it in all points...

[Martin Galese]

Honestly, I think those are the main two.  You might have said open source software and hardware, but that seems to be a diminished advantage at this point. And you could always go with the B9Creator if that was a big concern.

To be fair, I understand resin to require more care in handling than PLA or ABS.  You need to wear gloves when working with uncured resin.  And I know at least some kinds really can stink.  Given the office environment Formlabs seems to be targetting, I'd hope they'd have a low VOC formulation that addresses that.  There are also almost sure to be mechanical differences that might matter to an application.

Also, resin supply may be more iffy -- and not just more expensive -- than ABS and PLA.  There are a ton of vendors for filament at this point, and lots of different materials coming onto the market.  Formlabs intends to provide their resin at a good price and with the promise of more options, but we'll have to see how well that works out.

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/JWkhaAAtg0sJ.

To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.

--
Martin Galese

[RocketSled]

Yeah.  I agree.  As SLA starts to proliferate, Resin costs will come down.  

I have a 4-Axis mill and a Rep1.  They're each good for their own things.  You can combine them, in fact, and I have machined some prints to improve accuracy. But there are things neither can do, that an SLA will.  I'll be adding this to my repertoire once the prices come down and the options go up.

[Andrew]

Hey Kai,

Thanks for the detailed response.  The following is a bit long, so if anybody wants to skip it here's the TL;DR:

I'm a vfx artist not a cad modeler. People like me will be coming at 3d printing from a different perspective, with different ways of solving problems. Tinkercad is a good idea and I wish y'all GREAT SUCCESS!

RE: Open Source

I've come into the 3d printing market from a visual effects background.

The VFX industry has a history of excellent software made by brilliant people just up and disappearing. Not because the software was bad, quite the opposite.  Usually the company makes a poor strategic move or gets acquired, and the software goes off the market. 

This is the main reason why I mentioned Open Source.  Too many great ideas & implementations in graphics software just disappear, taking years to reappear in another package, if at all.  I've long wished that the companies owning the source would just open it, and walk away. At least the community could breathe new life into the software, or at a minimum keep porting it to modern operating systems.  Like John Carmack's open source game engines, there is much young programmers could learn from studying the code. 

Another aspect is that most 3d software is expensive and never becomes cheaper. Maya, Max, Houdini, Z-Brush, etc. are still priced way out of the average consumers budget.  The 3d printers now cost less than the software to make stuff for 'em!  If there was an OS alternative, at least people would have something to work with.  (To the Blender lovers, I have no hate in my heart for it. I just find the workflow kinda nuts and counterintuitive… Maybe its more intuitive compared to Houdini.)

RE: CAD modeling.

Again, I come into this ecosystem from a visual effects background.  We rarely use CAD software.  Everything you just described about Dasault and the ACIS kernel, is totally new information to me.  Apologies if this is common knowledge to most group members.   I know SUB-D, poly modeling, and if I think back hard enough, I might even be able to make some trimmed NURBS models.  The only "solid modeling" software I've used, aside from 20 minutes with Tinkercad, is openSCAD (just to try out the boolean operations.)  My workflow and my modeling experience is probably very different than people who work with Solidworks or other software that is designed around fabrication of practical, real world objects.

I think there are probably a lot of folks like me who are interested in, or just starting out with 3d printing.  Technical, but artistic minded. People who have worked in Max, Maya, Houdini, Z-Brush, Modo, & Softimage.  One of the things we were all taught long ago, is "booleans are bad!" (http://zheng3.wordpress.com/2012/05/16/cross-disciplinary-confusion/). It is rarely important in VFX that geometry is water tight.  I've got dozens of old models that I've been prepping and printing.  Most of these models are either dead polygons, or sub-d surfaces.  They are made this way because it is what we were taught.  The methods are perfectly appropriate for the original end goal, but can be challenging to make printable.

Many young people are learning Z-Brush and MudBox today.  And with 3d printers hitting the $2000 range (and less) they are going to buy printers and try to their models. Because of how they have been created, these models do not always print well (w/o lots of tweaking.) Functionality that hasn't been necessary  in the past (better booleans), will rapidly become essential.  Maybe I'm a bit too out of touch, and the kids are already out there somewhere discussing and solving these problems. (let me know if you find em!)

Having said all that, this is why I mention "better booleans" (http://www2.acae.cuhk.edu.hk/~cwang/pubs/TVCGMeshBoolean.pdf), "flawless shrink wrap" (no link to point to!), and previs of g-code (something combining this http://www.pleasantsoftware.com/developer/pleasant3d/  and this http://www.thingiverse.com/thing:6236.)

Finally, to answer your question regarding shrink wrap.

When the boolean nightmares get the best of me, I start dreaming of a "shrink wrap" function that could just encapsulate a collection of geometry using a parametric closed surface. Something that would initially conform to the outer most surfaces, locking on to the vertices, and interpolating where the surfaces intersect.  The surface flow and poly counts will not be the same between the two surfaces, so it would be great if the wrap could adapt/match its resolution between the two. No too different from how a boolean union retains the exact external polygons where the geom isn't intersecting, and does its best where they do.

Most "shrink-wrap"  tools I've seen are simple and usually work like this:

http://ticket01.com/multitool/docs/  (Check out the shrink wrap graphic)

Unfortunately, with the ones I've tried out, the end result is nothing like the original mesh in terms of surface appearance and topology.

The benefits of using a shrink wrap type approach vs boolean unions, is that you could potentially avoid the tricky problems of self intersection, non-manifold geom, and non-closed surfaces.   Just collide with the exterior surface, figure out the mesh density from the polygons/vertices, and don't bother trying to make a fillet between the two surfaces. Just create a hard seam along the intersection boundary of the meshes.  Does this exist? Would the math be easier to implement vs 'better booleans'? Can you add it to tinkercad? :)  

RE: Dev Costs

I totally understand that this type of software requires deep expertise, smart UX/UI folks, and lots of capital to fund development.  I don't mean to trivialize this monumental effort. I have many friends who's day jobs are doing exactly that.  They work for big software houses that charge thousands of dollars for their products.  They can charge these high rates because they are effectively sponsored by large vfx houses and academic institutions.  To sell the software for less but maintain profits would require a larger market. 3d printing going mainstream could help make this happen, and I suspect you guys already understand this or you wouldn't of made Tinkercad. 

thanks

-Andy

[Andrew]

On Thursday, September 27, 2012 10:48:32 PM UTC-7, Andrew wrote:

Many young people are learning Z-Brush and MudBox today.  And with 3d printers hitting the $2000 range (and less) they are going to buy printers and try to their models. Because of how 

Oops,  I meant to say, Z-Brush and Modo... Though some people prefer Mudbox over Z-brush.

[Elbot]

The only thing keeping me from buying it is the cost of the resin. Let's say you print a pair of high heels with your Makerbot for $20. Those same heels would cost you $60 if printed with resin. For $60, I'd rather just buy shoes at the shoe store. Also, size does matter. You can print a pair of women's 7 & 1/2 heels on your Replicator 1. You'd have to do it in 2 halves and glue it together if done on the Form 1.

Another example would be iphone 5 cases. On a Replicator, it would cost $2.50 (including depreciation on your bot, filament, and electricity). On the Form 1, it would cost $7.50. Again, I could buy an iphone case on ebay for $7.50. However, it would be difficult to buy a cool iphone 5 case for $2.50 including shipping on ebay. It would be difficult to buy a pair of heels for less then $30 at the shoe store, too.

It is more economical to print out your own stuff than to buy it in a store with a Replicator 1, but not so with a resin printer due to the high (triple) cost of the resin. Also, the build size is smaller, too.

I could print out an AR-15's lower receiver, grips, and stock on my Replicator 1, but the Form 1 is not big enough. It is also cheaper to print these out than to buy them at the store. Not so for a resin printer.

You could print out high dollar items at lower cost with a Replicator 1, but that is not possible for a Form 1 due to small build size and high resin cost. The Form 1 is not for the home/hobbyist crowd, but for the engineer crowd. When the price of resin comes down, then home / hobbyist users would get them.

[Elbot]

I just read your post from John Meacham about uv curable polymers. The Form 1 uses an acrylic polymer, so it is not the nylon polymer from bucktown polymers. For the bucktown nylon polymer to work, you'll need to have the nylon stick to the raising platform while not sticking to the liquid tray. Trouble is, nylon sticks to different stuff than acrylic. So, I'm not sure if the cheaper bucktown polymers will work with the Form 1. Also, the Bucktown nylon stinks more than the Form 1 acrylic. Stinkiness could be cured with this print: http://www.thingiverse.com/thing:28268

If someone tells me that the bucktown polymer nylon will not stick to the tray, then I'm buying the Form 1, even if it won't stick to the upper platform because you can always stick a nylon panty hose to the top platform to get the nylon print to stick to the rising platform. However, there's no easy mod to get it to work if it sticks to the liquid tray because it is not supposed to stick to the liquid tray.

So, anybody know?

Thanks.

[Aaron Double]

Look around their site, they have a 0 VOC version as well.

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/RKVdwsbz38gJ.

[Kai Backman]

Hey Andrew,

Thanks for the long thoughtful response, it was an really interesting read. I'll answer a few points but I'd rather not derail the thread by talking about Tinkercad, if you want to continue talking just mail me directly or start a new thread here.

You can actually import arbitrary and bad triangle data into Tinkercad, currently in just STL but we'll add more as we go. We then turn this into a proper editable solid behind the scenes. A super robust way of fixing your old data is to import them and then just export the resulting STL. It won't match the original geometry exactly but it's guaranteed to be watertight and non-intersecting. This achieves pretty much what your shrink wrap tool would.

I actually also have a background in classical poly modeling with sub-d, poly and tools like Maya. One thing we wanted to completely hide from the user was the distinction of a polygon model and a solid model. In Tinkercad everything is a solid. You only get watertight and non-intersecting models out that are OK for printing. In some sense our goal has been to let people who have little experience or are used to the ease of polygon modeling to quickly get started with solids. You don't have to do any special effort to make sure things work OK, as long as it looks OK on the screen the printer will try to output it. We have been talking a lot about incorporating more sub-d style tools, for some reason those seem quite easy to understand to most people, I'd love to talk more about it if you are interested.

That said I really just wanted to respond to the open source question you had, before founding the company I thought seriously about building an OS CAD instead. I just wanted to highlight our experience as one viewpoint why an OS CAD will take a while to happen, much like the OS airplane avionics system will take a while to run on any airplane that isn't experimental.

Oh, and totally love Pleasant 3D, you are right we should add a visualization layer like that.

Take care,

  Kai 

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/ggHVblktVxkJ.

To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.