nearly bought one

Now too many months ago I nearly bought an Up! 3D printer. I'd even received some sample prints and at 2 microns there is a definite grain. Easy to sort out on a large flat surface but on a freight car truck or say, a building facade or some neat little detail, you will get these grains or layers.

With Makerbot now going to 1 micron, this is big a step in the right direction. This will greatly reduce those pesky layer effects.

I think the dunking method is going to produce waaaay better results. Dunking isn't the correct term, but it does explain what the process is. Mind you,this method is in its' infancy and the costs are still very high. What ever way the crowd tilts it still remains a very exciting time. The article is spot though as far as economies of scale goes. If you want a million you go the ol' fashioned way. Well not all the way. You can still print out a master, cut a mold and cast a million of the rubber duckies. I would like to make one off city building rears and if they could offer me a machine that had minimal layering, I'd have got that Up! machine. No-one makes shallow relief examples of non descript building rears.

Still, I visit 3D sites on a regular basis to sit,read and watch it all unfolding.

cost

oh and the 2 to 3k is CHEAP.

Maker-bot only went to 100 micron steps

That's 0.004" height steps which is pretty much the present  (cheaper) 3D printer industry standard. And they don't say what the horizontal resolution is.

You'd need better than 0.001" resolution to make good smaller sized HO parts.

"Cheap" 3D printing

If you're looking for a "cheap" 3D printer, check out Printrbot. ( http://www.printrbot.com)

These are kits. Made of 3D printed parts, laser-cut wood, and some fairly common hardware. As do-it-yoursef, "hobbyist" type machines they obviously can't do 1 micron resolution. But at $400 for the smallest size (4" x 4" x 4"), and $700 for the largest (8" x 8" x 6"), they're definitely affordable.

I bought the Printrbot PLUS. And while I haven't finished assembling it yet, so far it's gone together fairly easy. I'm looking forward to "printing" some N scale buildings. We'll see how they turn out.

Printrbot

I'm looking forward to that, too.  Please let us see what comes out, good, bad or whatever.

Me too!

I can imagine all sorts of interesting possibilities, and ways to deal with the inevitable grain.  For example, one could print things like windows and doors which, since they're basically flat, should have less grain to them.  That would go a long way toward making custom structures practical.  Car sides, especially wooden ones (where grain is there anyway) might be another good application.

I'm looking forward to your report on the results.

3D printing

I worked in the R&D Dept. at Maytag and they had a SLA machine that used liquid medium and produced both the model and support material. The support material was removed after the model was complete and then the model was placed in a ultraviolet chamber to cure. After the curing the model makers would make any tweaks to the model and return the model to the UV chamber to reharden any newly exposed surfaces. This material was judged to be a carcinogenic and any exposed surface had to be hardened. The price of this machine and its materials was highly cost prohibitive and would only pay off to a manufacturer on making a preproduction model to check coordinates for CNC cutting of dies.

Two very different basic materials in use

The technology difference is "thermoplastic" versus "thermosetting". The machines the crowd is speaking about on the forum use thermoplastic material. The rapid prototyping machine at Maytag uses thermosetting material.

Don't confuse "thermo" with meaning only heat. It can also refer to any mechanism that causes a state change. Heat, UV, catalyst, solvent, pressure, and other mechanisms are used to force state change.

Thermoplastic materials can be converted back and forth between solid and liquid states. Thermosetting materials can only go one way once - from liquid to solid. UV initialized polymerization (conversion from liquid to solid) is a popular method with thermosetting materials. UV cured printing inks being a prime example.

Should I, or shouldn't I ? . . . .

hope someone here will have insight.

I'm working in TT scale (1:120), of all things. I use 2D and 3D software daily, I have tabletop Sherline machines (CNC mill, CNC lathe), and a casting shop that includes heat controlled pressure pots and an unused spin-casting machine. (I've yet to move to the casting shop for learning. So little time. . . .)

My attention has been captured by the new offering, the MakerBot Replicator 2. I'm considering it.

(Hands twisting around one another) oh, what to do. What to do?! The CNC machines do beautiful work but are time consuming (set up as well as run time).

I'm not sure that 100 micron resolution (as advertised for the MakerBot R2) is good enough for 1:120 scale. For example, rivets on freight car sides, if finish sanding is required, smoothing around printed rivets -- is that possible? Although, recently, I CNC cut the roof of the short hood for a GP7/9 modular model project. I cut the side radius in .001" increments and could not see or feel "grain." On the top flat surfaces, tool marks were barely visible (I used a .018" dia. end mill) but they could not be felt with finger. The rivets look fantastic. Sanding might be required there, too, on the flat surfaces, unless a primer coat covers up the tool marks.

It will take a major effort with a tire iron for me to pry open my wallet for a MakerBot Replicator 2, but I'm researching the machine now. Any comments anyone here would care to make, I'd love to have them. I'm after as much understanding about 3D printing for hobby manufacturing as I can find. (And, I must consider 3D printing vs. CNC cutting.)

Thanks much,

Brian Chapman / Evansdale, Iowa

3D printing vs CNC

You already have the CNC machines, why would you want to spend money on another machine. If you are making the molds with your CNC why spend more money on another machine?

From the sounds of 3D it takes time to make a model. In my opinion I would use the CNC machines to make what ever you are making and wait for the 3D printers to advance some more.

My 2¢, worth about as much as...........

Bernd

#D Model making

As to why you would want to make resin model before cutting a CNC mold is simple, the cost of the mold which is generally done in tool steel for longevity and very costly tool wise, where as making a resin model and tweaking it is much cheaper than cutting a new mold. I have seen at Maytag a well done CAD drawing for a mold was off on one of its coordinates by 2/1000th of an inch and turned mold done in aluminum into junk when they put it on an injection molding machine, instead of a window they got strike, breaking the mold. The SLA machine I was talking about uses a laser to harden the liquid resin, they also had a thermo plastic model making machine that used two rolls of plastic on for model and one for support. Several auto manufacturers use SLA to produce prototype parts such as intake manifolds and engine shrouds. The manifolds are produced in a high temp resin and are actually put on an engine to see how they perform, any tweaking is done on the drawing and in about 6 to 10 hours they have the revised model to test.

Appels vs Oranges

I should have been more specific on the kind of mold. I was thinking RTV for the mold not aluminum or tool steel. RTV is a bit cheaper than tool steel.

Bernd

Take a deep breath . . .

<

Andy, horizontal resolution . . . something I don't believe I've seen mentioned in 3D printer chats. This would be the difference, for example, between 220 sandpaper and 2000 sandpaper. Appropriate analogy? That is, surface smoothness is the issue?


<

<

Every time an announcement pops up, like MakerBot's R2 unveiling, I have the powerful emotional reaction that I'm being left behind. It takes me a day to calm down. MakerBot's claim that post-printing finishing is not required with its MBR2 is not correct for our hobby, though. A .004 resolution won't cut it.

My mill can do .001 resolution easily; I haven't tried to do better because I haven't seen the need to try. You're right, Bernd, I'll await further 3D printer developments.

Back to my USRA Mikado and GP7/9 projects, making 'em the old fashioned way 

-Brian Chapman / Evansdale, Iowa

Holding on ........

Yes, 

this is something that I'm watching carefully.

I want to make 9 particular passenger cars with smaller than available windows in N scale, but cutting the windows (14 per side ) out by hand will not produce consistent results. My plan was to take Autocad plans to a lazer cutter, and let him do the work.

As I am only into building my layout now, I thought that I could wait a year or so, until the 3d printers come down in price and up in spec. ( this is going to happen. ).

Then, picture how it would be to see the models you've always wanted get created in front of your eyes?

A little off topic

Fidel,

You might consider making your windows out of etched brass.Etching is easy to do and sounds like it would suffice for your project.

Rob Teed