Building Rapid Prototype 3D Models from Zbrush Files Tutorial

For the record I am not pushing the 510, just applied for a grant and it seems to be the best option for the Smithsonian to get into the 3D printing game.

Hey Fork,
Zcorp has been around quite awhile and their machines have been popular as concept modelers. I have not been able to consider their machines for my work as the resolution & accuracy are not up to what my clients require. This is not to say that their machines aren’t good though. It’s hard to argue that when the high level of detail or accuracy is not important that the cost savings is not very attractive. When this is the case many machines come into the light as possibilities. I guess it comes down to “if the shoe fits wear it”. Out of curiosity, what would the Smithsonian 3D printing application be?

I met ZCorp, Dimension, and another wax printing corp at Siggraph. ZCorps printer was the least desirable regarding the output quality. Top was the Wax printer, which took ages and could not print large items. But the quality was immense. Dimension with their ABS plastic were pretty cool, but then again, you have the ridges which depend on the resolution. I heard wondersome things about the printers which form the models out of a polymer bath. But I have never laid hand on such a model and thus can’t speak for or against it. I am having rather good results with 3 and more axis cnc machines now. But adapt my workflow to the tool a lot. Considering cutter geometry and tool path is going a long way. But it’s not the salvation for a general purpose method.

Shhark- I am a sculptor and make exhibits and scientific models. The main reason would be to 3D scan an artifact, print out a replica then safely make a mount from the 3D print. This is one example of many…

Can anyone recommend an RP machine for $50,000 that performs better than zcorp? We need large work envelope, inexpensive material cost and nice resolution. The wax printer are great for jewelry scale but not for bigger stuff.

I run a small 4 axis CNC machine at home but still learning. I am hoping the zbrush folks see this great potential for CNC and RP work.

excellent recopilation!! thanks for your work, Josh

Fork, you should also look at foam cutter. Those are fairly inexpensive and can yield great results if used for the core of models. The application of a thin skin with detail could be done manually and yield excellent results. That would cut a LOT of time of the process.

Here is an exotic use for foam cut parts. It’s the lost foam (like lost wax) method. I make parts for my 6axis machine with this technique and it works great. I thank the neighborhood for all my Aluminium for this haha.



Hey Fork,
You might want to check into used Objet machines or even a new Objet Eden 250. Objet used to offer refurbished machines with warranties in the past & may still do so. The Dimension printer is an FDM machine (fused depostion model) & costs between $19k-$33k. This is what Hiroshi had his parts made on (see here). Machine cost is affordable (relatively speaking) but the part resolution is terrible & requires significant post finishing.

I think the process lemonnado is referring to with the polymer bath is either SLA (stereolithography) or EnvisionTec as they are the only processes I know of that build like that. SLA has been around since the early years of RP but the machines are crazy expensive, have expensive lasers that need regular calibration and replacement, and require a controlled environment & many peripheral stations for cleaning and post curing of the parts. Definitley not “office friendly” but the part quality is still very acceptable depending on the application and the part cost can be much less than Objet when building large models.

The Invision is made by the same company as the SLA machines. I had the same parts made on it & the Objet machine to compare side by side. The dimensional accuracy was unacceptable for our applications and the material has a wax ingredient that makes for a terrible painting or primer surface. It’s resolution is also much lower than the Objet but the machine cost is very attractive. They initially offered the Dimensions for around $75k until their potential customers realized how sub par the output was for that price point that they dropped the price over half!

Another very interesting technology is the Perfactory by EnvisionTec. I had parts built to review and they were better than most but not as good as Objet. Accuracy was a little off but resolution (.001"-.004" layers) was generally very good and they have a very impressive set of material choices available. They are a German company & I cannot comment as to what their current support for U.S. would be. As of the last time I spoke with them there were zero service bureaus in the United States so getting parts made by a vendor in the States is not an option. Their smaller machines go for around $85K.

Best thing may be to contact as many machine manufacturers as possible and request samples of a file you provide so you can compare all of them side by side.


Josh Harker
Paradigm Development Group

Lemannado, thanks for the advice and cool link.

Shhark, thanks for all of the great information your provided about RP, much appreciated. btw nice work on your website.

man as much as i’d love to do this being addicted to zbrush and all its just too expensive for the size, a crack habit would be cheaper.
I"m curious though how much of the cost is really about the materials over trying to pay for the machine itself.

I’m not sure if a crack habit & ZBrush are all that different. Both keep you up all night & you end up with that blank stare after too much of it. :smiley: Although I haven’t pimped myself out for the new Wacom Cintiq yet your point is well taken.

I would definitely say a great deal of the cost in 3D printing is covering operating expenses. How much do the actual raw materials cost that go into making your computer? Maybe $10 if that. It’s all in manufacturing & the technology to pull it off not to mention the cost of doing business. Then there’s the cost of software. The CD’s it comes on costs pennies but the development behind it has a price.

3D Printing technology currently is not geared for the home user or hobbyist. It has however definitely revolutionized product development & I dare say how fine artists may develop their craft but it comes with a cost. PC World just had an article (see here) showing prices of technology from 1988 in todays dollars. A 150MB hard drive comes in at $8755! Printing costs will come down & quality & options will get better but it will take time. :wink:

As I mentioned earlier in this thread, I am considering offering a discounted group build for ZB users. Still have to work out the details but there would be limited model size, material choice, polygon count, delivery time, etc. Maybe try to do it once a month or so. Anyway the point would be to give us burgeoning digital sculptors a chance to try out 3D printing at a more affordable price. :smiley: If there’s enough interest (particularly from ZBC or Pixologic) I’ll put it together. Chime in if you’re interested.


Josh Harker
Paradigm Development Group

I suppose i would be interested if i finish a model that i think would be worth printing :wink:
Can you elaborate on why the poly count effects the price?
I did understand what you’re saying about operating costs, its just too bad, cause i would think that if it was overall cheaper, then companies would have more people interested and they would make the same amount of money etc. its funny cause i remember well over 10yrs ago a newspaper article saying how we’d have these devices at our homes like our normal printers, i suppose one day that will happen but it still seems very far off.

Thanks for the fascinating look at “lost foam” casting. (too bad my HOA would never allow glowing ingots in the dirt… but it’s fun to imagine appalling the fascists. :smiley:

thank you for such comprehensive work in your posts. From my own tedium with Googling this subject a year or so ago, I can tell just looking at this thread, that y’have it in hand. I’m lurking.

RE: 2.5D Routing
Asking a noob question of the assembled experience here=

Could a Zscript be capable of bookmatching a ZB file before export, so I could at least run model halves through something like a CarveWright?

With CW, I’m not talking scrolls and low relief-- with bookmatching, I’m inverting a mould-making process in carving both positive halves, true? Since CW will accept what appears (in the video) to be 4x16 stock of some length, apparently I could attempt to batch cut a scene of models and join them. I’d need tests for poly count limits but ZB does amazing work at low counts.

If not, maybe I could develop a register jig for work-in-turn, and coax CW to rout in the round. This is from a few days of mulling… Comments?

Don’t waste your time with a CarveWright… Research the forums carefully before you buy one. All I say is cheap ‘flex shaft driven high speed tool’.
There are other alternatives which cost about 1000-1500$ more and have hundred times more precision and are made 100% in the US and not in one of the foreign countries which flood us with junk. If interested, PM.

And yes, you can invert a model and create a mold. Then machine it into something to wither receive wax for further investment casting or, what I actually tried with success, directly a aluminum/zinc alloy (die cast metal).



Polycounts don’t really effect price but if I’m going to offer a group build then multiple models will have to share the same tray so they can be built together. The Objet software (Objet Studio) can only handle 250MB per build. Large polycounts mean large files sizes which mean fewer models that could be on a tray per build. That is why decimation, retopology, and careful modeling practices all help when creating your model. So polycount is really only applicable to me offering a group build.

Josh, I think you would have a lot of success for that group build when you would offer ‘compartments’. Make it fixed cubic inche ‘slots’ with a fixed amount of polygons. Someone could then step up and buy a group slot with a certain amount of polygons and would know that he will receive a model of that size made with the file which contained so many polygons. To much choices usually ruin a concept. I don’t know the envelope of the system you will use for that idea but if it would be 10x2x2inches then you could offer 5 slots with 2x2x2 inches volume and 50MB polycount for a fixed group price.
Just a thought…


You seem to know your stuff… pretty much right along the lines of what I’m thinking. Might have 3 size/price tiers. Early setups look like maybe 1x2x2, 2x2x2, & 2x2x4 “slots” as you called them. Prices would be relative to which slot you pick. Once we fill up a tray then we print & ship. Right now I’m really trying to work on how to get the price down as much as reasonably possible. Polycounts will probably be capped @ 25k or 25MB file size which happen to generally coincide. That means I could run 10 models on one build & split the build cost up between them all. I’ll start running #'s next week & hopefully get something workable together soon. :idea:

Thanks, Lemo I will PM; really frantic right now so I’m putting it off, but I knew I couldn’t afford even a cheapie multi-axis CNC. Besides, most work may be slip-cast porcelain or plaster. lol–I didn’t think of SIGGRAPH for alternatives!.. I appreciate your help.

I am reminded of an experience from Lemonnado’s idea of slots-- High-fire pottery kilns have been gang-filling space for a wide range of objects and customers, selling space by cubic inches and temperature required.

Like short-run offset litho offering a certain ink and paper of the day, one potter was offering special pricing if the customer could wait (e.g., Saturdays she ran a filled “Cone 6” kiln). For your purposes, aggregating hobby/amateur projects may be a low risk way to cross-train your crew?

I think we are up to something really cool here. Thanks a lot Shhark for putting energy into this. If this works, maybe there is a chance of ratcheting it up a notch further in the future and create a subscription. That would enable warranted batch cycles and take more risk of your shoulders and more continuity for anyone else involved. And… honestly… even if there is no strong demand… everyone can come up with ‘something’ to not loose the paid slot…

But I guess it would be great to get that first ever batch going. Communities like this have proven to be very enthusiastic in the begin and then to numb to follow up. Specially when it involves money. Freeloading is always appreciated but charging even the littlest amount for a fair compensation will be condemned and frowned upon; I am curious how this will work out ;).

Just brain storming here…

I must say I haven’t got a clue what you guys are talking about. I hope there will be some easy to follow summaries with pictures now and then? :cool:

I agree about the $ issue. That’s really what I’m trying to work out more than anything. More than likely I will be able to offer a deeply discounted price for Objet parts, but I certainly can’t give them away & the price may still be higher than some other processes. Although the Objet parts are superior I understand that cost is a big factor & some may not take into consideration the value for the resolution over simply getting just any printed model, no matter how poor the quality.

I’m open to creative thinking about how to organize & ultimately pay for educating the ZB community & establishing a well understood and reliable pipeline for using ZB for 3D output. I wonder if Pixologic as a company would have any motivation in participating in some kind of way? Anyway I’m committed to giving it a try and letting the chips fall where they may.

I promise I’ll lay it all out in the simplest terms with as little techno or industry speak as possible. I expect there may be some hand holding the first time around & am prepared to handle that as best I can. I understand that while many may have heard of the processes most are not familiar with the details. That is why I’d like to give everyone a chance to try it out with some kind of introductory step. Hopefully we can crack an egg here & get something special happening with expanding how we’re all applying ZB! :slight_smile: