Welcome to the SeeMeCNC Community

https://www.youtube.com/watch?v=556VmPXWfTg

Interesting idea that should be perfect for deltas.

Brian wrote:https://www.youtube.com/watch?v=556VmPXWfTg

Interesting idea that should be perfect for deltas.

Interesting concept, if you go to their site you must fill out your life's history
to get a quote, when I see that it means it is too expensive for me. If the price
is not listed up front, I say bye, bye.

Eaglezsoar wrote:

Brian wrote:https://www.youtube.com/watch?v=556VmPXWfTg

Interesting idea that should be perfect for deltas.

Interesting concept, if you go to their site you must fill out your life's history
to get a quote, when I see that it means it is too expensive for me. If the price
is not listed up front, I say bye, bye.

I thought you guys would be interested in seeing the process in the video. I didn't tell you to sign up for anything.

I like what they are doing, and think it could be something any of the G-code tools could reproduce. I'm no expert, but someone with the right experience could probably make something like this.

[/quote]I thought you guys would be interested in seeing the process in the video. I didn't tell you to sign up for anything.

I like what they are doing, and think it could be something any of the G-code tools could reproduce. I'm no expert, but someone with the right experience could probably make something like this.[/quote]

Indeed it is an interesting process and now that you mention it it may be worth looking into. Probably the first thing to do is to see if any patents have been granted to them.
Although you didn't tell anyone to sign up for anything, I am of a curious nature and I wanted to see what they wanted for their software. If they are that closed on the price, the odds
are very good that they obtained a patent somewhere along the line but it doesn't hurt to check. I think you can check for patents online but I've never done that sort of thing. I'll try to
figure out how to do that because I am curious.

I didn't have to look to hard, they have patent pending plastered on their website.
If you watch the video it really is cool what they are doing but they intend on selling it
no open source with these guys.

It's about time that we see this kind of slicing! Too bad it might be 20 more years before we get to print like this since they applied for a patent.

There was a thread last year that showed a hexapod printer printing tangential to curved surfaces. Maybe that is enough prior art. Printing with a fixed-angle nozzle will require a very pointy nozzle with no flat on the bottom in order to get good surface finishes. Maybe I'll be manufacturing those nozzles.

But the idea of using the Z axis during prints isn't new. I've written a bunch of gcode by hand to print 'wigs' out of nylon. Basically you just start extruding with the nozzle touching the build surface, then raise the Z axis. When you get to the length you want for the hairs, you retract the filament, move the delta platform over a millimeter or so, then plunge onto the bed and start the process over. The result is a bunch of hairs in an array, just like your scalp. I experimented for a while with ways to get the nozzle to not hit the previously extruded hairs. I guess that's slightly off topic compared to the contour printing that they're doing, but it seemed in the same category of printing with more than 2 axes.

With any luck the patent will be denied. I get so sick of asshats that try to patent math. Just because they can get away with it doesn't mean it's not despicable.

g.

geneb wrote:With any luck the patent will be denied. I get so sick of asshats that try to patent math. Just because they can get away with it doesn't mean it's not despicable.

g.

Well said.

geneb wrote:With any luck the patent will be denied. I get so sick of asshats that try to patent math. Just because they can get away with it doesn't mean it's not despicable.

g.

Nice wording, I agree.

I can see the point of software patents if it's to protect specific software, but patenting concepts, especially ones that aren't novel (such as this) is a major problem since it slows down the progress of an entire industry.

I bet their company will have been bought by Stratasys by July.

Software is NOT a physical thing. It never has been and it never will be. You cannot (or should not) patent the manifestation of math. That's all software is.

Copyright is enough of a legal protection for software.

g.

I think this is very cool.

On a separate note, giving someone the benefit of the doubt at least once is a good thing. They might be protecting it since they worked their buts off for years and years to get it going. They might open-source the code after its patented. Who knows. A patent is a claim but how it is claimed up to the owner.

I got a virus warning after going to the site. I'm good with the idea and will take a look later!

geneb wrote:Software is NOT a physical thing. It never has been and it never will be. You cannot (or should not) patent the manifestation of math. That's all software is.

Copyright is enough of a legal protection for software.

g.

Thanks for clearing that up for me. It didn't quite hit me until now what the difference is. Actually just to be sure about this...

Copyright will prevent people from stealing and commercializing your software (giving you exclusive rights), while a patent gives you a monopoly on something- in this case the mathematical interpolation of the Z axis of a 3d printer (which has been used extensively before).

Am I on the right track with the statements above?


And JohnStack, I gave them the benefit of the doubt until I read their team info. It looks like they have a history of starting and selling businesses. They already sold a business to 3d Systems, and I'm guessing that they knew they could sell this one too. This is exactly the kind of stuff that the huge names in 3d printing are looking to purchase. They see the storm of consumer level printers coming mess up their market dominance, so buying every piece of printer related IP they can is a way for them to avoid it.


EDIT;
I just re-noticed that they use the term "FDM" on their youtube video description. I remembered this from Wikipedia:

The term fused deposition modeling and its abbreviation to FDM are trademarked by Stratasys Inc. The exactly equivalent term, fused filament fabrication (FFF), was coined by the members of the RepRap project to give a phrase that would be legally unconstrained in its use.

Looks like they have a buyer in mind?

I'm not so sure that MasterCAM or SurfCAM could not be set up to do this kind of tool path right now, If you think in terms of a micro ball end mill machining out a cavity, the cavity actually being the additive part we are making. Thing is these programs command a hefty price tag. Our Printers are easily capable of the task, the code is the next horizon.

Check this out: http://www.joelonsoftware.com/items/2013/07/22.html

g.

geneb wrote:Check this out: http://www.joelonsoftware.com/items/2013/07/22.html

g.

An interesting read, perhaps we do have the power to stop some of the asshats.

Here's the deal on patents, etc - at least from my point of view....

Realistically, you can complain all you want about patents - to no particular end - except to promote change in the existing system. People have complained for years. I've been on teams where we've worked for months on something only to find out someone got a patent so similar to what we were doing, we killed the project. Hundreds of thousands of dollars. Do I like them? No.

Do I like them? No - but they are the present system and to protect my interests, I'll play the game - since it is the game. When the game changes, so will my actions. In other words, changing this system is as likely as anyone changing High Frequency Trading on wall Street - there's too much at stake for players much larger than ourselves.

I saw Joel's post last year. The USPTO asking a private individual or company to establish a software site? Not likely. More likely they said "If you want to assist, you're in the private sector, go ahead." This has been alive for six months and there are only 2200 submissions..... Unfortunately, it's going to die a slow quiet death unless people promote it.

Neat to see it done, but it could be accomplished right now with some free tools. Setup is a bit of a pain, but cambam could generate the code required to do what was shown in the video. Heck, a faster way would be to use the pencil toolpath of meshcam. Unfortunately the accomplishment of doing that path with either of those tools would be a bit of a hack, but certainly possible right now.

The question is how is the patent worded? If it is specific to 3D printing (as in additive printing) and the software produces G-code then anyone could write software to create tool paths if they don't describe the end use (this avoids process patents). So software patents for 3D printing tech could be avoided by designing the software for any kind of G-code output. Or use a different equation if the patent is a method patent. The nice thing about the patent is that you have the equation or method to avoid.

Patents have their place, but you need to be organized to help badly granted patents can be squashed.

Also, the patent system is now "first to file" and not "first to invent" anymore. This was a recent change to the patent process. So prior art may not come in to play as much as it used to. I don't know.

If there is a question of if a method or process is patented then get a hold of the filed patent(s) and see how it can be avoided. I think most OSS will be ignored unless there is commercial gain involved anyway. If a company gets a bad name for patent trolling in the printer community it will sink their ship.

I despise software patents. Things like Samba could never exist in a world where companies can get away with patenting software. I see the necessity of copyright, but not patents. The amount of prior art that goes into any kind of significant software is immense. The labor needed to assemble that prior art has value, but if one person can do it, why not another? To implement someone else's idea (as opposed to plagiarizing someone else's code) requires your own labor, and I think the value is there, in the labor.

I also have to wonder if this inventor might be exposing himself to liabilities from other patent holders. Did he invent Cartesian 3D printing? Hahaha, no. What about 3D toolpaths? Naturally not! There is plenty of prior art, with non-layer-based CNC milling machines and the like.

In milling you have the exact same problem with the exact same solution. It was done in CAM software around 25 years ago.

Yes, it would take some clever math (linear algebra with some heuristics, I assume) but the output is just G-code. The printer doesn't know or care if it's building something layer by layer, or following a 3-D curved surface. All it knows is a series of discrete poses.

Mac The Knife wrote:I use Meshcam for 3d machining on my minimill. Once I get a printer, I'll have to try it. I also assume that with a delta printer, some one with the coding no how, could probably get the "effector" plate to follow the shape of the model, instead of staying parallel with the table.

A delta printer by definition cannot rotate the effector. You would need a Stewart platform to be able to add the 4th and 5th axis.

It has been done however, and the proof can be seen on YouTube.

Mac The Knife wrote:I use Meshcam for 3d machining on my minimill. Once I get a printer, I'll have to try it. I also assume that with a delta printer, some one with the coding no how, could probably get the "effector" plate to follow the shape of the model, instead of staying parallel with the table.

No clever math will allow you to do that with the current design, it has exactly 3 degrees of freedom, and none of those influence head rotation. Regardless of the motor positions, the head will be level to the table.

There are delta designs with 6 independently driven arms (None used for printing AFAIK) they have full 6DOF and therefore rotational freedom.

The current design is good enough to do shallow curves. Adding two axes (horizontal effector rotation and vertical tool rotation) would make the envelope much more comprehensive.