Patent Infringement (Stratasys)

Note that I'm in the U.S. (patent laws seem to be a bit different here, but they follow the same WTO rules, as far as I know). Also note that I have only a basic understanding of law, and no personal experience.

I'm interested in building a RepRap and using parts that I make as components in machines that I will use to manufacture products that I can sell for a profit.

I understand that Stratasys has a series of patents protecting their FDM techniques, and that the RepRap has a design similar to their machines. Everyone seems to be okay with this because most people here are interested in research and recreational/hobby use.

What is the extent of the patent infringement by the use of a standard Darwin RepRap? Which parts of the design infringe on which patents? Is there perhaps a way to redesign it such that its use as described could be legal?

As far as I know, whether the idea was "invented" independently or copied doesn't matter in the eyes of the law. (Also, I'm not so sure you guys "invented" the RepRap totally independently of the patented FDM process anyway.)

Lastly, what is/are the ending date(s) of the relevant patent(s)? Stratasys' patents seem have filing dates from 1989 (or possibly 1988) to 2007, but I'm not sure exactly which ones apply, and I'm not sure what their terms are (they are not listed). I assume the terms are the standard 20-year terms starting on the filing date, but I could be wrong.


Oh, and does it matter if the patent describes an apparatus, as opposed to a method? That is, if someone has patented an apparatus (like patent # 5939008 [www.google.com]) for performing a certain function, and I come up with a different apparatus performing the same (or a similar) function, would the patent "protect" against the use (and/or creation) of my device?
I would think that the patent would be irrelevant in such a case; am I wrong?

Disclaimer: I am not a lawyer. I don't even play one on TV.

That being said, the Big One is patent 5121329, which covers the general concept of FDM. There isn't really a way around that one. That is the one due to expire at the end of October of 2009. There are many other patents, but they are pretty much as a rule either more specialized (avoidable) or not applicable to RepRap. However, I haven't read all of them, just scanned through the titles.

And I don't think anyone ever claimed that FFF is new. In fact it was only relatively recently that I heard anyone call it that, as opposed to FDM.

Hm...

I wonder if I could just do it as a hobby until the patent expires, and then make it commercial. That's what I was planning to do anyway (since it would take a while to learn about all the stuff I need to learn about - probably a few years or more).

And it's not like the parts I would make and use would be a major factor in the profitability... they don't even have to be precise, I just want them to be because I'm a bit obsessive about certain things. I could make them out of wood/etc, and skip the whole RepRap thing and patent infringement risk, but I've never had any luck with making stuff out of wood (the pieces always come out sloppy looking, and it bothers me, even if it functions the same).



Also, I'm curious as to whether the sale of Darwin kits by Bits From Bytes is infringement... maybe it doesn't apply the same way because he's in Europe? If that's true, maybe I should move (either to a similar location, or, better yet, a place not part of the WTO).
Or perhaps it's not infringement because he's just selling a cartbot... which is not an FDM machine. He also sells an extruder, but the extruder head (and extrusion method) by itself is not covered by the patent...
I suppose if he used a Darwin to RepRap the parts, he would be infringing, but since he's using laser-cut acrylic, it should be fine.
[Note that I'm not complaining about Ian Adkins' actions at all; I think his kits are great for the RepRap community. I am merely trying to find solutions to my own problems by drawing from others' experience (ironically, this is sometimes described as patent infringement; but that's not the intended meaning here).]



Then again, I could just give up the commercial bit of it and self-sufficiently use my devices. It's not the preferred option, however.

I find it tremendously amusing that so many people building Reprap machines harbour the notion that somehow they're going to be able to go commercial somehow, despite of the fact that by definition Reprap is a self-replicating, virally diffused technology.

I've been trying to come up with a way of getting an interesting industrial scale revenue stream out of reprap. It just doesn't happen, for me anyway.

Forrest Higgs Wrote:
-------------------------------------------------------
> I find it tremendously amusing that so many people
> building Reprap machines harbour the notion that
> somehow they're going to be able to go commercial
> somehow, despite of the fact that by definition
> Reprap is a self-replicating, virally diffused
> technology.
>
> I've been trying to come up with a way of getting
> an interesting industrial scale revenue stream out
> of reprap. It just doesn't happen, for me anyway.
>

I agree that once printing goes mainstream it will quickly become difficult to establish a business printing parts for others. But certainly in the meantime there may be a good deal of money to be made in that way. And even afterward there will still be a business in printing things for those who are late adopters. This probably be a very competitive and low-margin business.

However, RepRap also opens up more opportunities, like creating designs for profit and what Joshua is referring to - printing parts for your own custom manufacturing machines and then selling the products from them (note that there is no requirement to disclose the designs for these machines especially if you are only printing them for your own use, making them more difficult to reverse engineer). These are huge unexplored potential revenue streams, and are imho much more interesting than the vanilla "rapid prototyping" industry they displace!

Yeah, I think rapid prototyping is something that should be localized. Stratasys has the right idea, selling the machines instead of just RP'ed objects. The whole point of rapid prototyping is that it's rapid. Waiting a day or more for an object because it has to be shipped is just silly.

By the way:
Forrest Higgs Wrote:
-------------------------------------------------------
> I find it tremendously amusing that so many people
> building Reprap machines harbour the notion that
> somehow they're going to be able to go commercial
> somehow, despite of the fact that by definition
> Reprap is a self-replicating, virally diffused
> technology.

I hope you weren't including me in that group. I have no intention of selling RP machines or RP'ed objects. I'm just interested in making objects for myself which will indirectly help create a profit. The only reason I'm not purchasing an RP machine or having someone with an RP machine make my objects is that I think it would be more expensive (money-wise, not time-wise).

Forrest, I am seriously evaluating ways to make money with the RepRap. My intention is to create a balanced situation. I'm getting a lot of 'fun' out of it, I'm learning a lot from making it, I gain unique production capacity, but I don't want to consume my savings because I have a fascination about world changing technology (not enough resistance!). Given its potential, there has to be a balanced way. Currently I am using up my savings, I removed myself further from my company to be able to work more on this project. This is not ideal, but it will have to do for now.

I am aware of the effects that self replicating technology has, such as making profiting from the machine eventually impossible. I think this is true at a certain point, when there are quite a lot of RepRaps. Before that, there will be a scarcity of parts. But whether it's RepRap parts or something entirely else is a matter of how things develop. With such a versatile machine you can find a market. Even if more people have such a machine, you can identify potential markets before they do. I also think about profit models to be able to support sustainable growth of RepRap, or stimulate adoption (subsidized, or via micro-credit) in under-developed countries.

If I spend as much time on my RepRap as I want to, it would be nice to have a revenue flow to be able to keep doing so. With my companies, I've always combined work that started as a hobby with profit. This ensures that you can keep doing your hobby.

This goal of keeping things balanced is important to me. Besides that, I think it is just as important that I can produce inventions with a professional touch. Would also be nice if further down the road there would be projects to work on, such as an open source vacuuming-robot.

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]

hehe, you stepped on a few toes, Forest...

Demented

Demented Chihuahua Wrote:
-------------------------------------------------------
> hehe, you stepped on a few toes, Forest...
>

Well, there we are then. It bothers me that people get fixated on the Reprap machine and imagine themselves making big chunks of money from selling parts and the like when it is a "means" not an "end".

On my website homepage, I have a quote from Terry Pratchett which gives you an inkling of the vistas that open up to you when you think about Reprap as a means, viz,

...the only tools a dwarf needed were his axe and some means of making fire. That

Forrest Higgs Wrote:
-------------------------------------------------------
> The promise that I see with Reprap is that by the
> time you've built one you basically have the
> skills to build the rest of a modern, clean
> workshop that fits into the places we have to live
> these days.

the majority (probably 80%) of the public at large has neither the time nor the inclination to build a modern workshop, even if it is easy, clean and small. To that segment of the population, RepRap will be important because it *is* their workshop. Unlike traditional machine tools RepRap should soon become easy to use and more generalized, and not require much expertise to operate.

By contrast, there is a large minority - and I think nearly everyone active in the community right now fits in this category - that will be interested in continually expanding their manufacturing capabilities. That's your market for machine-tools-in-a-shoebox. I'm excited for this development too - I want a CNC lathe with instructions that are along the lines of: "Print out parts A, B, and C, and assemble in the manner shown in figure 1. Purchase a standard RepRap electronics set, and connect your motors at attachment points 1, 2, and 3..."

> Think of the robotics clubs at schools. The only
> way they can get things done is to make a team and
> use public school workshops. Drop that scale to a
> bright youngster's bedroom and you see what a new
> Cambrian Explosion of creative energy is about to
> be unleashed. Who knows where we'll go this time?
>

Amen. I'm very lucky to have access to many of these types of resources, but most people aren't. I've also been blessed with a relatively large piece of property in the not-yet-overcrowded suburbs of Seattle where you can still throw out your arms without hitting a neighbor. I've never understood how people manage to live in places like NYC or Tokyo or London... there isn't any room to do anything!

Kyle Corbitt Wrote:
-------------------------------------------------------
>
> the majority (probably 80%) of the public at large
> has neither the time nor the inclination to build
> a modern workshop, even if it is easy, clean and
> small.
>
That lot is already dead from the shoulders up. I'm not worried about them beyond their ability to get in the way.

Kyle Corbitt Wrote:
-------------------------------------------------------
>
> the majority (probably 80%) of the public at large
> has neither the time nor the inclination to build
> a modern workshop, even if it is easy, clean and
> small.
>
That lot is already dead from the shoulders up. I'm not worried about them beyond their uncanny ability to get in the way.

I would like to point out that companies can and do make substantial amounts of money using open source software; I see no reason to believe that simply because it can build itself (much as software can simply be copied/built from openly available source) and anyone who has one can have another there will not be significant possibilities to make money off of the RepRap. It certainly contrains the model of "build lots of repraps and sell them" somewhat, but not even as much as you might think; an injection-moulded copy could reasonably be cheaper than an RPed one, a company might sell support/service for repraps such that people simply do not have to know how to fix or use them and can just spend a bit more and get their parts (Which could be worthwhile depending on what time is valued at anyways)... and then there's the just flat-out going for selling parts and/or designs that are possible and cheap via RepRaps. Just because the material cost for running the iTunes store is tiny per song doesn't mean that the charge per track doesn't get paid and go somewhere, and if RepRap gives us for material goods what a speaker gives us for music...

I would be very surprised indeed if people did not find ways of making good money off of the RepRap. The whole project really would be halfway to failure if they couldn't.

Forrest Higgs Wrote:
-------------------------------------------------------

> A situation like this existed for a few decades in
> the late 19th century when the bicycle became a
> common mode of transportation. Little shops
> dedicated to making bicycles sprang up in back
> sheds of numberless houses. People with a little
> bit of money equipped them with the crude, and
> this is very important, cheap machine tools of
> that era and a Cambrian Explosion of creativity
> was unleashed. By 1910 cars and trucks were on
> the roads and airplanes in the skies, all because
> of these backyard and small-town workshops. Check
> out the early careers of people like Karl Benz,
> Wilbur and Orville Wright, Henry Ford and the
> like. You'll find that it pretty much all began
> with the small machine shops necessary to build
> bicycles, something just about everybody who could
> scrape the money together was doing.
>

I think it's funny you mention the bicycle Forrest. I see the complete production of a bicycle by a RepRap descendant as a significant milestone considering people will be printing their own laptops long before the first bicycle is produced (at least that's my guess based on the starting substrates).

One change I see the RepRap effecting eventually is to make living in remote areas and even islands significantly more tolerable. If you have a Santa Claus machine down at the village center it won't matter that you're 10,000 miles from anywhere unless you need something exceptionally rare or at the very cutting edge of technology (which will still require industrial temperatures, pressures, tolerances, etc.)

rodzite Wrote:
-------------------------------------------------------
> I think it's funny you mention the bicycle
> Forrest. I see the complete production of a
> bicycle by a RepRap descendant as a significant
> milestone considering people will be printing
> their own laptops long before the first bicycle is
> produced (at least that's my guess based on the
> starting substrates).

Bicycles are the sort of product that are pretty uniform and already ubiquitous enough worldwide that it will be really hard to RP one for a competitive price (economies of scale). Customization/replacement of broken parts would be a good way to enter the bicycle market.

> One change I see the RepRap effecting eventually
> is to make living in remote areas and even islands
> significantly more tolerable. If you have a Santa
> Claus machine down at the village center it won't
> matter that you're 10,000 miles from anywhere
> unless you need something exceptionally rare or at
> the very cutting edge of technology (which will
> still require industrial temperatures, pressures,
> tolerances, etc.)

Agreed, especially if you can get internet access.

I'm acquainted with someone who has made several research expeditions to Antarctica in the international labs down there. She got really excited when I told her about RepRap because of the potential it has for allowing someone to make replacement parts for failed pieces in remote regions. When you're in the middle of nowhere the fact that a certain component is readily available in the industrialized world is little comfort, I guess.

Quote

it will be really hard to RP one for a competitive price

But that could be applied to practically anything the reprap makes, surely? Aside from custom parts which cannot be obtained commercially, everything else can be made cheaper and faster using conventional production methods. 'Course, the printed version doesn't have huge wads of tax and middleman profit slapped upon it...

Also, speaking as an inhabitant of somewhere where bikes tend to be pretty expensive (it was cheaper for me to buy a second hand bike on ebay, and travel 100 miles to collect it than it was to buy one locally), 'competetive pricing' is a relative value.

I read rodzite's comment as being more about the fact that it would very exciting if we *could* print a bike, not that we *should*. Bikes are a world away from the soft plastic things we're making now, after all; the ability to make such a thing implies a very capable printer indeed.

Ru Wrote:
-------------------------------------------------------
>> it will be really hard to RP one for a competitive
>> price
>
> But that could be applied to practically anything
> the reprap makes, surely? Aside from custom parts
> which cannot be obtained commercially, everything
> else can be made cheaper and faster using
> conventional production methods.

But you see, "custom parts which cannot be obtained commercially" is such an incredibly huge and unexploited market that it may well dwarf the current mass-consumption one. The "long tail" applies well here.

There's also the fact that a RepRap is an RP device, and RP stands for Rapid Prototyping. Not manufacturing. Prototyping.

Though there are some instances where a technique similar to RP is used in manufacturing, prototyping methods are usually more expensive and less efficient per unit than mass production manufacturing methods.

Joshua Merchant Wrote:
-------------------------------------------------------
>
> Though there are some instances where a technique
> similar to RP is used in manufacturing,
> prototyping methods are usually more expensive and
> less efficient per unit than mass production
> manufacturing methods.
>
Depends on how long the production run is and how complex the object is that you want to manufacture. The shorter the production run and the more complex the object the better the economics of using 3D printing look.

Forrest Higgs Wrote:
-------------------------------------------------------
> Depends on how long the production run is and how
> complex the object is that you want to
> manufacture. The shorter the production run and
> the more complex the object the better the
> economics of using 3D printing look.


True, and that's why I included such exceptions in the rule, but most of the time it doesn't apply.

It should, however, be noted that even with some non-RP methods of manufacturing, RP can still be useful. For example, if you wanted to manufacture a complex object which requires a mold, you could create the mold using RP methods (directly or indirectly). It would be much faster to cast each subsequent piece than to extrude each one, layer by layer.
Though, this really isn't RP-style manufacturing; it is rapidly creating a prototype and then casting duplicates of it.

In any case, if you're creating an object that is so complex that each unit must be RP'ed, then you really should be asking yourself why it is so complex and why a simpler part can't be implemented. That alone should take care of most such instances. For the ones it doesn't take care of, consider inventing a new manufacturing technique. Sometimes, though, this invention process is more expensive than just RPing the batch and moving on.

Edited 1 time(s). Last edit at 08/08/2008 03:16PM by Joshua Merchant.

The economics of 3D printing are entirely different depending on which end of the supply chain the printer is. For a company or individual to produce lots of things the same it will always be more economical for them to tool up with hard tooling and produce things very cheaply with some form of molding process than it will be to RP them. But things produced cheaply cannot be bought cheaply by the time you add the manufacturer's profit, transport to the shop, the shop's profit and your transport costs to take it home.

On the other hand for many people to make the same object on their own RepRap will almost always be cheaper. Even for reasonable substantial objects the plastic and electricity is only pennies.

We can see this with RepRap itself. I have made all the plastic parts for about $20. Even if you tooled up an injection molding process you could not afford to sell a set of parts, including profits, amortized tooling, packaging and shipping, for $20.

IMHO the only way to make money with RepRap is to sell downloads of STL files or make small numbers of very bespoke items and sell them for a high price.

Edited 1 time(s). Last edit at 08/08/2008 04:00PM by nophead.

---

[www.hydraraptor.blogspot.com]

The act of making the end product is called Rapid Prototyping. The act of making molds, jigs, or other things associated with product production is called Rapid Tooling. Either way, it saves a siginifcant amount of time and money over conventional methods as long as production runs are less than a given amount.

From "Rapid Prototyping-Laser-based and Other Technologies" by Venuvinod and Ma 2004 pg 11-12

"""
How expensive is RP? The answer depends upon the part material, the desired part accuracy, the RP technology used, and how well the equipment has been tuned for the particular job. Hopkinson and Dickens recently compared the costs of producing thermoplastic components weighing in the range 3.6-760g by injection molding and SL. ... Note that while the unit cost associated with injection molding falls steeply with increasing production volume, that associated with SL is insensitive to production quantity. This is because the cost of SL has no upfront tooling costs, i.e., its cost is almost entirely made up of the machine and material costs. ...the break-even quantity could be increased from 7500 to 27000 through tuning. Clearly, with further reduction in machine and material costs, RP can start competing in high volume production too.
"""

That is with a commercial machine. The machine costs for us are truly trivial and thus make production more economically viable and competitive for much higher part runs than we would normally think. RepRap is on much firmer ground than we imagine from a competitive standpoint.

Demented

It might break even cost wise at 7500 but FFF is so slow compared to injection moulding, the only way to make 7500 of something would be to have 1000's of RepRaps. No problem if they distributed in people's homes but at $400 each we are talking millions of $ for a company to be producing things in volume with them.

---

[www.hydraraptor.blogspot.com]

Yeah, but how long does it take to set up an injection mold? How much time, manpower, technical ability? I believe they are talking about commercial machines here using Laser Sintering so the print speeds are much faster than our RepRaps right now. Obviously you can deduct some from that number to account for this. Even if you reduce it by a power of 10, you still have quite a respectable number for this technology. And, of course, this only gets better with time.

Demented

Quote

IMHO the only way to make money with RepRap is to sell downloads of STL files or make small numbers of very bespoke items and sell them for a high price.

You can get robotic machine quilting machines and these are in many ways similar to Rapid Prototyping machines. They use a Cartesian robot to move a sewing machine along a 2D path as it sews the pattern onto the fabric.

The patterns used to drive these machines are not much more than a simple vector graphic (downloadable over the internet), however, some of these can sell for up to $15(aud). And you can get literally thousands of people buying them each year.

It takes a few hours to design and create these files, so in a week you could easily make 20 or 30 of them.

Some simple maths later (average $10, making 20 in one week, and around 1000 people buying them):
This means that doing this you would stand to make around 200,000 in one year, for 1 weeks initial work (and setting up and maintaining an online shop that only deals with electronic media, no actual material goods).

Think of this in terms of a RepRap. You might develop a "pattern" (stl file) for a particular object. You might sell this stl file for $10 to $15 on an on line website. You might then have a market of millions for it.

This is where the powr of the RepRap is for commercial use, not selling the end product, but selling the stl files.

Edtharan Wrote:
-------------------------------------------------------
>
> Think of this in terms of a RepRap. You might
> develop a "pattern" (stl file) for a particular
> object. You might sell this stl file for $10 to
> $15 on an on line website. You might then have a
> market of millions for it.
>
> This is where the powr of the RepRap is for
> commercial use, not selling the end product, but
> selling the stl files.
>
ROTFLMAO! Yeah, right. That will work for maybe two days till the stl file shows up on Pirate Bay.

Files can still be sold over the internet, even with the prevalence of p2p sites. iTunes still has a successful business model.

But you're right, I wouldn't bet my future income off of it. Best to find some value added.

Quote

ROTFLMAO! Yeah, right. That will work for maybe two days till the stl file shows up on Pirate Bay.

Yet people can already make a living (and do) selling patterns for Quilting Machines, despite sites like pirate bay.

Actually, would it be possible to create a "Watermark" in an stl file where parts of the internal structure (internal supports, etc) reflected an ID number without impacting the overall integrity? Each stl file could then be automatically adjusted at the point of sale to include this "watermark". Just the presence of such a device would discourage piracy (but of course never eliminate it).

AFAIK stl files only represent the outer sides/faces of a object so internal structures are 'not allowed', but may possible
this would probably depend on how strict the renderer or printing software is
Any logos, codes etc on the outer surfaces would only be limited by imagination

I think major reasons for piracy are over-charging and lack of convenience (place of the 4 marketing P's) or availability. People do pay for music if prices are realistic, perhaps not all of them all the time, but more if the value added is big. You could also envision a system where creating products for personal use are always free, but when you earn money with it, there's a small fee per part. That would allow more designers to go through the trouble of designing parts.

If you look at how the world should be, you might think: I want people to only develop parts open source and only as a voluntary think without them earning a penny. You want the have designers that just enjoy developing something (for fun (seeing people like what you made) and perhaps earning some respect). You should also have those that design for a living. It would allow much more designs to come online and allows a revenue stream to people that actually create value (so the majority goes to the artists, not the distributors or agents).

I wonder if digital watermarks in 3D design files could be more robust than ordinary 2D graphical watermarks.

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]