Hacker News new | past | comments | ask | show | jobs | submit login
Is 3D printing being held back by an invalid patent?
190 points by K0balt 38 days ago | hide | past | favorite | 42 comments
Parts made using the FFM (FDM) process are stronger in some directions than in others. This is due to imperfect adhesion between printed layers. This directional or anisotropic structural strength is a significant limiting factor for the structural integrity of 3d printed parts made with the common FFM process.

A method to drastically reduce this effect is described in an expired Stratasys patent, US1997/5653925A. As of this writing, I have not seen this feature implemented in any of the common open-source slicers. I would have expected this now-public-domain knowledge to have made its way into the slicers we use since the patent expired in 2017.

With some investigation, I believe this may be because of the 2023 patent US2023/11813789B2, to my knowledge only implemented as a proprietary in-house product of a 3d printing service house which appears to make the same claims illustrated in the 1997 Stratasys patent.

I am not expertly versed in patent law, but this patent would seem to have been granted in error. The fact that the 2023 patent actually references the Stratasys patent from 1997* makes the issuance of this patent even more baffling. At any rate, this seems like a low-hanging opportunity for a significant improvement in print strength that could benefit millions of users.

*The 2023 patent references the 1997 Stratasys patent as “teaching a process for adjusting the deposition rate .. to provide a predetermined porosity rate” but neglects to mention that the core claim of the patent was the staggered bead heights claimed.

In referencing the 1997 patent the 2023 work also gives a single-digit error in the patent number, as well as other “minor” errors in other patent numbers referenced in the patent. These misleading references could have impeded the examiner’s role in determining the validity of the patent.

https://patentimages.storage.googleapis.com/08/f6/2b/d3d9964...

https://patentimages.storage.googleapis.com/ea/e7/3c/2b836c9...

Looking up the 2023 patent filer leads to this:

https://www.addmangroup.com/




This comment is not legal advice. This comment is merely for educational purposes.

The quote that you have about the 1997 patent, namely "U.S. Pat. No. 5,659,925 teaches a process for adjusting the deposition rate..." is from the specification. That is written by either the inventors or more likely the attorneys filing the application. The specification is not legally enforceable or binding. It is supposed to disclose enough information to let somebody reproduce the claimed invention. As a matter of practice, the specification can say almost anything it wants, including non-enforceable things.

Only the claims are legally enforceable. I suggest taking a closer look at the claims and seeing if these are problematic for your application or not. You may want to consult an attorney if you have further questions.

You can look up the patent examiner's reasoning for allowing the application here:

https://patentcenter.uspto.gov/applications/17667081

Click on "Documents and Transactions" and look for "Notice of Allowance and Fees Due". Also look for the "Non-Final Rejection" for additional information from the examiner from an earlier version of the application.

I should note that it does not appear that the patent examiner considered the 1997 patent directly relevant because they did not cite it in any of the office actions or specifically the PTO-892 forms (search the documents page for 892 to see what I'm talking about). However, they likely did at least look at it since it is listed in the specification.

Again, if you have further questions, you really should consult an attorney. This is not a good forum for this kind of discussion, in my opinion.


  > The quote that you have about the 1997 patent, namely "U.S. Pat. No. 5,659,925 teaches a process for adjusting the deposition rate..." is from the specification. 

  > I should note that it does not appear that the patent examiner considered the 1997 patent directly relevant because they did not cite it... However, they likely did at least look at it since it is listed in the specification.
Again that quote from the specification gives the wrong patent number, so we don't know that the patent examiner looked at it.


Wow, great catch! I just copied the text idly without reading it closely. I agree that the examiner may not have looked at the discussed reference (US5653925A).

(Disclaimer: this is not legal advice.)


testing


> This comment is not legal advice.

Serious question, why do lawyers always do this? No other profession seems to feel the need to disclaim their dialog.

Is there some precedent for attorneys discussing their area of expertise and unwittingly becoming someone’s counsel?


No problem, I'll give you a serious answer.

First of all, I'm not a lawyer. I'm a mechanical engineer.

And to answer your question, yes, many attorneys are in fact worried about giving legal advice online [1], especially the fact that it may imply some sort of legal representation being extended. This is why I added the disclaimer at the top and why I emphasized that this is not the right forum for this and that the original poster should in fact get an attorney (which strongly implies that I am not their attorney).

This is also why I did not discuss the specifics of the claims being made in the different patents, because that could imply that I am giving specific legal advice. I am not giving any legal advice. Instead, I just discussed broadly how the patent system works in this particular context, especially in regards to what parts of a patent are legally enforceable and how examiners examine patents (and how you can look up what they did).

[1] https://en.wikipedia.org/wiki/IANAL

(Extended disclaimer for the sake of emphasis: I am not a lawyer. I am not your lawyer. This is not legal advice. My comments in this thread are merely educational in nature.)


If you were a lawyer then you'd be bound by bar standards and such. But you're not, so why the disclaimer? Is there some legal jeopardy a non-lawyer places themselves in by talking about the law? If so that sounds positively dystopian.

More generally I am entirely supportive of not giving unsolicited advice of any sort. Not for legal reasons, but because it's presumptuous, rude, and often even pompous.


I believe atrettel wanted to be transparent about the nature of their reply more than protecting themselves from jeopardy. Also, this is not unsolicited as OP has asked a specific question to the community.


I wanted to be both transparent and protect myself by adding the disclaimer. I perhaps should have added that I am not a lawyer to make it clearer.


It is usually illegal to give legal advice or represent someone without being a lawyer, so this disclaimer is meant to protect yourself against claims that you're doing so. I think.


> It is usually illegal to give legal advice or represent someone without being a lawyer

This is not true. It is usually illegal to give someone the impression that you are a lawyer when you are giving legal advice (it is fraud.) So careful people lead by making sure that no one has the impression that they are a lawyer when they are giving legal advice, and lawyers who are just talking off the top of their head and not being paid warn you that they are not giving you advice as a lawyer (i.e. they are not working at this moment as a lawyer.)

People who are not lawyers give each other legal advice all the time. It's a good thing if the advice is good, and a bad thing if the advice is bad, but not a crime.


Engineers are often bound by similar standards and organizations. Depending on the situation they may be licensed and PE, etc. We are often held accountable for our actions and statements in exactly the same fashion as lawyers are. Therefore we are trained, and it's good form to, be very clear about the circumstances around any advice we give.


This makes sense to me and other comments have covered why plenty well, but what does frustrate me is the constant insistence that nothing should ever be discussed in public and you must consult your own lawyer privately for every bit of basic information. That's just not realistic for the majority of people and even plenty of small businesses with how much lawyers cost.


Risk.

Whenever you provide legal advice, you take on risk. And big or small, there is always the possibility that that risk might end your career and cause years of training to go up in smoke.

Lawyers are experts in risk management, which is why you see them trying to minimize risk and exposure in every way possible.

(Lawyer here.)


Have you ever considered wearing a helmet in day-to-day life? I'm just curious how far flights of fancy might go for people dealing with risk management constantly.


I bet you'd find a higher percentage of lawyers' kids wearing helmets biking than average. And fewer of them owning trampolines.


When I was a kid, I really wanted a trampoline.

Now that I'm an adult and video websites exist, I'm REALLY glad I never got one. I almost certainly would have broken my back or next while attempting a flip, or broken an ankle by having more than one person on it.


There are several other professions which do it. Doctors and financial advisers immediately spring to mind. It's bound to happen in any tightly regulated profession where the professional is expected to give potentially life altering advice as part of their services.


Relevant video on the subject, discussing brick layers, and the patents:

https://youtu.be/9IdNA_hWiyE


The real problem is that patent system seems ripe for abuse. Even if the "invention" is not a valid patent, companies want to patent it anyway, as to force the other side to expend resources to prove it's not valid.

There's a couple Bambu Lab patents that probably should not have been granted with prior art in open source.


> There's a couple Bambu Lab patents that probably should not have been granted with prior art in open source.

The open source world should come together and get them invalidated.


It's not a problem unless Bambu Lab starts suing.

Bambu Lab's biggest competitor isn't Prusa. It's Stratysys.

Stratysys shipments dropped in the past several quarters after the introduction of the Bambu printers about 2 years ago.


This post seems very well researched. It's great that OP has brought this up. If we could push further the state of the art in 3D printing, by simply no longer adhering to an (now proven to be invalid) patent, it's a no-brainer to do so.


Non linear slicing is also patent protected but there are some projects that modify slicers to do it.

I did some experiments with bricklaying using a macro in S3D and in all honesty it’s not worth it.

It doesn’t work all the time causes a lot of slicing edge cases on complex models and the cost in print time outweighs any strength gains.

For properly dried filament and a heated chamber the adhesion gains are within the margin of error.

I can also quite confidently say stratasys also currently does not implements it in GrabCAD* which is their slicers/CAM software and I’m not aware of any of their partner slicers/CAMs that do it either.

*There is a chance that they do it as a post processing step in their printers especially the super high end ones either globally or for select locations within the model but I’ve handled quite a few FDM prints on Stratasys machines and I’ve never actually seen it in a production model.


Some related discussion in here:

Ask HN: Why did consumer 3D printing take so long to be invented?

https://news.ycombinator.com/item?id=42079086


I'm not a lawyer, but have 20+ patents. It's my understanding that if patent A references patent B, it means the examiner thinks that B is not prior art for the claims in A.


For personal stuff you don't have to care about patents anyway. If the slicer can be modded to include this feature then there's nothing stopping people from using it. I'm not sure if just adding the feature in slicers would violate the patent, or using the feature commercially.

Also, if you're printing commercially, I would just go ahead and violate the patent. They are not going to find out, and if they do then just stop using the feature or negotiate a license for the patent.


Since you wrote

> [...] is a significant limiting factor for the structural integrity of 3d printed parts

I went and checked. There is a youtube called MyTechFun[0] who does proper material testing, and being a patreon I have access to the entire table of his results. The comparison I made is tensile strength vs layer adhesion, which is the same test (basically pull until it breaks), but the sample is printed lying down or standing up. I expressed layer adhesion as a percentage of tensile strength, and sorted by this.

The best three I will show here (tensile and adhesion are in kg):

  Name, Tensile, Adhesion, Percentage
  YXPolyer Nylon[1], 63.1, 66.4, 105
  PolyMaker PLA Pro[2], 56.55, 50.6, 89
  PolyMaker CoPA[3], 66.3, 56.6, 85
While the tensile strength is higher for all (except one), it does not seem to be quite as limiting a factor as your claim makes it out to be. We can find percentages of >70% for all major filament types. But to be fair, most of them seem to cluster around 40-50%. While this is not great, given the choice of materials I think it can be worked around quite easily.

[0] https://www.youtube.com/@MyTechFun/videos

[1] https://youtu.be/Q6v4xYrkOnU

[2] https://youtu.be/2dgzwCvSO2k

[3] https://youtu.be/cQb-hbr1KYY


I'm an attorney who litigates patents. I also write about IP litigation at https://ipde.com.

A patent can be invalidated by a court during litigation or through procedures at the U.S. Patent Office, such as inter partes review or an ex partes reexamination.

When asking whether an earlier patent renders a later patent invalid, the fact finder will generally look at anticipation and obviousness. Anticipation means that every element of the claims is literally present (explicitly or inherently) in the earlier patent, under 35 U.S.C. § 102. Obviousness means that the claimed invention as a whole would have been obvious to a person of skill in the art, under 35 U.S.C. § 103.

How do you figure that out? You have to look at each claim individually. If the fact finder determines that the earlier patent discloses or renders obvious every element of the claim, then the claim is invalid.

This comparison is easier said than done, because it is going to depend on claim construction, meaning how the claims are interpreted. In theory, they are interpreted according to how a person of skill in the art would read them, but in practice there is an art to claim construction, and it can be difficult for anyone to predict how a Court or the U.S. Patent Office will construe claims (especially since it is often judge- or examiner-dependent).

I won't get further into burdens of proof and legal standards, but both anticipation and obviousness tend to be a high bar.

From a very quick review of the two patents you posted, it looks like claim 1 the second patent claims, for example, "lowering the vertical axis of the bead height of a bead disposed between a bead on either side thereof." From a quick look at the specification, this may refer to staggering the vertical size of the beads themselves in each row.

It looks like the earlier patent discloses staggering the alignment of the columns of beads, but I don't see (from a quick glance) whether or not it discloses staggering the size of the beads. If it doesn't disclose that element of claim 1 (however the court or PTO interprets it), then it can't anticipate that claim of the second patent. Of course, it could still render it obvious if, for example, there is another prior art patent that discloses varying the beads in the claimed way, and if a person of skill in the art would have been motivated to combine them (assuming the combination also meets all of the other elements).

Each claim in the second patent will need to be looked at individually. Even if claim 1 were invalid, for example, other independent claims may not be.

Keep in mind that everything in this post is hypothetical and I am not your attorney. If you want to try to invalidate the patent, you'll need to get your own attorney involved or otherwise get one to take up the cause, including to take a real look at the patents and prior art and to figure out how best to proceed. It is not a small task.


Nor would I assume, cheap. If the community wanted to fundraise to get the patent invalidated, what sort of money are we talking about? How much if they decide to contest the invalidation? $10,000,? $100,000? $1,000,000? I'm sure it depends on a lot of factors, but getting an idea of how many zeros we're talking about would be useful.


Again speaking generally, attempting to invalidate a patent is going to cost more than $10k, but $100k should be enough to do an ex partes review petition. If you want to do inter partes review, which is much more involved but may also have a higher chance of success, it's more expensive. The total cost will depend on what attorneys you use and how they go about it (which also impacts the likelihood of success).


> 35 U.S.C. § 102. Obviousness means that the claimed invention as a whole would have been obvious to a person of skill in the art, under 35 U.S.C. § 103.

This is the part that always gets me. Having read through over a couple thousand patents, it is my opinion that the vast majority of them are obvious to anyone skilled in the art. Some of them are so ridiculous that a university student about to finish a relevant degree would consider the claims obvious.

This bothers me deeply because patents that should not have been granted force us to play a patent arms race. If the other side has an axe, you have to have an equal or better axe...and it gets nastier from there.

In some industry sectors you'd be crazy to put out a product without ensuring you have enough legal weapons of IP war to protect yourself from other IP as well as slow down or eliminate copy cats and competitors who will gladly take advantage your your "R" (Research) at zero cost. In "R&D" the "R" is usually the most expensive phase. Once you know what you are building the "D" tends to be simpler, shorter and costs significantly less.

It stands to reason that getting a patent should become more and more difficult over time. As more is invented the "art" and those skilled in it become more sophisticated. Which means the rate of true invention should (not fake invention) should come down to an asymptotic level. We should see less true patents per year, not more fluff patents per year.


Apart from the patent issues.

There has been some tinkering with this technique in the community.

From what I could gather this sort of bricklayer interlocking does basically nothing to help part strength.

And you can not do it on the outside walls because they would not be smooth.

Which means you would need to transition half way in the wall from stacked to interlocked which is possible but needs a very well tuned printer laying lines a fraction of the nozzle diameter.

All in all does not look like it's worth the effort.


The only thing that matters are the claims. You have to read and analyze them in order to understand where the legal IP fence is being built. If you are not used to reading and dissecting patents the claims can be difficult to understand.


I’m not sure but this may be of interest to the Electronic Frontier Foundation https://www.eff.org/pages/legal-cases


Can't you request a separate patent if you're making substantial improvements to it? No idea if that's what happened here, but I'm just saying it's not surprising that a new patent would reference old patents.

I assume that for most things the current tech is durable enough. If it needs to be more durable, going with cast, forged, or even mim metal instead of geometry changes would be better. If you really want geometry stuff, just design one yourself. Use three nozzles - a primary vertical down, two trailing secondaries opposing each other on slightly doward angles and slightly above the primary. Put all the nozzles on the same head but with steppers driving eccentrics to adjust height for each nozzle for areas where only a combination or single nozzle can be used. You'll effectively build a three layered wall with offset layers. Simultaneous printing allows better adhesion by row, and the staggered design provides increased interconnectivity between each strand.


The purpose of mentioning the older patents is to argue that they are not relevant as prior art. Whether that is true is another story.


Can't you request a separate patent if you're making substantial improvements to it? No idea if that's what happened here, but I'm just saying it's not surprising that a new patent would reference old patents.

Edit: sorry, seems to be a dupe.


I think the issue is that there doesn’t appear to be anything substantially different in the newer patent this YouTube video covers it pretty well https://youtu.be/9IdNA_hWiyE


disclaimer, I am not an expert in patent litigation but have been involved in and hold several patents) also, hold my beer and watch this.

Having read the claims on both patents (US2023/11813789B2, US1997/5653925A) , it still seems pretty sus to me on both sides of the fence.

The basic concept of of the geometric layer pattern is identical in the claims and illustrations in both patents, but the expired patent mentions this in the context of "density" and "porosity", while the newer patent describes the identical process for for the effect of interlayer adhesion. This -might- be enough to make a meaningful distinction.

The new patent fills in a little bit more detail on the process, and is much more narrow in scope. But...the expired patent claims -do- specifically mention strength in terms of density in claim 6: "the pressurizing Supply source is adjusted to provide a porosity in the article of between 1% and 9%, so as to form an article of a strength sufficient to maintain the integrity of the article."

Claim 6 refers to claim 3 which refers back to the primary claim of modifying density through the feed pressure and geometric distribution (specifically brick-like, as depicted in figure 2 - preferred embodiment) deposition of the filament.

In all, it is surprising to me that the patent was granted as is. I have had filings rejected for much weaker similarities to existing art when basing a patent on new applications of existing processes... and this is arguably not even a new application.

The described process (hexagonal or brick-like bead geometry) is identical to the process shown in figure 2 of the expired patent and described in the preferred embodiment.

What is missing from the expired patent is a specific claim on the geometry of the preferred embodiment given in figure 2. This could be construed in a sense to say that the preferred embodiment itself was not the subject of the patent, i guess?

In any rate, the 2023 patent clearly does not describe any actual innovation, but rather an attempt to lock down a competitive advantage by locking everyone else out of an invention that has been clearly documented since 1997 and released into the public domain for 6 years prior to this patent.

Although the patent holder appears to be using the patent in a proprietary, in house process, this still smacks a bit of troll dung.

It doesn't take a rocket scientist to see that there was exactly zero innovation here, and the patent applicant was clearly aware of the prior art, perhaps even being inspired by ambitious nonspecificity of the expired patent. His figures 1 and 2 (the primary basis of the new patent) could just have well been lifted directly from the Stratasys patent.

This might be one for the courts. What we need is someanon to fork slic3r with this feature added and dump it on torrent.


So there's two uses for patents. One is to keep others from doing what you're doing. The second is to keep anyone else from getting a patent on it, and then using it to keep you from doing what you're doing.

This could be (currently) used in the second way. ("Defensive" patent, as it is sometimes called.)

The problem is that patents can change hands, and in the hands of a troll, this patent could be a problem.


That would certainly be a less clown-world explanation. Thanks for the insight.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: