Making reverse engineered Altium designs public

[eecook]

Hi All,

So, my friend has reverse engineered this comercial board from schematics down to the layout. He had a blast during the experience and would like to start doing it regularly and post the designs for everybody to download.

He is worried about the legal aspect, basically getting sued.
Is there any way to know which product/s is unlikely to generate legal problems?

Thanks,

Simon.

[TiN]

Obsolete, out of support products (esp when original manufacturer gone) may be easier to get away with , than modern current tech.
But correct answer is always "it depends".

[janekm]

This may actually fall into the category of "it's no fun getting sued for doing something legal".

Reproducing a circuit schematic from scratch has been found to not be covered by copyright: https://www.lexology.com/library/detail.aspx?g=ecbb24af-4a0a-4971-9022-2edfeac253eb

Doesn't stop a large company's legal team suing you vexatiously if they really don't like you doing it.

[hans]

I think this is the sort of thing that no company admit they do, every engineer at a company admit they do (in person that is), but the real details stays behind closed doors because of legal consequences.

For sure it sounds like a blast to do, but I don't think it's easy to go public with the work. Unless you think your legal firing power is more than that of a small company even.

[TerraHertz]

So, my friend has reverse engineered this comercial board from schematics down to the layout. He had a blast during the experience and would like to start doing it regularly and post the designs for everybody to download.

Would 'your friend' like to talk about the methods used to reverse engineer boards?
It's a topic I find quite interesting, since I do it now and then too.
Hard enough just with single layer PCBs. Pic is an example, with 'done' markings on traces and components, using photoshop. The schematics are still messy hand-drawn first rough version, not tidied up at all. It's an LCD screen backlight power supply. Not of much interest to anyone, I think.

Two layer PCBs are OK too, though usually it's worthwhile removing large components during the process. I take photos of both sides, overlay and align in photoshop using partial transparency, and create more layers for working annotations.
Boards with internal layers used only for power and ground, are still not too hard.
Boards with traces on multiple internal layers - I don't have a practical method.

Why don't you post an example here?

[NiHaoMike]

I remember reading that it's possible to find reverse engineered schematics of iPhones and a few other very popular consumer products that commonly show up in repair shops. Perhaps the key is to make everything completely anonymous when releasing it? And do it from a public hotspot and/or VPN just to be safe?

[ebastler]

Reproducing a circuit schematic from scratch has been found to not be covered by copyright: https://www.lexology.com/library/detail.aspx?g=ecbb24af-4a0a-4971-9022-2edfeac253eb

Thanks for the link. The court ruling cited there speaks about both, circuit diagrams and layouts.

Copyright does cover the graphical rendition of circuit diagrams and PCB layouts.

• A reverse-engineered circuit diagram will most likely look very different from the original. (You probably haven't even seen the original one -- otherwise, what would there be to "reverse engineer"?) No problems with copyright in that case.
• For the reverse-engineered PCB layout, one may be tempted to use the same component and trace arrangement as for the original. (Maybe even based on scanning or X-raying an original PCB.) If one does that and publishes the result, it would most likely be a copyright violation. In the court case cited above, the court's reason for not finding a copyright violation was that the PCB layout was quite different from the original in several aspects; and that is the approach I would recommend for reverse-engineering.
  

In addition, there is the matter of patents. Aspects of the board you reverse-engineer may be covered by patents, and it may be difficult to find out whether that is the case. But as long as you don't start making or distributing a product based on the design, you would not violate any patent. When you publish the reverse-engineered design, it is probably a good idea to point out the possibility of 3rd party patents in a disclaimer, so the publication cannot be seen as "inducing infringement" by others. This all assumes that the reverse-engineering is done to satisfy technical curiosity or learn about the solution, rather than wanting to copy it for commercial gain.

EDIT: The OP does not mention any firmware. If firmware were part of the original design, that poses the biggest obstacles to publishing a reverse-engineered design. Binary copies of programs are universally considered to be protected by copyright, to my knowledge. Publishing a disassembled version should still violate the copyright, if it can be used to assemble an identical binary. I don't see a legally "safe" way to publish the firmware as part of a reverse-engineering project.

[donotdespisethesnake]

Copyright does cover the graphical rendition of circuit diagrams and PCB layouts.

No it doesn't, that's a common myth. Generally speaking, copyright does not apply to "useful articles" such as engineering drawings, although there are specific exceptions, e.g. 'vessel hull designs'.

[ebastler]

Generally speaking, copyright does not apply to "useful articles" such as engineering drawings, although there are specific exceptions, e.g. 'vessel hull designs'.

Would you have a reference for that? I don't think it is generally correct.

For "trivial" diagrams, or if there is essentially "only one way" to render something, copyright protection may indeed not apply. But if creativity went into the creation of a diagram or layout, e.g. to make the functionality very clear by chosing a clever arrangement of components on the schematic, copyright does protect the rendition IMO.

I can't find a ruling specific to circuit diagrams right now, but this article summarizes the arguments from a related court case: http://seattlecopyrightwatch.com/copyright/admitted_copier_of_diagram_on_the_hook_for_copyright_infringement/

EDIT: Found this blog post by a law firm. They make a pretty sweeping statement along the lines of "program code, PCB layouts, and schematics are all protected by copyright", without mentioning any limitations. That's probably over-simplifying matters, and they may be trying to drum up some business with this position... But I am pretty sure that the simple "not applicable at all" isn't correct either.
http://garson-law.com/ask-dr-copyright-protect-your-software-circuit-boards-and-mask-works/

[David Chamberlain]

I think it really depends on the product. If it's an iPhone then you can bet you will be sued to an inch of your life. On the other hand if it's a reference design from a chip maker then that gets a lot more blurry. I think it's the latter because you mentioned 'Altium designs' but I could be wrong.

If it is a reference design then consider that the chip manufacture publishes those designed to aid engineers in how to layout and use said chips. For example SMP products datasheets usually have a recommended PCB layout and unless you know more then the people who designed the chip (I'm not one of them) you just directly copy the layout. The chip manufacture of course OWNS the copyright to that layout but they are not making bucks if they sue every tom&dick for directly copying it.

The twist comes though when you then on-sell and market your product with that design. Consider the chip manufactures design is meant to put the component in the best possible light to entice engineers to purchase it. If you use their design and do something counter to that idea then you may expect some back lash.

So a given design is either Intellectual Property (in the case of the iPhone) and worth something to a company or a means to promote a products that the design is based around.

[ebastler]

So a given design is either Intellectual Property (in the case of the iPhone) and worth something to a company or a means to promote a products that the design is based around.

That's a step back from the more differentiated discussion we had started above. "Intellectual property" is a general term, which encompasses both, copyright and patent protection.

Many electronics designs (certainly including the iPhone) will be covered by patents. But that does not make it illegal to publish details about the designs. Only using the designs commercially, to offer products or processes, will be limited. In fact, the whole patent system is built around the concept that new ideas should be documented and published (to advance technical progress), but that the inventors should be given a time-limited protection in using them commercially (to reward innovation).

Schematics and/or layouts also may or may not be covered by copyright -- in my understanding, depending on how much "creativity" went into them, beyond implementing pure technical requirements. In that case, publishing the original schematics and layouts or reproductions thereof may constitute a copyright violation, even if one does not build and commercialize the actual product. But if you redraw the schematics and layout with a different arrangement of components and connections/traces, you will not violate the copyright. Copyright only protects the "artistic" rendition of the design, not its functionality.

[David Chamberlain]

Actually I'm going to argue that Intellectual Property is less about Copyright and Patents (that I never mentioned) and more about the degree of financial heft an angry corporation may put towards OP's 'friend' - no really it's not OP who did the copying for sure

Copyright is a given right to the original creator of a work and does not need to be applied for. It is one of those special natural justice things like freedom of religion (god help us). To prove in court that I was the copyright owner of a work means to provide evidence that it was not so derivative to be construed as a breach of another copyright. That is to say I did not just rip of someone else work and call it my own. How much evidence I need to prove that depends on the drunkenness of the presiding judge and phone-it-in attitude of the apposing council.

A Registered Design Mark is a way to put a time stamp on the creation of a design to aid in the resolution of any future copyright disputes. But however you cut it a copyright is week protection in electronics considering how many ways a PCB can be set out.

A patent, seeing as you asked, is not something you can apply to a PCB layout due to it not being definable as a Patentable invention, further details of that are here.
https://en.wikipedia.org/wiki/Patentability

So the natural question is how could Apple get a patent on a PCB layout? Well that is a whole different kettle of fish that in some ways paints the picture of a corrupt US patent system.

[ebastler]

David -- not sure whether we keep saying the same things here, just stressing different points or driving for clarification of different aspects? My main reason for arguing is that I want to answer the OP's question -- what can 'his friend' safely publish and what not?

A patent, seeing as you asked, is not something you can apply to a PCB layout due to it not being definable as a Patentable invention, further details of that are here.
https://en.wikipedia.org/wiki/Patentability

Two comments here:

(a) I am pretty sure that specific aspects of a PCB layout are patentable -- where the layout itself implements certain functionality. Think of various antenna geometries realized via PCB traces, for example.

(b) In the more typical case where the graphical layout is not patentable, the circuit functionality it implements may well be patented. So if you clone the PCB, populate and sell it, you are violating the patent.

[David Chamberlain]

Well OP has really not provided any detailed information that could be used to answer OP's question so at this point it's all just a bit of fun and a whole lot of speculation. I'll admit I usually read and respond to the OP's question without trawling through the evolving thread so it's possible I have missed the direction it has gone in.

My aim was just to point out the basics of product protection as I understand it and to point out that while we all like to think the rules are solid, it really comes down to who your up against.

So to your points a) Patents (should) provide protection to processes not designs, so if some unique process is developed around say putting copper on an FR4 for 5G applications and was novel then perhaps yes. b) again patents may very well apply to the individual components but their placement in a wider functional circuit I would say no.

As I mentioned my idea of what a patent is is completely idealistic and the true Patent system seems to no longer have any resemblance to it's original intent.

So my actual advice to OP, just make the thing and wait for the cease and desist letter.

[amyk]

Release anonymously and preferably host in a country/service that doesn't care about legal threats.

[Richard Crowley]

Hasn't this general topic been discussed several times by Louis Rossman?  Didn't Apple use their lawyers to try to stop him from showing schematic diagrams in his videos because Apple deemed them legally "protected" (by some undisclosed theory)?  I don't think the schematics were actually Apple documents, but rather reverse-engineered documents created by 3rd party sources.  Rossman stood up to them and they apparently backed down because it appears to be back to business as usual on his channel.

[TerraHertz]

So my actual advice to OP, just make the thing and wait for the cease and desist letter.

He didn't say anything about making physical copies. Just releasing schematics and board layouts. Presumably for fun, and to enable servicing by owners of the equipment.

The kind of information that honest, moral companies like HP and Tektronix used to publish by default, in those beautiful and very detailed service manuals. Before the lawyer c*nts took over from the worthy original company founders. Who were true Engineers, trying to better humankind's lot.

[Richard Crowley]

Back perhaps before many here were born, we had "Sams Photofact" documents. They were quite comprehensive component-level service manuals for a wide variety of consumer gear. Back from the days of firebottle, point-to-point wiring.  I believe they were reverse-engineered, independent 3rd party documentation, without support or participation from the manufacturer.  Apparently they are still going at: https://www.samswebsite.com/

[ebastler]

Well OP has really not provided any detailed information that could be used to answer OP's question so at this point it's all just a bit of fun and a whole lot of speculation. I'll admit I usually read and respond to the OP's question without trawling through the evolving thread so it's possible I have missed the direction it has gone in.

Yes, you might have missed something. The points I tried to make repeatedly, as advice to the OP, are:

• Don't publish direct copies of the schematics or layout, draw your own versions and make them somewhat different from the originals (graphically, not functionally). That avoids copyright infringement.
• Don't sell products or processes, only publish design information. That avoids patent infringement.

If he stays within the above constraints, the OP (or his friend...) should be fine. The only other pitfall I can think of is if the original information was obtained under a confidentiality agreement or other contract that explicitly limits what one can do with it.

[eecook]

It's roughly the same method I use. For 4 layers I just sand paper the pcb (with water)...if you are gentle and wait long enough, the inner layers will reveal themselves.

[eecook]

EDIT: The OP does not mention any firmware. If firmware were part of the original design, that poses the biggest obstacles to publishing a reverse-engineered design. Binary copies of programs are universally considered to be protected by copyright, to my knowledge. Publishing a disassembled version should still violate the copyright, if it can be used to assemble an identical binary. I don't see a legally "safe" way to publish the firmware as part of a reverse-engineering project.

Indeed, I am not interested in firmware.

[eecook]

When you publish the reverse-engineered design, it is probably a good idea to point out the possibility of 3rd party patents in a disclaimer, so the publication cannot be seen as "inducing infringement" by others. This all assumes that the reverse-engineering is done to satisfy technical curiosity or learn about the solution, rather than wanting to copy it for commercial gain.

I like that, good idea.

[eecook]

I am basically curious (it's me, I was just kidding) as to how the community would react if all of a sudden, there's a proper schematic and layout files for some popular product. Would people download it? Would they even care at all? Would they ask for more? Would they be angry at me using some convoluted ethical argument? Who would be interested?, Etc
But I am not going to move forward if there's trouble lying ahead. Just wanted have a better understanding of the situation. Maybe it would be less risky and equally useful to release only schematics.

[julianhigginson]

Sounds really cool!
I'd be keen to see schematics of anything interesting that you share.

[eecook]

Here's a sample view of a reverse engineering job I did where all 4 layers where revealed, aligned and fake coloured using GIMP.