Re: [EE] Reverse engineering a product

Thanks Denny.
So far, they got someone to CNC-mill the boards down layer by layer and 
they've got images of all internal layers.
They found a company who will reverse it (similar to what you mention) 
to BOM and gerbers, but you're right in that some info will still be 
missing, such as specific types of caps, maybe resistor wattage, cap 
ESD, etc.
So the only thing left to do from there is to reverse engineer to 
schematics, so that they can make other changes later.

Cheers,
-Neil



On 7/26/2025 2:49 PM, Denny Esterline wrote:
> I completely agree with other people's comments on this thread. As
> described, this seems like a liability nightmare and more than a bit
> shady. Even if everything else is on the up and up, not telling the
> client about this is its own red flag.
>
> All that said, from an actually doing the work standpoint, I would
> probably take a functioning unit. Desolder every component making note
> of location. Make every effort to measure every parameter you can and
> recreate the BOM.
> Clean the board and scan it.
> Remove each layer and scan it again. (chemistry, sandpaper, milling
> machine as needed)
>
> Then I'd look to put the scans back together in layers in some graphics program.
> Use those graphics directly as the new board files.
> You'd likely have to create drill files and component location files
> by hand. Hopefully the graphics program would give you the tools to
> measure those locations.
>
> After a lot of work you could come up with something that is very very
> similar, but not exactly what they started with.
>
> Would that be good enough for your customer? Would that be good enough
> for their client? I can't answer either of those questions.
>
> If you go through with this and take any of my suggestions, let it be
> this one: Get a lawyer involved in the contract and make sure you
> don't take on any liability for this.
>
> -Denny
>
> On Fri, Jul 25, 2025 at 10:50 PM Neil <[email protected]> wrote:
>> Hi all,
>>
>> I'm in discussion with an entity who wants a couple products reverse
>> engineered. It's actually their products (and have their brand), but
>> they've parted ways with their CEM/design house who now doesn't want to
>> hand over the design & assembly files. And they need to be the same as
>> they are currently, else it may re-trigger a re-evaluation/approval by
>> one of their major clients. What they need is gerbers, BOM, and assembly
>> files so they can continue to make the product.
>>
>> For one product, I have a schematic, list of component part numbers (not
>> calling it a BOM as there are no quantities), copper layout for all
>> layers, and a sample assembled product.
>> For the other product, all I have is the copper layout for all layers,
>> and a sample assembled product.
>>
>> I'm trying to figure out what methods are available to reverse-engineer
>> these products back to gerber and assembly files so they can continue to
>> have it made by another CEM.
>> For one product I can pretty much re-design the board from the
>> schematic, but not sure how to figure out the component parameters (not
>> just values like 0.1uf, but also tolerances, thermal characteristics,
>> etc). One is a very tight board with lots of mosfets so thermal design
>> will be important.
>> For the other product, I can come up with a non-valued schematic (so I
>> know where say capacitors and resistors would be), but won't easily know
>> what the values are. Can all of these be measured?
>>
>> The products may be able to be re-engineered, but I expect re-interating
>> prototypes to finalize the design would be somewhat different and would
>> take longer.
>>
>> Any suggestions here? Are they any design houses that specialize in this?
>>
>> Cheers,
>> -Neil.
>>
>> --
>> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
>> View/change your membership options at
>> https://mailman.mit.edu/mailman/listinfo/piclist

-- 
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
https://mailman.mit.edu/mailman/listinfo/piclist