Is there anything like Thingiverse for electronic circuits?

[CharlieWorton]

Thingiverse is a site which features many thousands of 3D models which other users can download and print for free on their 3D printers.  For example: some guy in Wyoming has a plastic part break in his 2007 Sedona, so he designs and prints a plastic replacement part on his 3D printer.  He also posts the file on Thingiverse with a descriptive title and keywords (Sedona, door guide, 2007) and a year later, another guy in Calcutta - who also owns a 2007 Sedona with a busted door guide - finds it on Thingiverse, does the happy dance, downloads it and prints it.  He fixes his door.  The world becomes better.

Is there anything like that for electronic circuits?

Ya see, I'm scooting right along in my fabrication facility, but my electronics knowledge hasn't yet progressed to a level where I can actually design anything.  So, I'm like all dressed up and no where to go.

Any cool place where I can download useful schematics?

Thanks - Charlie

[jimdeane]

Tagged to find out, because that's a damn good idea.

[Dave]

Well, there's Instructables, but you would be hard pressed to find a semi-decent circuit on there. The vast majority are pure shit and the problem with beginners is that they aren't able to recognize a good circuit, so the site is as good as useless.

[hendorog]

Really great idea. Electronics needs some disruption and that would certainly do it.

Would need to be independent of any electronics tool vendor though.

[jpanhalt]

For that approach, one would need to index the circuit types.  Problem is, there is no consistent index system that I have seen for electronic circuits, particularly circuit fragments.

To explain, in chemistry, there are rules for naming and for more than 100 years, there have been name indexes.  However, one may not know the name or may be looking for something similar that would have a different name.   For that, chemistry has indexes based on empirical formula, molecular weight (usually high resolution MW), and even the number and type of bonds (usually to carbon).   The latter might be attractive for electronics.

With our current ability to index even written statements and laws, one might think electrical engineers would see the need. 

John

[Syntax_Error]

For that approach, one would need to index the circuit types.  Problem is, there is no consistent index system that I have seen for electronic circuits, particularly circuit fragments.

This is close.

[jpanhalt]

Nope, not it.   You have to know what the circuit does:



I am talking about an index that is based on the structure, not purpose.  Purpose is like looking up the name of a chemical.

John

PS:   This site has the most archaic image handling I have seen in the 21st century.   How do you insert inline and image easily?

[jpanhalt]

To further explain: 

Suppose you draw a circuit or circuit fragment in Eagle. You get a net list.  Can you search on that list (minus coordinates)?

John

Edit:  On a lark, I searched Google for "C double bonded to N".   Try it.   You will get a name.  Can you do that for a resistor and capacitor in series or in parallel?

[FrankBuss]

Edit:  On a lark, I searched Google for "C double bonded to N".   Try it.   You will get a name.  Can you do that for a resistor and capacitor in series or in parallel?

Well, if you want to solder your own circuits, you should have some base knowledge, otherwise you would hire an engineer to do it for you. For the rest you could try this site:

https://electronics.stackexchange.com/

The rating helps a lot compared to sites like Instructables. Another thing I do is Google image search with the electronic keywords I'm searching for and usually you get good schematic results.

[jpanhalt]

Well, if you want to solder your own circuits, you should have some base knowledge, otherwise you would hire an engineer to do it for you.

Fortunately the forefathers of chemistry didn't have that attitude.  Physics, biology/medicine, and law caught up.   Maybe it is time for engineers to join our modern society and not be so parochial.

John

[technix]

I personally have been publishing open source designs on Github. Maybe someone should get on with it and build this Thingiverse for Electronics website?

[FrankBuss]

Well, if you want to solder your own circuits, you should have some base knowledge, otherwise you would hire an engineer to do it for you.

Fortunately the forefathers of chemistry didn't have that attitude.  Physics, biology/medicine, and law caught up.   Maybe it is time for engineers to join our modern society and not be so parochial.

I don't think this would make sense. Sure, you could invent a naming scheme to convert a netlist to a unique name. But this wouldn't make sense, because there are so many more possible components than in chemistry and by reading the name, it would be difficult to imagine the circuit. For example "op-amp input plus signal is input, op-amp input minus connected to op-amp output, op-amp supply plus connected to positive supply, op-amp supply minus connected to GND, op-amp output is output" would be difficult to understand and it doesn't say what it does. Instead there are lists of op-amp standard circuits and this kind of circuit is known as "buffer".

Actually, while writing this response, maybe it makes sense It would need some software support, for example a netlist format with coordinates for easier visualization, because auto-layout programs couldn't show it in a way that engineers are used to see the circuit diagrams, but then you could search for netlist fragments or whole netlists in a database. And each stored netlist would have a unique name for what it is used for, or what the common name is, e.g. "op-amp buffer". Same like in chemistry, H2O=water.

[brozier]

This question has been asked on Stackexchange funnily enough.

https://electronics.stackexchange.com/questions/5424/is-there-a-directory-of-open-source-hardware-projects

I knew about some PCB fabs sharing pcb designs which is pretty cool...

[sleemanj]

Fritzing sort of has something like that http://fritzing.org/projects/

There's no reason you can't use Thingiverse, it's not just for 3d print things although that's the bulk of it's projects.  A lot of the projects there have electronic gubbins.

Of course there is hackaday.io I personally find it a bit.. the browsing navigation is a bit naff because no categories, but you can at least look at the most recently updated to get started https://hackaday.io/projects?sort=updated

[alexanderbrevig]

I like this idea too. For me I'd like it to be searchable for what you want. And it would be great if it was just blocks, things like 'lipo charger', 'nrf52' and 'headphone amp'.

Furthermore it should have some kind of approval system, where peers review it and marks it as working,untested,outdated,reference etc.

I can make it, would we use it? Anyone want to help?

[jpanhalt]

I don't think this would make sense. Sure, you could invent a naming scheme to convert a netlist to a unique name. But this wouldn't make sense, because there are so many more possible components than in chemistry and by reading the name, it would be difficult to imagine the circuit.

First, I mentioned the netlist as an example, not as the answer.  It is the only example I am familiar with in electronics in which a circuit is described without means of a function and without drawing a picture.  At risk of trying to explain by example, consider norbornane:



Its name is bicyclo[2.2.1]heptane.  The "bicyclo" denotes two rings, and the numbers are the number of non-common (non-bridgehead)atoms in each ring.  In this particular case, it is unambiguous, but in tricyclooctane isomers and many other compounds, you need to define the bridgehead atoms.  Moreover, using the simple rules for absolute configuration (i.e., R and S nomenclature), one can describe stereo isomers in 3D space without a picture.  I see a clear analogy between the chemistry system of naming and a type of netlist.

Second, as for the number of "components,"  I would like to see your data on that.   How many different components are there?  Remember, a single compound "cetane" (hexadecane) has over 10,000 isomers, and every one of them can be uniquely named following current rules (https://pubchem.ncbi.nlm.nih.gov/compound/hexadecane).

The chemistry naming system is not complex and can easily be learned by high school students without nuch backgrond in chemistry.   They do need to know or be able to find the atomic numbers for elements.

For example "op-amp input plus signal is input, op-amp input minus connected to op-amp output, op-amp supply plus connected to positive supply, op-amp supply minus connected to GND, op-amp output is output" would be difficult to understand and it doesn't say what it does. Instead there are lists of op-amp standard circuits and this kind of circuit is known as "buffer".

That is my whole point: I am saying there needs to be a searchable index of circuits that is independent of function.  In essence, the index will let you find out what a circuit does and possibly other ways to do the same thing.  Of course other indexes of function are also useful.  Extending the chemistry analogy, you can search on phenacylbromide  (old name: alpha-bromoacetophenone).  Every chemist knows what it does, but you don't need to know that to look it up and find out.  You can also search on what it does (lachrymators).

Actually, while writing this response, maybe it makes sense It would need some software support, for example a netlist format with coordinates for easier visualization, because auto-layout programs couldn't show it in a way that engineers are used to see the circuit diagrams, but then you could search for netlist fragments or whole netlists in a database. And each stored netlist would have a unique name for what it is used for, or what the common name is, e.g. "op-amp buffer". Same like in chemistry, H2O=water.

Glad you agree it may be feasible and will require a lot of work and revision.  Chemical naming has been revised at least twice in my life.   Fortunately, the revisions are not that hard to remember when you are forced to do that.

John

[MosherIV]

Sorry to go back but I would like to answer some of the OPs points :

Is there anything like that for electronic circuits?

No and Yes.
There are sites dedicated to electronic circuits. Some are good and some are bad.
Just search the internet for circuits eg "DIY PSU schematic" will come up with many sites with schematics, some sites are blog where they walk you through the design, build and results.
Most circuits are presented and it is up to the user to make and understand the circuits.

It is NOT like 3D printing where you can just download a file and just print the thing. Some level of understanding IS required.

if you want to solder your own circuits, you should have some base knowledge, otherwise you would hire an engineer to do it for you.

Precisely the point.

I would however add that just searching and looking at the circuits and try to understand them helps a lot.
I have in the past and still do start any investigation by looking at what is out there and understand what schematics are out there and how the circuits work for myself.

If you just want to download and build the circuit - you will be better off looking at DIY kits that you can buy.

[tronde]

http://www.epanorama.net/ is still alive. Maybe something for you there.

[kasbah]

Sorry for replying this old topic but I have been building Kitspace for quite a while and "Thingiverse for electronics" really describes it best. Kit Space presents a project in a way that makes it easier to understand what the project is and allows you to order the parts you need with a few clicks. I spent a lot of time making bills of materials transferable so you can order someone else's project's parts at your preferred retailer.

The site itself is open source and on GitHub. Right now you still need to use Git and GitHub/GitLab or similar to add a project but I am working on a direct upload feature.

[rrinker]

 There certainly are for specific applications - like this one for model railroads. Though with the wide variety of circuits, you can see many of them have much more general use.

http://www.circuitous.ca/CircuitIndex.html

[Nusa]

To further explain: 

Suppose you draw a circuit or circuit fragment in Eagle. You get a net list.  Can you search on that list (minus coordinates)?

John

Edit:  On a lark, I searched Google for "C double bonded to N".   Try it.   You will get a name.  Can you do that for a resistor and capacitor in series or in parallel?

Did you even try? A search for any combination of series, parallel, resistor, and capacitor gives meaningful results for that combination.

[Someone]

I don't think this would make sense. Sure, you could invent a naming scheme to convert a netlist to a unique name. But this wouldn't make sense, because there are so many more possible components than in chemistry and by reading the name, it would be difficult to imagine the circuit.

First, I mentioned the netlist as an example, not as the answer.  It is the only example I am familiar with in electronics in which a circuit is described without means of a function and without drawing a picture.  At risk of trying to explain by example, consider norbornane:



Its name is bicyclo[2.2.1]heptane.  The "bicyclo" denotes two rings, and the numbers are the number of non-common (non-bridgehead)atoms in each ring.  In this particular case, it is unambiguous, but in tricyclooctane isomers and many other compounds, you need to define the bridgehead atoms.  Moreover, using the simple rules for absolute configuration (i.e., R and S nomenclature), one can describe stereo isomers in 3D space without a picture.  I see a clear analogy between the chemistry system of naming and a type of netlist.

Second, as for the number of "components,"  I would like to see your data on that.   How many different components are there?  Remember, a single compound "cetane" (hexadecane) has over 10,000 isomers, and every one of them can be uniquely named following current rules (https://pubchem.ncbi.nlm.nih.gov/compound/hexadecane).

Its easy to come up with the floor for just passive components, E24 over 3 component types of R,C,L with those spanning 12 decades each is already 1000 basic parts. The problem then is the orthogonal specifications of maximum voltage, current, power etc for each of those types and the search space explodes before you get into the complexities of non linear and active parts. Electronic circuits are more akin to simple organisms than molecules.

In essence, the index will let you find out what a circuit does and possibly other ways to do the same thing.

Just designing a low pass filter has so many implementations, each of which requires you to understand at minimum its interactions with the stages before and after it that you can't walk into this work and just make do with idealised functions.

[jpanhalt]

I question whether there are more possibilities in electronics than in chemistry.  Simple octane has 24 isomers, and that does not include ring structures and isomers as we see in spiral structures, such as DNA.   Add in the number of elements in just the second and third row of the periodic chart, and the number is immense.  Then, there are even more elements to consider.  DigiKey lists 7,072,345 distinct parts.   It is proud of that number.   There are billions of humans and virtually each of them is distinct chemically, and we are not the most numerous species by a long shot.

One of the things I like about Horowitz and Hill is their classification of 2-terminal and 3-terminal devices.  I guess in electronics there are no "one-terminal" devices as there is in chemistry (e.g., methane).  It only takes someone with that genius to organize and index electronics.   

Medicine and law had similar arguments, but now each of those disciplines is well indexed.   Electronics just needs the attention it deserves.