are drawing normal schematics a dying art?

[SiliconWizard]

I am a bit disappointed that regular schematic design has not adopted color.

Please, though, consider the *very* common red/green color deficiency that affects a good percentage of the male population.  Some colors are extremely hard for me and others to tell apart, and it's not just red and green.  At least make sure there is redundant information on the schematic so I can figure it out without being able to differentiate all the colors.

Good point. Also, I've seen a lot of abuse of colors for schematics. They should not be used to replace proper layout of schematics using "reasonable" symbols. I personally find colored wires/busses awful. An acceptable use of colors is giving IC symbols a colored background - like the common light yellow - but that's about it.

Also, color printers may not be a lot more expensive - although in a company's context, most printers are laser and reasonably reliable printers for corporate use are still more expensive in color... but also the printing cost per sheet is a lot higher. If you have never worked in a company with accountants telling people to stop printing in color unless it's absolutely necessary, you're lucky.

[Bassman59]

looking at some circuit designs online it looks to me like a lot of ppl are using this Fritzing thing,i find them horrid,look like sommat out of a 2yr olds scribbleing book,whats ppls views on it?,give me a proper schematic any day!!.

Fritzing diagrams remind me a lot of your grammar, I can barely read it. I can't stand Fritzing diagrams myself but they are really only used by beginners, mostly maker types, any real engineer and most mid range hobbyists use proper schematics, they haven't gone anywhere.

Seriously, the poster complains about Fritzing but his post is written in Authentic SMS Gibberish.

Something about motes and beams in eyes here, I think.

[james_s]

Seriously, the poster complains about Fritzing but his post is written in Authentic SMS Gibberish.

Something about motes and beams in eyes here, I think.

I wouldn't have said anything except I'm fairly sure I remember him popping into a thread with nothing more than a complaint about some minor grammatical mistake, by someone who was not a native English speaker.

SMS gibberish made some sense back in the days of flip phones and 160 character limits, but with modern smartphones that almost everyone has you have got to be exceptionally lazy to not at least try to use something resembling proper grammar. With autocorrect, auto suggest and support for much longer messages there is no need for abbreviations that make it harder to read.

[eti]

looking at some circuit designs online it looks to me like a lot of ppl are using this Fritzing thing,i find them horrid,look like sommat out of a 2yr olds scribbleing book,whats ppls views on it?,give me a proper schematic any day!!.

I know it's common to type in social media acronyms - I do it myself, and it's a hangover from the times when sending SMS cost money (here in the UK, it no longer does in a per unit sense), but in this context, I feel it would behove you to type words and sentences in their full forms. After all, it's not 1999, and you're not trying to save 10 pence by shortening the contents of a two-SMS, 20 pence worth of SMS to fit them into one 

[thm_w]

For laptop schematics and such where components have hundreds or thousands of pins, I can see the utility. But using that style for something much simpler is adding complexity where there shouldn't be any. When both ends of a net are on the same page, try to connect them with an actual wire.

If I have a SWD/Clk debug signal from micro going off to a debug connector, I'm not going to bother running the wires off to it on the side of the page.
Kinda prefer organized blocks in the same area, eg connectors all along the left side of the page.

But, these are schematics for myself, I get how its harder to understand whats going on. And you can definitely get caught doing this if you forget to hook up a net and don't run a schematic check.

[VK3DRB]

Altium allows you to you harnesses in schematics. Harnesses are OK if used wisely and the engineer does not go over-board ('scuse the pun). I have seen harnesses grouping harnesses grouping harnesses! And with harnesses' signal names changing between sheets. The engineer who created it made such a mess, there were Altium harness errors he left in the design, then he left the company. I ended up fixing the errors, and in doing so I found two instances where where one  digital output was driving two GPIO digital inputs in a processor.

I do use harnesses myself but rarely. Altium made a mess of harnesses in my opinion. Changes to harness names is not trivial.

I think that engineer is like some sharp coders who make a point of writing pointers to arrays of pointers which point to pointers so that the appointed engineer points out he doesn't see the point.

Yeah i am not a fan of how harnesses work in Altium. They are a great idea, but poorly executed....

...That being said i still found this style of schematic used in confusing ways where parts of a circuit are cut up in a nonsense way and spread across two or more modules, so you have to constantly flip pages to figure out what is going on.

Exactly. Great idea but poorly executed. Problem is when Altium creates something down the wrong path, they cannot easily fix it because it breaks earlier schematics. Backwards compatibility can sometimes be a curse.

I had an ex-colleague who worked for a company in Canada where they have one page per IC and even one page per transistor. He told me they had perfected stupidity to a fine art. As you say, a good schematic minimises page flipping and maximises comprehension. Page flipping slows down the productivity of debug technicians in Production. Years ago some TV schematics had oscilloscope waveforms in various circuits to aid fault diagnosis... they were brilliant because they considered the plight of the user (TV tech). In fact, TV sets often came with the schematics tucked away inside the set. 

Net labels with sensible labelling should proliferate in my opinion because they aid track layout work so we know exactly what each track is for to minimise crosstalk, consider currents etc, without having to refer to back the schematic all the time. Some net labels can be very small if they are benign to the end user, eg: "TP23".

[james_s]

One page per transistor? Unless the circuit contains a very unusual ratio of transistors to passives that sounds extremely tedious.

[Berni]

One page per transistor? Unless the circuit contains a very unusual ratio of transistors to passives that sounds extremely tedious.

In guessing the people there got so used to working with ICs and not much with discrete analog anymore. So perhaps a transistor is simply there for making a open collector output or a switch. So its 1 transistor doing 1 job.

I do sometimes do a single chip on a page. This is mostly on things like MCUs and FPGAs where that page caries the massive chip symbol/s and out of that schematic i have ports with sensible signal names. That way the top level schematic just show a "VIN_SENSE" going into the MCU block, but if you want to know exactly what pin that goes to you can just open it and see what pin it hooks into. This information usually quickly gets turned into a .h C header file, so its no longer needed afterwards.

The one annoying thing to clearly show are UART lines. There is no strictly defined master/slave since both nodes are symmetrical while you do have to cross over the RX and TX lines for it to work. So naming the nets becomes tricky. This is where the Altium direction arrows on the schematic sheet ports help, but sometimes i have UART going into a connector. I started drawing extra direction arrows there too in order to show what way the data is flowing.

[RJSV]

A slightly off-topic, is the old phenomena: The Drafting Dept. And the folks doing that work, 40 hpw.
   My approach, in a small business, used the schematic and other supposedly in-house packaging documents, in marketing. So, a sub-system package might include cables (schematic), plus lots of views of enclosures. That makes for a market-ready document and as a bonus, helps keep marketing dept. realistic in promises made (i.e. to the customer).
   The old-school: DRAFTSMAN that's what I got started thinking, and that (was) quite a role, back in da day.

[Zero999]

One page per transistor? Unless the circuit contains a very unusual ratio of transistors to passives that sounds extremely tedious.

In guessing the people there got so used to working with ICs and not much with discrete analog anymore. So perhaps a transistor is simply there for making a open collector output or a switch. So its 1 transistor doing 1 job.

When transistors were expensive, some designs did have a lot of passive components per transistor.

A single transistor LED flasher I designed just for fun. Yes I know it's possible to use a transistor in reverse bias mode, but it doesn't work at low voltages.

  phaseshift LED flash.asc

[Terry Bites]

See my post:  Can anyone still use a pencil?
https://www.eevblog.com/forum/projects/can-anyone-still-use-a-pencil/msg3810782/#msg3810782
It must be a shotage of envelope backs an fag packets.

[VK3DRB]

...The one annoying thing to clearly show are UART lines. There is no strictly defined master/slave since both nodes are symmetrical while you do have to cross over the RX and TX lines for it to work. So naming the nets becomes tricky. This is where the Altium direction arrows on the schematic sheet ports help, but sometimes i have UART going into a connector. I started drawing extra direction arrows there too in order to show what way the data is flowing.

The TX and RX pin on a chip's schematic symbol should always be in reference to that chip. The TX and RX lines on a board connector should be in reference to the board. Some people name the net labels with the reference in the name eg: "MCU-TX" meaning it is a TX from the MCU chip.

In my opinion, a very good idea is to use direction arrows in the device symbols. With Altium, for example, a microcontroller's symbol will have the TX line shown as an output and the RX line the receiving chip as an input. This makes it so much easier to read - no confusion and no ambiguity! The library symbol stays the same, but the instance on the schematic is modified showing the respective directions. The only risk here is if you carelessly do a symbol update from the library, it can be overwritten.

[penfold]

The RX and TX ‘issue’ is one reason I quite like keeping the number of ‘functions’ per sheet lower, because naturally in the hierarchy it gives an additional planning stage to define signal directions, altium makes it clear enough if there’s any violations that aren’t visible. So along with sensible naming, pin directions and annotations (fewer components… more room to annotate!) it makes everything easy to review, rather than “oh that’s not right, what was it meant to be” it’s “is what I intended to do correct and does it match what I drew”, doesn’t necessarily eliminate page turning, but it gives page turning a very defined objective to prove or disprove a single statement

[Berni]

The TX and RX pin on a chip's schematic symbol should always be in reference to that chip. The TX and RX lines on a board connector should be in reference to the board. Some people name the net labels with the reference in the name eg: "MCU-TX" meaning it is a TX from the MCU chip.

Yes in a lot of cases it is possible to have a sensible "Master" on that UART bus. In those causes i prefer to name the lines as being TX from the MCUs point of view. For example a MCU talking to a GPS receiver or Bluetooth module or something.

But sometimes things become tricky. I had cases where two MCUs talk to each other on the same board (It was some small MCU doing time critical stuff) or you have two MCUs on separate boards talking to each other over a connector or cable. For clarity you want to maintain naming of the connector pins on both sides so that its easy to follow the signal from schematic to schematic. Yet both are a MCU talking UART to some pins on a connector.

So the directional schematic ports in Altium are really nice for making this more clear. Where i have a UART bus just going into a connector (or something else without direction) i manually add some graphical arrows to it, so its clear what direction its flowing just by looking at the connector.

But this is just 1 small part of it. There are plenty worse ways to make schematics into a confusing mess.

[AlienRelics]

Oh, thanks, I did not know it is called Fritzing

No, it is called a pictorial diagram. This particular program is called Fritzing.

It looks like a great way to explain to non-technical people where to put what wire 
It is a totally different thing that schematics

Very true, and sad. I blame Make: Magazine for destroying the art of drawing clear, cogent schematics. They have these bastardized half-schematic, half-pictorial diagrams with an IC being shown as a block with 14 pins and no labels. It requires you to break out the datasheets and redraw the circuit to figure out what is going on.

I guess the question is, do you just want to assemble other people's projects? Or learn how they work so you can change them and even design your own?

[fourfathom]

It looks like a great way to explain to non-technical people where to put what wire 
It is a totally different thing that schematics

Very true, and sad. I blame Make: Magazine for destroying the art of drawing clear, cogent schematics. They have these bastardized half-schematic, half-pictorial diagrams with an IC being shown as a block with 14 pins and no labels. It requires you to break out the datasheets and redraw the circuit to figure out what is going on.

I guess the question is, do you just want to assemble other people's projects? Or learn how they work so you can change them and even design your own?

Make: Magazine isn't intended for engineers, techs, or other people who are already skilled in the art.  It's for those who want to get their feet wet, or are looking for ideas (although it's true that I occasionally pick up an issue to see what's going on).  When I was starting out (at the age of nine or ten) I had no clue how those tube things worked, and I couldn't have designed a radio to save my life.  But I did have simple schematics (which might as well have been Fritzing diagrams) which showed me how to re-wire war-surplus airplane radios so I could use them at home.  This was how I got started, and I see Make as providing a similar pathway for people who are interested in electronics.  Perhaps they will eventually graduate to schematics, but anything that helps them get over the initial hurdles is OK with me.

[AlienRelics]

Make: Magazine isn't intended for engineers, techs, or other people who are already skilled in the art.

Which is precisely why their schematics should be as clear and easy to understand as possible.

[Berni]

Make: Magazine isn't intended for engineers, techs, or other people who are already skilled in the art.

Which is precisely why their schematics should be as clear and easy to understand as possible.

Yes but what you find more readable might not be the more readable to them.

When they have a transistor in there hand they have a black thing with 3 pins on it. They don't even know the correct order of pin numbering for the package, let alone what each pin on a transistor schematic symbol is called. They need to be able to find the datasheet to find the pinout etc..

On the other hand if you show someone a drawing of a TO-92 and draw lines from each pin, then pretty much any sensible person will be able to wire it up from that drawing alone, even if they have no electronics experience at all.

Even just planning your breadboard layount can take a bit of skill to lay things out nicely compact without boxing yourself in while at the same time keeping jumpers short and neat.

Yes it's infuriating for me to read such diagrams. But for someone who is just trying to replicate the build they are indeed easier. Perhaps after they get a few projects working they will get enough interest in electronics to learn how to read proper schematics and to learn how things he put together actually work.

[rsjsouza]

In my early days of electronics, the pictorial diagrams were everywhere in the magazines I used - I personally don't see any bastardization of anything, as one is not really replacing the other.

Despite the magazines still showed the schematics and principle of operation, the visual correlation between the actual part and the pictorial diagram was an incredible tool to get a successful build. And that is what it is: a tool to help people build their projects.

[SL4P]

In my limited experience, the absolute best technical documentation I was exposed to - both verbal and diagrammed - was from AMPEX in the 1970s.

Everything from theory of operation, detailed block diagrams - fabulously accurate and readable schematics, and no colour or photos unless it was in a supporting role.

Mind you, the technical docs for a video tape recorder was contained in five 4-inch thick ring binders.
They were so good, I’d take one home, and read them overnight through my traineeship!

[AlienRelics]

As I said already, do you want to just replicate projects someone else designed? Then pictorial away.

Do you want to actually learn how things work so you can start modifying designs, and designing your own? A clear schematic is a wonderful tool.

I did a class in reading and drawing schematics for some local makerspaces. All of my examples of how not to do it came from Make: Magazine.

Older datasheets are great! They designed new ICs and they are trying to get people to use them. So they put a lot of effort into including reference designs and examples, with special emphasis on clarity.

I'm not telling anyone else what to do, just expressing my opinions.

[fourfathom]

As I said already, do you want to just replicate projects someone else designed? Then pictorial away.

Do you want to actually learn how things work so you can start modifying designs, and designing your own? A clear schematic is a wonderful tool.

I did a class in reading and drawing schematics for some local makerspaces. All of my examples of how not to do it came from Make: Magazine.

Older datasheets are great! They designed new ICs and they are trying to get people to use them. So they put a lot of effort into including reference designs and examples, with special emphasis on clarity.

I'm not telling anyone else what to do, just expressing my opinions.

And my opinion is that some/many/most/me for sure, get their initial exposure to electronics by replicating projects others had designed.  I found it fascinating, and began modifying, and understanding at a primitive level, those designs.  Eventually it turned into my passion, which led to a great career.  Now that I'm retired, it's still a passion, and my main hobby.

I don't know where I would have ended up had I not started out copying other peoples work.  Probably as an old guy playing bass in a bar-band for tips (one of my other hobbies, except for the tips).

[eti]

If you want to build a circuit you SHOULD take the time and effort to sit down and work out how it functions and what the components are doing. If you're blindly following a guide, slotting modules together or doing the electrical equivalent of "dot to dot" kids books, then don't you DARE go and show off your copied device to someone, boasting and accepting praise. People ARE VERY lazy nowadays - they want the rewards and adulation without the hard work (look at society too - Instagram nonsense etc - everyone's constantly out having a huge "party", and yet they've got nothing of any substance to be celebrating, if ANYTHING - it's an attention seeking generation)

[rsjsouza]

That is very myopic. People learn differently and get their motivation in many different ways. The process is a continuum of successes and failures and the learning comes along the way, not a pre-requisite.

For a kid like me starting to get the taste for electronics at the early age of 7 or 8, a successful build was incredibly rewarding and a feat on itself, which led to more builds and deeper understanding. Without the "blind copy" there was always a chance my interests would go elsewhere - at the time I was also interested in mechanics, but the lack of hobby magazines and the difficulty in assembling things without standard and mass produced parts available on a store were a huge stumbling block.

I am not the only one. Many of my youger co-workers started copying and assembling circuits and code using Arduino and simpler stuff and are starting to make a career out of it.

[vk6zgo]

At least people draw proper point to point connections with it. Not like the army of new generation dummies who drops parts on the schematic sheet and terminates every single pin with ports.

This.

Also, modern board designs are 5 ICs, 100 decoupling capacitors, 500 nets, and 3 resistors.  So much of circuit design has moved into IC design that I know people who just say "I'm a circuit designer" to mean "analog IC designer."  It's pretty understandable that few board level schematics look like the beautiful drawings in the end of an old HP service manual. 

Finally, I have to say looking at some fritzing schematics to see what it was all about that I am a bit disappointed that regular schematic design has not adopted color.  Many schematics will only ever be viewed on a monitor and color printers are hardly more expensive than black and white.  Color coded nets definitely could certainly be abused but it could also be a powerful tool for communication of intent if done properly.

Back in the 1960s, Schematics with different colours were used by some manufacturers, but personally, I found them hard to read.
They disappeared after a while, so I mustn't have been the only one of that opinion.