Electronics-Related.com
Forums

Can we PLEASE stop using these shitty symbols?

Started by Tim Williams May 22, 2017
> I like the insulated-gate-bipolar symbol because it's instantly, >intuitively obvious which way the current will flow. I'm not so much >interested in the semiconductor physics as feeling how the circuit >works.
The current flows as it does because of the device physics and bias conditions, not because of a magic symbol. I and the other right-thinking Little-Endians prefer the symbols that better express the way the devices actually work, over a range of conditions wider than the Big-Endian symbol can. Cheers Phil Hobbs
On 28/05/17 10:22, whit3rd wrote:
> On Monday, May 22, 2017 at 9:07:57 PM UTC-7, Clifford Heath wrote: >> On 23/05/17 12:23, Tim Williams wrote: > >>> I mean, there is absolutely, positively no work lost, using electronic >>> formats, with symbols as complicated, as baroque, as you can dream of. > >> A drawing short-cut is also a visual one. >> I personally quite like the simplified symbols. >> They're easy on the eye. They don't reflect reality, >> but neither does any other symbol, in any alphabet. > > That's SO untrue. There are entire written languages based on > pictographs
We have different definitions of "reflect reality". Obviously there are similarities, and schematic attempts at representations, but they're still just symbols, mnemonics.
On a sunny day (27 May 2017 19:24:49 -0700) it happened Winfield Hill
<hill@rowland.harvard.edu> wrote in <ogdcdh02btc@drn.newsguy.com>:

>pcdhobbs@gmail.com wrote... >> >> The BJT-ish symbols are faster to draw, so Big-Endians >> prefer them. Their deviation from the correct view >> naturally impairs their hand-eye coordination as well >> as their aesthetic judgement. > > Bah, humbug. I don't believe in a single symbol, and > use the full symbol whenever it's useful. My preference > has only a little to do with being faster to draw, but > lots to do with reducing drawing clutter, so it can be > more quickly comprehended. A clear simple drawing, > without leaving out anything critical, is very useful.
There are many aspects and most paint only part of the elephant. There is more to those symbols then just the design process there is also the service process, having been repairing all sorts of high - and low end stuff for many years, you want the technician who has to repair it to be able to understand the circuit too. It is not just about inside chips, he could not care less. So the requirements are different. These days they do not supply circuit diagram anymore, not sure about service manuals, I have seen people just swap out units. What a waste (financially too), repairing at component level is well, environmentally friendly. Sometimes I hear about those pencil drawing I make, 'cannot read it' apart from people trying to read it with a Nipkow disk or similar, those are all things I draw so I can build it, and most important 10 years later repair it (hundreds of wires often, if one does break on both sides you are in trouble). So I use it to build it, and I use it to repair it, trouble shoot if there is any. For the IC designer 'to repair it' has no meaning and he can make his own symbols. Jim Thompson could have used 'D' for DMOS and not 'O' (as in inverter), but he is free as nobody else in the outside world will use his drawings. I should be using some graphics program to draw stuff, but that requires a peesee, and I have no peesee on when I am soldering, and last time I used a printer was to print a map of some place hidden in the boonies where I had to pick something up. Whiteboards are too big, paper is patient, part of the last circuit diagram got dissolved as it was under a wet coffee cup. Luckily there was enough space left: http://panteltje.com/pub/took_a_byte_IMG_6210.JPG
pcdhobbs@gmail.com wrote...
> > I and the other right-thinking Little-Endians prefer the symbols > that better express the way the devices actually work, over a > range of conditions wider than the Big-Endian symbol can.
I don't think you use many MOSFETs in your designs, certainly not power MOSFETs. In contrast, other designs increasingly using many power MOSFETs, so clutter and clarity is an issue. These days it's hard to get a discrete MOSFET with an available substrate pin, and none of the power MOSFETs have one. This limits the value of a symbol highlighting the substrate path. If the substrate diode comes into play, it's more useful to actually show the diode as part of the symbol. I've seen many cases where the full-fledged classic MOSFET symbol has been supplemented with the diode. This may be redundant,* but it comes about because the designer knows that many aren't always aware of the diode aspect of the substrate part of the symbol, and he wants to be sure diode action is not overlooked. Ahem, would right-thinking Little-Endians eschew showing the diode? * One other aspect, this diode is drawn in place even though there's a chance of confusion, is this another required part? -- Thanks, - Win
Jan Panteltje wrote...
> > last time I used a printer was to print a map ... > Whiteboards are too big, paper is patient, part of the > last circuit diagram got dissolved as it was under a > wet coffee cup. Luckily there was enough space left: > http://panteltje.com/pub/took_a_byte_IMG_6210.JPG
Snork! I assume you have one of the cheap all-in-one ink-jet printers. They're nice to save scans of your valuable pencil sketches, and print copies that you can mistreat with solder burns, or whatever. Copies are useful for recording measurement notes as well. I got an HP 7525 for $125 at Costco, good features, full duplex on printing and scanning, etc. Just be careful with the ink usage, they make $$ back there. -- Thanks, - Win
"whit3rd" <whit3rd@gmail.com> wrote in message 
news:d0a99d43-2f87-41b0-8ce2-fcb5b25034a2@googlegroups.com...
> What I'd really like to see, is a good symbol for a Hall effect sensor. > The one for thermocouple is really a nuisance, it looks too much like > just a joint in a wire. Which it IS, but it's a SPECIAL joint, and the > wire > types matter, and they ARE NOT IDENTIFIED in that silly symbol. So, how > can I get the polarity right?
Thermocouple could have the wires in different colors. Or for B&W, one outlined and the other solid. Which one is which would inevitably have to be labeled, since it comes down to metal type. Convention could be one is positive (when the temperature difference is positive), much like the Brit electrolytic symbol (one electrode hollow to show polarity). Which, come to think of it, is a pretty good way to show polarity of an aluminum electrolytic, because the anodized surface is hollow -- porous. :) Hall effect is usually drawn as an isometric rectangle, with connections to the sides. It is what it is, and you can indeed measure the Hall effect in any conductive material connected this way; but the special part, that the material has low charge carrier density (a semiconductor), is the kicker that makes it usefully sensitive. How you want to indicate that, is up to you. *shrug* :) Tim -- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: http://seventransistorlabs.com
"John Larkin" <jjlarkin@highlandtechnology.com> wrote in message 
news:3l5jicdcuuhpfjui8r4pnp8mdc6mbt3n7f@4ax.com...
> You are missing the big picture: It doesn't matter. The electrons > don't care what you drew on the schematic or what words you used to > describe anything.
So the electrons magically know which way 'round the source and drain are? Shall I build your circuit and the OP circuit and connect both to an unlimited current supply? Which one is right, John? :-) Tim -- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: http://seventransistorlabs.com
"Jim Thompson" <To-Email-Use-The-Envelope-Icon@On-My-Web-Site.com> wrote in 
message news:74jjiclbb94ugamkikov1shahjaeb71out@4ax.com...
> Out of curiosity (>:-} I examined all the schematic entry software I > have on hand and compared symbols (Cadence Virtuoso from an Image, I > don't personally own a copy)... > > <http://www.analog-innovations.com/SED/MOS_Symbol_Comparison.pdf> > > Note that Cadence Virtuoso _does_ use an arrow to mark sources, but > then doesn't indicate body diode polarity :-(
Oh! My! God! Surely you can't get any more wrong than that! Just to add to the mix: Altium and Multisim use the "right" kind -- with substrate and source shown, and tied. Some also add the redundant antiparallel diode, but oh well. Tim -- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: http://seventransistorlabs.com
<pcdhobbs@gmail.com> wrote in message 
news:f3c21660-1ce9-4c0d-8509-010397e4e5c3@googlegroups.com...
>> I like the insulated-gate-bipolar symbol because it's instantly, >>intuitively obvious which way the current will flow. I'm not so much >>interested in the semiconductor physics as feeling how the circuit >>works. > > The current flows as it does because of the device physics and bias > conditions, not because of a magic symbol. >
The confusion is whether the reader will notice that the arrow terminal is indicating P or N channel. So which way does the current flow, as it does because of the device physics and bias conditions, _based on what the user purchased_ because of a magic symbol? :-) Tim -- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: http://seventransistorlabs.com
"Jim Thompson" <To-Email-Use-The-Envelope-Icon@On-My-Web-Site.com> wrote in 
message news:j5gkic1oek6bsk784tsgbvi5gsti8q5s57@4ax.com...
>>The convention that p-type devices have their sources/emitters up, and >>n-types down, helps a lot. But sometimes transistors work better >>sideways. > > Only snag... often, in analog applications, current flows both ways.
Like my kneejerk reaction to that LDO with the "lefty" symbol and the antiparallel diode... it's a diode both ways?!? No, wait.. oh... @*%$ Tim -- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: http://seventransistorlabs.com