How to Draw Schematic Symbols Simply and Easily(1)
There is a lot of discussion about drawing schematic symbols. It's important to make your schematic symbols understandable. Sometimes pre-made symbols in a computer-aided design (CAD) package are fine, but most symbols are not ideal. Make sure your package can easily create symbols because you may have to redraw each individual component and create a new one. The tens of thousands of symbols included in CAD software are just the basis for you to redraw them.
A good schematic should have a predictable signal flow. This flow direction requires the input section to be on the left and top, and the output section to be on the right and bottom. Of course, this is not a piece of iron, but if you want other engineers to understand your schematic at a glance, it is very important to follow this rule. If I yell at you loudly, "What is the difference between doing this?" This grammatical structure is obviously confusing, but if I say it from right to left, "What is the difference?" Can understand. While many semiconductor companies make a lot of money and provide a lot of support, many times they focus on the inside of the chip and can't do the right schematic flow (Figure 1).
Figure 1: The schematic symbols currently drawn by many companies mimic the pin diagram of the component, not the signal flow.
The six inverter U1 in Figure 1 is not very practical. It combines six inverters into one symbol, and has input and output on the left and right sides. The pin length does not need to be that long. The U2 symbol is slightly better, the input is on the left and the output is on the right. A person like me doesn't like a colorful background, because after six black and white copies, the yellow color turns black, so you can't see anything. The U3 I created consists of different components (heterogeneous components), including six identical components and a seventh component representing power and ground. Exclusion RP1 is a very stupid drawing, and it is easy to make a mess of the schematic when these resistors should be in different positions on the schematic. RP2 shows the role of heterogeneous components at this time.
Some semiconductor companies use ANSI symbolic logic devices, which are apparently invented by people who lack linear thinking in analysis, rather than graphical thinking in the eyes of analog engineers (Figure 2).
Figure 2: Many engineers don't like ANSI/IEEE logical notation, which is simply unhelpful and harmful.
Showing the actual logical symbols is slightly better. The components included in the CAD software package are basically useless. A good practice is to split the component into two. A better approach is to separate the power supply so that it does not clutter the signal flow. What the simulation engineer wants most is to draw a little pattern inside the component that shows its function.
For multi-element packages (such as many logic gates), the schematic symbols need to be decomposed because you rarely use all of these components in the same place in the schematic. This principle also applies to two-way or four-way op amps. The symbol of the component can be de Morgan equivalent symbol (Figure 3). I really admire engineers who can understand the circuit work through Boolean expressions, but I still like the graphical representation - the bits in the D latch can be imagined through the graphics, or the multiplexer is given Input pin.
Figure 3: As early as 1995, OrCAD 9 allowed the use of De Morgan equivalent symbols to represent NAND gates.
Altium/CircuitStudio allows users to assign different "modes" to components to accomplish the same task. It is very convenient if you want to draw an op amp symbol with a "pin negative" mode. If there is no equivalent symbol, if you want to flip a component vertically, you will also put the positive power supply on the bottom and put the ground on the top. By calling the de Morgan equivalent symbol, you can swap the input pins while keeping the power and ground positions unchanged. Another way to solve this problem is to make a heterogeneous component (U6) with an independent power supply. Now you can flip the op amp vertically and place the negative pin on it.
The schematic program of a certain era appeared in such a period: there are about 40 14-pin logic chips on the PCB, each chip is equipped with a decoupling capacitor, plus a card edge connector. In 1985, DOS OrCAD could not even draw triangles. This was the limitation of that era and the thing that needed to be worried in that era. At the time, many companies felt that there was only one power supply on the PCB, VCC (two "C" stands for "common collector" because all of these logic gates feed power to the collectors of many transistors). Therefore PCB only needs VCC and ground. The CAD company's programmers even thought that there was no need to display the power pins on the chip. They just invented the "zero length" pin, and then the layout designer would connect all the pins of the same name together. The programmer thinks that it is simply stupid for the engineer to use the schematic diagram of the last generated netlist.
Speaking of the ground, the "common" or "return" is actually more appropriate, unless your circuit is connected to the ground pin of the wall socket (Figure 4). I admit that this is just a personal preference, but I like the American-style power and resistance symbols, there is a circle on the transistor and MOSFET, and the MOSFET clearly indicates the N-channel or P-channel type.
Figure 4: Various component symbols such as ground, power, resistor, transistor, and MOSFET
There are more information in the Part two: How to draw schematic symbols simply and easily(2).
Next Post: ELECTRONIC ASSEMBLY COMPANIES