How to Draw Schematic Symbols Simply and Easily(2)
I have met a professor who will give you a failing judgment if he sees that you have a geodetic symbol on the car radio schematic. A car chassis is a different symbol, whether Altium calls it the earth or the triangle symbol you use on most PCB board, it means the common or the return. My personal preference is to use arrows to represent the power supply. I have never encountered an engineer who likes the resistance of European paintings like R1 and R2. Even the variable resistance symbol R3 in Altium has no meaning unless it has three feet. Or short the two feet together on the package. I also like the circles on the transistors, the short pins, the letters N or P to clearly show the type of MOSFET, and the gate pins that help to show the type of the tube, the type of P-channel that can be flipped so that the source is on top, Because more positive power is also on. I really appreciate the Altium/CircuitStudio display body diode.
In modern pcb designs, the problem with invisible power and ground pins is that the circuit often burns out when the power supply connection to the layout package is incorrect. I often burn. This is a very serious problem, because you may have multiple pcb layers with power, and it is very difficult to re-do the PCB or even rebuild the prototype. For this reason, many of us will explicitly draw the power pins. There are three ways to implement a multi-element package like a four op amp (Figure 5). The first method is that you can draw power pins on each component. The second method is to draw only the power pins on one of the components. Make sure to put all unused components on the schematic. The third method is to design the quad op amp into a heterogeneous package of five components, including four independent op amps and a separate power and ground pin component. The advantage of this method is that you can put the power supply and ground components and all decoupling capacitors together. The downside is that you may have forgotten to put power and ground components, and the resulting disaster is that the device is not powered, not the wrong power supply. One trick is to use the power pin as the first electronic component in the package so that when you place the component, the first one is the power supply. In any case, you should put all the components in the schematic to properly bias the unused components to prevent them from oscillating.
Figure 5: Do not use zero-length pins for power and ground
Instead, it is best to draw a power pin on each component of U1. You can also draw power pins on only one component of the package, but make sure all components are placed so you don't forget to connect the power supply (U2). The U3 package uses a separate "component" to draw power and ground. The advantage of this is that you can flip the op amp and flexibly place the negative pin above or below the positive pin, depending on your circuit needs.
These heterogeneous components were found in Cadence's OrCAD more than a decade ago, and this method can also break the connector into pieces. This is also done to maintain the signal flow of the schematic Diagram , ensuring that each wire is connected to the correct connector (Figure 6). Now you can make sure that your schematic flow is from left to right, making it easier for other engineers to understand, and it will make it easier to understand when you look at it in 5 years.
Figure 6: If you draw the connector only as a symbol, it will make the schematic messy (a)
By using the heterogeneous component functions in OrCAD, or the component "mode" in Altium/CircuitStudio, you can decompose the connector so that the flow of the schematic is clearer and easier to understand (b).
Another consideration is how to draw complex components such as switching power chips. Even if you move the input to the left and the output to the right, it is still difficult to understand how this component works. In this case, you can draw a simple diagram in the symbol box to indicate the function of this component. Not necessarily the block diagram in the data sheet, just a simple statement to remind you and others what this component does.
There are other conventions for schematic symbols, which are more of a preference than a good design principle. I really like to surround the transistor with a circle. It needs to be reiterated that the transistors painted by semiconductor engineers have no circles. I think the circle is very useful. Again, I really like to make a small jump when the line crosses. This leads to another important rule: there are no 4-way nodes. I have seen a schematic that was faxed, and I can't see if the wires are just crossed rather than connected. As a result, I guessed it wrong, which was a waste of my day. If all schematics are jumpered, the "no 4-way nodes" rule is less important. To my delight, the latest version of Altium/CircuitStudio can display jumpers and automatically prevent the generation of 4-way nodes (Figure 7).
Figure 7: Old people like me like to use jumper when there is no connection between the lines.
It should be noted that the 4-way node is a contraindication in the schematic. Altium/CircuitStudio has the option to generate jumpers, as well as the ability to eliminate cross-junctions by setting trace offsets, as shown in the GND connection of this chip. Note that the left side of the library component is the output, and the right side is the input, as opposed to what you think.
My approach is to redraw the symbol of the component using the rules entered on the left (Figure 8). I also used separate power and ground symbols to reduce clutter. After all, we are concerned with signal flow. Most engineers understand the internal functions of the 555 timing chip. But if you don't know, or if you think the person reading the schematic doesn't know, then you can draw some or all of the block diagram inside the component. Altium/CircuitStudio allows you to place images on schematic symbols, so I found a good 555 timer block diagram on the web, and after some minor adjustments I put it into the schematic symbol. I have to follow their pinout structure, so there are some jumps on the schematic (Figure 9).
Figure 8: Modify the 555 timer in Figure 7, placing the input on the left and the output on the right, so that the schematic flow is clearer. A separate power and ground symbol eliminates the clutter of the traces.
Figure 9: You can draw a block diagram inside the component to show its function.
This can be as simple as displaying an open collector output, or as complex as displaying the internal functions of a switching power supply chip. Some CAD packages allow you to paste images into component symbols.
There is a key point here. You can use the entire schematic to represent the internal functions of the component, or you can make the schematic simpler if you don't care about the internal functions of the component. My idea is to draw some content inside the component, such as the open collector output, but it is important to keep the entire schematic clear and organized, and people seem to understand it.
In college, John often joked that power transistors should be drawn a little thicker. We all sneered at the time, but now I really like to display the TO-3 giant packaged transistors with larger symbols (Figure 10). Being a simulation engineer has to accept the principle of importance, and larger transistors are more important and easier to draw.
Figure 10: Everyone can see that the transistor on the right is a power transistor
Schematic symbol preferences are like music preferences, they are very personal. This is a style of your work as an engineer. Things like jumpers and circles on transistors aren't very important, and things like input on the left and top, and output on the right and bottom are more important. We are all arguing about how to handle buses with both inputs and outputs. I think the local symbol is very important. There is an application note online, and the note says that if you connect it to the ground as suggested by the symbol, it is possible to burn out the diode.