Section 2.2. Hardware Basics

2.2. Hardware Basics The next step in understanding the hardware is to take a look at the schematic. A schematic is a drawing comprised of standardized symbols to represent all of a circuit's components and connections. The schematic gives the details of the hardware, showing the individual components represented in the block diagram, how the components are interconnected, and, most importantly, where to put the oscilloscope probe to see what's going on with a particular circuit. On most projects, it is not necessary for you to understand how every electrical circuit on the board operates, but you do need to understand the basic operation of the hardware. Along with the user's guides or manuals for the board, it is also helpful to collect the datasheets for all major components on your board. The datasheet is a complete specification of a particular hardware component, including electrical, timing, and interface parameters. Often the hardware engineer has already collected the datasheets; if so, your work is partly complete. You might want to take a look at the other information available for a particular component, because there are often separate hardware and software documents, especially for more complex devices. For example, a processor often includes a Programmer's Guide in addition to the other literature. These documents can give you valuable information for using various features of the processor; they occationally even provide code snippets. There are also application notes that address particular issues associated with a specific component. It is a good idea to look for any errata documents for all devices. The device's errata will give you a heads-up on any issues regarding the way a device operates, and, more importantly, workarounds for these issues. It's a good idea to periodically check for updates of the board components' information. This will save you the frustration of chasing a problem that was fixed in the latest datasheet. All of this information is an asset when you are trying to understand a new piece of hardware. You can generally find these documents on the manufacturer's web site. 2.2.1. Schematic Fundamentals Before we take a look at a schematic, let's go over some of the basics of schematics. Figure 2-3 shows some of the basic schematic symbols you will come across, although there might be some variations in symbols from schematic to schematic. The first column gives the name of the particular component; the second column shows the reference designator prefix or component name; and the third column shows the symbols for the related component. Figure 2-3. Basic schematic symbols You may notice that two symbols are shown for the diode component. The symbol on the right is for a light emitting diode (LED), which we will take a look at shortly. The symbols for ground and power can also vary from schematic to schematic; two symbols for power and ground are included in Figure 2-3. In addition to VCC, the reference designator commonly used for power is VDD. Since many circuits use multiple voltage levels, you may also come across power symbols that are labeled with the actual voltage, such as +5 or +3.3, instead of VCC or VDD. The power and ground symbols are typically placed vertically, as shown, whereas the other symbols in Figure 2-3 might show up in any orientation. A reference designator is a combination of letters, numbers, or both, which are used to identify components on a schematic. Reference designators typically include a number to aid in identifying a specific part in the schematic. For example, three resistors in a schematic might have reference designators R4, R21, and R428. The reference designators are normally silkscreened (a painted overlay) on the circuit board for part identification. Along with the reference designators, certain components (such as capacitors, inductors, and resistors) are marked by their associated values. For example, in Figure 2-4, resistor R39 has a value of 680 W. The values for some components on a schematic are written in a way to aid clarification. For example, a resistor with a value of 4.7 kW has its value written as 4K7. A resistor with a value of 12.4 W is written as 12R4. Using this method, it is easier to understand the value of the component should the decimal fail to print properly. You will also notice that integrated circuit symbols are not included in this figure. That is because IC schematic representations vary widely. Typically, a hardware engineer needs to create his own schematic symbol for the ICs used in the design. It is then up to the hardware engineer to use the clearest method possible to capture and represent the IC's symbol. The reference designator for ICs can vary as well. Typically, the letter U is used. The Arcom board schematic, however, uses the reference designator IC. IC symbols also include a component type or part number used by the manufacturer. The component type is often helpful when you need to hunt for the datasheets for the parts of a particular design. Descriptive text might save you the trouble of deciphering the markings and codes on the top of a specific IC. Now that we have an introduction to the symbols used in a schematic, let's take a look at a schematic. Figure 2-4 is a partial schematic of a fictional board. In this figure, we show some of the connections on the processor. Figure 2-4. Example schematic The italic labels and associated arrows are not part of the original schematic. These are included to point out particular aspects of the schematic. We wanted to note this because quite often text notes are included in the schematic for clarification or layout concerns. One such clarification note in Figure 2-4 is the label OUTPUT PORT on port PL1. The processor is the main component in this schematic. The symbol for the processor has a reference designator IC12, which is located at the top of the symbol on this schematic. The component type of the processor is PXA255 and is located at the bottom of the symbol. The processor's pins and their associated pin numbers run along the sides of the symbol. For example, bit 0 of the data bus is named D0 and is located on pin number 5 of the processor. You will also notice that some pins, such as P1.1/rts0 pin number 102, have a bar over the pin name. This indicates that the signal is active low. This means a logic level of 0 will activate the funtionality of this signal, whereas a logic level of 1 will deactivate the function. The opposite type of operation is active high. Active low functionality can also be indicated by a forward slash (/) or tilde (~) placed either before or after the signal name. The signal is then typically pronounced "not RTS0" or "RTS0 bar." Some component manufacturers represent an active-low signal with the prefix "n" in front of the signal name, such as nRESET. The wire connecting the different components together is called a net. Two nets that connect create a junction. On the schematic, a junction point is indicated by a dot, as you can see in Figure 2-4 on the RESET pin of the processor. In contrast, when two nets cross, but are not connected, there is no junction. This is shown where the net connected to the LED D3 crosses the net /RTS1. We say that pins not connected to any nets are no connects or open. No connects are often represented on a schematic with a small cross at the end of the net. Examples of no connect pins are shown on the processor pins A21 through A25. Sometimes IC manufacturers will label no connect pins NC. Related signals, such as data signals or address signals, are represented on a schematic by a thicker line called a bus net. For example, the data bus is labeled D[0..15] (in other schematics the data bus might be labeled [D0:D15]), which means the data bus net is made up of the signals D0 through D15. The individual signals are shown connecting to the processor data pins. Similarly, the address bus net is labeled A[1..20] and is made up of the signals A1 through A20. Nets connected to a bus net still need to be individually labeled. If each net in a schematic were connected to the desired location, after a while it could create quite a rat's nest. [*] Having nets cross over each other is typically avoided in order to maintain clarity in the schematic. To facilitate this clarity, the hardware engineer can use net labels to assign names to the nets. The convention is that nets marked with the same net label are connected, even if the engineer did not actually draw a line connecting them. [*] Incidentally, "rats nest" is the term used to describe the connection of nets made during layout. Once you see the initial stage of a layout, you'll understand how this name was derived. For example, in Figure 2-4, a portion of the connector with the reference designator PL1 is shown. (Incidentally, connectors and jumpers often use the reference designator J.) Because the net connected to pin number 23 of the connector PL1 is labeled A2, and the net connected to the processor's pin number 43 is labeled A2, they are connected even though the hardware engineer did not run a line to represent the A2 net connected from the processor over to PL1. In order to aid in testing the hardware and software, hardware engineers often include test points. A test point is a metallic area on the finished board that provides access to a particular signal to aid in the debugging or monitoring of that signal. One test point, with the reference designator TP11, is shown in Figure 2-4 on the RESET pin of the processor. With the move to smaller and smaller IC packages and smaller pins, test points are a necessity for debugging and also aid in production testing. Also, it is impossible to probe on any pins of a ball grid array (BGA) package part, because all of the pins are contained under the IC. In this case, a test point helps greatly. In cases where a schematic cannot fit onto a single page, there must be a way to interconnect nets from one page to another. This is the job of the off-page connector. Off-page connectors can be used for individual nets or bus nets. For example, the off-page connector of the data bus is D[0..15]. This is the exact same off-page connector name used on the memory page to connect the processor's data bus to the RAM's data bus. We have found a couple of ideas to be useful for off-page connectors. These might be useful mainly for the hardware engineer working on the schematic, but other people on the team should know about them, too. First, it is helpful if the signal type (input, output, or bidirectional) is properly represented by the appropriate off-page connector. Thus, in Figure 2-4, the data bus off-page connector indicates that these are bidirectional signals; the CPU_RESET off-page connector indicates that this signal is an input to the processor; and the TX1 off-page connector indicates that this signal is an output from the processor. Another helpful idea is to add a little text note next to each off-page connector with the page number(s) where that particular net is used. This might not make sense for a 5-page schematic, but flipping through 20 pages of schematics can be a nightmare. Additional tips can be found in the December 2002 Embedded Systems Programming [] article "Design for Debugability," which can be found online at http://www.embedded.com. [] Embedded Systems Programming magazine has changed its name to Embedded Systems Design. When you take a look at the full set of schematics, you will notice that there is a block at the lower-righthand corner of each page. This is the title block; every schematic we have come across has some version of this. The title block has useful information about the schematic, such as the date, the designer's name, the revision, a page number and description of the schematic on that page, and often a list of changes made. At this point, we have a solid understanding of the system components that make up our platform and how they are interconnected. Let's next see how to get to know the hardware.