How to make a good PCB

 

• We all know that PCB is to turn the designed schematic diagram into a real PCB Board. Please do not underestimate this process. There are many things that work in principle that are hard to implement in the project, or what others can achieve, but others cannot. Therefore, it is not difficult to do a PCB, but it is not easy to do a PCB.
  The two major difficulties in the Microelectronics field are the processing of high-frequency and weak signals. In this regard, the PCB production level is particularly important. The same principle design, the same components, the PCB produced by different people has different results. How can we make a good PCB? Based on our past experience, I would like to share my views on the following aspects:
  
  I. Clear design objectives
  To accept a design task, we must first clarify its design objectives: General PCB, high-frequency PCB, small-signal processing PCB, or PCB with both high-frequency and small-signal processing, if it is an ordinary PCB, as long as the layout and wiring are reasonable and tidy, the mechanical size is accurate, if there is a medium load line and long line, it is necessary to use a certain means to deal with, reduce load, to enhance the drive for the long line, the focus is to prevent the long line reflection.
  When there is a signal line over 40 MHz on the board, special consideration should be given to these signal lines, such as crosstalk between lines. If the frequency is higher, there will be more stringent restrictions on the length of the wiring. According to the network theory of the distribution parameter, the interaction between the high-speed circuit and its connection is the decisive factor, you cannot ignore this feature during system design. As the door transmission speed increases, the opposition on the signal line will increase accordingly, and the crosstalk between adjacent signal lines will increase in a proportional manner. Generally, the power consumption and heat dissipation of High-Speed Circuits are also very high, high-Speed PCB should be paid enough attention.
  When the board has a millivolt or even microvolt level weak signal, special attention is required for these signal lines. Because the small signal is too weak, it is very vulnerable to interference from other strong signals, blocking measures are often necessary. Otherwise, the signal-to-noise ratio will be greatly reduced. So that useful signals are overwhelmed by noise and cannot be effectively extracted.
  The commissioning and testing of the Board should also be considered in the design phase. factors such as the physical location and isolation of the test points cannot be ignored, because some small and high-frequency signals cannot be directly added to the probe for measurement.
  In addition, we also need to consider other related factors, such as the number of layers of the Board, the encapsulated shape of components, and the mechanical strength of the Board. Before making PCB, you should be aware of the design objectives of the design.
  
  II. Measure the test taker's knowledge about the layout and wiring requirements of the components used.
  
  We know that some special components have special requirements for layout and wiring, such as the analog signal amplifiers used by Loti and APH. The analog signal amplifiers require stable power supply and small ripple. To simulate a small signal, try to stay away from the power device. On the Oti board, the Small Signal Amplification part also has a special shielding cover to shield stray electromagnetic interference. The glink chip used on the ntoi Board adopts the ECL process, causing high power consumption and heat dissipation. Special consideration must be given to the heat dissipation problem during layout. If natural heat dissipation is used, the glink chip should be placed in a place where the air flow is relatively smooth, and the heat generated cannot have a big impact on other chips. If the Board is equipped with speakers or other high-power devices, it may cause serious pollution to the power supply.
   
  Iii. Component layout considerations
  One of the first factors to consider the layout of components is the electrical performance. Put components with close connections together as much as possible, especially for some high-speed lines, the layout should be as short as possible, power signals and small-signal devices must be separated. When the circuit performance is met, components must be neatly arranged and beautifully arranged to facilitate testing. The mechanical size of the Board and the position of the socket must also be carefully considered.
  The transmission delay time on the grounding and interconnectivity in high-speed systems is also the first factor to be considered during system design. The transmission time on the signal line has a great impact on the overall system speed, especially for the High-Speed ECL circuit, although the integrated circuit block itself is very fast, however, the system speed can be greatly reduced due to the increase in latency caused by the use of common interconnectivity (latency of about 2 NS every 30cm lines) on the baseboard. for example, a shift register, a synchronization counter is recommended to be placed on the same plug-in board. Because the transmission delay time of clock signals to different plug-in boards is not equal, the shift register may cause a master error, if it cannot be placed on a board, synchronization is the key. The clock lines connecting from the public clock source to the plug-in board must be of the same length.

• 1st floor: bunny2007-12-07

  Iv. Wiring considerations
  With the completion of otni and star optical fiber network design, more boards with high-speed signal lines above MHz will be designed in the future. Here we will introduce some basic concepts of High-speed cables.
  
  1. Transmission Line
  Any "long" signal path on a printed circuit board can be considered as a transmission line. If the transmission delay time of this line is much shorter than the signal rising time, the main reflection produced during the signal rising will be drowned. There is no longer a forward, reverse, or ring. For most current MOS circuits, the increase time is much longer than the delay time of line transmission, so the length of the line can be measured in meters without signal distortion. For fast-speed logic circuits, especially Ultra-High-Speed ECL
  For an integrated circuit, because the edge speed increases rapidly, if there are no other measures, the length of the line must be greatly shortened to maintain the integrity of the signal.
  There are two ways to make the high-speed circuit work on a relatively long line without serious waveform distortion. TTL uses the toggle diode clamping method for the rapidly falling edge, so that the impulse is flushed to an electric flat with a lower voltage drop of a diode than the ground potential, which reduces the back-flushing margin and allows a punch at the slower rising edge, however, it is in the Level "H" state, the circuit's relatively high output impedance (50 ~ 80Ω. In addition, because the level "H" state has a high degree of anti-disturbance, the back-flushing problem is not very prominent. For HCT series devices, the improvement will be more obvious if the combination of the diode clamp and the serial resistance connection method is used.
  When fan-out is performed along the signal line, the TTL shaping method described above is somewhat inadequate at a high bit rate and a fast edge rate. Because there are reflection waves in the line, they will tend to be synthesized at a high rate, resulting in serious signal distortion and reduced anti-interference capabilities. Therefore, to solve the reflection problem, we usually use the line Impedance Matching Method in the ECL system. This method can control the reflection and ensure the integrity of the signal.
  Strictly speaking, for conventional TTL and CMOS devices with slow edge speeds, transmission lines are not very much needed. transmission lines are not always required for High-Speed ECL Devices with fast edge speeds. However, when transmission lines are used, they have the advantages of predicting the connection latency and controlling reflection and oscillation through impedance matching. 1
  There are five basic factors that determine whether transmission lines are used. They are: (1) system signal speed, (2) link distance, (3) capacitive load (fan-out), (4) resistance load (line connection mode); (5) Percentage of repeat and overhead allowed (degree of reduction of AC immunity ).
  2. Transmission Line types
  (1) Coaxial Cables and twisted pair wires: they are often used for connection between systems. The characteristic impedance of the coaxial cable is generally 50 Ω and 75 Ω, And the twisted pair is usually 110 Ω.
  (2) strip on the printed board
  A strip line is a band Guide (signal line). It is separated from the ground plane by a dielectric. If the thickness, width, and distance between the line and the ground plane are controllable, the characteristic impedance is also controllable. The Characteristic Impedance z0 of the microline is:
  
  Medium: [er is the relative dielectric constant of Printed Board Media
  6 is the thickness of the dielectric layer
  W is the line width
  T indicates the wire thickness.
  The transmission delay time of a line with a unit length depends only on the dielectric constant and is independent of the width or interval of the line.
  (3) cables in printed boards
  A copper strip is a dielectric line placed in the middle of a two-layer conductive plane. If the thickness and width of the wire, the dielectric constant of the media, and the distance between two Conductive Planes are controllable, the characteristic impedance of the wire is also controllable. The characteristic impedance of the wire B is:
  
  Formula: B is the distance between two geographic panels.
  W is the line width
  T indicates the wire thickness.
  Similarly, the transmission delay time per unit of length of the dummy line is independent of the width or spacing of the wire. It only depends on the relative dielectric constant of the media used.
   

• 2nd floor: bunny2007-12-07

  3. End-to-End Transmission Line
  When the receiving end of a line is connected with a resistor with the same line characteristic impedance, the transmission line is called a parallel connection. It is mainly used to obtain the best electrical performance, including driving Distributed Loads.
  Sometimes, in order to save power consumption, the peer resistor is connected to a 104 capacitor to form an AC connection circuit, which can effectively reduce DC losses.
  A resistor is concatenated between the drive and the transmission line, and the terminal of the wire is no longer connected to the resistor. This method is called series connection. Over-rush and ringing over a long line can be controlled by series damping or series connection technology. The series damping is a small resistance (generally 10 ~ 75 Ω). This damping method is suitable for connection with the controlled line of the Characteristic Impedance (such as the floor wiring, the ground-less plane circuit board and most winding wiring.
  The sum of the value of the series resistance and the output impedance of the circuit (Driving door) is equal to the characteristic impedance of the transmission line. the series connection end has the disadvantages that the terminal can only use the total load and long transmission delay. however, this can be overcome by using redundant series-connected transmission lines.
  4. Non-End Transmission Line
  
  If the line delay time is much shorter than the signal rise time, you can use the transmission line without the connection or connection, if the two-way delay of a non-end connection (one round-trip time of the signal on the transmission line) is shorter than the rising time of the pulse signal, the backlash caused by non-terminal connections is about 15% of the logic swing. The maximum length of the route is approximately:
  Lmax <TR/2tpd
  Formula: TR is the rising time
  TPD indicates the transmission delay time per line length.
  5. Comparison of several Connection Methods
  Both the connection at the parallel end and the connection at the serial end have their own advantages. either or both of them are used depends on the interests of the designer and the requirements of the system.
  The main advantage of the connection at the parallel end is that the system speed is fast and the signal transmission is completely non-distorted online. The load on the long line will not affect the transmission delay time of the driver door of the long line, but will not affect its signal edge speed, but will increase the transmission delay time of the signal along the long line. When driving a large fan-out, the load can be distributed along the branch line, rather than the terminal that must aggregate the load set online as in the connection.
  The series connection method enables the circuit to drive several parallel load lines. The delay time Increment Caused by the capacitive load of the series connection is about twice as large as that of the corresponding parallel connection, however, due to the capacitive load, the edge speed slows down and the drive door delay time increases. However, the crosstalk of the series end connection is smaller than that of the parallel end connection, the main reason is that the signal transmitted along the connection end is only 1/2 of the logic amplitude, so the switch current is only half of the switch current of the parallel end, and the signal energy is small crosstalk.
  

• 3rd Floor: bunny2007-12-07

  V. PCB layout technology
  When making a PCB, it is determined by the maximum operating frequency, the complexity of the circuit system, and the Assembly density requirements. When the clock frequency exceeds MHz, it is best to use multiple layers. If the operating frequency exceeds 350 MHz, it is best to use PTFE as the printed circuit board of the media layer, because its high-frequency attenuation is smaller, the parasitic capacitor is smaller, and the transmission speed is faster, due to the high z0 power consumption, z0 has the following requirements on the cabling of printed circuit boards:
  (1) try to keep a large interval between all parallel signal lines to reduce crosstalk. If there are two signal lines that are close to each other, it is best to take a ground line between the two lines, which can shield the line.
  (2) Avoid sharp turns when designing a signal transmission line to prevent sudden changes in transmission line characteristic impedance from reflection. Design a circular arc line with a certain size as much as possible.
  (3) the width of the printed line can be calculated based on the characteristic impedance calculation formula of the above-mentioned microline and cable. The characteristic impedance of the microline on the printed circuit board is generally 50 ~ Between 120. To obtain a large characteristic impedance, the line width must be narrow. However, very fine lines are not easy to create. Considering a variety of factors, it is generally appropriate to select an impedance of around 68 Ω, because the selection of Characteristic Impedance of 68 Ω can achieve the optimal balance between the delay time and power consumption. A 50Ω transmission line will consume more power. A large impedance can reduce the power consumption, but it will increase the transmission delay. The negative capacitance will increase the transmission delay time and decrease the characteristic impedance. However, the intrinsic capacitance of a line segment with low characteristic impedance is large, so the transmission delay time and characteristic impedance are less affected by the load capacitance. An important feature of transmission lines with proper end-to-end connection is that the branch-to-line short-term has no effect on the line delay time. When z0 is 50Ω. The length of the branch short line must be less than 5cm.
  (4) For the dual panel (or layer-4 cables in the 6-board), the cables on both sides of the circuit board must be perpendicular to each other to prevent mutual sensing of main crosstalk.
  (5) If the Printed Board is equipped with large current devices, such as relays, indicator lights, and speakers, it is best to separate the ground wires to reduce noise on the ground line, the ground wires of these large current devices should be connected to an independent ground bus on the plug-in board and backplane, and these independent Ground Wires should also be connected to the ground points of the entire system.
  (6) If there is a small signal amplifier on the board, the weak signal line before amplification should be far away from the strong signal line, and the line should be as short as possible, if possible, it should be shielded by the ground line.

How to make a good PCB