Categories

PCB Routing Technology: How Much Impact Does Right-Angle Routing Have on Signal Transmission?

Jun 17th,2026 157 Views

01 Layout


10 Rules for Component Layout:

  1. Follow the layout principle of "large first, then small; difficult first, then easy", that is, important unit circuits and core components should be laid out first.
  2. Refer to the schematic block diagram during layout, and arrange the main components according to the main signal flow direction of the single board.
  3. The arrangement of components should be convenient for debugging and maintenance, that is, no large components should be placed around small components, and sufficient space should be left around components that need debugging.
  4. For circuit parts with the same structure, adopt "symmetrical" standard layout as much as possible.
  5. Optimize the layout according to the standards of uniform distribution, balanced center of gravity, and beautiful layout.
  6. Through-hole components of the same type should be placed in one direction in the X or Y direction. Polarized discrete components of the same type should also strive to be consistent in the X or Y direction to facilitate production and inspection.
  7. Heating elements should generally be evenly distributed to facilitate heat dissipation of the single board and the whole machine. Temperature-sensitive components except temperature detection components should be far away from components with large heat generation.
  8. The layout should try to meet the following requirements: the total wiring is as short as possible, and the key signal lines are the shortest; high-voltage and high-current signals are completely separated from low-current and low-voltage weak signals; analog signals are separated from digital signals; high-frequency signals are separated from low-frequency signals; the spacing between high-frequency components should be sufficient.
  9. The layout of decoupling capacitors should be as close as possible to the power pins of the IC, so that the loop formed between the power supply and the ground is the shortest.
  10. When laying out components, appropriately consider placing devices using the same power supply together to facilitate future power supply separation.

02 Routing


(1) Routing Priority


Key signal lines first: Key signals such as analog small signals, high-speed signals, clock signals and synchronization signals are routed first.

Density priority principle: Start routing from the device with the most complex connection relationship on the single board. Start routing from the area with the densest connections on the single board.

Notes:

a. Try to provide dedicated routing layers for key signals such as clock signals, high-frequency signals, and sensitive signals, and ensure their minimum loop area. If necessary, manual priority routing, shielding and increasing safety spacing should be adopted to ensure signal quality.

b. The EMC environment between the power layer and the ground layer is poor, so avoid arranging interference-sensitive signals.

c. Networks with impedance control requirements should be routed as much as possible according to the line length and line width requirements.

(2) Four Specific Routing Methods


1. Clock Routing


Clock lines are one of the factors that have the greatest impact on EMC. Fewer vias should be made on clock lines, avoid parallel routing with other signal lines as much as possible, and stay away from general signal lines to avoid interference to signal lines. At the same time, stay away from the power supply part of the board to prevent mutual interference between the power supply and the clock.

If there is a dedicated clock generation chip on the board, no routing is allowed under it, and copper should be laid under it, and special ground splitting can be performed if necessary. For many chips with reference crystal oscillators, no routing is allowed under these crystals either, and copper should be laid for isolation.


2. Right-Angle Routing


Right-angle routing is generally a situation that should be avoided as much as possible in PCB routing, and has almost become one of the standards for measuring the quality of routing. So how much impact does right-angle routing have on signal transmission? In principle, right-angle routing will change the line width of the transmission line, resulting in impedance discontinuity. In fact, not only right-angle routing, but also obtuse-angle and acute-angle routing may cause impedance changes.

The impact of right-angle routing on signals is mainly reflected in three aspects:

First, the corner can be equivalent to a capacitive load on the transmission line, slowing down the rise time;

Second, impedance discontinuity will cause signal reflection;

Third, EMI generated by the right-angle tip.

3. Differential Routing


See also: Altium Designer -- Differential Routing and Impedance Matching

Differential signals are increasingly used in high-speed circuit design, and the most critical signals in the circuit often adopt differential structure design. Definition: Popularly speaking, the driving end sends two equal and opposite signals, and the receiving end judges the logic state "0" or "1" by comparing the difference between the two voltages. The pair of traces carrying differential signals is called differential routing.

Compared with ordinary single-ended signal routing, the most obvious advantages of differential signals are reflected in the following three aspects:

a. Strong anti-interference ability, because the coupling between the two differential traces is very good. When there is noise interference from the outside, it is almost coupled to the two lines at the same time, and the receiving end only cares about the difference between the two signals, so the external common-mode noise can be completely canceled out.

b. Can effectively suppress EMI. For the same reason, due to the opposite polarity of the two signals, the electromagnetic fields they radiate to the outside can cancel each other out. The tighter the coupling, the less electromagnetic energy is released to the outside world.

c. Accurate timing positioning. Since the switching change of the differential signal is located at the intersection of the two signals, unlike ordinary single-ended signals which rely on high and low threshold voltages for judgment, it is less affected by process and temperature, can reduce timing errors, and is also more suitable for circuits with low-amplitude signals. The currently popular LVDS (Low Voltage Differential Signaling) refers to this small-amplitude differential signal technology.

For PCB engineers, the most concerned thing is how to ensure that these advantages of differential routing can be fully exerted in actual routing. Maybe anyone who has been exposed to Layout will understand the general requirements for differential routing, that is, "equal length and equal distance".

Equal length is to ensure that the two differential signals always maintain opposite polarities and reduce common-mode components; equal distance is mainly to ensure that the differential impedance of the two is consistent and reduce reflection. The "as close as possible principle" is sometimes also one of the requirements for differential routing.

4. Serpentine Routing


Serpentine routing is a type of routing often used in Layout. Its main purpose is to adjust the delay to meet the system timing design requirements.

Designers must first have such an understanding: Serpentine routing will destroy signal quality and change transmission delay, so it should be avoided as much as possible during routing. However, in actual design, in order to ensure that the signal has sufficient hold time, or to reduce the time offset between signals in the same group, it is often necessary to deliberately perform winding.

Notes:

  • For paired differential signal lines, generally route in parallel, make as few vias as possible, and when holes must be drilled, drill both lines together to achieve impedance matching.
  • A group of buses with the same attributes should be routed side by side as much as possible to be as equal in length as possible.
  • Vias drawn from SMD pads should be as far away from the pads as possible.
  • =20mil


(3) Common Routing Rules


1. Routing Direction Control Rule


That is, the routing directions of adjacent layers form an orthogonal structure. Avoid routing different signal lines in the same direction on adjacent layers to reduce unnecessary inter-layer crosstalk.

When this situation is difficult to avoid due to board structure limitations (such as some backplanes), especially when the signal rate is high, consider using ground planes to isolate each routing layer and ground signal lines to isolate each signal line.

2. Routing Open Loop Inspection Rule


Generally, routing with one end floating (Dangling Line) is not allowed, mainly to avoid the "antenna effect", reduce unnecessary interference radiation and reception, otherwise it may bring unpredictable results.


3. Impedance Matching Inspection Rule


The routing width of the same network should be consistent. Changes in line width will cause uneven characteristic impedance of the line, and reflection will occur when the transmission speed is high. This situation should be avoided as much as possible in the design.

Under certain conditions, such as the lead-out lines of connectors and the lead-out lines of BGA packages, it may be impossible to avoid changes in line width, and the effective length of the inconsistent part in the middle should be minimized as much as possible.


4. Routing Length Control Rule


That is, the short line rule. In the design, the routing length should be as short as possible to reduce interference problems caused by too long routing, especially for some important signal lines such as clock lines, be sure to place the oscillator very close to the device. For driving multiple devices, decide which network topology to adopt according to the specific situation.


5. Chamfering Rule


Acute angles and right angles should be avoided in PCB design to generate unnecessary radiation, and the process performance is also poor.


6. Component Decoupling Rule


A. Add necessary decoupling capacitors on the printed board to filter out interference signals on the power supply and stabilize the power supply signal.

In multilayer boards, the requirements for the position of decoupling capacitors are generally not too high, but for double-layer boards, the layout of decoupling capacitors and the wiring method of the power supply will directly affect the stability of the entire system, and sometimes even relate to the success or failure of the design.

B. In double-layer board design, generally the current should be filtered by the filter capacitor first before being supplied to the device.

C. In high-speed circuit design, the correct use of decoupling capacitors is related to the stability of the entire board.


7. Component Layout Partitioning/Layering Rule


A. Mainly to prevent mutual interference between modules with different operating frequencies, and at the same time try to shorten the routing length of the high-frequency part.

B. For hybrid circuits, there is also a method of arranging analog and digital circuits on both sides of the printed board respectively, using different layers for routing, and isolating them with a ground layer in the middle.

8. Ground Loop Rule


Minimum loop rule, that is, the loop area formed by the signal line and its return path should be as small as possible. The smaller the loop area, the less external radiation and the less external interference received.

9. Power and Ground Layer Integrity Rule


For areas with dense vias, attention should be paid to avoiding the mutual connection of holes in the hollowed-out areas of the power and ground layers, forming a division of the plane layer, thereby destroying the integrity of the plane layer, and further increasing the loop area of the signal lines in the ground layer.

10. 3W Rule


To reduce crosstalk between lines, ensure that the line spacing is large enough. When the center-to-center spacing of lines is not less than 3 times the line width, 70% of the electric field can be kept from interfering with each other, which is called the 3W rule. If 98% of the electric field is not to interfere with each other, a spacing of 10W can be used.

11. Shielding Protection


Corresponding to the ground loop rule, it is actually to minimize the loop area of the signal, which is more common for some important signals such as clock signals and synchronization signals.

For some particularly important and particularly high-frequency signals, copper coaxial cable shielding structure design should be considered, that is, the routed lines are isolated up, down, left and right with ground lines, and also consider how to effectively combine the shielding ground with the actual ground plane.

12. Routing Termination Network Rule


In high-speed digital circuits, when the delay time of PCB routing is greater than 1/4 of the signal rise time (or fall time), the routing can be regarded as a transmission line. To ensure that the input and output impedance of the signal correctly matches the impedance of the transmission line, various matching methods can be adopted. The selected matching method is related to the connection mode of the network and the routing topology.

A. For point-to-point (one output corresponding to one input) connection, you can choose source series matching or terminal parallel matching. The former has a simple structure and low cost, but has a large delay. The latter has a good matching effect but a complex structure and high cost.

B. For point-to-multipoint (one output corresponding to multiple outputs) connection, when the network topology is daisy chain, terminal parallel matching should be selected.

When the network is a star structure, you can refer to the point-to-point structure. Star and daisy chain are the two basic topologies, and other structures can be regarded as deformations of the basic structures, and some flexible measures can be taken for matching.

In actual operation, factors such as cost, power consumption and performance should be taken into account. Generally, perfect matching is not pursued, as long as the interference such as reflection caused by mismatch is limited to an acceptable range.

13. Routing Closed Loop Inspection Rule


Prevent signal lines from forming self-loops between different layers. Such problems are prone to occur in multilayer board design, and self-loops will cause radiation interference.

14. Routing Branch Length Control Rule


Try to control the length of the branch. The general requirement is Tdelay <= Trise/20.

15. Routing Resonance Rule


Mainly for high-frequency signal design, that is, the routing length must not be an integer multiple of its wavelength to avoid resonance.


16. Isolated Copper Area Control Rule


The appearance of isolated copper areas will bring some unpredictable problems. Therefore, connecting isolated copper areas with other signals helps to improve signal quality. Usually, isolated copper areas are grounded or deleted.

In actual production, PCB manufacturers add some copper foil to the vacant parts of some boards, which is mainly to facilitate the processing of printed boards and also has a certain effect on preventing warpage of printed boards.

17. Overlapping Power and Ground Layer Rule


Different power layers should avoid overlapping in space. Mainly to reduce interference between different power supplies, especially between power supplies with large voltage differences, the problem of overlapping power planes must be avoided as much as possible. When it is difficult to avoid, a middle ground layer can be considered.

18. 20H Rule


Since the electric field between the power layer and the ground layer is changing, electromagnetic interference will radiate outward at the edge of the board. This is called the edge effect.

The solution is to shrink the power layer inward so that the electric field is only conducted within the range of the ground layer. Taking one H (the dielectric thickness between the power supply and the ground) as a unit, if it is shrunk by 20H, 70% of the electric field can be confined within the edge of the ground layer; if it is shrunk by 100H, 98% of the electric field can be confined inside.

(4) Others


For single and double-layer boards, power lines should be as thick and short as possible. The width requirements of power lines and ground lines can be calculated according to the maximum current of 1A corresponding to a line width of 1mm. The loop formed by the power supply and the ground should be as small as possible.


To prevent coupled noise on the power line from directly entering the load device when the power line is long, decouple the power supply before entering each device. And to prevent mutual interference between them, independently decouple the power supply of each load, and ensure filtering before entering the load.


Maintain good grounding during routing. As shown in the figure below.


03 DDR Routing Rules


(1) First understand the composition of DDR2 signals: The DDR2 chip model I use is: MT47H64M16HG Package:


Pin Definitions


Table 3: FBGA 84-Ball - x16 and 60-Ball - x4, x8 Descriptions

SymbolTypeDescription
A12:0, A[13:0] (x4, x8)InputAddress inputs: Provide the row address for ACTIVATE commands, and the column address and auto precharge bit (A10) for READ/WRITE commands, to select one location out of the memory array in the respective bank. A10 sampled during a PRECHARGE command determines whether the PRECHARGE applies to one bank (A10 LOW, bank selected by BA[2:0]) or all banks (A10 HIGH). The address inputs also provide the op-code during a LOAD MODE command.
BA[2:0]InputBank address inputs: BA[2:0] define to which bank an ACTIVATE, READ, WRITE, or PRECHARGE command is being applied. BA[2:0] define which mode register, including MR, EMR, EMR(2), and EMR(3), is loaded during the LOAD MODE command.
CK, CK#InputClock: CK and CK# are differential clock inputs. All address and control input signals are sampled on the crossing of the positive edge of CK and negative edge of CK#. Output data (DQ and DQS/DQS#) is referenced to the crossings of CK and CK#.
CKEInputClock enable: CKE (registered HIGH) activates and CKE (registered LOW) deactivates clocking circuitry on the DDR2 SDRAM. The specific circuitry that is enabled/disabled is dependent on the DDR2 SDRAM configuration and operating mode. CKE LOW provides precharge power-down and SELF REFRESH operations (all banks idle), or ACTIVATE power-down (row active in any bank). CKE is synchronous for power-down entry, power-down exit, output disable, and for self refresh entry. CKE is asynchronous for self refresh exit. Input buffers (excluding CK, CK#, CKE, and ODT) are disabled during power-down. Input buffers (excluding CKE) are disabled during self refresh. CKE is an SSTL_18 input but will detect a LVCMOS LOW level after VDD is applied during first power-up. After VRef has become stable during the power-on and initialization sequence, it must be maintained for proper operation of the CKE receiver. For proper SELF REFRESH operation, VRef must be maintained.
CS#InputChip select: CS# enables (registered LOW) and disables (registered HIGH) the command decoder. All commands are masked when CS# is registered high. CS# provides for external bank selection on systems with multiple ranks. CS# is considered part of the command code.
LDM, UDM, DMInputInput data mask: DM is an input mask signal for write data. Input data is masked when DM is sampled HIGH along with that input data during a WRITE access. DM is sampled on both edges of DQS. Although DM balls are input-only, the DM loading is designed to match that of DQ and DQS balls. LDM is DM for lower byte DQ[7:0] and UDM is DM for upper byte DQ[15:8].
ODTInputOn-die termination: ODT (registered HIGH) enables termination resistance internal to the DDR2 SDRAM. When enabled, ODT is only applied to each of the following balls: DQ[15:0], LDM, UDM, LDQS, LDQS#, UDQS, and UDQS# for the x16; DQ[7:0], DQS, DQS#, RDQS, RDQS#, and DM for the x8; DQ[3:0], DQS, DQS#, and DM for the x4. The ODT input will be ignored if disabled via the LOAD MODE command.
RAS#, CAS#, WE#InputCommand inputs: RAS#, CAS#, and WE# (along with CS#) define the command being entered.
DQ15:0 DQ3:0 DQ7:0I/OData input/output: Bidirectional data bus for 64 Meg x 16. Bidirectional data bus for 256 Meg x 4. Bidirectional data bus for 128 Meg x 8.
DQS, DQS#I/OData strobe: Output with read data, input with write data for source synchronous operation. Edge-aligned with read data, center-aligned with write data. DQS# is only used when differential data strobe mode is enabled via the LOAD MODE command.
LDQS, LDQS#I/OData strobe for lower byte: Output with read data, input with write data for source synchronous operation. Edge-aligned with read data, center-aligned with write data. LDQS# is only used when differential data strobe mode is enabled via the LOAD MODE command.
UDQS, UDQS#I/OData strobe for upper byte: Output with read data, input with write data for source synchronous operation. Edge-aligned with read data, center-aligned with write data. UDQS# is only used when differential data strobe mode is enabled via the LOAD MODE command.
RDQS, RDQS#OutputRedundant data strobe: For x8 only. RDQS is enabled/disabled via the LOAD MODE command to the extended mode register (EMR). When RDQS is enabled, RDQS is output with read data only and is ignored during write data. When RDQS is disabled, ball B3 becomes data mask (see DM ball). RDQS# is only used when RDQS is enabled and differential data strobe mode is enabled.
VDDSupplyPower supply: 1.8V±0.1V.
VDDQSupplyDQ power supply: 1.8V ±0.1V. Isolated on the device for improved noise immunity.
VDDLSupplyDLL power supply: 1.8V ±0.1V.
VREFSupplySSTL_18 reference voltage (VDDQ/2).
VSSSupplyGround.
VSSDLSupplyDLL ground: Isolated on the device from VSS and VSSQ.
VSSQSupplyDQ ground: Isolated on the device for improved noise immunity.
NCNo connect: These balls should be left unconnected.
NFNo function: x8: these balls are used as DQ[7:4]; x4: they are no function.
NU-Not used: For x16 only. If EMR(E10) = 0, A8 and E8 are UDQS# and LDQS#. If EMR(E10) = 1, then A8 and E8 are not used.
NUNot used: For x8 only. If EMR(E10) = 0, A2 and E8 are RDQS# and DQS#. If EMR(E10) = 1, then A2 and E8 are not used.
RFUReserved for future use: Row address bits A13 (x16 only), A14, and A15.

Schematic Diagram

MT47H64M16HG

Data lines and address lines: Data lines: DQ[0-15], DQS, DM, (clock signal) CK/CK# Address lines: A[0-15], BA[0-2], (control signal) CS/WE/RAS/CAS, CKE, ODT

About Maxipcb
Maxipcb empowers innovators to turn cutting-edge technologies into reality.
We offer one-stop solutions for design, simulation, testing, PCB manufacturing, component procurement and SMT assembly, enabling efficient development, rapid deployment and risk control across the full product lifecycle.
Serving the world in communications, industrial automation, aerospace, automotive, semiconductor and beyond, we build a safer, more connected future together.