MIPI D-PHY is intended for use in mobile devices including cameras, displays, and unified protocol interfaces. This standard is only supported in the HP I/O banks of the Kintex UltraScale+ and Virtex UltraScale+ FPGAs, and Zynq UltraScale+ MPSoCs. Support for this standard in UltraScale devices is in adherence to the MIPI alliance interface specifications. Typical MIPI D-PHY cores consist of four differential data lanes and one differential clock lane, as shown in This Figure.
MIPI D-PHY PCB guidelines for impedance, trace length, spacing, skew, and termination are shown in Table: MIPI D-PHY PCB Trace and Skew Guidelines.
Parameter |
Specification |
---|---|
Trace type |
Stripline |
Differential impedance |
95±10% W |
Trace length |
|
Spacing in pair |
2X trace width |
Spacing to other MIPI pairs |
5X trace width |
Spacing to other non-MIPI pairs |
6X trace width |
Skew between P/N per pair |
±2 ps |
Skew between clock and data |
±8 ps |
Skew between data and data |
±8 ps |
Termination at receiver |
100W differential(3) |
Notes: 1.For distances greater than 4 inches, receiver equalization must be used (CTLE). It is strongly recommended that the PCB (data-to-data and data-to-clock) be deskewed even if the deskew feature in the IP is used. 2.A reference stackup used for the longer trace lengths (4 to 10 inches) can be provided upon request. Operation at 2.5 Gb/s is supported as long as the channel loss is kept under 9 dB. 3.When FPGA/MPSoC is the receiver, 100W on-chip termination is recommended via the DIFF_TERM attribute. See the MIPI_DPHY section in UltraScale Architecture SelectIO Resources User Guide (UG571) [Ref 10]. |