ds925 Zynq Ultrascale Plus
ds925 Zynq Ultrascale Plus
Summary
The AMD Zynq™ UltraScale+™ MPSoCs are available in -3, -2, -1 speed grades, with -3E devices having the
highest performance. The -2LE and -1LI devices can operate at a VCCINT voltage at 0.85V or 0.72V and are
screened for lower maximum static power. When operated at VCCINT = 0.85V, using -2LE and -1LI devices, the
speed specification for the L devices is the same as the -2I or -1I speed grades. When operated at
VCCINT = 0.72V, the -2LE and -1LI performance and static and dynamic power is reduced.
DC and AC characteristics are specified in extended (E), industrial (I), automotive (Q), and military (M)
temperature ranges. Except the operating temperature range or unless otherwise noted, all the DC and AC
electrical parameters are the same for a particular speed grade (that is, the timing characteristics of a -1 speed
grade extended device are the same as for a -1 speed grade industrial device). However, only selected speed
grades and/or devices are available in each temperature range.
The XQ references in this data sheet are specific to the devices available in XQ Ruggedized packages. See the
Defense-Grade UltraScale Architecture Data Sheet: Overview (DS895) for further information on XQ Defense-
grade part numbers, packages, and ordering information.
All supply voltage and junction temperature specifications are representative of worst-case conditions. The
parameters included are common to popular designs and typical applications.
This data sheet, part of an overall set of documentation on the Zynq UltraScale+ MPSoCs, is available on the
AMD website at www.xilinx.com/documentation.
DC Characteristics
Absolute Maximum Ratings
Table 1: Absolute Maximum Ratings
AMD Adaptive Computing is creating an environment where employees, customers, and partners feel welcome and included. To that end, we’re
removing non-inclusive language from our products and related collateral. We’ve launched an internal initiative to remove language that could
exclude people or reinforce historical biases, including terms embedded in our software and IPs. You may still find examples of non-inclusive
language in our older products as we work to make these changes and align with evolving industry standards. Follow this link for more
information.
AC Voltage Overshoot1 % of UI2 at –40°C to 100°C3 AC Voltage Undershoot1 % of UI2 at –40°C to 100°C3
VCCO + 0.30 100% –0.30 100%
VCCO + 0.35 100% –0.35 90%
VCCO + 0.40 100% –0.40 78%
VCCO + 0.45 100% –0.45 40%
VCCO + 0.50 100% –0.50 24%
VCCO + 0.55 100% –0.55 18.0%
VCCO + 0.60 100% –0.60 13.0%
VCCO + 0.65 100% –0.65 10.8%
VCCO + 0.70 92% –0.70 9.0%
VCCO + 0.75 92% –0.75 7.0%
VCCO + 0.80 92% –0.80 6.0%
VCCO + 0.85 92% –0.85 5.0%
VCCO + 0.90 92% –0.90 4.0%
VCCO + 0.95 92% –0.95 2.5%
Notes:
1. A total of 200 mA per bank should not be exceeded.
2. For UI smaller than 20 µs.
3. For the -1Q devices, the upper temperature limit is 125°C. For the -1M devices, the temperature limits are –55°C to 125°C.
Table 7: VIN Maximum Allowed AC Voltage Overshoot and Undershoot for HP I/O Banks
AC Voltage Overshoot1 % of UI2 at –40°C to 100°C3 AC Voltage Undershoot1 % of UI2 at –40°C to 100°C3
VCCO + 0.30 100% –0.30 100%
VCCO + 0.35 100% –0.35 100%
VCCO + 0.40 92% –0.40 92%
VCCO + 0.45 50% –0.45 50%
VCCO + 0.50 20% –0.50 20%
VCCO + 0.55 10% –0.55 10%
VCCO + 0.60 6% –0.60 6%
VCCO + 0.65 2% –0.65 2%
VCCO + 0.70 2% –0.70 2%
Notes:
1. A total of 200 mA per bank should not be exceeded.
2. For UI smaller than 20 µs.
3. For the -1Q devices, the upper temperature limit is 125°C. For the -1M devices, the temperature limits are –55°C to 125°C.
Table 8: VPSIN Maximum Allowed AC Voltage Overshoot and Undershoot for PS I/O Banks
AC Voltage Overshoot1 % of UI2 at –40°C to 100°C3 AC Voltage Undershoot1 % of UI2 at –40°C to 100°C3
VCCO_PSIO + 0.30 100% –0.30 100%
VCCO_PSIO + 0.35 100% –0.35 75%
VCCO_PSIO + 0.40 100% –0.40 45%
VCCO_PSIO + 0.45 100% –0.45 40%
VCCO_PSIO + 0.50 75% –0.50 10%
VCCO_PSIO + 0.55 75% –0.55 6%
VCCO_PSIO + 0.60 60% –0.60 2%
VCCO_PSIO + 0.65 30% –0.65 0%
VCCO_PSIO + 0.70 20% –0.70 0%
VCCO_PSIO + 0.75 10% –0.75 0%
VCCO_PSIO + 0.80 10% –0.80 0%
VCCO_PSIO + 0.85 8% –0.85 0%
VCCO_PSIO + 0.90 6% –0.90 0%
VCCO_PSIO + 0.95 6% –0.95 0%
Notes:
1. A total of 200 mA per bank should not be exceeded.
2. For UI smaller than 20 µs.
3. For the -1Q devices, the upper temperature limit is 125°C. For the -1M devices, the temperature limits are –55°C to 125°C.
To achieve minimum current draw at power-on, the recommended power-on sequence for the low-power
domain (LPD) is listed. The recommended power-off sequence is the reverse of the power-on sequence.
1. VCC_PSINTLP
2. VCC_PSAUX, VCC_PSADC, and VCC_PSPLL in any order or simultaneously.
3. VCCO_PSIO
To achieve minimum current draw at power-on, the recommended power-on sequence for the full-power
domain (FPD) is listed. The recommended power-off sequence is the reverse of the power-on sequence.
The recommended power-on sequence to achieve minimum current draw for the GTH or GTY transceivers is
VCCINT, VMGTAVCC, VMGTAVTT OR VMGTAVCC, VCCINT, VMGTAVTT. There is no recommended sequencing for
VMGTVCCAUX. Both VMGTAVCC and VCCINT can be ramped simultaneously. The recommended power-off sequence
is the reverse of the power-on sequence to achieve minimum current draw. If these recommended sequences
are not met, current drawn from VMGTAVTT can be higher than specifications during power-up and power-down.
PS I/O Levels
Table 12: PS MIO and CONFIG DC Input and Output Levels
PL I/O Levels
Table 14: SelectIO DC Input and Output Levels For HD I/O Banks
Table 15: SelectIO DC Input and Output Levels for HP I/O Banks
Table 16: DC Input Levels for Single-ended POD10 and POD12 I/O Standards
VIL VIH
I/O Standard1, 2
V, Min V, Max V, Min V, Max
POD10 –0.300 VREF – 0.068 VREF + 0.068 VCCO + 0.300
POD12 –0.300 VREF – 0.068 VREF + 0.068 VCCO + 0.300
Notes:
1. Tested according to relevant specifications.
2. Standards specified using the default I/O standard configuration. For details, see the UltraScale Architecture SelectIO Resources User Guide
(UG571).
VICM (V)1 VID (V)2 VILHS3 VIHHS3 VOCM (V)4 VOD (V)5
I/O Standard
Min Typ Max Min Typ Max Min Max Min Typ Max Min Typ Max
SUB_LVDS8 0.500 0.900 1.300 0.070 – – – – 0.700 0.900 1.100 0.100 0.150 0.200
LVPECL 0.300 1.200 1.425 0.100 0.350 0.600 – – – – – – – –
SLVS_400_18 0.070 0.200 0.330 0.140 – 0.450 – – – – – – – –
SLVS_400_25 0.070 0.200 0.330 0.140 – 0.450 – – – – – – – –
MIPI_DPHY_ DCI_HS9, 10 0.070 – 0.330 0.070 – – –0.040 0.460 0.150 0.200 0.250 0.140 0.200 0.270
Notes:
1. VICM is the input common mode voltage.
2. VID is the input differential voltage (Q – Q).
3. VIHHS and VILHS are the single-ended input high and low voltages, respectively.
4. VOCM is the output common mode voltage.
5. VOD is the output differential voltage (Q – Q).
6. LVDS_25 is specified in Table 23.
7. LVDS is specified in Table 24.
8. The SUB_LVDS receiver is supported in HP I/O and HD I/O banks. The SUB_LVDS transmitter is supported only in HP I/O banks.
9. High-speed option for MIPI_DPHY_DCI. The VID maximum is aligned with the standard’s specification. A higher VID is acceptable as long
as the VIN specification is also met.
10. When operating at data rates of 1.5 Gb/s to 2.5 Gb/s, the minimum VID is 0.040V. These data rates, outlined in Table 72 are supported for
XC and XA devices only.
Table 18: Complementary Differential SelectIO DC Input and Output Levels for HD I/O Banks
VICM (V)1 VID (V)2 VOL (V)3 VOH (V)4 IOL IOH
I/O Standard
Min Typ Max Min Max Max Min mA mA
DIFF_HSTL_I 0.300 0.750 1.125 0.100 – 0.400 VCCO – 0.400 8.0 –8.0
DIFF_HSTL_I_18 0.300 0.900 1.425 0.100 – 0.400 VCCO – 0.400 8.0 –8.0
DIFF_HSUL_12 0.300 0.600 0.850 0.100 – 20% VCCO 80% VCCO 0.1 –0.1
DIFF_SSTL12 0.300 0.600 0.850 0.100 – (VCCO/2) – 0.150 (VCCO/2) + 0.150 14.25 –14.25
DIFF_SSTL135 0.300 0.675 1.000 0.100 – (VCCO/2) – 0.150 (VCCO/2) + 0.150 8.9 –8.9
DIFF_SSTL135_II 0.300 0.675 1.000 0.100 – (VCCO/2) – 0.150 (VCCO/2) + 0.150 13.0 –13.0
DIFF_SSTL15 0.300 0.750 1.125 0.100 – (VCCO/2) – 0.175 (VCCO/2) + 0.175 8.9 –8.9
DIFF_SSTL15_II 0.300 0.750 1.125 0.100 – (VCCO/2) – 0.175 (VCCO/2) + 0.175 13.0 –13.0
DIFF_SSTL18_I 0.300 0.900 1.425 0.100 – (VCCO/2) – 0.470 (VCCO/2) + 0.470 8.0 –8.0
DIFF_SSTL18_II 0.300 0.900 1.425 0.100 – (VCCO/2) – 0.600 (VCCO/2) + 0.600 13.4 –13.4
Notes:
1. VICM is the input common mode voltage.
2. VID is the input differential voltage.
3. VOL is the single-ended low-output voltage.
4. VOH is the single-ended high-output voltage.
Table 19: Complementary Differential SelectIO DC Input and Output Levels for HP I/O Banks
VICM (V)2 VID (V)3 VOL (V)4 VOH (V)5 IOL IOH
I/O Standard1
Min Typ Max Min Max Max Min mA mA
DIFF_HSTL_I 0.680 VCCO/2 (VCCO/2) + 0.150 0.100 – 0.400 VCCO – 0.400 5.8 –5.8
DIFF_HSTL_I_12 0.400 x VCCO VCCO/2 0.600 x VCCO 0.100 – 0.250 x VCCO 0.750 x VCCO 4.1 –4.1
DIFF_HSTL_I_18 (VCCO/2) – 0.175 VCCO/2 (VCCO/2) + 0.175 0.100 – 0.400 VCCO – 0.400 6.2 –6.2
DIFF_HSUL_12 (VCCO/2) – 0.120 VCCO/2 (VCCO/2) + 0.120 0.100 – 20% VCCO 80% VCCO 0.1 –0.1
DIFF_SSTL12 (VCCO/2) – 0.150 VCCO/2 (VCCO/2) + 0.150 0.100 – (VCCO/2) – 0.150 (VCCO/2) + 0.150 8.0 –8.0
DIFF_SSTL135 (VCCO/2) – 0.150 VCCO/2 (VCCO/2) + 0.150 0.100 – (VCCO/2) – 0.150 (VCCO/2) + 0.150 9.0 –9.0
DIFF_SSTL15 (VCCO/2) – 0.175 VCCO/2 (VCCO/2) + 0.175 0.100 – (VCCO/2) – 0.175 (VCCO/2) + 0.175 10.0 –10.0
DIFF_SSTL18_I (VCCO/2) – 0.175 VCCO/2 (VCCO/2) + 0.175 0.100 – (VCCO/2) – 0.470 (VCCO/2) + 0.470 7.0 –7.0
Notes:
1. DIFF_POD10 and DIFF_POD12 HP I/O bank specifications are shown in Table 20, Table 21, and Table 22.
2. VICM is the input common mode voltage.
3. VID is the input differential voltage.
4. VOL is the single-ended low-output voltage.
5. VOH is the single-ended high-output voltage.
Table 20: DC Input Levels for Differential POD10 and POD12 I/O Standards
Table 21: DC Output Levels for Single-ended and Differential POD10 and POD12 Standards
Table 22: Definitions for DC Output Levels for Single-ended and Differential POD10 and POD12
Standards
AC Switching Characteristics
All values represented in this data sheet are based on the speed specifications in the AMD Vivado™ Design
Suite as outlined in the following table.
2023.1 Device
1.03 XCZU3TCG, XCZU3TEG
1.30 XCZU1CG, XCZU1EG, XCZU2CG, XCZU2EG, XCZU3CG, XCZU3EG, XCZU4CG, XCZU4EG, XCZU4EV, XCZU5CG, XCZU5EG,
XCZU5EV, XCZU6CG, XCZU6EG, XCZU7CG, XCZU7EG, XCZU7EV, XCZU9CG, XCZU9EG, XCZU11EG, XCZU15EG,
XCZU17EG, XCZU19EG
XAZU1EG
XQZU3EG, XQZU5EV, XQZU7EV, XQZU9EG, XQZU11EG, XQZU15EG, XQZU19EG
1.31 XAZU2EG, XAZU3EG, XAZU4EV, XAZU5EV, XAZU7EV, XAZU11EG
Switching characteristics are specified on a per-speed-grade basis and can be designated as Advance,
Preliminary, or Production. Each designation is defined as follows:
• Advance Product Specification: These specifications are based on simulations only and are typically available
soon after device design specifications are frozen. Although speed grades with this designation are
considered relatively stable and conservative, some under-reporting might still occur.
• Preliminary Product Specification: These specifications are based on complete ES (engineering sample)
silicon characterization. Devices and speed grades with this designation are intended to give a better
indication of the expected performance of production silicon. The probability of under-reporting delays is
greatly reduced as compared to Advance data.
• Product Specification: These specifications are released once enough production silicon of a particular
device family member has been characterized to provide full correlation between specifications and devices
over numerous production lots. There is no under-reporting of delays, and customers receive formal
notification of any subsequent changes. Typically, the slowest speed grades transition to production before
faster speed grades.
For more specific, more precise, and worst-case guaranteed data, use the values reported by the static timing
analyzer and back-annotate to the simulation net list. Unless otherwise noted, values apply to all Zynq
UltraScale+ MPSoCs.
Table 27 lists the production released Zynq UltraScale+ MPSoC, speed grade, and the minimum corresponding
supported speed specification version and Vivado software revisions. The Vivado software and speed
specifications listed are the minimum releases required for production. All subsequent releases of software and
speed specifications are valid.
Table 27: Zynq UltraScale+ MPSoC Device Production Software and Speed Specification Release
Table 27: Zynq UltraScale+ MPSoC Device Production Software and Speed Specification Release
(cont'd)
Table 27: Zynq UltraScale+ MPSoC Device Production Software and Speed Specification Release
(cont'd)
Speed Grade
Symbol Description Units
-3 -2 -1
FAPUMAX Maximum APU clock frequency 1500 1333 1200 MHz
FRPUMAX Maximum RPU clock frequency 600 533 500 MHz
FGPUMAX Maximum GPU clock frequency 667 600 600 MHz
Speed Grade
Symbol Description Units
-3 -2 -1
FCSUCIBMAX Maximum CSU crypto interface block frequency 400 400 400 MHz
Speed Grade
-3E -2I
Memory -2E/-2LE -1I/-1M/-1Q
Package DRAM Type Units
Standard
-1E -1LI
Speed Grade
-3E -2I
Memory -2E/-2LE -1I/-1M/-1Q
Package DRAM Type Units
Standard
-1E -1LI
Speed Grade
Memory Standard Mode -3 -2 -1 Units
Max Max Max
NV-DDR1 5 200 200 200 Mb/s
4 166.6 166.6 166.6 Mb/s
3 133.3 133.3 133.3 Mb/s
2 100 100 100 Mb/s
1 66.6 66.6 66.6 Mb/s
0 40 40 40 Mb/s
Notes:
1. The PS NAND memory controller interface for NV-DDR switching characteristics meets the requirements of the ONFI 3.1 specification.
Speed Grade
Memory Standard Mode -3 -2 -1 Units
Max Max Max
SDR1, 2 5 50 50 50 Mb/s
4 40 40 40 Mb/s
3 33.3 33.3 33.3 Mb/s
2 28.5 28.5 28.5 Mb/s
1 20 20 20 Mb/s
0 10 10 10 Mb/s
Notes:
1. The PS NAND memory controller interface for SDR switching characteristics meets the requirements of the ONFI 3.1 specification.
2. The NAND controller reference clock frequency maximum is 83 MHz.
PS Switching Characteristics
PS Clocks
Table 34: PS Reference Clock Requirements
Speed Grade
Symbol Description Units
-3 -2 -1
TLOCKPSPLL PLL maximum lock time 100 100 100 µs
FPSPLLMAX PLL maximum output frequency 1600 1600 1600 MHz
FPSPLLMIN PLL minimum output frequency 750 750 750 MHz
FPSPLLVCOMAX PLL maximum VCO frequency 3000 3000 3000 MHz
FPSPLLVCOMIN PLL minimum VCO frequency 1500 1500 1500 MHz
Speed Grade
Symbol Description Units
-3 -2 -1
FTOPSW_MAINMAX FPD AXI interconnect clock maximum frequency 600 533 533 MHz
FTOPSW_LSBUSMAX FPD APB bus clock maximum frequency 100 100 100 MHz
FGDMAMAX FPD-DMA controller clock maximum frequency 600 600 600 MHz
FDPDMAMAX DisplayPort controller clock maximum frequency 600 600 600 MHz
FLPD_SWITCH_CTRLMAX LPD AXI interconnect clock maximum frequency 600 500 500 MHz
FLPD_LSBUS_CTRLMAX LPD APB bus clock maximum frequency 100 100 100 MHz
FADMAMAX LPD-DMA maximum frequency 600 500 500 MHz
FAPLL_TO_LPDMAX APLL_TO_LPD maximum frequency 533 533 533 MHz
FDPLL_TO_LPDMAX DPLL_TO_LPD maximum frequency 533 533 533 MHz
FVPLL_TO_LPDMAX VPLL_TO_LPD maximum frequency 533 533 533 MHz
FIOPLL_TO_FPDMAX IOPLL_TO_FPD maximum frequency 533 533 533 MHz
FRPLL_TO_FPDMAX RPLL_TO_FPD maximum frequency 533 533 533 MHz
PS Configuration
Table 39: Processor Configuration Access Port Switching Characteristics
PS Interface Specifications
PS Quad-SPI Controller Interface
Table 41: Generic Quad-SPI Interface
PS USB Interface
Table 43: ULPI Interface
PS DAP Interface
Table 50: DAP Interface
PS UART Interface
Table 51: UART Interface
PS Trace Interface
Table 53: Trace Interface
PS-GTR Transceiver
Table 56: PS-GTR Transceiver DC Specifications
Speed Grade
Symbol Description Units
-3 -2 -1
FGTRMAX PS-GTR maximum line rate 6.0 6.0 6.0 Gb/s
FGTRMIN PS-GTR minimum line rate 1.25 1.25 1.25 Gb/s
Table 61: PS-GTR Transceiver Reference Clock Oscillator Selection Phase Noise Mask
Offset
Symbol Description1 Min Typ Max Units
Frequency
PLLREFCLKMASK PLL reference clock select phase noise mask at 100 – – –102 dBc/Hz
REFCLK frequency = 25 MHz
1 KHz – – –124
10 KHz – – –132
100 KHz – – –139
1 MHz – – –152
10 MHz – – –154
PLL reference clock select phase noise mask at 100 – – –96 dBc/Hz
REFCLK frequency = 50 MHz
1 KHz – – –118
10 KHz – – –126
100 KHz – – –133
1 MHz – – –146
10 MHz – – –148
PLL reference clock select phase noise mask at 100 – – –90 dBc/Hz
REFCLK frequency = 100 MHz
1 KHz – – –112
10 KHz – – –120
100 KHz – – –127
1 MHz – – –140
10 MHz – – –142
PLL reference clock select phase noise mask at 100 – – –88 dBc/Hz
REFCLK frequency = 125 MHz
1 KHz – – –110
10 KHz – – –118
100 KHz – – –125
1 MHz – – –138
10 MHz – – –140
PLL reference clock select phase noise mask at 100 – – –86 dBc/Hz
REFCLK frequency = 150 MHz
1 KHz – – –108
10 KHz – – –116
100 KHz – – –123
1 MHz – – –136
10 MHz – – –138
Notes:
1. For reference clock frequencies not in this table, use the phase noise mask for the nearest reference clock frequency.
In each of the following LVDS performance tables, the I/O bank type is either high performance (HP) or high
density (HD).
• For the input/output registers in HP I/O banks, the Vivado tools limit clock frequencies to 312.9 MHz for all
speed grades.
• For IDDR in HP I/O banks, Vivado tools limit clock frequencies to 625.0 MHz for all speed grades.
• For ODDR in HP I/O banks, Vivado tools limit clock frequencies to 625.0 MHz for all speed grades.
The following table provides the maximum data rates for applicable memory standards using the Zynq
UltraScale+ MPSoC memory PHY. Refer to Memory Interfaces for the complete list of memory interface
standards supported and detailed specifications. The final performance of the memory interface is determined
through a complete design implemented in the Vivado Design Suite, following guidelines in the UltraScale
Architecture PCB Design User Guide (UG583), electrical analysis, and characterization of the system.
Table 74: Maximum Physical Interface (PHY) Rate for Memory Interfaces
• TINBUF_DELAY_PAD_I is the delay from IOB pad through the input buffer to the I-pin of an IOB pad. The delay
varies depending on the capability of the SelectIO input buffer.
• TOUTBUF_DELAY_O_PAD is the delay from the O pin to the IOB pad through the output buffer of an IOB pad.
The delay varies depending on the capability of the SelectIO output buffer.
• TOUTBUF_DELAY_TD_PAD is the delay from the T pin to the IOB pad through the output buffer of an IOB pad,
when 3-state is disabled. The delay varies depending on the SelectIO capability of the output buffer. In HP
I/O banks, the internal DCI termination turn-on time is always faster than TOUTBUF_DELAY_TD_PAD when the
DCITERMDISABLE pin is used. In HD I/O banks, the on-die termination turn-on time is always faster than
TOUTBUF_DELAY_TD_PAD when the INTERMDISABLE pin is used.
• TOUTBUF_DELAY_TE_PAD is the delay from the T pin to the IOB pad through the output buffer of an IOB pad,
when 3-state is enabled (i.e., a high impedance state).
• TINBUF_DELAY_IBUFDIS_O is the IOB delay from IBUFDISABLE to O output.
• In HP I/O banks, the internal DCI termination turn-off time is always faster than TOUTBUF_DELAY_TE_PAD when
the DCITERMDISABLE pin is used.
• In HD I/O banks, the internal IN_TERM termination turn-off time is always faster than TOUTBUF_DELAY_TE_PAD
when the INTERMDISABLE pin is used.
I/O Standard
Description VL1, 2 VH1, 2 VMEAS 1, 4 VREF 1, 3, 5
Attribute
LVCMOS, 1.2V LVCMOS12 0.1 1.1 0.6 –
LVCMOS, LVDCI, HSLVDCI, 1.5V LVCMOS15, LVDCI_15, 0.1 1.4 0.75 –
HSLVDCI_15
LVCMOS, LVDCI, HSLVDCI, 1.8V LVCMOS18, LVDCI_18, 0.1 1.7 0.9 –
HSLVDCI_18
LVCMOS, 2.5V LVCMOS25 0.1 2.4 1.25 –
LVCMOS, 3.3V LVCMOS33 0.1 3.2 1.65 –
LVTTL, 3.3V LVTTL 0.1 3.2 1.65 –
HSTL (high-speed transceiver logic), class I, 1.2V HSTL_I_12 VREF – 0.25 VREF + 0.25 VREF 0.6
HSTL, class I, 1.5V HSTL_I VREF – 0.325 VREF + 0.325 VREF 0.75
HSTL, class I, 1.8V HSTL_I_18 VREF – 0.4 VREF + 0.4 VREF 0.9
HSUL (high-speed unterminated logic), 1.2V HSUL_12 VREF – 0.25 VREF + 0.25 VREF 0.6
SSTL12 (stub series terminated logic), 1.2V SSTL12 VREF – 0.25 VREF + 0.25 VREF 0.6
SSTL135 and SSTL135 class II, 1.35V SSTL135, SSTL135_II VREF – 0.2875 VREF + 0.2875 VREF 0.675
SSTL15 and SSTL15 class II, 1.5V SSTL15, SSTL15_II VREF – 0.325 VREF + 0.325 VREF 0.75
SSTL18, class I and II, 1.8V SSTL18_I, SSTL18_II VREF – 0.4 VREF + 0.4 VREF 0.9
POD10, 1.0V POD10 VREF – 0.2 VREF + 0.2 VREF 0.7
POD12, 1.2V POD12 VREF – 0.24 VREF + 0.24 VREF 0.84
DIFF_HSTL, class I, 1.2V DIFF_HSTL_I_12 0.6 – 0.25 0.6 + 0.25 06 –
DIFF_HSTL, class I, 1.5V DIFF_HSTL_I 0.75 – 0.325 0.75 + 0.325 06 –
DIFF_HSTL, class I, 1.8V DIFF_HSTL_I_18 0.9 – 0.4 0.9 + 0.4 06 –
DIFF_HSUL, 1.2V DIFF_HSUL_12 0.6 – 0.25 0.6 + 0.25 06 –
DIFF_SSTL, 1.2V DIFF_SSTL12 0.6 – 0.25 0.6 + 0.25 06 –
DIFF_SSTL135 and DIFF_SSTL135 class II, 1.35V DIFF_SSTL135, 0.675 – 0.2875 0.675 + 0.2875 06 –
DIFF_SSTL135_II
I/O Standard
Description VL1, 2 VH1, 2 VMEAS 1, 4 VREF 1, 3, 5
Attribute
DIFF_SSTL15 and DIFF_SSTL15 class II, 1.5V DIFF_SSTL15, 0.75 – 0.325 0.75 + 0.325 06 –
DIFF_SSTL15_II
DIFF_SSTL18_I, DIFF_SSTL18_II, 1.8V DIFF_SSTL18_I, 0.9 – 0.4 0.9 + 0.4 06 –
DIFF_SSTL18_II
DIFF_POD10, 1.0V DIFF_POD10 0.5 – 0.2 0.5 + 0.2 06 –
DIFF_POD12, 1.2V DIFF_POD12 0.6 – 0.25 0.6 + 0.25 06 –
LVDS (low-voltage differential signaling), 1.8V LVDS 0.9 – 0.125 0.9 + 0.125 06 –
LVDS_25, 2.5V LVDS_25 1.25 – 0.125 1.25 + 0.125 06 –
SUB_LVDS, 1.8V SUB_LVDS 0.9 – 0.125 0.9 + 0.125 06 –
SLVS, 1.8V SLVS_400_18 0.9 – 0.125 0.9 + 0.125 06 –
SLVS, 2.5V SLVS_400_25 1.25 – 0.125 1.25 + 0.125 06 –
LVPECL, 2.5V LVPECL 1.25 – 0.125 1.25 + 0.125 06 –
MIPI D-PHY (high speed) 1.2V MIPI_DPHY_DCI_HS 0.2 – 0.125 0.2 + 0.125 06 –
MIPI D-PHY (low power) 1.2V MIPI_DPHY_DCI_LP 0.715 – 0.2 0.715 + 0.2 06 –
Notes:
1. The input delay measurement methodology parameters for LVDCI/HSLVDCI are the same for LVCMOS standards of the same voltage.
Parameters for all other DCI standards are the same for the corresponding non-DCI standards.
2. Input waveform switches between VL and VH.
3. Measurements are made at typical, minimum, and maximum VREF values. Reported delays reflect worst case of these measurements.
VREF values listed are typical.
4. Input voltage level from which measurement starts.
5. This is an input voltage reference that bears no relation to the VREF/VMEAS parameters found in IBIS models and/or noted in Figure 1.
6. The value given is the differential input voltage.
VREF
Output RREF
X16654-072117
X16640-072117
Parameters VREF, RREF, CREF, and VMEAS fully describe the test conditions for each I/O standard. The most
accurate prediction of propagation delay in any given application can be obtained through IBIS simulation, using
this method:
1. Simulate the output driver of choice into the generalized test setup using values from Table 79.
2. Record the time to VMEAS.
3. Simulate the output driver of choice into the actual PCB trace and load using the appropriate IBIS model or
capacitance value to represent the load.
4. Record the time to VMEAS.
5. Compare the results of step 2 and step 4. The increase or decrease in delay yields the actual propagation
delay of the PCB trace.
Table 87: Global Clock Input to Output Delay Without MMCM (Near Clock Region)
Table 88: Global Clock Input to Output Delay Without MMCM (Far Clock Region)
Table 91: Global Clock Input Setup and Hold With 3.3V HD I/O Without MMCM
Table 91: Global Clock Input Setup and Hold With 3.3V HD I/O Without MMCM (cont'd)
Table 92: Global Clock Input Setup and Hold With MMCM
Table 92: Global Clock Input Setup and Hold With MMCM (cont'd)
Table 94: Input Logic Characteristics for Dynamic Phase Aligned Applications (Component Mode)
VCMOUTDC Common mode output voltage: DC When remote RX is terminated to VMGTAVTT/2 – DVPPOUT/4 mV
coupled (equation based) GND
When remote RX termination is VMGTAVTT – DVPPOUT/2 mV
floating
When remote RX is terminated to mV
VRX_TERM2
VCMOUTAC Common mode output voltage: AC coupled (equation based) VMGTAVTT – DVPPOUT/2 mV
RIN Differential input resistance – 100 – Ω
ROUT Differential output resistance – 100 – Ω
TOSKEW Transmitter output pair (TXP and TXN) intra-pair skew (all packages) – – 10 ps
CEXT Recommended external AC coupling capacitor3 – 100 – nF
Notes:
1. The output swing and pre-emphasis levels are programmable using the attributes discussed in the UltraScale Architecture GTH
Transceivers User Guide (UG576), and can result in values lower than reported in this table.
2. VRX_TERM is the remote RX termination voltage.
3. Other values can be used as appropriate to conform to specific protocols and standards.
+V P
Single-Ended
Peak-to-Peak
N Voltage
0
X16653-072117
+V
Differential
0 Peak-to-Peak
Voltage
–V P–N
Differential peak-to-peak voltage = (Single-ended peak-to-peak voltage) x 2
X16639-072117
Table 97 and Table 98 summarize the DC specifications of the GTH transceivers input and output clocks in Zynq
UltraScale+ MPSoCs. Consult the UltraScale Architecture GTH Transceivers User Guide (UG576) for further details.
Table 100: GTH Transceiver Dynamic Reconfiguration Port (DRP) Switching Characteristics
Table 102: GTH Transceiver Reference Clock Oscillator Selection Phase Noise Mask
Offset
Symbol Description Min Typ Max Units
Frequency
QPLLREFCLKMASK1, 2 QPLL0/QPLL1 reference clock select phase noise 10 kHz – – –105 dBc/Hz
mask at REFCLK frequency = 312.5 MHz
100 kHz – – –124
1 MHz – – –130
CPLLREFCLKMASK1, 2 CPLL reference clock select phase noise mask at 10 kHz – – –105 dBc/Hz
REFCLK frequency = 312.5 MHz
100 kHz – – –124
1 MHz – – –130
50 MHz – – –140
Notes:
1. For reference clock frequencies other than 312.5 MHz, adjust the phase-noise mask values by 20 × Log(N/312.5) where N is the new
reference clock frequency in MHz.
2. This reference clock phase-noise mask is superseded by any reference clock phase-noise mask that is specified in a supported protocol,
e.g., PCIe.
Electrical
Protocol Specification Serial Rate (Gb/s)
Compliance
CAUI-10 IEEE 802.3-2012 10.3125 Compliant
nPPI IEEE 802.3-2012 10.3125 Compliant
10GBASE-KR1 IEEE 802.3-2012 10.3125 Compliant
40GBASE-KR IEEE 802.3-2012 10.3125 Compliant
SFP+ SFF-8431 (SR and LR) 9.95328–11.10 Compliant
XFP INF-8077i, revision 4.5 10.3125 Compliant
RXAUI CEI-6G-SR 6.25 Compliant
XAUI IEEE 802.3-2012 3.125 Compliant
1000BASE-X IEEE 802.3-2012 1.25 Compliant
5.0G Ethernet IEEE 802.3bx (PAR) 5 Compliant
2.5G Ethernet IEEE 802.3bx (PAR) 2.5 Compliant
HiGig, HiGig+, HiGig2 IEEE 802.3-2012 3.74, 6.6 Compliant
OTU2 ITU G.8251 10.709225 Compliant
OTU4 (OTL4.10) OIF-CEI-11G-SR 11.180997 Compliant
OC-3/12/48/192 GR-253-CORE 0.1555–9.956 Compliant
TFI-5 OIF-TFI5-0.1.0 2.488 Compliant
Interlaken OIF-CEI-6G, OIF-CEI-11G-SR 4.25–12.5 Compliant
PCIe Gen1, 2, 32 PCI Express base 3.0 2.5, 5.0, and 8.0 Compliant
SDI3 SMPTE 424M-2006 0.27–2.97 Compliant
Electrical
Protocol Specification Serial Rate (Gb/s)
Compliance
UHD-SDI3 SMPTE ST-2081 6G, SMPTE ST-2082 12G 6 and 12 Compliant
Hybrid memory cube (HMC) HMC-15G-SR 10, 12.5, and 15.0 Compliant
MoSys Bandwidth Engine CEI-11-SR and CEI-11-SR (overclocked) 10.3125, 15.5 Compliant
CPRI CPRI_v_6_1_2014-07-01 0.6144–12.165 Compliant
HDMI3 HDMI 2.0 All Compliant
Passive optical network (PON) 10G-EPON, 1G-EPON, NG-PON2, XG-PON, and 2.5G- 0.155–10.3125 Compliant
PON
JESD204a/b OIF-CEI-6G, OIF-CEI-11G 3.125–12.5 Compliant
Serial RapidIO RapidIO specification 3.1 1.25–10.3125 Compliant
DisplayPort3 DP 1.2B CTS 1.62–5.4 Compliant
Fibre channel FC-PI-4 1.0625–14.025 Compliant
SATA Gen1, 2, 3 Serial ATA revision 3.0 specification 1.5, 3.0, and 6.0 Compliant
SAS Gen1, 2, 3 T10/BSR INCITS 519 3.0, 6.0, and 12.0 Compliant
SFI-5 OIF-SFI5-01.0 0.625–12.5 Compliant
Aurora CEI-6G, CEI-11G-LR up to 11.180997 Compliant
Notes:
1. The transition time of the transmitter is faster than the IEEE Std 802.3-2012 specification.
2. For PCIe operating modes supported in each device package refer to Interlaken 150G LogiCORE IP Product Guide (PG212).
3. This protocol requires external circuitry to achieve compliance.
+V P
Single-Ended
Peak-to-Peak
N Voltage
0
X16653-072117
+V
Differential
0 Peak-to-Peak
Voltage
–V P–N
Differential peak-to-peak voltage = (Single-ended peak-to-peak voltage) x 2
X16639-072117
The following tables summarize the DC specifications of the clock input/output levels of the GTY transceivers in
Zynq UltraScale+ MPSoCs. Consult the UltraScale Architecture GTY Transceivers User Guide (UG578) for further
details.
Table 112: GTY Transceiver Dynamic Reconfiguration Port (DRP) Switching Characteristics
Table 114: GTY Transceiver Reference Clock Oscillator Selection Phase Noise Mask
Offset
Symbol Description1, 2 Min Typ Max Units
Frequency
QPLLREFCLKMASK QPLL0/QPLL1 reference clock select phase noise 10 kHz – – –112 dBc/Hz
mask at REFCLK frequency = 156.25 MHz
100 kHz – – –128
1 MHz – – –145
QPLL0/QPLL1 reference clock select phase noise 10 kHz – – –103 dBc/Hz
mask at REFCLK frequency = 312.5 MHz
100 kHz – – –123
1 MHz – – –143
QPLL0/QPLL1 reference clock select phase noise 10 kHz – – –98 dBc/Hz
mask at REFCLK frequency = 625 MHz
100 kHz – – –117
1 MHz – – –140
CPLLREFCLKMASK CPLL reference clock select phase noise mask at 10 kHz – – –112 dBc/Hz
REFCLK frequency = 156.25 MHz
100 kHz – – –128
1 MHz – – –145
50 MHz – – –145
CPLL reference clock select phase noise mask at 10 kHz – – –103 dBc/Hz
REFCLK frequency = 312.5 MHz
100 kHz – – –123
1 MHz – – –143
50 MHz – – –145
CPLL reference clock select phase noise mask at 10 kHz – – –98 dBc/Hz
REFCLK frequency = 625 MHz
100 kHz – – –117
1 MHz – – –140
50 MHz – – –144
Notes:
1. For reference clock frequencies not in this table, use the phase-noise mask for the nearest reference clock frequency.
2. This reference clock phase-noise mask is superseded by any reference clock phase-noise mask that is specified in a supported protocol,
e.g., PCIe.
Electrical
Protocol Specification Serial Rate (Gb/s)
Compliance
CAUI-4 IEEE 802.3-2012 25.78125 Compliant
28 Gb/s backplane CEI-25G-LR 25–28.05 Compliant
Interlaken OIF-CEI-6G, OIF-CEI-11GSR, OIF-CEI-28G-MR 4.25–25.78125 Compliant
100GBASE-KR4 IEEE 802.3bj-2014, CEI-25G-LR 25.78125 Compliant1
100GBASE-CR4 IEEE 802.3bj-2014, CEI-25G-LR 25.78125 Compliant1
50GBASE-KR4 IEEE 802.3by-2014, CEI-25G-LR 25.78125 Compliant1
50GBASE-CR4 IEEE 802.3by-2014, CEI-25G-LR 25.78125 Compliant1
25GBASE-KR4 IEEE 802.3by-2014, CEI-25G-LR 25.78125 Compliant1
25GBASE-CR4 IEEE 802.3by-2014, CEI-25G-LR 25.78125 Compliant1
OTU4 (OTL4.4) CFP2 OIF-CEI-28G-VSR 27.952493–32.75 Compliant
OTU4 (OTL4.4) CFP OIF-CEI-11G-MR 11.18–13.1 Compliant
CAUI-10 IEEE 802.3-2012 10.3125 Compliant
nPPI IEEE 802.3-2012 10.3125 Compliant
10GBASE-KR2 IEEE 802.3-2012 10.3125 Compliant
SFP+ SFF-8431 (SR and LR) 9.95328–11.10 Compliant
XFP INF-8077i, revision 4.5 10.3125 Compliant
RXAUI CEI-6G-SR 6.25 Compliant
XAUI IEEE 802.3-2012 3.125 Compliant
1000BASE-X IEEE 802.3-2012 1.25 Compliant
Electrical
Protocol Specification Serial Rate (Gb/s)
Compliance
5.0G Ethernet IEEE 802.3bx (PAR) 5 Compliant
2.5G Ethernet IEEE 802.3bx (PAR) 2.5 Compliant
HiGig, HiGig+, HiGig2 IEEE 802.3-2012 3.74, 6.6 Compliant
QSGMII QSGMII v1.2 (Cisco System, ENG-46158) 5 Compliant
OTU2 ITU G.8251 10.709225 Compliant
OTU4 (OTL4.10) OIF-CEI-11G-SR 11.180997 Compliant
OC-3/12/48/192 GR-253-CORE 0.1555–9.956 Compliant
PCIe Gen1, 2, 3 PCI Express base 3.0 2.5, 5.0, and 8.0 Compliant
SDI3 SMPTE 424M-2006 0.27–2.97 Compliant
UHD-SDI3 SMPTE ST-2081 6G, SMPTE ST-2082 12G 6 and 12 Compliant
Hybrid memory cube (HMC) HMC-15G-SR 10, 12.5, and 15.0 Compliant
MoSys bandwidth engine CEI-11-SR and CEI-11-SR (overclocked) 10.3125, 15.5 Compliant
CPRI CPRI_v_6_1_2014-07-01 0.6144–12.165 Compliant
Passive optical network (PON) 10G-EPON, 1G-EPON, NG-PON2, XG-PON, and 2.5G-PON 0.155–10.3125 Compliant
JESD204a/b OIF-CEI-6G, OIF-CEI-11G 3.125–12.5 Compliant
Serial RapidIO RapidIO specification 3.1 1.25–10.3125 Compliant
DisplayPort DP 1.2B CTS 1.62–5.4 Compliant3
Fibre channel FC-PI-4 1.0625–14.025 Compliant
SATA Gen1, 2, 3 Serial ATA revision 3.0 specification 1.5, 3.0, and 6.0 Compliant
SAS Gen1, 2, 3 T10/BSR INCITS 519 3.0, 6.0, and 12.0 Compliant
SFI-5 OIF-SFI5-01.0 0.625 - 12.5 Compliant
Aurora CEI-6G, CEI-11G-LR All rates Compliant
Notes:
1. 25 dB loss at Nyquist without FEC.
2. The transition time of the transmitter is faster than the IEEE Std 802.3-2012 specification.
3. This protocol requires external circuitry to achieve compliance.
Table 120: Maximum Performance for Interlaken 12 x 12.5 Gb/s Protocol and Lane Logic Mode
Designs
Table 121: Maximum Performance for Interlaken 6 x 25.78125 Gb/s and 6 x 28.21 Gb/s Protocol and
Lane Logic Mode Designs
Table 122: Maximum Performance for Interlaken 12 x 25.78125 Gb/s Lane Logic Only Mode Designs
Table 125: Maximum Performance for PCIE4C-based PCI Express and CCIX Designs for ZU3T Devices
TPS_PROG_B PL program pulse width 250 250 250 250 250 ns, Min
Internal Configuration Access Port
FICAPCK Internal configuration access port (ICAPE3) 200 200 200 150 150 MHz, Max
DNA Port Switching
FDNACK DNA port frequency (DNA_PORT) 200 200 200 175 175 MHz, Max
STARTUPE3 Ports
FCFGMCLK STARTUPE3 CFGMCLK output frequency 50.00 50.00 50.00 50.00 50.00 MHz, Typ
FCFGMCLKTOL STARTUPE3 CFGMCLK output frequency ±15 ±15 ±15 ±15 ±15 %, Max
tolerance
TDCI_MATCH Specifies a stall in the startup cycle until the 4 4 4 4 4 ms, Max
digitally controlled impedance (DCI) match
signals are asserted
Notes:
1. The TPOR specification begins when the last of the monitored supplies (VCCINT, VCCAUX, VCCBRAM) reaches 95% of its recommended
operating condition voltage.
2. The POR override (POR_OVERRIDE pin tied to VCCINT) is applicable only when the monitored supplies ramp within the specified time.
Revision History
Date Version Description of Revisions
05/16/2023 1.24 Revised ICCOMIN for XCZU3T from 96 mA to 50 mA in Table 10. Updated Table 25 to Vivado Design Suite
2023.1 v1.03 for the XCZU3TCG and XCZU3TEG devices.
Added note about Quad-SPI feedback clock MIO[6] pin to Table 41 and Table 42.
02/28/2023 1.23 Added the XCZU3T device and SFVD784 package throughout the data sheet. Updated Table 25 to Vivado
Design Suite 2022.2.2. Added note about PCIe operating modes to Table 107. Updated introductory
paragraph in Integrated Interface Block for PCI Express Designs. Added Table 125.
12/01/2022 1.22 Removed mention of I/Os being tristated at power-on from PS Power-On/Off Power Supply Sequencing.
Replaced "die" with "rank" for LPDDR4 DRAM type in Table 30.
06/14/2022 1.21 Added the XAZU1 device throughout the data sheet in the SBVA484, SFVA625, and SFVC784 packages.
Updated Table 25 to Vivado Design Suite 2022.1.
In Note 1 of Table 37, clarified that TPOR is TPOR,MAX.
AUTOMOTIVE PRODUCTS (IDENTIFIED AS "XA" IN THE PART NUMBER) ARE NOT WARRANTED FOR USE
IN THE DEPLOYMENT OF AIRBAGS OR FOR USE IN APPLICATIONS THAT AFFECT CONTROL OF A
VEHICLE ("SAFETY APPLICATION") UNLESS THERE IS A SAFETY CONCEPT OR REDUNDANCY FEATURE
CONSISTENT WITH THE ISO 26262 AUTOMOTIVE SAFETY STANDARD ("SAFETY DESIGN"). CUSTOMER
SHALL, PRIOR TO USING OR DISTRIBUTING ANY SYSTEMS THAT INCORPORATE PRODUCTS,
THOROUGHLY TEST SUCH SYSTEMS FOR SAFETY PURPOSES. USE OF PRODUCTS IN A SAFETY
APPLICATION WITHOUT A SAFETY DESIGN IS FULLY AT THE RISK OF CUSTOMER, SUBJECT ONLY TO
APPLICABLE LAWS AND REGULATIONS GOVERNING LIMITATIONS ON PRODUCT LIABILITY.
Copyright
© Copyright 2015–2023 Advanced Micro Devices, Inc. AMD, the AMD Arrow logo, Vivado, Zynq, and
combinations thereof are trademarks of Advanced Micro Devices, Inc. AMBA, AMBA Designer, Arm,
ARM1176JZ-S, CoreSight, Cortex, PrimeCell, Mali, and MPCore are trademarks of Arm Limited in the US and/or
elsewhere. PCI, PCIe, and PCI Express are trademarks of PCI-SIG and used under license. The DisplayPort Icon
is a trademark of the Video Electronics Standards Association, registered in the U.S. and other countries. Other
product names used in this publication are for identification purposes only and may be trademarks of their
respective companies.