Electrical Validation For DDR Io Interface: Internship Report
Electrical Validation For DDR Io Interface: Internship Report
Electrical Validation For DDR Io Interface: Internship Report
submitted by
4 Signal Integrity 28
4.1 Typical Silicon Problem . . . . . . . . . . . . . . . . . . . . . . . . 28
4.1.1 Crosstalk . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
1
4.1.2 Frequency Dependent Loss . . . . . . . . . . . . . . . . . . 30
4.1.3 Need of proper Termination . . . . . . . . . . . . . . . . . 32
4.1.4 Reflections . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.1.5 Jitter in High Speed Serial Link . . . . . . . . . . . . . . . 33
5 JEDEC Specifications 34
5.1 Setup and Hold Times: . . . . . . . . . . . . . . . . . . . . . . . . 35
5.1.1 DQ/DQS: . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.1.2 Clk/Cntl: . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.1.3 Clk/Cmd: . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.2 Clock Jitter: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.3 Clock Vix: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.4 tDQSS (DQS latching rising transitions to associated clock edges) : 36
7 Results 40
7.1 DLL Characterization . . . . . . . . . . . . . . . . . . . . . . . . . 40
7.1.1 Delay range and Step Size . . . . . . . . . . . . . . . . . . 40
7.1.2 Code Vs. Delay . . . . . . . . . . . . . . . . . . . . . . . . 41
7.1.3 Integral Nonlinearity . . . . . . . . . . . . . . . . . . . . . 42
7.1.4 Differential Non Linearity. . . . . . . . . . . . . . . . . . . 42
7.2 Signal Integrity Validation . . . . . . . . . . . . . . . . . . . . . . 43
7.2.1 DQ-DQS setup and hold time measurement . . . . . . . . . 43
7.2.2 Clock-Command setup and hold time measurement . . . . . 45
7.2.3 Clock-Control setup and hold time measurement . . . . . . 47
7.2.4 TDQSS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
7.2.5 Clock Vix . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
7.2.6 Jitter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
2
List of Figures
3
7.14 Clock Jitter at 800MHz. . . . . . . . . . . . . . . . . . . . . . . . . 52
7.15 Clock Jitter at 1066MHz. . . . . . . . . . . . . . . . . . . . . . . . 52
4
SYNOPSIS
The latest trend in mobile computing and media computing platforms is to integrate
chipsets and micro processor in to a single chip solution or System-on-chip (SOC).
Increasing the performance of memory interfaces with minimal power is one of the
major challenges faced by chipset designs. As a result of this new DDR memory
standards like DDR2 and DDR3 evolved. This revolution leads to complex design
which imposes severe validation requirements, pre-silicon validation and postsili-
con validation. But complete system level verification of a complex SoC at 45nm
or below is not feasible in pre-silicon, in system at-speed post-silicon validation has
become an essential step in design in verification and optimization of the design
before the product is shipped to the customer. Using Bench setup with high-speed
oscilloscope and advanced probing techniques, the signal are captured and signal
integrity timing and quality checks are done. Debug is done by the appropriate soft-
ware tools and scripts specified by Intel.
Ten years ago a typical Intel chipset had about 50 thousand gates. Today, these core
platform components have as many as 4 million gates with increase in Intel desktop
processors transistor numbers from about 3 million to over 40 million As the pro-
cessor complexity has grown, the bugs found in Post-Silicon have increased. The
complexity of creating the conditions to cause the bugs to appear and the complexity
of debugging them have increased as well.
The project dealt with Post-Silicon validation by devising new ways of acquiring
the required signals under stress conditions across Process, Voltage and Tempera-
ture usually indicate as PVT corners. The complexity to acquire the required signals
increases with advancements in the process technology. The latest 45nm (nanome-
ter) technologies posses great challenge to the Post-Silicon validation team as the
new generation processors and chipsets have more functions, operate at faster fre-
quencies and must be compatible with far more hardware devices and software com-
ponents than ever before with even reduced number of observation pins.
5
Internship Details
Team: HIP-EV-DDR
This team works on Electrical Validation of DDR Interface.
Role:
Masters Intern in HIP-EV-DDR team.
Responsibilities:
• To understand SOC architecture of mobile computer platforms.
6
• Worked on various debug activities
• Automated procedures
7
Company Profile
About Intel
Intel pushes the boundaries of innovation so our work can make people’s lives more
exciting, fulfilling, and manageable. And our work never stops. We at Intel, never
stop looking for the next leap ahead in technology, education, culture, manufactur-
ing, and social responsibility. And we never stop striving to deliver solutions with
greater benefits for everyone. - INTEL
For past 40 years, Intel Corporation has developed technology enabling the com-
puter and Internet revolution that has changed the world. Founded in 1968 by
Robert Noyce and Gordon Moore to build semiconductor memory products, In-
tel introduced the world’s first microprocessor in 1971. Initial mission was to de-
sign and manufacture complex integrated circuits or silicon chips. The company
8
was first called NM Electronics and later became INTEL (for INTegrated ELectron-
ics). Today, Intel supplies the computing and communications industries with chips,
boards, systems, and software building blocks that are the ”ingredients” of comput-
ers, servers and networking and communications products. These products are used
by industry members to create advanced computing and communications systems.
Intel’s mission is to do a great job for our customers, employees, and stockholders
by being the pre-eminent building block supplier to the worldwide digital economy.
Intel, the world’s largest chip maker, is also a leading manufacturer of computer, net-
working, and communications products. Intel’s Web site serves geographies around
the globe localized content including product information, solutions and strategies,
programs, and support. Main areas of operation include Software and Hardware
Design and Development, Sales and Marketing, Venture Capital (a strategic in-
vestment initiative to support related technology enterprise) and Intel Innovation
in Education. Today, the Bangalore center has the distinction of housing a number
of mission critical projects for the company. Design of the next generation 32-bit
server and Intels 100% e-corporation initiative support are especially significant in
this regard. The design center, which houses the cream of the companys knowledge
workers, boasts of the highest software quality within Intel and has more than 250
invention disclosures and 335 patent filings (2004) to its credit. Donation, Spon-
sorship, and Grant Information, Intel receives many requests to help support very
worthy and important causes. In order to focus our giving and maximize the results
achieved, Intel established a set of criteria to guide donations, grants, and sponsor-
ship commitments. These criteria generally emphasize math and science education,
and technology literacy programs. Intel is also involved in community volunteering
program that coordinates a variety of employee volunteer programs in support of
education, environmental stewardship and safety, youth development, and commu-
nity service. Its goal is to be a great asset to the community; by being a responsible
corporate citizen. Some of the programs include blood donation camps, eco-watch
for improving the environment in our communities through energy conservation and
biodiversity projects, projects to build awareness in health, hygiene, nutrition, life
skills among underprivileged people and many more such activities.
• Architecture and Silicon Technology
In the age of Moore’s Law, Intel has delivered architecture and silicon tech-
nology with amazing transistor counts as many as two billion and growing.
With steady gains in energy efficient performance, and innovative uses of new
materials, our innovations continue to enable industry leading firsts.
• Micro-architectures
By continuously introducing new micro-architectures, Intel is creating great
leaps in energy efficient performance that enable new form factors at home,
in the office, and on the go.
9
• Silicon Technology
Transforming the way we experience computing at the silicon level, Intel is
developing new technologies that use innovative materials for ever smaller,
faster, and more energy efficient processors and processor technology.
• Multi-core Technology
Developing new levels of system intelligence, energy efficiency, and process-
ing performance, Intel multi-core technology enables improved computing
experiences while paving the way for the evolution of Tera-scale computing.
• Product Technologies
Intel’s leading edge notebook and desktop products range from processor to
processor technologies that are designed to work together to deliver a great
computing experience for home, business, and on the go. Discover your next
great technology experience with Intel inside.
• Manufacturing
Operating 24/7 in plants around the world, Intel’s manufacturing processes are
precision tuned to perform with maximum efficiency and quality to produce
fast, smart, and more energy efficient technologies.
• Fab
With 15 wafer fabrication plants (fabs) worldwide, Intel’s manufacturing pro-
10
cesses employ exceptional flexibility on a global and virtual network. By de-
sign, these facilities are consistently sharing information to improve product
performance while further fine tuning the manufacturing process.
11
Chapter 1
12
sign and continues throughout as Pre-Silicon and Post-Silicon validation. With any
semiconductor device, Pre-Silicon design verification and Post-Silicon validation
are critical to the functional and operational quality of the finished product. This sit-
uation holds especially true for SOC (System-On-Chip) devices, in which a single
defect in any one function could necessitate a re spin. The term bug has been used to
describe design problems. There are three categories of bugs: functional, electrical,
and manufacturing/yield.
• Functional bugs, or logic bugs, are where the logic of a design is improperly
implemented.
• Electrical bugs are where the functional behavior of the circuit is correct, but
it fails to operate correctly in some part of the specified operating region (i.e.
at some voltage, temperature or frequency).
Manufacturing/yield bugs occur when a circuit is sensitive to some variable in the
manufacturing process and changes its operation as a result of manufacturing varia-
tions.
13
System Validation (SV) Methodology
The purpose of the System validation is to perform
• Chipset logic functionality validation.
• Validate and stress Chipset specific features using synthetic testing
validation board custom designed for SV use that include debug hooks for all inter-
faces. In addition it uses
• Special test cards designed specifically for SV tests
• AGP, PCI, USB, LPC, LAN, AC97
• Internally developed SV software running under DOS
• Memory Stress tests
• The two key benefits of this approach are:
• Controllability of tests and test coverage.
• Lack of dependency on internal or external drivers for O/S specific support.
14
1.2.3 Post-Silicon Validation Challenges
The challenges faced in Post-Silicon validation are:
• Signal Integrity.
• Design Complexity.
• Involve system level integration where some components are not ready at Pre-
Silicon stage, for example system level stress test on the device where the test
involves multiple peripheral transactions which increase over a time.
• Combination of above, for example, running system level stress test on the
device at different temperature variation and verify if there is any flash data
retention issue observed.
One of the initial stages of validation is actual mechanical probing, where a very
small mechanical probe is used to measure a signal on the device. This is a very
time consuming and error-prone activity due to the incredibly small dimensions in-
volved in todays devices. It is possible to determine when voltage transitions occur
on a given node and provide oscilloscope traces of individual signals relative to each
other which are extremely useful in debugging timing-related failures. This is useful
to allow debug of either functional or electrical failures through the reconstruction
of signal waveforms on a device for further comparison to simulation and data col-
lected through scan and clock manipulation.
15
Chapter 2
Double data rate (DDR) or double pumped, dual-pumped, and double transition is
used in applications where fast data transmission is needed, such as memory access
and first-in first-out (FIFO) memory structures. DDR stands for Dual Data Rate.
Dual data rate means data is sampled at both the rising and falling edge of the clock.
Its an interface that connects between MCH and system memory. In mobile it started
off from DDR, gave birth to DDR2 and now the latest DDR3. DDR uses both edges
of a clock to transmit data, which facilitates data transmission at twice the rate of
a Single Data Rate (SDR) architecture using the same clock speed by transferring
data on both rising and falling edges of the clock signal This method also reduces
the number of I/O pins required to transmit data.
16
layer between the Data and the centralised System Clock (CLK).
DDR1s Single-ended Data Strobe refers to the characteristic of having only a
single rising and falling signal wave. This single-ended Data Strobe is tracked by the
Data (DQ) for more effective signalling it has greater fundamental strength in toler-
ating Process-Voltage-Temperature (PVT) variation and issues caused by crosstalks,
echoes and signal reflection, which tend to distort the waveform or induce timing
inaccuracy.
17
Figure 2.2: DDR2 clocking Routing.
18
Figure 2.3: DDR3 clocking Routing.
• DQS/DQS# [7:0] :- Differential strobes required to latch the data. Each strobe
validate 7 data lines. Data setup and hold is wrt strobe.
• CKE :- Clock enable for enabling/disabling memory clock, mainly used dur-
ing power down mode or the self refresh mode.
• CS :- Chip select signal. These are used for selecting the rank of the DIMM.
19
• ODT :- On Die Termination. Used to improve signal integrity.( overshoot
undershoot etc). ODT is asserted for DQ, DM, DQS.
• A[15:0] :- Address lines. Row and Column address lines are multiplexed.
• WE# :- Write enable asserted by the controller during memory write cycle.
20
Chapter 3
The Delay Locked Loop (DLL) is a feed back control system like Phase Locked
Loop (PLL) and is employed in high speed phase alignment circuits such as mi-
croprocessor and SDRAMs i.e., memory ICs in order to cancel the on-chip clock
amplication and buffering delays and improve the I/O timing margins. The supply
and substrate noise resulting from the switching of digital circuits affects the PLL
or DLL operation and results in output clock jitter which subtracts from the I/O tim-
ing margins. With shrinking margins for jitter, DLLs are becoming substitutes for
PLLs based on Voltage Controlled Oscillators (VCO) where no clock synthesis is
required.
21
Figure 3.1: Phase Relationship between DQ-DQS.
22
Figure 3.3: DLL Block Diagram.
tree), with the phase of the clock signal (CLK IN) just inside the clock input pad as
shown in figure /refdllic.
The clock signal from the clock input pad is usually the most convenient clock
for latching data signals just inside the data input pads. Without active synchroniza-
tion, the phase of the two clocks will likely be different due to the uncertain and IC
specific propagation delay through the clock tree buffer network. The role of the
DLL is to adjust the output tap of a programmable digital delay line at the root of
the clock tree so that the phase at the leaf nodes of within ASIC are synchronized at
all the flip-flip inputs.
23
3.2 DLL Functional Blocks
DLL block diagram is shown in figure /refbd.
• An RC low-pass filter
24
• A charge-pump and a capacitor
PI design is a dual input differential buffer which uses the same symmetric loads
as all the core. The current sources of the two differential pairs are thermometer
controlled elements. The thermometer codes are generated by a X bit long up/down
shift register which is controlled by the peripheral loop FSM. By changing the ther-
mometer code, the FSM adjusts in a complementary fashion the currents of the two
input differential pairs resulting in a mixing of the two input clock phases.
25
3.3 Characterization of Delay Locked Loop on Test
chip
3.3.1 Delay per PI code
To find delay per PI, consider the below calculation The code given to Finite State
Machine (FSM) when incremented by one produces a delay of (1/8)* 45 degrees,
in other words it produces T/64 seconds of delay per PI where T is the Time period
of DIMM clock. For 800 MHz DIMM frequency, time period (T) is (1/800) which
equals 1.25 ms Thus the delay per PI code should be 1.25 ms / 64 equal to 19.53
ps (pico seconds).That is each code increment delays the DIMM clock by 19.53
pico seconds As per above case the output clock will be delayed by 19.53 pico
seconds with respect to input reference clock (RefCLK). Similarly for 1066 and
1333 MHz the delay per PI code equals 14.65 ps and 11.7 ps respectively. During the
DLL validation process, the DLLs are also checked for per code delay consistency.
Because of the earlier mentioned capacitive capacitance of the differential pair input
transistors in the Phase Interpolators, delay produced in non linear with code.
26
one code apart. Differential non-linearity is a measure of the worst case deviation
from the ideal one code step. The INL and DNL deviations from ideal case are also
computed for all modules across all PVT corners.
27
Chapter 4
Signal Integrity
The term Signal Integrity (SI) addresses two concerns in the electrical design aspects
the timing and the quality of the signal. Does the signal reach its destination when
it is supposed to? And also, when it gets there, is it in good condition? The goal
of signal integrity analysis is to ensure reliable high-speed data transmission. In a
digital system, a signal is transmitted from one component to another in the form
of logic 1 or 0, which is actually at certain reference voltage levels. At the input
gate of a receiver, voltage above the reference value Vih is considered as logic high,
while voltage below the reference value Vil is considered as logic low shows the
ideal voltage waveform in the perfect logic world, whereas figure 4.2 shows how
signal will look like in a real system.
More complex data, composed of a string of bit 1 and 0s, are actually continuous
voltage waveforms. The receiving component needs to sample the waveform in
order to obtain the binary encoded information. The data sampling process is usually
triggered by the rising edge or the falling edge of a clock signal. It is clear from the
diagram that the data must arrive at the receiving gate on time and settle down to
a non-ambiguous logic state when the receiving component starts to latch in. Any
delay of the data or distortion of the data waveform will result in a failure of the
data transmission. Imagine if the signal waveform in Figure 4.2 exhibits excessive
ringing into the logic gray zone while the sampling occurs, then the logic level
cannot be reliably detected.
28
Figure 4.1: Ideal and Real waveforms at receiving end.
problems:
• Reflection Noise
Due to impedance mismatch, stubs, vias and other interconnect discontinu-
ities.
• Crosstalk Noise
Due to electromagnetic coupling between signal traces and vias.
• Power/Ground Noise
Due to parasitics of the power/ground delivery system during drivers simul-
taneous switching output (SSO). It is sometimes also called Ground Bounce,
Delta-I Noise or Simultaneous Switching Noise (SSN).
Besides these three kinds of SI problems, there are other Electromagnetic Com-
patibility or Electromagnetic Interference (EMC/EMI) problems that may contribute
29
to the signal waveform distortions. When SI problems happen and the system noise
margin requirements are not satisfied the input to a switching receiver makes an
inflection below Vih minimum or above Vil maximum; the input to a quiet receiver
rises above Vil maximum or falls below Vih minimum; power/ground voltage fluc-
tuations disturb the data in the latch, then logic error, data drop, false switching, or
even system failure may occur. These types of noise faults are extremely difficult to
diagnose and solve after the system is built or prototyped. Understanding and solv-
ing these problems before they occur will eliminate having to deal with them further
into the project cycle, and will in turn cut down the development cycle and reduce
the cost. In the later part of this chapter, we will have further investigations on the
physical behavior of these noise phenomena, their causes, their electrical models for
analysis and simulation, and the ways to avoid them.
4.1.1 Crosstalk
Crosstalk is the coupling of energy from one line to another via:
• The mutual inductance will induce current on the victim line opposite of the
driving current (Lenzs Law).
• The mutual capacitance will pass current through the mutual capacitance that
flows in both directions on the victim line.
30
Figure 4.3: Graph for Cross talk.
Copper Loss:
Copper has resistive loss as does any conductor. At high frequencies, the internal
inductance of conductors pushes the current to the outer surfaces. This effect is
called skin effect.
This phenomenon decreases the effective area available for current flow and so
increases the effective resistance. It is as is there is only a thin layer on the surface
of the conductor that is involved in high-frequency current flow, and the thickness
of this layer is called skin depth.
Dielectric Loss:
In dielectric the relative dielectric constant is thought to be due to such things as
the physical distortion of molecules, the reorientation of molecules, the changing
31
of the scope of electron orbits, etc. each case has a stimulus and response. When
calculating response of the system to fields through dielectric, the imaginary part of
the dielectric constant shows up as a loss.
4.1.4 Reflections
The mismatches between transmission line segments are caused by manufacturing
tolerances, connectors, vias and other factors. Reflections adversely affect signal
integrity in two ways:
• Wave reflections cause less energy to propagate forward to the receiver.
• Wave reflections can cause the equivalent of multi-path distortion, whereby
the reflected signal arrives at the receiver some time later than the main signal.
Sources of Reflections: Z - Discontinuities
• PCB Zs
• Connector Zs
• Vias (through) Zs
• Package Zs
• Termination Zs
The best way to control wave reflections is to minimize impedance discontinu-
ities between the driver and receiver.
32
Figure 4.6: TDR impedance profile.
33
Chapter 5
JEDEC Specifications
JEDEC Solid State Technology Association, formerly known as Joint Electron De-
vice Engineering Council (JEDEC) or Joint Electron Device Engineering Councils,
is the semiconductor engineering standardization body of the Electronic Industries
Alliance (EIA), a trade association that represents all areas of the electronics in-
dustry in the United States. JEDEC is the leading developer of standards for the
solid-state industry. Almost 3,300 participants, appointed by some 295 companies
work together in 50 JEDEC committees to meet the needs of every segment of the
industry, manufacturers and consumers alike. The publications and standards that
they generate are accepted throughout the world. All JEDEC standards are avail-
able online, at no charge. JEDEC was founded in 1958 as a joint activity between
EIA and NEMA to develop standards for semiconductor devices. This early work
began as a part numbering system for devices which became quite popular in the
’60s. For example, the 1N4001 rectifier diode and 2N2222 transistor part numbers
came from JEDEC. These part numbers are still popular today. JEDEC later devel-
oped a numbering system for integrated circuits, but this did not gain acceptance
in the semiconductor industry. JEDEC’s adoption of open industry standards (i.e.,
standards that permit any and all interested companies to freely manufacture in com-
pliance with adopted standards) serves several vital functions for the advancement
of electronic technologies. First and foremost, such standards allow for interoper-
ability between different electrical components. JEDEC standards do not protect
members from normal patent obligations. The designated representatives of JEDEC
member companies are required to disclose patents and patent applications of which
they personally are aware (assuming that this information is not considered propri-
etary). JEDEC patent policy requires that standards found to contain patents, whose
owners will not sign a standard JEDEC patent letter, be withdrawn. Thus the penalty
for a failure to disclose patents is retraction of the standard. Typically, standards will
not be adopted to cover technology that will be subject to patent protection. In rare
circumstances, standards covered by a patent may be adopted, but only on the under-
34
standing that the patent owner will not enforce such patent rights or, at a minimum,
that the patent owner will provide a reasonable and non-discriminatory license to
the patented technology.
5.1.1 DQ/DQS:
5.1.2 Clk/Cntl:
5.1.3 Clk/Cmd:
35
5.2 Clock Jitter:
Min Max
DDR2 0.5xVDD - 0.175 0.5xVDD + 0.175
DDR3 0.5xVDD - 0.175 0.5xVDD + 0.175
Min Max
DDR2-667 -0.25 times of tCK (avg) 0.25 times of tCK (avg)
DDR2-800 -0.25 times of tCK (avg) 0.25 times of tCK (avg)
DDR3-800 -0.25 times of tCK (avg) 0.25 times of tCK (avg)
DDR3-1066 -0.25 times of tCK (avg) 0.25 times of tCK (avg)
36
Chapter 6
6.3 Probes:
Differential probes are required to measure the Differential signals and single ended
signals as well. These probes should be chosen according to the BW requirement.
37
6.4 Software Tools
6.4.1 Jitter Analysis Application Tool
The Jitter measurement application consists of two products: Jitter Analysis Ad-
vanced and Jitter Analysis Essentials. These products are applications that enhance
basic capabilities of some Windows-based oscilloscopes from Agilent. These jitter
analysis applications include the following features:
• Select and configure multiple measurements on more than one waveform
• Display statistical results for up to six measurements
• Perform random and deterministic jitter analysis including
• Show results as plots
• Save statistical results to a data log file
• Save individual data points to a measurement results file
• Save the worst case waveforms to .wfm files
6.4.3 INFINISCAN
With the help of infiniscan zones on the oscilloscope can be defined such that the
signal waveforms crossing that particular zone can be completely eliminated from
the screen. This helps in generating write eye, read signals from the screen can be
eliminated so that their appearance should not disturb write eye.
38
6.4.5 Internal Post processing tool
This is the tool that is used to capture waveforms from oscilloscope for use on other
post processing applications. Featuring an easy to use interface, this makes it simple
to transfer waveform data from Tektronix or Agilent oscilloscope to PC. This will
take through the steps of capturing first set of data. The first step involves getting
system and oscilloscope to talk to each other. The second step involves tuning the
signals display on the oscilloscope to get the maximum wave resolution for tool
to analyze. Finally the third step is to use tool to transfer the waveforms over the
destination.
39
Chapter 7
Results
40
7.1.2 Code Vs. Delay
For this Code Vs. Delay Characterization, code is varied from 0 to 63 and the
corresponding delay is recorded from the scope. The above graph shows one of the
finding.
41
7.1.3 Integral Nonlinearity
42
7.2 Signal Integrity Validation
7.2.1 DQ-DQS setup and hold time measurement
43
Figure 7.5: Setup and Hold Time of DQ-DQS at 1066 MHz.
44
7.2.2 Clock-Command setup and hold time measurement
45
Figure 7.7: Setup and Hold Time of Clock-Command at 1066 MHz.
46
7.2.3 Clock-Control setup and hold time measurement
47
Figure 7.9: Setup and Hold Time of Clock-Control at 1066 MHz.
48
7.2.4 TDQSS
49
Figure 7.11: Tdqss(Clock-DQS) at 1066 MHz.
According to specification the clk and DQS should be exactly alligned. TDqss
is the delta between Clk and DQS, it should be within the JEDEC specification. For
this measurement Clk and DQS are probed differentially.
50
7.2.5 Clock Vix
This defines the measure of the crossover of the Clock and Clockbar signal(Clock
is a differential signal). For this measurement clock is probed single ended with re-
spect to ground.
51
7.2.6 Jitter
For this measurement the Clock signals are probed differentialy, and the inbuilt
jitter tool from the Agilent is used.
52
Bibliography
53