Electricity: measuring and testing – Fault detecting in electric circuits and of electric components – For fault location
Reexamination Certificate
2001-06-29
2002-12-31
Nghiem, Michael (Department: 2861)
Electricity: measuring and testing
Fault detecting in electric circuits and of electric components
For fault location
Reexamination Certificate
active
06501278
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to testing and characterizing circuit boards, as well as the components mounted thereon. More particularly, the present invention relates to apparatus and methods which are used to characterize and test the performance of printed circuit boards and various components using a differential signal interface at frequencies above about one Gigahertz.
BACKGROUND INFORMATION
As circuit design clock speeds increase, and data rates approach one billion transfers per second, the radio frequency properties of each signal as it crosses a circuit board become all-important. Because high-speed signals traveling along a transmission line tend to reflect energy backward (toward their source) as they encounter a change in impedance, and the amount of reflected energy depends on the magnitude of the impedance change, it is often the reflected energy due to impedance changes that makes the difference between a circuit that works and one that does not.
Within conventional dynamic RAM circuits, a single data bit or pulse typically travels down a bus transmission line alone, and reaches its destination, such that any reflections caused by an impedance mismatch dissipate before the next bit is launched. In higher data rate environments, this is not necessarily the case.
For example, in some recent data bus circuitry designs, data is transported so rapidly that up to three bits of information may be in transit between a source and the destination along a particular circuit trace (such as between a memory controller and the memory it operates) at a given instant in time. At such speeds, the second and third bits must pass through any reflections generated by the first bit due to an impedance mismatch, such as that which may occur at the junction where the bit enters a connector. Such reflections degrade signal margins and cause timing errors, leading to data transmission failures. Therefore, in these high-speed signal propagation situations, it is absolutely essential to minimize impedance variations encountered within the circuit board and data processing circuitry.
One way of verifying the impedance characteristics of a particular component is to observe signal reflections from the component when high speed test signals are launched onto the associated circuit board. Test “coupons” (sample test areas) are therefore incorporated into circuit board designs to facilitate connection to impedance measurement tools. Such coupons are used to verify behavior of the printed circuit board (PCB) itself, as well as various attached circuit packages, sockets, and connectors.
The industry-standard tool used for circuit board impedance measurements is the Time Domain Reflectometer (TDR). TDR instruments make use of fast system risetimes and a high bandwidth to resolve features of circuit boards and their attached circuitry. A TDR typically takes measurements by using a probe to send a known pulse down the transmission medium (in this case a circuit board trace or differential pair of traces) and capturing the reflections that result. The heart of the TDR is an extremely high-bandwidth (20 GHz) sampling oscilloscope and fast-risetime sampling head with an integral step generator.
Therefore, in setting up for actual circuit board impedance measurements, the TDR probing approach is of great concern. Impedance matching is critical. In dense high-speed circuit areas, the space available to firmly attach a TDR probe is very limited. Yet probe connection integrity is essential. Any compromise in these areas will inevitably appear as an inaccurate TDR reading.
SMA connectors are sometimes used to couple TDR equipment to the circuit board for testing. However, this type of connector installation is usually less than satisfactory, since it contributes to an impedance mismatch between the probe and the PCB, and usually requires about one square inch of board surface area for attachment. Thus, the use of SMA connectors is often not an option for the designer.
Specially-designed microprobes are also available for TDR measurement applications. However, the conventional microprobe contact connection also contributes to impedance mismatch at high frequencies, such that microprobes are also less than ideal as a signal coupling device for test signals above about one Gigahertz.
Another difficulty involves launching differential test signals into a particular circuit or board design. With differential signals, two conductors are required: the first carries a true signal value, while the second carries the inverted signal value. Ideally, the differential system leads to complete cancellation of emitted noise. The receiver then rejects common-mode noise, and the result is a substantial improvement in both signaling speed and reduced emissions due to electromagnetic interference. High-speed designs tend to incorporate differential signaling, rather than single-ended methods, because higher data rates are possible, susceptibility to EMI is less, and power consumption is also typically less. However the problems encountered when testing high-speed differential interfaces are in effect doubled with regard to impedance mismatch problems and connector surface area requirements. While some approaches have been tested for launching single-ended, high-speed signals into circuit boards for testing purposes, no uniformly satisfactory apparatus or method is known for the convenient application of high-speed differential signals necessary to properly characterize high-speed differential interface circuitry.
Thus, there is a need in the art for test structure apparatus and methods which can be easily applied to launch differential test signals into circuit boards and components for high-speed signal performance characterization purposes. Such apparatus and methods should provide an optimal impedance match to minimize reflections, while using a minimum of circuit board real estate for connection to the circuitry under test.
REFERENCES:
patent: 5012213 (1991-04-01), Tsai
patent: 5219827 (1993-06-01), Higaki et al.
Winslow, T.A., “Component Modeling for PCB Design”,Microwave, Application Notes,61-63, (Mar. 2000).
Intel Corporation
Nghiem Michael
Schwegman Lundberg Woessner & Kluth P.A.
LandOfFree
Test structure apparatus and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Test structure apparatus and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Test structure apparatus and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2998269