Design for testing: Difference between revisions

Revision as of 14:28, 4 March 2008

Latest revision as of 17:07, 23 February 2025

(64 intermediate revisions by 46 users not shown)

Line 1:

{{Short description|IC design techniques that include testability features}}

'''Design for Test''' (aka "Design for Testability" or "DFT") is a name for [[Integrated circuit design|design]] techniques that add certain testability features to a [[integrated circuit|microelectronic]] hardware product design. The premise of the added features is that they make it easier to develop and apply manufacturing tests for the designed hardware. The purpose of manufacturing tests is to validate that the product hardware contains no defects that could, otherwise, adversely affect the product’s correct functioning.

'''Design for Testtesting''' (akaor "Design'''design for Testability"testability''' or "('''DFT"''') isconsists a name forof [[Integratedintegrated circuit design|design]] techniques that add certain testability features to a [[integrated circuit|microelectronic]] hardware product design. The premise of the added features is that they make it easier to develop and apply manufacturing tests forto the designed hardware. The purpose of manufacturing tests is to validate that the product hardware contains no manufacturing defects that could, otherwise, adversely affect the product’sproduct's correct functioning.

Tests are applied at several steps in the [[Semiconductor fabrication|hardware manufacturing]] flow and, for certain products, may also be used for hardware maintenance in the customer’s environment. The tests generally are driven by [[Automated testing|test programs]] that execute in [[Automatic test equipment|Automatic Test Equipment]] (ATE) or, in the case of system maintenance, inside the assembled system itself. In addition to finding and indicating the presence of defects (i.e., the test fails), tests may be able to log [[diagnostic]] information about the nature of the encountered test fails. The diagnostic information can be used to locate the source of the failure.

Tests are applied at several steps in the [[Semiconductor fabrication|hardware manufacturing]] flow and, for certain products, may also be used for hardware maintenance in the customer’scustomer's environment. The tests generally are generally driven by [[Automated testing|test programs]] that execute inusing [[Automaticautomatic test equipment|Automatic Test Equipment]] (ATE) or, in the case of system maintenance, inside the assembled system itself. In addition to finding and indicating the presence of defects (i.e., the test fails), tests may be able to log [[diagnostic]] information about the nature of the encountered test fails. The diagnostic information can be used to locate the source of the failure.

DFT plays an important role in the development of test programs and as an interface for test application and diagnostics. [[Automatic test pattern generation]], or ATPG, is much easier if appropriate DFT rules and suggestions have been implemented.

In other words, the response of vectors (patterns) from a good circuit is compared with the response of vectors (using the same patterns) from a DUT (device under test). If the response is the same or matches, the circuit is good. Otherwise, the circuit is not manufactured as intended.

DFT plays an important role in the development of test programs and as an interface for test applicationapplications and diagnostics. [[Automatic test pattern generation]], or (ATPG,) is much easier if appropriate DFT rules and suggestions have been implemented.

== History ==

DFT techniques have been used at least since the early days of electric/electronic data processing equipment. Early examples from the 1940s/50s are the switches and instruments that allowed an engineer to “scan”"scan" (i.e., selectively probe) the voltage/current at some internal nodes in an [[analog computer]] [analog scan]. DFT often is associated with design modifications that provide improved access to internal circuit elements such that the local internal state can be controlled ([[controllability]]) and/or observed ([[observability]]) more easily. The design modifications can be strictly physical in nature (e.g., adding a physical probe point to a net) and/or add active circuit elements to facilitate controllability/observability (e.g., inserting a [[multiplexer]] into a net). While controllability and observability improvements for internal circuit elements definitely are important for test, they are not the only type of DFT. Other guidelines, for example, deal with the [[Electromechanics|electromechanical]] characteristics of the interface between the product under test and the test equipment. Examples are guidelines for the size, shape, and spacing of probe points, or the suggestion to add a [[Tri-state buffer|high-impedance state]] to drivers attached to probed nets such that the risk of damage from back-driving is mitigated.

Over the years the industry has developed and used a large variety of more or less detailed and more or less formal guidelines for desired and/or mandatory DFT circuit modifications. The common understanding of DFT in the context of [[Electronicelectronic design automation|Electronic Design Automation]] (EDA) for modern microelectronics is shaped to a large extent by the capabilities of commercial DFT software tools as well as by the expertise and experience of a professional community of DFT engineers researching, developing, and using such tools. Much of the related body of DFT knowledge focuses on digital circuits while DFT for analog/mixed-signal circuits takes somewhat of a backseat.

== Objectives of DFT for microelectronics products ==

Line 15 ⟶ 19:

Most tool-supported DFT practiced in the industry today, at least for digital circuits, is predicated on a ''Structural test'' paradigm. Structural test makes no direct attempt to determine if the overall functionality of the circuit is correct. Instead, it tries to make sure that the circuit has been assembled correctly from some low-level building blocks as specified in a structural [[netlist]]. For example, are all specified [[logic gate]]s present, operating correctly, and connected correctly? The stipulation is that if the netlist is correct, and structural testing has confirmed the correct assembly of the circuit elements, then the circuit should be functioning correctly.

Note that this is very different from ''[[Acceptance test|functional testing]]'', which attempts to validate that the circuit under test functions according to its functional specification. This is closely related to the [[functional verification]] problem of determining if the circuit specified by the netlist meets the functional specifications, assuming it is built correctly.

One benefit of the Structural paradigm is that test generation can focus on testing a limited number of relatively simple circuit elements rather than having to deal with an exponentially exploding multiplicity of functional [[State (computer science)|state]]s and state transitions. While the task of testing a single logic gate at a time sounds simple, there is an obstacle to overcome. For today’stoday's highly complex designs, most gates are deeply embedded whereas the test equipment is only connected to the primary [[Input/output]]s (I/Os) and/or some physical test points. The embedded gates, hence, must be manipulated through intervening layers of logic. If the intervening logic contains state elements, then the issue of an exponentially exploding [[state space]] and state transition sequencing creates an [[Computational complexity theory|unsolvable problem]] for test generation. To simplify test generation, DFT addresses the accessibility problem by removing the need for complicated state transition sequences when trying to control and/or observe what’swhat's happening at some internal circuit element.

Depending on the DFT choices made during circuit design/implementation, the generation of Structural tests for complex logic circuits can be more or less [[Automatic test pattern generation|automated]] or self-automated.<ref>{{cite web|url=http://www.eng.tau.ac.il/~bengal/SCI_paper.pdf|title=Self-correcting inspection procedure under inspection errors|author=Ben-Gal I., Herer Y. and Raz T.|publisher=IIE Transactions on Quality and Reliability, 34(6), pp. 529-540.|year=2003|access-date=2014-01-10|archive-date=2013-10-13|archive-url=https://web.archive.org/web/20131013171945/http://www.eng.tau.ac.il/~bengal/SCI_paper.pdf|url-status=dead}}</ref><ref>[http://www.eng.tau.ac.il/~bengal/SCI_paper.pdf] {{Webarchive|url=https://web.archive.org/web/20131013171945/http://www.eng.tau.ac.il/~bengal/SCI_paper.pdf |date=2013-10-13 }}</ref> One key objective of DFT methodologies, hence, is to allow designers to make trade-offs between the amount and type of DFT and the cost/benefit (time, effort, quality) of the test generation task.

Another benefit is to diagnose a circuit in case any problem emerges in the future. It is like adding some features or provisions in the design so that devices can be tested in case of any fault during its use.

== Looking forward ==

One challenge for the industry is keeping up with the [[Moore's law|rapid advances in chip technology]] (I/O count/size/placement/spacing, I/O speed, internal circuit count/speed/power, thermal control, etc.) without being forced to continually upgrade the test equipment. Modern DFT techniques, hence, have to offer options that allow next -generation chips and assemblies to be tested on existing test equipment and/or reduce the requirements/cost for new test equipment. As a result, DFT techniques are continually being updated, such as incorporation of compression, in order to make sure that tester application times stay within certain bounds dictated by the cost target for the products under test.

== Diagnostics ==

Especially for advanced semiconductor technologies, it is expected some of the chips on each manufactured [[Wafer (electronics)|wafer]] contain defects that render them non-functional. The primary objective of testing is to find and separate those non-functional chips from the fully functional ones, meaning that one or more responses captured by the tester from a non-functional chip under test differ from the expected response. The percentage of chips that fail test, hence, should be closely related to the expected functional yield for that chip type. In reality, however, it is not uncommon that all chips of a new chip type arriving at the test floor for the first time fail (so -called zero-yield situation). In that case, the chips have to go through a debug process that tries to identify the reason for the zero-yield situation. In other cases, the test fall-out (percentage of test fails) may be higher than expected/acceptable or fluctuate suddenly. Again, the chips have to be subjected to an analysis process to identify the reason for the excessive test fall-out.

In both cases, vital information about the nature of the underlying problem may be hidden in the way the chips fail during test. To facilitate better analysis, additional fail information beyond a simple pass/fail is collected into a fail log. The fail log typically contains information about when (e.g., tester cycle), where (e.g., at what tester channel), and how (e.g., logic value) the test failed. Diagnostics attempt to derive from the fail log at which logical/physical ___location inside the chip the problem most likely started. By running a large number of failures through the diagnostics process, called volume diagnostics, systematic failures can be identified.

In some cases (e.g., [[Printedprinted circuit board]]s, [[Multimulti-Chipchip Modulemodule]]s (MCMs), embedded or stand-alone [[Memory (computers)|memories]]) it may be possible to repair a failing circuit under test. For that purpose, diagnostics must quickly find the failing unit and create a work- order for repairing/ or replacing the failing unit.

DFT approaches can be more or less diagnostics-friendly. The related objectives of DFT are to facilitate/ or simplify failfailure data collection and diagnostics to an extent that can enable intelligent failure analysis (FA) sample selection, as well as improve the cost, accuracy, speed, and throughput of diagnostics and FA.

== Scan design ==

The most common method for delivering test data from chip inputs to internal ''circuits under test'' (CUTs, for short), and observing their outputs, is called scan-design. In scan- design, registers ([[Flip-flop (electronics)|flip-flop]]s or [[Latch (electronic)|latches]]) in the design are connected in one or more [[scan chain]]s, which are used to gain access to internal nodes of the chip. Test patterns are shifted in via the scan chain(s), functional [[clock signal]]s are pulsed to test the circuit during the "capture cycle(s)", and the results are then shifted out to chip output pins and compared against the expected "good machine" results.

Straightforward application of scan techniques can result in large vector sets with corresponding long tester time and memory requirements. [[Test compression]] techniques address this problem, by decompressing the scan input on chip and compressing the test output. Large gains are possible since any particular test vector usually only needs to set and/or examine a small fraction of the scan chain bits.

The output of a scan design may be provided in forms such as [[Serial Vector Format]] (SVF), to be executed by test equipment.

== Debug using DFT features ==

In addition to being useful for manufacturing "go/no go" testing, scan chains can also be used to "debug" chip designs. In this context, the chip is exercised in normal "functional mode" (for example, a computer or mobile- phone chip might execute assembly language instructions). At any time, the chip clock can be stopped, and the chip re-configured into "test mode". At this point, the full internal state can be dumped out, or set to any desired values, by use of the scan chains. Another use of scan to aid debugdebugging consists of scanning in an initial state to all memory elements and then go back to functional mode to perform system debugdebugging. The advantage is to bring the system to a known state without going through many clock cycles. This use of scan chains, along with the clock control circuits areis a related sub-discipline of logic design called "Design''design for Debug"debug'' or ''design for debuggability''.<ref>[https://archive.today/20130122014800/http://www.edn.com/article/CA6451246.html?nid=3673 "Design for Debugabilitydebugging: the unspoken imperative in chip design".]

article by Ron Wilson, EDN, 6/21/2007</ref>

== See also ==

* [[JTAGBuilt-in self-test|BIST]]

* [[Automatic test pattern generation]]

* [[BIST]]

* [[Design for X]]

* [[Fault grading]]

* [[Iddq testing]]

* [[BISTJTAG]]

== References ==

{{Reflist}}

''Electronic Design Automation For Integrated Circuits Handbook'', by Lavagno, Martin and Scheffer, ISBN 0-8493-3096-3 A survey of the field of [[electronic design automation]]. This summary was derived (with permission) from Vol I, Chapter 21, ''Design For Test'', by Bernd Koenemann.

{{refbegin}}

* [http://focus.ti.com/lit/an/ssya002c/ssya002c.pdf IEEE Std 1149.1 (JTAG) Testability Primer] A technical presentation on Design-for-Test centered on JTAG and Boundary Scan

* ''VLSI Test Principles and Architectures'', by L.T. Wang, C.W. Wu, and X.Q. Wen, Chapter 2, 2006. Elsevier.

* ''Electronic Design Automation For Integrated Circuits Handbook'', by Lavagno, Martin and Scheffer, {{ISBN |0-8493-3096-3}} A survey of the field of [[electronic design automation]]. This summary was derived (with permission) from Vol I, Chapter 21, ''Design For Test'', by Bernd Koenemann.

{{refend}}

==External links==

* [http://www.corelis.com/education/Tips_DFT_Considerations_Boundary_Scan_Chain.htm Boundary-Scan Chain Design]

* [http://www.corelis.com/education/Tips_DFT_Considerations_Board_Level_Design.htm Board Level Design]

* [https://www.xjtag.com/about-jtag/design-for-test-guidelines/ Design for Testability Guidelines]

[[Category:Electronic design automation]]

[[Category:Hardware testing]]

[[Category:Design for X]]