Study On Increasing The Length Of Test Groups Using Of Don't Care Bits

With the advances in manufacturing technology, the complexity of integrated circuits(ICs) is increased rapidly, and ICs test become increasingly difficult. Currently, the technology which named Design-for-Testabiliy(DFT) is a good way to solve the test problem. The fundamental idea of DFT is to consider the test issue while designing, in order to make the ICs to be tested easily. Build-In Self-Test (BIST) is a commonly used DFT technologie.Recently, a BIST scheme using test patterns applied by Circuit-under-Test (TPAC) is proposed. In this approach, CUTs are no longer only regarded as test objects, but also a sort of available resources. When the first test vector is applied to the CUT, the responses of the interior nodes are captured by flip-flops and are fed back to the inputs of the CUT as the next test vector.The don't care bits in the MinTest test set and the unknown bits in the response vectors of interior nodes are researched in this article. In this paper, a identity rule of unknown bits is proposed. By the rule, all response values of the interior nodes can be calculated.Based on the rule, this paper presents a method to make full use of don't care bits in the test vectors to increase the length of test groups in TPAC. The response vectors of interior nodes are generated from a given circuit and its test vectors which include don't care bits. Directed graphs are built based on some heuristic rules and a longest public path is found by using depth-first search algorithm. Through analysis, don't care bits can be used to delay the speed of alternative nodes to reduce, but unknown bits will accelerate the reduction of alternative nodes. Fortunately, the proportion of unknown bits in response vectors is lower than the proportion of don't care bits in test vectors. So, the application of don't care bits in TPAC is feasible.Experimental results on the ISCAS'85 benchmark circuits demonstrate that the number of vectors in the largest test groups increase by an average of 23.9%, and the test groups are reduced 15.9%.