Definition
Redundancy repair is the technique foundries use to salvage chips that contain a small number of manufacturing defects. Rather than scrap a die because one memory cell or logic block is faulty, designers build in spare rows, columns, or functional units. During wafer test, the locations of defects are found, and the chip is reconfigured so accesses to a bad element are silently redirected to a good spare. A die that would have been a total loss becomes a fully working part.
How the swap is made permanent
The redirection has to survive forever, so the repair address is burned into hardware. Traditional DRAM repair blows tiny links with a focused laser beam while the wafer is still unpackaged — the classic "laser fuse" process. Modern parts increasingly use electrically programmable fuses that can be blown after packaging, even by high voltage in the field. Either way, once the fuse encodes which spare replaces which defect, the decode logic routes around the flaw on every subsequent access.
The yield payoff and its limits
Redundancy repair is one of the most powerful weapons against defect density: it directly converts otherwise-killer defects into recoverable ones, lifting the number of salable dies per wafer. The trade-off is overhead — spares occupy area whether used or not, and a repair often retires a whole row or block to mask a single bad bit. The economics of large mining ASIC arrays lean on the same idea: structure the hash cores so a localized flaw disables a small block that a spare or down-bin can absorb, rather than the entire chip.
Redundancy repair is the reason a few defects no longer mean a dead chip. See the fuses that lock the repair in at e-fuse, and the yield math it fights at defect density.
In Simple Terms
Redundancy repair is the technique foundries use to salvage chips that contain a small number of manufacturing defects. Rather than scrap a die because one…