But I am not interested in matching resistors. What I want is a resistor that stays stable, even when extra stress is applied on the package. So if I have 160 unit fingers, I place 2x40 units horizontal, and 2x40 units vertical. These two groups are placed cross quad. The number of units is chosen in sutch a way the that the final result is a perfect square. The different orientation of the units cancels out the effects of stress.
I have seen this before, but I have nothing to back it up. Maybe someone knows a paper or book that references this?
- - - Updated - - -
But I am not interested in matching resistors. What I want is a resistor that stays stable, even when extra stress is applied on the package. So if I have 160 unit fingers, I place 2x40 units horizontal, and 2x40 units vertical. These two groups are placed cross quad. The number of units is chosen in sutch a way the that the final result is a perfect square. The different orientation of the units cancels out the effects of stress.
I have seen this before, but I have nothing to back it up. Maybe someone knows a paper or book that references this?