Nope. 3point14 has it right. A3 and A6 are the same design as the other segments, but they have a fault which means their data has to be read differently
Warning 4.4mb pdf
https://authors.library.caltech.edu/91565/1/1069808.pdf
Section 1.2 Describes the operation
Section 2:4:2 Para 2 Describes the problem.
Essentially, there are a couple of defective linear variable differential transformers (LVDT) on the position sensors of mirror segments A3 and A6. Each LVDT has two coils to cancel out differential thermal effects. On A3 and A6, one of the two coils on each LVDT is faulty.
The engineers had to work out a way of getting accurate readouts using only one coil in each LVDT. So they developed a different readout procedure that could use only the one good coil in each LVDT."
Even though this was discovered back in 2018, it would have put a huge delay on the whole mission because the mirror would have to be disassembled to get the the components to be replaced (remember in 2018 they were very close to what they thought was their launch date).