Hey Laurence,
These methods are not as accurate as a bar gauge. See tolerance in the laser manual. Using a square is not accurate enough, and a square of this size and at same time of sufficient (and certified) precision would cost many hundreds of dollars/euros. The best method is the compare diagonals method, no measuring is needed, just a bar gauge to compare them. It is important to measure between the point where the chrome-plated, hardened steel hollow shafts go into the black anodized axis feet, and not on the base of the machine’s feet, because the steel shaft’s position can be adjusted inside the black anodized axis ends, on each end. Also it is important to consider that the equal diagonals method only works in a parallelogram, but not in a trapezoid, which means, with the bar gauge, I would first measure if the Y rails on the left and on the right have the exact same length (to exclude that assembling the rails in the factory had a deviation).
Generally the question on this issue is: Have you ensured that your machine is rectangular (“squared”) (bar gauge) and coplanar (“not twisted”) (fishing line method).
In both cases if it is not accurately adjusted, not only you can get parallelograms instead of rectangles, but also workpieces with a twisted surface. Surfacing the wasteboard will not help against lack of coplanarity (twisted machine base), the machine will only reproduce the twist on the workpiece surface. Also if any of both is not accurately adjusted, the Y movement can block.
Finally you can also check the tramming of the milling motor. One of its adjustments is made with the forward/back tilt of the two steel hollow shafts by adjusting their adjustment bolts in the black anodized axis ends (see Support: Tramming - Front To Back for details). If the tilt differs on one end of the axis from the other end, the milling motor will tilt from one end to the other and make the line it mills with your v-bit slightly deviating from a right angle.