This can be seen in this photo comparing the inside of the shell to the frame with the decoder board.
It was also determined in the same way that the shorter board would fit in this location.
I have used this technique of reversing the position of the boards before to fit this type of decoder in the Atlas Classic RS1.
With the dynamic brake section removed and the shell on the frame it was also determined that there would be no clearance for the wires between the top of the frame and the shell so a groove on the top of the frame was going to be needed.
This is really important. If wires are squeezed between the inside of the shell and the frame, the shell may not be seated properly on the frame causing body mounted couplers to be too high. There is also the chance that the wires could become pinched and eventually short on the frame.
Now on to the installation
At this point, the mechanism is re-assembled and wired.
Track testing at this point should have the loco running in normal direction in DC and reverse direction in DCC.
In this photo the shell has been installed and the positions of the lens and the LED on the PC board can be seen.
Because of the reversed position of the boards, CV29 needs to be programmed for 07 if using 2 digit addressing or 27 if using 4 digit addressing.