It's nice to use servos, but how to fasten them on or under the table? How to transmit their movement to the turnouts? How can you activate a switch to feed the frog? There are several possible answers. The simplest, but also the most expensive, is to buy specially adapted components, such as the TAM or ESU servos supplied with mounting chassis and hardware. I preferred a more personal, but easily adaptable solution.
Apparently, most brands chose the classic solution already used for conventional motors (Peco, Tortoise), that of a simple rod crossing the table and directly driving the turnout mobile sleeper.
In the case of a servo, this rod is fixed to the rudder bar and thus describes a rotational movement in a vertical plane. The main disadvantage I see with this solution is to require a hole in the table, large enough to allow the rod to move with clearance, ie a diameter of about 10 mm. And this hole must be made before laying the track, or you risk damaging a turnout already in place. Another disadvantage is that, in order to leave a certain flexibility to the transmission, the rod must be long enough; the height under the table may be significant.
I preferred the solution of the lever rotating in a horizontal plane, forming a two arm crank, and pivoting in a hole of small diameter (1.2 to 1.5 mm). This is actually the one recommended for Fulgurex motors. The mounting brackets of the servo are horizontal, which also facilitates the setting. Note that, for mechanical reasons, the axis of rotation of the crank must be (approximately) coincident with that of the servo rudder. Also note that it is absolutely not necessary that the two arms of the crank are in the same plane. The bottom one can be oriented at will to facilitate the implantation of the motor.
As you can see from this drawing, I chose to fasten the servo on an epoxy PCB. Advantages: good rigidity, and ability to create a circuit to connect a frog feeding switch. On the other hand, I wanted to reuse the Peco switches that I had left of the first installation with the same brand motors. But it will be seen that microswitches, cheaper and probably more reliable, can be used.
The servo is fixed to the board by sheet metal screws (the supplied screws are often too short), with spacers to adjust the height of the rudder bar. Depending on the servo type, the mounting brackets are not located at the same height. Fixing the board on the table will be done by three wood screws, also with spacers. The chosen solution required me a lot of milling work on the board, to adapt the Peco switch and the servo itself.
The Peco PL-13 switches, originally designed to fit on PL-10 motors, do not have any fixing points. I fix them by their pods, welding them on the board. But this obliges me to mill a rectangular hole to receive their casing, which will thus be perfectly maintained. The position of the switch with respect to the servo must be sufficiently precise for the through hole of the control rod to be aligned with that of the rudder bar. Rudders are not not standardized, neither in terms of distance between the perforations nor for the fixing to the servo shaft. So they are unfortunately not interchangeable. This means that the distance between servo and switch must be adjusted according to the available models.