At Thingiverse.com a 3D printable design for a solenoid engine starter switch can be found.
Unfortunately the design for the axes is missing and the knob must be pushed in also for the CONT and FLT switch positions, while this normally is only required for the GRD position (as confirmed by someone who has access to commercial simulators used by airlines).
While building the Thingiverse design I noticed a problem: the TDS-08A solenoids that I purchased were missing the spring that is required to make the design work.
In the 200 pc spring collection that I used for several cockpit parts I found a spring that fitted, but it is too weak to push the switch to the OFF position with the solenoid unpowered because of the rotary switch clicks.
This can be fixed by removing the spring and balls from the rotary switch as I did for some other cockpit switches, but this makes it difficult to turn the switch to the exact CONT and FLT positions.
Also the impulse of the powered solenoid in that case pushed the switch past the OFF position to the CONT position.
Another problem that I noticed is that the solenoid plunger sometimes rotates a bit causing the screw in the switch axis to get stuck to it.
I tried to solve the switch positioning problems by various designs for the panel side part of the switch:
These couldn’t solve the problems with the solenoid however: I needed a way to limit the outer position of the plunger.
While looking at the solenoid I noticed that it has screw holes on both sides, which brought me to the idea to attach an L-profile to the solenoid that limits the outer plunger position.
Initially I assumed that a 3D printed L-profile ‘solenoid stop’ would be too fragile but there is enough space to add some reinforcements and by printing the part sideways the filament is following the L-shape.
With the first prototype of this solenoid stop I realized that by extending it a bit further away from the solenoid and adding a slot it could also prevent the plunger from rotating.
The ideal rotary switch would have stops between OFF, CONT and FLT positions and no stop between GRD and OFF so that the solenoid spring can easily return the switch to the OFF position and the other positions have a solid click.
I have tried to remove the GRD-OFF stop inside the rotary switch, but it was difficult to do this accurately.
Using the CNC mill would have been an option, but I wanted a solution that can be used without having access to a CNC.
One option was to create slots in the panel side part of the switch.
This sort of worked, but didn’t give a nice click.
In the end I realized that I could use the spring and balls that have been removed from the rotary switch: the springs and balls are placed in holes in the panel side part of the switch and lock to small notches in the outer axis.
During the first test with spring holes that only had an opening at the inner side I felt that this gave a nice click, but it appeared impossible to place the springs and balls and insert the axes.
This has been solved by extending the spring holes to the outer side so that the balls and springs can be placed from the outside and locked in position after the axes have been inserted.
An additional advantage of this is that it is possible to experiment a bit with the correct length of the spring for the desired click force.
I’ve also designed a new knob for the switch with a D-shape axis and hole for a square nut to tightly screw the knob to the axis.
For the three M5 screws connecting the panel side part with the rotary side part I used 50 mm long screws that I had left from another project.
For the rotary switch I used this.
I first assembled the starter switch with the spring and balls still in the rotary switch, rotated both the rotary switch and starter switch axis to the rightmost (seen from the panel side) FLT position and then drilled the 3 mm hole via the existing hole in the outer axis through the rotary axis.
In this way the clicks of the rotary switch prevent its axis from moving while drilling the hole. This is important for correct alignment of the starter switch positions with the rotary switch positions.
Then I removed the rotary switch, opened it to remove the spring and balls and assembled it again.
In the 200 pc spring collection mentioned above there are a number of compression springs that are only slightly different:
For the inner axis the second spring from the above picture is used which has a slightly larger diameter than the left one which is just enough to avoid that it disappears in the upper part of the outer axis. The diameter is approximately 7.15 mm (9/32 inch), the length 12.7 mm (1/2 inch) and the wire diameter 0.5 mm (= ~0.02 inch).
For the solenoid the third spring in the picture is used. This one has a diameter of approximately 9.52 mm (3/8 inch), length of 15.8 mm (5/8 inch) and wire diameter of 0.5 mm (= ~ 0.02 inch)
For the cross axis through the inner axis I originally used a 3 mm metal rod, but the hole for this made the 5 mm 3D printed inner axis too fragile, so later I have replaced this by a 2 mm metal rod of approximately 22 mm length.
An instruction video how to interface it with SimVim to X-Plane will follow later.