Concentric rotary encoder – band switch

In a Boeing 737 NG cockpit a number of concentric shaft rotary encoder / band switch combinations are found, e.g. the Heading / Bank Angle selection combination in the MCP and the MINS and BARO combinations in the EFIS panel.
One way to build this combination is to drill a hole completely through the shaft and the body of a band switch so that a smaller shaft can go through it that is connected to the axis of a rotary encoder behind the band switch.
I used band switches of brand Alpha, similar to this one. If you buy them make sure that the axis is a fully round (not D-shape) plastic, 6 mm diameter; the knob mentioned below below only fits on a knurled axis, which is 6 mm instead of 6.35 mm.
The used rotary encoders are similar to this one, the shop where I bought my ones unfortunately doesn’t sell them anymore but there are many other shops selling the same ones.

Drilling this hole is a precision job, especially if the band switch shaft has 6 mm diameter and an inner shaft of 4 mm is used so that a knob with 4 mm hole exactly fits. This knob fits exactly in this bigger knob for the outer shaft.
For the inner shaft a 4 mm brass rod from a DIY shop is used.

With inspiration of a YouTube video I found the following way which appeared to be not as difficult as I expected. Using the described method I modified a total of 5 band switches in approximately 1 hour.

Needed are a drilling stand, an extra drill chuck (which came with my hammer drill) and a vise. I used a piece of the brass rod to align the drill and the extra chuck:

 

The first attempt resulted in this:

I was a bit puzzled where the part of the spring came from, so I opened the band switch to investigate.

The spring that pushes two metal balls in the serrated body of the switch to make it click runs straight through the shaft, so if you drill through the shaft without removing the spring first you run into trouble. But it appeared quite easy to open the switch, you only have to be careful when pulling the shaft out otherwise the two metal balls will fly away.

With the shaft removed, it’s much easier to drill the hole, because the internal end of the shaft has a center hole which makes it easy to center the drill.

But before drilling the shaft, first drill the hole in the switch body, because a massive shaft is less fragile when clamped in the chuck.

Next drill the hole in the shaft. Be careful not to damage the slider contacts.

I managed to drill a hole of 4.5 mm trough the 6 mm shaft, but better have a couple of reserve band switches available, because it requires a bit of practice to get the hole also exactly centered at the other end of the shaft.
Switches with knurled shaft are a bit thinner than non-knurled ones, so the latter are a bit easier, but the knob for the outer axis mentioned above only fits on the knurled axis.

Next the spring was cut in two pieces and the 4 mm brass rod inserted as a stop in the middle for the two spring pieces.

Putting the two balls back in place is a tricky job, but there is a small opening in the switch body that makes it easier: rotate the axis so that this opening exactly aligns with the hole for the spring, place the spring and first ball and rotate the axis to the other spring hole to place the second spring and ball.
Next pull the shaft completely towards the top of the switch body.

Picture to be updated to show the small opening aligned with the spring hole.

 

Now the switch body can be closed again.

In this picture the small opening is visible that makes placing of the spring and ball easier.

This was the first and most difficult part, next I have designed a 3D printed frame (STL format) to connect the rotary encoder and the band switch together and a small cylinder (STL format) to connect the 4 mm brass rod with the rotary encoder axis.