Well it had to happen I guess. After 22 years of faithful service with only a few minor issues, one of my old bullet proof 022 switches suffered a fatal flaw. It quit switching into the turnout position and one controller bulb was dead. This usually means either a wiring issue, controller issue, or an electrical issue within the switch machine. A quick test involving shorting each outer electrical post to the center one confirmed my worst fears; an internal machine issue. Yikes! This means removal of the switch from the layout, with tearing up of track, moving a ton of stuff, the works! Ugh!
From this picture you can see why I was not enthusiastic about the prospect.
After feeling sorry for myself for a while and mulling over options, I decided to try something I've never done before; an in situ repair. If you are able to remove the two screws holding the switch machine to the bottom of the switch you can easily remove the machine. The trouble is the screws are only accessible from the underside! So I decided to tackle the repair from beneath the layout.
First it took about 1/2 hour just to move stuff from under the layout to get to the right area. This was not a bad thing because there were several long-lost treasures found along the way. Then I made a template of the baseplate from a spare that was on hand. I used this to mark the location of the screws and used a 1" hole saw to cut access holes, per below. Luckily the screws were right where they ought to be! The hole sawing was pretty straightforward, there being only 1/2" of homosote topped with 3/16" of cork roadbed.
After removing the screws all I had to do was loosen the adjoining switches and tracks, and lift the switch up slightly about 1/4" and slide the switch machine out. I replaced it with my ready to go spare, and bingo, all was well and good in trainland once again. Took maybe 2 hours total, and I got to unearth some cool treasures in the process!
Below are the sliding contacts within the base of these switches. As luck would have it when I connected power to the failed switch machine on the workbench, it worked flawlessly in both directions! What?? How can this be? Undaunted I cleaned the contact points up nice and shiny anyway per below and oiled it up sparingly. Now it will be the new "spare" for the next switch failure.
So in summary, this unorthadox repair saved me a lot of back-breaking work leaning over the edge of the layout pulling up track and switches. I got lucky because it so happened that both screws were accessible from below, and not obstructed by framework members. I am sure not all would be this easy to get to. But it's good to know that this may be an option for the future.
Rod