DCC Controlled Uncoupler

One of the magical things about DCC is that we have been freed from the constraints of relying on the layout itself while we are controlling our trains. Sound systems and full controls have moved into the locomotives themselves. It was obvious to me from the first time that I heard about this new DCC control system that it could even include the uncouplers themselves, if I could just figure out how to actuate them. I use Kadee couplers in HO guage, so it was natural that I would try and actuate them. I searched on the net and saw some schemes for operating couplers using relay coils, but they seemed to fill one whole end of a locomotive and be designed to operate european couplers. I even found one web site that stated flatly that Kadee couplers could not be automated. Obviously there had to be something smaller that would do the job. I toyed with the idea of small solonoids disguised as air tanks, but that proved unworkable as well. In the process I discovered just how little empty space there was inside of an HO switch engine.

I had previously obtained some Muscle Wire® to try and actuate a semaphore signal. That was in the days before I obtained my 12" = 1' version. My first thought was to simple use the wire to pull on the knuckle. There are several problems with that. First is that the wire only changes length about 3-5%. To get enough motion would take half the length of a locomotive. Another problem was that if you just pull straight back on a coupler it will swing to one side. You don't want that to happen. A bit of fussing showed that to pull on a lever that will swing the knuckle without moving the coupler itself you have to pull from directly over the coupler mount pin. Actually that information helped, because I now knew that the actuator had to sit right under the coupler pocket. I did some measurements of the force needed to open a coupler reliably and discovered that the muscle wire had 10 times what I needed even in the next to the smallest size. The operating current was also less than a decoder could put out, so it could be directly driven from the decoder. Hey, were on a roll here! The extra force immediately made me start thinking in terms of a mutliplication arm that would trade force for distance. After a few false starts, the swinging "T" bar was dreamed up, and the first one I built worked! (after a fashon) It actually worked for nearly a half hour befor it died. The next two that I built are still operating 3 years later.

The actuators are built on a PC board base. The length just fits between the bottom steps, and the width is from the back of the running board of my Proto-2000 SW-9, to the back of the coupler pocket, (about 1/4") which just clears the brake rigging on the trucks. The actuator mounts under (bottom side when Loco is on rails) the coupler pocket using the normal coupler mounting screw. I put the screws into my Dremel tool and filed them into a countersunk flathead shape and counter sank the hole in the PC material, so they do not stick up above the surface of the PC board and catch on things. I actually used 1/16" thick fiberglass PC material, but removed half the thickness over the coupler poclets to provide more clearance.

These actuators require about 1/2 volt at 180-200ma. to operate. The DCC decoder functions supply 10 or more volts. This excess voltage needs to be dissipated safely. The electrical connections and current limiter are also described.

Click on the engine to continue.