Jim Mietlicki writes: Anyone who builds with Erector knows that when you try to follow the instructions in the manual, the models just don’t seem to work right. I honestly think A. C. Gilbert was deliberately unclear in his model instructions so that a young boy would have to observe the model, figure out what was going wrong, and then find a solution to make them work.

Both A. C. Gilbert and Frank Hornby of Meccano said the basis of their construction systems was to teach kids how to build things. As the models got more complex, these inherent problems served as a powerful teaching tool. If you look at the Erector Ferris Wheel—the simplest of the Amusement Park models—if you run them with the pulleys set up as the manual shows, you’ll discover they run far too fast and the string band is likely to slip. But observe a real working Ferris Wheel, and you see that motive force is always applied at the outer rim of the wheel. Make this change to the Ferris Wheel model and the wheel runs at a reasonable and consistent speed, once you then realize that a simple clutch mechanism is needed for consistent pressure on the string band.

When you have to figure this out for yourself, you actually learn about the whys and hows of mechanical operation. That’s a far more instructive and a valuable lesson than being told what to do.

If you are of a mind to try the Walking Robot, it’s a time consuming build. My version of the model is based on the 1956-1959 Walking Robot, and any 12½ set from this period could be used to build the model. Let me go through my modifications:

1. The motor mechanism itself is not complex—just 2 BTs on a motor gearbox that work the 12-inch axles that push the feet back and forth. The feet are where the problems arose.
2. The manual show 4-inch axles, through flat car trucks holding the NX pulleys and CJ gear for each foot. Only one shows an external P-37, with the other points on those axles braced internally. But the model above the feet involves a lot of steel and weight. As the sides of the feet made with EY girders and the flat car trucks are not braced at the bottom, the weight tends to cause them to flare out, making the feet all uneven. As I don’t believe there are any axles from Gilbert between 4-inch and 5-inch, I make my own to allow for the P-37s to be installed outside the flat car trucks, which serves to brace the bottom of the feet in place.
3. The manual also shows that the inner side of the feet held up with one flat car truck and a flanged wheel. Not enough support to keep the feet steady. I used 2 flat car trucks with P-7s as wheels placed on the next holes behind the axles on which the NX pulleys are mounted for stability.
4. The O pawl ratchets continue to be a problem as they tend to stick. So I removed the O pawl ratchets, reinstalled same on a ½-inch screw for they axle pivot instead of a ¼-inch, with washers greased with lithium on both sides of the pawl. Fixed in place with the double square nuts locked together and carefully set to allow just the necessary play. O pawls not slipping on and off the CJs with no binding, and it’s working just as it should. VICTORY!

March 30, 2022 – Continuing saga of the Walking Robot, I think I finally have it licked. After moving the position of the treads, and moving the flat truck and wheels so one supports each side of each foot — and later changing back from P-7 wheels to the original flanged wheels — one foot was working perfectly while the other was still lagging. When they are not both moving in alternating unison, the robot doesn’t move. I considered whether it could be a difference in force exerted by the levers. These pass from the BTs on the gearbox down through the CH right angles at the bottom of the main body, and down through the O pawls at the bottom of the legs to create the leverage that moves the feet.

I moved the CH on the side where the foot was lagging back one hole, figuring it might increase the force from that lever. And it worked! The feet are now working as they should. Unfortunately, this change has resulted in the axle-based lever protruding down too far and hitting the rear of that foot. So now I have to remove the head to slide it out and adjust the P-37 on that lever down about 1/4 of an inch. Will this do end the quest? I believe it will, but of course we shall see when it is done.

May 5, 2022 – One more modification on my Walking Robot. The robot is now moving without any issues, although I did change the gearing to get it back closer to what’s shown in the manual. While my original plan had him moving faster, it also made him less stable. Slowing him down fixed the fear of falling. But then there was an issue with the lighted eyes. The manual shows the necessary battery with the Erector battery holder mounted vertically inside the robot’s head on a fiber strip. That certainly works. But inserting or removing the D battery, while not impossible, is rather difficult. And leaving the battery in place when the guy isn’t in use is not ideal. To remedy this, I eliminated the D battery and went with a AA, and built a battery holder that’s mounted at the top of the head in horizontal position. This was done by forming a battery holder from a broken strip I had and some brass strip, using brass for a clip for the negative pole on the battery. One end of the strip is bent up to hold the contact for the positive pole that will connect to the socket wires for the eyes. The other is bent downward to serve as a contact for negative ground and is positioned there the battery holder used in the original would be located.

The use of the robot’s nose with the inner O pawl that turns with the nose to create the negative ground does not change, but contacts the strip protruding downward that is actually part of the AA battery holder. To hold this contraption to the side of the head I made two brackets using two-stage epoxy putty, which allowed for the forming on U shaped brackets so the battery slips in easily into the U’s. I won’t say these brackets are pretty, but the two-stage putty is not conductive, serving the same purpose as the fiber strip used in the original instruction. And as the battery is at the top of the head in a horizontal position. Just remove the top of the head and it can be easily inserted and removed.