An investigation of potential mechanisms for puzzle boxes.
I don’t remember when I first became aware of puzzle boxes, or the idea that I could make my own. In fact, I don’t think I’ve even seen one in action. But I have been interested in them now for a few years, and as an amateur woodworker (emphasis on amateur), I have an interest in making them. What I’ve lacked for the longest time is a clear set of designs, ideas for working mechanisms. While I haven’t looked at the plans others have made, I have nevertheless been captivated by puzzle boxes to the extent that my engineering mind has concocted its own plans.
Figure A: Catch
What I lay down here is a kind of methodology for the family of mechanisms I have in mind. This family shares a number of attributes, the most important of which is the catch. The catch works a bit like those 3 digit combination bike locks that you find attached to a cable, where you have to line up the numbers to form a straight gap that will allow the pin to slide out. In the case of this puzzle box mechanism, the catch need only pass through one such gap after rotating through sufficient range and is fully engaged when rotated between 270 and 359 degrees clockwise or counterclockwise (see figure A). Rotational motion is impeded by from one to six barriers, which are themselves moved by solving the mechanism for each side. Complexity of the puzzle can be scaled up or down and depends entirely on the number of catch barriers, the number of steps and combinations of the side mechanisms, and the degree of independence that each side mechanism has.
Catch Barriers
Beneath the catch ring are the barriers that prevent rotational movement of the catch. Of these, one barrier is fixed to force rotation to a particular direction, either clockwise or counterclockwise. At its initial (locked) setting, the catch should be barricaded on both sides to allow little or no rotational motion. The remaining catch barriers should travel along paths made accessible by the movement of the side mechanisms. I have toyed with a number of possibilities for the shape of the catch barriers, but a simple sliding bar seems to be the easiest to implement and provides the greatest stability against the rotational motion of the catch.
Side Mechanisms
The side mechanisms control the release of the catch barriers and take the form of a series of sliding bars. The configuration of these bars provides the greatest input to the complexity of the final puzzle. Configuration choices include degrees of independence, ranging from completely independent (simple) to completely dependent (complex); number of paths (usually one per side, but could be more in case of dependence between sides); degree of path constraint, ranging from fully constrained (each bar moves in a particular order) to loosely constrained (while the correct solution requires movement in a particular order, the bar may move the same distance in either direction from fully locked position); and the mechanism action, either hidden action (each bar controlled by a knob installed flush against the outside of the box so that the internal motion can only be guessed) or visible action (each bar controlled by a visible slider so that the internal motion is obvious). There may be (and indeed probably are) many more configuration options available, but let me attempt to describe the ones I have outlined here, providing diagrams where possible.
Figure B: Independent Mechanics
Completely Independent (figure B) – In this configuration, the motion required to solve each side is not dependent on the motion of other sides. It is by far the easiest to design, implement, and solve.
Partially Dependent (figure C) – In a partially dependent scenario, the solution of one side is dependent on the solution of at least one other side. This adds a layer of complexity to the design, and may either make the solution easier (if the side requiring a prior solution is immobile until the solution is provided) or harder (if the side requiring prior solution merely requires it for its own final solution).
Wholly Dependent (not pictured) – If the sides are wholly dependent, this means that each side must be solved in a particular order for the final puzzle solution to present. In the most complex interpretation, each side may have a partial solution that unlocks the next successive side and requires steps to be completed on the preceding sides before a final solution is achieved.
Figure C: Adjacent Dependency
Number of Paths – Since the side mechanism is comprised of sliding bars, it can be constructed in such a way that multiple paths lead to a side solution. While this can reduce the complexity of the puzzle, it adds some complexity to the design. One path should be sufficient for most applications, but it occurred to me as a potential option, so I am including it to be thorough.
Fully Constrained (see figure B again) – If the side mechanism is fully constrained, then on each side the only movable bar will be the first required to move. Movement then proceeds sequentially until all bars have been moved and the catch barrier is removed. This is a very simplistic approach and correspondingly easy to solve. It also has the benefit of being easy to design.
Loosely Constrained (figure C also works for this) – Where full constraint means that the solution path is easy to determine, loose constraint means it takes trial and error. This is because the bars may move in either direction, approximately the same distance about the fully engaged position, without revealing which direction was correct. The solution still requires the bars to be moved in a certain order.
Figure D: Hidden Mechanism
Hidden Mechanism (figure D) – In this scenario, the internal movement is obscured by external knobs set flush against the surface of the box sides. Inside, movement is achieved by using gears and gear teeth on the bars such that the bars move up or down, left or right depending on the bar orientation and the direction in which the knobs are turned. Gears are affixed to the internal box walls to minimize erroneous movement. The space for the bars must be engineered as precisely as possible to prevent free play (which could lead to jamming). Some provision might be required to keep bars from moving once supports have been removed.
Visible Mechanism (figure E) – Push rods act to slide the bars along externally visible channels, either in a straight line or along angular paths. Virtually any configuration is possible here, and it contributes well to complexity.
Figure E: Visible Mechanism
One final possibility that suggests itself is the use of a strong magnet to move a completely hidden mechanism (such as the sliding bars). To do this effectively, however, and retain the puzzle-like qualities, the design should not be prone to excessive scratching, lest the solution be guessable by merely observing the surface. I will leave it to anyone who stumbles across this posting to provide ideas on achieving this.
Conclusion
Now that I’ve outlined some possibilities, the next step is to pick one or more of these mechanisms and build the box. The most effective approach may be a combination of mechanisms to add variety; after all, variety provides the challenge. I should also note that it might be wise to take some steps toward disguising the fact that the box is in fact a puzzle box, if the goal is to effectively hide things. For mere amusement, however, no such steps are explicitly necessary.
I am intensely interested in the thoughts of others regarding this approach. Does it look viable? If I manage to build something from this, would anyone want the plans and/or a finished box? Leave your thoughts in the comments and let’s talk about it.
GooglePlus
Twitter