7

Devil's Pitchfork

Intermediate-Advanced Transfer (manipulation) Fundamental group of configuration space

Overview

A ring sits on the left prong of a three-pronged fork. Ball-stops on the prong tips prevent the ring from being lifted off. A cord connects the ring to the base of the center prong — too short to allow a direct transfer. But the center prong is subtly shorter than the others, and this dimensional asymmetry is the key to everything.

Components

Part Material Dimensions
Base 4mm steel rod, U-shaped 200mm wide, forms the base connecting three prongs
Left prong 4mm steel rod, vertical 100mm tall
Center prong 4mm steel rod, vertical 80mm tall (20mm shorter)
Right prong 4mm steel rod, vertical 100mm tall
Ball-stops 8mm steel balls Welded to tips of all three prongs
Ring Welded steel O-ring 50mm OD
Cord 5mm braided nylon, closed loop 300mm circumference

The cord loop connects the ring to the base of the center prong via a small hole drilled at the prong's base.

Setup

Devil's Pitchfork -- Setup A three-pronged fork with ball-stops. Left and right prongs are 100mm, center prong is 80mm (shorter). A ring sits on the left prong with a cord attached to the center prong base. Devil's Pitchfork -- Setup o O = 8mm ball-stop Left: 100mm Center: 80mm Right: 100mm RING on left prong cord attached here
Setup: three-pronged fork with ring on left prong, cord to center prong base

The ring is on the left prong, resting against the ball-stop. The cord runs from the ring down to the base of the center prong. The cord is too short to allow the ring to be lifted over any ball-stop.

Objective

Move the ring from the left prong to the right prong. The ring cannot be removed from the assembly entirely.

The Topology

The set of possible positions for the ring (its configuration space) is not simply connected. The ring can be at various heights on each prong and in the channels between prongs, but the ball-stops and cord length create barriers that partition the space.

The configuration space has a non-trivial fundamental group: there exists a loop in configuration space (a sequence of moves that returns to a seemingly identical state) that is not contractible to a point. The ring must trace a specific element of this fundamental group — corresponding to looping the cord over the center prong — before the transfer becomes possible.

In physical terms: you must change the cord's topology (its relationship with the center prong) as a precondition for changing the ring's position. The puzzle has two layers — cord configuration and ring position — and the first must be solved before the second.

Configuration Space Intuition

The configuration space of this puzzle is the set of all possible positions for the ring AND all possible states of the cord. It's not just where the ring can go — it's the combination of ring position and cord geometry.

Before looping the cord over the center prong, the configuration space is 'disconnected' between left and right — there is no continuous path of ring-and-cord states that moves the ring from the left prong to the right prong. The cord constraint creates an invisible wall in configuration space.

After looping the cord over the shorter center prong, the configuration space changes. The cord now acts like a pulley, redirecting the constraint so the ring can reach the right channel. Think of it as unlocking a door: the cord loop over the center prong is the key turn, and the ring transfer is walking through the door.

flowchart LR
    subgraph Before["BEFORE (cord straight)"]
        direction LR
        L1["Left\nring here"] ---|"wall\n(blocked)"| R1["Right\nring can't reach"]
    end
    subgraph After["AFTER (cord looped over center)"]
        direction LR
        L2["Left\nring here"] -->|"door\n(open)"| R2["Right\nring can reach"]
    end

For the general theory of configuration spaces, see Topology Primer: Configuration Spaces.

Physical Intuition

What you feel in your hands: before the cord loop, the ring slides freely up and down the left prong but hits an invisible wall when you try to move it right. The cord goes taut and stops you. After looping the cord over the center prong's ball-stop, you'll feel the cord go slack in the right channel — suddenly there's room. The ring slides right as if a barrier has been removed. That moment of sudden slack IS the configuration space changing topology.

Solution

Phase 1: Reconfigure the cord

  1. Slide the ring down to the base of the left prong, between the left and center prongs
  2. Pull the cord to create maximum slack near the top of the center prong
  3. The center prong is 20mm shorter than the others — the cord can just barely reach over the center prong's ball-stop
  4. Loop the cord bight over the center prong's ball-stop
  5. The cord is now looped over the center prong, changing its effective geometry

Cord Geometry Change

Devil's Pitchfork -- Cord Geometry (Before and After) Side-by-side comparison showing how looping the cord over the center prong's ball-stop changes the effective reach of the ring, enabling it to reach the right prong. BEFORE (cord straight) can't reach 100 80 100 AFTER (cord looped) can reach! cord loops over center 100 80 100 Looping the cord over the shorter center prong gains enough slack to move the ring
Before: cord straight, ring can't reach right. After: cord looped over center prong, ring can reach right

Phase 2: Transfer the ring

  1. With the cord looped over the center prong, the ring now has a different effective "reach" in the right channel
  2. Slide the ring from the left channel, under the cord's new configuration, into the right channel
  3. Work the ring up the right prong to the ball-stop position

The key is that the cord loop over the center prong creates a pulley-like redirection of the constraint, giving the ring enough freedom to reach the right side.

Devil's Pitchfork -- Solved The same three-pronged fork. The cord has been looped over the shorter center prong, which adds enough slack for the ring to travel across; the ring now sits on the right prong, having been transferred from the left. Devil's Pitchfork -- Solved cord looped over center prong was here RING transferred Looping the cord enlarges the reachable region — the ring crosses over.
Solved: cord looped over the center prong, ring transferred from the left prong to the right prong

Why It's Tricky

The height difference is dismissed. The center prong being 20mm shorter looks like an imprecision or aesthetic choice. Solvers do not recognize it as the critical functional feature. It is, in puzzle design terms, a "hidden in plain sight" element.

Meta-level thinking required. Most puzzles have one layer: move the object. This puzzle has two layers: first reconfigure the constraint (the cord), then move the object (the ring). Solvers who focus only on the ring will never solve it because the ring's movement is impossible under the cord's initial configuration.

The precondition is invisible. The cord's state change (looping over the center prong) doesn't look like "progress." It looks like a meaningless intermediate step. Solvers who accidentally achieve it may undo it, not realizing they were one step away from the solution.

Lesson

Configuration space has its own topology. Sometimes you must change the constraints before you can move the constrained object. The structure of the problem changes as you manipulate it.

Common Mistakes

  1. Trying to force the ring over a ball-stop. The ball-stops (8mm) are larger than the ring's passage — forcing will damage the puzzle. The ring is not meant to go over any prong tip.

  2. Ignoring the center prong's height difference. The center prong being 20mm shorter is the entire puzzle. If you haven't used the height difference, you haven't started solving. Look at the prong tips — one is shorter. That's your clue.

  3. Undoing the cord loop accidentally. After looping the cord over the center prong (Phase 1), solvers often accidentally undo the loop while attempting Phase 2. Keep tension on the cord to maintain the loop, and work the ring with your other hand.

Construction Notes

  • The height difference (20mm) must be precisely calibrated:
    • Cord circumference (300mm) minus the path from ring (at base) to center prong base (~60mm each way = 120mm) leaves 180mm of slack
    • The cord needs to travel from ring position, up to center prong tip (80mm), over the ball-stop, and back — about 160-170mm
    • This leaves only 10-20mm of margin — tight enough to be non-obvious, loose enough to be physically achievable
  • Test with prototypes. Build with adjustable prong heights (use set screws) and fine-tune the cord length
  • Weld all joints cleanly; the ring must slide freely between prongs without catching
  • The ball-stops must be centered and round — any flat spot or misalignment will let the ring slip past, breaking the puzzle
  • Deburr the hole at the center prong base where the cord attaches