Force balancing

What problem it solves

A four-bar linkage in motion has a moving centre of mass. As the crank rotates, that centre of mass accelerates — and Newton requires that acceleration be supplied by a force from the ground frame. That force is the shaking force: the unwanted reaction transmitted from the mechanism into whatever it is bolted to.

In an engine, the shaking force is what makes the block vibrate. In an oscillating sewing-machine head it is what walks the machine across the table. In a balance-shaft drive it is the entire reason the balance shaft exists. In a printer carriage at high cycle rate it is what fatigues the mounting bracket and breaks it open after 10⁶ cycles.

The classical fix is force balancing: add counterweights such that the moving centre of mass is held stationary in the ground frame as the mechanism cycles. If the centre of mass does not move, no net force is required to support it. The shaking force vanishes — at least in principle, and at least at the harmonic frequencies the analysis covers.

The math (sketched)

For a planar four-bar with link masses m_i at link centroids r_i, the position of the system centre of mass is a function of the crank angle θ. Decomposing that position into a Fourier series in θ yields a constant term (which contributes nothing dynamically) plus first-harmonic, second- harmonic, third-harmonic … terms.

Two counterweights — one on the crank with mass-radius product (m_a, r_a), one on the rocker with (m_c, r_c) — provide four scalar design knobs. Setting the first-harmonic terms of the centre-of-mass motion to zero is two algebraic equations in those four knobs. The remaining two knobs can be used to balance higher harmonics, minimise total counterweight mass, or stay within a packaging envelope.

The result is closed-form, not optimisation-based, when the goal is first-harmonic balance. Higher-harmonic balance, minimum-mass balance, or balance under additional constraints can be cast as a small optimisation problem.

Reduction across harmonics

The marketing line on this is 100% shaking-force reduction across harmonics 1× through 6× on the chapter benchmark, and that is real:

• 1× — fundamental, the largest and most dangerous component. Closed-form balanced exactly to zero.
• 2× through 6× — higher harmonics. Often also eliminated by the same counterweight choice for symmetric four-bars; in asymmetric cases the designer trades among them.

Beyond 6× the harmonic content is usually below the noise floor of the mechanism’s structural response — there is no audience for a 7× correction.

Force balance vs torque balance

Force balancing kills the force transmitted to the frame. It does not necessarily kill the torque. A force-balanced four-bar can still generate a non-zero shaking moment that rocks the frame about its mounting axis. Moment balance is the dual problem and uses different counterweight placement; in the chapter we treat it separately.

In practice many production mechanisms force-balance and accept some residual moment — the moment is far easier to absorb in a stiff mounting than the force is.

Co-design — the platform’s headline differentiator

Counterweight masses are not free. They take packaging volume, they cost material, they increase rotational inertia, and they may push the transmission angle floor down or pull the coupler curve away from its design target. Force balancing in isolation can produce a balanced mechanism that fails on every other axis.

The platform’s co-design chapter addresses this directly: a single optimiser sweeps over link lengths and counterweight masses on a Pareto front of (geometry cost) vs (balance cost). The engineer picks a point on the front rather than balancing first and discovering later.

This is the differentiator that separates the platform from “another MBSD library”: no other open-source platform ships joint co-design of geometry and counterweights with the dynamics engine validating each candidate inside the same loop.

See also

• Four-bar linkage
• Burmester theory — geometric synthesis comes first; balance comes second.
• Practica case study — the full pipeline including dynamics validation.