This work reviews wheel load application methods in single-wheel testbeds (SWTBs), used widely to study terramechanics. Normal force oscillations have been reported in SWTBs, and visible in time-series data when provided. This has usually been attributed to grousers, but is also visible for wheels without grousers suggesting other system dynamics (e.g. friction in the vertical axis) are the cause. Furthermore, accelerations related to periodic vertical displacements of the wheel (whether caused by grousers or otherwise) cannot explain more than a small fraction of the amplitude of normal force variations. In this work, a 4-bar mechanism was developed that balances the normal load and ensures that a constant (not oscillating) normal force is achieved throughout terramechanics experiments. System behaviour with and without the 4-bar mechanism was assessed, highlighting the need for a 4-bar mechanism or other method of balancing normal loads if a constant normal force is desired. No significant effects on average experimental results were observed due to normal force oscillation, but artifacts were observed in the time-series data. Comparisons of results with a system where wheel load is applied pneumatically and characterization of friction in each testbed provide further support to the conclusion that normal force oscillations are mainly caused by friction in the vertical axis.