To improve matters, a variable capacitor is often added in parallel with the resistance of the attenuator, resulting in a
compensated attenuator probe such as the one shown in Fig. 1. In Chapter 19, it was demonstrated that a square wave can be generated by a summation of sinusoidal signals of particular frequency and amplitude. If we therefore design a network such as the one shown in Fig. $ 1 $ that will ensure that $ V_{ \text {scope }} $ is $ 0.1 V_{i} $ for any frequency, then the rounding distortion will be removed, and $ V_{ \text {scope }} $ will have the same appearance as $ V_{i} $.
Fig. 1: Compensated attenuator and input impedance
to a scope, including the cable capacitance.
Applying the voltage divider rule to the network of Fig. 1,
$$\mathbf{V}_{\text {scope }}=\frac{\mathbf{Z}_{s} \mathbf{V}_{i}}{\mathbf{Z}_{s}+\mathbf{Z}_{p}} \tag{1}$$
If the parameters are chosen or adjusted such that
$$R_{p} C_{p}=R_{s} C_{s}$$
the phase angle of $ \mathbf{Z}_{s} $ and $ \mathbf{Z}_{p} $ will be the same, and Equation (1) will reduce to
$$\mathbf{V}_{\text {scope }}=\frac{R_{s} \mathbf{V}_{i}}{R_{s}+R_{p}}$$
which is insensitive to frequency since the capacitive elements have dropped out of the relationship. In the laboratory, simply adjust the probe capacitance using a standard or known square-wave signal until the desired sharp corners of the square wave are obtained. If you avoid the calibration step, you may make a rounded signal look square since you assumed a square wave at the point of measurement. Too much capacitance will result in an overshoot effect, whereas too little will continue to show the rounding effect.
Do you want to say or ask something?