A true ground potential is something that only exists on paper or in simulations. In the real world there is no such thing as a true ground which in test and measurement leads to ground loop errors. Ground loops present problems when measuring low level signals such as thermocouple measurements. When measuring voltages in circuits where the DMM and the device-under-test are both referenced to a common earth ground, a ground loop is formed. As shown in the figure, any voltage difference between the two ground reference points (Vground) causes a current to flow through the LO measurement lead. This causes an error voltage (VL) which leads to inaccuracies in the DMM’s measurement.
When considering ground loops just in terms of DC, as long as Ri is a large value (meaning air between the two potentials) the error will be fairly insignificant when measuring mV and up. Agilent DMMs such as the 34401A, the 34410A, and the 34411A have a Ri of 10 Gohm at 80% humidity. 80% humidity is high for a lab environment so in most settings the actual Ri is much greater than 10 Gohms. Error caused by DC ground loops can be further reduced by keeping the ground path of low level signals as short as possible.
The bigger source of noise and error from ground loops is the AC component. The DMMs impedance to ground is lower with AC because of the capacitive component, Ci, in parallel with Ri. The capacitive component results from the windings in the transformer inside the DMM. Referring to the Z calculation at the bottom of the figure, as the frequency increases the Z isolation of the DMM to ground begins to decrease. Now in most low frequency settings the ground loop noise is from the power line so it is 60 or 50 Hz. The effect of AC power line ground loop noise can be reduced by setting the DMM’s measurement integration time to 1 or more power line cycles (for 60 Hz that is 16.67 ms). If your testing environment consists of high frequency signals, high speed digital signals, or noisy components like relays or motor it is best to put any sensitive voltage measurements on a separate ground potential if possible.
For the ground loop Wikipedia page click here