In low voltage and high current applications, it is common to use a measuring instrument capable of 4-wire sensing when it is desired to minimize the effects of resistance between connections (such as wire resistance, alligator-to-pin resistance, alligator-to-alligator resistance and so on) due to a significant voltage drop consequently being induced on those paths. However, disaster can strike when sensing is implemented erroneously; increasing both the magnitude of error (instead of the measurement accuracy) and the number of wrinkles on ones forehead. Such disasters can be avoided if one has a firm grasp on the underlying principle of how sense lines work and if one takes the utmost caution in the practice of such a witty method.
The difference between 4-wire and 2-wire measurement
What is 4-wire measurement anyway and how does it differ from the ordinary 2-wire measurement? 4-wire sensing or 4-terminal sensing as Wikipedia describes it is “an electrical impedance measuring technique that uses separate pairs of current-carrying and voltage-sensing electrodes to make more accurate measurements than the simpler and more usual two-terminal (2T) sensing”. Further annotation by Wikipedia explains the concept as “…a pair of sense connections (voltage leads) which are made immediately adjacent to the target impedance, so that they do not include the voltage drop in the force leads or contacts.” Take for example a 5V source connected to a 2A load through metallic meter sticks. If the meter sticks each have a 1 ohm resistance, then the potential developed at the load would only be 1V. But, by introducing a feedback path (the sense lines) that could say to the 5V source “Hey! I’m only receiving 1V here, so you better increase to 9V?” then the correct/desired voltage level will be impressed at the load. It is self-evident that a 2-wire measurement set-up can’t achieve this self-correcting mechanism due to the absence of the feedback path.
How the 4-wire set-up works
So the 4-wire method is an effective set-up for compensating drops on low resistance paths. But how does it do it electrically? Below is a simple schematic of a 4-wire set-up.
Figure 1 Simplified schematic representation of 4-wire sensing.
As with nearly all feedback paths in electronics, a 4-wire schematic would have the ever ubiquitous operational amplifier. Due to the op-amps impressively high input impedance, it serves as a perfect candidate for sampling voltages in your circuit without having a significant effect. This characteristic is effectively taken advantage of in a 4-wire set-up because of the implication that there will be an infinitesimal voltage drop at R3 and R4. Voltage drops at the sense lines are highly undesirable because that is where the true developed voltage is sampled.
From a block level perspective, the schematic looks like 2 voltage followers, one voltage follower tracking the high end and the other voltage follower tracking the low end (or ground potential).
A danger in 4-wire sensing
The 4-wire set-up has to be practiced with utmost caution in lieu of the possible scenario demonstrated in Fig. 2
Figure 2 4-wire set-up with a floating sense pin.
What happens if one of the SENSE pins is accidentally left floating (like when a cold solder breaks loose or an alligator clip snaps free)? The SENSE lines are now telling the source to keep increasing its voltage because it is just reading a high-impedance (i.e. approx. 0V reading). Eventually, the FORCE lines (R1 and R2) will keep increasing the bias level until the pre-set current limit is reached, possibly destroying the DUT (device under test), or worse, damaging the supply (though a supply is rarely damaged this way if it was manufactured for industry use).
One way I can think of to avoid damaging the supply is by connecting a switch between the SENSE and FORCE lines. See Fig. 3
Figure 3 Possible solution to avoid damaging the supply when a floating SENSE line occurs.
The PROT_EN signal creates a short between the force and SENSE lines when the supply voltage reaches a threshold (maybe on a level that corresponds to the pre-set current limit).