Why You're Measuring Resistance and What It Actually Tells You
Most people approach resistance testing wrong. They treat it like a checkbox exercise — connect the leads, read the number, move on. That's why boards get shipped with marginal continuity and motors end up drawing 30% more current than they should. A proper resistance test reveals more than just "connected or not." It shows contact resistance in relays, wire gauge mismatches, degraded solder joints, and early-stage insulation breakdown before a full shorts test even triggers. I've spent years debugging electronics through resistance measurements alone. The trick isn't the multimeter; it's knowing what voltage and current to use during the test and understanding how temperature skews your reading. A J-type thermocouple wire and a copper trace of the same length will read nearly identical at room temperature, but under load those values diverge enough to cause real issues in precision circuits.
How To Do A Resistance Test Properly
Start by selecting the right test method for what you're measuring. Four-point Kelvin sensing is the gold standard for low-resistance work — anything under about 10 ohms. Two-wire measurements introduce lead resistance into your result, which matters when you're trying to verify a ground plane connection that should read near zero. If you don't have a Kelvin setup, you can compensate by measuring the leads first and subtracting, but that adds a step where errors creep in. Here's the practical sequence I use: First, power down and discharge the circuit. I know that sounds obvious, but I once blew a $40 DMM fuse on a board that still had 12 volts across a capacitor bank. The capacitor looked dead from the outside. Always probe the rails with a high-impedance voltmeter before switching to resistance mode.
Second, set your meter to the appropriate range. If you're checking a fusible trace on a PCB that should read close to 0 ohms, start on the lowest ohms range. If you're testing insulation resistance between two isolated traces, you need the megaohm range. Don't jump around trying to find the right scale — pick it upfront based on what you're expecting. Third, apply consistent probe pressure and contact points. I've seen technicians get different readings on the same joint simply because they pressed harder with one hand than the other. For through-hole components, touch both leads to the same pin if you're measuring component body resistance, or to adjacent pins if you're checking trace continuity. Be methodical about where the probes land. Fourth, account for parallel paths. This is the most common mistake. If you measure resistance across a component that's still soldered into a circuit, you're measuring the component in parallel with whatever else is connected to those nodes. A 10k resistor sitting next to a 1k pull-down will read roughly 909 ohms, not 10k. Lift one leg of the component, or better yet, isolate the test point entirely for an accurate reading.
Get the Full Details
I ran into this exact problem last year on a power supply revision. The design called for 0.1 ohm sense resistors on the primary side, but my two-wire measurement kept reading 0.3 ohms. I assumed the resistors were defective and pulled three spares from the same reel. All three read the same. Only after desoldering one end did I discover the copper pour beneath the pads was providing a parallel path through the ground plane. The actual sense resistor was fine. The board layout was the issue, and it cost us two full prototype runs to catch.
The Details Beginners Miss
Resistance changes with temperature, and not everyone factors that in. Copper has a temperature coefficient of about 0.393% per degree Celsius. A trace reading 1.00 ohms at 25°C will read roughly 1.19 ohms at 75°C — a 19% shift that can look like a failure if you don't expect it. This matters especially when testing connectors or relay contacts that warm up during operation. Another thing nobody mentions: test current matters. A DMM in resistance mode typically sources between 0.1mA and 1mA depending on the range. Some components behave differently at different test currents. A varistor, for instance, will show wildly different readings at 0.5mA versus 5mA because its resistance is nonlinear by design. A forward-biased diode in the test path will also skew results since the meter is applying a small voltage that may not be enough to turn the diode on — or might forward-bias it partially. Understanding your meter's test conditions prevents misreads. For insulation resistance testing, especially on harnesses and motor windings, you need a megohmmeter or a DMM with a dedicated Hi-Fi/IR test function. Standard resistance mode won't apply enough voltage to detect leakage paths through contaminated PCBs or degraded wire insulation. A 500V DC hipot test reveals issues that a 0.5V ohmmeter reading completely misses. The industry standard for motor windings is typically 1 megohm minimum at 500V DC, though some specifications require 100 megohms for critical applications.
Common Pitfalls and When This Method Fails
Resistance testing cannot detect intermittent opens. A wire with a broken strand that makes contact only when bent will test fine at rest. If the application involves vibration or flexing, complement resistance testing with a micro-ohm shake test — slowly bend the harness or board while monitoring resistance in real time. Even a basic DMM with a relative mode and a 1-second update rate will show you the moment contact breaks. Another limitation: resistance testing won't catch partial shorts. If two traces are 50k ohms apart instead of open, your DMM in the megaohm range will read that, but if they're 500 ohms apart due to carbon tracking or conductive contamination, you might not notice unless you know what value to expect. Always compare against a known-good reference board when possible. I keep one unmodified prototype of every design I work on specifically for this reason. For high-current applications, also remember that contact resistance in connectors and terminals dominates the total resistance budget. A single mating cycle of a spade connector adds roughly 0.5 to 2 milliohms depending on quality. Over 50 cycles, that can double. If your design specifies a max voltage drop across a connector, measure it under load, not just at rest with a DMM. A 5-amp current through 10 milliohms of contact resistance produces 50mV of drop and 250mW of heat — negligible on a breadboard, significant in a sealed enclosure.
The bottom line is that resistance testing is fast and cheap when done correctly, but it only tells you about steady-state DC conditions. It won't replace functional testing, thermal cycling, or signal integrity analysis. Use it as a screening tool, not a validation method. I check resistance on incoming harnesses, during assembly, and before shipping, and it catches maybe 60% of the defects I'd otherwise ship. The rest show up in the field or during burn-in. That's the tradeoff — nothing is free.
