Continuity testing is one of those things you think you know until you're staring at a dead board at 11 PM.
The multimeter has been doing this for decades. You put the probes on two points and listen for a beep or watch the needle move. That's the basic idea. The actual execution is where people mess up. I'm going to walk through what actually works and what goes wrong when you're in the field. Start by setting the dial to the continuity symbol, which looks like a series of sound waves or a diode symbol depending on your meter. Most modern digital multimeters will have this as a dedicated setting. If yours doesn't, the resistance ohms setting works too, though you won't get the beep and you'll need to interpret the numbers yourself. Touch the probes together first. You should hear a beep or see near-zero resistance on the display. This is called verification and skipping it is the most common mistake I see. People skip it because they're confident their meter works, then spend twenty minutes chasing a dead battery in the multimeter before realizing the problem was never the circuit.
Once verified, power down the circuit completely. I can't stress this enough. Testing continuity on a live circuit will damage your multimeter and possibly the board you're working on. There's no room for ambiguity here. Remove batteries, unplug power supplies, and wait a few seconds for capacitors to discharge. On larger boards with power supply capacitors, that wait time should be closer to thirty seconds. Place one probe on each test point. Keep light pressure. The metal tips of your probes need good contact with the trace or pad you're measuring. Alligator clip test leads help here if you're working on a crowded board where holding both probes steady with your fingers is fighting an uphill battle. One hand on each probe is fine for quick checks, but anything more involved and your hand will shake within a minute. If the circuit is continuous, the meter beeps or shows very low resistance, typically under fifty ohms. Some meters will show a range from zero up to roughly two hundred ohms and still count as a valid connection depending on the wire gauge and trace width. A thick copper trace on a PCB might read three ohms. That's still continuous. A thin wire spanning several feet might read ten ohms. Also continuous. The beep tells you continuity exists. The resistance number tells you how much resistance is in that path, which is a separate question.
Here's something most guides don't mention. Continuity testing on a PCB with ground planes can give you false positives. You might probe two points that appear connected because they're both tied to the same ground layer, but there's no actual signal path between them. The beep means electrical continuity, not functional continuity. I learned this the hard way on a custom audio board where a ground plane was acting as a return path for everything. Measured continuity everywhere, spent hours tracking down why the circuit wouldn't work, and the issue was a broken trace on a different layer that the continuity test couldn't see because the ground plane was masking it. For multi-layer boards, continuity testing from the top side alone is insufficient. You need to either access test points on the other layer or desolder one component to isolate the trace you're checking. This adds maybe fifteen to twenty minutes to your troubleshooting time but saves you from the scenario where your meter says everything is fine and the circuit still doesn't work.
Get the Full Details
Common problems and edge cases
Breakout traces in printed circuit boards are the continuity killer. A trace looks intact on the surface but has cracked underneath the solder mask from thermal stress or flexing. The resistance reading will jump around as you move the probe slightly, settling at a value somewhere between continuity and open circuit. If you're seeing erratic readings in what should be a solid connection, switch to resistance mode and look at the number rather than relying on the beep. The beep threshold on most meters is fixed and won't warn you about marginal connections. Another issue is testing through solder joints. Fresh solder gives reliable contact. Old, cold, or cracked solder joints create high-resistance connections that the continuity beep might not catch. I usually scrape a tiny spot on the pad with a hobby knife to remove oxidation before probing, which takes about ten seconds per point and eliminates one variable from the equation. When you're testing continuity through connectors rather than directly on traces, remember that connector pins can have internal resistance from corrosion or loose contacts. A reading of eighty ohms through a connector might mean the connector itself is degrading, not that your wire is broken. If the number seems high, disconnect the connector and test the wire separately to isolate where the resistance is coming from.
High-frequency circuits behave differently during continuity tests. The multimeter uses a low test current, typically one to five milliamps, at DC. A trace that measures as continuous at DC might have significant impedance at RF frequencies due to parasitic inductance and capacitance. This only matters if you're troubleshooting high-speed digital or RF designs. For power supply traces and general-purpose wiring, the DC continuity test is accurate enough. I once spent an afternoon troubleshooting a motor control board where the continuity test showed every trace was fine. The problem turned out to be a thermal pad under a MOSFET that had cracked during reflow. The component was electrically continuous but thermally isolated, causing the part to overheat and fail within seconds of powering up. Continuity testing doesn't tell you about mechanical integrity or thermal paths. It only tells you whether electrons can flow between two points at DC. Knowing what the test cannot do is as important as knowing what it can do.
What the resistance number actually means
When your multimeter shows a resistance value during a continuity check, pay attention to it. A reading below five ohms is a solid short. Between five and fifty ohms is acceptable for most power and signal traces. Above one hundred ohms on a trace that should be a direct connection usually indicates a problem, even if your meter still beeps. Some meters have a beep threshold around one hundred to two hundred ohms, which means they'll beep for connections that are borderline at best. If you're checking continuity on long wires, expect resistance proportional to length. Twenty-gauge wire has approximately five ohms per thousand feet. A fifty-foot run of that wire will read about two and a half ohms. That's normal. Don't misread it as a fault just because the number isn't zero. Zero resistance only exists in theoretical models and superconductors. Real copper has real resistance. The test lead resistance itself adds to your measurement. Standard multimeter leads add about zero point two to zero point five ohms. If you need precision, measure the leads by touching them together, note the reading, and subtract that from your subsequent measurements. On most bench work this correction is unnecessary. On production testing where you're verifying wire harnesses that need to be near-zero resistance, it matters. I keep a small card with my lead resistance printed on it taped to my multimeter case so I don't have to think about it each time.
If your multimeter has a relative mode, use it. Press the delta or relative button after touching the probes together and the meter will zero out the lead resistance. This is cleaner than writing numbers on a card and doing mental subtraction. The relative function stores the lead resistance value in memory and subtracts it from every reading until you press the button again. Be careful not to bump the button during testing, which resets the reference and throws off all your subsequent measurements.
When continuity testing fails
There are situations where a multimeter continuity test simply won't work. SMD components on densely packed boards are nearly impossible to probe without damaging nearby parts. The pads are too small, the spacing is too tight, and probe slip can create bridges between adjacent pins that weren't meant to be connected. In these cases, a continuity test from the component side of the board or using fine-pitch probe tips is the only option, and even then you're working with reduced accuracy. Optoisolators and semiconductor junctions will also throw off continuity readings. If you probe across a diode, LED, or transistor junction, the multimeter might show an open circuit in one direction and a low resistance in the other, or sometimes no beep at all. This is normal behavior for semiconductors. The test current from the multimeter isn't enough to forward bias many junctions. If you're trying to verify whether a component is connected correctly, remember that semiconductors don't behave like wires and the continuity test isn't designed to characterize them. For boards with active components still soldered in place, continuity tests can feed current through unintended paths. A trace that appears open because a diode is blocking the current might actually be fine. Desolder at least one leg of the component being tested to isolate it from the rest of the circuit. This is standard practice and it takes about thirty seconds per component, which is faster than diagnosing why your readings don't match the schematic.
If you're dealing with a board that has multiple power domains or isolated grounds, continuity between those domains might be intentionally absent. Verify the schematic before assuming a missing continuity reading indicates a fault. I've corrected my own assumptions twice in the last year because I forgot to check whether two ground points were supposed to be connected or were separated by an isolation barrier.
Practical workflow
Set up your workspace with good lighting and a magnifying lamp if you're working on small components. A cheap LED desk lamp pointed at the right angle reveals traces and solder joints that are easy to miss in ambient light. Place your multimeter within easy reach so you're not rotating the dial blindly while holding probes in your other hand. The dial should be set to continuity before you touch the board, not after. Work systematically. Start from the power input and trace each net outward. Mark components or traces as you verify them with a pencil or digital checklist. This prevents double-testing and helps you track where you stopped when you need to come back later. A clean checklist is faster than a messy one because you're not guessing whether you already checked a particular trace. Document unusual readings. If a trace reads forty ohms when it should read near zero, note it. That forty ohms might be a cold solder joint, a partial trace break, or an intentional current-sensing resistor you forgot about. Writing the number down prevents you from forgetting it when you come back to the board three days later.
The multimeter is a fundamental tool and continuity testing is one of its simplest functions, but simple doesn't mean effortless. The difference between a quick diagnosis and a frustrating afternoon usually comes down to whether you verified your meter first, powered down the circuit properly, and understood what the beep actually represents. Everything else is detail work.
