How to use a multimeter: the sequence that protects the meter and the reading

As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.

The order of operations that keeps the meter alive

Write the sequence out once and it becomes automatic.

• Name the quantity first. Volts, amps, ohms, continuity, capacitance, frequency, temperature. Saying it out loud prevents the commonest error, which is measuring the right thing with the wrong function selected.
• Set the dial, then set the jacks. These are two separate acts and both have to agree. A dial on amps with leads in the voltage jacks reads nothing. A dial on volts with leads in the current jacks reads nothing and then blows a fuse the moment you find real voltage.
• Predict the answer before you look. Roughly twelve volts. Roughly 230. A few ohms. This single habit catches reversed polarity, the wrong decade and the wrong function, because a prediction that misses by a factor of a thousand tells you the setup is wrong rather than the circuit.
• Connect the black lead first, remove it last. On anything with real energy behind it this limits what happens if a probe slips.
• Read, then disconnect, then return the dial to a voltage range. Leaving the instrument parked in a safe state is what protects the next measurement.

Naming the quantity is less trivial than it sounds on meters that measure more than the basics. The Fluke 116 publishes a built in thermometer covering -40 to 400 degrees Celsius and a microamp range specifically for flame sensor work, so on that instrument the first decision genuinely includes temperature and microamps rather than just the usual three.

Volts in parallel, amps in series, and the fuse that pays for mistakes

The two connection rules are consequences of how each function is built, which is why they cannot be bent.

A voltmeter has to sit across two points without changing what happens between them, so its input impedance is made very high, commonly around ten megohms. At that impedance it draws microamps and the circuit behaves as if nothing were attached. You can therefore clip it anywhere, in parallel with anything, and it is safe.

An ammeter has to carry the current it is measuring, so it must become part of the loop, and to avoid disturbing the circuit its resistance is made very low, often a small fraction of an ohm. That is the whole problem. Place a deliberately near zero resistance across a battery or a mains supply and the only things limiting the current are the internal shunt and the fuse. A meter in current mode across a car battery is a bolt spanner across the terminals with a fuse in series.

Two practical consequences follow. Break the circuit and insert the meter in the gap, which means the measurement needs a physical interruption somewhere, and plan where that will be before you start. And understand your own instrument’s fusing, because the milliamp jack and the high current jack are frequently protected differently. If breaking the circuit is awkward or the current is large, measure the magnetic field instead, as described in the comparison of multimeters and clamp meters.

Reading the accuracy specification off your own display

Accuracy on a digital meter is written as a percentage of the reading plus a number of counts, and both halves matter. The percentage term scales with the measurement; the counts term is a fixed number of least significant digits and dominates at the bottom of a range.

Work an example. Suppose a specification of plus or minus 0.015 percent of reading plus 2 counts, and a displayed value of 10.0000 volts on a bench instrument resolving to 100 microvolts. The percentage term contributes 1.5 millivolts. The counts term contributes two steps of 100 microvolts, so 0.2 millivolts. Total uncertainty about 1.7 millivolts, and the last displayed digit is therefore not meaningful on its own. Now take the same specification at 0.1000 volts on the same range: the percentage term is 15 microvolts while the counts term is still 200 microvolts, so the fixed term is now thirteen times larger than the proportional one. This is why you always select the lowest range that will hold the reading.

The bench figures come from the Fluke 8808A, which states a five and a half digit dual display and a basic direct voltage accuracy of 0.015 percent. The point of the table is the pattern rather than the numbers: handheld listings routinely publish resolution and omit accuracy, and resolution is the specification that looks impressive while accuracy is the one that decides whether the reading means anything. Choosing between instruments on that basis is the subject of what true RMS means and our handheld meter notes.

Continuity, resistance and the habit of shorting the leads first

Touch the probes together before any resistance measurement and read what the display says. It will not be zero. Ordinary test leads contribute somewhere around a tenth to half an ohm, plus whatever the probe contact adds, and that offset is in every subsequent reading.

Against a 10 kilohm resistor it is irrelevant. Against a 0.5 ohm shunt, a motor winding or a battery cable it is most of the measurement. Either subtract the lead resistance by hand or use the relative or zero function if the meter has one, which stores the current reading as the new zero.

Where lead resistance genuinely cannot be tolerated, the answer is four wire measurement: one pair of leads carries the test current and a separate pair senses the voltage right at the component, so the voltage drop in the current leads never enters the result. The Fluke 8808A publishes a four wire ohms technique along with a dedicated direct current leakage measurement, which is the kind of capability that distinguishes a bench instrument from a handheld rather than a specification handhelds do badly.

Two further traps on the resistance function. The circuit must be isolated, because the function works by injecting a known current and any external voltage corrupts it. And a component measured in place reads as itself in parallel with every other path across it, so the number will be lower than the component’s true value and the only fix is to lift one end.

Measuring current without breaking into the circuit

A clamp sidesteps the series problem entirely by measuring the magnetic field a current creates. Nothing is interrupted, nothing flows through the instrument, and the current range jumps by two orders of magnitude.

The POROMETISTO CM06 clamp publishes 600 amps alternating and direct current, 600 volts alternating and direct, true RMS measurement, capacitance, frequency, duty cycle, resistance, continuity and temperature, plus non contact voltage detection and a low impedance mode to suppress stray voltage readings. It also publishes an inrush current function with a measurement time of about 100 milliseconds, stated as applying to alternating current only.

That inrush figure is worth understanding because it is a time domain measurement hiding inside a meter. A motor starting draws several times its running current for a fraction of a second while the rotor is stationary and there is no back electromotive force to oppose the supply. An ordinary averaging meter smooths that peak away and shows something close to the running current, so a breaker that trips on starting looks like a mystery. A function that captures a 100 millisecond window catches the peak and turns the mystery into a number. The variable frequency drive mode in the same listing solves a related problem by filtering the high frequency switching content that otherwise makes a drive output unreadable.

Building the sequence into muscle memory

Five habits cover almost every mistake people make with these instruments.

• Park the meter on a voltage range when you put it down. The next person to pick it up, including you, will assume that is where it is.
• Predict every reading before you take it. The prediction is the error check, and it costs nothing.
• Short the leads before any resistance or continuity work. It also confirms the leads themselves are not broken, which is a surprisingly common fault.
• Prove the meter on a known live source before trusting a zero. A flat battery, a blown fuse or a broken lead all look exactly like a dead circuit.
• Select the lowest range that holds the reading. The fixed counts term in the accuracy specification punishes you for sitting at the bottom of a wide range.

None of that is about the instrument being expensive or cheap. A disciplined sequence on a modest meter produces trustworthy numbers, and a careless sequence on a reference grade one produces confident nonsense, which is considerably more dangerous. Our notes on mid range handhelds go through what the listings in that class actually publish.

Frequently Asked Questions