Measurement The multimeter is the tool that converts guessing into knowing, and it is the one piece of kit worth spending money on besides the iron. Most of its value comes from four modes and one habit.

buy this firstmethod included

Tier 0 · Groundthe multimeter

Four modes that cover almost everything

Module 0.1

Voltage, current, resistance, continuity

How to connect, and what each is really telling you

Measurement	Connect	Power	Note
Voltage	Parallel — across the thing	On	90% of what you will ever do
Current	Series — break the circuit, meter in the gap	On	Move the red lead to the A socket first
Resistance	Across the component	Off	Lift one leg, or other paths skew the reading
Continuity	Probe two points	Off	The beep — best friend for breaks and shorts
Diode test	Across a diode, both ways	Off	Shows Vf one way, open the other

Why voltage is measured in parallel and current in series: a voltmeter has very high internal resistance so it barely disturbs the circuit, while an ammeter has near-zero resistance so it can be inserted into the path. That difference is exactly why swapping them is destructive.

The classic meter mistake. Leaving the leads in the current sockets and then measuring voltage places a near-short across your supply — blown fuse at best, damaged meter or supply at worst. Build the habit now: after any current measurement, immediately return the red lead to the voltage socket. Nearly everyone does this once; the habit is what stops it being twice.

Buying advice: auto-ranging saves constant fiddling, a continuity beep is essential, and the meter must be fused on the current ranges — an unfused cheap meter is a genuine hazard. Anything mid-range from a known brand is ample; you are not doing calibrated metrology.

Module 0.2

What an oscilloscope adds — and why not yet

Honest advice about the expensive tool

A multimeter shows you a value averaged over time. A scope shows you voltage against time, which reveals everything a meter smooths away: ripple on a supply, the shape of a PWM signal, ringing when a transistor switches, the bounce of a switch contact, and whether a bus is actually clocking.

Why to wait: a scope answers questions you cannot yet frame. Buying one on day one usually means an unused instrument and less money for a decent iron. The useful sequence is multimeter → build things → hit a problem the meter cannot explain → then buy the scope, knowing what you need it for.

Cheaper interim option: a USB logic analyser costs around £10 and decodes UART, I²C and SPI directly. For microcontroller work — which is most of what you will do — it answers more of your real questions than a budget scope, because your problems will be digital protocol problems rather than analogue waveform problems.

Checkpoint 0I can measure voltage, current, resistance and continuity correctly, I understand why the connections differ, and I have the lead-socket habit.

Tier 3 · Buildmethod

A systematic measuring method

Module 3.1

Walking the circuit

This works before you own a scope, and beats swapping parts

1Measure the supply rail, under load. Not the wall adapter's label — the actual voltage at the board while it is running. A surprising number of faults end here.

2Check ground continuity everywhere it should exist. Beep from the board's ground to the supply's negative, and between any two boards that must talk.

3Predict, then measure, each node. Write down what theory says each point should be, then measure. Work input to output.

4The first node that disagrees with your prediction is where the fault is, or immediately before it.

5Only then suspect a component. Most faults are wiring, power, or a wrong assumption.

The step people skip is the prediction. Measuring without a prior expectation gives you numbers but no verdict — 2.1V means nothing until you have decided it should have been 3.3V. Writing the expected values on the schematic before you probe is what turns measurement into diagnosis.

Two specific techniques worth knowing

Measuring current draw as a health check: break the supply line and put the meter in series. Knowing your circuit's normal current means an abnormal figure instantly tells you something is shorted (too high) or not running (too low).
Voltage drop across a known resistor as an indirect current measurement — no need to break anything. Measure across it, divide by its value. This is how you check LED current without disturbing the circuit.

Drill 1

A circuit works when probed with the meter but fails when you take the probes away. What does that suggest?

Something is floating. A voltmeter has very high but not infinite resistance, which can be enough to pull a genuinely undefined node to a stable level — so the act of measuring changes the result. It points at a missing pull-up or pull-down, an unconnected input, or a broken track. This class of fault, where observation changes behaviour, is confusing until you have seen it once.

Checkpoint 1I predict node voltages before probing, walk a circuit systematically, and can measure current both directly and via a known resistor.