Electrical lab safety

An electrical measurements lab is where circuit theory meets real hardware. On paper, a resistor has exactly the value printed on it, a wire has no resistance and a voltmeter draws no current. On the bench, none of that is true, and the job of a measurement engineer is to know by how much it is not true and whether that matters. This course teaches the instruments, the habits and the arithmetic that let you trust a number you read off a meter or a screen. It starts where every lab starts: with how to work on energized circuits without getting hurt and without destroying the equipment.

In an ABET-accredited American electrical or computer engineering program, the measurements lab usually runs in the fall of the sophomore year, alongside Circuit Analysis and Digital Logic. It supplies the bench skills that every later lab assumes: Electronics I builds amplifiers on the breadboard, Signals and Systems samples them, and the power, control and communications courses all hand you an oscilloscope on the first day and expect you to use it well. ABET's student outcomes include the ability to develop and conduct appropriate experimentation, analyze and interpret data, and use engineering judgment to draw conclusions, and this course is where you first practice them.

The examples follow Cedarline Controls, the invented Madison, Wisconsin company from ELE-210, ELE-240 and ELE-250 that designs controllers for traffic signals and small machines. Amara Osei, an engineer intern, runs the company's test bench. This term she is verifying the first prototype boards of the pedestrian crossing controller: its 24 V supply, its LED signal drivers, its sensor inputs and the microphone amplifier of its accessible pedestrian signal. Leo Tremblay, a sophomore co-op student, has just joined her, and each lesson follows a task she gives him.

What electricity does to a body

Current, not voltage, is what injures. Voltage matters because it drives current through the resistance of the body, and that resistance varies enormously. Dry, intact skin may present tens of thousands of ohms at low voltage; wet or broken skin, or a contact through a cut, can drop the total hand-to-hand or hand-to-foot path to around 1,000 Ω or less. Above roughly 50 V the skin itself can break down, so the resistance falls just when the voltage rises. The current that results is set by Ohm's law, I = V/R, and its effects depend on its size, the path it takes, its frequency and how long it lasts.

Two things in that table deserve attention. First, the dangerous range starts at tens of milliamperes, which is a tiny fraction of the current a lab supply or a wall outlet can deliver. A 120 V outlet behind a 15 A breaker will happily push 100 mA through you without tripping the breaker, because breakers protect wiring, not people. Second, the let-go threshold is lower than the fibrillation threshold. A person who grips a live conductor and cannot let go stays in contact, the skin heats and its resistance falls, the current rises and the exposure lengthens. That is why a ground-fault circuit interrupter, which trips at about 5 mA of imbalance between line and neutral, is fitted to bench outlets in many teaching labs.

Hazards beyond shock

Shock is the hazard everyone thinks of first, but the lab has others. An arc flash happens when current jumps through air between conductors, usually because a tool, probe tip or dropped screw shorts two points with a large available fault current behind them. The arc can reach temperatures of thousands of degrees in milliseconds, vaporize copper, throw molten metal and cause severe burns at a distance. A student lab bench supply limited to a few amperes cannot produce a serious arc flash, but a building's 600 V or 480 V distribution panel can, and the measurement habits you build on the bench go with you to those places. Burns also come from ordinary heat: a resistor dissipating a watt or two can reach well over 100 °C, and a soldering iron tip runs at 300 to 400 °C.

Stored energy is the hazard most often forgotten. A capacitor holds charge after the supply is off. The bulk capacitor in a mains-powered supply can hold hundreds of volts for minutes, and a large electrolytic on a low-voltage board can deliver enough current to weld a screwdriver tip. Electrolytic capacitors connected backwards, or run above their voltage rating, can vent violently. Lithium batteries can catch fire if shorted, punctured or overcharged. Finally, there are ordinary workshop hazards: flying wire clippings, which is why eye protection belongs in an electronics lab, solder fumes from flux, which need ventilation, and trailing cables across the floor.

Rules that prevent most incidents

Most lab injuries and most destroyed instruments come from a short list of mistakes, and a short list of habits prevents them. Wire the circuit with the supply off and connect the supply last. Check the wiring against the schematic before applying power, preferably with a second person. Set the current limit on the bench supply before you connect the load. Keep one hand away from the circuit when probing anything above about 30 V, so that a current path through the chest is less likely. Remove rings, watches and metal bracelets. Never work alone on circuits connected to mains voltage. Discharge capacitors through a suitable resistor, and check with a meter that they are discharged, before touching the board. When something smells hot or smokes, switch the supply off first and investigate second.

These habits are not just good practice. OSHA's electrical rules require employers to protect workers from electrical hazards and to train qualified persons, and NFPA 70E, Standard for Electrical Safety in the Workplace, sets out how to assess shock and arc flash risk and choose protective equipment for work on or near energized equipment. Cedarline's test lab follows a written procedure based on NFPA 70E, and Leo's first task is to read it and sign that he has.

Measurement categories and meter ratings

A meter is part of the circuit while you measure, and a meter that fails can become the shock or arc hazard. IEC 61010-1, the international safety standard for measurement equipment, sorts measurement locations into categories according to the size of the voltage transients, or brief spikes, that can appear there. Lightning strikes and switching of large loads send transients down power lines. Near the service entrance, where the utility supply arrives, those transients are large and the fault current available is huge. Inside a piece of equipment plugged into an outlet, they are smaller and limited by the wiring in between.

Each meter and each set of test leads is marked with a category and a voltage, for example CAT III 600 V, and the rating applies to the whole measuring set: a CAT IV meter with CAT II leads is a CAT II measuring set. The category matters more than the voltage. A CAT III 600 V meter is designed to survive a larger transient than a CAT II 1000 V meter, because the category reflects what the meter can survive in that location. A properly rated meter also has fuses on its current inputs that can interrupt a high fault current. One of the most common serious meter accidents is leaving the leads in the current jacks, where the meter is nearly a short circuit, and then touching them across a voltage source. A cheap fuse may explode instead of interrupting the arc.

Lockout and verifying the circuit is dead

Work on mains-connected equipment begins by making it safe. Switch it off, unplug it or lock out its breaker, and then prove it is dead with a meter you have just checked on a known live source. This live-dead-live test catches the dangerous case where the meter itself has failed, a blown fuse or a broken lead, and reads 0 V on a live circuit. In a teaching lab you will rarely work inside mains equipment, but the habit transfers directly: before touching a board, confirm with the meter that its supply rail reads zero and that its large capacitors have discharged.

# Body current and capacitor discharge estimates for lab safety checks
def body_current(volts, ohms):
    return volts / ohms

thresholds = [(0.001, "perception"), (0.010, "let-go threshold region"),
              (0.050, "fibrillation possible")]

def describe(i):
    label = "below perception"
    for limit, name in thresholds:
        if i >= limit:
            label = name
    return label

for v in (24, 48, 120):
    for r in (1500, 50000):
        i = body_current(v, r)
        print(f"{v:4d} V through {r:6d} ohms: {i*1000:7.2f} mA  ({describe(i)})")

# A 470 uF capacitor charged to 35 V, bled by a 10 kilohm resistor
import math
C, V0, R = 470e-6, 35.0, 10e3
tau = R * C
t_safe = tau * math.log(V0 / 2.0)    # time to fall below 2 V
print(f"Time constant {tau:.1f} s; below 2 V after {t_safe:.1f} s")
print(f"Stored energy {0.5*C*V0**2:.3f} J")

  24 V through   1500 ohms:   16.00 mA  (let-go threshold region)
  24 V through  50000 ohms:    0.48 mA  (below perception)
  48 V through   1500 ohms:   32.00 mA  (let-go threshold region)
  48 V through  50000 ohms:    0.96 mA  (below perception)
 120 V through   1500 ohms:   80.00 mA  (fibrillation possible)
 120 V through  50000 ohms:    2.40 mA  (perception)
Time constant 4.7 s; below 2 V after 13.5 s
Stored energy 0.288 J

The capacitor part of the program matters for Cedarline's board. Its 24 V input has a 470 µF bulk capacitor rated at 35 V, and the board has no bleeder resistor. With nothing connected, the capacitor can hold its charge for a long time. Amara adds a 10 kΩ resistor to the bench fixture so the voltage falls below 2 V in about 13 seconds, and she tells Leo to read the rail with a meter before he touches the board after a test. The energy stored is under a third of a joule, too little to hurt a person, but more than enough to vaporize a fine probe tip if it is shorted.

Protecting the equipment

Safety in a lab also means protecting the instruments, the circuit and the measurement. Most damage comes from three causes. The first is connecting a meter in the wrong mode, especially measuring voltage with the leads in the current jacks. The second is exceeding an input rating: an oscilloscope input may be rated for a few hundred volts at low frequency but much less at high frequency, and a 1X probe gives no protection at all. The third is ground: on most bench oscilloscopes the outer shell of every input connector is tied to the mains safety ground. Clipping the probe's ground lead to a point that is not at ground potential creates a short circuit through the scope's ground wiring. Lesson 7 returns to this, because it is behind many destroyed boards and probes. Static electricity is the fourth, quieter cause: a few kilovolts from a walk across a carpet can damage a MOSFET gate without any visible spark, so sensitive parts are handled on a grounded mat with a wrist strap.