Electrical installations
Here electrical technology stops being a calculation and becomes a responsibility. Each component of an installation protects something different, and mixing up what protects what is the origin of most electrical accidents.
01What protects what
This is the idea that organizes the whole topic, and it is worth fixing in mind before any calculation:
| Device | Protects | Against |
|---|---|---|
| Thermal-magnetic breaker | The conductor | Overload and short circuit. Keeps the cable from heating beyond what its insulation can withstand. |
| Fuse | The conductor and the equipment | Mainly short circuits. It is fast and has a high breaking capacity. |
| Residual-current device (RCD) | The person | Direct or indirect contact: current that flows to ground. |
| Grounding (earthing) | The person | Metal enclosures ending up live after an insulation fault. |
| Thermal overload relay | The motor | Prolonged overload, phase loss, locked rotor. |
| Motor protection circuit breaker | The motor and its supply line | Both: adjustable thermal + magnetic trip, in a single device. |
| Surge arrester | The equipment | Transients caused by lightning or switching. |
The thermal-magnetic breaker looks after the cable; the RCD looks after the person. They are independent and neither replaces the other: an installation can have perfect circuit breakers and still electrocute someone, and it can have an RCD and still catch fire because of an undersized cable. That is why the standard requires both.
02Selecting conductors
Choosing the cross-section of a cable means answering three questions, and all three must be satisfied: the largest of the resulting cross-sections is the one you use.
The cable must not exceed the maximum temperature of its insulation (70 °C for PVC, 90 °C for XLPE). The value depends on the installation method: it is not the same in free air as inside a conduit with other circuits.
The voltage at the farthest point must remain acceptable. The AEA (Argentine Electrotechnical Association) sets 3 % for final circuits and 5 % from the meter. On long runs, this condition takes precedence over the previous one.
Regardless of the calculation: 1.5 mm² for lighting, 2.5 mm² for outlets and 4 mm² for special circuits. Never less, even if the calculation allows it.
| Cross-section | Approximate current-carrying capacity | Typical protection | Common use |
|---|---|---|---|
| 1.5 mm² | 15 A | 10 A | Lighting circuit (IUG) |
| 2,5 mm² | 21 A | 16 A | General-purpose outlets (TUG) |
| 4 mm² | 28 A | 20 to 25 A | Special circuits, electric water heater, air conditioner |
| 6 mm² | 36 A | 32 A | Supply to sub-distribution boards |
| 10 mm² | 50 A | 40 A | Main line of a large dwelling |
Indicative values for PVC-insulated copper in conduit. The table that counts is always the one in the AEA 90364 standard for the specific installation method, together with the cable manufacturer’s.
Single-phase 220 V circuit, 16 A, with the farthest outlet 20 m from the distribution board.
- By current: 16 A fits within 2.5 mm² (21 A allowable).
- By voltage drop with 2.5 mm²: ΔV = 2 × 20 × 16 × 0.0175 / 2.5 = 4.48 V, which is 2.04 % of 220 V. It meets the 3 % limit.
- Had 1.5 mm² been used: ΔV = 11.2/1.5 = 7.47 V = 3.4 %. It does not comply, and besides, that cross-section is not permitted for outlets.
- Minimum regulatory cross-section for TUG: 2.5 mm².
Result: 2.5 mm² with 16 A protection. And if the same circuit were 45 m long, the drop with 2.5 mm² would be 10.1 V (4.6 %): you would have to move up to 4 mm².
The condition is Iload ≤ Iprotection ≤ Icable allowable. Putting a 25 A thermal-magnetic breaker on a 1.5 mm² cable leaves the cable unprotected: it can heat up enough to damage the insulation without the breaker ever tripping. It is the most dangerous and most common mistake in improvised installations, and the reason why replacing a breaker with a larger one because it “keeps tripping” is exactly what you must not do.
03Overcurrent protection
A thermal-magnetic breaker has two mechanisms inside, each for a different problem:
Acts on overloads: currents somewhat above the rated value, sustained. It is slow on purpose—seconds to minutes—to tolerate motor starting. It depends on the ambient temperature: in summer it trips sooner.
Acts on short circuits: currents tens of times the rated value. It cuts in milliseconds, before the energy of the arc can damage the installation.
| Curve | Trips between | For what load |
|---|---|---|
| B | 3 a 5 · In | Resistive loads: lighting, heating, outlets without motors. |
| C | 5 a 10 · In | General use. Tolerates the starting current of small motors and switch-mode power supplies. |
| D | 10 a 20 · In | Large motors, transformers, welders: loads with huge starting peaks. |
- The breaking capacity (3 kA, 6 kA, 10 kA) is the maximum short-circuit current the device can interrupt without being destroyed. It depends on how close the utility’s transformer is: the closer, the more fault current available.
- Selectivity means that on a fault only the protection closest to the problem acts, not the main one. It is achieved by grading currents and times: the main one always larger and slower than the branch ones.
The fuse is faster on large short circuits and has a higher breaking capacity, but it has to be replaced and someone always ends up fitting one of higher rating—or a piece of wire. The thermal-magnetic breaker can be reset, but it ages with every trip. In motor circuits both are used: aM-type fuses for the short circuit and a thermal overload relay for the overload.
04The residual-current device and grounding
It is the only device designed to protect the person, and it works on a very simple idea: all the current that enters through one conductor has to return through the other. If it does not return, it is going somewhere else, and that somewhere can be a person.
| Sensitivity | What it protects | Where it is used |
|---|---|---|
| 30 mA | People: direct and indirect contact | Mandatory in every residential installation, on all circuits. |
| 10 mA | People, with an extra margin | Bathrooms, laboratories, operating rooms, children’s circuits. |
| 300 to 500 mA | Property: prevents fires caused by leakage current | Head of industrial installations, selective with the 30 mA ones. |
Because it is the threshold below which a 50 Hz current passing through the body does not cause ventricular fibrillation in the time it takes the protection to act. For a sense of scale: 1 mA can be felt, 10 mA already prevents letting go of the conductor (tetanization), 30 mA affects breathing and from 50 mA there is a risk of fibrillation. The RCD does not prevent the shock: it prevents it from lasting long enough to kill.
It connects all exposed metal parts—a motor housing, a refrigerator, a distribution board enclosure—to a buried ground rod. If a line conductor touches the casing, the current finds that path instead of waiting for a person, and the RCD detects the leakage and trips.
Without grounding, the RCD still protects—it acts when the person touches—but the first line of defense is lost: the fault is not detected until someone finds it with their body.
Every RCD has a T button that creates an artificial leakage through an internal resistor. It must be tested once a month: the mechanism is electromechanical and can stick over the years without giving any sign.
If pressing T does not trip it, the RCD is faulty and must be replaced. That it “has never tripped” does not mean it works.
In the Argentine TT scheme, the neutral is grounded at the utility’s transformer, not at the house. Bonding neutral to ground at the user’s board makes the return current split between the two paths: the RCD sees a permanent difference and trips, or—worse—stops protecting. The green-yellow protective conductor carries no current in normal operation: only during a fault.
05Switching and motor control
A motor is not operated with an ordinary switch: you must be able to start and stop it from several places, protect it from overload and guarantee that it does not restart on its own when power returns. The contactor and the thermal overload relay take care of that, governed by a control circuit separate from the power circuit.
| Element | Function | Practical detail |
|---|---|---|
| Contactor | Switch operated by an electromagnet | Its coil (220 V, 24 V) is controlled with low-current pushbuttons. When the voltage is lost, it opens: hence the protection against restart. |
| Auxiliary contacts | Signaling and holding | NO (normally open) and NC (normally closed). The NO in parallel with the start pushbutton is the holding contact. |
| Thermal overload relay | Protecting the motor from overload | It is set to the motor’s nameplate current. Its NC contact is in series with the contactor coil. It also detects the loss of a phase. |
| Motor protection circuit breaker | Thermal + magnetic in one device | Replaces the fuse + thermal relay combination and is adjusted with a dial. |
| Pushbuttons | Start and stop commands | Start NO (green) and stop NC (red). Stop is always normally closed: if the pushbutton wire is cut, the motor stops, which is the safe outcome. |
| Interlock | Preventing impossible operations | In a reversing circuit, each contactor opens the other’s circuit with an NC contact: the two can never close together and cause a short circuit between phases. |
They are drawn separately, and that separation is the key to understanding any industrial panel. The power diagram carries the three phases, the motor protection breaker, the main contacts of the contactor, the thermal overload relay and the motor: large currents and thick lines. The control diagram works with one phase and neutral (or 24 V) and contains pushbuttons, auxiliary contacts, the coil and the indicator lamps. The same contactor appears in both diagrams, identified with the same reference —KM1, for example. Drawing standards are covered in schematic circuit diagrams.
06In the lab
Set up a long run of thin cable (for example 30 m of 1 mm² on a reel) feeding a known load. Measure the voltage at the source and at the far end with the load connected and disconnected. Compare the difference with the calculated value. It is the check that the formula is not a paper formality.
On a teaching board with a 30 mA RCD, press the test button and verify the trip. Then, with a calibrated high-value resistor between line and ground (for example 10 kΩ, which gives 22 mA), check that it does not trip, and with 4.7 kΩ (47 mA) that it does trip. Measure the approximate time with a stopwatch. The whole lab is done on a protected teaching board and under supervision.
Wire the control circuit of Figure 2 with one contactor and two pushbuttons, without connecting the motor. Verify the holding contact, then remove the supply and restore it: the contactor must not pull in again on its own. Finally, add the NC contact of the thermal overload relay in series and simulate its trip.
With two contactors, wire the reversing circuit by swapping two phases on the second one. Add the electrical interlock—the NC of each in series with the other’s coil—and check that it is impossible to energize both at once. Only then connect the motor and test.
- Disconnect all voltage sources.
- Lock out the disconnecting device and tag it.
- Verify that voltage is absent with an instrument, and verify before and after that the instrument works.
- Ground and short-circuit where applicable.
- Delimit and mark the work area.
They are applied in that order and without skipping any. The third is the one most often omitted and the one that prevents the most accidents.
07Common mistakes
| Symptom | Usual cause |
|---|---|
| “It kept tripping, so I fitted a bigger breaker” | The cable was left unprotected. If the breaker keeps tripping, the circuit is overloaded or there is a fault: split the load or increase the cross-section, never enlarge the breaker. |
| The RCD trips for no apparent reason | Real leakage accumulated from several devices (each switch-mode power supply contributes some), moisture in a junction box, or a neutral bonded to ground downstream. |
| The RCD does not trip when the test button is pressed | It is faulty. Replace it: it cannot be repaired. |
| You get a shock touching the enclosure of an appliance plugged into an outlet | Insulation fault without grounding. The appliance has voltage on its casing and waits for someone to touch it. |
| The lamp at the far end of the shed is dimmer | Voltage drop from an insufficient cross-section on a long run. |
| A motor’s breaker trips on starting | Wrong curve: a C or D curve is needed to tolerate the starting peak. |
| The motor does not start and the thermal relay looks “fine” | Thermal overload relay tripped and not reset, or its NC contact open in the control circuit. |
| When power comes back, the machines start by themselves | The control has no contactor holding circuit, just an ordinary switch. It is a dangerous situation, which is why control is always done with a holding contact. |
| Short circuit when reversing the rotation | The interlock between the two contactors is missing. |
08Self-assessment
What does a thermal-magnetic breaker protect and what does an RCD protect?
The thermal-magnetic breaker protects the conductor against overload and short circuit. The RCD protects the person against currents diverted to ground. They are different functions and both are needed.
A 25 m, 20 A, single-phase 220 V circuit with 2.5 mm² cable: does it meet the voltage-drop limit?
ΔV = 2 × 25 × 20 × 0.0175 / 2.5 = 17.5/2.5 = 7 V, which is 3.18 %. It exceeds the 3 % allowed: you have to move up to 4 mm², which brings the drop down to 4.4 V (2.0 %).
Why can’t you put a 25 A breaker on a 1.5 mm² cable?
Because that cable carries about 15 A: current above its limit could flow for hours without the protection acting, degrading the insulation until it causes a fire. The condition is Iload ≤ Iprotection ≤ Iallowable.
What is the difference between curves B, C and D?
The magnetic trip threshold: B between 3 and 5 times the rated current, C between 5 and 10, D between 10 and 20. The larger the starting peak of the load, the higher the curve must be.
How does an RCD detect a leakage?
It adds up magnetically, in a toroid, the current going in and the current coming back. If they are equal, the resulting field is zero. If return current is missing, a field appears that induces a voltage in a sensing winding and trips the mechanism.
Does an RCD protect if there is no grounding?
Yes, when the current goes to ground through the person: it detects the imbalance and trips. What is lost without grounding is early detection of the fault: with grounding, the leakage occurs as soon as the line touches the casing, and nobody has to touch it.
Why is the stop pushbutton normally closed?
So that any break in the pushbutton wiring—a cut wire, a loose contact—causes the motor to stop rather than making it impossible to stop. This is the fail-safe design criterion.
What is the holding contact in a control circuit?
An NO auxiliary contact of the contactor itself connected in parallel with the start pushbutton. When the coil is energized, that contact closes and keeps the circuit on even after the pushbutton is released. Because it depends on the coil voltage, after a power outage the motor does not restart on its own.
What is the interlock between two contactors for?
To prevent both from closing at the same time. In a reversing circuit that would short-circuit two phases: each contactor opens the other’s coil circuit with an NC contact.
How often should an RCD be tested, and why?
Once a month, with its test button. It is an electromechanical mechanism that can jam over time without warning, and the only way to know it still works is to make it trip.