The Repair LibraryRead · Learn · Master

Power System Failure Modes

How power systems actually break — dead rails, high ripple, wrong voltage, sag under load, overheating, sequencing and battery faults — and the input-to-load, test-under-load method that localizes them.

IntermediateMedium Risk28 min read

What You Will Learn

  • You will learn the common ways power systems fail — dead rail, high ripple, wrong voltage, sag under load, overheating, sequencing/supervisor faults, battery faults, and intermittents.
  • You will learn the systematic power-diagnosis method: work from the input toward the load, checking each stage and rail.
  • You will learn to measure both voltage and ripple and to test under load, and to compare against a known-good reference.
  • You will learn the most common power culprits and to work safely around mains, charged capacitors, and lithium.

What You Will Be Able To Do

  • You will be able to recognize and name the common power failure modes from their symptoms.
  • You will be able to diagnose a power system from the input toward the load, stage by stage and rail by rail.
  • You will be able to measure voltage and ripple, test under load, and compare to a known-good reference.
  • You will be able to link a power symptom to its likely culprit and work safely around the chapter's hazards.

Required Tools

No physical tools required. This is a conceptual section.

Section Overview

This section closes the power chapter the way Section 5.5 closed analog and Section 6.5 closed digital: by turning everything in it into a diagnostic framework. You've learned how power is made — the supply chain (7.1), linear (7.2) and switching (7.3) regulators, batteries (7.4), and rail sequencing (7.5) — and now you'll learn how power fails and how to find the fault. Power failures fall into a recognizable set: a dead rail, high ripple, a wrong voltage, sag under load, an overheating regulator, a sequencing or supervisor fault, a battery fault, and the intermittent. Just as important is the method: work from the input toward the load, checking each stage and rail, measuring both voltage and ripple, testing under load — and comparing to a known-good reference. This is also where the "check power first" theme from the analog and digital capstones comes home, because power problems are the most common root cause of dead and misbehaving equipment, and this is the power system those checks point at.

Why This Matters

Power diagnosis is the most frequently-needed repair skill there is. A huge share of "dead," "won't boot," and "misbehaves" faults are power problems — a rail that's missing, wrong, noisy, or sagging — which is exactly why the analog and digital chapters put "check power first" at the top of their methods. This section gives you the framework to act on that: knowing the handful of ways a power system fails lets you name what you're seeing, and the input-to-load method lets you localize it to a stage or a rail instead of guessing. It applies to essentially all mains- and battery-powered equipment, and it ties the whole chapter together: a dead rail is a failed part from 7.1–7.3, high ripple is a tired capacitor, a held-in-reset board is a sequencing fault from 7.5, a swollen pack is the lithium hazard of 7.4. This is the payoff section — the framework that makes power repair systematic, and it carries the chapter's highest safety stakes.

Required Prerequisites

No consumables required. The exercises are reasoning about symptoms and diagnostic method; nothing is consumed.

  • Optional: a multimeter (DC and AC ranges), an oscilloscope for ripple and switching-node work, an ESR meter for testing capacitors, and a current-limited bench supply and load for safe testing
  • A known-good identical unit or board makes an invaluable reference for comparison
  • No special hardware is required; the framework stands on reasoning — but any hands-on power work demands the safety rules below

Real-World Applications

These failure modes and this method are the daily bread of repair, because power is where so much goes wrong. A completely dead device is usually a power problem — a blown fuse, a failed rectifier or regulator, a dead rail — not a failed processor. Equipment that hums, glitches, or resets under load very often has a dried-out electrolytic capacitor letting ripple through. A gadget with poor battery life or that won't charge is a battery or charging fault. A multi-rail board that won't boot is frequently one missing rail holding the system in reset. And the maddening "works until it warms up" is a thermal intermittent, classically a cracked solder joint. In every case the technician's edge is the same: recognize the mode, then work from the input toward the load to find the first stage or rail that's wrong. This is among the highest-value and most safety-critical skills on the bench.

Common Challenges

  • Assuming the load is at fault. A dead or misbehaving device is more often a power problem than a failed main chip; confirm the power system first, exactly as the analog and digital capstones advise.
  • Testing only at no load. A supply can read perfect with nothing connected and collapse under load; sag and many marginal faults appear only when current is drawn.
  • Reading only DC voltage. A rail can be the right average voltage yet carry excessive ripple; you have to look at the ripple (with a scope), not just the DC number.

Safety Notes

Risk Level: Medium. Power diagnosis means working on live supplies and stored energy, which carries the chapter's real hazards. Read these carefully.

Professional Tips Before Starting

  • Suspect power first. When a device is dead or misbehaving, confirm the power system before chasing anything downstream — it is the single most common root cause, which is why every diagnostic method in this handbook starts there.
  • Measure voltage and ripple, and test under load. A rail is only good if it is the right voltage, clean (low ripple), and holds up when the circuit draws current; check all three, using a scope for ripple and a load to reveal sag.
  • Keep a known-good reference and the right tools. An identical working unit turns "is this normal?" into a direct comparison, and an ESR meter finds the tired capacitor that causes so many power faults — often without even removing it.

Recognizing and Diagnosing Power Faults

The Power Failure Modes

Power systems fail in a recognizable handful of ways, each tied to a stage from earlier in the chapter:

  • Dead rail / no output. A rail is missing entirely — a blown fuse, an open transformer or a dead rectifier diode (Section 7.1), a failed linear or switching regulator (Sections 7.2, 7.3), or an open inductor. The whole downstream section is dead.
  • High ripple or noise. The rail is roughly the right voltage but carries excessive ripple — almost always a tired, high-ESR smoothing or output capacitor (Sections 7.1, 7.3, and the aging from 3.7). This is the single most common power fault.
  • Wrong voltage. The rail is present but at the wrong level — a failed regulator, or a bad feedback or set resistor on an adjustable regulator (Section 7.2).
  • Sag under load. The rail is correct at no load but droops when the circuit draws current — an undersized or failing supply, or a weak capacitor. It only shows under load. A related effect is a brownout: a momentary dip in the input (a weak source, a sagging battery, or an overload) that pulls one or more rails below their valid range, which a voltage supervisor (Section 7.5) responds to by holding or re-asserting reset.
  • Overheating / thermal shutdown. A regulator runs too hot and may cycle its output on and off — too much (Vin − Vout) × I for its heatsink on a linear regulator (Section 7.2), or an overloaded switcher.
  • Sequencing / supervisor fault. On a multi-rail board, a rail is missing or late, or the system is held in reset because a supervisor never sees all rails good (Section 7.5) — a fully-powered board that sits inert.
  • Battery fault. A swollen cell, a pack that won't charge, or tripped protection (Section 7.4) — often the protection correctly refusing an unsafe condition.
  • Intermittent power. A rail that comes and goes with temperature or flexing — classically a cracked solder joint or a bad connection.

The Method: Input Toward the Load

Power diagnosis follows the flow of energy: work from the input toward the load. Start at the input — is the supply plugged in or the battery charged, and is the fuse intact (and if blown, why)? Then move through each stage or rail in order: on a linear supply, follow the conversion chain (transformer → rectifier → smoothing → regulator, Section 7.1); on a switching supply, scope the switching node to confirm it's running (Section 7.3); on a multi-rail board, check each rail and the power-good and reset lines in turn (Section 7.5). At each point, ask whether the signal is what it should be — the same actual-versus-expected logic as the analog and digital capstones. The fault is the first stage or rail where the output is wrong: the first missing rail, the first stage whose output isn't there or isn't clean. Working input-to-load keeps you from chasing a downstream symptom back to its real upstream cause.

Voltage, Ripple, and Testing Under Load

Three measurement habits catch most power faults. First, at each rail, measure the voltage with a multimeter and confirm it's correct. Second, look at the ripple with an oscilloscope (or the meter's AC range for a rough read) — a rail can be the right average voltage yet carry excessive ripple from a tired capacitor, which a DC reading alone completely misses. Third, and critically, test under load: a supply that reads perfect at no load can sag or collapse when the circuit draws current, so check the rail while it's actually powering its load (or with a bench load applied). These connect to two quality measures worth naming: load regulation (how much a rail's voltage changes from no-load to full-load) and line regulation (how much it changes when the input voltage varies) — a supply with poor load regulation is exactly the one that sags under load. Voltage-plus-ripple-plus-under-load is the core of thorough power measurement.

The Right Tools and Known-Good

Match the tool to the question, and lean on comparison. A multimeter gives DC rail voltages fast; an oscilloscope reveals ripple, switching activity, and timing; an ESR meter measures a capacitor's equivalent series resistance — often in-circuit — and is the fastest way to find the dried-out electrolytic behind so many power faults, since a cap can read the right capacitance yet have ruinously high ESR. A current-limited bench supply lets you power up a suspect board safely, limiting damage if a rail is shorted. And a known-good reference — the same rail on an identical working unit — turns "is this ripple normal?" into an instant side-by-side answer. The right tool plus a known-good comparison is what makes power diagnosis fast instead of a guess.

The Common Culprits

A short list of culprits dominates power faults. Far and away the most common is the dried-out electrolytic capacitor — its capacitance falls and its ESR rises with age (Section 3.7), causing high ripple, sag, instability, and hum across supplies of every kind; it is the first thing to suspect in aging equipment. Next come a failed regulator (linear or switching), a blown fuse (and always ask why it blew — a shorted rectifier or a downstream short, not just the fuse itself), a bad connection or cracked solder joint (the usual source of intermittents), and, in portable gear, a worn or protected battery. A genuinely failed load chip is far down the list — which is precisely why "check power first" is the universal opening move.

Common Mistakes

  • Blaming the load before the power. Confirm the rails first; a dead device is usually a power fault, not a failed main chip.
  • Only reading DC, only at no load. Check ripple too, and test under load — many faults hide from a no-load DC measurement.
  • Replacing a fuse without finding why it blew. A blown fuse is a symptom; a shorted rectifier or downstream short will just blow the new one — find the cause.
  • Overlooking ESR. A capacitor can measure the right value yet have high ESR; an ESR meter catches the tired electrolytic that a capacitance check misses.

Troubleshooting Guidance

Make the equipment safe first — unplug, discharge, and verify the main capacitors are down, and never touch a damaged lithium cell (the consolidated 7.1 and 7.4 hazards). Then diagnose from the input toward the load. Confirm the input: mains present or battery charged, and the fuse intact — if it's blown, find why before replacing it, because a shorted rectifier or a downstream short will blow the replacement. Then walk the stages and rails: for no output, follow the chain to the first dead stage (fuse, transformer, rectifier, regulator, inductor); for high ripple — hum, glitching, resets — suspect a high-ESR smoothing or output capacitor and confirm with an ESR meter or scope; for a wrong voltage, suspect the regulator or its feedback/set resistor; for sag under load, test with the load applied (poor load regulation from a weak supply or capacitor); for overheating/shutdown, check the (Vin − Vout) × I dissipation and heatsinking or an overload; for a held-in-reset multi-rail board, find the rail that isn't good and the supervisor waiting on it; for a battery complaint, inspect for swelling and check cell voltages and protection; and for an intermittent, look for a cracked joint or connection that changes with heat or flexing. At every rail, measure voltage and ripple and test under load, and compare to a known-good unit whenever possible. Because power is the most common root cause of dead and misbehaving equipment, confirming a clean, correct, stable set of rails first — the check-power-first habit made concrete — is usually the fastest route to the real fault.

Verification & Testing Methods

Check your understanding before moving on:

  • [ ] Name the common power failure modes and the symptom that identifies each.
  • [ ] Explain the input-to-load diagnostic method and why the first wrong stage or rail is the fault.
  • [ ] Explain why you must measure ripple (not just DC voltage) and test under load, and what load regulation means.
  • [ ] Name the most common power culprit and the tool that finds it, and state the safety steps before live power work.

Then try the practice exercises below — reasoning about power symptoms and method, no live mains work required.

Practice Exercises

  1. Name the mode (5 minutes, reasoning). For each symptom, name the failure mode and a likely stage: (a) a rail reads the right voltage but the device hums and resets under load; (b) a device is completely dead and the fuse is blown; (c) a multi-rail board is powered but inert, most rails present; (d) a regulator gets very hot and its output cycles on and off.
  2. Why under load? (5 minutes, reasoning). Explain why a supply can read perfect at no load yet fail when the device runs, and what "load regulation" describes. How would you test for it?
  3. Order the checks (10 minutes, reasoning). A mains device is dead. List, in order from the input toward the load, the stages you would check on a linear supply, and state what a blown fuse tells you and why you must find why it blew.
  4. Pick the culprit and the tool (10 minutes, reasoning). Aging equipment hums and has excessive ripple on its main rail. Name the most likely culprit, the tool that confirms it fastest (and why a plain capacitance check can miss it), and the two safety steps you'd take before probing inside.

These core ideas — the power failure modes, the input-to-load method, measuring voltage and ripple, testing under load, and the common culprits — are tested in the Chapter Quiz at the end of this chapter, where a score of 80% is required to continue.

Key Takeaways

  • Power systems fail in a recognizable set: a dead rail, high ripple, wrong voltage, sag under load, overheating/thermal shutdown, sequencing/supervisor faults (held in reset), battery faults, and intermittents — each tied to a stage from Sections 7.1–7.5.
  • Diagnose from the input toward the load: check the input and fuse first, then each stage or rail in order; the fault is the first stage or rail that is wrong.
  • Measure both voltage and ripple, and test under load — a rail can read the right DC voltage yet carry excess ripple or sag under current; load regulation and line regulation describe how a rail holds up against load and input changes.
  • Match the tool to the question: a multimeter for voltage, an oscilloscope for ripple and switching activity, an ESR meter for the tired capacitor, and a current-limited supply for safe bring-up; a known-good unit is the fastest comparison.
  • The most common culprit by far is a dried-out, high-ESR electrolytic capacitor (Section 3.7); then a failed regulator, a blown fuse (find why it blew), a bad connection, and a worn or protected battery — a failed load chip is far down the list.
  • Safety is paramount: mains primaries and charged capacitors (Section 7.1) and lithium fire risk (Section 7.4) all apply — unplug, discharge, verify, and never work on damaged lithium before diagnosing power.

Skills Learned

  • You can now recognize and name the common power failure modes from their symptoms.
  • You can now diagnose a power system from the input toward the load, stage by stage and rail by rail.
  • You can now measure voltage and ripple, test under load, and compare to a known-good reference.
  • You can now link a power symptom to its likely culprit — the dried electrolytic above all — and pick the right tool.
  • You can now apply the consolidated mains, charged-capacitor, and lithium safety rules before live power work.

Glossary Additions

  • brownout — a temporary dip in supply voltage (from a weak source, a sagging battery, or a momentary overload) that can drop one or more rails below their valid range; a voltage supervisor typically responds by holding or re-asserting reset so the system does not run on bad power.
  • load regulation — how much a supply's output voltage changes as the load current varies from little or no load to full load; poor load regulation is exactly the fault behind a rail that reads correct at no load but sags when the device draws current.
  • line regulation — how much a supply's output voltage changes when its input voltage varies; good line regulation means the output stays steady despite a changing input, such as a drooping battery or a fluctuating mains.
  • ESR meter — a test instrument that measures a capacitor's equivalent series resistance (ESR), often in-circuit; because a tired electrolytic can retain much of its capacitance while its ESR climbs, an ESR meter finds the dried-out capacitor behind high ripple and many power faults faster than a capacitance measurement can.

Suggested Next Sections

Must read next:

  • Schematic Symbols and Conventions — the opening of Chapter 8 (Reading Schematics and Datasheets), which turns from building and diagnosing circuits to reading the documentation that describes them — the symbols, conventions, and datasheets a technician lives by.

Recommended: