The Repair LibraryRead · Learn · Master

Power Sequencing and Enable Logic

A board with a dozen rails cannot simply switch them all on at once. Many chips require their rails to come up — and go down — in a defined order, or they latch up, draw excessive current, or are damaged; so a modern board orchestrates its power, bringing each rail up in turn, waiting for each to be good before enabling the next. That orchestration is a system of signals: an enable that turns each regulator on, a power-good that reports it has reached voltage, and a sequencer or power-management chip that ties them together into the correct order. When it works, the board powers up cleanly. When one link breaks — an enable never asserted, a power-good that never goes high, a rail that fails and stalls the chain — the board simply will not start, and no amount of staring at a dead rail explains why until you follow the sequence. This section is about that logic: why rails come up in order, how enable and power-good signals turn rails on and gate the next, what a power sequencer and a soft-start do, and how to diagnose the board that will not start because one rail is waiting on another that never comes. Follow the sequence, and a dead board that should have powered up reveals exactly where the chain broke.

IntermediateMedium Risk22 min read

What You Will Learn

  • You will learn to explain why a board's rails must come up in a particular order.
  • You will learn to read enable signals and recognise a rail held off by a missing enable.
  • You will learn to read power-good signals and recognise how a power sequencer gates the rails.
  • You will learn to read soft-start and inrush behaviour at turn-on.
  • You will learn to diagnose a board that will not start because a rail waits on one that never comes.

What You Will Be Able To Do

  • You will be able to explain why a board's rails must come up in a particular order.
  • You will be able to read enable signals and recognise a rail held off by a missing enable.
  • You will be able to read power-good signals and recognise how a power sequencer gates the rails.
  • You will be able to read soft-start and inrush behaviour at turn-on.
  • You will be able to diagnose a board that will not start because a rail waits on one that never comes.

Required Tools

  • An oscilloscope to capture the turn-on sequence of the rails
  • A multimeter to check enable and power-good signal levels
  • A schematic showing the sequencing, enables, and power-good signals
  • A known-good board to compare the startup sequence against
  • A logic probe or scope with several channels for the signals

Section Overview

A board with a dozen rails cannot switch them all on at once — many chips require their rails to come up, and go down, in a defined order or they latch up, draw excessive current, or are damaged — so a modern board orchestrates its power, bringing each rail up in turn (understanding-power-rails-and-distribution). That orchestration is a system of signals this section reads. Rails come up in order for a reason. A defined sequence protects the chips and lets a board start cleanly, so the order in which rails rise is a designed requirement, not an accident, and a rail out of order or missing stalls the start (understanding-power-rails-and-distribution). Each regulator is turned on by a signal. An enable signal is the control input that turns a regulator on — a rail waits, off, until its enable is asserted — so a rail reading zero may simply never have been enabled rather than being faulty. Each rail reports when it is up. A power-good signal reports that a rail has reached its voltage, and a power sequencer — a controller or power-management chip — waits for each rail's power-good before enabling the next, tying the enables and power-goods into the correct order (voltage-measurements-in-diagnosis). Rails ramp up gently. A soft-start is the controlled, gradual ramp of a rail's voltage at turn-on, which limits the inrush current that charging the rail's capacitance would otherwise draw, so a rail rises smoothly rather than slamming up. And a broken link stalls the whole start. When an enable is never asserted, a power-good never goes high, or a rail fails and stalls the chain, the board simply will not start, and the fault is the first missing link in the sequence (voltage-measurements-in-diagnosis). Follow the sequence, and a dead board that should have powered up reveals exactly where the chain broke.

Why This Matters

A board that will not start is one of the hardest faults to face, because a dead board gives little to measure — and when the cause is a broken power sequence, following the enable and power-good logic is the only way to find where the start stalled (understanding-power-rails-and-distribution). This matters because dead rails have an order: when several rails are dead, they did not all fail — the sequence stalled at one point and everything after it never got its turn, so the order matters more than the count (understanding-power-rails-and-distribution). This matters because a rail off is not a rail failed: a rail reading zero because its enable was never asserted looks identical to a failed regulator on a meter, so understanding enable logic stops you replacing a good part that was simply never turned on. It matters because power-good gates the chain: a rail whose power-good never goes high halts the sequencer, so every rail after it stays off, and the fault is the one rail that could not report good, not the many that waited (voltage-measurements-in-diagnosis). It matters because the sequence is diagnosable: capturing the turn-on order on a scope shows exactly which rail rose, which did not, and where the chain broke, turning a dead board into a located fault. And it matters because sequencing faults masquerade as everything: a board that will not start, resets, or hangs at power-on is often a sequencing fault, so recognising the pattern points at the enable and power-good logic rather than deeper in the board (voltage-measurements-in-diagnosis). Follow the sequence of enables and power-goods, and the dead board tells you which link failed.

Required Prerequisites

  • Understanding Power Rails and Distribution — Section 5.1 introduced the power tree and rail dependencies; this section adds the sequencing and enable logic that determines the order those dependent rails come up in.
  • Voltage Measurements in Diagnosis — Section 3.2 taught reading a node against expected; enable and power-good are logic-level signals read the same way, and the rails are read for presence in sequence.
  • A schematic with the sequencing and enable logic — to see which enable and power-good ties which rails (understanding-power-rails-and-distribution)
  • An oscilloscope with several channels — to capture several rails and signals rising together at turn-on
  • A logic probe or fine scope probes — to read the enable and power-good signal levels
  • A notebook to record the startup order — to log which rail rose, which did not, and when
  • A known-good board or its startup timing — to compare the turn-on sequence against normal
  • A board with a multi-rail sequenced supply — to watch the rails come up in order (understanding-power-rails-and-distribution)
  • An oscilloscope able to capture several channels — to see the sequence and its signals at once
  • A board that will not start from a sequencing fault — to find where the chain stalls
  • A power-management chip or sequencer board — to study enables, power-goods, and their order (voltage-measurements-in-diagnosis)
  • A known-good identical board — to compare the startup timing and signal levels against
  • A schematic of the sequencing logic — to map the enable and power-good chain

Real-World Applications

Reading the sequence is how a dead-at-power-on board is finally diagnosed. A technician with a board that will not start captures the rails on a scope at power-on and sees the sequence stall at one rail that never rose, pointing straight at it (voltage-measurements-in-diagnosis). A repairer facing several dead rails recognises they all sit after one stalled rail in the sequence and fixes the one that broke the chain, not the many that waited (understanding-power-rails-and-distribution). Someone about to replace a regulator reading zero checks its enable, finds it never asserted, and traces back to why the enable never came rather than blaming the regulator. A technician whose board hangs at power-on finds a rail whose power-good never goes high, halting the sequencer and the boot. And a repairer of a supply that trips on inrush sees a soft-start failing and the rail slamming up, drawing an inrush the supply cannot take (understanding-power-rails-and-distribution). The failures this prevents: replacing a good regulator that was never enabled, chasing many dead rails instead of the one that stalled the chain, and missing a power-good or soft-start fault behind a board that will not start.

Common Challenges

  • Treating dead rails as independent. A stalled sequence leaves everything after it dead — read the order, not just the count (understanding-power-rails-and-distribution).
  • Confusing off with failed. A rail never enabled looks like a failed one — check the enable before condemning the regulator.
  • Not capturing the sequence. A single measurement misses the turn-on order — capture the rails rising together on a scope (voltage-measurements-in-diagnosis).

Safety Notes

Risk Level: Medium. Reading the sequence, enables, and power-goods is done on a powered board with a meter and scope, so it is live work and this section is Medium risk.

Professional Tips Before Starting

  • Capture the sequence. A dead board's fault is often in the turn-on order — scope the rails rising together at power-on (voltage-measurements-in-diagnosis).
  • Check the enable first. A rail off may never have been enabled — read the enable before suspecting the regulator.
  • Find where the chain breaks. Everything after a stalled rail is dead — find the first link that failed, not the last (understanding-power-rails-and-distribution).

Power Sequencing and the Signals That Orchestrate It

Recap and Frame

Section 5.1 mapped the rails and their dependencies; this section reads how they are turned on in order, and the frame to hold is that a board's power is orchestrated by enable and power-good signals, so a board that will not start is diagnosed by following that sequence to the broken link (understanding-power-rails-and-distribution). Power is sequenced, not switched. Rails must come up in a defined order to protect the chips, so a board turns them on in turn rather than all at once, and the order is a designed requirement (understanding-power-rails-and-distribution). Signals do the orchestrating. An enable turns each regulator on, a power-good reports each rail is up, and a sequencer ties them together, so the startup is a chain of signals as much as of rails. A rail off is not a rail failed. Because a rail waits for its enable, a rail reading zero may simply be un-enabled, so the enable logic must be read before a regulator is blamed. A broken link stalls the chain. One enable that never comes, one power-good that never goes high, or one rail that fails halts the sequence and leaves everything after it dead, so the fault is the first broken link. And the sequence is captured to see it. Scoping the rails and signals rising together at turn-on shows exactly which link broke, turning a dead board into a located fault (voltage-measurements-in-diagnosis). Hold the frame — power is an orchestrated sequence of enables and power-goods, and a dead board's fault is the first link that failed — and the board that will not start becomes diagnosable.

Why Rails Come Up in Order

The reason for all the sequencing logic is that many chips genuinely require their rails in a particular order, so understanding why order matters frames the enable and power-good signals that enforce it (understanding-power-rails-and-distribution). Know the chip requirements. Many complex chips — processors, FPGAs, memory — specify that their rails come up, and go down, in a defined order and within a time window, or internal structures conduct wrongly, latch up, or are stressed, so the order is a hard requirement of the parts. Understand the harm of the wrong order. Powering a chip's rails out of order can forward-bias internal diodes, trigger latch-up that draws destructive current, or leave the chip in an undefined state, so a board that does not sequence correctly can fail to start or damage itself (understanding-power-rails-and-distribution). Know the dependency drives the order. A rail derived from another must come up after it, and a rail that enables another must be good first, so the power tree's dependencies set much of the required order, tying this section to the last (understanding-power-rails-and-distribution). Recognise both up and down sequencing. The order matters at shutdown as well as startup — rails must fall in a controlled order too — so a sequencing fault can appear at power-down as well as power-up. Know the timing windows. Some sequences require not just an order but a timing — a rail up within so long of another — so a sequence that is out of order or out of time can both cause a fault, which is why the turn-on is captured with timing. See why it must be orchestrated. Because the order and timing are strict and the rails are many, the board needs active logic — enables, power-goods, a sequencer — to enforce them, so the orchestrating signals exist to meet the chips' requirements. The chip requirements, the harm of wrong order, the dependency-driven order, up-and-down sequencing, the timing, and the need for orchestration understood — and why rails come up in order is clear. Grasp why order matters, and the enable and power-good logic makes sense.

Enable Signals — Turning Rails On

The first orchestrating signal is the enable: each regulator is turned on by an enable input, so reading enables is how you tell a rail that is off from a rail that has failed (voltage-measurements-in-diagnosis). Understand the enable signal. An enable signal is the logic-level control input that turns a regulator on or off — a rail sits off, producing nothing, until its enable is asserted to the on state — so the enable is what starts each rail, and its absence keeps a rail dark. Read a dead rail's enable first. A rail reading zero is checked at its enable before its regulator is suspected — if the enable is not asserted, the regulator is doing exactly what it is told, and the fault is upstream in whatever should have driven the enable (voltage-measurements-in-diagnosis). Know where enables come from. An enable may come from another rail's power-good, a sequencer, a system control signal, a pull-up to a standby rail, or a processor, so tracing back an un-asserted enable finds why the rail was never turned on. Read the enable's logic level. An enable is a logic signal with an on and off state, so it is read against its expected asserted level — and an enable stuck in the off state, floating, or at the wrong level explains a rail that never comes up (voltage-measurements-in-diagnosis). Distinguish a held-off rail from a failed one. A rail off because its enable is not asserted, and a rail whose regulator has failed with the enable properly asserted, look identical on a meter but have opposite causes, so the enable reading tells them apart — and because enable polarity varies by part (active-high on some regulators, active-low on others), the enable is read against its own expected asserted state, not assumed high. Follow a missing enable back. When an enable is not asserted, the fault is traced back along whatever should have driven it — a missing upstream power-good, a stalled sequencer, a control line — to the real origin. The enable signal understood, read first on a dead rail, its source known, its level read, held-off told from failed, and a missing enable traced back — and enable logic is grasped. Read the enable, and a dark rail tells you whether it was ever asked to turn on.

Power-Good and the Sequencer

Enables turn rails on, but something must decide when to assert each enable, and that is the job of power-good signals and a power sequencer working together to bring the rails up in order (voltage-measurements-in-diagnosis). Recall the power-good signal. A power-good signal is an output from a rail or its regulator that goes high when the rail has reached its correct voltage, reporting that the rail is up and ready, so power-good is how a rail announces it is good (voltage-measurements-in-diagnosis). Understand the power sequencer. A power sequencer is a controller — a dedicated sequencing chip, a power-management chip, or logic — that brings a board's rails up in the required order by asserting each rail's enable, waiting for that rail's power-good, and only then enabling the next, so it enforces the sequence. Read the gate power-good provides. Because the sequencer waits for each power-good before the next enable, a rail whose power-good never goes high halts the whole sequence — everything after it stays off — so a stuck-low power-good stalls the start (voltage-measurements-in-diagnosis). Find where the sequence stalls. Following the chain — enable asserted, rail rises, power-good goes high, next enable asserted — until a step fails locates exactly where the sequence stalled and which rail broke the chain. Know the power-management chip's role. On many boards a power-management chip integrates several regulators and the sequencer, so its enables, power-goods, and internal sequencing are read together, and a fault in it can stall many rails at once (understanding-power-rails-and-distribution). Read the supervisor and reset too. A voltage supervisor or power-on reset may hold the system in reset until the rails are good, so a reset stuck asserted because a rail is not good is part of the same sequencing picture. The power-good recalled, the sequencer understood, its gating read, the stall located, the power-management chip's role known, and reset tied in — and the sequencer and power-good logic is grasped. Follow the enable-then-power-good chain, and the point it stops is the fault.

Soft-Start and Inrush at Turn-On

Beyond the order of the rails is how each one rises, and soft-start — the controlled ramp of a rail at turn-on — shapes the inrush current and is a startup behaviour that can itself fault (understanding-power-rails-and-distribution). Understand soft-start. A soft-start is the deliberate, gradual ramp of a rail's voltage from zero to its final value over a controlled time at turn-on, rather than an instant jump, so the rail rises smoothly under the regulator's control. Know why soft-start limits inrush. Charging a rail's bulk and decoupling capacitance instantly would draw a large inrush current, so ramping the voltage slowly limits that inrush to a manageable level, protecting the supply and preventing a turn-on current spike from tripping protection (understanding-power-rails-and-distribution). Read the soft-start ramp. Scoping a rail at turn-on shows its soft-start ramp — a smooth rise over a defined time — and a ramp that is too fast, too slow, or stepped can indicate a soft-start component or control fault. Recognise an inrush-tripping fault. A failed soft-start that lets a rail slam up draws a large inrush that can trip the upstream supply's current limit or foldback, so a board that trips at turn-on may have a soft-start fault, not a short (understanding-power-rails-and-distribution). Connect soft-start to sequencing. The soft-start time is part of the sequence timing, so a rail that ramps too slowly may miss its timing window and stall the sequence, tying soft-start to the ordering logic. Mind inrush at the system level. The whole board draws an inrush as all its capacitance charges at turn-on, so an inrush-limiting circuit at the input, and the per-rail soft-starts, together shape the turn-on current the supply must survive. Soft-start understood, its inrush-limiting role, its ramp read, an inrush-trip recognised, connected to sequencing, and system inrush minded — and soft-start and inrush are grasped. Read how a rail ramps up, and a turn-on trip or a stalled sequence may explain itself.

Diagnosing a Board That Will Not Start

The methods come together on the hardest case — a board that will not start — where following the sequence of enables and power-goods to the first broken link is what finds the fault (voltage-measurements-in-diagnosis). Capture the turn-on sequence. Scope several rails and their signals rising together at power-on, so the whole startup — which rail rose, which did not, and in what order — is seen at once rather than one rail at a time (voltage-measurements-in-diagnosis). Find the first rail that fails to rise. Read the captured sequence for the first rail that does not come up — everything after it is a symptom of its failure — so the first broken link, not the last dead rail, is the fault (understanding-power-rails-and-distribution). At the broken link, check enable then power-good. At the rail that failed to rise, check whether its enable was asserted — if not, the fault is upstream in what should have enabled it; if it was, the rail's own regulator or a load failed to reach voltage, and its power-good never came (voltage-measurements-in-diagnosis). Distinguish stall causes. A stall can be a missing enable, a rail that cannot reach voltage (a weak supply or a short loading it), a power-good that never asserts, or a soft-start too slow — and reading the signals tells which, so the stall is diagnosed, not guessed. Follow it to the origin. The broken link is traced to its cause — the upstream signal that never came, the shorted rail that could not rise, the failed sequencer — so the repair addresses the origin, not the symptom (understanding-power-rails-and-distribution). Confirm the fix restarts the sequence. After a repair, the turn-on is captured again to confirm the sequence now completes and the board starts, closing the loop from a dead board to a working one. The sequence captured, the first failure found, enable-then-power-good checked, the stall cause distinguished, traced to its origin, and confirmed by a clean restart — and the board that will not start is diagnosed. Follow the startup to its first broken link, and the dead board gives up its fault.

Common Mistakes

  • Replacing a rail's regulator without checking its enable. A rail off may never have been enabled — read the enable first (voltage-measurements-in-diagnosis).
  • Chasing every dead rail equally. Everything after a stalled rail is a symptom — find the first rail that failed to rise (understanding-power-rails-and-distribution).
  • Not capturing the sequence. A single reading misses the turn-on order and timing — scope the rails rising together at power-on.
  • Blaming a short for a turn-on trip. A failed soft-start can trip the supply on inrush — read the ramp before assuming a short (understanding-power-rails-and-distribution).
  • Ignoring the power-good chain. A stuck-low power-good stalls the sequencer — check that each rail reports good before the next enables.

Troubleshooting Guidance

Sequencing problems come down to a missing enable, a stalled power-good, or an un-captured sequence. If a board will not start: capture the turn-on sequence on a scope and find the first rail that fails to rise (voltage-measurements-in-diagnosis). If a rail reads zero: check its enable before its regulator — an un-asserted enable means the fault is upstream. If several rails are dead: they sit after one stalled rail — fix the one that broke the chain, not the many that waited (understanding-power-rails-and-distribution). If a rail's power-good never goes high: the sequencer halts there — find why the rail cannot reach voltage or report good. If the board trips at turn-on: a soft-start may have failed and the rail slams up drawing inrush — read the ramp, do not assume a short. If a board hangs in reset at power-on: a voltage supervisor may be holding reset because a rail is not good — find the rail it is watching (voltage-measurements-in-diagnosis). If the sequence is out of time, not just order: a rail ramping too slowly can miss its window — check the soft-start timing. The throughline: capture the sequence, find the first broken link, and check its enable and power-good before its regulator.

Verification & Testing Methods

Confirm you followed the sequence to the broken link:

  • [ ] I captured the turn-on sequence on a scope and identified the first rail that failed to rise (voltage-measurements-in-diagnosis).
  • [ ] I read the failed rail's enable signal first, telling a rail never enabled from a regulator that failed with its enable asserted.
  • [ ] I followed the power-good chain and the power sequencer to see where it stalled — a stuck-low power-good halting the next enable.
  • [ ] I read the soft-start ramp where a board tripped at turn-on, distinguishing an inrush trip from a short.
  • [ ] I traced the first broken link to its origin and confirmed a repair by a clean restart of the sequence (understanding-power-rails-and-distribution).

Then try the practice exercises below — sequencing-diagnosis practice on powered boards; scenarios differ from the quiz.

Practice Exercises

  1. Capture the sequence (5 minutes, hands-on). Scope several rails of a multi-rail board rising together at power-on, and record the order and timing in which they come up (voltage-measurements-in-diagnosis).
  2. Read the enables (5 minutes, hands-on). For each rail, read its enable signal and confirm it is asserted when the rail should come up, spotting any rail held off by an un-asserted enable.
  3. Follow the power-good chain (5 minutes, hands-on). Trace the enable-then-power-good chain through the sequencer, finding where it would stall if one rail's power-good never went high (understanding-power-rails-and-distribution).
  4. Diagnose a no-start (5 minutes, reasoning). For a board that will not start, decide how you would find the first broken link and whether it is a missing enable, a stalled power-good, a rail that cannot rise, or a soft-start fault (voltage-measurements-in-diagnosis).

These core steps — why rails come up in order, reading enable signals, the power-good and sequencer chain, soft-start and inrush, and diagnosing a no-start — are tested in the Chapter Quiz at the end of this chapter, where a score of 80% is required to continue.

Key Takeaways

  • Many chips require their rails up and down in a defined order, so a board sequences its power rather than switching it all at once — and a rail out of order or missing stalls the start (understanding-power-rails-and-distribution).
  • An enable signal turns each regulator on, so a rail reading zero may simply never have been enabled rather than failed — the enable is read before the regulator is condemned.
  • A power-good reports a rail is up, and a power sequencer waits for each power-good before enabling the next, so a stuck-low power-good halts the sequence and leaves every later rail dead (voltage-measurements-in-diagnosis).
  • A soft-start ramps a rail up gradually to limit inrush, so a failed soft-start can slam a rail up and trip the supply on inrush, mimicking a short at turn-on.
  • A board that will not start is diagnosed by capturing the turn-on sequence and finding the first broken link — a missing enable, a stalled power-good, a rail that cannot rise — then tracing it to its origin (voltage-measurements-in-diagnosis).

Skills Learned

  • You can now explain why a board's rails must come up in a particular order.
  • You can now read enable signals and recognise a rail held off by a missing enable.
  • You can now read power-good signals and recognise how a power sequencer gates the rails.
  • You can now read soft-start and inrush behaviour at turn-on.
  • You can now diagnose a board that will not start because a rail waits on one that never comes.

Glossary Additions

  • enable signal — the logic-level control input that turns a voltage regulator on or off: a rail sits off, producing nothing, until its enable is asserted to the on state, so the enable is what starts each rail and its absence keeps a rail dark. Because a rail reading zero because its enable was never asserted looks identical on a meter to a regulator that has failed, the enable is read first on a dead rail — if it is not asserted, the regulator is doing exactly what it is told and the fault lies upstream in whatever should have driven the enable. An enable may come from another rail's power-good, a power sequencer, a system control line, a pull-up to a standby rail, or a processor, so tracing an un-asserted enable back along its source finds why the rail was never turned on.
  • power sequencer — a controller — a dedicated sequencing chip, a power-management chip, or discrete logic — that brings a board's power rails up in the required order by asserting each rail's enable, waiting for that rail's power-good to confirm it has reached voltage, and only then enabling the next rail. The power sequencer enforces the ordering that many chips require, and it is central to diagnosing a board that will not start: because it waits for each power-good before the next enable, a rail whose power-good never goes high halts the whole sequence and leaves every rail after it off, so following the enable-then-power-good chain until a step fails locates exactly where the sequence stalled. On many boards a power-management chip integrates several regulators and the sequencer together, so a fault within it can stall many rails at once.
  • soft-start — the deliberate, gradual ramp of a power rail's voltage from zero to its final value over a controlled time at turn-on, rather than an instant jump, so the rail rises smoothly under the regulator's control. Soft-start's purpose is to limit inrush current: charging a rail's bulk and decoupling capacitance instantly would draw a large surge, so ramping the voltage slowly holds that inrush to a manageable level and prevents a turn-on spike from tripping the upstream supply's protection. A soft-start that has failed can let a rail slam up and draw an inrush that trips the supply's current limit or foldback, so a board that trips at turn-on may have a soft-start fault rather than a short; and because the soft-start time is part of the sequence timing, a rail that ramps too slowly can miss its window and stall the sequence.

Suggested Next Sections

Must read next:

  • Diagnosing Regulator and Converter Faults — Section 5.5 goes inside the parts that make the rails: the linear regulators and switching converters and their feedback, and how to diagnose the specific failures — dead, wrong output, oscillating, or shut down — that leave a rail faulty.

Recommended: