The Repair LibraryRead · Learn · Master

USB Power Delivery Diagnostics and Charging Fault Isolation

The chapter's stealthiest faults carry no crack to photograph: the device that takes five volts and never raises a contract, the laptop that fast-charges from one brick and crawls from its twin, the port that negotiates on Mondays. This section is the bench that makes those failures visible. Its instruments are three: the pass-through power meter that reads the negotiation as numbers — offered profiles, requested contract, settled volts and amps — instead of inferring it from an icon; the trigger board that requests Power Delivery profiles on command, testing a charger's honesty and a cable's identity with no device in the loop at all; and the breakout board that brings the connector's own pins to probe points, where the configuration channel's resistors and continuity answer for themselves. The method is the chapter's split, now with instruments: the charger proven against the trigger board, the cable proven by swap and by marker interrogation, and only then the device suspected — its channel path traced from breakout to board, its detection resistors measured, and the charge-path silicon behind the port accused last, with evidence, and routed to the board-level bench that owns it. Along the way the section names the negotiation's honest impostors: the counterfeit charger whose label promises profiles its silicon never offers, the cooked cable whose marker answers wrong, the moisture lockout that refuses charge on purpose, and the dead-flat battery that will not negotiate until the default has fed it. Every case closes the same way — the contract restored and read on the meter, the fault named in the record, and the customer shown numbers instead of guesses. The parts cannon never fires here; the meters make it unnecessary.

IntermediateMedium Risk22 min read

What You Will Learn

  • You will learn the invisible fault families — five volts without a contract, partial negotiations, chargers that lie, cables that answer wrong, and lockouts that refuse on purpose.
  • You will learn the three instruments — the pass-through meter reading contracts, the trigger board testing sources and cables alone, and the breakout board probing the connector's own pins.
  • You will learn the instrumented split — charger proven first, cable proven second, and the device suspected only after both testify.
  • You will learn the device-side trace — channel continuity from breakout to board, detection resistors measured, and the charge-path silicon accused last with evidence.
  • You will learn the diagnostic close — the contract restored and read, the fault named in the record, the customer shown numbers.

What You Will Be Able To Do

  • You will be able to read a negotiation on the pass-through meter and name what capped it.
  • You will be able to prove a charger's honesty and a cable's identity with the trigger board, no device in the loop.
  • You will be able to run the instrumented split and route every invisible fault to charger, cable, or device.
  • You will be able to trace a device's channel path and detection resistors and accuse the charge-path silicon only with evidence.
  • You will be able to close diagnostics with the restored contract read, recorded, and shown.

Required Tools

  • A PD-aware pass-through USB power meter — one that lists profiles and contracts, not bare volts and amps
  • A PD trigger board — profiles requested on command, no device in the loop
  • A USB-C breakout board — the connector's pins brought to probe points
  • A multimeter for channel continuity and detection-resistor measurements
  • A staged reference set: honest chargers and marked cables whose behavior is known

Section Overview

The chapter's stealthiest faults have no crack to photograph, and this bench makes them visible with instruments instead of guesses (usb-c-platform-overview-and-fault-landscape). The meter reads the deal. The pass-through USB power meter turns the negotiation into three numbers — offered, requested, settled — and any gap between them names a suspect (usb-c-port-replacement-and-board-level-rework). The trigger board isolates the source. A PD trigger board requests profiles on command with no device in the loop — the charger's honesty and the cable's identity proven or convicted alone (usb-protocol-analyzers). The breakout reaches the pins. A USB-C breakout board brings the connector to probe points, where channel continuity and detection resistors answer on the multimeter (board-level-charging-repair-the-m92t36-and-bq24193). The split runs instrumented. Charger proven first, cable second, and the device suspected only after both testify — with the charge-path silicon accused last and routed to the board bench that owns it. The impostors get named. Counterfeit chargers, cooked markers, moisture lockouts, and dead-flat batteries — the honest explanations checked before any teardown. And the close shows numbers. The contract restored, read, recorded. The meter, the trigger, the breakout, the split, the close — the diagnostics bench entire.

Why This Matters

Every other bench in this chapter repairs what it can see, and this one exists because the worst charging faults are invisible (usb-c-platform-overview-and-fault-landscape). This matters because the parts cannon loads itself here: a device that will not fast-charge invites a port swap, then a cable, then a board — and the meter would have named the counterfeit charger in ninety seconds, before the first unnecessary part (usb-c-port-replacement-and-board-level-rework). This matters because the accessories lie with confidence: a charger's label is a claim, a cable's thickness is a costume, and only instruments interrogate either one under oath (usb-protocol-analyzers). It matters because the device-side suspects are ordered by price: a channel trace and two resistor measurements cost minutes, and the charge-path silicon they exonerate or accuse costs a board-level job — the cheap evidence must come first (board-level-charging-repair-the-m92t36-and-bq24193). It matters because some refusals are features: the moisture lockout that will not charge a damp port is protecting the customer, and the bench that reads the warning sells a drying wait instead of a teardown. And it matters because numbers close arguments: the customer shown offered-requested-settled understands the fault, the fix, and the invoice — and the record that holds those numbers answers every callback. Read the deal, prove the accessories, trace the cheap paths, accuse with evidence — and the invisible tier becomes the bench's most efficient work.

Required Prerequisites

  • A labeled reference charger per wattage tier, retired from customer duty — to give the trigger board a known-honest baseline (usb-c-platform-overview-and-fault-landscape)
  • Marked and unmarked reference cables, permanently tagged — to make cable verdicts comparative instead of theoretical (usb-protocol-analyzers)
  • Fine probe tips and grabber clips — to work the breakout's points without slips (board-level-charging-repair-the-m92t36-and-bq24193)
  • A printed offered-requested-settled log card per ticket — to file the numbers the close will show
  • A pass-through USB power meter — to watch healthy negotiations until the anomalies stand out (usb-c-platform-overview-and-fault-landscape)
  • A PD trigger board with selectable profiles — to interrogate every charger in the shop until the honest ones are known (usb-protocol-analyzers)
  • A USB-C breakout board — to probe channel lines and resistors on donors where slips are free (usb-c-port-replacement-and-board-level-rework)
  • A known-counterfeit or bargain charger from the drawer — to see a label lie on instruments before a customer pays for one
  • A donor device with a charging complaint — to run the whole instrumented split end to end (board-level-charging-repair-the-m92t36-and-bq24193)

Real-World Applications

The diagnostics bench turns the queue's longest arguments into its shortest tickets. A technician handed a laptop that crawls on its brand-new bargain charger puts the charger on the trigger board, requests the profiles its label promises, watches half of them refused, and sells an honest charger with the printout attached — no teardown, no parts (usb-protocol-analyzers). A bench facing a phone that takes five volts and never negotiates — its port rework already proven perfect at the gate — traces the channel path from breakout to board, finds one detection resistor reading open, and routes the case to board-level with the measurement attached instead of a guess (board-level-charging-repair-the-m92t36-and-bq24193). A counter fielding a tablet that refuses to charge at all after a rainstorm commute reads the moisture warning the customer dismissed, explains the lockout is the tablet protecting itself, and books a dry-out instead of the port replacement the customer requested (usb-c-platform-overview-and-fault-landscape). And a shop testing a game console that fast-charges on the bench but not at the customer's desk interrogates the customer's own cable on the trigger board — the marker answers wrong, cooked by a cheap charger's history — and the cable retires with its confession printed (usb-c-port-replacement-and-board-level-rework). The failures this prevents: a port swapped to fix a counterfeit charger, a board condemned over an open resistor nobody measured, a teardown sold against a moisture lockout doing its job, and a callback that no record of numbers could have answered.

Common Challenges

  • The instruments are only as good as the references. A trigger board reading an unknown charger against an unknown cable proves nothing — the difficulty is the discipline of the staged reference set: known-honest chargers, tagged cables, retired from customer duty (usb-protocol-analyzers).
  • The numbers need a baseline to be anomalies. Offered-requested-settled means nothing to a bench that never watched healthy negotiations — the difficulty is the practice hours on working devices before the broken ones arrive (usb-c-platform-overview-and-fault-landscape).
  • The silicon tempts the shortcut. The charge-path IC is the exotic suspect, and accusing it feels like expertise — the difficulty is running the cheap evidence first, because the expensive accusation is wrong more often than it is impressive (board-level-charging-repair-the-m92t36-and-bq24193).

Safety Notes

Risk Level: Medium. Powered measurements, trigger boards that summon real wattage, and probe tips near fine pins — the bench is electrical now, and its manners must be too.

Professional Tips Before Starting

  • Interrogate the shop's own chargers first. An afternoon with the trigger board turns the drawer into a reference library — and the counterfeits it finds were failing customers already (usb-protocol-analyzers).
  • Log offered-requested-settled on every charging ticket. Healthy or not, the numbers cost seconds — and the file of normals is what makes the anomalies obvious (usb-c-platform-overview-and-fault-landscape).
  • Keep the impostor list at the counter. Counterfeit, cooked cable, moisture lockout, dead-flat battery — four questions asked at intake save four teardowns a month (board-level-charging-repair-the-m92t36-and-bq24193).

The Diagnostics Bench — The Meter, The Trigger, The Breakout, The Split

Recap and Frame

The chapter built the ladder to this rung: the platform section taught the negotiation, the first-line tier cleared the mechanical suspects, and the rework bench proved the solder — what remains is the invisible tier (usb-c-platform-overview-and-fault-landscape). The gate's hard cases arrive here. A rework that passes magnification twice, powers sanely, and still will not negotiate is this section's opening customer — solder exonerated, fault alive (usb-c-port-replacement-and-board-level-rework). The tool chapter supplied the deep end. The protocol analyzer that decodes Power Delivery messages frame by frame is the escalation past this bench — most cases close on the meter and trigger long before frame-level decoding is worth the setup (usb-protocol-analyzers). The board bench waits downstream. The charge-path ICs behind the port — this section accuses them with measurements, and that section repairs what the accusation names (board-level-charging-repair-the-m92t36-and-bq24193). What is new is instrumented sight. Three tools that turn contracts, sources, cables, and channel paths into numbers — and the split that orders their testimony. Hold the frame — cases from the gate, escalation to the analyzer, accusations to the board bench — and the invisible tier becomes routine.

The Meter — Offered, Requested, Settled

The pass-through meter is the bench's stethoscope, and its whole skill is reading three numbers as a story (usb-c-platform-overview-and-fault-landscape). The read assumes a meter that speaks Power Delivery. One that lists the source's capability messages and shows the negotiated contract — not the bare volt-amp display of the cheapest inline meters; the bench buys the PD-aware tier once and never regrets it. Offered is the charger's testimony. The profiles advertised on attach — read live as the meter enumerates them — and the first check is the label: a charger offering less than it printed has already confessed — on a single-port unit, or a multi-port brick interrogated with its other ports empty, because shared-budget units reshuffle their offers per port (usb-protocol-analyzers). Requested is the device's side. What contract the device asked for — a device requesting the ceiling is healthy and hungry; one that never asks despite an honest offer has a fault behind its port. Settled is the deal. The volts and amps actually flowing, watched under load — because a contract that settles high and sags under draw is a cable, connection, or tired-brick story — the reference-cable and reference-charger swaps split it in minutes — not a negotiation one (usb-c-port-replacement-and-board-level-rework). The gaps name the suspects. Offer below label: the charger. Offer honest, no request: the device's channel or silicon. Request granted, settle poor: the copper path or a tired brick's output stage. No gaps and still slow: the fault is past the negotiation — battery, charge circuit, thermal policy — and the meter has just saved a teardown by saying so (board-level-charging-repair-the-m92t36-and-bq24193). Offered, requested, settled — the meter entire. Three numbers, read in order, and the argument is over before it starts.

The Trigger and the Breakout — Sources and Cables Under Oath

The meter reads deals the device makes; the trigger board makes deals itself, which is what isolates the accessories (usb-protocol-analyzers). The trigger interrogates chargers. Profiles requested one by one against the label's promises, no device in the loop — a charger that refuses its own printing is convicted alone, with nothing else to blame (usb-c-platform-overview-and-fault-landscape). The trigger tests cables by proxy. The same charger through a reference cable and then the suspect, with the capability list read on a trigger board or meter that shows each profile's current and not just its voltage: a five-amp offer that drops to three amps when the suspect cable carries it has interrogated the marker without a word — cooked, counterfeit, or absent, the cable's identity failed under oath. The requested rail is respected. A triggered twenty volts is real supply — pointed only at loads meant to take it, never left live across a bench (usb-c-port-replacement-and-board-level-rework). The breakout brings the device's pins to court. With the connector's lines at probe points, the channel path's continuity from pin to board answers on the multimeter, and the detection resistors measure against their expected values — present, open, or wrong — the cheap evidence that convicts or exonerates the port-side path before any silicon is named, with expected values learned from the board diagram and a known-good donor rather than the spec sheet, because detection resistors implemented inside live silicon read differently unpowered than a discrete resistor would (board-level-charging-repair-the-m92t36-and-bq24193). And the analyzer waits past all of it. Frame-level decoding for the case that survives every cheaper instrument — rare, real, and worth its setup only then. Trigger for sources, trigger-by-proxy for cables, breakout for the device's own pins — the interrogation entire. Nothing testifies in company; the instruments hear every suspect alone.

The Split and the Impostors — Order Against Error

The instruments only convict when the split orders their testimony, and the split's first law is unchanged: cheap and likely before dear and exotic (usb-c-platform-overview-and-fault-landscape). The charger goes first because it lies most. Counterfeits and tired bricks fail more devices than every board fault combined, and the trigger board convicts them in minutes without a screw turned (usb-protocol-analyzers). The cable goes second because it fails silently. Markers cook, identities corrupt, and the swap-plus-proxy test costs ninety seconds. The impostors get their questions before any teardown. The moisture lockout announced in the device's own warning — a drying wait, not a repair; the dead-flat battery that will not negotiate until the five-volt default has fed it for a while — a patience test, not a fault; both asked about at intake, both capable of consuming an afternoon of false diagnostics (usb-c-port-replacement-and-board-level-rework). Only then the device, in price order. Channel continuity at the breakout, detection resistors on the multimeter, and the charge-path silicon accused last — with the measurements attached — routed to the board bench that owns the repair (board-level-charging-repair-the-m92t36-and-bq24193). And the close shows the numbers. Offered-requested-settled on the restored contract, the fault named in the record, the customer shown the story instead of told it. Charger, cable, impostors, device, close — the split entire. The accused must be the last thing left, not the first thing guessed.

Common Mistakes

  • Accusing the device before the accessories testify. The port swapped, the board blamed, the counterfeit charger still in the customer's bag — the trigger board hears the charger first, every case (usb-protocol-analyzers).
  • Reading the icon instead of the numbers. Charging equals fine, until the crawl comes back — offered-requested-settled on the meter, logged on every ticket (usb-c-platform-overview-and-fault-landscape).
  • Triggering wattage into things not meant to take it. Twenty volts requested into a five-volt load for curiosity — the requested rail is real supply and is pointed accordingly (usb-c-port-replacement-and-board-level-rework).
  • Treating the lockout as a fault. The moisture warning dismissed and the port torn down damp — some refusals are the device protecting the customer; read before prying.
  • Skipping the cheap measurements to name the exotic suspect. The charge-path IC accused by vibe while an open resistor sat unmeasured at the breakout — silicon is accused last, with numbers attached (board-level-charging-repair-the-m92t36-and-bq24193).

Troubleshooting Guidance

The diagnostics bench troubleshoots by testimony order: source, cable, impostors, device. If a device charges slowly everywhere: the meter's three numbers first — an honest offer never requested points into the device; an offer below the label convicts the charger on the spot (usb-c-platform-overview-and-fault-landscape). If the fault follows one charger: the trigger board requests its label's profiles — refused profiles end the case without a screwdriver (usb-protocol-analyzers). If the fault follows one cable: the proxy test — reference charger, reference load, suspect cable in the middle, the capability list read with each profile's current showing — and the five-amp offer that drops to three names the marker. If the device refuses all charge suddenly: the impostor questions before any tool — moisture warning on screen, battery run truly flat, either one masquerading as hardware (usb-c-port-replacement-and-board-level-rework). If five volts flows and no contract ever rises: the breakout — channel continuity pin to board, detection resistors against the donor-and-diagram expected values — and the case routes to board-level with measurements, not adjectives (board-level-charging-repair-the-m92t36-and-bq24193). If every instrument passes and the fault persists: the analyzer's frame-level decode — the rare case that earns the deep tool. The throughline: every suspect testifies alone, in price order, and the teardown is the last resort with the best evidence.

Verification & Testing Methods

Confirm the diagnostics bench end to end before the invisible tier tests it:

  • [ ] I read negotiations as three numbers on the USB power meter — offered, requested, settled — and name the suspect any gap implies.
  • [ ] I prove chargers alone on the PD trigger board against their labels, and cables by proxy, before any device is suspected.
  • [ ] I probe channel continuity and detection resistors at the USB-C breakout board with clips and the diagram, treating triggered rails as live supply.
  • [ ] I ask the impostor questions — counterfeit, cooked cable, moisture lockout, dead-flat battery — before any teardown is scheduled.
  • [ ] I accuse the charge-path silicon last, with measurements attached, and close every case with the restored contract read, recorded, and shown.

Then try the practice exercises below — instrument drills on the shop's own hardware and donors; scenarios differ from the quiz.

Practice Exercises

  1. Read five healthy negotiations (5 minutes, power meter, assorted devices and chargers). Charge whatever the bench owns through the pass-through meter in five combinations, logging offered-requested-settled for each — then write one line per combination naming which number would move first under each fault family: counterfeit source, cooked cable, dead channel, tired battery (usb-c-platform-overview-and-fault-landscape).
  2. Interrogate the drawer (5 minutes, PD trigger board, three chargers). Request each charger's labeled profiles one by one on the trigger board — no device in the loop — marking each profile honored or refused, then run one suspect cable as a proxy test between a proven charger and a proven load with the capability list read current-and-voltage, and file the drawer's honest-reference list where the counter can see it (usb-protocol-analyzers).
  3. Trace a donor's channel (5 minutes, breakout board, multimeter, donor device). With the donor unpowered, probe channel continuity from the breakout's points toward the board and measure the detection resistors against the diagram's and a known-good donor's expected values, logging each as present, open, or wrong — then write the one-line routing verdict the measurements support: port-side path, board-level silicon, or exonerated (board-level-charging-repair-the-m92t36-and-bq24193).
  4. Write the impostor card (3 minutes, desk). Draft the counter's four-question card — counterfeit charger, cooked cable, moisture lockout, dead-flat battery — with the intake question and the instrument or wait that answers each — then perform the close once: copy one of Exercise 1's logged reads onto the printed offered-requested-settled log card and deliver its close line with those numbers as if the customer were at the counter (usb-c-port-replacement-and-board-level-rework).

These core steps — the metered read, the trigger interrogation, the channel trace, and the impostor card — are tested in the Chapter Quiz at the end of this chapter, where a score of 80% is required to continue.

Key Takeaways

  • The negotiation reads as three numbers on the USB power meter — offered, requested, settled — and every gap between them names a suspect while no gap points past the negotiation entirely (usb-c-platform-overview-and-fault-landscape).
  • The PD trigger board hears sources alone — profiles requested against the label with no device in the loop — and hears cables by proxy, which is how accessories get convicted without a screw turned (usb-protocol-analyzers).
  • The USB-C breakout board brings the connector's pins to court: channel continuity and detection resistors measured cheaply, before any silicon is named — and triggered rails are treated as the live supply they are (usb-c-port-replacement-and-board-level-rework).
  • The impostors get asked before the teardown: the counterfeit label, the cooked marker, the moisture lockout protecting on purpose, and the dead-flat battery that negotiates only after the default has fed it.
  • The split's verdict order is price order — charger, cable, impostors, channel path, silicon last with measurements attached — and the close shows the customer numbers instead of guesses (board-level-charging-repair-the-m92t36-and-bq24193).

Skills Learned

  • You can now read a negotiation on the pass-through meter and name what capped it.
  • You can now prove a charger's honesty and a cable's identity with the trigger board, no device in the loop.
  • You can now run the instrumented split and route every invisible fault to charger, cable, or device.
  • You can now trace a device's channel path and detection resistors and accuse the charge-path silicon only with evidence.
  • You can now close diagnostics with the restored contract read, recorded, and shown.

Glossary Additions

  • USB power meter — the pass-through instrument that sits between charger and device and turns a Power Delivery negotiation into readable numbers: the profiles offered, the contract requested, and the volts and amps actually settled and flowing — a read that requires the PD-aware tier of meter, one that decodes capability messages rather than displaying bare volts and amps. The meter's skill is gap-reading — an offer below the charger's label convicts the source, an honest offer never requested points behind the device's port, a healthy contract that sags under load names the copper path, and three clean numbers on a still-slow device point past the negotiation entirely, toward battery, charge circuit, or thermal policy. Logged as offered-requested-settled on every charging ticket, the meter's numbers are the diagnostic record that answers callbacks and the close that shows customers a story instead of telling them one.
  • PD trigger board — a small tool that speaks Power Delivery on its own behalf, requesting specific profiles from a charger with no device in the loop: the instrument that hears sources testify alone. Against a charger, it requests the label's promises one by one and marks each honored or refused — convicting counterfeits and tired bricks in minutes. Against a cable, it works by proxy: the same proven charger and load with the suspect cable in the middle, read on a board or meter that shows each profile's current and not just its voltage — where a five-amp offer dropping to three amps interrogates the e-marker without a word. Its one discipline is respect for what it summons — a requested twenty volts is real supply, pointed only at loads meant to receive it and never left live across a bench.
  • USB-C breakout board — a passive adapter that brings a USB-C connector's pins out to labeled probe points, giving a multimeter access to lines no probe tip can safely reach in a live cavity. At the diagnostics bench it carries the device-side trace: configuration channel continuity from connector toward board, detection resistors measured against their expected values — present, open, or wrong, with the expectations learned from the board diagram and a known-good donor, since resistors implemented inside live silicon read differently unpowered — the cheap evidence that convicts or exonerates the port-side path before any charge-path silicon is accused. Probe work at the breakout runs with grabber clips and the board diagram open, and its verdicts route cases with measurements attached: port-side path, board-level silicon, or exonerated.

Suggested Next Sections

Must read next:

  • Charging Port Repair Across Device Families — Section 7.5 closes the chapter with the tour: the same connector failing the same ways in phones, tablets, laptops, handhelds, and consoles — what changes per family, what never does, and how the chapter's four benches price and route each one.

Recommended: