Premise: this article is only about intermittent charging, and the basic checks are already done
This article starts from one explicit premise: the basic checks are already done, but charging is still intermittent. Basic checks here mean trying another genuine or compliant charger, trying another wall socket, restarting the computer, and checking battery health. If any of those steps has not been done, the advanced reasoning in this article does not apply, and the basic checks should be done first, because everything that follows is built on the foundation that the basics have already been excluded.
The subject of this article is very specific: charging that is intermittent rather than completely dead. The symptoms in scope are charging that cuts in and out, charging that stops when the cable is moved, charging that only works when the plug is held firmly in place, charging that runs for a while and then stops, a charging light that flickers on and off, and the same charger and cable working sometimes and failing at other times. The key word is intermittent. A fault that is completely open usually behaves consistently, while an intermittent fault usually has one variable that changes with contact, temperature, load, or negotiation state. The job of this article is to find that variable, layer by layer.
This article is not a general charging guide. It does not cover every cause of a laptop that will not charge at all, it is not a continuation of the previous USB-C dock article, and dock power, display output, and USB expansion problems are outside its focus. It also does not cover a battery that drains quickly as such; it covers only the charging process itself cutting in and out. Readers whose symptom is actually no charging icon at all, no light at all, or unusually fast battery drain should check the basic guides first rather than applying the advanced conclusions here directly.
There is only one correct way to read this article: change one variable at a time, work through the layers in order, and write down what you observe. Changing the cable, the charger, and the socket all at once may fix the problem without revealing which item was responsible, and if nothing improves, it leaves nothing useful to hand to a repair technician. Each section ends with observations worth recording, and taking that record to a diagnostic appointment is far more useful than simply saying that the laptop will not charge.
The layered framework: port, cable, charger, negotiation, battery, embedded controller, mainboard
The most common error in intermittent charging diagnosis is jumping straight from a symptom to a conclusion: movement breaks charging, so the port must be broken; charging is slow, so the battery must be broken; the light flickers, so the mainboard must be broken. The correct framework works from the outside inward, with each layer confirmed or excluded by a test that can be repeated, before moving to the next layer. The order is not arbitrary; it follows repair cost and testing difficulty from lowest to highest. External accessories are the easiest to swap, settings are the easiest to change, the port can be inspected without opening the machine, and the mainboard is considered last.
The table below is the route map for the whole article. Later sections expand each layer with its mechanism, its test method, and its misreading traps. Readers may jump to the layer they have not yet excluded, but they should not skip an untested layer and declare a mainboard fault.
| Layer | What is checked | Core question | How it is excluded |
|---|---|---|---|
| L0 Control | Charging while powered off | Does it still cut out when powered off | Stable powered-off charging for more than 10 minutes |
| L1 Charger | Charger power and sharing | Is the power enough, and is it being shared | Single-port genuine charger comparison |
| L2 Cable | Cable and plug | E-marker, conductors, plug wear | Another known good compliant cable |
| L3 Port | Physical port condition | Tongue, springs, solder joints, dirt, oxidation | Visual inspection plus movement comparison |
| L4 Negotiation | CC and PD contract | Does negotiation repeatedly fail or fall back | Power meter readings and voltage behaviour |
| L5 Settings | Charge limits and power management | Is a limit pretending to be a fault | Switch each limit off and retest |
| L6 Sleep | Sleep and wake behaviour | Does it only fail after waking | Powered-off, asleep, and woken comparisons |
| L7 Battery | Battery and protection board | Fuel gauge, protection board, temperature sensing | Battery report plus percentage behaviour |
| L8 EC and firmware | Embedded controller state and firmware version | Is the controller stuck, or is a version at fault | Manufacturer reset or update procedure |
| L9 Mainboard | Charger IC and surrounding circuit | Input, control, sensing loop | Bench measurement by qualified personnel |
The framework carries one further rule: every layer conclusion must rest on an observation. An observation can be a power meter number, a BIOS display, a battery report figure, or a record of which action triggers a break and which does not. A conclusion without an observation is only a guess, and guessing in the wrong direction costs more than a wrong accessory purchase. It can turn a simple port problem into worsening solder damage through continued use at a forced angle, or send a pure settings problem to a bench teardown it never needed.
Port physics: why moving the cable breaks charging
Inside a USB-C receptacle sits a tongue carrying the contact points. When a plug is inserted, spring contacts inside the plug press against both faces of the tongue and form the electrical connection. This is a friction fit, not a screwed joint, so a bent, cracked, or thinned tongue produces uneven contact pressure; fatigued or deformed plug springs produce a plug that goes in but does not grip; and a loose receptacle shell gives the whole plug extra room to move. Repair community descriptions agree: a cracked or broken centre tongue, bent or missing contact pins, and separated receptacle solder joints are common forms of physical port damage, and a port that still fails after cleaning usually points to damage in the receptacle itself rather than dirt alone.
Physical wear causes intermittent charging not by breaking the whole circuit, but by making contact resistance vary with position, angle, and vibration. First, the effective contact area shrinks: once the tongue or springs are worn, contact resistance rises. Second, contact pressure falls short: once spring force drops, a small movement of the plug can lift one pin temporarily. Third, fretting takes hold: repeated micro-movement inside the receptacle wears the contact surfaces and thickens the oxide layer, pushing contact resistance higher still. Fourth, a vicious circle forms at high current: higher contact resistance causes local heating, metal expansion accelerates oxidation, and contact becomes still less stable. This circle explains why a fault that at first only appears on movement later appears even when nothing moves. Fifth, pins fail selectively: USB-C has many pins with different jobs for power, ground, and configuration. If only the configuration pin loses contact, charging may show up as negotiation failure or fallback; if only one power or ground group loses contact, the fault may appear only at high current and vanish at low current.
This matters for diagnosis: movement breaking charging does not prove that the whole port is broken, and it does not prove that the charger is broken. Whether the fault sits in the receptacle, the plug, the cable, or the mainboard solder joints can only be settled by layered exclusion, not by feel. The evidence is how much the plug moves inside the receptacle, how charging stability varies with angle, and what changes when the cable or port is swapped. Feel alone can only say that something seems loose; evidence must say how far it moves in which direction before charging breaks, and in which direction it does not break.
Durability figures need careful reading. USB Type-C specifications and connector maker data commonly use 10,000 mating cycles as a durability floor, with specified insertion and extraction force ranges. That figure is a floor under specification test conditions, not a promise that failure arrives at cycle 10,001, and not a promise that everyday use always reaches 10,000 cycles. Real life depends heavily on insertion angle, sideways force, cable pulls during a drop, dust, and oxidation. Sideways force damages the tongue and solder joints far more than straight insertion and removal. Mating cycle count is therefore not diagnostic evidence; movement and comparison results are.
Older or some business machines still use a round DC barrel jack, whose mechanism differs from USB-C but produces the same intermittency. The centre pin of a barrel jack loosens and shifts under long-term insertion and sideways force; the receptacle is fixed to the mainboard by solder joints or connected by a cable to a small board, and once a solder joint cracks, moving the plug makes the circuit connect and disconnect. Inspection looks for cracked, burnt, or discoloured solder joints, and moving the plug while measuring voltage fluctuation with a multimeter can confirm a mechanical fault. Such measurement must be done by qualified personnel, and readers should not open a machine to measure a mainboard themselves.
Dirt, oxidation, and physical damage can look alike but demand different handling. Compacted dust or lint prevents the plug from seating fully and leaves a visible gap; on powered-off visual inspection, dark compacted material is visible at the bottom of the receptacle. Oxidation or corrosion darkens or discolours the contacts and produces on-and-off charging that improves after cleaning but keeps returning. Tongue, spring, or solder damage shows up as obvious plug wobble, charging that only works at one angle, and no improvement from cleaning. Cleaning without distinguishing these three wastes time and risks pressing a still-intact tongue until it breaks.
USB-C CC pins and the four-step PD negotiation: why charging cuts in and out
USB-C charging does not deliver full power the moment a plug is inserted. Before negotiation is complete, the power source supplies only a safe initial voltage, generally 5V, and only then negotiates higher voltage and current through PD messages on the configuration channel. This design is itself a safety mechanism: the system must first confirm what the other end is, how much current the cable can carry, and how much power the device wants, before raising voltage and current. If any step in that conversation fails, the system stays at low power, falls back, or renegotiates repeatedly, and the user sees charging that cuts in and out.
The configuration channel has several key jobs. First, it detects insertion: the pull-up resistor on the source side meets the pull-down resistor on the device side, and only then does the system know that something has been plugged in. Second, it determines plug orientation: which of the two configuration lines carries a signal decides which pin group is used. Third, it advertises initial current capability: before PD negotiation begins, the pull-up resistance can advertise default USB power, 5V at 1.5A, or 5V at 3A. Fourth, it can supply VCONN on the unused line to power the E-marker chip inside the cable. If the configuration pin loses contact, oxidises, or is lifted by a bent tongue, the system repeatedly believes the device has been unplugged and plugged in again, and every reinsertion forces detection and negotiation to start over, with the charging light flashing on and then off before negotiation completes. This is the mechanical basis of movement triggering renegotiation.
The standard PD explicit contract process has four steps. First, Source Capabilities: the charger lists the voltage and current combinations it can supply as a set of power data objects, for example 5V, 9V, 15V, and 20V. Second, Request: the laptop chooses one combination from that list according to its own sink capability and replies with a request data object. Third, Accept: the source confirms that the request is valid and replies with Accept. Fourth, PS_RDY: the source completes its voltage transition, confirms that the voltage is stable, and sends PS_RDY to signal that power may now be delivered under the contract, at which point the explicit contract is formally established. During the transition, the device must keep its draw low until Accept and PS_RDY have been received, and only then may it return to normal draw.
If the configuration line breaks at any step, the process fails in a characteristic way: a missing acknowledgement triggers a repeat of Source Capabilities; a missing Request means negotiation failure; an Accept without PS_RDY means the supply is never confirmed ready; and a break after the contract is formed may trigger a hard reset back to the safe 5V state and a fresh negotiation. The user sees a charging icon that alternates between charging and not charging, or a power meter voltage jumping between 5V and 20V. High-power charging demands more configuration stability than low-power charging: a cable or port that seems fine at 5V may fail repeatedly at 20V, 28V, or 48V. The extended power range mode of PD 3.1 has keep-alive requirements, with the device expected to maintain communication with the source; if communication breaks, the source can hard reset to 5V.
Negotiation failure and fallback produce four intermittent patterns worth matching against experience. The table below sets out each pattern, its mechanism, and what the user sees.
| Pattern | Mechanism | What the user sees |
|---|---|---|
| Repeated renegotiation | Configuration contact fails mid-negotiation | Charging light flickers, meter voltage jumps |
| Fallback to 5V | Negotiation fails, system returns to safe voltage | Light on but extremely slow charging, battery may fall while in use |
| Fallback to a lower combination | Charger, cable, and device capabilities intersect at a lower combination | Charging works but clearly slower than the genuine combination |
| Intermittent hard reset | Communication breaks in extended power range mode, source returns to 5V | High-power charging suddenly breaks and restarts from 5V |
E-marker chips, PD 3.0, and PD 3.1: why a chipless cable fails at high power
An E-marker is a small chip hidden inside a USB-C cable plug. Its job is to declare to both ends the maximum current the cable can carry, the data speeds and functions it supports, and whether it supports extended power range modes. The power source can query the cable, confirm its capability, and only then decide how much current it may deliver. Put simply, an E-marker does not make a cable faster; it gives the system permission to deliver that much power. Without that declaration, the system must assume for safety that the cable can carry only a lower current.
A standard USB-C cable without an E-marker is generally limited to 3A, which is about 60W at 20V. Power above 60W requires a 5A E-marker cable, and 240W extended power range operation requires a 5A cable with an E-marker that supports extended power range. This directly explains a common intermittent symptom: using the same 100W charger with an unmarked cable, a laptop falls from 100W to 60W, or its battery falls during heavy load even while plugged in. The user may conclude that the charger or the battery has failed, when in fact the cable declaration has forced a fallback. Worse, a poor-quality cable may claim 100W while its conductors are thin and its E-marker data is inaccurate, so the system negotiates from false information, voltage sags under high current, heating follows, protection trips, and charging cuts in and out.
PD 3.0 and PD 3.1 compare as follows. PD 3.0 standard power range tops out at 100W, with fixed voltages of 5V, 9V, 15V, and 20V, and a programmable power supply mechanism spanning 3.3V to 21V in fine steps. PD 3.1 adds an extended power range with fixed voltages of 28V, 36V, and 48V, topping out at 240W, and an adjustable voltage supply mechanism spanning 15V to 48V. Programmable power supply lets the source adjust voltage in fine steps to match the device charging algorithm, reducing conversion loss and heat, while adjustable voltage supply serves extended power range operation. The key warning is that PD 3.1 does not mean every device supports 240W; the charger, the device, and the cable must all support the same extended power range level before it can be used.
| Item | PD 3.0 | PD 3.1 |
|---|---|---|
| Maximum power | 100W (20V at 5A) | 240W (48V at 5A) |
| Fixed voltages | 5V, 9V, 15V, 20V | Plus 28V, 36V, 48V |
| Adjustable voltage | Programmable power supply: 3.3V to 21V | Programmable power supply plus adjustable voltage supply: 15V to 48V |
| Cable requirement | Above 60W needs a 5A E-marker cable | Extended power range needs an extended power range E-marker cable |
Several further cable traps deserve separate treatment. USB-C is only a plug shape, not a capability statement; charging ability, data speed, and video output must be checked separately, and the port diagram in the manual decides which port supports charging. A 240W cable can still carry data at USB 2.0 speed only, so it may charge at full rating while struggling with large file transfers, external storage, or displays; conversely, a very fast data cable may still be rated for only 60W of charging. Fast file transfer therefore does not prove that a cable can charge a laptop, and fast phone charging does not prove that it can charge a laptop either. Some cables omit data conductors to save cost or thickness, keeping only charging conductors; such a cable may charge a phone without trouble while PD negotiation or E-marker communication with a laptop is affected. The practical test is always the same: a result matrix plus a power meter, and if a fully marked E-marker cable restores stability with the same charger and laptop, the old cable is the prime suspect.
Charger fallback: underpowered chargers, dual-port sharing, overheating, and genuine identification
If a laptop is designed for 90W and its charger can only supply 65W, several outcomes are possible. The system may charge under a 65W contract at a slower rate. Under heavy load, such as gaming, rendering, or simultaneous charging and high-performance work, 65W may be insufficient for system draw plus battery charging at the same time, so the battery falls while plugged in, producing the familiar plugged-in-but-discharging symptom. Some models display a BIOS or system message warning that charger power is insufficient. Manufacturer support documents note that slow charging or discharging while plugged in can be caused by a charger that cannot meet device demand, and that the remedy is to use the charger and cable the device was designed for. A dock or hub also consumes power, so the power reaching the laptop through a dock is less than the charger rating; for example, a 130W charger through a dock may deliver only 90W to the laptop. The same principle applies to dual-port and multi-port chargers.
A multi-port charger rating usually describes the total across all ports, not the power each port can deliver simultaneously. Once a second device is plugged in, the charger redistributes power, and the laptop port contract may fall immediately. As a published example, a 65W dual-port charger may deliver 65W on a single port, fall to 45W on the main port with 20W on the second port when a second device is added, and fall further when a third device is added. For intermittent charging, this matters in a very practical way: a user charging a laptop who plugs in a phone as well may see laptop charging power drop suddenly, possibly triggering renegotiation, with a brief charging interruption visible as a flickering icon. If laptop load is high, the reduced power may be insufficient, the battery starts to fall, and the user concludes that charging is broken when the cause is only sharing. The diagnostic rule is therefore strict: when testing intermittent charging, unplug every other load from the charger first and leave only the laptop, so that sharing is excluded.
Overheating fallback has two sides. A charger that overheats reduces its output power or pauses output to protect itself, and poor ventilation, covering by bedding or a bag, or sustained heavy load accelerates overheating. Inside the laptop, high temperatures at the charger IC, battery, or mainboard cause the system to reduce charging current or pause charging until temperatures fall. Manufacturer documents also list high-temperature protection blocking charging among the causes of a plugged-in-but-not-charging state. The typical overheating pattern is distinctive: charging is normal when cold, becomes intermittent after the machine has been used for a while and has warmed up, recovers after a cooling pause, and is worst in summer, without air conditioning, or when the machine is used on a bed or on a lap. That pattern differs from physical looseness, which follows movement, while overheating follows time, temperature, and load. Distinguishing them requires recording all three variables together: whether anything moved, how long the machine had been running, and how hot it felt.
Genuine identification mechanisms deserve their own section. Taking a Dell round-plug charger as an example, beyond positive and negative conductors there is a centre pin connected to an identification chip inside the charger, which declares charger power and model to the laptop. The BIOS checks this information; if identification fails, the BIOS reports that charger power and model cannot be determined, and the laptop may refuse to charge the battery, throttle the system to match an unconfirmed supply, or show the charger state as Unknown. If the centre pin bends, breaks, or oxidises, or if the matching contact inside the receptacle loosens, the identification signal becomes intermittent, producing a laptop that sometimes recognises the charger and sometimes does not. The mechanism differs from USB-C configuration detection, but the logic is identical: an unstable identification signal makes the system unwilling to use full power, so charging becomes intermittent or the system slows down. For Dell models, diagnosis should include checking charger state in the BIOS and watching whether moving the plug changes that state from normal to Unknown.
Reading a power meter: how voltage, current, and power patterns map to layers
A USB-C power meter sits between the charger and the laptop and displays voltage, current, and power in real time, with higher-end models also showing the PD protocol, power data objects, and E-marker information. Normal readings have clear characteristics: voltage starts at 5V after insertion, rises to the target voltage such as 20V once negotiation completes, and then stays stable; current changes smoothly with battery level and system load rather than repeatedly dropping to zero; and power, equal to voltage multiplied by current, roughly matches the charger rating and laptop demand. Test conditions must be held constant: the same laptop, the same charger, and the same state, ideally idle, without a large program running at the same time, because load changes alone make wattage jump. Three sets of numbers should be recorded: the stable value at rest, the instantaneous value when the cable is moved, and whether wattage falls before dropping to zero just before a break.
Need a laptop repair expert?
Free diagnosis · Transparent pricing · No fix, no fee · 180-day warranty
Abnormal readings map to layers as follows. Voltage jumping repeatedly between 5V and 20V points to unstable configuration negotiation, including the port, the cable, and the configuration pin. Voltage stable at 20V with current repeatedly falling to zero and recovering points to intermittent disconnection in the charger IC, MOSFET, or protection board. Voltage stable but current persistently very low, for example 0.1A to 0.3A, points to fallback, trickle charging, a full battery, a charge limit setting, or temperature protection. Voltage falling clearly under load, for example from 20V to below 17V, points to excessive cable resistance, insufficient charger power, or high contact resistance. Readings jumping immediately when the cable is moved point to a physical contact problem in the port, cable, or solder joints. Readings returning fully to normal after a cable swap point to the original cable. Readings remaining abnormal after both cable and charger swaps point to the laptop side, including the port, mainboard charging circuit, and battery.
| Meter behaviour | More likely layer | Next step |
|---|---|---|
| Voltage stuck at 5V, low current, power far below charger rating | PD negotiation failure, cable without E-marker limited, charger profile mismatch | Retest with an E-marker cable rated high enough |
| Voltage and current intermittently zero, movement triggers it | Cable or port contact interruption | Localise by movement test: plug or cable body |
| Voltage stable, current rising and falling, power fluctuating but not zero | System load change, battery near full reducing current, charging management intervening | Check battery report and system charging state, do not blame the cable at once |
| Voltage suddenly falls, power falls but current does not reach zero | High cable resistance, high contact resistance, oxidised port causing voltage drop | Inspect the port and run a cable comparison |
| Immediate zero on connection, charger light off, abnormal machine heating | Possible short circuit or internal supply protection triggered | Stop use at once and send for inspection, do not keep replugging |
| Abnormal readings persist after cable and charger swaps | Laptop side problem | Move to port inspection, battery, and mainboard layers |
The movement test done properly, and cleaning rules
The purpose of a movement test is not to shake the plug until it breaks, but to confirm with the smallest possible movement whether movement is a trigger. First place the machine on a stable table, plug the charger firmly into a wall socket, and let the cable hang naturally without tension or a sharp bend. Record the resting state for two to three minutes: whether the charging icon flickers, whether the charging light goes out, and how the battery percentage moves. Then make only small movements, one at a time, pausing to observe after each: gently move the plug up and down without twisting sideways or rocking hard; gently push in or pull out by about one to two millimetres without fully unplugging; and gently move the cable segment near the plug, since bends in the cable body are where internal breaks most often occur. After each movement, record which movement triggered a break and which had no effect.
Reading the result is straightforward. If only moving the plug breaks charging while rest is completely stable, port contact or plug wear is the prime suspect. If only moving one segment of the cable body breaks charging while moving the plug does nothing, an internal break in that cable segment is the prime suspect. If charging breaks without any movement at all, movement should no longer be blamed, and attention should turn to the battery, mainboard, and software or firmware layers. Movements that must never be used include rocking, twisting, or prying the plug hard, since USB-C tongues and barrel jack solder joints are fragile, and violent testing can turn simple poor contact into a cracked solder joint; probing the port with metal objects; and testing a cable that is knotted or repeatedly bent at the same point. Manufacturer documents remind users to grip the plug rather than pulling the cable body when unplugging, because pulling at bends is a common cause of cable damage.
A port inspection checklist follows, to be done only after powering off, unplugging the charger, and letting the machine cool, using strong light plus a magnifier or phone macro mode rather than guesswork. For a round DC jack, check whether the inner pin is bent, crooked, broken, or pushed inward; whether the inner wall is cracked, loose, or deformed; whether the plug wobbles clearly from side to side after insertion; whether the housing around the receptacle shows scorching, melting, or odour; and whether insertion depth matches what it used to be, since a plug that cannot seat fully may be blocked by debris. For a USB-C receptacle, check whether the centre tongue is bent, crooked, cracked, or chipped; whether the gold contacts on both faces of the tongue are blackened, green, or whitened, which suggests oxidation or corrosion; whether lint, dust, food fragments, or bag fibres are inside; and whether the plug seats with a firm feel rather than sliding out easily or wobbling widely. Where a machine has two USB-C ports, compare how the two plugs feel, but remember that a charging-capable port and a non-charging port may legitimately differ in ability, so a difference alone does not prove a faulty port.
Looseness should be described concretely when reporting it for repair, not summarised as very loose. A useful description states how far the inserted plug can move up, down, or sideways, in which direction movement breaks charging, whether the plug slides out on its own, and whether it must be held by hand to charge. Looseness does not automatically prove that the receptacle itself is faulty: a worn plug, or lint preventing full seating, can imitate looseness, so debris should be excluded before the receptacle is blamed.
Cleaning, done correctly, follows a short sequence: power off and disconnect; use short bursts of compressed air with the can upright, never tilted, with the receptacle facing downward so debris falls out rather than being blown deeper; remove stubborn lint only with a non-metallic tool along the inner wall, avoiding the centre tongue and without prying; and if oxidation marks remain, wipe the receptacle edge very lightly with a cotton swab carrying only a tiny amount of isopropyl alcohol, without pouring liquid in, and wait until completely dry before powering on or charging. Cleaning rules that must never be broken include never using a metal pin, paper clip, tweezers, or sharp tool to pick at pins or the tongue, which risks short circuits, scratched contacts, and a bent tongue; never cleaning while powered on or plugged in, when short-circuit risk is highest; never blowing with the mouth, since moisture accelerates corrosion; never pouring water, large amounts of alcohol, or any cleaning agent into the receptacle; never forcing a bent tongue straight again, which is very likely to snap it and leave fewer repair options; and never continuing to plug and unplug after seeing scorching, melting, odour, sparks, or abnormal heating, which calls for immediate stoppage and inspection. This article does not provide detailed do-it-yourself internal cleaning steps; internal inspection belongs to qualified repair personnel.
Light and icon misunderstandings, and how Windows and macOS report charging
Misreading the charging light is among the most common diagnostic errors. A light on means only that voltage is present or that the charger IC is active; it does not prove that a PD contract has been formed, that charging current is sufficient, or that the battery is actually charging rather than the system merely being powered. System documents distinguish slow charging, trickle charging, and failure to charge, showing that internal judgements of charging capability can differ from what the eye sees in a light. The correct method is to read power meter figures together with system battery state and battery percentage movement, all three at once, before judging whether charging is normal.
Windows users should distinguish three states carefully. Charging plus a rising percentage is genuinely charging. Plugged in but not charging, or plugged in with a percentage that does not move, means only that Windows has detected an external supply; it does not prove that charger power is sufficient or that the charging circuit is normal, and the message alone does not prove a battery driver fault. Charging with a falling percentage means input power is below system draw, which is charging but too slowly. For a battery report, running the battery report command as administrator prints a report path, usually in the user folder, and opening that HTML file in a browser shows the figures that matter: design capacity against full charge capacity, recent usage, and battery capacity history. The design against full charge gap reflects wear, with a gap of roughly two to three tenths after years of use commonly described as ordinary wear, while a larger gap points more strongly toward replacement; this article sets no hard percentage threshold, because model, cycle count, and environment all matter. The recent usage timeline shows whether a plugged-in battery intermittently falls, which can be matched against movement test times. The capacity history shows whether capacity has declined steadily or changed suddenly after one update or event. The report has limits: it shows battery capacity and usage history, but not charger model, and it cannot prove maximum charger output, so testing with a known good compliant charger remains more useful than reading percentages alone.
| Icon or message | Real meaning | Common misreading |
|---|---|---|
| Plugged in, charging | System believes it is charging, but not necessarily at full speed | Assumed to mean full-speed charging |
| Plugged in, not charging | May be a charge limit, full battery, temperature protection, or insufficient charger power | Always assumed to mean a dead battery |
| Slow charging | System judges charging power below expectation | Assumed to mean a broken charger, when the cause may be cable fallback |
| Not charging | System judges that charging cannot proceed | Assumed to mean a broken mainboard, when the cause may be a setting or identification problem |
| Charging but percentage falling | Input power below system draw | Assumed to mean no charging at all |
At the driver layer, the battery and mains adapter entries in Device Manager report battery and charger state to Windows, and the common procedure is to uninstall them and restart so Windows reinstalls them automatically. Some models carry extra limits, so the model manufacturer guidance should be confirmed before copying a generic procedure. The test standard is practical: if the charging icon recovers after driver reinstallation but actual percentage behaviour stays intermittent, the problem is not in the driver layer; if nothing changes at all, repeated reinstallation should not be continued, and attention should move to hardware layers.
Charge limit settings that pretend to be faults
A charge that stops at the same percentage every time is very often a charge limit setting rather than a fault. Manufacturer battery health features deliberately hold charging at a set level to extend battery life, for example a primarily mains-powered mode holding charging at a lower ceiling. A user who sees charging stop at the same level every time and never reach full charge may conclude that the battery is worn out or the port is broken. The correct check order is to open the battery management software for the machine brand, one item at a time, look for an active limit, switch the limit off temporarily or set it back to full charge, and retest overnight. If full charge returns after the limit is switched off, the earlier stop was a setting, not a fault, and no battery or port replacement is needed.
| Brand or system | Setting concept | Typical limit behaviour |
|---|---|---|
| Lenovo | Battery charge threshold, conservation mode | Holds at a lower level, does not reach full charge |
| Dell | Primarily mains-powered and related battery settings | Charging stops at a set ceiling, BIOS settings also relevant |
| ASUS | Battery health charging balance mode, maximum lifespan mode | Stops at a lower ceiling depending on mode |
| Windows | Smart charging where the model supports it | Automatically sets a lower charging ceiling |
| macOS | Optimised battery charging | Learns usage habits, sometimes holds at a lower level |
| BIOS generally | Battery charge configuration, custom ceiling | A custom low ceiling prevents reaching full charge |
Embedded controller, fuel gauge error, protection board, and temperature sensing
The embedded controller is a small microcontroller on the mainboard that stays powered long-term and manages low-level functions including power sequencing, fans, keyboard, and battery communication. The BIOS handles booting and system settings, while the embedded controller handles continuous power management before and after booting. Their firmware is usually updated together, and manufacturer documents note that an embedded controller update follows a BIOS update immediately, with the screen staying black and the power light blinking during the process, and power must not be interrupted. When the embedded controller misbehaves, charging logic may stick in a wrong state, for example believing the battery is full or believing the charger has been unplugged; battery communication may break intermittently, so the system sometimes sees the battery and sometimes does not; and temperature or current sensing data may be wrong, triggering incorrect protection actions. Embedded controller failure does not necessarily mean the controller chip itself has burnt out; more often it is a stuck firmware state, an incomplete reset, or a BIOS and controller version mismatch. The general idea of an embedded controller reset is to remove power and press a reset hole for a period or hold the power button, resetting controller state, but the exact operation differs by brand and must follow the official guidance for that model.
The fuel gauge chip inside the battery estimates remaining charge from coulomb counting and voltage curves. Long periods of shallow charging and discharging within a narrow range, or long periods plugged in without battery use, let the estimate drift away from true capacity, producing percentage jumps such as a sudden fall from 40 percent to 20 percent; a display that still shows charge followed by immediate shutdown on unplugging; a display of full charge with very short usable time; or charging that stalls at one percentage. Recalibration can only correct the reported capacity; it cannot restore lost true capacity, and if the battery is already worn, the reported capacity after calibration may fall, because the number becomes accurate. Manufacturers provide a battery gauge reset procedure using an official tool and a full charge and discharge cycle to relearn full and empty points, a process that may take several hours and should not be interrupted. The key distinction is that fuel gauge drift usually appears together with percentage jumps and disappears when the battery is replaced; if the problem persists after battery replacement, the mainboard charging circuit or embedded controller should be suspected, not the battery itself.
The battery protection board handles overcharge protection, cutting charging when any cell voltage runs too high; overdischarge protection, cutting output when voltage runs too low; overcurrent and short-circuit protection, cutting off on abnormal current; overtemperature protection, cutting off on high temperature; and cell balancing, keeping series cell voltages consistent. Its failure modes include cold or cracked solder joints making the board connect intermittently; ageing of the protection MOSFET raising internal resistance so that voltage drop under high current triggers protection; and cell imbalance letting one cell reach an overcharge or overdischarge threshold early, so the whole pack repeatedly cuts off. Typical intermittent protection board behaviour is charging that stops at one percentage and resumes after cooling; breaking under heavy load but working under light load; and a battery that appears and disappears in the system. Such problems cannot be solved by swapping chargers or cables alone and require battery-level inspection.
Temperature sensing failure deserves separate treatment. Battery packs and mainboard charging circuits usually measure temperature with a thermistor, and the charger IC sets charging current from temperature: too cold, and charging current falls or charging pauses to avoid damage to lithium cells charged at low temperature; too hot, and charging current falls or charging pauses to avoid thermal runaway; and a broken, shorted, or poorly connected sensing line makes the charger IC read a false temperature, so it pauses charging in error. Intermittent temperature sensing often follows movement or temperature change: moving a battery cable or mainboard makes the temperature reading jump and charging follows; heating makes the reading abnormally high and charging pauses; cold behaviour is normal and hot behaviour fails. Diagnosis should check whether the system or inspection tool battery temperature is plausible, for example not extremely hot or cold at room temperature, and whether the reading jumps with movement.
Battery swelling is a safety matter and must be flagged separately. Lithium polymer batteries can swell, and a swollen battery may affect machine performance and internal parts; manufacturer guidance is to stop use, disconnect the charger, let the battery discharge, and have it handled by authorised repair. Externally visible signs that need no opening include a raised keyboard or trackpad that feels different when pressed, a bottom cover that will not close, one side of the machine bulging, hinge pressure, abnormal gaps, a battery report showing a sudden capacity fall, and abnormal machine heating. The safety rules are strict: never press, bend, or pry a swollen battery with tools; never try to force a swollen battery back into the machine; and never puncture, drop, or strike a battery. On seeing swelling, stop use and send the machine for inspection, without further movement or battery removal tests.
Mainboard charger IC, MOSFET, and sense resistor: measurement layers
Taking a common charger controller family as an example, the charging circuit divides labour as follows. The charger IC controls charging current in a closed loop, adjusting pulse width from battery voltage, charging current, and charger current. MOSFETs driven by the charger IC convert charger voltage into battery charging voltage and current, and other MOSFETs select the system power source between charger and battery. The sense resistor, a low-value resistor in series with the charger input and battery output paths, lets the charger IC infer actual current from the tiny voltage difference across it. Charger detection checks whether charger voltage is sufficient; if detection voltage falls below an internal reference, the system judges charger power insufficient, switches to battery power, and enters a low-power sleep mode. This division matters because a fault in any one of the three, charger IC, MOSFET, or sense resistor, can make charging current uncontrolled or misjudged, and any of them can present as intermittent charging.
| Fault location | Mechanism | Intermittent behaviour |
|---|---|---|
| Charger IC itself damaged | Internal pulse control, reference voltage, or protection logic fails | Charging sometimes present, sometimes absent, current unstable |
| Charger IC supply unstable | Supply pin voltage fluctuates, bypass capacitor fails | Normal when cold, intermittent when hot |
| Charger IC to controller communication broken | Communication bus intermittently breaks | System sometimes sees the charger, sometimes does not |
| Charger detection abnormal | Divider resistor value changes, leakage | System wrongly judges the charger unplugged, repeatedly switching charger and battery supply |
| MOSFET resistance rising | Ageing or partial damage raises conduction resistance | Works at low current, fails at high current, heats severely |
| MOSFET drive abnormal | Gate drive signal unstable | MOSFET switches on and off, charging current follows |
| Sense resistor value drifting | Value changes from heat, oxidation, or solder problems | Charger IC reads current wrongly, limits early, charging slow or intermittently limited |
| Sense resistor solder joint cracked | Voltage difference reading intermittent | Charger IC control loop oscillates |
Measurement tools must be layered. A multimeter serves two distinct purposes that must not be confused: powered-off measurement of resistance and continuity, after shutdown, disconnection, and battery removal, checking continuity from receptacle pins to mainboard solder joints for open or short circuits; and powered-on measurement of voltage, with the charger connected, measuring receptacle solder joints, charger IC input, and battery-side voltage to see at which layer voltage disappears. Voltage fluctuation while moving the plug points to a mechanical fault. An oscilloscope shows waveforms and suits cases where charger IC pulse waveforms are unstable, MOSFET gate drive signals disappear intermittently, configuration line PD communication waveforms carry noise or breaks, or voltage falls for microseconds or milliseconds too fast for a meter to show. An oscilloscope is not needed for every intermittent charging case; it becomes the next layer when meter and multimeter readings show nothing abnormal yet the problem persists. The safety warning stands: powered-on mainboard measurement carries short-circuit risk and must be done by qualified personnel in an electrostatic-safe environment, and readers should not open a machine to measure a mainboard themselves.
The mainboard exclusion logic runs from outside inward, confirming voltage and signals layer by layer: charger output normal excludes the charger; cable plus meter readings normal exclude the cable; receptacle solder continuity normal excludes the receptacle; charger IC input voltage normal excludes the input path; charger IC output or battery-side voltage abnormal points to the charger IC, MOSFET, sense resistor, or battery protection board; and charger IC input and output both normal with abnormal system reports points to controller communication, fuel gauge, or temperature sensing. Every layer conclusion must rest on measurement data, never on reasoning of the form that a new battery did not help so the mainboard must be at fault.
Sleep, wake, Modern Standby, and BIOS and controller firmware
Traditional sleep powers down most parts and keeps only memory, while Modern Standby keeps the system in a low-power but still partly awake state for instant wake, notifications, and background updates. Their effects on charging differ. In traditional sleep, system draw is extremely low, almost all charger power goes to charging, and charging behaviour is more stable. In Modern Standby, the system still has background activity and higher draw; if background programs repeatedly wake the system, charging power is shared with system draw and may even produce the illusion of battery falling while plugged in. After waking, the system returns from low power to full speed, instantaneous draw rises, the charger IC may need to redistribute power, and charging current fluctuates briefly. System documents note that newer versions add protection: if Modern Standby power draw is found excessive, most wake sources are disabled, leaving only the power button or lid opening to wake the machine, preventing background waking from draining the battery.
Intermittent charging in sleep must be separated into three cases. Charging normal during sleep but intermittent after waking points to wake-related load, drivers, or controller state, and calls for comparison across sleep, wake, and powered-off charging. Intermittent during sleep itself but normal when powered off points to sleep power management or Modern Standby background activity, with powered-off charging as the control. Intermittent even when powered off has nothing to do with sleep and points to hardware layers including the port, cable, and charging circuit. The diagnostic rule repeats here: powered-off charging is the cleanest control group, and if it fails intermittently, hardware is very likely at fault without further circling through system settings.
| Case | Behaviour | How to separate it |
|---|---|---|
| Normal asleep, intermittent after waking | Problem follows wake load, drivers, or controller state | Compare asleep, woken, and powered-off charging |
| Intermittent asleep, normal powered off | Problem follows sleep power management or background activity | Powered-off charging as control |
| Intermittent powered off | Problem unrelated to sleep, points to hardware | Powered-off failure largely excludes system power management |
Firmware handling rules are: record the current BIOS and controller versions first; download firmware only from the manufacturer official site for the exact model; keep the machine on stable power during an update and never interrupt it; retest powered-off, powered-on, and sleep charging after updating; and if the problem worsens after an update, consider firmware rollback only under manufacturer guidance, never with unofficial tools. Firmware updating is not a cure-all: if the problem is physical looseness, a cracked solder joint, or a damaged charger IC, firmware cannot repair hardware. Conversely, if a problem appears only in one firmware version and disappears on rollback, firmware is implicated. One special warning applies: if the current symptom is itself intermittent power loss, the risk of power interruption during an update is higher than usual, so supply stability should be confirmed by inspection personnel before updating, or the machine should be sent for inspection rather than updated at home. Important data should be backed up before updating, as covered in the data safety section.
Hong Kong context: humidity and oxidation, commuting wear, and third-party accessories
Hong Kong climate is humid, and laptop charging port metal contacts kept long-term in high humidity oxidise more readily. An oxide layer raises contact resistance and makes previously normal contact unstable, particularly where a machine sits long-term near a window, an air conditioner outlet, or a damp room; where a port has been exposed to water, sweat, or drinks, since residue can accelerate corrosion even after wiping dry; and where a plug stays inserted for long periods, letting an oxide layer thicken on the contact faces until intermittent disconnection begins one day. Repair community cleaning guidance notes that oxide and oil residue cannot be removed by compressed air alone and need light wiping with a small amount of alcohol. This section is a mechanism inference from environment, not measured data: this article claims no humidity figure causing any failure rate, and no particular model as especially prone to oxidation in Hong Kong. Whether oxidation is present must rest on visual inspection for blackened, green, or whitened marks.
Common Hong Kong usage patterns include carrying a laptop to work or school daily, plugging in in the morning and unplugging at night, at least twice a day; using the machine on transport, in cafes, and in shared workspaces, where insertion angle is not always vertical; keeping the machine in a backpack where the cable is pressed, pulled, and snagged by other items; and using the machine while charging, with the cable bent long-term. These patterns accelerate tongue, spring, and solder wear. Damage from sideways force is worst when a machine is used on a bed or on a lap, where the cable pulls sideways hardest. Manufacturer documents also warn that pulling the cable body rather than the plug, and incorrect cable wrapping, damage chargers and cables. Practical wear reduction is simple: grip the plug and insert and remove straight, never at an angle; never close the machine or pack it away with the cable still plugged in; and keep one cable at home and one at the office to reduce daily repeated insertions of the same cable. These are usage suggestions, not repair conclusions.
The Hong Kong market offers many third-party cables and chargers with wide price differences. Realistic trade-offs for buyers include genuine chargers costing more but matching the device best in power, PD combinations, and identification; compliant third-party chargers with complete PD combinations, E-marker cables, and overheat and overcurrent protection that can in theory work normally; and unmarked, uncertified, or falsely rated cables and chargers that most readily produce negotiation failure, fallback, and intermittent disconnection at high power. This article evaluates no particular brand or shop and provides no purchase links. The diagnostic rule is to run a comparison test with a genuine or compliant charger plus a compliant E-marker cable first, so that third-party factors can be excluded; if the problem disappears in that comparison, the original third-party cable or charger is implicated, not the laptop itself.
Shared socket voltage instability stories must be verified, not treated as fact. Hong Kong standard mains supply is 220V at 50Hz, which is a standard value, not a measured value at every socket at every moment. Claims that a particular cafe or shared workspace has unstable voltage and therefore damaged a machine can only be listed as user reports or unverified until measured data exists, and must not be written as a cause. User-level verification means testing the same laptop, cable, and charger at a home socket and at the suspected socket and comparing meter readings and intermittent frequency; changing to another socket position to exclude one loose socket or an ageing power strip; switching to a charger plugged directly into a wall socket if a power strip is suspected; and recording time, place, socket position, and test results to hand over at inspection. Voltage instability should not be used as a conclusion to demand compensation or assign blame without qualified measurement, and hearing that one place has a problem does not prove that a particular intermittent fault came from there, since port, cable, battery, and mainboard possibilities must still be excluded one by one.
Misdiagnosis traps: the directions most easily guessed wrong
| Misdiagnosis | Apparent evidence | Real direction | Verification |
|---|---|---|---|
| Replacing only the battery | Slow charging, jumping percentage | Fuel gauge, protection board, and charging circuit still unexcluded | Problem persists after replacement, return to layered table |
| Replacing only the charger without checking the port | Still breaks after charger swap | Tongue, spring, or solder looseness | Meter readings jumping on movement |
| Light on means normal | Light on but battery falling | Only 5V safe voltage or trickle charging | Meter plus percentage movement |
| PD fallback mistaken for battery failure | Never fills, slow speed | Cable without E-marker, dual-port sharing, configuration negotiation failure | Compliant cable and single-port charger comparison |
| System icon misreading | Plugged in but not charging | Charge limit, temperature protection, insufficient power | Switch each limit off and retest |
| Percentage only, no meter | Slow percentage movement | Fallback, not battery failure | Meter reading comparison |
| One cable only tested | One cable faulty | Multiple compliant cables needed for comparison | Result matrix |
| One socket only tested | Loose socket, ageing power strip | Wall socket comparison | Charger directly into wall socket |
| Temperature factor ignored | Normal cold, intermittent hot | Overheating fallback, not loose port | Record time, temperature, load |
| Sleep factor ignored | Normal asleep, intermittent after waking | Power management or drivers, not charger IC | Powered-off, asleep, and woken comparisons |
Repair, replacement, and upgrade decisions, and price wording
Decision principles are fourfold. Work from outside inward: exclude charger, cable, and settings first, then consider the port, and only then the mainboard. Decide only on data: every decision should rest on meter readings, BIOS state, or inspection results, not on feeling. Compare repair cost with remaining value: if repair cost approaches replacement cost, or the mainboard shows multiple ageing faults, replacement deserves serious consideration. Put data safety first: before any teardown or mainboard work, confirm that data is backed up.
Receptacle soldering or replacement is a reasonable first choice where plug movement is clearly abnormal and movement breaks charging immediately; where cable and charger swaps change nothing but moving the receptacle improves charging; where visual inspection shows a bent tongue, cracked solder, or a loose receptacle; where meter readings jump with movement while other layers test normal; and where the rest of the machine works normally and its remaining value justifies repair. Receptacle soldering or replacement charges follow the written quotation after inspection. Mainboard-level cases include all of receptacle, cable, charger, and battery testing normal while charging stays intermittent; charger IC input normal with output abnormal; MOSFET or sense resistor measurements abnormal; controller communication abnormal or temperature sensing abnormal; and powered-off charging intermittent, excluding system and sleep factors. Mainboard chip-level repair pricing follows the table below.
Replacement deserves consideration where the mainboard shows multiple ageing, corrosion, or burn marks leaving further hidden risks after repair; where charger IC, controller, and PD controller faults appear together; where an ageing machine needs battery, screen, and keyboard work at the same time, bringing total cost close to replacement; where the manufacturer has discontinued that model mainboard, making parts scarce and repair slow; and where user needs for performance, battery life, or ports have outgrown the old machine, so repairing charging alone does not meet the need. Replacement decisions should never rest on the charging problem alone, but on whole-machine condition, data migration cost, and actual user need. Upgrade considerations belong only with replacement or major repair: if replacing, prioritise port durability, sufficient charger power, and PD support matching future need; if undertaking major repair, check battery health, cooling, and solder joints around the port for other ageing signs; and never buy a higher-power charger blindly for faster charging, since charger power must match device PD capability, or it will simply fall back and charge no faster.
| Item | Wording in this article |
|---|---|
| Battery | Battery from HK$580 |
| Screen | Screen from HK$1,280 |
| Mainboard chip-level repair | Mainboard chip-level repair from HK$880 |
| Keyboard | Keyboard from HK$480 |
| Port soldering, port replacement, charger IC replacement, controller repair, and all other items | As per written quotation after inspection |
Pre-service checklist and data safety reminders
The information to provide at inspection is: how often symptoms appear, how many times per day and for how long each time; which action triggers a break, including moving the cable, moving the machine, opening a particular program, or machine heating; comparison results across powered-off, powered-on, and sleep charging; which charger, cable, and socket were tried; power meter readings if available; BIOS charger state, normal or Unknown for Dell models; and whether BIOS or controller firmware was recently updated, or the machine was dropped or exposed to water. This information is not a diagnosis; it only helps inspection personnel locate the right layer of the layered table quickly.
Items that must be taken to inspection include the cable normally used, even if it is suspected faulty, so personnel can run comparisons; the charger normally used; a second cable or charger if one exists, so a result matrix can be run directly; and a photograph of the socket or power strip if the problem only appears at one socket or strip. Taking only the machine and saying the home accessories were left behind leaves inspection personnel unable to reproduce the reported intermittency, and the inspection result may come back as normal on site, wasting an inspection.
| Pre-service item | Bring or prepare | Reason |
|---|---|---|
| Genuine charger and usual cable | Take both | Comparison testing needs the original combination |
| Second cable and charger if available | Take both | Result matrix needs multiple combinations |
| Symptom record | Frequency, triggers, three-state comparison | Fast location of the right layer |
| Power meter readings if available | Resting value, movement value, pre-break value | Separating negotiation layer from hardware layer |
| BIOS charger state | Normal or Unknown | Separating identification layer problems |
| Battery report highlights | Design capacity, full charge capacity, cycle count | Separating battery layer problems |
| Charge limit settings | Which limit was on, whether switching it off helped | Excluding false faults |
Signals that should stop self-testing at once and trigger immediate inspection are abnormal heating near the charging port, burning odour, housing deformation, a loose port or internal rattling; visible oxidation, green corrosion, water marks, or odour at the port; no encryption recovery key at hand while a BIOS change is needed; no response at all after a firmware update power interruption; and a charger light going out immediately on connection, abnormal machine heating, or burning smell. On any of these, stop self-testing and send the machine for inspection. In summary: intermittent charging should not trigger an immediate new charger, new cable, battery blame, or mainboard blame. First lock the layer from the symptom: movement breaking charging points to port and cable first; jumping voltage points to configuration negotiation first; fallback to 5V points to E-marker and sharing first; normal cold and intermittent hot points to temperature first; failure only after waking points to power management first; failure even powered off points to hardware first. Change one item at a time, settle data safety before changing anything, and test all states completely after each change. With complete evidence, repair, replacement, or upgrade finally has a correct answer. Pricing information is at pricing, inspection arrangements are at diagnostic, and further details are at pricing and diagnostic.
Laptop King | Room 1209, Dalton Plaza, Shek Mun, Sha Tin | Tel +852 2186 8010 | WhatsApp +852 6558 6806 | Hours 12:00–19:00
Frequently asked questions
Q1: I tried another cable and it still fails, does that exclude the cable?
A1: Not necessarily. If the second cable is also a USB 2.0-only charging cable, also lacks an E-marker, or is also a long passive cable, the cable test changed no key variable. A valid cable test must use a cable of a clearly stated specification and a different grade, such as the short genuine cable supplied with the device or a certified short cable, before the cable can be treated as excluded.
Q2: The charging light is on, does that mean charging is normal?
A2: No. A charging light usually reflects only that voltage is present or that the charger IC is active, not the full state of the PD contract. Safe 5V voltage alone, a contract that forms and then breaks, or trickle or slow charging can all leave the light on. Power meter readings plus system battery state plus percentage movement, all three together, are needed to judge.
Q3: Moving the cable breaks charging, does that mean the whole port is broken?
A3: No. Movement breaking charging only proves that movement is a trigger, not where the fault sits. It may be receptacle tongue or spring wear, plug wear, an internal break at a cable bend, or a cracked solder joint. A small-movement test localises it by segment: only moving the plug breaking charging points most to the port, while only moving one cable segment breaking charging points most to that cable segment.
Q4: The same charger charges a phone quickly but charges a laptop intermittently, does that mean the laptop is broken?
A4: Not necessarily. Phone power demand is low and laptop power demand is high, so a cable or charger that is stable at low power may fail at high power through E-marker limits, conductor resistance, or sharing fallback. Test with a genuine or compliant charger plus a compliant E-marker cable on the laptop, and do not infer laptop charging ability from phone charging behaviour.
Q5: Plugging a phone into a dual-port charger suddenly slows laptop charging or makes it flicker, is that a fault?
A5: Not necessarily a fault, and very likely sharing fallback. A multi-port charger rating is a total across ports, so adding a second device redistributes power, the laptop port contract may fall immediately, and charging may briefly interrupt during renegotiation. When testing intermittent charging, unplug every other load from the charger first and leave only the laptop, so sharing is excluded.
Q6: Charging stops at the same percentage every time, does that mean the battery is broken?
A6: Not necessarily, and very often it is a charge limit setting. Check brand battery management software, Windows Smart Charging, macOS optimised charging, and BIOS charge settings one by one, switch the limit off temporarily or set it back to full charge, and retest overnight. If full charge returns, it was a setting, not a fault; if the stop persists at the same percentage with abnormal battery report capacity, battery or charging management problems deserve investigation again.
Q7: Powered-off charging is also intermittent, does that exclude software problems?
A7: Largely yes. When powered off, system draw is extremely low, and power management, drivers, and background programs are all out of the picture; intermittent powered-off charging points very likely to hardware layers including the port, cable, and charging circuit. This control test needs no tools and is worth doing before anything else.
Q8: Sleep is normal but charging fails after waking, does that mean the port is broken?
A8: Not necessarily, since this is a separate failure mode more likely pointing to power management or re-enumeration layers than permanent hardware damage. Compare powered-off, asleep, and woken charging, and check driver and firmware wake fixes, rather than declaring hardware damage directly.
Q9: The meter shows stable 20V but current repeatedly falls to zero, which layer does that point to?
A9: It points to intermittent disconnection in the charger IC, MOSFET, or protection board layer, not the negotiation layer. Stable voltage proves negotiation has succeeded, while current falling to zero proves the charging path itself breaks intermittently. That pattern needs battery-layer and mainboard-layer inspection, not another cable or charger swap.
Q10: What are the three most important things to prepare before inspection?
A10: First, a symptom record covering frequency, triggers, and powered-off, asleep, and woken comparisons; second, the genuine charger and usual cable taken together so a result matrix can be run; and third, verified data backup plus confirmed disk encryption recovery keys at hand before any BIOS change or firmware update. Abnormal heating, burning odour, housing deformation, or sparks mean stop use and send for inspection at once.
Need Professional Laptop Repair?
Laptop King has 15+ years of repair experience in Shek Mun, Sha Tin. Free diagnosis, transparent pricing, no fix no fee. 180-day warranty for your peace of mind.