1. Scope First: This Article Covers Only the Antenna, Coaxial, and Routing Layer
This article applies to one narrow situation: a laptop whose Wi-Fi is fast at times and slow at others, drops when moved slightly farther away, misbehaves only at a certain lid angle, or performs noticeably worse on one band, where the reader wants to understand how the antenna layer fails, how layer-by-layer exclusion narrows the suspicion, how to avoid common misdiagnoses, and what to record before inspection. The antenna layer means the radio-frequency path that begins at the antenna element receiving the signal, continues through the coaxial cable, and ends at the connector on the wireless card, together with how that path is routed through the hinge and along routing guides. This article explains how that path fails, how external comparisons narrow it down, how misdiagnosis is avoided, and what the reader can do at user level.
Three neighbouring layers appear here only as boundary comparisons and are never rewritten. The wireless card layer concerns the card's own chips, its radio chains, or its slot contact; if the antenna layer is intact but the card layer is faulty, replacing the cable will not help, and this article mentions the card only when separating layers, without expanding into card specifications or replacement. The driver and system layer is a software layer covering driver versions, system power management, and band preference settings; a physical break in the antenna layer cannot be repaired by reinstalling drivers, and this article mentions drivers only in the exclusion order, never as a tutorial. The router and broadband layer is the far-end and environment layer covering router placement, channel interference, and broadband speed itself; an antenna-layer fault characteristically follows the machine, remaining poor when the environment and network change, rather than following the router, and this article mentions routers only during comparisons, never expanding into router settings or broadband buying advice.
A hard boundary must be stated before anything else, because the whole value of this article rests on it. The user level is limited to external comparison tests and settings checks: distance comparison, angle comparison, band comparison, other-device comparison, network comparison, time and event comparison, and a symptom log. Internal inspection is left entirely to inspection: opening the machine, opening the screen, removing the base cover, unplugging internal cables, unplugging antenna connectors, replacing an antenna as a do-it-yourself job, and inspecting a connector for oxidation or cable damage are all out of bounds. Manufacturers' own replacement procedures set the same boundary, requiring safety information to be read first and restricting the work to authorised technicians. This article follows that boundary rather than working around it, and no wording of the form "just open it and look" appears anywhere in it.
2. What Form the Antenna Takes, and Where It Sits: A Flat Built-in Antenna Whose Position Varies by Model
Laptop built-in antennas are mostly flat in form, hidden inside the casing, not exposed, and therefore not easily damaged by external impact, while also allowing the device to be made thinner; the price is that space must be reserved inside the casing between the ground plane and the radiating element, which sets a limit on miniaturisation. One common form is the planar inverted-F antenna: a radiating plate parallel to a ground plane, connected to the ground plane through a shorting plate or shorting pin, with a separate feed point supplying power; the plate size relates to the operating frequency range, and size is roughly proportional to wavelength. It must be stated clearly that not every laptop uses the same form; individual models may use other forms, such as monopole, loop, or printed-circuit-board trace antennas, and the model's own service manual is the authority. This reservation matters because every later discussion of position and routing obeys the same principle: it varies by model.
Position must not be generalised either. Repair guides describe the laptop display module as a sandwich containing the screen, the webcam, and the wireless antenna, clamped between the front bezel and the top cover, with the hinge fixing the display module to the base while allowing it to open and close. Manufacturer service manuals, however, prove that position is not uniform: in some models the wireless local-area-network antenna module sits in the palm-rest assembly and is removed from the palm rest during replacement, while in other models the antennas sit at the display assembly base or palm rest, with left and right antennas and their cables removed together from the palm rest. This article therefore writes that position varies by model, listing the display bezel or top-cover area alongside the palm-rest module; it must never be written that the antenna is always at the top of the screen or always in the left and right bezels. A metal top cover or shielding shell can affect radiation, but that belongs to the model's design, and without that model's official information no claim may be made that a particular metal shell blocks a particular amount of signal.
One system-level idea must be established first: antenna gain already includes cable loss, which means the antenna plus the coaxial cable must be treated as one whole system, not judged from the antenna element alone. Manufacturer guidance requires the antenna paired with a wireless card to be installed inside the laptop or tablet host platform and to meet a minimum separation distance from the user. That requirement already states the point: the antenna layer is not an isolated part but a system of antenna element, cable, routing, connector, and installed position, and a failure anywhere in that system can degrade overall reception without the antenna element itself being faulty.
| Position wording | How to write it | Assertion that must never be written |
|---|---|---|
| Display bezel or top-cover area | The display module sandwich may contain the antenna, depending on model | The antenna is always at the top of the screen |
| Palm-rest module | In some models the antenna module sits in the palm-rest assembly | The antenna is always in the palm rest |
| Display base and palm-rest boundary | In some models left and right antennas with cables come out of the palm rest together | All models route their cables the same way |
| Metal cover or shield effect | A matter of model design; consult that model's official information | A particular metal shell blocks a particular amount of signal |
3. Routing: The Path from the Antenna Element Through the Hinge to the Card
The significance of routing is that the signal received by the antenna element must travel through the coaxial cable to the card at the base of the machine; once routing leaves its original guides, is pinched by a cover, or is rubbed by the hinge, what is affected is not the antenna element itself but the transmission path. That is precisely the boundary between this article and the card layer: the card layer concerns the card itself, while this article concerns the road before the card. Manufacturer service manuals repeatedly stress that the routing path must be noted during disassembly and followed along the original routing guides during reassembly, with tape used to secure the cable; one model's procedure even lists removing the tape that secures the auxiliary cable first and then drawing two cables out of the palm-rest guides. Repair guides likewise warn that at least one cable runs from the base through the hinge into the display module, and that the routing position must be planned in advance so the cable is not damaged.
The hinge area deserves its own section because it satisfies two conditions at once: it moves constantly, and its space is narrow. Every opening and closing of the lid changes the load on cables passing through the hinge, and the routing guides and tape exist so that a cable in narrow space keeps its intended path instead of protruding, being pinched by a cover, or preventing a panel from seating. In other words, routing is not a matter of draping a cable across from one end to the other; it is a designed path with guides, fixings, and reserved bending space, and departure from that path can itself become a failure mechanism, as the next sections explain one mechanism at a time.
For the reader, the practical meaning of the routing section is only one thing: angle and position observations have a mechanism behind them. Because the cable passes through the hinge, a change of lid angle changes its load; because the cable has a fixed path, a drop or sustained pressure afterwards may have changed that path. Observing an angle relationship, however, does not entitle the reader to open the machine and confirm the routing; confirming routing is inspection work, and the reader's work is to record angle, position, and events accurately so inspection can focus quickly.
4. MIMO and Diversity Reception: Why Two Antennas Are Not One in Use and One Spare
MIMO, multiple-input multiple-output, means transmitter and receiver using several antennas at once and exploiting multipath propagation to raise throughput and reliability; compared with a single-antenna system, the defining feature of MIMO is that each antenna has its own independent radio chain behind it, not merely that there are more antennas. Two uses must be kept apart and never confused. Spatial diversity sends or receives redundant information along different spatial paths in parallel, raising reliability and coverage without directly multiplying speed. Spatial multiplexing sends several independent data streams at once on the same channel, which the receiver recombines into one stream, raising capacity; 2x2 means two streams in parallel, and the number denotes how many data streams can be sent and received simultaneously, with the weaker side, device or router, limiting actual performance.
An official manufacturer community reply states that a 2x2 wireless card is designed to need two antennas for best performance; with only one antenna connected, the card may still operate, but not in its designed complete state. That sentence explains the central symptom pattern of this whole article. The consequences of one antenna or coaxial path failing may only be described with "may" wording, never with a fixed percentage. Spatial multiplexing falls from two streams back to one, peak transmission capability drops, and the effect is felt most in high-speed scenes at close range; diversity reception loses one path for comparison, so stability drops under weak signals, long distances, and highly reflective environments, producing fluctuating speed and intermittent disconnection; but this does not mean no Wi-Fi at all, because the remaining stream can still connect. That is the mechanism behind why a machine with one failed side can still get online yet drop whenever it is moved.
Four writing traps must be named one by one. First, never write that one failed side means speed halves; sources support only a fall from multiple streams to a single stream with capability reduced, and no source gives a fixed ratio, since the ratio depends on the router, distance, band, and environmental reflections, so writing a number would be invention. Second, never write that speed reveals which side failed; user-level observation cannot localise a fault to one particular cable. Third, never describe a 1x1 model as having half its antenna broken, since 1x1 has only one stream by design and must not be confused with a 2x2 failure. Fourth, diversity and multiplexing must be discussed separately; reliability gains must not be written as speed multiplication, and capacity gains must not be written as coverage gains.
| MIMO concept | Meaning | What one failed path may bring |
|---|---|---|
| Spatial diversity | Redundant information in parallel over different spatial paths | Reduced stability under weak signals and reflective environments |
| Spatial multiplexing | Several independent data streams at once on one channel | Reduced peak capability, felt most in close-range high-speed scenes |
| 2x2 | Two streams in parallel | Falls back to one stream, still connected but not in complete state |
| 1x1 | Only one stream by design | Must not be described as half broken |
A manufacturer guide's antenna table lists the same flat antenna's peak gain at the Main position and at the Aux position by antenna location, noting that all gains include cable loss; a MIMO column sits alongside Main and Aux, showing that the two are ports of one MIMO system rather than two independent systems. Manufacturer replacement procedures likewise list disconnecting the Main and Aux antennas as steps and require reconnecting by triangle marks, showing that wrong alignment or incomplete seating is itself a failure mode. One model's procedure writes a white cable to the Main hollow-triangle mark and a black cable to the Aux solid-triangle mark; another manufacturer's manual extract writes a first white antenna cable to the first Main terminal and a second black cable to the second Aux terminal. These may only serve as conceptual examples that a Main and Aux distinction exists, and must never be written as universal for all machines.
Colour and marking are not universal rules, and the point cannot be repeated too often. The white-equals-Main and black-equals-Aux wording above belongs only to individual model procedures; other manufacturers or models use numeric labels, for example discussions showing a black cable marked with a number, or different colour combinations. This article must never write that white is always Main and black is always Aux; it must write that the model's own service manual marking is the authority. Turning one model's cable colours into a universal rule is one of the easiest errors in this topic to make and one of the hardest to notice, because it sounds specific and professional while actually generalising from a single case.
The greatest trap is the by-band claim, which must be rejected in the clearest language. A manufacturer community reply states explicitly that wireless antenna position has no correspondence with the 2.4 GHz or 5 GHz bands, and that a single-band card also has two antennas; both antennas are generally dual-band designs rather than one handling one band each. Bluetooth attribution is also conflicted: one reply says Main handles Wi-Fi and Aux handles Bluetooth, while another says a 2x2 card's Wi-Fi needs Main plus Aux together, with Aux supporting both Wi-Fi and Bluetooth. The two statements contradict each other, so this article cannot pick one as a universal conclusion; the correct wording is that which cable Bluetooth shares depends on the card and model design, judged by official specifications, and this article does not need to expand the Bluetooth layer. The mechanism for speed loss or disconnection when one side fails should be explained in the MIMO and diversity language of the previous section, that is, one stream fewer, never in by-band language such as "the 5 GHz antenna broke".
| Claim | Correct or not | Reason |
|---|---|---|
| Main and Aux are two MIMO ports | Correct | The guide lists them alongside MIMO, with gains including cable loss |
| White is always Main, black is always Aux | Wrong | Only individual model wording; markings vary by model |
| Main handles 2.4 GHz, Aux handles 5 GHz | Wrong | A manufacturer reply explicitly denies any position-to-band correspondence |
| Main handles only Wi-Fi, Aux only Bluetooth | Cannot be stated as a conclusion | Published statements conflict; depends on card and model design |
| One failed side means one stream fewer | Correct | Explained in MIMO and diversity language |
6. Band Differences: Why 2.4 GHz, 5 GHz, and 6 GHz Behave Differently
Band differences are the physical background of layer-by-layer exclusion, and without them every band comparison later in this article would misdiagnose. The higher the frequency, the shorter the wavelength, the lower the diffraction, and the harder it is to get around obstacles; wavelength size determines antenna size, and an antenna element's length is tuned to wavelength, so no arbitrary length receives all bands well. An antenna supporting only 2.4 GHz cannot radiate a 5 GHz signal effectively, and vice versa; a dual-band antenna is a single element tuned to resonate at two bands, and its gain must be examined separately for each band. This tuning idea is important: an antenna is not a universal receiving wire but an element tuned to a particular frequency range, and damage that changes the element's or cable's electrical state can affect different bands to different degrees.
| Band | Frequency range | Approximate wavelength, calculated | Penetration and coverage character |
|---|---|---|---|
| 2.4 GHz | About 2400-2483.5 MHz | About 12.5 cm | Longest wavelength, relatively best wall penetration and diffraction, farther coverage, but few channels and much interference |
| 5 GHz | About 5150-5850 MHz | About 6.0 cm | Shorter wavelength, lower diffraction, shorter effective range at the same transmit power than 2.4 GHz, more channels and higher speed |
| 6 GHz, Wi-Fi 6E and Wi-Fi 7 only | About 5925-7125 MHz | About 5.1-4.2 cm, calculated from the two ends of the range | Weakest wall penetration, but cleaner spectrum and widest bandwidth |
Two further practical facts about bands must be used carefully. First, 2.4 GHz has only three non-overlapping channels, namely channels 1, 6, and 11, so neighbour interference is severe in dense environments, while 5 GHz has more channels and relatively less interference. Second, a manufacturer community reply notes that the 5 GHz band is generally less prone to interference but has lower range and penetration, and that at a longer distance from the router or with walls in between, 2.4 GHz may be the better choice. Together these form the physical basis of the misdiagnosis traps in a later section: a distant 5 GHz weakness can be normal physics, and a noisy 2.4 GHz band can be neighbour interference, and neither is antenna evidence on its own.
7. Coaxial Cable Structure: Four Layers from Inside Out, Each with an Electrical Job
From inside out, the four layers and their jobs are as follows. The first layer is the inner conductor, the centre conductor, carrying signal current, generally copper wire, solid or stranded, and it is the signal's main channel. The second layer is the dielectric, the insulating layer, separating the inner conductor from the shield and maintaining a fixed spacing; spacing and dielectric properties determine characteristic impedance, so once spacing changes, impedance changes. The third layer is the shield, the outer conductor, braided copper mesh, aluminium foil, or a combination of both; it has a dual job of confining the signal's electromagnetic field between inner conductor and shield to prevent signal leakage, while blocking external electromagnetic interference from entering; under normal conditions the outside of the shield stays at ground potential while the inner conductor carries signal voltage. The fourth layer is the jacket, a protective plastic layer resisting mechanical damage, abrasion, and moisture; the jacket itself carries no signal, but once it breaks, the inner layers lose their protection.
Wi-Fi radio-frequency coaxial systems generally use a 50-ohm characteristic impedance, which is the shared language of system design: antenna, cable, connector, and card end must sit in the same impedance system for power to transfer effectively, and once impedance changes, reflection and loss follow. This idea explains why a flattened cable can perform badly without being broken: flattening changes the spacing between inner conductor and shield, spacing change means impedance change, and impedance change means increased reflection and loss, so the signal is not cut off but already degraded. Cable loss relates to frequency, and an individual product specification lists a loss figure only at one particular frequency; that figure belongs to that product and that frequency and must never be generalised to all laptop cables or all bands. In writing, the four-layer skeleton is enough for structure; diameter, length, and loss figures must name their product, or they must not be written at all.
| Layer | Name | Electrical job | What damage does |
|---|---|---|---|
| First, innermost | Inner conductor | Carries signal current, the main channel | Micro-stress and partial strand breakage raise resistance points |
| Second | Dielectric | Separates inner conductor from shield, holds fixed spacing, sets characteristic impedance | Deformation, thinning, or micro-cracking changes impedance and raises reflection |
| Third | Shield, outer conductor | Confines signal leakage, blocks external interference | Broken braid, wrinkling, or foil cracking damages shielding continuity |
| Fourth, outermost | Jacket | Resists mechanical damage, abrasion, and moisture | Breakage leaves inner layers unprotected |
8. Connector Concepts: Micro-Coaxial Connectors in Concept Only, No Model Table
At the laptop card end, a miniature coaxial socket and plug are generally used: the plug connects to the micro-coaxial cable, and the socket is surface-mounted and soldered to the circuit board. That conceptual sentence is already enough to support this article's discussion, because the connector layer's point is not model numbers but three concepts. First, the connector and the cable termination are themselves part of radio-frequency performance, not an ordinary plug, and manufacturers emphasise solderless micro-coaxial termination techniques for keeping termination performance consistent. Second, different connector generations differ entirely in height, cable diameter, and frequency ceiling, and cannot be treated as interchangeable. Third, many miniature connectors on the market look alike in shape, a wrong size will not insert or will sit unstably, and different batches of the same card or different sources can describe the connector model differently.
For exactly that reason, this article may only write the concept, and must never write that a reader's machine definitely uses a particular connector model; readers must check that model's service manual or the original cable's markings. Still less may it write a specific mating-cycle limit as a universal specification, since an individual connector's mating life belongs to its product specification and varies by model. The connector layer's failure modes are left to the next section: incomplete seating raises contact resistance and reflection, making the signal intermittent, and slight vibration or lid movement can change the contact state and cause intermittent disconnection. One boundary must be drawn here first: confirming and handling a connector is inspection work, and readers must never unplug a connector themselves to test it, as the next section states again.
9. Damage Mechanisms: Six Mechanisms, One at a Time
9.1 Hinge Repeated Opening and Closing Wear, Fatigue
Display connecting cables run through the hinge, and opening and closing bends the cable repeatedly; general repair discussion notes that long-term bending and straightening can damage conductors inside a cable and even cause a short circuit. That statement addresses the display cable bundle as a whole, but a coaxial cable passes through the same hinge path, so the mechanism carries over, and writing must attribute it to the hinge path rather than asserting that every machine fails this way. The general mechanism of bending too tightly is that the outside is stretched and the inside compressed, stranded conductors may develop micro-stress and partial strand breakage raising resistance points, the dielectric thins on the outside and is compressed on the inside and may micro-crack over time, and the shield may wrinkle or break, damaging shielding continuity and interference immunity. More opening cycles and a smaller bending radius accumulate fatigue faster; but how many cycles cause failure has no source, so no cycle count or year count may be written. Cable assembly failure literature likewise notes that repeated bending can cause fatigue damage until the cable or jacket fails, and that bending too tightly can violate the minimum bending radius and cause an open circuit. Only the direction of the mechanism is stated here, not a specific radius figure, because the figure depends on the cable chosen.
9.2 Crushing and Pinching, Routing Departure and Cover Pinching
Manufacturer manuals repeatedly require cables to follow guides and be secured with tape, precisely to prevent a cable from protruding, being pinched by a cover, or preventing a panel from seating; discussion also reflects that if Wi-Fi and display cables do not hug the frame, the base cover may not close and the cable risks being pinched. The general coaxial mechanism is that pressure deforms the dielectric, changes the spacing between inner conductor and shield, thereby changes characteristic impedance, and causes reflection and loss; a jacket cracked by pinching also exposes the inner layers. The writing point is that crushing need not break the cable immediately: the appearance may be only slightly flattened, yet electrically it has already changed, which explains a signal already degraded without a visible break. Crushing differs from fatigue in its time pattern: fatigue usually worsens gradually, while crushing can appear suddenly after one disassembly, one drop, or one episode of pressure; but a time pattern is only a clue, not a diagnosis, and only inspection can separate them.
9.3 Shield Damage Causing Interference, Leakage and Ingress in Both Directions
Once the coaxial shield has a gap, including broken braid strands, a cracked foil layer, or looseness at the connector transition, radio-frequency energy can leak from inside out and can also enter from outside in; shielding effectiveness is a system property, the cable design sets the ceiling, and the bending radius, pressure, and moisture after installation decide how much remains. Shield damage need not present as a complete loss of signal; it may present as normal near the router and unstable slightly farther away, particularly poor on certain bands or in certain environments, and more obviously interfered with near other electronic equipment. This article may only use "may" wording and must never write it as a diagnostic standard. The most easily misdiagnosed aspect of shield damage is that its presentation resembles environmental interference, while environmental interference in turn resembles shield damage; the two must be separated with other-device comparison and a distance log, never settled from a single impression.
9.4 Connector Loosening and Incomplete Seating
Manufacturer procedures list disconnecting the antenna and reconnecting it as separate steps and require alignment by triangle marks on reconnection, showing that wrong alignment or incomplete seating is itself a failure mode; repair guides also warn that the display module and base should be separated before handling cables through the hinge, so that connectors are not pulled. An incompletely seated connector means raised contact resistance and increased reflection, with the signal present at times and absent at others; slight vibration or lid movement can change the contact state and cause intermittent disconnection. This mechanism can explain a pattern of pressing the body or moving the screen and dropping briefly, but this article cannot write it as the only explanation, because hinge cable fatigue can equally cause intermittent faults. The boundary must be repeated: confirming and handling a connector is inspection work, and readers must not unplug connectors themselves to test, since the test itself can pull hinge-routed cables or damage a connector.
9.5 Connector Oxidation and Corrosion, Low Confidence, Cautious Wording Required
Coaxial leakage literature notes that corroded or loose connectors are often the strongest leakage points; but that literature addresses distributed coaxial systems, not laptop micro-coaxial connectors, and cannot be copied across word for word. Connector literature notes that possible causes of corrosion or oxidation include high temperature, high humidity, and corrosive gases; the phenomenon is base metals such as nickel and copper corroding and forming deposits on the surface, which can cause poor contact; another environmental parameter analysis notes that relative humidity above 80 percent is one of the main causes of electrical breakdown, and that long-term high humidity can cause moisture absorption, corrosion, and cracking. The writing treatment may only be conditional: if a connector's metal contact surface is damp, oxidised, or contaminated, contact resistance may rise; rates of oxidation, how long dampness takes to cause failure, and which environment certainly causes oxidation must never be written as figures. If oxidation is to be made the main cause, a first-hand source specific to laptop micro-coaxial connectors must first be obtained; otherwise it stays in the to-be-verified column and must not enter the main text as a conclusion.
9.6 Cable Ageing, Materials Only, Direction Only, No Year Count
Jacket, dielectric, and shield are all physical materials, and long-term heat, pressure, bending, and moisture degrade their elasticity and structure; ageing itself is not an on-off fault but a gradual fall in the cable's tolerance of bending, crushing, and interference. Current research has found no ageing-life data specific to laptop Wi-Fi coaxial cables, such as a number of years or a number of opening cycles, so any year count, cycle count, or life figure must not be written. Ageing may only serve as background explaining why an older machine is more likely to fail at the hinge, never as a diagnostic conclusion; diagnostic conclusions must rest on step-by-step exclusion. The relationship between ageing and fatigue must be stated clearly: fatigue is damage accumulated by repeated bending, ageing is the material itself degrading over time and environment, the two can add together and make the same old cable at the hinge more fragile, but their combination does not justify deriving a lifetime.
| Mechanism | Possible presentation, all in "may" wording | Assertion that must never be written |
|---|---|---|
| One antenna or coaxial path failed | Speed falls in high-speed scenes, unstable under weak signals, intermittent disconnection, still connected | Speed halves, a particular band disappears completely |
| Hinge fatigue or crushing | Disconnection or recovery as the angle changes, only at one angle | It must be the hinge, no need to check elsewhere |
| Shield damage | Particularly poor at distance or in interference, possibly normal nearby | It must be interference, not a cable problem |
| Connector loosening | Breaks on vibration or movement, intermittent from changing contact | Re-seating once will certainly cure it |
| Antenna element itself deformed or displaced | Directional coverage change, particularly poor in one direction | A metal shell must be the culprit |
| Connector oxidation or corrosion | Contact resistance may rise, presentation may be intermittent | Dampness for a given time certainly causes failure, visible oxidation confirms diagnosis |
| Cable ageing | Gradually falling tolerance of bending, crushing, and interference | A given number of years certainly causes failure, an old machine means ageing caused it |
10. Layer-by-Layer Exclusion: Seven External Comparisons, Each with Its Interference Factors
The principle must be stated first: layer-by-layer exclusion can only raise or lower suspicion of a layer and can never diagnose at user level. Every comparison has interference factors, and writing must state them in the same passage, or the writing misleads. The value of exclusion lies in narrowing and recording, not in conviction; a well-written log lets inspection focus faster, but the log itself is not a diagnosis.
| Comparison | How to do it, user level | What a result leaning toward the antenna layer looks like | Interference factors and counter-examples, which must be written in the same passage |
|---|---|---|---|
| Distance comparison | Same network, same room: first record close to the router, then gradually move farther or behind one wall; use another device at the same place as a comparison | Acceptable close up, sharply worse or disconnected slightly farther away, while other devices at the same far place still work | A far place is weak by itself; router placement and wall material also matter; normal close up does not mean the antenna is healthy |
| Angle comparison | At the same place, slowly change the lid angle, pause each time, and observe whether signal or connection changes in step, recording angle and change | Signal regularly improves or worsens in step with lid angle | Hand or body shielding, a moved resting position, and changing nearby interference can cause similar fluctuation; angle testing itself is not a diagnosis and must not involve forcing the hinge |
| Band comparison | In settings, confirm whether the current connection is 2.4 GHz, 5 GHz, or 6 GHz; repeat the same distance comparison on two bands, only within what the user can choose in system settings | Only 2.4 GHz visible, no 5 GHz or 6 GHz, or 5 GHz or 6 GHz dropping much faster slightly farther away while 2.4 GHz remains usable | 5 GHz penetrates walls worse by nature, so a distant 5 GHz weakness is not antenna evidence by itself; a router not enabling that band or a device not supporting it produces the same appearance |
| External adapter comparison, concept only | At the same place, compare with an external USB wireless adapter, discussing concept and interpretation only, without teaching model selection | External performance clearly more stable than built-in | The external adapter bypasses the internal card and internal antenna together, so it can only point to the built-in path and cannot separate card from antenna; external antenna position and gain also differ |
| Other-device comparison | Same place, same network, same time, compare with a phone or another laptop | Only this laptop is poor, other devices normal | Other devices differ in antenna design, placement, and supported bands; one phone alone cannot exclude a router problem |
| Network comparison | Repeat the same observation on another known-good network or at another place | The problem follows the machine after changing network or place | Two networks differ in settings, bands, and interference environment; only a reference, never a diagnosis |
| Time and event comparison | Record whether any drop, impact, repair, disassembly, or sustained pressure preceded the first appearance | The problem appeared after a drop or repair, or worsened gradually with opening cycles | An event and a fault may be coincidence only; no event does not exclude antenna ageing or connector problems |
11. A Reference Line for Close-Range Signal: Use Carefully, Never as a Diagnostic Standard
Need a laptop repair expert?
Free diagnosis · Transparent pricing · No fix, no fee · 180-day warranty
A manufacturer community reply once offered an experience-based reference: close to the router, signal strength is generally expected around 0 to minus 30 dBm, and if an analysis tool measures around minus 50 or minus 70 dBm at such a close distance, the antenna may be involved, with a recommendation to contact the system manufacturer. The limits of use must be written into the main text and the figure must never be written alone. First, this is an experience-based reference in a community reply, not a guaranteed value for all models, and different antennas, routers, and measuring tools give different results. Second, writing may only say that an abnormally weak signal even close to the router is worth suspecting the antenna layer and sending for inspection, and must never write that minus 50 dBm equals a faulty antenna. Third, the more robust user-level recording method is a combination of signal percentage or bars plus distance plus wall count plus band, rather than condemning a machine on a single dBm figure. This passage demonstrates the scale of the whole article: a figure may be quoted, but only together with its limits, since a bare figure misleads.
12. What a Symptom Log Should Record: The Most Useful Part to Bring to Inspection
The purpose of a log is to let inspection focus faster, not to let the reader diagnose alone. The end of this article should state that the log should be brought to inspection, not used to judge alone. Every field below is observable by the user, involves no privacy, and involves no disassembly, and filling each field in is more useful than writing a pile of adjectives.
| Log field | How to record it | Format example only |
|---|---|---|
| Date and time, duration | How long the problem has lasted, sudden or gradual | Began on a certain day, worsening gradually |
| Place and distance | Living room or bedroom, approximate distance from the router, how many walls or doors in between | Bedroom, one wall in between |
| Current band | 2.4 GHz, 5 GHz, or 6 GHz | Currently connected on 5 GHz |
| Signal bars or percentage | Bars or percentage, together with distance, wall count, and band | Far-place bars clearly fewer than near-place bars |
| Felt connection speed | Whether video stalls, without writing speed-test figures as conclusions | Video pauses intermittently |
| Lid-angle change | Whether the signal changes in step, recording only yes, no, or unsure, without precise angles | Signal worsens in step at one angle |
| Other devices at the same place | Normal, equally poor, or not tested | Phone normal at the same place |
| External adapter comparison result | Better, no difference, or not tested, noting that external testing cannot separate card from antenna | External better, but cannot separate |
| Events before the problem | Any drop, impact, repair, or sustained pressure, yes, no, or unsure | One drop beforehand |
| User-level exclusions already performed | Router restart, computer restart, network change, place change, and whether anything changed afterwards | Pattern unchanged after restart |
| Conditions of appearance | Whether it appears only on one band, at one place, or at one angle | Only at a far place and on one band |
13. Hong Kong Context: How Humidity, Commuting, Drops, and Dense Housing Amplify a Weakness
13.1 Humidity: A Background Factor in Connector Ageing, Not a Diagnosis
The Hong Kong Observatory's 1991-2020 climatological normals give an annual mean relative humidity of 78 percent, but that figure is only a Hong Kong outdoor climate background, not indoor humidity and certainly not humidity inside a machine. Connector literature notes that possible causes of corrosion or oxidation include high temperature, high humidity, and corrosive gases; the phenomenon is base metals corroding and forming deposits on the surface, which can cause poor contact; another environmental parameter analysis notes that relative humidity above 80 percent is one of the main causes of electrical breakdown, and that long-term high humidity can cause moisture absorption, corrosion, and cracking. The careful wording available to this article is that Hong Kong's long-term dampness is a reasonable background factor in connector ageing, but dampness itself is not a diagnosis, and only inspection can confirm whether a particular machine has a connector problem. It must never be written that Hong Kong humidity causes a particular proportion of antenna failures, that a particular district fails more easily, that humid-season weather certainly breaks antennas, or that visible oxidation confirms a diagnosis, since the user level should not open a machine to inspect at all.
13.2 Daily Commuting and Opening: The Concept of Bending Fatigue
Micro-coaxial bending and twisting tests are designed precisely for the use case of a display connected to a main board through a hinge, and are assessed by performance change after bending cycles; cable fatigue literature likewise notes that repeated bending can fatigue a cable until it fails. Usable scenarios, without figures, are daily commuting, opening several times a day, sustained pressure in a backpack over a long period, or other items pressing against the hinge area, all of which raise the mechanical load on the routing area; but this article must never derive a figure such as a certain number of openings per day causing a break after a certain number of years, because no measured lifetime source for that model exists. Writing must also warn that deliberately opening and closing rapidly or forcefully for a test is the wrong approach; angle comparison needs only slow observation within the normal range of use. The other side of commuting is pressure: a backpack packed full for a long time can press on the hinge and routing areas, but pressure, like fatigue, is only a direction of suspicion, never a diagnosis.
13.3 Drops: One Event Can Be a Starting Point, But Not Evidence
A manufacturer community case reports a user eventually discovering that the cable connecting the wireless card had been cut when the laptop was accidentally dropped; even reinstalling drivers and moving close to the router had not helped beforehand, and this is a user account offered only as a scenario example. Another manufacturer's community contains user accounts of an antenna wearing through near the hinge and then giving an extremely weak signal, and of a hinge breaking the outer shielding so that a signal could only be received very close to the router. Both are forum user accounts with low evidential strength and may only be written as "users have reported similar scenarios", never as a common fault or a diagnostic standard. The writing principle is that a signal problem after a drop is worth sending for inspection, but the drop itself does not prove the antenna layer, since the card, main board, screen, and hinge structure may also be affected at the same time, and inspection must judge them together. The value of time and event comparison lies precisely in writing the drop into the log so inspection knows which layers to check, rather than letting the reader decide which layer failed.
13.4 Dense Housing: How Concrete Walls Plus Neighbour Wi-Fi Amplify an Antenna Weakness
On the physical layer, the higher the frequency, the higher the material transmission loss; a predictive model using a heavy concrete wall as an example lists a large gap between its 2.4 GHz and 5 GHz attenuation figures, but those figures are model examples depending on material and thickness and must never be presented as measured values for Hong Kong flats. Concrete walls, columns, and brick walls common in Hong Kong flats all push a distant signal below the usable threshold sooner. On the interference layer, 2.4 GHz has only three non-overlapping channels, so overlapping neighbour networks in dense housing worsen interference; 5 GHz has more channels but penetrates walls worse, so each band has its own weakness. The correct way to write the amplification effect is that when an antenna is already weak, it may barely work in a normal environment, and once a concrete wall plus neighbour interference is added, the weakness is amplified into a pattern of normal near the router and dead in the next room. Conversely, dense housing alone can make a healthy antenna look poor in a far room, so "dead in the next room" alone must never be used to conclude an antenna fault. Writing must never include Hong Kong flat wall-thickness figures, signal data for a particular estate or building type, or any relationship between neighbour count and failure rate, since all of those lack sources and must not be written.
14. Boundary Comparison Once More: Three Layers, Why Separation Matters
Separation matters because the three layers can present similarly while requiring entirely different handling. If the antenna layer is intact but the card layer is faulty, replacing the cable will not help; if the antenna layer is physically broken, reinstalling drivers will not repair it; and an antenna-layer fault characteristically follows the machine, remaining poor when environment and network change, while a router-layer fault characteristically follows the environment and improves with a different machine. These three comparison sentences are enough for readers to understand why inspection must separate layers, without expanding the three layers into tutorials. The comparison table exists to stop readers from heading in the wrong direction: do not treat a problem that follows the machine as a router problem, do not treat a problem that follows the environment as an antenna problem, and do not treat a software-layer problem as a physical-layer problem. Every wrong direction wastes time and money and can complicate an originally simple problem, for example by opening the machine yourself, after which inspection must first separate which damage was the original fault and which was caused by opening.
| Layer | Direction of the pattern | What replacing cable, card, or router does |
|---|---|---|
| Antenna, coaxial, and routing layer | Follows the machine, poor across environments and networks | Replacing the cable may help, but only after inspection confirms |
| Wireless card layer | Card or slot problem, healthy antenna layer still fails | Replacing the cable will not help, separation required |
| Driver and system layer | Software layer, physical break not repaired by reinstalling | Reinstalling will not repair a physical break |
| Router and broadband layer | Follows the environment, better with a different machine | Replacing the router will not repair a problem that follows the machine |
15. Common Misdiagnoses and Traps: Twelve Items, Each with Why the Error Happens and the Correct Understanding
Trap 1: Still Weak After Replacing the Card, Reinstalling Drivers, and Changing Routers
The misdiagnosis logic is assuming that replacing one layer excludes all layers. The correct understanding is that if the problem sits in the antenna layer, replacing the card, reinstalling drivers, and changing routers all still pass through the same antenna path, so the weak-signal pattern may remain exactly as before; but the reverse also holds, since a persisting problem does not prove the antenna, because settings, interference, wall material, and router placement not yet excluded can equally keep the problem alive. Writing must be two-directional and must never present a persisting problem after replacement as an antenna diagnosis.
Trap 2: Lid Angle Affects the Signal
The misdiagnosis logic is assuming that improvement on moving the screen means a software problem, or that an angle problem must mean a screen problem. The correct understanding is that the antenna and its cable pass through the hinge area, so an angle change alters cable load and antenna position, and a signal changing in step with angle is a clue worth recording; but body shielding, a moved resting position, and changing interference can cause similar fluctuation, so angle testing can only feed a log, never a diagnosis. The trap is that forcing the hinge, flapping the lid rapidly, or opening the hinge yourself to prove the point must never be done.
Trap 3: Only 2.4 GHz Visible, No 5 GHz or 6 GHz
The misdiagnosis logic is writing directly that the 5 GHz antenna is broken. The correct understanding is that possible causes include the antenna layer, a router not enabling that band, a device not supporting it, and driver or system settings. This article expands only the antenna-layer possibility and treats the rest as boundary reminders without tutorials. The checking sentence writing must add is to confirm the router has 5 GHz or 6 GHz enabled and whether other devices at the same place can see 5 GHz or 6 GHz, then write the result into the log for inspection.
Trap 4: Normal Near the Router, Dead Slightly Farther Away
The misdiagnosis logic is assuming that normal nearby means the machine is healthy and the router or broadband must be at fault. The correct understanding is that when antenna gain falls, a strong nearby signal can still mask the weakness, and once distance grows or a wall intervenes, the signal can fall sharply below the usable threshold; a manufacturer community reply also uses close-range signal as a preliminary distinguishing clue. But a distant weakness can also come from walls, interference, and router placement, so other-device comparison and a distance log must be read together.
Trap 5: One Antenna Failing Causes MIMO or Diversity Failure
The misdiagnosis logic is assuming that still receiving Wi-Fi means the antenna is healthy. The correct understanding is that 2x2 needs two usable paths; when one fails, the device may still connect but loses multi-stream or diversity benefits, showing as speed that cannot rise and particular instability at distance or under interference. The user level cannot tell which path failed and cannot measure how many streams were actually negotiated, so writing may only say "may" and "worth sending for inspection". The trap is writing a speed-drop ratio or claiming that connection speed reveals which side failed.
Trap 6: Looking at Bars Only
Bars are a relative estimate, and different devices and systems use different bar scales; distance, wall count, and band should be recorded together. Looking at bars only compares figures from different scales directly, and the comparison itself is already invalid.
Trap 7: Testing Only Once
Interference changes over time, so one good or one bad result cannot settle anything; observations should be repeated at the same place at different times. The biggest problem with a single test is not the small sample but that interference itself changes: one good result may only mean interference was low at that moment, and one bad result may only mean interference was high at that moment.
Trap 8: Moving the Machine During the Test
Distance, angle, and body shielding changing together destroy the comparison; only one variable may change at a time. This trap is very common: walking to a far place while also turning the lid angle and changing hand position changes three variables at once, leaving no way to know which variable caused the change.
Trap 9: Over-Reading the External Adapter Test
An external adapter improving only points to the built-in path and cannot separate card from antenna. Writing an external test as an antenna diagnosis is one of the jumps this article most wants to prevent, because it skips the card layer entirely.
Trap 10: Treating "Tried Everything" as Proof of the Antenna
Before settings, interference, walls, and placement are excluded, other possibilities remain. Having tried many things only means more layers have been tried, not that untried layers have been excluded.
Trap 11: An Undamaged Appearance Means the Cable Is Healthy
Crushing and shield damage can be outwardly inconspicuous, contradicting the appearance argument in the same way the ninth section explains: crushing and shield damage can be outwardly inconspicuous. Appearance was never a user-level exclusion tool, and still less can appearance exclude the antenna layer.
Trap 12: Forcing the Hinge, Flapping Rapidly, or Opening the Hinge to Test
All of these are dangerous or out-of-bounds suggestions and must never be written as test methods. Angle comparison needs only slow observation within the normal range of use, and any test technique beyond normal use can create new damage and make inspection harder to separate.
16. Repair, Replace Antenna, or Replace Machine: A Decision Framework with No Price Derivation
The pricing red line must be observed exactly as written: antenna, coaxial, and routing charges are always subject to a written quotation after inspection, and no antenna price, inspection fee, part price, or labour figure of any kind may be invented. The only confirmed starting prices are four, mentioned only where relevant and never applied to antennas: battery from HK$580, screen from HK$1,280, mainboard chip-level from HK$880, and keyboard from HK$480. Those four figures serve in this article only as examples of what cannot be applied, and must never be used to derive whether antenna repair is worthwhile or to write any analogy such as antenna repair costing about as much as a battery or screen. The decision framework therefore cannot use a sentence of the form "antenna repair costs about this much, so it is worth repairing"; it can only use the structure of obtaining a written quotation first and then comparing dimension by dimension below. To understand the pricing range first, see pricing; to understand the inspection process, see diagnostic service.
Directions worth considering for repair grow stronger the more items apply. First, the rest of the machine is in good condition, with screen, keyboard, battery, and performance still meeting daily needs, and the problem concentrated in the signal layer. Second, the signal problem has relatively clear clues, such as angle-related, distance-related, or band-related logs, so inspection can focus more easily. Third, the model still has corresponding antenna or routing parts available, noting that antenna part numbers vary by model and by touch or non-touch configuration and cannot be generalised; writing must never list specific part numbers as buying guidance. Fourth, the user still needs mobile Wi-Fi often, including commuting, different rooms, and different places, where an external adapter as a long-term substitute is inconvenient or unsatisfactory. Fifth, the written quotation falls within a range the user accepts, with the range judged by the user, since this article sets no monetary threshold.
Directions worth considering for comparing a replacement machine seriously grow stronger the more items apply. First, besides the antenna, several other items are ageing or faulty at once, such as battery, screen, hinge, and keyboard, so the combined repair total may approach replacement considerations, but no conclusion may be drawn before a written quotation exists. Second, the model has been discontinued for many years, so corresponding antenna or routing parts are hard to find or require a longer wait, written only as depending on model supply, never with a specific discontinuation year as a general rule. Third, the hinge or screen structure itself is also damaged, so replacing the antenna alone may not solve the root cause of routing pressure; that judgement must be made by inspection and must never be inferred at user level. Fourth, daily needs already exceed the old machine's performance, for example new systems or software requirements, so even a repaired signal leaves the overall experience insufficient. Fifth, the user mainly works in one fixed place, and an external solution already meets needs stably, so mobility demand is low; in that case the written quotation plus actual needs should be compared first rather than deciding immediately.
The decision process should follow this order. First, perform external comparisons and write the symptom log, without rushing to replace parts. Second, send the machine for inspection, so inspection separates the antenna layer, card layer, driver or system layer, and router or environment layer. Third, after obtaining a written quotation, compare dimension by dimension using the two directions above; before a quotation exists, any conclusion that repair is or is not worthwhile is only a preliminary direction, not a decision. Fourth, if the decision is not to repair, general matters such as data backup and old-machine disposal still deserve consideration, and this article notes them only as a reminder sentence without expanding. Fifth, if the decision is to repair, the same comparison should be repeated after completion at the same place, same distance, and same band, confirming whether the improvement matches the problem pattern in the log; if the pattern has not changed, the machine should be sent for inspection again rather than replacing parts repeatedly. The point of this process is its order: log before inspection, inspection before quotation, quotation before decision, and any decision skipping a middle step is premature.
| Decision dimension | Leaning toward repair | Leaning toward seriously comparing replacement |
|---|---|---|
| Rest of the machine | Screen, keyboard, battery, and performance still meet needs | Several ageing or faulty items at once |
| Clarity of clues | Angle, distance, and band logs clear, inspection can focus | Clues confused, layers not yet separated |
| Parts supply | Model still has corresponding antenna or routing parts | Discontinued for many years, parts hard to find or slow, depending on model |
| Mobility need | Frequent mobile Wi-Fi use, external substitute unsatisfactory | Mainly fixed place, external solution already stable |
| Structural condition | Hinge and screen structure undamaged | Hinge or screen structure also damaged, antenna alone may not solve the root cause |
| Overall performance | Overall experience sufficient once signal is repaired | Daily needs already exceed the old machine's performance |
17. Preparing for Inspection and Checking Afterwards: What to Bring, What to Say, How to Verify
Before inspection, readers should prepare three things: first, the symptom log, filled in according to the template in the twelfth section, emphasising pattern rather than conclusion; second, the machine and its original charger, since inspection needs to verify under similar conditions and a different charger can affect performance; third, a list of exclusions already performed, including router restart, computer restart, network change, place change, and external adapter comparison, together with whether anything changed afterwards. On the inspection form, the exclusions performed and their results should be written out, so inspection does not have to retrace blind alleys. When speaking at inspection, describe patterns rather than conclusions: normal nearby, dead farther away, trouble only at one angle, particularly poor on one band, rather than saying the antenna is broken and must be replaced; inspection's responsibility is separation, and the reader's responsibility is accurate observation.
After inspection, verification needs a method too. If the decision is to repair, the same comparison should be repeated after completion at the same place, same distance, and same band, confirming whether the improvement matches the problem pattern in the log: for example, a far place that used to drop should become stable, and an angle that used to fail in step should no longer change in step. If the pattern has not changed, the machine should be sent for inspection again rather than replacing parts repeatedly; an unchanged pattern is itself important information, telling inspection that separation may be incomplete or that another layer remains unhandled. To understand the pricing range first, see pricing; to understand the inspection process, see diagnostic service.
18. Understanding Charges and Shop Information: A Starting Price Is Not This Case's Quotation
The understanding of charges must be stated once more, because this is the part most easily misread. The only confirmed starting-price categories are four: battery from HK$580, screen from HK$1,280, mainboard chip-level from HK$880, and keyboard from HK$480, each only a starting price for its own category, and a starting price is not this case's quotation; every other item is subject to a written quotation after inspection. The screen starting price must never be written as an antenna price, the battery starting price must never be written as an antenna price, and no analogy of any kind may be used. Readers should decide using the written quotation, not using starting prices; the purpose of starting prices is only to show readers where to look for the pricing range, not to tell readers in advance what this case will cost.
The shop information is as follows: Room 1209, Dalton Plaza, Shek Mun, Sha Tin, telephone +852 2186 8010, WhatsApp +852 6558 6806, opening hours 12:00-19:00. To understand the pricing range first, see pricing; to understand the inspection process, see diagnostic service. Before inspection, fill in the log from the twelfth section and bring the machine and its original charger, so inspection can focus as quickly as possible.
19. Conclusion: A Way of Thinking About the Antenna Layer
A way of thinking about the antenna layer can be summed up in four sentences. First, antenna plus cable plus routing plus connector form one whole system, not an isolated part. Second, two antennas are not one in use and one spare: one stream fewer can still stay online, but with lower speed, instability, and intermittent disconnection. Third, layer-by-layer exclusion can only narrow suspicion and can never diagnose at user level; every comparison must state its interference factors in the same passage, or it misleads. Fourth, the decision order is log, inspection, written quotation, and dimension-by-dimension comparison; before a quotation exists, any conclusion that repair is or is not worthwhile is only a preliminary direction. Following these four sentences, readers can avoid most misdiagnosis traps, hand an accurate log to inspection, and let inspection separate layers, focus, and quote, rather than replacing parts on feeling. To understand the pricing range first, see pricing; to understand the inspection process, see diagnostic service.
20. Frequently Asked Questions
Q1: If a laptop Wi-Fi antenna fails, does that mean no internet at all?
No. A 2x2 machine normally has two antenna paths, and only together do they deliver multi-stream transmission and diversity reception. If one path fails, the other can still connect, so the typical pattern is still online but unable to reach full speed, unstable at a slightly longer distance, and intermittently dropping, rather than an instant total loss of connectivity. The reverse is equally true: still receiving Wi-Fi does not prove the antenna is healthy, because a single working path can hide a failed one. User-level observation can only record this pattern of being connected but unstable; it cannot identify which cable failed, and no fixed percentage drop can be derived from speed, because no source supports a fixed ratio.
Q2: Do MAIN and AUX mean one antenna handles 2.4 GHz and the other handles 5 GHz?
No, and this is the most common misunderstanding. MAIN and AUX are two antenna positions, two ports of a MIMO system, and both antennas are generally dual-band designs rather than one handling one band each; even a single-band card has two antennas. Individual service manuals may state a colour and triangle-mark correspondence for that model, but that is only that model's wording and must never be generalised into a rule that a white cable always means Main and a black cable always means Aux. Which cable Bluetooth shares also depends on the card and model design, and published statements conflict, so no single version can be written as a universal rule; the model's official specification is the authority.
Q3: Why does the Wi-Fi get better and worse as I change the screen angle?
Because the antenna and its cable pass through the hinge area, changing the lid angle changes the mechanical load on the cable and the antenna's relative position, so a signal that improves and worsens in step with the angle is a clue worth recording. That clue is not a diagnosis: hand and body shielding, a changed resting position, and changing nearby interference can produce similar fluctuations. The correct method is to move slowly within the normal range of use, pause at each angle, observe, and record whether the change is synchronised. Never force the hinge, flap the lid rapidly, or open the hinge yourself to prove the point. An angle-related log goes to the inspection, where the examiner can distinguish cable fatigue, crushing, a loose connector, or another layer.
Q4: It works next to the router but drops in the next room. Does that prove the antenna is faulty?
Not on this point alone. When antenna gain falls, a strong nearby signal can still mask the weakness; once the distance grows or a wall intervenes, the signal can fall sharply below the usable threshold, so normal nearby and dead in the next room is indeed a pattern the antenna layer can produce. But a healthy antenna in a dense flat suffers the same way from concrete walls, neighbour interference, and router placement, and will also look worse in a far room. Other-device comparison and a distance log must therefore be combined with it: if other devices are fine at the same place and time while only this laptop struggles, suspicion of the antenna layer rises; if other devices struggle too, the environment and the far end must be suspected first.
Q5: I only see 2.4 GHz and no 5 GHz. Does that mean the 5 GHz antenna is broken?
No. Because MAIN and AUX are not divided by band, seeing only 2.4 GHz cannot be translated into a failed 5 GHz antenna. Possible causes include the antenna layer, a router that has not enabled that band, a device that does not support that band, and driver or system settings; this article expands only the antenna-layer possibility and treats the rest as boundary reminders. The practical checking order is to confirm the router has 5 GHz or 6 GHz enabled, then check whether other devices at the same place can see that band, and finally write the result into the log for inspection. One more physical fact must be stated first: 5 GHz penetrates walls worse than 2.4 GHz by nature, so a distant 5 GHz drop while 2.4 GHz still works can happen with a normal antenna and is not antenna evidence on its own.
Q6: An external USB Wi-Fi adapter works better. Does that prove the internal antenna is faulty?
No, it only points to the built-in path. An external USB adapter bypasses both the internal card and the internal antenna at once, so an improvement only narrows suspicion to the built-in path and cannot separate the card layer from the antenna layer at all; the external antenna's position and gain are also different variables. The correct reading is to write the external-adapter result into the log, explicitly noting that it cannot distinguish card from antenna, and leave the separation to inspection. The reverse also holds: no difference with an external adapter does not prove the built-in path is healthy, because interference, settings, and the far-end layer can affect internal and external performance together, so no single comparison can serve as a diagnosis.
Q7: I replaced the card, reinstalled drivers, and changed routers, and it is still weak. Must it be the antenna?
No. The reasoning is only half right: if the fault truly sits in the antenna, coaxial, or routing layer, replacing the card, reinstalling drivers, or replacing the router all still pass through the same antenna path, so the weak-signal pattern may remain exactly as before. But a persisting problem after replacement does not prove the antenna either, because settings, interference, wall material, and router placement that have not been excluded can equally keep the problem alive. A persisting problem after replacement is therefore only one clue that raises antenna-layer suspicion, and it must be read together with distance, angle, band, and other-device comparisons, with the final separation left to inspection rather than decided from a replacement history.
Q8: Hong Kong is so humid. Does that mean antenna connectors oxidise easily?
Only as a background factor, never as a diagnosis for an individual machine. The Hong Kong Observatory's 1991-2020 climatological normals give an annual mean relative humidity of 78 percent, but that figure is an outdoor climate background, not indoor humidity and certainly not humidity inside a machine. Connector literature notes that high temperature, high humidity, and corrosive gases can corrode base metals and form deposits that cause poor contact, yet deriving a failure rate, an oxidation speed, or a cause for one particular machine from a climate average is entirely unverified. At user level the connector must not be opened to look for oxidation anyway; any connector condition must be confirmed by inspection.
Q9: The antenna looks unbroken with no cracks. Does that mean the cable is fine?
No. Crushing and shielding damage can be outwardly inconspicuous: a cable squeezed flat may look only slightly deformed, yet its dielectric is already deformed, the spacing between inner conductor and shield has already changed, and its characteristic impedance and loss have already changed; a broken shielding braid or a cracked foil layer may sit under an intact-looking outer jacket while interference immunity has already fallen. This is exactly why a cable can look unbroken and still perform badly. The reverse also applies: a slightly scuffed jacket does not prove the inner layers are damaged. Visual judgement is not a user-level tool at all, because the cable is hidden inside the machine and cannot be seen without opening it, so appearance must never be used to rule anything in or out.
Q10: Why do 5 GHz and 6 GHz seem to fail before 2.4 GHz?
Two qualitative mechanisms explain it, but neither can be quantified. First, the higher the frequency, the shorter the wavelength, so a physical defect of a given absolute size occupies a larger proportion of the wavelength, and high-band performance therefore tends to deteriorate first. Second, coaxial loss rises with frequency, so the high bands are also more sensitive to cable damage. Both points are qualitative mechanisms only, and no measured source supports a statement such as several times worse, so no multiple may be written. The bands' own physics must also be remembered: 5 GHz and 6 GHz penetrate walls worse than 2.4 GHz by nature, so worse distant performance can be normal physics rather than damage, and only a same-distance, same-place band comparison plus other-device comparison can make the comparison meaningful.
Q11: Can I open the machine myself to check whether an antenna connector is loose?
No. The user level is limited to external comparison tests and settings checks: distance comparison, angle comparison, band comparison, other-device comparison, network comparison, time and event comparison, and a symptom log. Opening the machine, opening the screen, removing the base cover, unplugging internal cables, unplugging antenna connectors, replacing an antenna as a do-it-yourself job, or inspecting a connector for oxidation or cable damage are all out of bounds; manufacturers' own replacement procedures likewise require reading safety information first and restrict the work to authorised technicians. Opening the machine yourself can pull cables routed through the hinge, crush routing, or damage a connector, and can muddle evidence that inspection could otherwise have separated layer by layer. What the reader should deliver is an accurate log, not an opened machine.
Q12: Why should I not rely on signal bars alone or on a single test?
Because signal bars are a relative estimate, and different devices and different systems use different bar scales, so bars alone mislead; and because interference changes over time, a single test, good or bad, cannot settle anything. The robust method is to record distance, wall count, band, and bars or percentage together, repeat the observation at the same place at different times, and change only one variable at a time: when testing distance, do not also change the angle and position; when testing angle, do not also walk several steps away, or the comparison collapses. The value of a log lies in its pattern: the same place, the same angle, and the same band failing repeatedly is a useful clue, while scattered single numbers, however tidy, are not evidence.
Q13: How should pricing be understood? What is the difference between a starting price and this case's quotation?
The starting price must not be confused with this case's quotation, and that distinction is essential. The only confirmed starting prices are four: battery from HK$580, screen from HK$1,280, mainboard chip-level from HK$880, and keyboard from HK$480, each belonging to its own category only. Antenna, coaxial, and routing charges are always subject to a written quotation after inspection, and the battery, screen, mainboard, or keyboard starting prices must never be used to estimate or analogise antenna costs. The decision order must be: write the log, send the machine for inspection and layer separation, obtain the written quotation, and only then compare dimension by dimension using this article's framework; before a quotation exists, any conclusion that repair is or is not worthwhile is only a preliminary direction, not a decision. To understand the pricing range first, see pricing; to understand the inspection process, see diagnostic service.
Q14: What can readers do themselves, what must they never do, and what should they bring to inspection?
Everything the reader can do sits at user level: external comparison tests covering distance, angle, band, other devices, network, and time and event comparison; settings checks covering which band is currently connected and the signal bars or percentage; a symptom log following this article's template, recording place, distance, wall count, band, angle changes, other-device performance, external-adapter results, and whether any drop, impact, or repair preceded the problem; and bringing that log, the machine, and its original charger to inspection, with the exclusions already performed and their results written on the inspection form. What must never be done includes opening the machine, opening the screen, removing the base cover, unplugging internal cables, unplugging antenna connectors, replacing an antenna yourself, or forcing the hinge or flapping the lid rapidly for a test. The shop is at Room 1209, Dalton Plaza, Shek Mun, Sha Tin, telephone +852 2186 8010, WhatsApp +852 6558 6806, opening hours 12:00-19:00. To understand the pricing range first, see pricing; to understand the inspection process, see diagnostic service.
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.