GPU Broke After a PSU Swap? Modular Cable Reuse, Wattage & Wiring Diagnosis Guide

2026-11-13 | GPU Repair | 34 minutes | HK GPU Fix Editorial Team

GPU Broke After a PSU Swap? Modular Cable Reuse, Wattage & Wiring Diagnosis Guide

Key Takeaways

A graphics card that worked before a power supply swap and fails afterwards usually points to something that changed during the swap. This guide separates five suspicion layers — reused modular cables, wattage and transient spikes, wiring, the new unit itself, and disturbance to other parts — and walks through seven diagnostic steps to find which one is responsible.

What This Problem Is — and What It Is Not

Picture the sequence: the computer worked normally. Games ran, the desktop was stable, the graphics card was recognised every time. Then the power supply was replaced — for more wattage, for a quieter unit, for a newer standard, or because the old one was suspected of failing. After the swap, something is wrong. There is no picture at all, or the screen goes black under load, or the computer restarts during a game, or the card is no longer recognised, or a message appears asking for the PCIe power cables to be connected.

That sequence is the problem this guide covers, and the sequence itself is the most important piece of evidence in the whole diagnosis. A power supply swap does not change one thing. It changes several things at once: the unit, the cables, the connections, and everything that was touched inside the case while the swap was carried out. Any one of those changes can be the cause, and they need to be separated before any of them can be blamed.

Four conditions must all be true before the steps below apply:

  1. The computer worked normally before the power supply was replaced. Games ran, the desktop was stable, and the graphics card was recognised.
  2. The problem appeared after the swap, in close time order with it — at the first start after the swap, or at the first heavy load after the swap.
  3. The only deliberate change was the power supply. If a graphics card, memory, or motherboard was replaced at the same time, the cause cannot be separated by this guide alone, because two changes were made at once.
  4. The computer is a desktop computer. A laptop has no replaceable internal power supply in this sense; an external laptop charger is a different device with a different diagnosis.

The first condition deserves care, because it is the one most often remembered wrongly. If the computer was already restarting occasionally, or already going black under load, or already failing to recognise the card now and then before the swap, the swap is likely a coincidence rather than a trigger. A power supply is often replaced precisely because the computer was already misbehaving, and when the misbehaviour continues afterwards, the new unit was never the cause. Before going further, it is worth asking honestly whether the computer was truly normal before the swap, or whether the swap was itself an attempt to fix a problem that was already there. If it was already there, the right path is a general power-delivery diagnosis, not a swap-triggered one.

Why does the timing matter so much? Because a graphics card does not normally wear out at the exact moment a different part is replaced. When a working card fails immediately after a power supply swap, the most likely explanation is that something in the swap changed what the card receives, or changed the physical state of the card and its connections. The swap is not proof that the new power supply is faulty. It is proof that something changed, and the diagnosis that follows is about finding which change it was.

A note on the structure of this guide: the five suspicion layers, the seven steps, and the six misconceptions are the whole structure. Every section below refers back to them, and the numbers stay the same throughout — five layers, seven steps, six misconceptions — so that a result at any step can be traced back to the layer it tests.

The Five Suspicion Layers

A power supply swap changes the power delivery chain in several places at once. The five layers below must be kept separate. Mixing them together is how a cable problem gets blamed on wattage, or a wiring mistake gets blamed on a faulty unit, and how a new power supply gets returned when nothing was wrong with it.

Layer One: Reused Modular Cables — the Most Dangerous Layer

A modular power supply cable has two ends, and the two ends follow different rules. At the device end — the end that plugs into the graphics card, the motherboard, or a drive — the pin layout is standardised. The same pin position carries the same voltage whichever manufacturer made the cable. At the power supply end, there is no universal standard. Each manufacturer decides its own pin layout for the back of its own units. The same pin position can carry 12V on one manufacturer's unit and 5V, or a ground connection, on another's.

The most common mistake during a power supply swap is made to save effort: the old cables are left routed inside the case, the old unit is unplugged from the back of the case, and the new unit is plugged into the same cables. The cables look identical. They fit. The computer may even start. But if the pin layout at the power supply end is different, 12V can be delivered into a 5V position, and the result can be permanent damage to the graphics card, the motherboard, the power supply itself, or a drive. The danger of this mistake is that it gives no warning. It fits, it starts, and then damage is done.

The same brand does not guarantee compatibility. Corsair Type 3, Type 4, and Type 5 modular cables are not interchangeable with each other, and even the 24-pin layout differs between them. Two units from the same brand can be built by different original design manufacturers, with completely different internal backplane pin layouts. A cable that is correct for one unit can be wrong for another unit carrying the same brand name on the box.

The rule that follows from this is absolute: when the power supply is replaced, the whole cable set is replaced with it. Every cable that came in the new unit's box is used; no cable from the old unit is left in the case, not even a cable that looks unused or that serves a drive rather than the graphics card. The risk is not limited to the graphics card cables. A wrong cable on any connector can damage the part at the other end.

It is also worth understanding why this mistake can appear to be harmless for a while. A wrong cable does not always cause immediate, visible damage. In some cases the damage is gradual: a component receives the wrong voltage on a pin that tolerates it briefly, or a pin that is only lightly loaded, and the harm accumulates over days or weeks before the card fails. "It worked for a while with the old cables" is therefore not evidence that the old cables were correct. It is evidence that the damage took time to show.

Layer Two: Wattage and Transient Spikes

A graphics card does not draw power evenly. At idle it draws very little. In the instant it jumps from idle to full load, its current demand rises sharply in a very short time, forming a transient power spike. For modern high-end cards, that spike can reach two to three times the card's rated power draw. An RTX 4090, for example, draws around 450W continuously, while measured spikes have been recorded above 500W, with some tests recording spikes above 1000W. The ATX 3.0 and ATX 3.1 standards were designed in part to cope with these spikes, requiring a power supply to tolerate brief overloads of up to 200%.

A wattage mistake during a swap can take two forms. The first is simple: the new unit's rated wattage is lower than the system needs. An 850W unit replaced by a 650W unit, in a system with an RTX 4090 class card, is an example. The second is subtler: the wattage number looks sufficient, but the unit follows an older standard, ATX 2.x, and was not designed for the spike behaviour of a modern card. When a spike arrives, the unit's over-current protection trips, and the whole computer restarts or goes black.

The distinction matters, because the number on the box is not the whole story. A wattage figure is a continuous rating. A transient spike is a brief event that can be two to three times that rating. A unit can be adequate on paper and still trip its protection the first time a game loads a demanding scene. This is why a symptom that appears only under load — normal at idle, black screen or restart during a game — leans towards this layer, while a symptom that appears at the first start, before any load, leans elsewhere.

The official guidance is consistent on this point. NVIDIA's own guidance for an RTX 4090 system calls for an 850W unit or higher, and the ASUS ROG power supply calculator recommends 850W to 1200W for an RTX 4090 depending on the processor, and 1200W or higher with a native 12V-2x6 connector for an RTX 5090. Those figures are recommendations for the whole system, not for the card alone, and they already include headroom for spikes. A unit chosen by matching its number exactly to the system's steady demand, with no headroom, is a unit chosen without room for the spike.

Layer Three: Wiring — Not Seated, Wrong Connector, or Missing a Cable

A power supply swap requires every power cable in the case to be disconnected and reconnected. That creates far more opportunity for a wiring mistake than ordinary use does. The common forms are: a PCIe power cable not pushed fully home, with the latch not clicked shut; a 6-pin cable in an 8-pin socket, or the reverse; a 12VHPWR or 12V-2x6 connector not pushed fully in; or a missing cable, where a card that needs two 8-pin connectors has only one connected.

Some wiring mistakes announce themselves. A graphics card that detects a missing PCIe power connection can show a warning on screen asking for the power cables to be connected. But not every wiring mistake produces a warning. A cable that is seated almost fully, or a 12VHPWR connector that is a millimetre short of full insertion, can pass enough power for idle and the desktop, and then fail under load, when the card draws far more current through a connection that is not making full contact. The symptom then is normal at idle and a black screen or restart during a game — the same outward symptom as a wattage problem, with a completely different cause.

The way to think about this layer is that a connection is either fully made or it is not. There is no useful middle state. A latch that has not clicked, a connector that can still be pushed further, or a cable that wiggles in its socket is a connection that has not been made properly, even if the computer starts. Every power connector should be checked by hand, after the power is off, for full seating and a latched clip.

Layer Four: The New Unit Itself — Fault, Age, or Protection Tripping

A new power supply can itself be the problem. It can have a manufacturing fault. It can be a second-hand unit whose capacitors have aged, so its output is no longer stable under load. It can be a counterfeit or unbranded unit whose stated wattage is not what it can actually deliver. Or its protection circuitry can be unusually sensitive, tripping on a transient spike that a different unit would ride through.

The signature of this layer is a process of elimination. If the cables are the new unit's own set, the wattage is adequate with headroom, and every connector is fully seated and correct — and the problem still disappears the moment the old unit is put back — then the new unit itself is the remaining suspect. This layer cannot be diagnosed by looking at the unit. It is diagnosed by the comparison test in Step 2, which removes the unit and its cables from the system entirely and observes what changes.

One caution belongs here: a second-hand power supply carries a history that cannot be seen from the outside. Capacitor ageing, dust, heat exposure, and previous overloads all reduce a unit's ability to deliver stable power under load, while leaving its label unchanged. A second-hand unit that worked in a lighter system can fail in a heavier one, not because it is broken in an obvious way, but because its capacity has quietly fallen below what the new system needs.

Layer Five: Something Else Was Disturbed During the Swap

Replacing a power supply means working inside the case: reaching past the graphics card, moving cables, unplugging and replugging connectors, and sometimes removing the card to get access. In the course of that work, something else can easily be disturbed. The graphics card can be loosened in its slot. The gold fingers can shift position. Another cable — the processor power cable, or the motherboard 24-pin — can be left not fully seated. A cooler can be knocked out of position.

The signature of this layer is that the symptom has no direct relationship to power delivery, and that putting the old power supply back does not fix it. A computer that stops at a particular stage of starting, or a card that is no longer recognised at all, or a symptom that is unchanged by the old-unit comparison in Step 2, points here. The fault is not in the power supply. It is in something that was moved while the power supply was being changed.

This layer is easy to miss, because attention is naturally fixed on the part that was deliberately changed. But the case was opened, hands were inside it, and cables were moved. Any of those actions can disturb a connection that was previously fine, and a disturbed connection can produce a symptom that looks exactly like a power problem.

Taken together, the five layers give the diagnosis its shape. The cable layer is about which physical cables are in use. The wattage layer is about whether the new unit can meet the system's demand, including its spikes. The wiring layer is about whether every connection was made fully and correctly. The unit layer is about whether the new unit itself can deliver what it promises. And the disturbance layer is about what else was touched while the swap was carried out. Each of the seven steps below tests one or more of these layers, in an order that clears the most dangerous and the most decisive tests first.

The Seven Diagnostic Steps

Work through these steps in order, and change only one thing at a time. The order is part of the diagnosis. Step 2, the old-unit comparison, is the most decisive test in the guide, and it comes early for that reason. Do not skip ahead to wattage calculations or cable purchases before it has been done. And before any step that involves opening the case or reconnecting cables, switch the power off at the wall and wait for the system to discharge. Power cables must never be moved while the system is live.

Step 1: Record Before and After

Write down two sets of facts before anything is changed. First, the state before the swap: was the computer fully normal, or did it already have a problem? Which games ran, and for how long, without trouble? Second, the state after the swap: which symptom appeared, and when does it appear — at the first start, on entering Windows, or only under load in a game?

The judgement this step produces is the foundation of everything that follows. If the computer was fully normal before the swap and the problem appeared immediately afterwards, the timing itself is strong evidence that the swap is the trigger, and the diagnosis should stay inside the five layers above. If the computer already had a problem before the swap, the swap is likely a coincidence, and the diagnosis should leave this guide for a general power-delivery investigation. Record the symptom's timing precisely, because "black screen at first start" and "black screen only during a game" point at different layers, and a vague memory of "it broke after the swap" cannot separate them.

This note is also what will be brought to an in-store inspection if the diagnosis ends at the stop line in Step 7, so make it while the details are fresh. Include the old unit's model and wattage, the new unit's model and wattage, which cable set is in use, and whether the problem appears at start-up, in Windows, or under load.

Step 2: Put the Old Power Supply Back — the Decisive Comparison

Refit the old power supply together with its own full cable set, and observe whether the symptom disappears. This is the single most powerful test in the guide, and its result divides the whole diagnosis in two:

One safety rule governs this step. Before it is attempted, check for any smell of burning, any visible scorch mark, any melted connector, or any abnormal heat. If any of those is present, do not switch the computer on again. Further starts risk turning repairable damage into permanent damage. The machine should be brought in for inspection instead.

The old-unit comparison also needs to be a clean comparison. The old unit must be refitted with its own cables, not with the new unit's cables, and every connector must be fully seated. A comparison that mixes old and new cables tests nothing, because the cable layer has not been isolated. One variable at a time is the discipline that makes this test decisive rather than confusing.

Step 3: Compare the Cable Sets — New Cables Against Old Cables

Check which cable set is currently in use, and whether it is the set that came in the new unit's box:

This step should be done even when the old-unit comparison in Step 2 has already pointed at the new unit or its cables, because it names which of the two is responsible. A problem that disappears when the new unit's own cables replace reused old cables was a cable problem, not a unit problem, and the new unit itself may be perfectly sound.

Step 4: Compare the Wattage

Check the new unit's rated wattage against the system's demand:

For reference, the connector ratings that frame this calculation are 75W for a 6-pin PCIe connector, 150W for an 8-pin PCIe connector, and 600W for a 12VHPWR connector, under the PCI-SIG PCI Express Card Electromechanical Specification and the Intel ATX 3.0 design guide. Those are the rated delivery limits of the connectors themselves, not a promise that any particular unit can sustain them under a spike.

A wattage comparison that finds the new unit adequate does not clear this layer entirely. A unit can be adequate on paper and still trip its protection on a spike if its protection is unusually sensitive, or if it is a second-hand unit whose capacity has fallen. That possibility is what Step 6, the isolation test, is for. But a wattage comparison that finds the new unit clearly too small is decisive on its own, and no further testing is needed before the unit is replaced with one of adequate rating.

Step 5: Check Every Power Cable, One by One

With the power off at the wall and the system discharged, check every power cable in turn:

This step must only be done with the power off. Power cables must never be moved while the system is live. A connector that can still be pushed further, or a latch that has not clicked, is a connection that has not been made properly, even if the computer starts. Reseat each connector deliberately, and confirm the latch by feel as well as by sight.

If a connector shows a scorch mark, a melted housing, or a bent pin, do not reuse it. A damaged connector is a fault in itself, and continuing to use it risks further damage. The cable set should be replaced with the new unit's own set, and the mating socket on the card or the board should be inspected before the computer is started again.

Step 6: Isolate a Single Variable

If the steps above have not found the cause, isolate one variable at a time:

The discipline of one variable at a time is what turns a test into evidence. A test that changes two things at once produces a result that cannot be attributed to either of them, and the next time the problem appears, there will be no record of which change mattered. Keep a note of each single change and its result, in the same record started in Step 1.

Step 7: Cross-Test and the Stop Line

Two cross-tests complete the diagnosis:

The stop line is reached when any of the following is true: there is a smell of burning, a visible scorch mark, or a melted connector; the old-unit comparison has been done and the symptom remains with the old unit in place; or all seven steps have been worked through, one variable at a time, and the problem is still present. At that point, stop switching the computer on and off. Bring the machine in with the record from Step 1: the old and new units' models and wattages, which cable set is in use, whether the computer was normal before the swap, the result of the old-unit comparison, and when the symptom appears. In-store inspection is charged at the site's published three-tier fees of HK$400–2,800, quoted after inspection. Inspection takes place at Room 1209, Dali Plaza, Shek Mun. The shop can be called on 2186 8010, or reached on WhatsApp at 6558 6806. There is no home-visit service.

A Quick Comparison: The Five Suspicion Layers at a Glance

Suspicion layerWhat changes in the swapHow to test itWhat points at it
Reused modular cablesThe old unit's cables stay, the new unit plugs into themCheck whether the cables in use came in the new unit's boxOld cables in use; symptom persists with new cables ruled out
Wattage and transient spikesA lower rating, or an older ATX standardCompare the new rating with the card maker's recommendation plus spike headroomNormal at idle, black screen or restart only under load
WiringEvery cable disconnected and reconnectedReseat each connector with power off; count the cables the card needsLatch not clicked; a cable missing; warning about PCIe power
The new unit itselfA different unit with its own historyOld-unit comparison in Step 2; cross-test in Step 7Old unit restores normal operation; cables, wattage, wiring all checked
Disturbance to other partsHands and cables moved inside the caseOld-unit comparison leaves the symptom unchangedSymptom unchanged by the old unit; card not recognised; start-up stalls

Boundaries: Problems That Look Similar but Are Not This One

A diagnosis is only as good as its boundaries. Several nearby problems share a power supply, a swap, or a black screen, but each belongs to a different guide, and the steps above will not fix them:

The principle is to isolate one thing at a time. A symptom that appears only after a swap, only under load, only with the new cables, or only with the old unit refitted is evidence — and the evidence points in different directions depending on when it appears. The seven steps are arranged so that each result narrows the field, and the note kept along the way is what turns a frustrating failure into a diagnosis someone else can continue.

Six Common Misconceptions

Misconception 1: "The card failed after the new power supply went in, so the new power supply must be faulty."

No. A failure after a swap only shows that something changed during the swap. It does not show that the new unit itself is faulty. Reused cables, a wiring mistake, or something disturbed inside the case can all be the cause. Until the old-unit comparison in Step 2 has been done, no single part should be blamed. The comparison is what separates the unit from the cables and the wiring around it.

Misconception 2: "The cable fits, so it must be the right cable."

No. A modular cable fitting into its socket only shows that the plastic housing has the right shape. It says nothing about the voltage on each pin at the power supply end, where there is no universal standard. A wrong cable can deliver 12V into a 5V position and permanently damage hardware. When the power supply is replaced, the whole cable set must be replaced with it, and the cables that came in the new unit's box are the only ones that should be used.

Misconception 3: "Cables from the same brand must be compatible."

No. The same brand can use different series, different factories, and different pin layouts. Corsair Type 3, Type 4, and Type 5 modular cables are not interchangeable with each other, and even the 24-pin layout differs between them. Two units from one brand can be built by different original design manufacturers with completely different backplane pin layouts. The brand name on the box is not evidence that a cable is correct for a particular unit. The model is what matters, and the cables in the unit's own box are the reference.

Misconception 4: "The wattage number is high enough, so the power delivery must be enough."

No. A wattage figure is a continuous rating. A graphics card's transient spike can reach two to three times its rated power draw, and an older-standard unit can trip its protection on that spike even when its wattage number looks adequate. The ATX 3.0 and ATX 3.1 standards exist in part because of these spikes, and a unit built to an older standard may not ride through them. Wattage being enough on paper is not the same as the unit coping with the spike.

Misconception 5: "Replacing a power supply cannot affect the BIOS settings, so there is no need to check settings."

That is half right, and the half that is right is itself a diagnostic clue. Replacing a power supply does not reset the BIOS settings. That is different from replacing a processor, memory, or motherboard, each of which can reset settings. The distinction is useful: if the symptom looks like a settings problem — a card not recognised, or a resolution locked to a low value — and only the power supply was changed, the settings layer is not the first suspect. The power delivery layer and the physical disturbance layer should be checked first. The fact that settings were not touched narrows the search, rather than widening it.

Misconception 6: "Putting the old power supply back fixed it, so the new one is defective and should be returned at once."

Only half right. Putting the old unit back and seeing the problem disappear proves that the problem lies somewhere among three things: the new unit, the new cable set, or the new wiring. It does not prove that the new unit itself is faulty. In many cases the cause is reused cables or a wiring mistake, and the new unit itself is perfectly sound. Before returning anything, reconnect the new unit using its own cable set, with every connector fully seated, and run the comparison again. A return made on a cable mistake wastes the return and leaves the real cause unaddressed.

When to Stop and Bring the Machine In

Power problems after a swap are narrowed down by the seven steps above — often at Step 2, where the old-unit comparison either clears the card entirely or clears the new unit entirely, or at Step 3, where a reused cable set is found. If all seven steps have been worked through, one variable at a time, and the problem is still present — or if there is a smell of burning, a scorch mark, or a melted connector at any point — stop. Further starts risk turning repairable damage into permanent damage, and further guessing risks changing things that were never part of the problem.

Bring the machine in with the notes from Step 1 and the results of each step tried. In-store inspection takes place at Room 1209, Dali Plaza, Shek Mun. You can call 2186 8010, or reach the shop on WhatsApp at 6558 6806. Inspection is charged at the site's published three-tier fees of HK$400–2,800, quoted after inspection. There is no home-visit service; inspection is carried out in the shop.

One last point about expectations: a diagnosis that ends at the stop line is not a failed diagnosis. A note that says "normal before the swap; old unit restores normal operation; new unit's own cables in use; wattage checked; every connector reseated; symptom appears only under load" is a precise description of where the fault lives, and it is exactly what an in-store inspection needs to start from the right place instead of from the beginning.

FAQ

Q1: The card failed after I fitted a new power supply. Does that mean the new power supply is faulty?

No. A failure after a swap only shows that something changed during the swap. Reused cables, a wiring mistake, or something disturbed inside the case can all be the cause. Until the old-unit comparison in Step 2 has been done, no single part should be blamed. The comparison is what separates the new unit from the cables and the wiring around it.

Q2: The old cable fits the new power supply, so it must be the right cable. Is that right?

No. Fitting only shows that the plastic housing has the right shape. At the power supply end of a modular cable there is no universal pin standard, so a cable that fits can still deliver 12V into a 5V position and permanently damage hardware. When the power supply is replaced, the whole cable set must be replaced with the set that came in the new unit's box.

Q3: Cables from the same brand must be compatible with each other. Is that true?

No. The same brand can use different series, different factories, and different pin layouts. Corsair Type 3, Type 4, and Type 5 modular cables are not interchangeable with each other, and even the 24-pin layout differs between them. The brand name on the box is not evidence that a cable is correct for a particular unit; the model is what matters.

Q4: The wattage number is high enough, so the power delivery must be enough. Is that right?

No. A wattage figure is a continuous rating, while a graphics card's transient spike can reach two to three times its rated power draw. An older-standard unit can trip its protection on that spike even when its wattage number looks adequate. Wattage being enough on paper is not the same as the unit coping with the spike, which is why the ATX 3.0 and ATX 3.1 standards were designed in part for these spikes.

Q5: Putting the old power supply back fixed it, so the new one is defective and I should return it at once. Is that the right conclusion?

Not yet. Putting the old unit back proves only that the problem lies among the new unit, the new cable set, or the new wiring. It does not prove that the new unit itself is faulty. In many cases the cause is reused cables or a wiring mistake, and the new unit is perfectly sound. Before returning anything, reconnect the new unit using its own cable set, with every connector fully seated, and run the comparison again.

Q6: Replacing a power supply cannot affect BIOS settings, so there is no need to check settings at all. Is that correct?

Half right. Replacing a power supply does not reset BIOS settings, unlike replacing a processor, memory, or motherboard. That distinction is itself a diagnostic clue: if the symptom looks like a settings problem and only the power supply was changed, the settings layer is not the first suspect. The power delivery layer and the physical disturbance layer should be checked first.

Q7: How should the seven-step diagnosis be done, and what should I do first?

Step one is to record the before and the after: whether the machine was normal before the swap, which symptom appeared after it, and exactly when it appears. That step decides which road the rest of the diagnosis takes, so do not skip it and go straight to changing cables or reinstalling drivers. Step two, refitting the old unit, is the decisive test; after that, check the cables, the wattage, and each power cable in turn, narrowing layer by layer.

Q8: The old power supply has already been thrown away. How do I do the comparison?

Borrow a unit known to be good and run the comparison once, or bring the machine in for inspection, where the shop's test unit can be used. A borrowed unit must be fitted with its own full cable set — never leave the old cables in. Checking a power supply's output is an inspection job; do not measure a live circuit yourself.

Q9: Something else was disturbed when the power supply was swapped. How do I tell?

If the fault remains even with the old unit back in place, the problem is not in the power supply — move to the physical disturbance layer: check whether the graphics card is fully seated, whether its gold fingers have shifted, and whether the processor power cable and the motherboard 24-pin are fully seated. The signature of this layer is a symptom with no direct link to power delivery — for example, the machine stopping at one stage of start-up, or the card no longer being recognised.

Q10: When should I stop trying things myself and bring the machine in for inspection?

Stop when there is a smell of burning, a visible scorch mark, or a melted connector; when the old-unit comparison has been done and the symptom remains with the old unit in place; or when all seven steps have been worked through, one variable at a time, and the problem is still present. Bring your notes: the old and new units' models and wattages, which cable set is in use, whether the computer was normal before the swap, the old-unit comparison result, and when the symptom appears. In-store inspection is at Room 1209, Dali Plaza, Shek Mun, charged at the site's published three-tier fees of HK$400–2,800, quoted after inspection. You can call 2186 8010 or WhatsApp 6558 6806.

Related articles

Related services

If you need Power connector repair, see these service pages: