So Daniel sent us this one. He writes: Back in 2015 I bought three identical AOC monitors, nothing special, but identical, which turned out to be the whole point. Once you mix in even a slightly different model from a different year, the colours don't match and the brightness curves don't match, and it quietly drives you mad. Which is exactly why I've spent years dreading one of them dying. One now has.
Oh, the triple-monitor curse. I know this pain.
Two of them failed while under warranty, and he had them factory repaired. Both times the symptom was the same, an audible whine from inside the monitor. Both times it was, as far as he was told, a fairly straightforward component-level fix. He understood it as a failed transistor. And both times the monitor was shipped the length of the country and back for what was probably twenty minutes of bench work and a part worth pocket change. Now the warranty is long gone and he has no recourse at all. This feels like the textbook right-to-repair case. If he had a modest inventory of small electronic components on a shelf, could he have just done this himself? So, is component-level repair of consumer display electronics actually feasible for a competent amateur, and what would he need to stock?
That is a fantastic question. And the answer is yes, but with asterisks. Big asterisks. Some of them shaped like lightning bolts.
He has a whole list of things he wants us to cover. What is this even called, board-level versus component-level repair. What that whine was actually telling him, and he's self-aware enough to say he guessed transistor but wants the honest diagnosis. Safety, and he specifically says please give this real weight, not a throwaway line. The parts inventory, a concrete shopping list. The gear, soldering station, ESR meter, the works. The feasibility verdict, honestly, what proportion of failures can a determined amateur actually fix. And sourcing, with the note that he's in Israel and wants to know what that means for parts and lead times. Then he wraps up by asking what you'd tell someone who wants to try exactly one repair to find out whether this hobby is for them.
That is a properly thought-through question. He's done his homework.
He has. I will say, the man has three identical monitors and two have already failed. At some point you have to ask whether the monitors are the problem or whether AOC just had a rough year in 2015.
That's not entirely fair. Switching power supplies in consumer monitors, especially from that era, have a handful of known failure modes that show up across every brand. It's not an AOC problem, it's a physics problem. And the fact that they all failed the same way is actually good news for repairability. It means the failure is predictable.
Fair enough. So where do we start?
Let's start with what to call this. Daniel asked for the name, and the name matters because it tells you where you sit on the repair spectrum. At one end you've got module swapping. That's when you diagnose down to the power board or the main board or the panel, and you swap the whole board. No component-level work, just connectors and screws. That's what most repair shops actually do now, because labour costs more than boards. At the other end you've got reverse engineering, where you're tracing out circuits from a blank board with no schematic, probing signals with an oscilloscope, and essentially reconstructing the design to find the fault. That's engineering work. What Daniel is asking about sits right in the middle, and the industry term is component-level repair. You're not designing anything, you're not guessing. You're identifying a failed component on a known-good board and replacing it with an equivalent part. The board already worked once, the design is proven, you're just putting it back the way it was.
So it's the difference between replacing the engine and replacing a spark plug.
And here's the thing that makes it viable. The boring truth, as Daniel put it, is that the same handful of parts fail in the same handful of ways across a huge range of devices. Switching power supplies are everywhere. Your monitor, your TV, your laptop charger, your fridge, your washing machine. They all use the same topology, the same families of components, and they fail in the same three or four ways. Once you learn to fix one, you can fix dozens.
So let's talk about that whine. He said transistor, and he's clearly open to being corrected. What was actually happening inside his monitor?
I'm glad he asked, because this is where the diagnosis gets interesting. An audible whine from a monitor almost always comes from the power supply section, specifically the switching supply that converts mains AC to the various DC rails the monitor needs. In a 2015-era LED-backlit monitor, you've typically got a single board that handles both the main power conversion and the backlight drive. The whine is a mechanical vibration at the switching frequency or a harmonic of it, usually somewhere between one kilohertz and twenty kilohertz, right at the edge of human hearing. Something is vibrating that shouldn't be.
And his guess was a transistor.
It's not a bad guess, but it's probably not the transistor itself. A switching MOSFET doesn't whine. It either works or it fails short and blows the fuse, or it fails open and you get no power. What whines is a magnetic component, a transformer or an inductor, and it whines because the current waveform going through it has gone wrong. The most common cause, and I'd bet real money this was the case in his monitors, is degraded electrolytic capacitors on the output side of the supply.
Capacitors. The thing that costs twelve cents.
Sometimes less. Here's what happens. The switching supply has electrolytic capacitors on the secondary side to filter the output. Over time, especially in a monitor that runs warm for hours a day, the electrolyte inside those capacitors dries out. The equivalent series resistance, the ESR, goes up. When the ESR rises, the capacitor can't smooth the ripple current properly anymore. The ripple increases, the control loop starts hunting to compensate, the switching waveform gets jittery, and the transformer or output inductor starts singing. That's your whine. The capacitor hasn't failed completely yet, the monitor still works, but it's dying. Eventually the ripple gets bad enough that the output voltage sags, the monitor starts flickering or shutting down, and then it won't turn on at all.
So the transistor was probably fine the whole time.
Probably. The tech who did the repair might have replaced a capacitor and just told the service desk it was a transistor because that sounds more like a real repair. Or they replaced a capacitor and a transistor as a precaution. But the root cause was almost certainly electrolytics. The other possibility is a failing backlight boost converter. In an LED monitor, the backlight runs at a higher voltage than the main logic rail, maybe twenty-four to sixty volts depending on the panel, and there's a boost converter to generate that. If the output capacitors on the boost converter go bad, same story, whining inductor. You can tell them apart by where the whine changes. If it changes pitch when the screen brightness changes, it's the backlight supply. If it's constant regardless of what's on screen, it's the main supply.
That's genuinely useful. So before we get to the parts list, we need to talk about the thing Daniel specifically asked us to give real weight to. Safety. What's inside a monitor that can kill you?
This is the part where I stop being the enthusiastic nerd and become the retired pediatrician who has seen what electricity does to a human body. I am going to be very direct here. A switch-mode power supply has a section called the primary side. That's everything before the isolation transformer. On the primary side, the mains voltage comes in, gets rectified, and charges a large electrolytic capacitor, the bulk capacitor, to roughly one point four times the mains RMS voltage. On 230-volt mains, which is what Daniel has in Israel, that capacitor sits at around 325 volts DC. That is not a maybe, that is not an if. That capacitor is charged to 325 volts whenever the monitor is plugged in. And here's the part that catches people. When you unplug the monitor, that capacitor does not magically discharge. There is no bleeder resistor on every design, and even when there is, it can fail open. That capacitor can hold a lethal charge for minutes, hours, or in some poorly designed supplies, days.
So you unplug the monitor, you open it up, you touch the wrong thing, and you're having a very bad day.
You're having your last day. 325 volts DC across the chest is more than enough to stop a heart. So rule number one, and this is non-negotiable. Before you touch anything on the power supply board, you discharge that bulk capacitor. The right way to do this is with a resistor, not a screwdriver. A screwdriver across the terminals will discharge it instantly with a bang, a spark, and possibly molten metal in your eye, and it can damage the capacitor internally. Use a power resistor, something like a 10-kilohm, 5-watt wirewound resistor. Solder or clip insulated leads to it, and touch the leads across the capacitor terminals for at least ten seconds. Then measure the voltage across the capacitor with your multimeter to verify it's below about five volts. Only then is it safe to handle.
And if you don't own a 10-kilohm 5-watt resistor?
Then you buy one before you open the monitor. They cost about fifty cents. This is not optional equipment.
Got it. What else?
The one-hand rule. When you're working on anything that has ever been plugged into the wall, keep one hand in your pocket or behind your back. If you do accidentally touch something live, the current goes from your working hand to your feet, not across your chest from hand to hand. Across the chest is the kill path. Hand to foot hurts but you'll probably survive. This is standard practice for anyone who works on tube amplifiers, CRT displays, or mains power supplies, and it should be standard practice here too.
That's a simple rule. One hand in the pocket.
Next, the dim-bulb tester. This is a brilliant piece of safety gear that costs about five dollars to build. You take an incandescent light bulb, a 60-watt or 100-watt bulb, in a socket, wired in series with the mains live wire. You plug the monitor into the dim-bulb tester, and the tester into the wall. If the monitor has a short circuit, the bulb glows bright and limits the current so nothing explodes. If the monitor is healthy, the bulb might glow dimly for a moment as the bulk capacitor charges, then go dark. It's a current limiter, a short-circuit detector, and a smoke preventer all in one. Every time you power up a repaired supply for the first time, you do it through the dim-bulb tester.
And if the bulb glows bright and stays bright?
You have a short. Power down immediately and find it before you try again. The bulb saved your replacement parts from going up in smoke.
What about the isolation transformer he mentioned?
This is where we need to be precise about when you need one. An isolation transformer is essential if you're probing a live mains-referenced supply with an oscilloscope. Your oscilloscope probe ground is connected to earth ground. If you connect that ground clip to a point on the primary side of a non-isolated supply, you create a dead short from that point to earth, and you will destroy your scope, your probe, and possibly your face. An isolation transformer breaks that ground reference and makes it safe to probe the primary side. However, for the kind of repair Daniel is likely doing, replacing capacitors and diodes with the power off, he does not need an isolation transformer. He's not probing a live supply. If he ever advances to the point of doing live diagnostics with a scope, then yes, isolation transformer, no question. But for now, it's not on the must-buy list.
And the extra hazard in older CCFL-backlit monitors?
Important distinction. Daniel's monitors are 2015 models, which means they're almost certainly LED-backlit. The backlight voltage in an LED monitor is typically under a hundred volts, which is still dangerous but not in the same category as CCFL. CCFL backlights, the cold cathode fluorescent lamps used in older LCD monitors, run at somewhere between 600 and 1500 volts AC at high frequency. That inverter transformer will absolutely kill you. If you ever open a monitor and see thin glass tubes running along the edges of the panel, that's CCFL. Stop. The inverter section is lethal even when the monitor is off if the capacitors haven't discharged. For Daniel's LED monitors, the hazard is the mains bulk capacitor and the primary side of the supply. Still lethal, but contained to a specific section of the board you can identify and avoid or discharge.
So the line is, LED backlight supply you can work on if you discharge the bulk cap and follow the one-hand rule. CCFL backlight, think twice and maybe don't.
For a beginner, CCFL repair is past the line. The voltages in the inverter section are terrifying. Start with LED backlit units, flat-screen TVs and monitors from roughly 2012 onward.
Alright. Safety covered, and I appreciate you not soft-pedaling that. Now let's get to what Daniel actually asked for. The parts inventory. If he wants to stock a shelf so that when the next monitor whines, he can fix it that evening instead of shipping it across the country, what goes on that shelf?
This is my favourite part. Let's build a starter component inventory for switch-mode power supply repair. The good news is it's surprisingly affordable. The bad news is you have to be picky about what you buy and where you buy it, because counterfeit components are a real problem, especially for electrolytic capacitors and semiconductors. But we'll get to sourcing in a minute. Let's start with the parts.
Capacitors first, I assume.
Capacitors first. Electrolytic capacitors are the number one failure point in switching supplies, and you want low-ESR types rated for 105 degrees Celsius, not 85 degrees. The 105-degree rating matters because the inside of a monitor runs warm, and capacitor life halves for roughly every 10 degrees Celsius rise. A 105-degree cap will outlast an 85-degree cap by years in that environment. The brands to look for are Nichicon, Rubycon, Panasonic, and United Chemi-Con. These are the Japanese majors and they are the gold standard. For a starter kit, I'd stock the following values. These are the ones that fail over and over across monitors, TVs, and power supplies.
Give me the list.
For the secondary side filtering, you want low-voltage, high-capacitance parts. Stock 10 of each: 1000 microfarad at 25 volts, 470 microfarad at 25 volts, 2200 microfarad at 16 volts, and 1000 microfarad at 16 volts. These are the workhorse values on the output rails. For the primary side, you need higher voltage parts. Stock 5 each of: 47 microfarad at 50 volts, 100 microfarad at 50 volts, 22 microfarad at 50 volts. These are common on the primary-side startup and control circuitry. Also stock a few 2.2 microfarad and 4.7 microfarad at 50 volts for the small electrolytics that dry out and cause startup failures. And then the big one, the bulk capacitor on the primary side. This is typically 100 to 220 microfarad at 400 or 450 volts. Stock maybe two of the most common value, 150 microfarad at 450 volts. They're bigger and more expensive, about three to five dollars each, but when they go, the monitor is completely dead.
What's the total on capacitors so far?
Roughly forty to fifty capacitors. If you buy genuine Japanese parts from a reputable distributor, you're looking at maybe thirty to forty dollars for the whole lot. That's the capacitor drawer. Next, semiconductors. For diodes, you want a selection of fast-recovery and Schottky rectifiers. The common through-hole packages are DO-201 and DO-41. Stock the UF4007, that's an ultra-fast 1000-volt 1-amp diode, maybe ten of those. The 1N4148 small-signal diode, a pack of twenty, they cost pennies. For Schottky diodes, the 1N5819, 40-volt 1-amp, and the SB540, 40-volt 5-amp. Five of each. These fail in the output rectification stage.
And the transistors he originally blamed?
MOSFETs do fail, especially when something downstream shorts and takes them out. The most common switching MOSFETs in monitor supplies are in the TO-220 package. I'd stock a few IRF840, that's a 500-volt 8-amp N-channel MOSFET, and a few IRFBC20 or similar 600-volt parts. Also the 2N7000 small-signal MOSFET in TO-92 for control circuits. Maybe five of each. For bipolar transistors, the 2N3904 and 2N3906, NPN and PNP small-signal, and the 2SC1815 and 2SA1015 which are common in Asian-designed supplies. Ten of each, they're practically free.
What else?
Optocouplers. These provide the isolation between the primary and secondary sides for the feedback loop. The industry standard is the PC817, a 4-pin DIP package. Stock five of them. They fail occasionally, and when they do, the output voltage goes unregulated and either sags or spikes. Shunt regulators, the TL431, three-pin TO-92 package. This is the reference voltage source for the feedback loop. It's in practically every supply. Stock ten, they cost about ten cents each. Fuses, glass cartridge fuses in the common sizes, 2-amp, 3.15-amp, 4-amp, and 5-amp, all 250-volt rated. A pack of each. And a few NTC thermistors for inrush current limiting, the kind that go in series with the mains input, typically 5 to 10 ohms cold. Maybe three of the 5-ohm 5-amp type.
So we've got capacitors, diodes, MOSFETs, bipolar transistors, optocouplers, shunt regulators, fuses, and thermistors. What's the total damage?
If you're buying from AliExpress or from a surplus dealer, you could put this whole kit together for maybe sixty to eighty dollars. From a proper distributor like Mouser or Digi-Key, maybe a hundred to a hundred twenty dollars. That's a fully stocked repair drawer that will cover maybe 80 percent of the component failures you'll encounter in consumer switch-mode supplies.
And what can't you stock?
Great question. You cannot stock the proprietary parts. The main scaler IC, the large QFP or BGA chip that handles the video processing, is custom to the monitor and only available as a board pull from a donor unit. The PWM controller IC, sometimes it's a standard part like the UC3842 or the SG6841, and you can stock those, but sometimes it's a custom house-numbered part and you're out of luck. The EEPROM that stores the monitor's firmware and EDID data, you can replace the chip but you can't replace the firmware unless you've backed it up. And the LCD panel itself, if the panel has a row or column driver failure, that's not repairable at the component level. That's a panel swap. But for the power supply, which is where the vast majority of failures happen, the parts are standard and stockable.
Alright. So now he has a parts drawer. What gear does he need to actually do the repair?
Let's start with the essentials and work up. First, a temperature-controlled soldering station. You cannot do this work with a fixed-temperature pencil iron from the hardware store. You need temperature control because different joints need different heat, and because you'll be working on boards with ground planes that suck heat away. The entry-level standard is the Hakko FX-888D, about a hundred dollars. The Chinese clone stations, the ones based on the Hakko T12 tips, are actually quite good now. Look for a T12-compatible station on AliExpress, search for T12 soldering station, roughly thirty to fifty dollars. They heat up fast, they hold temperature well, and the tips are cheap. Get a selection of tips, a chisel tip around 2.4 millimeters for through-hole work, and a fine conical tip for smaller joints.
And for desoldering?
You need at least two methods. A desoldering pump, the spring-loaded solder sucker type, is essential for clearing through-hole joints. The Engineer SS-02 is a Japanese-made pump with a silicone nozzle that seals well and lasts. About fifteen dollars. Also stock desoldering wick, sometimes called solder braid, for cleaning up pads and removing residual solder. Get a 2-millimeter wide roll, name brand like Chemtronics or MG Chemicals. The cheap no-name wick often has poor flux and doesn't work well. About five dollars a roll. Hot air is optional for a beginner but becomes essential if you ever need to remove a surface-mount IC. A basic 858D hot air station is about forty dollars on AliExpress. Not needed for the capacitor-and-diode work Daniel is likely to start with, but worth knowing about.
Now the tool he specifically mentioned, the ESR meter.
This is the diagnostic tool that changes everything. An ESR meter measures the equivalent series resistance of a capacitor in-circuit, without desoldering it. When an electrolytic capacitor dries out, its capacitance might still measure close to spec, but its ESR goes through the roof. A normal multimeter can't tell you that. An ESR meter can. You go through the board, probe each electrolytic, and the one with the high ESR is your culprit. It turns a guessing game into a five-minute diagnostic procedure.
So it pays for itself in saved time immediately.
On the first repair. There are dedicated ESR meters like the Peak Atlas ESR70, about eighty dollars, very good, very reliable. There are also the multi-function component testers, the little boards with a colour screen and a ZIF socket that you see on AliExpress for about fifteen to twenty dollars. Search for LCR-T4 or multi-function tester. These measure ESR, capacitance, inductance, and they'll identify transistors and tell you the pinout. They're not as precise as a dedicated ESR meter but they're astonishingly capable for the price. I'd say start with one of those, and if you find yourself doing a lot of this work, upgrade to the Peak or to a proper LCR meter later.
What else is on the bench?
A decent digital multimeter. Doesn't have to be a Fluke, though Flukes are wonderful. A Uni-T UT61E or UT139C is perfectly adequate for this work, about thirty to fifty dollars. You need voltage, resistance, continuity, and diode test. Capacitance measurement is nice to have but the ESR meter does that better. A bench power supply is useful for testing repaired boards outside the monitor, but not essential for a beginner. If you want one, a basic 30-volt 5-amp unit like the Korad KA3005D is about eighty dollars. Magnification. You need to read tiny part numbers and inspect solder joints. A headband magnifier with interchangeable lenses, about fifteen dollars, or a USB microscope for about thirty dollars. ESD protection. At minimum, an anti-static wrist strap and a grounded mat. About fifteen dollars for a basic set. Don't skip this. You can damage components with static discharge and not know it until the thing fails again a month later.
So if we add all that up for a beginner setup?
Soldering station, let's say the T12 clone at forty dollars. Desoldering pump and wick, twenty dollars. Multi-function tester for ESR, twenty dollars. Multimeter, forty dollars. Magnification, twenty dollars. ESD kit, fifteen dollars. That's about a hundred fifty-five dollars for a fully functional repair bench. If you add the Hakko instead of the clone, and a dedicated ESR meter, you're closer to three hundred. Either way, it's less than the cost of one new monitor.
And the parts drawer was another sixty to a hundred twenty dollars. So all in, we're talking somewhere between two hundred and four hundred dollars to be equipped for component-level repair on switching supplies for the foreseeable future.
The skills transfer to everything. Your TV, your laptop charger, your microwave, your washing machine control board. They all have switching supplies, and they all fail the same way.
Let's talk sourcing, because Daniel specifically asked about this, and he's in Israel. Where does he actually buy all this stuff?
This is where we have to be honest about the counterfeit problem. Counterfeit electrolytic capacitors are endemic. A counterfeit cap will have the right markings, maybe even a convincing sleeve, but inside it's a smaller, lower-voltage capacitor in a bigger can, or it's a used part that's been re-sleeved. It will fail in months, sometimes spectacularly. Counterfeit semiconductors are also common, MOSFETs with half the rated current, transistors that are actually a different part with the markings sanded off and re-printed. This is not a theoretical problem. It's the reason you don't buy electrolytic capacitors or power semiconductors from random AliExpress sellers unless you know exactly which sellers are trustworthy.
Where do you buy them?
For electrolytic capacitors, I strongly recommend buying from authorized distributors. Mouser and Digi-Key both ship to Israel. Shipping is about twenty to thirty dollars for a small order, and it takes about a week. Yes, the parts cost more than AliExpress, but you are getting genuine Nichicon or Panasonic capacitors from the factory, not from a bin in Shenzhen. For a capacitor kit that you're going to rely on for years, spend the extra money. The peace of mind is worth it. For things like resistors, ceramic capacitors, fuses, connectors, generic diodes like the 1N4148, AliExpress is fine. The counterfeiting problem is concentrated in electrolytics and power semiconductors. For MOSFETs and Schottky diodes, I'd also go with Mouser or Digi-Key, or use a trusted local source.
What are the local options in Israel?
For tools, KSP is good for multimeters and basic soldering equipment. They carry some Uni-T meters. For components, Lastiq and Robotix are the hobbyist electronics shops. They stock Arduino and Raspberry Pi accessories, some basic components, but their selection of power supply repair parts is limited. They're great for the tools, less so for a specific Nichicon capacitor in a weird value. For components specifically, your best bet for speed is to build a relationship with one of the industrial electronics suppliers in Israel. There are distributors who carry Vishay, Panasonic, and other major brands, but they often have minimum order quantities. For the one-off repair, Mouser and Digi-Key are the most practical. You pay shipping but you get exactly what you ordered and it's genuine.
The rule of thumb is, AliExpress for the cheap modules and generic passives, Mouser or Digi-Key for electrolytics and power semiconductors, local shops for tools and the screws you'll lose.
One more note on sourcing. If you're buying from AliExpress, the shipping to Israel is typically free but slow, two to four weeks. Plan ahead. Build your parts drawer before you need it, not when the monitor is already dead and you're staring at a blank screen.
Now we get to the feasibility verdict. Daniel asked honestly, what proportion of these failures can a determined amateur actually fix?
For switch-mode power supply failures in LED-backlit monitors, I'd say a determined amateur with the right tools and a basic parts stock can fix maybe 70 to 80 percent of them. The most common failures are electrolytic capacitors, and those are straightforward to diagnose with an ESR meter and straightforward to replace with a soldering iron. The next tier is diodes and MOSFETs, which are also replaceable with basic soldering skills. The tier after that is the PWM controller IC, which might be an 8-pin surface-mount part, doable with care and maybe some flux. Where it gets hard is when the failure is in a BGA chip, or when the board has been damaged by a short that burned a trace, or when the fault is intermittent and heat-dependent and you can't pin it down without an oscilloscope and a schematic.
Where do beginners usually go wrong?
Lifted pads. That's the number one beginner mistake. They apply too much heat for too long, or they pull a component before the solder is fully molten, and the copper pad lifts off the board. Now you have a broken connection and you need to run a jumper wire. It's fixable but it's demoralizing. The fix is patience and flux. Use extra flux, it helps the solder flow and reduces the time you need to apply heat. The second mistake is putting a capacitor in backwards. Electrolytics are polarized. The stripe on the can marks the negative lead. If you install it backwards, it will vent, possibly explosively, the moment you apply power. Take a photo of the board before you remove anything. Note the orientation of every polarized component. The third mistake is not testing before full reassembly. After you replace parts, power the board up through the dim-bulb tester with the monitor still open. If the bulb glows dim and the monitor shows signs of life, then you can reassemble. If the bulb glows bright, you have more work to do.
The economics? Daniel acknowledged this upfront. For a ten-year-old 21-inch monitor, the repair almost certainly isn't worth it on pure money terms.
It's not. A replacement 21-inch monitor costs maybe a hundred dollars. The parts for the repair might cost two dollars. But the labour, the tools, the learning curve, none of that pencils out if you're only fixing one monitor. The case for doing this is not financial. It's about e-waste. A monitor that goes to landfill is a monitor that will sit there for centuries. The plastics, the lead in the solder, the mercury in older backlights, none of that biodegrades. Fixing it keeps it out of the ground. It's about the absurdity Daniel mentioned, freighting a monitor the length of the country for a component worth a few shekels. The carbon footprint of that shipping is probably a thousand times the carbon footprint of manufacturing the capacitor. And it's about the skill. Once you can fix a monitor power supply, you can fix so many other things. You become less dependent on the replacement economy. That's the right-to-repair argument in a nutshell. It's not about saving money on this specific repair. It's about reclaiming the ability to fix your own stuff.
For someone who wants to try exactly one repair to find out whether this hobby is for them, what's the sensible first project?
Don't start with a monitor that's plugged into the wall. Start with something low-voltage. Find a dead router, a dead set-top box, a dead external hard drive enclosure. Something with a 12-volt or 5-volt wall-wart supply. Open it up, look at the board, find the electrolytic capacitors. Practice desoldering and resoldering them on a board that doesn't matter. Get a feel for how much heat it takes, how the solder flows, what a good joint looks like versus a cold joint. Do that three or four times on junk boards. Then, when you're comfortable, find a dead monitor. Ideally one that someone is throwing away. Check that it's LED-backlit, not CCFL. Open it, discharge the bulk capacitor properly, photograph everything, and start probing electrolytics with your ESR meter. I would bet real money that you find at least one capacitor with high ESR, and replacing it brings the monitor back to life. That first successful repair is an incredible feeling. You'll be hooked.
If he's hooked, what's the upgrade path?
After you've done a few capacitor replacements, you'll want an oscilloscope. A basic digital scope like the Rigol DS1054Z or the Siglent SDS1104X-E, about three hundred to four hundred dollars. That lets you actually see the switching waveform, check for ripple, verify that the supply is stable under load. It's the tool that takes you from component swapping to real diagnosis. But you don't need it on day one. Day one is a soldering iron, a multimeter, an ESR meter, and a dim-bulb tester.
To wrap this up. Daniel wanted a concrete shopping list, a realistic budget, and an honest verdict. Let's put the numbers together.
Tools, roughly a hundred fifty-five dollars for the basic setup, up to three hundred if you buy name-brand. Parts drawer, sixty to a hundred twenty dollars depending on where you source. Total startup cost, somewhere between two hundred and four hundred dollars. For that, you get the ability to diagnose and repair the most common failures in switch-mode power supplies across monitors, TVs, and a huge range of consumer electronics. The first repair that would otherwise have been a replacement, you've broken even. Everything after that is gravy. And you've kept a chunk of e-waste out of the ground.
The misconception. What's the single most common wrong belief people hold about this kind of repair?
That you need to be an electrical engineer with a full schematic and an oscilloscope to fix anything. The reality is that the same five components fail in the same five ways across nearly every switching supply ever made, and an ESR meter plus a multimeter will find them in minutes.
The second most common wrong belief is that the transistor is always the culprit.
It's almost never the transistor. It's the capacitor. It's always the capacitor.
Daniel, you've got your answer. Roughly two to four hundred dollars to set yourself up, a parts drawer that'll cover most failures, and a skill that transfers to practically everything with a power cord. Discharge the bulk cap, keep one hand in your pocket, and start on something that's already dead. Send us a picture when you bring that third monitor back to life. Thanks to our producer Hilbert Flumingtop. This has been My Weird Prompts. If you enjoyed this episode, leave us a review wherever you get your podcasts. We're back soon with more.
