Apart from that, just fix broken stuff. Practice like any other skill, like others have said.

Like decide what skills to learn based on what’s broken.

Need to solder something together? Buy the soldering iron. Need to figure out which capacitor you need to replace with? Learn how to identify capacitors and navigate digikey.com

* big caps that can contain high voltages even when a device is powered off, in TVs or microwaves.

* Know where a fire extinguisher etc is, keep them near the workbench.

* Keep the workbench clean, tidy and well-vented.

* Don't panic when mistakes are made. Slow down instead of doing a quick-fix.

* I like a big red button to power down everything instantly. Can be just a switch on a plug box.

* an isolation transformer - meaning you must now touch both conductors to get shocked instead of just the live one

* a foot switch to control the power supply - serves as a "dead man's switch" to cut power in case of an accident

(these are not mutually-exclusive, you can combine them)

Risks to consider in general are (also of the top of my head don't take as exhaustive):

* Electrocution

* Burning yourself

* Setting stuff on fire

* Fumes, both from soldering and overheating/burning plastic

There's a kind of balance between habits and awareness. Rely too much on awareness and you will miss some safety issue during a complicated repair. You need good habits, but rely too much on them and you won't notice when you finally make a mistake.

Those apply during repair processes. It's also possible to render a device unsafe to use, for example by damaging a li-ion battery or by a 'bodge' repair that circumvents a safety feature.

Reminds me of the comments I read on a video of somebody doing something unsafe - "safety squint".

A cheap multimeter and a cheap soldering iron/hot air station combo will get you very far in the hobby. I enjoy these channels:

MyMateVince (a guy who fixes many useful household items): https://www.youtube.com/user/mymatevince

StezStixFix: https://www.youtube.com/@StezStixFix

Electronics Repair School (more advanced, he fixes laptops, tvs, and sometimes gpus): https://www.youtube.com/@electronicsrepairschool

When the problem is not exactly the same, I'm just lost. There is not a lot of diagnosis videos on YouTube. All the videos are: "1. I observed this problem. 2. [???] 3. I'll walk you through soldering on the new components." skipping the most important step 2.

Same for car repair videos: "I see Problem X happening. Problem X usually means component Y has failed. Here's how to replace component Y. The end." If that doesn't work, you wasted money on the part and your time ripping apart your car and putting it back together.

In the channels that I suggested, all of them go into the repair not knowing what the fault actually is. They take the viewer through the whole diagnosis, and they (with the exception of Electronics Repair School) are not electronics technicians.

Once a person has seen enough different ways of diagnosing items (by watching videos or hands on trying), then faults in other items become easier to find.

Might not be the case for all the vids of course, but for those i watched i never had that "ah, that's how it works" gotcha moment...

He buys cheap crap, takes it apart, and usually infers a schematic. He also admires or critiques the designs. After a while you'll notice patterns.

1. Take things apart, and try to figure out how they work. Don't worry too much about getting them back together. But then, trying to reassemble it is good practice too. There's plenty of old electronic junk that you can explore before throwing it away. Build a mental catalog of how things work and are assembled. Among other things, this will help you when you need to come up with a strategy for getting some particular gadget open -- often 9/10 of the problem.

2. Like others have said, YouTube is great. People will make a video of a repair, or even just getting something open, for a few "likes." Also, most of them are just normal genuine people who aren't trying to be influencers, so it's kind of culturally refreshing. Many of my successful repairs started with YouTube, such as my clothes washer and dryer, refrigerator, lawn mower.

3. Many of my most cherished tools are what I call for "demolition," not electronics specific, like picks, pry bars, a slim knife that can get between things, stuff like that. Another set of cherished tools are my magnifying visor, small magnifying glass, bright flashlight, and a stereo inspection microscope that I got second-hand.

4. A lot of "electronic" failures are actually mechanical in nature, and you can just use your general troubleshooting rundown to figure them out. Switches, connectors, cables, etc.

5. Begin to get the hang of identifying parts, and how part numbering schemes work. It will help you in your search for spares.

6. These days I often ask my spouse for a opinion. She's a laboratory scientist, quite sharp and skeptical. I'll have a hypothesis, and she asks the one question that blows it out of the water.

Good luck! Don't get electrocuted or burn your house down.

For some reason I got my hands on a lot of old clocks as a child, and after getting the springs out, I never could get them back in. Turns out there's a special lathe you need to use to coil them up tightly enough. Only learnt that when I was about 20.

Beginner

* Get a Miniware TS101 USB C soldering iron, some solder, solder wick, flux

* Get a mediocre multimeter

* Do a little soldering assembly kit to learn how to solder, there are some good NASA manuals and videos to learn techniques and you won't need to do as good a job as they do

* Learn how to identify and replace broken capacitors

* Learn how to replace power and USB connectors

* ^ (a huge proportion of things that are fixable at all are going to be those two)

Intermediate

* Learn about static (ESD) safety, get the equipment and supplies to prevent frying things yourself

* Get an arduino or other low cost prototyping board and learn to program it and do little projects

* Find some broken consumer electronics with guides on https://www.ifixit.com/Guide and try to fix them, these things are often less about "electronics" and more about glue

* Read the parts of Zen and the Art of Motorcycle Maintenance about maintaining and fixing things

* Read https://www.eevblog.com and watch https://www.youtube.com/eevblog for information and inspiration

Advanced

* Get an oscilloscope and bench power supply, learn how to probe and analyze signals on boards

* Figure out yourself what next steps to pursue, you should be able to at this point, or at least to ask more specific questions

I've learned not to fear B+, but to give very healthy respect to anything more than 500 volts. (Only use 1 hand, keep the other behind you, always have a safety partner, etc)

I've learned to hate Silver Mica capacitors. I've learned how to track down the bad ones that cause the "crashing" sound in old radios.

I've learned that in very old electronics, you can let smoke out, and things will still work. I've gotten good at seeing where the smoke comes from.

You'll definitely need to learn to solder. I'd suggest starting with something like an Arduino starter kit to get a sense of how components actually interact.

But remember, if something is dead, you can't really make it worse. (Just be careful not to make yourself worse along the way)

Man, I'd say be very respectful to anything that might be running 120-240V, and be very sure about anything higher than that. Anything <50V can often just be fun tinker whatever assuming you're ok with breaking whatever you're playing with, up to a certain max amperage of your power source. High amperage but low voltage can still cause some serious messups if you don't have the right fuses in place. But I get other voltages are common in Europe.

AC, 120V will 'bite', 240V will make you shaky (remembers id10t maneuver yesterday, first in a decade). And burn you. In damp environments or with corrosion around, much worse things can happen.

That said, even something as simple as surge protector power strip is going to have a nice and fast circuit breaker in it. If you want to get fancier, get a GFCI extension strip from home depot. And if you get an isolation transformer, it'll certainly have a fuse or a breaker.

Probably implied here, but learning to desolder is huge too. And there are tricks to it that didn't come easily to me. When you buy your first iron, I think it's worth getting some flux and solder wick as well. Taking circuits apart cleanly and properly can make repairs so much easier.

Sure, you can desolder with a cigarette lighter and some luck, but you’ll be much happier with the iron.

Start with something like The Art of Electronics by Horowitz and Winfield.

An example of the kinds of EE stuff you don't heave to know to be an effective troubleshooter of basic issues, might be, oh, say, what it means for the poles of a transfer function lie in the left half of the s-plane.

A stereo that loses the right channel unless you push on the board? Broken solder joint.

Visibly exploded caps? Bad caps.

Some basic understanding of electricity and current flow and the high-level concepts will help, but a detailed education is more for building your own than just repairing existing mass-produced setups.

(As an aside the electrical knowledge in a good ham radio study guide would be a great place to start - https://home.arrl.org/action/Store/Product-Details/productId... )

Anyways, I took pictures of the components and described what I was seeing. It walked me through things really well and asked me to do tests and report back. It even told me how to use my specific multimeter after I took a picture of it. I ended up soldering a jumper cable from the console power supply line (not working) to the micro usb power line (working).

It actually works now and really saved me 30 bucks buying a new controller.

I know few people who repaired their washing machines just watching youtube videos. No skills and no knowledge. That’s the best error you can find - series error. All the appliances of the same manufacturer built in the 5 years will predictably fail and replacing 20 cent regulator revives the machine.

There are then random faults, that need deep knowledge and hours of debugging. They’re not economically viable to repair unless the machinery is very expensive. Good example is 400€ transistor replacement in €20k Tektronix probe.

Then there are things one shouldn’t touch - Tesla battery packs and open microwave ovens for example. Enough energy to kill or badly injure the unlucky hobbyists (in German): https://www.kosmo.at/tesla-akku-explodiert-mann-schwer-verle...

There are enough analog electronics to repair what requires deep analog design knowledge. Music instruments and radio equipment might fall into this category.

I would say, the repairs make no sense in the future that comes. With more and more electronics and programmable components the repairs are not economically viable. I am designing a motherboard with MPM54304 PMIC and a microprocessor. Both are programmed and without firmware sources and circuit diagram one will not able to recreate desired functionality. After product release I will organize workshop for my colleagues from service department and they will still come to me to discuss the repairs when some early faulty products will be send back by customers.

80% of the problems are a single part or loose connector, and they’re big enough and common enough that they’re easy to work on and have lots of YouTube help available.

Then you can progress to working on the control boards themselves instead of just swapping them.

If you have a garage and will do this, I highly recommend the harbor freight lift table: https://www.harborfreight.com/500-lb-capacity-hydraulic-tabl...

Once you get into it you can actually make some coin, free washer, $10-50 in parts, sell for $100 when working.

One upside to this approach is that as you improve it can pay for itself by reselling the fixed items.

You usually can combine them to make one working one - no need to even buy parts.

A few weeks trolling Facebook marketplace or Craigslist will often get you some that match.

https://youtube.com/@stezstixfix

The Nintendo Switch Lite is a fun piece of hardware because they're cheap to buy used/broken and there are many opportunities for component-level repair.

Some great written info (specific to the switch) here:

https://www.retrosix.wiki/first-stage-boot-short-checks

https://repair.wiki/w/Nintendo_Switch

I've learned a decent amount with some electronics adjacent hobbies, like 3d printing, diy sim racing stuff, mechanical keyboards. Mainly just copying things other people build. Enough that recently I was able to diagnose a broken transformer in an electronic theater chair power supply.

This channel is awesome:

https://www.youtube.com/@greatscottlab/videos

https://hifiaudio.com/

Their kits come with a helpful pamphlet. I have an old hi-fi with a few annoying component failures I’d like to fix.

I’d say they’re intermediate difficulty. I’m going to start with some soldering practice kits, lightbulb replacements, and pot deep cleans. I’m reasonably confident I can get that done without killing my stereo.

Find some popular piece of electronics with a fatal design flaw that’s trivially fixed with some basic rework that you can flip for $25-$1200. Do this a hundred times. Branch out.

This changes year to year but could be fixing joycons, reflowing bga’s, upgrading soldered memory on newer laptops, doing case swaps on consoles or smartphones, and as others have pointed out, lots of automotive, vintage collectible computer, or hifi work.

A plugin hakko soldering iron, a hot air reflow station, Amtech STIRRI-V3-TF flux, some copper braid, and Kester SAC305 lead free solder will get you 95% of the way there.

The financial component means you churn through hundreds of devices rather than dozens or less if you were to do it as a hobby.

My most recent fix was on a washing machine. After ruling out the simpler issues .. it turned out to be dry solder on the main relay on a rather hairy looking control board. Before this the drain pump had died .. it was amazing technology shipped home by someone on ebay for $20!

Although, I'm not sure you should start writing software by debugging code that doesn't work, or electronics by fixing something broken.

I think assembling some parts to make a basic thing that works is better.

So first, this book, and then find an equivalent of the 70s and 80s Radio Shack (US) or Tandy (EU) electronic breadboard and parts "build anything electronic" kits.

Then build stuff. Not Arduino stuff, more basic. The things those kits suggest making. Understand why they are the parts they are, in the wiring circuits they are.

Then tweak those to "make your own" to see if you know what the parts are and do and how they combine.

THEN start seeing if you can debug what's broken (and maybe tap into the nothing-but-YouTube suggestions here when it gets to specific things to fix, where seeing can be helpful).

RadioShack/Tandy used to have a series of handbooks like the "Getting Started in Electronics" book but focused more narrowly on classes of things you'd want to repair with both theory (how it works) and practice (therefore what to debug). They were blue covers and quite plain, if I recall.

You don't need courses, but readable/absorbable references to theory and practice for beginner tradespersons are actually great.

For troubleshooting power audio electronics Rod Elliott (sound-au.com) has a large number of useful articles both on simple theory (no poles and zeros) and on practical matters. Bob Cordell's and/or Doug Self's books on designing power amplifiers are good for a bit more "practical theory" (how the blocks are arranged, what goes wrong with poor design and/or a too-small budget for protection).

I don't have recommendations for radio frequency because I've not spent time with it. The ARRL (Amateur Radio Relay League) has several books that are well regarded, I believe.

Which may be only one factor, or maybe somebody just placed a diode backwards :\

So there can really be some unsung variables that make you think.

Until your hair falls out . . .

2.Soldering iron, for starting out I suggest spending a little more and getting a Hakko 888 instead of something cheaper.

3. Flux, Leaded solder, Braid.

4. Broken things you want to repair (for me it has been electronic musical instruments).

5. Practice, Patience, and hobby money.

6. Pay for Youtube Premium.

7. Ali Express Account.

It is hard to get a healthy amount of thermal mass with a small iron.

Hot air and tweezers, cheap is fine.

Understand atoms and valence electrons 1st.

Get acquainted with ohms law and…

Tackle series circuits, then parallel then series parallel with resistive loads only 2nd.

If you’re feeling ambitious move on to ac theory capacitors and inductors, transformers and get comfortable converting polar to cartesian and vice versa.

Typically diodes and then transistors next.

Thats pretty much it for the low level.

Find a book that covers this stuff and avoid maxwell and the physics approach unless at some point you want to suffer / get really deep.

Also, flux. Lots of flux and alcohol to clean it up. Flux is the key to good solder joints, that and getting comfortable creating jigs. Whatever you're soldering has to remain still while the solder sets. Let the heat and capillary action do the work.

I'm sure for someone experienced the benefit is marginal but having a bunch of probes and being able to see the waveform feels like cheating for an amateur like me.

They're not that bad. You can get a good entry-level oscilloscope from a company like Rigol or Siglent for ~$300.

A good, fast reacting multimeter is likely better thing to get than a cheap scope.

That said the cheapies might be usable for debugging audio stuff.

I am also very much in the belief system that you should not buy an expensive tool before you are frustrated with a cheap one, most of the time people get into random hobbies and bow out six months later with way too much crap on their hands, and often time tools are more for show than for real world use.

3.5: scrap PCBs for soldering practice that you don't necessarily want to repair.

There are a lot of good videos on Youtube, for one. Louis Rossmann has some good stuff (especially his older stuff), Dave Jones has some good repair videos (intermingled with a LOT of other stuff, though), and there are plenty of other channels dedicated to electronics repairs.

Having said that... you'll need to know at least a bit about electronics (qua electronics) before going terribly far with electronics repair. Much the same way that one could only get so far at debugging and fixing code, if one didn't know something about writing code in the first place.

So... there are, again, videos on Youtube. The Vocademy channel, for one, is a great resource for general electronics theory.

And books. Don't forget books.

I'd suggest, depending on your existing knowledge level, find a couple of books on basic electronics, load up some of those videos, get some components and perfboard and a soldering iron and some basic test equipment (a multimeter and an oscilloscope are a good start) and start building simple circuits. And gradually expand the range of circuits you build to become more complex over time. More or less simultaneously, start watching YT videos on "electronics repair". The thing is, there's a difference between knowing "the theory of electronics" and having the debugging / troubleshooting skills, intuition, judgment, etc. to diagnose faults. The two things are related, but are somewhat orthogonal.

At some point, find used junk non-working electronics, either stuff friends/family will give you, or stuff you dig out of a dumpster, or buy at thrift shops, or buy off of Ebay listed as "not working / for parts only" and start trying to fix things.

So basically... "learn theory / build stuff / fix stuff" in an iterative loop. That's about the best advice I can give.

I reckon it will be valuable in the future that comes.

I agree. That's one reason (not the only reason mind you, but one) that I've spent a modest amount of money over the last year or two, upgrading my electronics lab, in terms of tools, test equipment, etc. I mean, I do this stuff for a hobby, and it's been a nearly life-long thing for me anyway. But more and more recently, I find myself thinking that the ability to repair/hack/build electronic "stuff" will be a skill with serious value. I just wish I had more time to commit to it.

I think it helps if you have a specific niche you want to play in. I stick with the era before integrated circuits became dominant because troubleshooting those is next level from replacing blown caps or whatever. Motivation is generally that I want to hear the thing I’m fiddling with and that there is usually some mechanical part of that gear that makes them interesting.

Get a desolder gun when you get a soldering iron.

Also the TechMoan guy on YouTube is fun as he shows his failures repairing things along with the successes

The most basic core of practical electronics repair is that there are things that should be conductive and things that shouldn't be conductive and you need to figure out how to do either.

Start with simple thing: A heater, a drill, etc. You won't learn much if you take apart complex gadget like smartphone.

Also learn common failure mode: MLCC usually short itself, or old devices usually have dried out capacitor.

For repairing equipment, you dont need high-end equipment from fluke or keysight. a cheap $20 DMM and $30 soldering iron will get you really far.

Dont stick your hand into where you wont stick your D.

As Randy Fromm says, "the things that work the hardest fail the most"

Examples: MLCC (multi-layer ceramic capacitors) over time will often fail short.

Older devices with electrolytic capacitors (the large can-shaped things often situated by where the power input is) have a liquid electrolyte in them that evaporates (boils off?) over time. When it does, they lose their capacitance and the power supply stops being able to supply (good) power.

The point being, when something stops working, check these things first. It's like if your lawnmower dies on you, don't go pulling off the head to look at the piston. Check if there's gas in the tank first.

Finally, a DMM is a digital multimeter. This is your basic tool to measure voltage, current resistance, capacitance, et c.) You can't do much troubleshooting without one.

Hope this helps.

It's usually something stupid.

>It's usually something stupid.

Like trunk mounted car batteries in ICE cars.

Where the one long copper positive wire reaching up to the engine is longer & thinner than it should be, the short ground wire goes to some sheetmetal in the trunk, and that current eventually makes its way to the starter indirectly through a maze of non-copper pathways through the frame of the vehicle.

Then under ideal conditions the starter never sees more than 11 volts :(

Not even the full 11 volts when the lug in the trunk is painted over before connecting the terminal during assembly :\

https://www.youtube.com/@EEVblog

https://www.youtube.com/@NorthridgeFix

https://www.youtube.com/@electronicsrepairschool

https://www.youtube.com/@adriansdigitalbasement

https://www.youtube.com/@bitsundbolts

https://www.youtube.com/@necro_ware

https://www.amazon.com/gp/aw/d/1850106541

https://youtube.com/playlist?list=PL926EC0F1F93C1837&si=U4Jx...

you have to know how to make it before you can break it or repair it.

jk, but dammit, man, sometimes I wish that I had learned practical electronics in my school and college days.

in fact, iirc, there was a book by that exact name, practical electronics, by an author named bernard grob. it was in our school library.

I just googled. the book is still around.

https://www.amazon.in/Grobs-Basic-Electronics-Mitchel-Schult...

but it's very expensive for India these days.

also it my day had good Synergy with software development

>it my day had

that's speech recognition for you. some of the mistakes it makes are even funnier, particularly when a non English speaker is speaking into the mic, like me.

for example it automatically changed the "some of the" to samudra, which is Hindi and Sanskrit for sea or ocean. I've seen many more such errors in that category.

but I have to say that there are huge amounts of improvement in the quality of the recognition over the last some years, since I first started using a phone which had that feature.

Need electronics basics before you can even attempt to debug

Very good bet.

You still have to beat the odds a lot of times beyond that, since some things are just plain fubar and can't be fixed :\

Others which can just barely be fixed often rely on equal parts luck and skill, I'm the first to admit.

I like the idea of learning how to build for a bit, before migrating more seriously toward teardown & repair.

A really good start is to build a few simple hobbyist kits that run on 9 volt batteries.

You may just get blinking LED's or a buzzing speaker but everything is in the kit and you get the experience with individual components.

You can gain a lot that way without even trying to understand theory, but one of the best ways to develop understanding is to go back to a project you have built from the ground up, and then research each of the elements that you have already been successfully working with until you reach full understanding or at least a satisfactory plateau at that point. Simple discrete components are probably best to develop great confidence with, before trying to deeply study the internals of integrated circuit "chips". It can also be a lot more meaningful when it's your own project you are learning basic theory about, or even the most advanced equations. If you concentrate on a limited area, it sure is more within reach than learning everything you need for a degree.

The main "academic" thing that you can not skip is constantly learning how to better read & draw schematics.

You will definitely be more quickly able to "turn something that doesn't work" (never even was working!) into something that works if your success rate is nominal.

Every project like this is helpful as long as you are continuing to learn, Once you have built a simple kit you may later want to revisit the device and construct a "duplicate" from entirely recycled components that you must first remove from scrap circuit boards. If you do the math these can be some of the most affordable components :) Also flexes your un-soldering ability. All without electronic theory, but it's there for the asking any time you want to dig deeper.

You will use all that experience to better repair things when they need it, especially the simpler problems which may not be the most abundant but it's nice to have them under control. Gaining more intuition about electron flow in general through experience will be more rewarding through time.

While escalating your abilities when your ambition moves beyond 9 volts.

More care than needed will get you a lot further than less care.

But yes, if you are talking about replacing BGA packaged high-end chips in very complex devices, there's a threshold were it becomes very difficult. But even then, if we watch some of what the different folks on Youtube accomplish repairing smartphones, laptops, etc., we'll see that some of these guys can achieve some pretty impressive results even without fancy labs and complete schematics, etc. Not everything is repairable of course, but I don't think pursuing this path is exactly "tilting at windmills" either.

...is carrying some baggage there. What you need to learn for, say, a severe economic depression is different from the useful skills to survive a repressive government.

If you look at the things people built in the 1930s, there was a lot of repurposing, make do and mend, etc, to get some basic function restored or working. Barbed wire telephones, that sort of thing. Kind of like the jugaad approach. Whereas surviving repression showed up as home brew radios, mimeograph machines made from cookie tins, etc. For electronics, maybe an understanding of circuits, how to desolder and resue chips, what the popular ones do, etc.? You wont be fixing the finer points of the surface mounted control board of your fancy miele washing machine, but you might at least get the drum to fill, spin, and drain.

Most important equipment is a multimeter and soldering station (with desoldering supplies), good wire cutters and strippers (I like the self-adjusting wire strippers for everything except very large and very small gauge wire), so I'll recommend a couple of additional tips I've learned; first, don't get an auto-ranging multimeter, especially cheap ones are usually not very good. I would also recommend one with a higher "count" like 6000 count, meaning the voltage and resistance measurements ranges go up to 6, meaning you have more range before you have to switch to the next higher range.

Also, I want to really emphasize the importance of good soldering materials and technique - I prefer finer solder because it melts faster and requires less heat. It is easy to destroy some more sensitive items with too much heat. Learn to recognize a cold joint, make sure to use a good flux (I like liquid flux like kester 186). Keep a wet sponge at your station for cleaning the tip, remember to tin the tip, and use a fume extractor. Also, there are a variety of "third hands" you can get, I don't really like the small ones with two alligator clips and a magnifying glass, you can get ones with more clips that can hold more things, but a magnifying glass is really helpful (although it interferes with your depth perception) - I use a gooseneck magnifying glass with a built in light ring to help me solder.

Make sure you have heat shrink tubing and a heat gun. These are great for finishing wires and much more durable than wrapping with electrical tape (although high quality electrical tape like super 33 is much better than the cheap stuff)

Finally, when you're looking at a board, try to identify how the electricity will flow. Learn to identify the ground and vcc traces/planes on a pcb and test your theories with the multimeter (continuity mode on a powered off device is best for this). Build up a mental model of how the device works, and write down a circuit diagram. Seek to learn to identify common circuit patterns and what they are used for - decoupling caps, pull-up/pull-down resistors, ESD protection diodes, debouncing filters, level shifters, voltage dividers, op-amp configurations, serial headers (SPI, UART, JTAG, etc), transistor/mosfet/h-bridges. Learn to identify your ICs and look up the datasheets (I have found that taking a picture with flash with my phone is the best way to read hard-to-read IC labels) - the app notes will usually have a reference circuit with the necessary components, and you can use those to identify which components are related and why they're there. Good luck and have fun!

If you foresee supply chain or financial difficulties where you are, enough that people will need to keep old machines running...

* Question: Are you thinking for keeping your family's device's working, and maybe teaching them? Or for operating a primary business, or side business?

* If you're currently financially comfortable, you might want to quickly learn what tools and supplies you'll need for several(?) years, and see whether they're currently available and inexpensive where you are. (For example, temperature-controlled soldering irons, desoldering device or braid, a multimeter that has all the features you expect to need, misc. screwdrivers, many sizes and types of security bits, smartphone repair spudgers and suction cups, an assortment of various high-quality capacitors of various specs, an assortment of hookup wire, fuses. Some common parts, like mains power cords, are easy to find from unfixable devices.)

* Repairing power tools and small&large appliances, is a bit different than repairing old radios and televisions, is a bit different than new radios and televisions, is a bit different than older computers, is a bit different than newer computers.

* If this includes keeping old computers working, you might want to think about what kinds, and what parts will wear out and be difficult to cannibalize. For example, some laptop models will need their fans replaced eventually, and then maybe their keyboards. Laptops made 15(?) years ago will soon need new backlight tubes and/or inverter boards, or new panels (possibly LED-backlit, plus the electronics to drive them). Most PCs and laptops will need new thermal compound eventually, especially if you do some kinds of work on them. Fortunately, most PC parts can be cannibalized easily.

* Repairing smartphones requires special parts and supplies, and in some cases might be impossible. You might think ahead to which ones you expect to want to repair, watch (and try to youtube-dl) the tutorial videos, and

* For a business refurbishing old laptops, you might want to start saving repair PDFs now, in case the Internet gets fragmented. For Lenovo/IBM ThinkPad laptops, search for "Hardware Maintenance Manual" and the model numbers.

* Try to avoid IoT, and computers and software that depend on some company's servers to keep running. And move to Linux (such as Debian Stable) if you haven't already. WiFi routers should be running OpenWrt for security and longevity. And consider whether you want to focus on a particular ecosystem of smartphone and tablet (iOS, proprietary Android, "alternative firmware" Android) that you expect to be able to source enough devices for and keep working.