On the fine art (and profit) of electronics repair/scrapping

[Syonyk]

Does anyone else regularly repair electronics, either for themselves or for other people?

I've been doing this for years, and have recently "upped my game" a bit, with a nice soldering station and a (cheap) lab grade PSU for doing work and repair.  I've already got a vintage oscilloscope that works for me, though I rarely need it.

Over the years, I've learned that almost nobody works on electronics in the US, at any level.  If people use anyone, it's the local "cell phone repair shop" - but people don't even use those!

This leads to a very strong "Well, it's broken, so it's not worth anything. :(" mentality - which is entirely wrong!  In many cases, it's easy to repair a broken thing (either properly fixing it, or replacing the bad part), to work around the failure (I can't count how many laptops I've owned with "creative" power jacks), or, at worst, to sell off the parts for a tidy profit.

I've been doing this for years, and I genuinely have no idea how many thousands of dollars ahead I am because of it.  People will either pay me good sums of money (or whiskey - prefer that, lately) to fix a broken laptop/cell phone, or are willing to flat out give me a broken device because it doesn't work!  Most of the time, it's easy to get it running again with a few parts or repairs, and if not, I can part it out and make good money on that side of things.  A laptop keyboard is often worth as much as a full broken laptop, and a full screen assembly has often fed me for weeks!

Last night, I swapped out the micro USB port on a Nexus 7, and despite things going badly wrong, I was successful in the end - my tablet now charges off USB just as well as with Qi wireless charging.  I've got a $1500+ ebike that I effectively got for $300 ($600 counting tools), because I was willing to dive in and rebuild the battery pack.  I've had many, many laptops that I ran over the years I got cheap or free because they were broken, and I've made good money fixing stuff for people as well.

Does anyone else do this?  It's a really, really neat way to get fun technology toys for nearly nothing!

[Glenstache]

I have not purposefully used it as any type of side hustle, but have kept many of my own electronic devices going by fixing them myself. I've only been totally stumped once, and that was on a guitar amp that likely had a small crack in a circuit board somewhere that I couldn't locate; I was able to sell it on CL to someone more experienced than myself who was looking for a project. It's really fun and satisfying. Usually there is nothing to lose by failing other than a little time, so just going in and trying is totally worth it.

[onemorebike]

I'm now following this as just yesterday on was considering picking up a tablet/phone repair kit to fix the third device in our house to have its screen go down  in a year. This would be an additional justification for doing it,  maybe a side hustle as a device repair man. Any idea how much could be brought in doing this?

[Syonyk]

I really have no idea, but the prevalence of broken cell phones/tablets says "A good bit, just not that much per device."

It depends on how deep you want to go.  A couple hundred in setup costs (soldering station/hot air rework station, possibly a separate heat gun, some geekdrivers/spudgers/etc) will let you do pretty much anything, and then it comes down to how much profit you can make.

If you look at the prices of broken items on eBay, it's not going to be profitable to work there - the broken things almost always go for "value of a new one minus the value of the parts needed."  So it's something to do at a more local scale.

But battery replacements, screen replacements, and micro USB ports are decently easy to do once you learn how, and those are the common failure points for phones/tablets.  Laptops are a bit more varied, and it depends on if you want to go down the road of a full service IT shop, or just physical repairs.  I don't do IT work on laptops because I don't want to, but I typically charge about $100 for a deep physical repair (fan replacement, resoldering power jacks, etc), and at least in my neck of the woods, people are willing to pay that on expensive laptops.

It does require knowing what you're doing and being willing to cover screwups, so learning on stuff that's free to you or cheap to you is a good idea.

I mostly do it because I enjoy it, and the profit is a side benefit.  But, I will say, I've been stocking Nexus 5 batteries, and I'm out of stock right now due to the popularity of that particular repair at work. :)

[vhalros]

I have a bricked Samsung Galaxy S4 that I have contemplated repairing. I think it is a faulty power switch, and that I would have to do the repair shown in this video: https://www.youtube.com/watch?v=2l_6fwzuIwA.  Any idea much practice it would take to become proficient enough to do that, starting with never having held a sodering iron?

[redattack34]

This is slightly off topic, but this seems like a good place to ask.

I have a bunch of computers in varying states of old and/or busted - some as much as 10-15 years old. I had been using the ones that worked for grid computing with BOINC until I realized that their scientific output per watt was terrible, so all I was really accomplishing is wasting power.

So my questions are:

-How old can computer parts get before there's no reasonable hope of selling them and it'd be better to recycle them instead?
-For the newer stuff, which websites would you recommend for selling them? Kijiji/Craigslist, eBay, or are there more specialized sites that would be better? (I'm new to selling things online, if that matters)
-I have one which is only a few years old, but doesn't work. Would I be better off fixing it and then selling it as a working machine, or selling the parts that do work? I think it's either the CPU or the motherboard that broke, but I'm not sure how to diagnose that except by replacing them and seeing if that fixes it, so any suggestions there would help as well.

[Syonyk]

I have a bricked Samsung Galaxy S4 that I have contemplated repairing. I think it is a faulty power switch, and that I would have to do the repair shown in this video: https://www.youtube.com/watch?v=2l_6fwzuIwA.  Any idea much practice it would take to become proficient enough to do that, starting with never having held a sodering iron?

How proficient are you in the rest of small electronics repair?  That's a very complex repair, as they go - I'd certainly do something along those lines, but I've been doing this for a long while.  I still wouldn't provide any guarantee of success, but it's something I'd be pretty confident taking on.  I think a hot air rework station would be nicer for that, though.

If you're good with the rest of it, spend a few hours playing with scrap circuit boards and a soldering iron and you should be fine.  If you haven't done anything like that, assume you'll fail entirely, and have fun. :)  The RoHS lead-free solder is kind of a pain to deal with. :(

-How old can computer parts get before there's no reasonable hope of selling them and it'd be better to recycle them instead?

Past 8-10 years, they're probably not worth much.  Or if they're an "obsolete" technology - AGP graphics cards, for instance.

-For the newer stuff, which websites would you recommend for selling them? Kijiji/Craigslist, eBay, or are there more specialized sites that would be better? (I'm new to selling things online, if that matters)

eBay for parts, Craigslist for full machines (shipping is annoying).

-I have one which is only a few years old, but doesn't work. Would I be better off fixing it and then selling it as a working machine, or selling the parts that do work? I think it's either the CPU or the motherboard that broke, but I'm not sure how to diagnose that except by replacing them and seeing if that fixes it, so any suggestions there would help as well.

If you don't have the parts laying around to test it, just sell it as-is.

[flyhyr]

I frequently repair my own family's electronic items so we don't have to buy new stuff.  Generally though, I don't fix things for other people because I don't want to become their defacto tech support.

Some of the things I've fixed recently:
1. A co-worker was giving away a 47" flatscreen TV because it was broken and I fixed it for $40 in discrete parts
2. Antilock brake module on my truck - no parts, just reflowed cracked solder joints
3. Automatic gate opener - replaced bad optical counter
4. Furnace/AC system - I wish I had time to debug down to the component level but ended up just getting a new mainboard.  My wife wasn't willing to live in a 40 degree house in the middle of winter while I debugged the board :)
5. Various electronic kids toys; not to mention I modify (reduce) the speaker volume of every new toy that comes into our house.

Most electronic items fail because a single component failed, and it will occasionally cause some downstream failures as well, but if you can troubleshoot to the root cause, then discrete components are usually cheap and easy to replace. 

[The Money Monk]

Most electronic items fail because a single component failed, and it will occasionally cause some downstream failures as well, but if you can troubleshoot to the root cause, then discrete components are usually cheap and easy to replace.

My question is how do you do the diagnostics on items like that? I am totally confident in my ability to learn to replace parts and have done some soldering before, but I have no idea how to figure out what parts need replacing.

Like with the garage opener for example, how did you determine that was the part that needed replacing?

[fiftyincher]

Lots of times you can search Google and someone has had the problem and a solution. I've replaced a few capacitors on various electronic items. For a couple bucks in parts, things are as good as new.

The right attitude to have - it's already broke or not working properly, what's the worst you could do?

[Syonyk]

That's always been my view. It's broken. The worst thing that can happen is that I spend some time and it's still broken. Usually it's much better. :)

[flyhyr]

Most electronic items fail because a single component failed, and it will occasionally cause some downstream failures as well, but if you can troubleshoot to the root cause, then discrete components are usually cheap and easy to replace.

My question is how do you do the diagnostics on items like that? I am totally confident in my ability to learn to replace parts and have done some soldering before, but I have no idea how to figure out what parts need replacing.

Like with the garage opener for example, how did you determine that was the part that needed replacing?

There are a couple of different troubleshooting techniques you can use:
1. Physical evidence - On the TV repair, I took the back off and after some careful looking, I saw a burned area under an inductor and a bulging capacitor next to it.  Further downstream, the control chips for the backlight master/slave boards showed a burn spot on the package near the VCC input pin (too much voltage due to insufficient pwr supply filtering due to bad capacitors) and some additional bulging caps.  I replaced the caps and control chips and it worked.

2. Discrete component functional tests - With a multimeter you can probe around to see if components are working as they should.  i.e. if current is flowing through a resistor, then there should be a corresponding voltage across the resistor.  you have to be careful because any component soldered down isn't necessarily testable by itself.  It's behavior will be influenced by other components in the circuit (ie. 2 equal value resistors wired in parallel will only show half the resistance when connected to a multimeter).   Power regulators are a notable exception.  A 3.3v regulator should always output 3.3v, assuming correct input voltage.

3. Functional block analysis - This is what I used with the gate opener (not a garage door opener, more of an articulating arm that opens and closes a driveway gate).  I knew that the main board must have a way of knowing where the arm was during it's motion.  There were a couple of limit switches on both the min and max extension that checked out OK with the multimeter.  After that, there was an optical counter wheel that would feed back pulses to the main board.  Since the gate would start to move, but would then hesitate and stop, I suspected the optical counter wheel was the issue.  I didn't have the portable test equipment available to verify the pulses so I made an educated guess that a faulty counter would cause the issue and just replaced it and it ended up working.

[The Money Monk]

Most electronic items fail because a single component failed, and it will occasionally cause some downstream failures as well, but if you can troubleshoot to the root cause, then discrete components are usually cheap and easy to replace.

My question is how do you do the diagnostics on items like that? I am totally confident in my ability to learn to replace parts and have done some soldering before, but I have no idea how to figure out what parts need replacing.

Like with the garage opener for example, how did you determine that was the part that needed replacing?

There are a couple of different troubleshooting techniques you can use:
1. Physical evidence - On the TV repair, I took the back off and after some careful looking, I saw a burned area under an inductor and a bulging capacitor next to it.  Further downstream, the control chips for the backlight master/slave boards showed a burn spot on the package near the VCC input pin (too much voltage due to insufficient pwr supply filtering due to bad capacitors) and some additional bulging caps.  I replaced the caps and control chips and it worked.

2. Discrete component functional tests - With a multimeter you can probe around to see if components are working as they should.  i.e. if current is flowing through a resistor, then there should be a corresponding voltage across the resistor.  you have to be careful because any component soldered down isn't necessarily testable by itself.  It's behavior will be influenced by other components in the circuit (ie. 2 equal value resistors wired in parallel will only show half the resistance when connected to a multimeter).   Power regulators are a notable exception.  A 3.3v regulator should always output 3.3v, assuming correct input voltage.

3. Functional block analysis - This is what I used with the gate opener (not a garage door opener, more of an articulating arm that opens and closes a driveway gate).  I knew that the main board must have a way of knowing where the arm was during it's motion.  There were a couple of limit switches on both the min and max extension that checked out OK with the multimeter.  After that, there was an optical counter wheel that would feed back pulses to the main board.  Since the gate would start to move, but would then hesitate and stop, I suspected the optical counter wheel was the issue.  I didn't have the portable test equipment available to verify the pulses so I made an educated guess that a faulty counter would cause the issue and just replaced it and it ended up working.

Cool, thanks for the info.