I know the title is wierdly structured. I’m typing this on my phone so apologies for any grammar mistakes. (Also I’ll throw in the excuse that it’s 11PM in it as well and I have Linux courses in the morning tomorrow haha.)
I’m actually interested in learning on how to make electronics devices such as tablets, smartphones, IoT devices, etc… Since I’ve worked in the IT industry before, I’ve always had an interest in creating/repairing hardware instead of just swapping parts out. So I decided to start learning how to make electronics with a book I purchased on Amazon, I forgot the whole title but it was something like: Electronics for Inventors and it’s a lot of stuff to sink in so I just read it from time to time.
But I was wondering for those magical wizards of electronics. Where did you guys learn to make electronics?
Thank you in advance,
I’m not a wizard, but I’m really interested in that too, so I bought myself an Arduino. It’s not like a modern computer, but you’d be surprised how many concepts you learn there are still used. For example, if I remember correctly, our GPS chip is connected through I2C and Arduino has that.
To take a step further, you can grab a C.H.I.P. from getchip.com. That’s already a real ARM computer, although low-spec’d. If you visit their forum, there are a lot of helpful people who can help you get started. I already have a C.H.I.P., but don’t have too much time so most of the time it just lays in my drawer. However, I started a project to “hack” it into a GBA case and it’s moving forward in snail speed. I learned a lot about LCDs and interfacing them… Also now I know how keyboard scan codes actually work, and these things can be useful
This takes a lot of time and patience, since you often need to order parts from China and you often get stuck, but it’s also a very good way of learning. Getting stuck and then getting yourself out of the situation is hands-down the most effective way of learning. In addition to that, you can move at your own pace and choose the topics you like, without following a predetermined path in a book.
But if you want to learn the theory fast, sithout getting your hands dirty and stuck from time to time, I think coursera.com or edx.com or other e-learning sites are better than any book.
I think @pauliunas has some great advice for creating Hardware and I would do it like He described.
But repairing Hardware is something that is impossible at this point. No company would every try to figure out what tiny component on for example a graphics card is broken, because that costs way more than buying a new one since there are thousands if not millions of parts on there which could be faulty at some point.
That is true for computers, but he can still repair less complex electronics, for example car radio or kitchen appliances
I think repairing refurbished parts is still being done even in the pc industry, obviously more & more stuff is within ICs & it is the complexity of identifying an issue that may cause it is not worth to repair a piece, but easily identifiable issues with passive elements can be taken care of. In praxis it is quite easy to check if the piece is completely fried or if the main elements correctly react to signals, if so, lets say a mobo can be fixed by replacing some smd resistors or a 10 cent power management chip. Obvious physical issues like damaged power connector or other connectors are the easiest repair that is done. It requires deep knowledge on electronics and serious equipment to reverse engineer to identify an issue yourself as obviously you dont have schematics or a service manual, but at the end of the day, we all would be surprised of what amount of issues can be fixed with a couple of cents of material, instead stuff gets thrown away. Your 5y old 1000$ 50 inch plasma TV may just need replacement of 5$ worth electrolytes in its PSU to work, unfortunately a contract service center may advise you rather to buy a new product or offer you to replace the PSU for 250$, the point is, if you replace, they may take your old PSU, repair it and it gets into their service parts stock…
Yup, most things can still be repaired. Even if they’re in an IC, that IC can be replaced instead of the whole part. But the hard part is identifying the issue, and it’s often not worth the effort. As @Alexander_Halbarth said, there are millions of parts in a computer, and unless something is obviously dead that can be seen by just looking (fried, rusty etc.), you will spend way too much time checking each and every component.
You are right, but as @pauliunas said: if you can see that there is a defective part you can replace the resistor, transistor, plug, whatsoever you can solder it off and replace it
But if you cannot see any damage with your eyes it takes you ages to just find out how the system should react to some signals, and then testing all of this…
Well I am also kind of interested in learning electronics. Soldering circuits and such are not a problem for me, done enough at school when I was young. I still have my extension cord from than, and it has never failed on me. But I want to take it a step further now. I am interested in making an adult size pedal car/cyclecar for daily commutes, one that can legal use bicycle lanes (max 1m width, if electrical assisted than limited to 250W electrical assist without autonomous function). I use my bike now, but on rainy days that is no pleasure. But for those cyclecars, I would like to figure lighting out myself, such as blinkers (in a cyclecar that seems a better option than waving my hand to show where I want to go and that way having possible problems for emergency braking). Just an on-off switch for the front and rear lighting is not a problem, but for blinkers I should find a way to automatically interrupt the signal instead of having to make it blink by using the on-off switch. Also a solution to combine multiple battery’s would be nice. On regular electric bikes they have a range of 60-130km, good enough for daily commute, but if I was planning a road trip I would at least need to combine 2, and then it would be easier if I don’t have to swap one for the other. Also would be nice to make my own dashboard without those modern touchscreen stuff. Just a classic speedometer would be nice. And than maybe I could also find a way for airco in the summer (well in fact just a fan spinning, no cooled air or such fancy things).
That is a common miss-conception, repairing is still done in the industry, if you come across a service manual, you will see that it includes plans and steps to perform, which easily tell you if a piece is worth to pursue for repair, many boards even are designed to include so called service connection points to streamline the process - connections points are clearly marked, you connect lets say a signal generator, oscilloscope etc. you just follow steps by the service manual & root cause, it may be minutes to identify a problem.
It is all about cost/benefit, storing service parts on stock and possibly running into a risk to not have parts available. You obviously cant go to your manufacturer after a couple of years and ask to manufacture 3y old spec boards…
It is of course different when boards are of a standard, if you dont have the original you replace with a newer model that will work - example graca for a desktop, but you cant do so for a notebook mobo or some model specific extension card for your server…
There are companies in the industry that do those tasks within the supply chain process for service, it is many times the same company for different large players, similar as Foxconn is for manufacturing…
That is what may happen if you send your laptop for a repair, they replace the board, but the old one may be tested, repaired and used for a future repair.
You would not be able to identify non IC damage by eye, all is smd nowadays and too working with tiny currents a visual fried piece is not what you see usually as a result of a failure.
Of course we are not talking about integrated cpus, memory etc, repair gets done mostly on the passive elements that each mobo still has 100s of them and that i guess too are statistically probably the primary source of a failure. Passive components are even many times designed in to act as a buffer, for example a large piece of PSUs that seem fried as their output nothing, may just need to have the varistor replaced on their input, 1 component (looks like a ceramic capacitor, is blue) for cents and thats it…
What you want is very easy to do with Arduino with almost no skills to blink lights, you can just write a while loop that turns on the lights, waits n milliseconds, turns them off, waits m seconds and starts again.
Combining multiple batteries is a bit more complex, but if you buy identical Li-Po batteries with identical capacities, you can just connect them in series and it should be OK. Just make sure they are exactly the same and have the same charge level, otherwise you night overcharge one of them and it would go boom.
But to get a blinker to work an arduino is a bit of an Overkill
There are cheap blinker Relais that just blink if they get power which are used in old cars and motorcycles
Thanks for the responses so far everyone. I should of been a bit more clear on the repair part. No way am I going to fix a graphics card with so many layers of PCB on those suckers. It’s more of the basic stuff, like PSU, radios, basic modems, etc. Perhaps one day I’ll try prototyping a new product but for now I rather get the basics down first.
I have never worked with arduino, but I might give it a shot.
For charging I would take them both out of the cyclecar and charge them, I think that would go fasten than trying to charge both with just 1 adapter. It is more the problem how to connect them in parallel without risking some boom. And I probably would want to connect the lights to those batteries instead of having a third individual battery for them. So a lot of things to think about. But first thing is finding the funds for that build
If you charge them separately, you will disrupt the charge balance and it’s not safe to use them together anymore. However, if you bond them permanently and leave it like that, it’s safe. That’s essentially how big batteries are made - if you take apart a laptop battery, you’ll likely find a bunch of separate cells similar to AA batteries soldered together
But that would also double the charge time. Than I think I would prefer to have to switch them out when necessary over having them coupled. But thanks for the advice, it clearly isn’t as easily is I had hoped. I have seen people taking apart laptop batteries and then replace the old cells with new cells, so I have seen those things, but that is something I don’t want risk
If you have a bigger battery (or more small batteries for that matter) you can charge them at higher current, so the charging speed is the same. It’s just that instead of 2 separate chargers you need one powerful charger.
That’s interesting. I’ve built quite a few analog guitar effects by finding schematics and prototyping my own circuit boards, and then building the case to install everything in. It’s fun, but I would really like to get more into the electronic theory part of it to the point where I can actually design my own schematics. And possibly get more into the digital processing side so I’m not limited just to analog effects.
Hint: Buy an electronic flasher module from the automotive parts store It will solve your blinker problems. Coming from a background of home mechanic work and car and mini-bike modding
But Arduino is more fun and it costs $10
Thanks for the advice. I was myself thinking about having a look at motorcycles systems. I guess those are close to automotive parts. Than I would have both a 12V system and a 36V systems (I am thinking about using bafang, and the set I am looking at comes with 36V battery). I could probably also use a 36V to 12V converter, but don’t know if that could screw the electronic regulation, so I should contact bafang about that. But that is for when I get on the project. For the moment I come at 3000€ for parts without the bafang system, the bafang system would easily add another 1000€ and than there are some parts that I still would need to find pricing on, such as internal gear such as pinion or effigear, and ofcourse the lighting and such. So could easily turn out 5000+€, much higher than I expected it to cost, and higher than I can afford. Even when dropping the electric assist it is to costly for me.
But certainly thanks for the advice!