Shopping list for electronics shop

I'm a complete noob in electronics. I have an old self-modified soldering iron, some colophony, some wick, some desoldering pump, that's all. I also have some dead electronics (a display with bulged caps, an old NVidia GeForce 8600GT GPU with bulged caps, a Samsung smart TV motherboard, replaced entirely (I currently connected the TV as a display), a dead USB-sata bridge, a e-book with shattered e-ink display, a dead wireless router (I was using Internet ... and suddenly that router has turned off, after I plugged it off and then on I found out it doesn't boot and its CPU just heats, and that's all, there is no bulged caps on its motherboard or any visibly burnt components) ) I'm thinking I should try to repair it. Because delivery is costly (and prices in online shops are very costly, orders of magnitude more than in physical shops, completely insane) I prefer to visit the shop only once. So I probably need some very cheap but good components. 1. first of all, caps (should I repair the GPU and then sell it, or merely throw it into trash?). 2. usb-uart bridge 3. probably some jtag to revive the router and here are some moar ideas: 4. maybe some sdr? 5. maybe some cheapest possible arduino-like board with some peripherials (the planned use case - to automate greenhouse vent based on internal temperature and speed of wind (when wind becomes strong it has ripped cellular polycarbonate sheets of the metal frame several times and damaged the automatic passive gas spring-based opening gear) got from a separate outdoor sensor)? 6. maybe something like meshtastic in 2 items, but I suspect the whole meshtastic thing is a big scam since without monopoly there will be contention, and with monopoly the whole thing becomes moot, just buy a cell phone. With monopoly or not, the law enforcement already tracks all the signal sources and can and will prosecute meshtastic users for providing network service without a license. Any advice on the the shopping list, including exact items?

>>88099 Get a breadboard and a multimeter (Fluke is the best but very expensive). Make sure that the multimeter has connectivity test. >maybe some cheapest possible arduino-like board I can recommend Arduino.

Already have a cheap multimeter (Mastech). A breadboard is surely a must. About original Arduino ... not sure... And about jtag ... I'm pretty sure Jlink is overpriced... and maybe some universal programator ... I heard of CH341A...

Chinese arduino "copies" are much cheaper, but its nice to pay those who develop open hardware if you can.

Arduino is boring soykaf where everything is already provided.

>Chinese arduino "copies" are much cheaper, but its nice to pay those who develop open hardware if you can. I don't mean Arduino copies at all. I mean boards kinda Arduino, but with preemptive multitasking, some RAM, a lot of gpio, a lot of interfaces (SPI, I2C, UART, USB, JTAG and so on), and maybe some RF capabilities (maybe including hacked bluetooth/wifi chips allowing making them into SDRs), something to which I can flash various open source projects making the device into various without hardware replacement, but the cheapest possible.

>>88137 Not really? At it's core, arduino is just a cheap microcontroller+programmer and a streamlined IDE. You can use any i2c/spi/uart peripheral, or just drive relays or whatever.

>>88138 I see. These are a lot more expensive though. SDR peripherals alone reached near 1000€ the last time I checked (~5 years ago). I don't know what you want to make, but preemptive multitasking is generally not needed, unless you're running MicroEJ or some bloat like that. A small RTOS and interrupt routines will take you farther on a cheaper chip. Perhaps an FPGA board would be suitable. They've gotten much cheaper in the last few years, and hardware design is fun.

>SDR peripherals alone reached near 1000€ the last time I checked I don't mean the professional market segment intentionally sold as "we sell SDRs, bribed FCC, buy it from us at premium prices". I mean ones that intentionally ship with already hacked wireless chips (see nexmon, internalblue, and so on) that can work as SDRs. Basically I need a cheapest possible (since i don't know how seriously I'm going to do it at all, possibly I don't need this board at all, possibly I'm incapable to fix anything because I'm a moron) devboard with loads of interfaces since I don't even know what I may want to do with this board in future. I still suspec everything will be wasted money and that I shouldn't touch electronics at all since it is costly and since I'm moron that during the last year has only vibecoded Python apps (those ones that I wouldn't have even touching it there was no vibecoding at all, I have a very long backlog of apps I need for many years, and, unfortunately, that backlog constantly grows) without any success.

>>88139 i meant software libraries. arduino is good for toddlers (not offensive!), but when you're going balls deep (seek employment) you'll need something less basic. it's good for example to work with stm32 because cubeide is brainrotten and hal libraries were written by demented retards.

Isn't platformio the proper hal?

>prices in online shops are very costly, orders of magnitude more than in physical shops It was always completely opposite to me, local electronics shops always were super expensive and had very little to offer compared to online stores. Every time I need something I just write it down and place the order when I collect enough components to qualify for free delivery. >I prefer to visit the shop only once Not possible in my experience. You will always need something you don't have, you cannot just buy enough stuff in advance. It's pain in the ass, especially if your country does not have any major distributor. Every time I need something, I just write it down and place a big order every two months or so when I need enough components to qualify for free delivery.

Maybe "not possible", but I still must to try.

>>88099 EEVBlog has done a couple of setup videos >https://yewtu.be/watch?v=HicV3Z6XLFA I think this is cool > Analog Discovery 3 One tiny box has a Logic Analyzer, Spectrometer, Signal Generator, Oscilloscope + more. Used for several college courses as a all in one lab setup, but limited to 30Mhz bandwidth, which you could compare to entry level oscopes at 100Mhz.

>>88362 Thank you for the video, I have only managed to find time to watch it. Too hypousy and exacerbated. And the problem is: too expensive and too first world, and too promotinal. I cannot spend a penny on the soykaf I don't need. To be honest I have the following set of beliefs (they can be wrong) 1. if the problem is hardware one, you cannot fix it, period. Well, you can probably, if you spend millions of USD and few years of own lifespan working full-time on that. I have experience of "fixing" USB-UART bridge that has started turning off after few years of continuos operation paired with a router (the one that is dead by now). I have suspected that HDD vibration caused problems with contacts. So I have resoldered all the connectors - zero success. Then I resoldered the JMicron chip ... no success, and alignment issues this time. I resoldered it again ... no success, and alignment issues again, and what worse - a trace delamination. Replaced the delaminated trace with a piece of small insulated wire ... but the trace that has delaminated was not the one with misalignment. Then I tried to fix alignment issues by applying some small force to the pin that mismatches the trace. And that pin then has fallen out. As knowing guys on forums have said, just JMicron chips are soykaf that overheat and then fail. So it seems if electronics fail, you cannot fix it, the cheapest way is to just buy a brand new one that is programed to fail after some time. Please note: all this soldering was done on my former employer's professional-grade equipment. 2. I absolutely hate analog electronics, it is non-local, one cannot analyze it element-by-element, the only way to reason about it is to convert it to a system of differential equations, and you cannot in principle design a circuit, it is an inverse problem, you cannot even predict behavior of a simple circuit like a Chua circuit or even a complex circuit with specified behavior like a CPU... Prove me wrong. I start suspecting that I should just throw all the e-junk I have collected in the hope to hack it somewhen into trash. It is all intentionally undocumented and complex just to guard company's market positions. One of my old moronic ideas I want to implement but I'm Dunning-Kruger-palgued moron, so everything is impossibly complex if you ever try to touch it: 1. try to fix the 8/32 ramips wireless router 2. resolder RAM and flash to allow modern OpenWRT bloatware to work (fuarrrk this "security", this router used to be running all a torrent client, a samba share, a web interface, a print server, a VPN server, a VPN client and a wireless client and a lot of things simultaneously, now it cannot even run simple OpenWRT without webui or anything else including WPS, likely these assholes have enabled some "security" feature like some sort of CFI that causes large bloat) 3. fix a e-ink reader by replacing its display 4. port coreboot to my mainboard 5. somehow figure out how to make an FPGA-based adapter allowing to plug DDR5 memory into DDR2 mainboard because any required amount of DDR2 memory is golden. Memory is always golden: everybody know the software will bloat to the point you will have to throw your system into trash because of lack of RAM, and that eventually that version of DDR will stop being manufactured, that will make that memory not merely golden, but antiquity, so everybody buys as much RAM as he can afford at all costs. So "enormous" amount of DDR2 can be replaced with "very modest" amount of DDR 5, if you can figure out how to do that. 6. I have a PCB from a broken monitor. Figure out how the big chip there transforms DVI into signals into signals for muxes and pixels. I hate hardware tinkering: it is costly, I cannot afford doing it. Unfortunately I cannot afford not doing it either.