File: buffer.png (61 KB, 1304x868)

61 KB PNG

I wanted to try to buffer an op-amp with a transistor, but the output voltage was always 0.7 V lower than the input. Not so if I connect the inverting input directly to the load.

>>2855858

That's how you need to do it. In a buffer configuration, an opamp will always try to keep the inputs at equal voltage. In other words, the inverting input needs to be connected to the point you want to be at the same voltage as the input you're buffering.

>>2855858
>inverting input directly to the load

this has been standard practice since the 1950's
any modifications to the output can be reversed by choosing an ulterior feedback point

>>2855858
Congratulations, you've figured out how to make non-trivial feedback loops using op-amps. For further learning, consider:
>using an op-amp to control a MOSFET voltage follower
>using a precision op-amp to control a high-power op-amp
>using an op-amp to control an inverting common-emitter amplifier
>using an op-amp to provide constant-current through a load via a pass-transistor
Have fun, but keep in mind the limitations of the op-amp, where they fall-short.

File: opamp vs npn buffers.png (55 KB, 1356x605)

55 KB PNG

At first I was hoping to use just a smol single-channel op-amp (LM321) and buffer its output with a simple NPN transistor circuit. But the lack of feedback and loss of output voltage made it a bit unusable for me. I wasn't experienced enough to design this circuit with discrete transistors. I switched to a dual op-amp (LM2904), so I could use the second op-amp to overcome the 0.7V buffering loss.

File: Capture.png (105 KB, 1127x472)

105 KB PNG

does anyone make a capacitor discharging adapter that fits standard DMM probes? like, a resistor between the positive and negative leads. fluke made a massively overpriced adapter years ago for """"ghost voltages"""" but it's been discontinued.
>make it yourself
i'm too lazy.

>>2855896
im sorry, can you explain exactly why the top circuit wont work for you?

File: IMG_E3605.jpg (1018 KB, 1456x1868)

1018 KB JPG

Hey all, very clueless Dad here trying to help out my toddler.

I got my son one of these cheap ass Chinese digital cameras awhile back and of course the lithium battery shit itself within a few months. I opened it up and found that it's a LITER 3.7V 600mAH battery, so I ordered a replacement off of amazon but it came with a connecter that's too big for the board.

In my infinite wisdom, I snipped the cords on both batteries and tried using two crimpers to attach the old connector to the new battery (I've never soldered before in my life). That actually did work for about a day until the negative cable came loose.

I'd like to just order a replacement battery with the right connector, but I've been looking at lists and diagrams online for an hour and cannot for the life of me figure out which one this is.

From what I can tell it's definitely a type of JST, but it's not a PH 2.0mm connector, that one is definitely too big. From this site I believe it's a ZHR 1.5mm connector? I can't tell for sure: https://www.lithium-polymer-battery.net/select-battery-connector/

I've got a couple pictures here, is anyone able to identify the exact connector? And if so, does anyone know where I can find a battery with one? All the ones I see on amazon are PH.

File: IMG_E3606.jpg (2 MB, 2557x1977)

2 MB JPG

>>2856003
Pic of the board where it's attached

https://www.youtube.com/shorts/OFZsz_QbV5o

can someone tell me what this nigga doin? it looks like some kind of flame thrower with a taser circuit to ignite alcohol maybe

File: cat-looks-inside-cat.png (79 KB, 498x337)

79 KB PNG

>trying to design FM-band radio front-end LNA for fun
>design meets spec
>simulate with 50 ohm load
>gain collapses
>even with CC input and output buffer

>>2856012
it would seem trying to impedance match by designing the input impedance of the buffer to be 50 ohms is retarded, and that I should just be using a transformer instead

File: dfn-osc1-singleopamp_schem-1.png (70 KB, 1014x665)

70 KB PNG

I hope this is just a simple misunderstanding but I can't get my head around the concept of "virtual ground".
In this circuit, the op amp is generating an output between 0 and 5 v, centered around the virtual ground of 2.5 V. But the speaker is connected to the 0 V ground, not the virtual ground. Doesn't that mean there is a DC bias to the speaker? Why doesn't C2 just charge up and then stay charged, blocking the signal?

>>2856060
>DC bias to the speaker?

C2 removes all biases, prejudices and ancient curses

>C2 just charge up and then stay charged

left side of C2 is jumping around, causing charges to discharge, and vice versa

>>2856060
In this circuit the speaker is biased above 0V, so it will always be dissipating power, even if there's no signal

File: file.png (286 KB, 915x1006)

286 KB PNG

tiny pot

>>2856060
you have to make a mental distinction between DC signals and AC signals in your circuit. capacitors appear as a very large resistance to DC signals (infinite, ideally), and as a small resistance to AC signals (inversely proportional to the frequency, so doubling a frequency will halve the resistance; the exact value is 1 / 2 * pi * frequency * capacitance).
looking into the node labelled "virtual ground" in your circuit, the bottom resistor R7 is in parallel with a capacitor. to DC, the bottom branch is R7 in parallel with a very large resistance, so the total resistance is approximately R7. to AC, the bottom branch is R7 in parallel with a very small resistance, so that node almost appears as a short to ground for AC signals. thats what is meant by virtual ground.
you say the output from the op amp is centered around 2.5V, so you can think of the output as a DC signal of 2.5v plus an AC signal centered at 0v. the capacitor C2 blocks the 2.5V, but allows the AC signal to pass through.
also, you cant drive a speaker with an op amp, they cant provide enough current.
>>2856088
>In this circuit the speaker is biased above 0V
lol wut?

File: only 20 cycles.gif (60 KB, 1237x171)

60 KB GIF

>>2856103
>tiny pot

tiny and fragile
dies after only 20 turns
poor celine

>>2856040
Or just follow the datasheet recommendations.

>>2856060
Could make that circuit simpler by tying the non-inverting input directly to a 50/50 voltage divider instead of having R3 and the virtual ground circuit, and just tying C1 to ground or Vcc or whatever. It doesn’t matter what rail you tie C1 to, because capacitors block DC.

Virtual grounds are useful when you have more than one circuit that needs to be referenced to a solid intermediate voltage rail, but this isn’t such a case.

>>2856088
Even if C2 weren’t there, the speaker is drawn as a piezo disc, and they block DC too. You might get a bit of distortion without C2, but it’s just a square wave buzzer.

>>2856103
Those cross drive slots are meant for flat-heads, right? Tiny Phillips drivers are too pointy.

>>2856103
>20 rotation cycle life

I know it's a trimpot, but lol

>>2856103
whats the advantage of making them this small? RF/microwave?

>>2855858
Who could have known that all pn junctions consume 0.7 volts?

>>2856166
>all pn junctions consume 0.7 volts
damn, i guess visible light LEDs will forever remain a pipe dream

Thoughts on CH340? I found mostly very old posts on the web complaining about some issues compared to FTDI which "just works". I bought several of them for like 40c a piece, and so far so good but I have only used them a little. If they prove to be reliable, I can then embed them on each board and use a USB connector instead of plugging and unplugging an FT232 breakout board.

>>2856212
Oh and also I got ch340N which has a built-in crystal so I realized that it may have been a mistake since an external crystal probably has much better accuracy?

>>2856212
I use them myself and I’ve never had an issue, though A I’m not using windows and so don’t have to worry about drivers, and B I never had to do anything timing critical. Apparently the best method is to use a native USB MCU programmed to be a USB serial bridge, as that lets you program it to not require a driver at all.

FTDI chips are expensive and have features you don’t really need, there’s a middle ground chip that I forgot about.

>>2856214
I really liked the look of the N model, until an anon on /mcg/ told me the potential downsides of clock drift. Still might get some for non-critical stuff.

>>2856219
I checked LCSC and CP2102/CP2104 cost about $2, actually a bit more expensive than the cheapest FT232 which is about $1.70. And CP2102/4 are in a QFN package that is hard to hand solder.

I am not doing anything timing critical either. I am going to use them on a coin battery powered board, vey low current, low clock frequency and therefore very low baud rate. My budget is about 30uA to push some quick data and go back to sleep at 0.8uA.

>>2856219
>I really liked the look of the N model, until an anon on /mcg/ told me the potential downsides of clock drift. Still might get some for non-critical stuff.
Apparently CH340 are really bad at RX (about 2%) regardless if you use an xtal or not or at least that's my interpretation of the datasheet. TX is better.

From the DS:
The baud rate error of CH340 UART reception allows not less than 2%, the baud rate error of
CH340G/CH340T/CH340R UART transmission is less than 0.3%, less than 1% for CH340C/CH340E/
CH340B.

CH340N is not even mentioned.

File: Datasheet version 3B.png (47 KB, 1480x164)

47 KB PNG

>>2856227
For sleep modes you can always use a small-signal MOSFET to turn off the peripherals. Well, you can't do that to the CH340N if you're relying on USB serial for your interrupts.

>>2856229
Some datasheets don't talk about all the units, I have two different ones downloaded for that reason, pic related. The C/E/B don't have XI/XO pins, so it stands to reason that the N shares the same specs.

The R sounds interesting, as it can do bidirectional infrared comms.

Some of their other chips like the CH341 and CH343 might also be promising, they also do CH55x low-cost native USB 8-bit MCUs, and the CH32Vx RISC-V MCUs.

>>2856002
The top one works well. At first I was going to use the lower one, but its output voltage is about 0.7V less than its input.

>>2856212
>Thoughts on CH340?
They'll work if that's what you have, but I would strongly suggest choosing the CH343 instead. The two biggest differences are that the CH343 works with the USB CDC-ACM class driver, so you plug it in and it just works no matter what OS you're dealing with, and the chips have serial numbers, so you can actually have persistent device names. With the CH340, on Linux you have to use the /dev/serial/by-path/ links (and that only works if the device is plugged into the same USB port each time), and on Windows I think you're just shit out of luck.

>>2856227
>I checked LCSC and CP2102/CP2104 cost about $2, actually a bit more expensive than the cheapest FT232 which is about $1.70. And CP2102/4 are in a QFN package that is hard to hand solder.
There are cheaper chips in the CH910x series, like the CH9101N in a SOP-8 package for $0.88 at qty. 1. WCH/Qinheng has a comparison table on their website that might be helpful:
https://special.wch.cn/en/produce/#/

File: ch340n.jpg (47 KB, 762x309)

47 KB JPG

Confused about D6 and D7. The ds recommends adding them when using separate power supplies. But none of the breakout boards have them. And those breakout boards don't really mandate how you power your uC. They provide 3.3V output or it could be powered separately. And no diodes.
I don't even understand how these diodes even work, wouldn't they just block the signal entirely? For example, TX on the CH340 side is pushing data to RX on the uC side, but there is D7 in a reverse direction between them?? I probably got polarity all wrong? I know idle is high.

>>2855858
how does that work

File: 20240930_205004.jpg (3.18 MB, 4032x3024)

3.18 MB JPG

Hello, I have an electric blankie I got from Aliexpress that I wanted to change the plug from EU to UK. Opened it up to find two smooth white wires. Tested both on the multimeter and they both gave the same reading 0.07 on 20V. What am I doing wrong? How do I find which wire is hot and which is neutral?

I asked my dad and he said it doesn't matter if there are only two wires and they don't have colours/ribbing, he said I can just wire them whichever way I want.

What can I do?

>>2856427
you didn't probe it for voltage before cutting it up?

File: 20240930_023306.jpg (2.08 MB, 2944x2208)

2.08 MB JPG

>>2856428
I cut it fully expecting it to be colour coded like any normal electrical product. Can you really blame me?

File: file.png (340 KB, 1531x1309)

340 KB PNG

well

>>2856427
The EU 2-pin plug is non-polar, so if you're wiring up a UK plug it doesn't matter which is which.

I'd recommend a 3-amp fuse instead of a 13-amp for that thin wire.

File: 20240930_221133.jpg (3.39 MB, 4032x3024)

3.39 MB JPG

>>2856427
>>2856428
>>2856430
I gambled. Blankie still works. Will just have to unplug it at source when it's all heated up so I don't set anything on fire. It's all good.

>>2856444
Checked. That's very reassuring, thank you. I'll buy some 3-amps later.

File: nichrome loop.jpg (18 KB, 680x680)

18 KB JPG

>>2856427
>which wire is hot and which is neutral?

question is silly
electric blanket is just a loop of resistive wire in a plastic sleeve
if you wire it one way, the live will be towards the left and neutral on the right
when the flip the blanket over, now live is on the right
there's no difference; no polarity is safer than the other
(however, if you have wifey, get a contactless voltage tester to make sure he/she stays on the bad side)

>>2856362
It buffers the op-amp output to drive a low resistance load (1 kΩ).

>>2856451
>there's no difference; no polarity is safer than the other
One side has the fuse and that should be where the live goes through, but yeah I guess you're right since the original plug wasn't fused. As a toxo catfag I will be sure to sleep on the danger side and keep my kitter safe.

>>2856262
yea, thats how BJTs work. that circuit is called an emitter-follower (aka common-collector); it has a gain of (almost) 1 and a DC offset somewhere around -600mV to -800mV.
consider reading chapter 2 of Horowitz's The Art of Electronics.

>>2856460
>keep my kitter safe

i can see the headlines
''starving kitten forced to eat owner's face off after electric blanket mishap''

>>2856345
I can only guess there are hidden internal pullup/down resistors.

>>2856427
I wouldn’t risk a name-brand electric blanket, let alone an AliExpress one. Take a piece of the fabric off to see what it does when heated with a flame, if it easily melts or combusts you’re fucked. Put a smoke alarm in your bedroom if you don’t have one already.

>>2855842
is it possible to retrofit a (tft) lcd panel from cold cathode lamps to led strips?
scavenging my shak I have found a couple of samsung monitors working but they use the lamps.

Also
what defines the resolution of a lcd screen?
is it a physic factor? or it's programmed into the device driver?
And is there a "one size fits all" method to use any lcd display with any source.
I mean a way to interconnect edp, dsi and all that? I mean the interfaces using fpc cable connectors?

Where can I consult more about the lcd display designs, the power sources, the panels, the ilumination and all that. I really like the toy.
Thanks

>>2856546
>cold cathode lamps to led strips?

absolutely
90% chance you'll damage it beyond repair in the process, tho
and not worth it now that LCDs at the thrift store have returned to single digit prices after the covid inflation

>what defines the resolution of a lcd screen?

resolution = 1 pixel for every 3 RGB cells

>>2856546
My local tv repairman said that he tried but the leds are too powerful and don't diffuse well over the picture.

>>2856614
The best methods use a lot of high-density low-brightness LED strips, but those are kinda expensive. You may also want to find LEDs that have a certain emission pattern. Alternatively you can use multiple layers of diffusors to spread the light out better, or even a fresnel lens array that collimates the light from each LED. I'm not sure how edge-lit LED backlights are constructed, maybe they'd be doable?

>>2856003
>>2856004
If you look to the left of the battery connector, there is 2 exposed pads, where you can solder the plain wires the one closer tot he PCB edge is the negative, the other one above it is the positive.. You don't need a replacement connector and even if you get one, I personally never managed to crimp these without damaging them. Solder that shit to the board and tape it down, so when the kid drops it, the pads won't get ripped off.

>>2855842
>>Principles (by increasing skill level):
>Mims III, Getting Started in Electronics
>Geier, How to Diagnose & Fix Everything Electronic
>Kybett & Boysen, All New Electronics Self-Teaching Guide
>Scherz & Monk, Practical Electronics for Inventors
>Horowitz and Hill, The Art of Electronics

Where would the wikibooks Electronics textbook from the github resources links sit on this list?

>>2856724
>Where

consider the wiki book has 3 pages on diodes
compare with Boylestad & Nashelsky - Electronic Devices and Circuit Theory (7th ed.) 1998-08-14
with 111 pages on diodes
and you quickly realize you're being short-changed thousands of mind-numbing formulas and mind-killing minutiae

File: decrapped.jpg (175 KB, 1200x900)

175 KB JPG

I love electronics, but never went to college. I landed my first tech job repairing digital cameras. I've repaired cameras for 12 years and fixed about 20,000 of them.

I want to start doing more design of electronics, and I am a huge supporter of free and open source software and hardware. So I think designing a free and open source camera makes a lot of sense. Theres tons of things Ive noticed over the years where cameras would benefit from being FOS.

My biggest issue is choosing an image sensor(s). I would require one with readily available documentation. I would also want one that allows its use in a commercial product without fees and NDAs. I also strongly prefer a minimum size of 1 inch, but would prefer 4/3, APS-C, and even FF. Resolution in the 4-12MP range

tl;dr any recommendations for an image sensor that I can use in a product that has freely available documentation, no NDA requirements, allows use in a product with little to no restrictions, is between 1inch and FF, and between 4 and 12 MP

File: file.png (192 KB, 991x1087)

192 KB PNG

>>2856839
cool idea. your best bet is to basically do what framework did: design multiple boards with different sensors. you wont be able to make one single board for every sensor; theres no standard pinout or anything like that.
from a quick search, i see that onsemi makes some OTS CMOS sensors. they look pretty nice too; pic rel is only $400. theres also some literally who company called AMS that has some stock on Mouser/Digikey, but i think those two are the only game in town.

>>2856839
Yeah, good luck bro. That's a big big task. Like genuinely multi year effort big.
Here's a open source camera from just a few days ago. https://www.youtube.com/watch?v=OkfzjmY9cF8

There's also the Chronos 1.4 High speed camera by Tesla500, there is videos of the full development on youtube.

>>2855944
there is those numerals, but what a gay and faggy question.

File: sda.jpg (197 KB, 577x985)

197 KB JPG

I haven't plugged a single thing into the breadboard yet and I already want to kms myself
How do you people do it

>>2856873
Either I breadboard as I go, or I rely on a software that actually has a netlist, like KiCAD. Even if I have to lay out the breadboard traces manually, it’s easier than trying to rearrange the components manually in a digital paint program.

There’s also fritzing, lmao.

>>2856839
Sounds cool. I'm not familiar with camera electronics, but I assume they have an ASIC to shuttle the data from the sensor into memory, likely encoding/compressing it too. You'd either need to find such an IC that has open documentation, or to find an FPGA to perform the same task.

I'd see if any of the image sensors mentioned by >>2856843 have appnotes that describe some possible supporting ICs.

>>2856873
I don't bother with shitboards anymore, I just do LTSpice to ground plane

File: 1709833617768380.jpg (164 KB, 1500x1450)

164 KB JPG

Are cheap ass chinese heat guns like this any good or am I asking for trouble?
https://www.amazon.co.uk/gp/product/B0BMPVXTFP

I don't have a big need for one but just figure it could come in handy one day, especially if I need to replace a HDMI port on a board or something. I don't really know how to use one either, though I do know how to solder. Worth buying or avoid? Works out around $15 with a voucher I have

>>2856920
That's for stripping paint and removing bumper stickers. It's what automotive mechanics use to fix their Xbox. Maybe you want a hot air rework station so you don't kill your electronics with static discharge.

File: 69228ED9-A44E-465A-9439-0(...).jpg (265 KB, 1016x1025)

265 KB JPG

>>2856873
Never seen anyone take the time to do a breadboard layout before.
Get on my level.

>>2856976
>grandpa's computer
https://www.youtube.com/watch?v=LfQ-Xg_MS3k
All of that just to play centipede? What happens if you bump into it?

>>2856974
Oh, kek
So it's of no use to me for electronics?

Isn't static discharge relatively low concern as long as the board isn't on or am I talking out my ass again?

>>2857070
Certain ICs can be wrecked easily by static discharges, especially when they're removed from the PCB. A heat gun will definitely desolder components but it's the wrong tool for the job. In the end you can do whatever you want.

>>2856427
>electric blanket from Aliexpress
>wants to change plug from EU to UK
A-anon...

>>2857090
I decided against it and cancelled the order, thanks anon

File: 1ECC8ECE-AED4-49D1-8B13-A(...).jpg (328 KB, 986x1295)

328 KB JPG

>>2856980
Look closer. Not for playing centipede. Although that was a good game.

>>2857202
>The OCD is strong with this one.

>>2857202
So much effort for shitty breadboard contacts.

>>2856976
I would want to do a ground penetrating radar scan of your backyard, or a thorough exam of your basement (if you have one).

>>2857070
I worked in IT for 26 years and did it as a hobby 7 before that, in my entire time working I never wore a static discharge thingy. zapped many a component with quite impressive static discharges, caused several reboots, and never saw anything ever get damaged by static.

I even did experiments with purposely trying to fry memory, hard drives, and video cards with static discharges. I never succeeded in killing any device.

In one instance I tried frying a hard drive connected to power and on, by dragging the 12v power connection randomly on the hard drives circuit board. could not kill it.


yet I have seen new out of the box computers hard drives fail on first boot. so...

>>2857202
what are you hiding anon?

ಠ_ಠ

>>2857311
>never saw anything ever get damaged by static.
Remove random components from the PCB and zap them once with at least 100V discharge, then replace and test. Do FETs, diodes and LEDs first, then caps and resistors.

>>2856839
>fixed about 20,000
no you didn't.

>>2857321
I zapped memory with a discharge of about 15,000 volts (based on length of spark). all the things I zapped were not in a computer at the time.

>>2856839
many 4/3s sensors are widely used in random astronomy cameras, and the 4/3 mount is kinda open, so ill look into that direction.
good luck, and start a repository or something we can follow

>>2857378
What I mean is that the PCB ground plane (and sometimes ESD-specific mitigation) is protecting the chips. If they're removed from the PCB so the pins are floating and you zap the chip it's highly likely to die. You can trim carbon film resistor values, trash p-n junctions, kill sensors, etc. all with static discharge. It's far less likely to happen with an assembled PCB, to your point.

File: ESD_9.jpg (162 KB, 800x505)

162 KB JPG

>>2857378
ESD blasts craters in the chip at the point of contact. Sometimes they're bad, sometimes they're not. It's not black magic, it's not some autistic all-or-nothing rule, it's just a matter of probability. You either fry the chip, damage something unimportant, or damage something in such a subtle way that it may only fail on you much later.
I'm all for not being a paranoid retard about ESD, but ignorance and superstition are equally stupid.

This manufacturer has a wealth of information about electronics on their website
https://www.shindengen.com/products/semi/column/basic/mosfet/mosfet_on_resistance.html

File: usb_hot_plate.jpg (5 KB, 259x194)

5 KB JPG

If you use hot plate for resoldering/repair is there some kind of limit on how many heat cycles I can run on PCB/componets? (Less is better, I get that)

File: Untitled.jpg (397 KB, 1710x984)

397 KB JPG

>>2856873
Halfway through and I feel like this has no hope in hell of working, maybe this was too ambitious to try and put on a breadboard tbdesu

>>2857455
Put the power supply and audio amp on a piece of protoboard, and the LFO section on another. Socket the opamps, stack the boards with standoffs, and connect them with pins & header. Use another piece of protoboard as a faceplate and mount for LEDs/switches/pots/jacks.
It's more trouble than it's worth using a breadboard to build circuits beyond a certain level of complexity.

>>2857420
yes but manufacturers wont tell you the exact numbers.
also its not about the cycles, its about the total amount of heat. a semiconductor device will degrade at room temperature if you wait long enough, but its not something to worry about. at typical preheater temps i wouldnt worry about it either desu.

File: 20241004_101950~2.jpg (840 KB, 2286x2879)

840 KB JPG

Will a pcb cleaner or iso clean up what I hope is melted glue around my dodgy solders? Did I melt my board bros?

File: 1589590388068.jpg (86 KB, 750x468)

86 KB JPG

>>2857518
RIP, Anon. The glue came out.
F

>>2857518
>melted glue
looks like cum to me. you jacked off on your board, didnt you? fuckin sicko.

File: a neat board requires ver(...).jpg (53 KB, 600x345)

53 KB JPG

>>2857455
>too ambitious to try and put on a breadboard

bullshit, you just did everything all wrong
- trim the resistor leads and stand them up vertically - they take up 10x less room
- the choice to use no wires is completely retarded, and is responsible for creating that disaster area
- if feeling overwhelmed on such a simple circuit, build one section at a time, test it, then continue

>>2857523
What do
>>2857525
You got me, it's a sickness

>>2857534
It's just flux from your solder. IPA should clean it up unless it's a mystery goo from that place with the Ramen noodle skyscrapers.

File: derf1.jpg (198 KB, 900x1376)

198 KB JPG

>>2857527
Against all odds, I finished and it does work, but breadboard was a mistake. I didn't want to trim any component legs in case I want to re-use them later - the idea was to make it on breadboard first, before committing to stripboard.
I think I may as well solder it up as a finished product though, as it would just be a pain to try and return these components to inventory, sort out all the resistors, etc.

>>2857548
>I didn't want to trim any component legs in case I want to re-use them later

nothing stopping you from using them later on stripboard, or PCB, or whatever
just solder them vertically
and if you leave the shorter leg a little long, then you can still use 'em horizontally later if that's you kink

File: Corporal 6-transistor radio.jpg (169 KB, 1200x1600)

169 KB JPG

>>2857554
>just solder them vertically

see how much space you save

File: 1574959933042.jpg (12 KB, 200x200)

12 KB JPG

>>2857555
>vertical resistors
just awful

>>2857557
skill issue

>>2857490
Thanks, bro. I guess PCB can handle the most and complicated semis the least.

>>2855842
How do I into making RF oscillators and sheet? I can't even properly calculate an NPN amplifier properly or keep a colpitt's oscillator going

File: receiver transmitter pair(...).jpg (130 KB, 1506x561)

130 KB JPG

>>2857627
>making RF oscillators

for the first time in 13 billion years, you can buy an Rx/Tx pair for less than the price of a pack of Hubba Bubba
and you wanna waste your precious time, talent, and bodily fluids trying to make one from scratch
sounds like the beginnings of a sad wasted life
not unlike my own

>>2857639
>sounds like the beginnings of a sad wasted life
>not unlike my own
Peas in a pod we are, anon
And I wouldn't have it any other way

>>2857627
It’s probably easier if you use those components with negative resistance areas, like tunnel diodes. Relying on negative and positive feedback with a phase delay to make a sine oscillator just sounds like a pain.

>>2857627
https://www.youtube.com/playlist?list=PL9Ox3wpnB0kqekAyz6blg4YdvoEMoJNJY

And don't even thing about using solderless breadboards for any of this or you will have a very bad time.

File: 1427428014811.jpg (133 KB, 780x818)

133 KB JPG

>>2857776
After a few days of researching oscillators, I have come to the same conclusion. Although I'm not sure if there's any way to find these components besides actually buying them from digikey etc.
I built a relaxation oscillator on a breadboard but it barely went above a few KHz. I built astable multivibrator and it was also very slow

I wish there was an way to build cheap and easy VCOs

>>2857935
You know what anon? I'll watch it all. I don't even care at this point

File: 1723614425491447.jpg (62 KB, 633x758)

62 KB JPG

>on a work trip to buddy's town
>bud is a police officer, desk job or something
>visit his station, go together for lunch afterwards
>during the station visit he gives me a tour
>walking by the store room, see a massive fluke multimeter on the shelf
>walk over and take a look because it looks like a really high end model
>it's an "insulation tester", not a multimeter

>>2857943
>insulation tester
Taser maintenance?

>>2857944
Torture device anon
Used to be more common in the old days

>chinkshit lab power supply breaks
>always displays 18V and 0A despite fiddling with V/I knobs
>disassemble to see what's wrong
>destroy plastic iec inlet due to it being physically fucking impossible to squeeze the tabs and pop out
>nigger rig everything up with an old extension cord soldered to the PCB
>non-standard retarded chink through-hole sizes and so 16 AWG stranded wire extension cord barely fits
>plug it in and everything works fine except for the mentally retarded temperature-dependent fan control
>(guess it fixed itself!)
>try reverse engineering fan control
>the usual chink nonsense of using an op amp and a diode to save 0.01 RMB on BOM
>still not sure why fan control isn't working
>realize that i need to buy a new power switch, new IEC inlet, 18AWG stranded wire, etc.
>impossible to even find the correct size IEC inlet on aliexpress due to it not having an integrated SPST switch
>realize it's literally less of a headache to throw it all in the trash and buy a new one

>>2857970
So like those russian hand crank phones but more modern?

>>2857989
>more modern
Yes. You need a subscription to GOVX to use it and it doesn't work on niggers. Powered by KangAI.

>>2857518
It honestly looks like leftover flux from what I'm able to see. Looks like your iron temperature is too low, though, you should crank it up more and reflow those joints.

>>2857202
This may look nice to autists but using mostly one plane increases capacitive coupling between signals but then again, circuits build on bread board shouldn't be high speed digital or sensitive analog circuitry anyway. So this is just a fun way to waste time I guess.

>>2856839
This list might be a good starting point:
https://wiki.apertus.org/index.php?title=Image_Sensors

>>2857377
If he's been repairing cameras for 12 years, and he worked 5 days a week 50 weeks a year, that's 5 * 50 * 12 = 3000 days of repairing cameras. If he repaired 7 cameras each day that's 21,000 cameras, so it doesn't seem unbelievable.

File: tr.jpg (2.39 MB, 4000x1800)

2.39 MB JPG

am I putting too much solder?
Also, bending the component legs so that they stay in the holes while the board is upside down seems to make it much more likely that I accidentally bridge two tracks, because the legs are already pointing perpendicular to the tracks. I feel like there has to be a better way.

>>2858183
nah
>bending legs
doesn't have to be a huge bend, 15 degrees or so works

>>2857982
Im in a same boat with chinkshit atx psu.
250w full of dry caps, known fault with 5vsb killing motherboards ect.

>>2858183
Bending legs will make your life miserable when you desolder components.

>>2858183
not too much solder, joints might be a touch cold, but good enough.

Those radioshack quality level boards are a pita. Have a look at something from twin industries:

https://twinind.com/index.php/products/prototyping-boards/

then look at the
"8100 Series Plated Through Holes - Single Power/Ground Plane"

I like the 8100-4565, and normally keep the ground plane on the component side. you can bridge a socket pin on the top side or loop a component leg down and back up for a ground connection. These boards seem to stand up to soldering heat a lot better, and the ground plane seems to speed prototyping.

Admittedly I went nearly full retard on smt and don't do much though hole anymore.

File: a perfect solder joint.jpg (48 KB, 640x375)

48 KB JPG

>>2858183
>am I putting too much solder?

yes, you're definitely in ''too much solder'' territory

File: Capture.png (118 KB, 1221x466)

118 KB PNG

for this module, am i supposed to leave both the SMA GND and pin 15 disconnected from the actual ground plane of the board?

>>2858278
RF ground is separate. Pins 15 & 16 are only for the SMA connector.

I only want to make simple projects in my free time, not complex ones. Even as my skills get better i actually only feel like making more simple projects

What is happening?

File: 1711205145415593.jpg (361 KB, 1066x800)

361 KB JPG

I came up with an idea to motivate me to learn soldering: simple shit like soldering batteries, salvaging components, etc. Ok so I was cleaning an mp3 player with IPA and, because I hadn't known back then, the IPA dissolved the foam gasket thingy around the LCD. Disintegrated it completely. Looks like some types of glue are very weak to IPA for some reason, while others need something more aggressive like naphtha.
Anyway now I want to replace it before I reassemble it. It was square shaped, 1mm thick and wide on all sides, see pic rel. The surface on which the foam was applied is typical plastic, like the one in your phone (the case side, not the LCD panel). What should I use?
So far I've identified these options:
>silicone (like the one you use in cars or your bathroom or kitchen)
>liquid electrical tape
>regular double sided tape (might be too strong? it also looks ugly because it's white)
>polystyrene

Looks kinda like pic rel but it's a different model. Same idea though.

>>2858293
silicone tape

>>2858295
looks good, thanks! I didn't think about that. Is there a smart method to cut it into super thin strips (1mm or 1/32 inch)? Or should I just use a razor and keep a steady hand?

File: cell phone repair kit glue.gif (95 KB, 1049x729)

95 KB GIF

>>2858293

you might wanna get the real deal instead of half-assing it like usual

>>2858296
I use an X-acto knife but a razor would work just as well. Clean the blade first.

File: 1725219001125025.png (170 KB, 293x319)

170 KB PNG

>>2858297
I thought about using glue, but this space is so tiny I'm afraid that when I install the LCD it'll just spread around and mess up the screen, and I'd have to clean it again and try some other method. Or is it more viscous than that?

I tried to measure the distance between the LCD and the display cover, and it's something like 0.5 mm. So the foam gasket thingy had to have been 1mm wide and 0.5mm thick when compressed. Or I guess 1/32-1/64th of an inch.
(again this isn't my pic but my model is almost identical. I couldn't find any good pics online for Sansa Clip Sport).


>>2858298
Good point, thanks. I'm using DE blades.

File: 1697881508228195.png (83 KB, 565x489)

83 KB PNG

>>2858299
I found this thing:
>tesa® 61395 is a double-sided black self-adhesive tape with a PET backing and tackified acrylic adhesive. Ideal for cover lens and touch panel mounting, it provides superior bonding and push-out performance. Its special acrylic adhesive and versatility make it perfect for various surfaces.
It's also pre-cut (2mm width)

>>2858257
I see that you recommended board just with holes and no strips, why is this thread so against stripboard? I'm still an amateur of course, but to me it seems preferable that stripboard gives you so many inter-component connections "for free", whereas with perfboard/solderable breadboard you need to make all the connections yourself. I can't imagine how fiddly it is wiring up an 8-pin IC socket like that.

>>2858320
Different Anon here. I think you should use whatever suits you best.

tldr: is there an way to reuse the scanner parts of a printer with some sort of open source software?
printer broke cuz it hit hard coded page limit, I tried to flash new eeprom data with .bins from internet but its still broken, its a printer+scanner combo (scanner refused to work to after the limit was hit too)

File: vuescan at pirate bay.gif (126 KB, 1081x824)

126 KB GIF

>>2858335
>open source software?

you might try Vuescan as it includes many drivers for scanners that dont rely on manufacturer's drivers
try the trial version, and, if it works, buy it or pirate it

>>2858293
Take the learning process a couple steps further and design a new case with 18650 battery for it.
>Hard mode: New display and switches

That's what's on my list.

>>2858018
> mostly one plane increases capacitive coupling
That’s why no electrical devices use PCBs I guess.
Also, no need for coupling capacitors anymore.
Got to get with the times and design circuits that are cubeoids and spheres to avoid the single plane pitfalls that were the reason PCBs never worked.

File: 1274684108802.jpg (31 KB, 512x512)

31 KB JPG

>>2858330
That's not how it works, we're supposed to have a fight about which way is best and flame/troll each other

File: SDC12428.jpg (1.67 MB, 3600x2400)

1.67 MB JPG

I've cleaned these pots with QD contact cleaner, they play clean for a little while the go back to having dropouts and static pops. Is Dexoit still the best way to lube these or are there other options out there now? Deoxit is surprisingly expensive

>>2858357
>18650
That would increase its mAh tenfold (from roughly 300mAh to 3000-4000). It would be awesome and comically overkill to have my little mp3 player last weeks. It would pair well with my e-reader. Yeah, a good idea but I'm an europoorfag and I can't afford a 3d printer or a cnc at this time. But I'll try it when I get the right tools.

>>2858367
> deoxit
Yeah marco reps was doing a thing on pure copper contacts and I learned that it’s not really a special lubricant (I thought it was conductive or something, I don’t know) but it’s just some kind of grease or mineral oil.
I’d clean them with contact cleaner first though-somethhing that doesn’t leave any residue.
Ideally, you’d probably want to take apart the pots, clean the contact surface and wipers, the slightly move/nudge the wiper so it’s on a fresh, non-worn area of the carbon track, lube it with something and put it back together.
Watchmakers can do shit like this, so why can’t you? Also good skills for repairing video game controllers.

File: file.png (71 KB, 391x990)

71 KB PNG

>>2858373
I dont see an easy way to remove these pots. The shaft has nut on the top-facing side, im guessing the black clamps on the side hold it in place for the proper orientation, but PC1/PC2 are in the way to simply slide back and out the pot

>>2858378
> not easy
I have an old sears radio and a creative labs speaker amp that need new pots and I could’t bring my self to disassemble the whole thing, take the PCBs out, desolder the pots, fix them (or replace them) and put it all back together.
But I think that’s what’s gonna make the difference between an actual repair, or a half-assed hack job.
Deoxit is like a temporary fix if you’re selling it on facebook market place and your moving to another country in a week when the deoxit stops working.

File: SDC12423.jpg (2.71 MB, 3600x2400)

2.71 MB JPG

>>2858381
I'll spend some time looking at the assembly of the unit to see if it could be done easier.
>if you’re selling it
still on the fence about this, I got it 13 years ago and fixed it up then, and its been in a storage unit for the last 9 years. I brought it back to my house last week, and aside from the pots it works as it should. In the last 9 years I've bought other stereos and record players, so I dont know how long I plan to keep it this. It is a nice unit though, a 1971 magnavox astrosonic

>>2858369
>I can't afford a 3d printer or a cnc at this time. Neither can I and that's my choice is OpenSCAD & JLC3DP.

Though I have to admit I got no idea where my Sansa Clip+ is let alone does it work at all.

>>2858304
it should probably go without saying but this stuff is absolutely tiny and a pain to work with. just something to keep in mind if you plan on using it.

File: Capture.png (24 KB, 816x564)

24 KB PNG

i have a 12v signal that i need to divide down to 0v-0.5v. obviously i could use a potentiometer but the signal absolutely cannot (like if the wiper leg skipped) go above 0.5v or bad things will happen to the downstream circuitry.

what's the best way to do this?

>>2858430
i guess i also mean to say, i tried dicking around with pic related but all of the cheap 10 turn potentiometers are 10k or 100k max.

>>2858418
yeah I gave up, can't even find that specific tesa in my country and I don't want to order it off amazon. But even if I had it, it'd probably be a nightmare to work with. I assume these manufacturers use robots to cut through these adhesives so neatly.

Sansa clip sport isn't even that water resistant to begin with. I squirted some ethanol on the case and it seeped through the keyboard and the edge of the display in just a few seconds.

>>2858397
Nice. I'll do some research on that stuff too, thanks. I have a laptop mobo and I've been thinking to turn it into a router or simple NAS.

Are you working on any personal projects where you use that stuff?

>>2858432
>Are you working on any personal projects where you use that stuff?
Not yet, while the learning curve isn't as steep as it is with "real" CAD software it's still considerable and I have to admit I've only played around a bit.

However, besides "Sansa Clip+ Supreme", I've been thinking about a truly portable lab PSU with RK6006 and a two or three high-watt USB-C outputs. Could be built wth a 20V tool battery relatively easy but hey, there would be no challenge in that.

>>2858183
pretty sure that pcb is oxidized to shit.
give your next one a scrub with fine steel wool or scotch bright right before getting started, that should fix the ugly blobs

File: u.jpg (1.38 MB, 1586x1574)

1.38 MB JPG

The answer may just be "it depends" but here goes: how important is it to ground the bodies of potentiometers?
This circuit is controlling a bunch of oscillators to make weird noises, and I notice that the sounds change even when I just touch some of the potentiometer bodies/shafts, not even adjusting them. The circuit also starts behaving differently after about a minute of being powered up, which suggests to me something is building up a charge somewhere.
This is just described as a "wacky noise maker circuit" which isn't terribly helpful for troubleshooting, I don't know if it sounds wacky in the right way. But I'm wondering if the problems will be alleviated if I mount all the pots in a grounded face plate/control panel.

>>2858448
The solder doesn't seem to wet the traces consistently, I will give that a try for my next board for sure

what problems arise when using resonant inductive coupling to transfer power? I don't mean wireless charging, I mean like powering a device strictly from induction.

>>2858455
very. every single one acts like an antenna and picks up mains.
good luck, it's a massive pain in the ass getting the solder to flow.

>>2858335
>printer broke cuz it hit hard coded page limit,
what the fuck

>>2858335
>hard coded page limit
Your honor, I hit my hard coded non criminal limit.

>>2858473
it actually "broke" two years ago, just now I was thinking about trying to repair it again and got an eeprom flasher, just because I wanted to use the scanner functionality, there's a whole business selling .bin eeprom images (mostly on russian forums) to reset the page counter that manufactures put on the firmware just to make them unusable after it prints 2500 or so pages, mostly on inkjets, I tried some .bin that I found but the shit just failed. anyways I was watching a video earlier talking about how canon also does this, but here's a rossmann video if you want to know more
https://www.youtube.com/watch?v=J_XiaMzzZUo

>>2858430
Usually you’d have a resistor tied to the 12V signal, then your potentiometer ends tied to that resistor and ground respectively. Size the two resistance values right and the wiper should output a voltage between 0V and 0.5V. If the wiper skips, the output is left floating, you could pull it down with a big value fixed resistor. 0.5V is very low for a maximum, what’s it going to? I’d suggest using some sort of diode (forward-biased schottky) to clip the output voltage, but in this case it’s not trivial. If I had a 5V ADC with a 2.5V voltage reference, a reverse-biased 3.7V zener diode would be nice and safe, for example.

>>2858455
High impedance potentiometer circuits can definitely capacitive couple noise from the person touching its shaft. Less so when touching a plastic knob. Grounding them is commonplace for hifi audio circuits, less important for lofi and synth stuff, but it’s a pain when you adjust something to sound good, then it changes once you take your hand away.

You also have to consider what will happen if you mount them to a metal panel and don’t tie that panel to a 0V rail somewhere, the crosstalk could be significant.

I see you too are making a stripboard circuit with a mess of wires coming off to pots and switches and such, did that a few months ago myself.

>>2858469
Feedback, efficiency, resonant topologies (with feedback) themselves are a pain, beam-forming if you’ve got multiple transmit coils, high-speed (synchronous?) rectification, it’s a lot of potential difficulty.

>>2858482
already knew these people were walking scum, but holy shit

>>2858493
>I see you too are making a stripboard circuit with a mess of wires coming off to pots and switches and such, did that a few months ago myself.
What kind of circuit?
I'm trying to make the "music from outer space" beginner level audio synth type projects but I don't think this one is working properly, I made the mistake of using some guy's stripboard layout off a forum without taking the time to fully understand it myself first, so I don't know how to troubleshoot it. I can't believe I'm having to face the consequences of my own actions.

>>2858513
Bass booster filter between my TV and amplifier so I could get the full experience when rewatching MS Gundam Unicorn. Was a dual-op-amp circuit with a dual-gang rotary switch and dual-gang pot for adjusting gain and frequency. The chosen frequencies for the switch aren’t as good as they could be, I may open it up and swap some different caps in. At the moment three modes are indistinguishable.

It’s in an enclosure with a relay that turns on the amp when the TV outputs power through its USB port.

Hmm, ordered a 936 / 900-series pencil to use with threaded inserts, but it has a different plug and thermistor resistance compared to my T12 station. I have a spare T12 controller I could likely swap out a resistor on to get it working with this unit, I guess that's worth doing? And it's a plug I can buy locally.

It has a larger microphone/aviation connector, and the pinout is:
>1-2: ~4Ω element
>3: connected to tip
>4-5: ~50Ω thermistor
4Ω is pretty low for a 75W iron too, I wonder if it will run fine on 24V? I'll be building a pair of old ATX PSUs into my bench for general-purpose 12V and 24V projects, so current won't be a problem. The 50Ω is damn low too.

>>2858543
>The 50Ω is damn low too

for a thermistor, sure
but kinda normal for a thermocouple

>>2856976
how do you diagnose a random lose connection on one of these

>>2857555
sovl

can i pull off a signal jammer to stop people from playing youtube loudly on their phones
i only know digital

File: Cldkem Logic Probe DTL TT(...).jpg (8 KB, 350x261)

8 KB JPG

>>2858559
>how

a sharp probe and a dull mind

>>2858563
thats really really illegal

>>2858621
can't be illegal in your home

>>2858626
you think its impossible for something to be illegal as long as you do it at home?

>>2858627
yeah usually there are exceptions for personal use

>>2858629
youre not gonna be able to scramble signals just inside your house.

>>2858563
transmit noise onto 2.4ghz band
gl hf

Would 1.8k pullup resistors (3.3v) work for i2c fast mode plus.
Read some datasheets etc but they are a lot more complicated than my question. Min value seems to be around 1k at 3.3v but the max seems crazy high unless my math is off or more likely I dont understand what Im trying to accomplish.

>>2858553
I'm fairly sure it's a thermistor, that's what a few people online said anyhow. The diy circuits use a voltage divider made with another low-impedance resistor to read temperature, as opposed to a high-impedance buffer. 50Ω is also really high for a thermocouple. Furthermore, when I hold it in my hand to heat it up, the resistance increases about 1Ω, regardless of which wire goes to my DMM's negative, if it were a thermocouple I'd see a value that's distorted by the voltage produced by the thermocouple, depending on the measurement polarity.

Also when I take it apart there's a little PCB inside with a thermistor symbol on the silkscreen.

I want to buy a replacement battery for a phone before I need it, in case it becomes unavailable later. I'm reading that lipos should be stored at 3.6-3.8V, but can self-discharge significantly if you don't check their voltage every 3 months or so. Especially if they have a BMS that's draining their power constantly.

Has anyone designed a long-term preservation circuit that keeps them topped up with minimal risk of damage, for years at a time? It would need to survive power outages, and so have a method of sensing battery voltage without backfeeding the sense circuitry when the sense circuitry loses power. I'm thinking a long-tailed pair made from a dual MOSFET package. Would there ever be a case where discharging would be necessary? Keeping them in a cool dry place is suggested, I'm not sure how far to go on this front either. Peltier cooler + desiccant in a sealed vessel seems over the top.

>>2858758
> battery top-up
> needs to survive power outages
The top-up would only *need* to run, like, once a month, no? Or you have 12-month power outages and you need to (a) move, and (b) get a small solar cell.

I see you’re on the right track though.

>>2858762
>The top-up would only *need* to run, like, once a month, no
Probably, but I was thinking that if I could get away without needing a microcontroller (or needing to remember to push a button on occasion) that would be nice. AFAIK there's no problem with keeping a (current-limited) 3.7V source directly connected to a lithium ion cell, so an op-amp or LM317 CC/CV circuit would be more than enough, which isn't really something you can say for lead-acid. That said, an MCU would be able to detect battery fault conditions, like if you kept trickle-charging it but the voltage never went up, so I'd probably add a thermistor and/or thermal fuse to trigger some cutoff if I wasn't using an MCU.

Power outages are just a concern for the safety of the electronics, or if someone unplugs the thing accidentally. If I ran it off an RCD then that would need to be manually reset after a power outage, but I should be able to avoid that situation.

On a similar note, I'm planning on making an electrochemical reactor to regenerate ferric chloride and pull copper dendrites back out of it. If this ever loses power, the copper cathode and any dendrites sitting on it will just redissolve. So it will be important to give such a circuit a battery backup to keep a small potential across the cell at all times. Keep the current low otherwise it makes chlorine gas.

File: YF-19.jpg (958 KB, 1848x2157)

958 KB JPG

pcb art

File: this is art, that's clipart.jpg (113 KB, 637x900)

113 KB JPG

>>2858779

please kindly sir, to distinguish between art and clipart

>>2857377
first 3 years doing point and shoots i averaged 12 a day.
say 9,000 for first 3 years.

remaining 9 years were DSLRs, lenses, and video cameras, so I averaged more like 4 a day.
So say 9,000 for last 9 years

so that puts me at 18,000

I put in plenty of OT and worked a bit over 12 years. My math here was super simplistic. Maybe I didn't quite hit 20k, or maybe I surpassed it. 20k is my perfectly reasonable round number estimate.

File: il_794xN.4183296514_j5wk.png (479 KB, 794x596)

479 KB PNG

any way i can disassemble this mic?
it's polish dynamic tonsil mdo-12
no screws, can't turn the grey head, did they super glue this or what

Dunno if this is the place to ask but I will anyway.
Does analog layout get any less boring and tedious? I started a 6mth internship in a research company's IC design department about a month ago and the work is making me want to blow my brains out. I'm realising a lot about myself and part of it is that what I'm doing now really isn't what captivated me with electronics when I was younger. Using CAD software and drawing a bunch of rectangles over and over is messing with my head. Considering asking my school to transfer me to some where more hands-on, working on PCBs and shit when I finish 3mths here so I can say I spent the time here at least. You boomers think that's smart? Or should I tough it out?
Cheers, zoomie.

>>2858901
>analog IC design
Do people still do that? Aside from like 15 people making minor improvements on op-amps working for TI and Analog Devices?
Your coworkers are going to be the best to ask about the variety of work in the future. It could be you've been left with the boring shit because you're an intern, or it could be that's just par for the course.
t. retail slob

File: usb-wiring-connection.png (1.42 MB, 1024x768)

1.42 MB PNG

I soldered 2 wires (d+, d-) to my audio interface cable (d+, d-), and other 2 wires (power/ground) i soldered to cable that runs to adapter power outlet/(5 volt/1 amp, same as USB 3). Audio interface works (led diode is lit on usb3 mode and can use it in when switched back to external power mode) but when i plug it into laptop usb 3 or 2 jack nothing happens.
Anyone know what i did wrong?

>>2858924
The 5V from your power supply was backfed into your USB port. Disconnect the red wire from the PC while you're using external power.

>>2858930
>Disconnect the red wire from the PC
it is not connected to anything (only white and green are connected to the interface) or you mean disconnect pin 1 and 4 by pulling it out?

File: 427399-anime-girls-circui(...).jpg (1.29 MB, 3840x2160)

1.29 MB JPG

>embed nichrome heating element in a small slab of white cement
>use as reflow plate
Tell me why this wouldn't work? A basic ass reflow plate costs 80$ in my third world shithole

>>2858940
nichrome would have hard time heating up a thick slab like that. it would probably work but if youre salvaging garbage anyway, grab an old heating element from an electric stove and just slap some piece of metal on top of it.

>>2858938
Sorry, I misunderstood. Maybe your D+ and D- got swapped or you're missing pullup/pulldown resistors.

>>2858901
why would you ever get a job in something you love

>>2858901
youre a bit overqualified for this place, fellow zoom-zoom, but i guess there really isnt anywhere more appropriate for this discussion on this site.
generally things get more boring and tedious the more you do them, not less. but even if you were having a blast i would strongly recommend trying out as many different specialties as you can while youre still young. i just turned 25 and im starting to worry that im too old to be hired as an entry-level VLSI engineer. its unlikely you find a job you love 100% of the time (and if you do, buyer beware, see below), but making an informed decision is nonetheless important. if you think youll like PCB design then you should definitely give it a go, or else you might spend the rest of your life meritlessly convincing yourself youre in the wrong field.
>>2858984
unironically this, i work 16 hour days + weekends when literally no one asked me to and theres no need for it. im like a whore that undercharges her clients because she likes getting fucked by dirty old men. im too scared to leave this job because im afraid ill never like a job as much as i like this one.

File: PXL_20241008_184904691~2.jpg (490 KB, 2688x1512)

490 KB JPG

hey guys u like my triangles?

where the fuck did i put my multimeter? please help very urgent

>>2859118
did you check the fridge?

>>2859012
are you seriously thinking about being too old for a job position at 25

>>2859130
every year you get older is one less year of potential productivity they can suck out of you

>>2859130
why would they hire a 25 y/o with no relevant experience when they could hire a 22 y/o with no relevant experience?
you also have to think about lost wages; youll be 25 making a 22 y/o’s wage, 40 making a 37 y/o’s wage, etc. it compounds.

>>2858940
>reflow plate costs 80$
I'd upcycle a discarded flatiron

>>2859126
found it thanks
it was in my bag where i left it

>>2858898
Screw under metallic plate?

Hello, I am Senator Jon Tester (D-MT) and I approve this general. I love testing shit.

>>2858877
Storing them in raw is a climate changer.

>>2858901
Digital is a subset of analog. You have someone elses dream job. Appreciate what you have.

>>2858954
> electric stove
You can get standalone counter-top frying pans from value village for $5. Y pay mor?

>>2859118
My dad borrowed it and blew the fuse again, sorry.

File: p8.jpg (433 KB, 1298x533)

433 KB JPG

I think I've reached the point where I can no longer keep stuff organized on the basis of which box it shipped in, what approaches do anons take to keep components organized? Do you use those plastic organizer units with drawers? Storage bins? Sort component properties by order of magnitude?
It's not sustainable to look through 40 strips of resistors every time I need a specific type...

>>2859188
For strips of SMDs, get a clear file book and fuse some dividers into it with your soldering iron. For everything else, those mini drawer racks. I like to use appropriately sized containers for project-specific storage, for example I've got all my hifi audio ICs stored in the same icecream tub, along with a salvaged gooseneck and some really thin coax.

I accidentally got a 2-wire inductive sensor for my 24V 3D printer instead of a 3-wire one. The datasheet states a voltage range of 6-36V, a maximum load current of 300mA, and a quiescent current below 10mA. Is there a good way of getting a solid signal from this without needing a power resistor? The mainboard has a 3.3V logic level that I ultimately need, but I can use a small-signal BJT to do logic level conversion.

I was trying to think of some sort of long-tailed pair circuit that runs on 3.3V but can compare signals above its rails, but it doesn't look like a simple 2-transistor circuit can do that. So I'll probably just use a comparator with a 3.3V pullup. I've given it a 910Ω resistor, and when not triggered the sensor is drawing 20mA through it and presenting only 3V across itself, dropping the other 19V across the resistor (gets a bit toasty). When I trigger the sensor, the voltage across the resistor drops to 0.7V. Giving it a 3.3kΩ resistor instead and it still seems to work fine, but I'll probably just get a 5W 1.2kΩ resistor to stick to the datasheet specs, and use a 12V voltage divider for the comparator.

How easy would it be to rig a range extender to an EV that kicks on when the battery drops below 50%? Basically, just want a gas-powered vehicle with no emissions testing, an electric transmission that never needs a new battery, or expensive engine repair.

>>2859253
If you're modding a hybrid vehicle, then it would be a software issue. No clue what that entails, but it's probably doable. If you're modding an electric vehicle from scratch, then the mechanical side is gonna be tough. Mounting the engine and its transmission components, then somehow getting that engine or its ECU to talk to an EV that was never designed to have such a thing.

Expensive engine repair is far rarer to require than expensive battery replacement, just stick to petrol. A petrol car should be expected to last 20 years minimum, and if you treat it half-decently and pick a reliable model (e.g. late 90s or early 2000s Toyota Corolla) I'd expect it to last over 40 years. EVs might last 20 years if you stick to within 20% and 80% of its battery, I guess that's doable for a strictly short-range commuter car, but they're going to be more expensive for what you get.

>emissions testing
Don't live in a kleptocratic dystopia.

>>2859255
>transmission components
"Transmission" would be an electric generator that charges the battery.

>>2859255
>expensive battery replacement
You'd never have to replace the battery if you simply use it as a capacitor to smooth the output of the generator.

>>2859188
>every time I need a specific type
a small box, organized by decade, for each type then?
still to this day don't know what those tiny drawers are for

File: 1634591926594.gif (82 KB, 900x770)

82 KB GIF

How useful of a skill is knowing electronics? Can you give me examples in your lives where this hobby has helped you and what exactly do you do for your projects?
Genuinely curious about it, not trying to trash your interests

File: asd1.jpg (249 KB, 1594x769)

249 KB JPG

How can I choose a decent speaker as a non-audiophile person
I'm generating a lot of bleeps and bloops with oscillator circuits and I want them to sound better than when they come out of a small, tinny-sounding speaker like this:
https://www.adafruit.com/product/3351
Do I just go for something with a larger diameter so it has better response at lower frequencies?

File: Society of St. Vincent de(...).jpg (89 KB, 960x960)

89 KB JPG

>>2859347
>a larger diameter

you're talking about drivers, not speakers
a speaker is 90% about the wood box, plus the equalizer, not the individual drivers
fortunately, getting adequate speakers cheap is as simple as a trip to your local thrift store

>>2859372

you can even bring along one of your bleep/bloop circuits to test them

no more of what you got back home to play your fuzzy warbles on
i bet you got, say, pitiful, portable picnic players.

>>2859340
There's a few main practical uses:
>repairing broken electronics
I've fixed a hifi amplifier that I use frequently now, probably other things I'm forgetting.
>designing novel electronics that you can't buy cost effectively
I made a high-power soldering iron that runs on an M18 tool battery, that isn't something you can buy anywhere. I'm also dabbling with a custom microphone circuit.
>getting work as a freelance
We're getting there, client has requested boards to work with a different kind of actuator, I've made the revisions and will be ordering the boards soon.

Specifically, simple audio and DC motor circuits are the kinds of things that are relatively easy to troubleshoot and design.

>>2859347
If you can find a datasheet, look for one that has a relatively flat frequency response in the frequency range you care about for your fundamental and its harmonics.

I've always wondered how many people get zapped by building utterly insane YT electronics projects.

https://www.youtube.com/watch?v=N4bBmjXcRgQ

>>2859347
any half decent $10 desktop speakers should do
>lower frequencies
save those for modules that pass the smoke/shriek tests and get headphones

File: Screenshot_20241010-14143(...).jpg (179 KB, 720x1123)

179 KB JPG

Would any Anons here be able to point me in the right direction to find a replacement board fan?
It's an ADDA AD0412HB-J93 12V 0.15A which has 3 pins. The board won't run unless it senses that the RPM is high enough and it's reading 0 RPM despite the fan spinning well.
I figure I'll try replacing the fan before the board ($2k+) but the only place I've found with it in stock is AliExpress (https://www.aliexpress.com/item/1005007731113749.html), mouser etc have it obsolete.
Is there a different search term I should be using? Alternative models that I could try instead?
Is there something I should be looking at instead of the fan since it seems odd for a tacho to die with only 5 years of operation doesn't it?

File: xxxxxx.jpg (48 KB, 1142x450)

48 KB JPG

>>2859449
Alright, time to see how good the $11 Aliexpress USB speakers are

>>2859590
Probably pretty good. The tweeters look real.
Most of the big brands have had fake tweeters for the last 20 years.

>>2859511
Usually the 3rd pin is pwm, not sense.
However, maybe you could give it a fake pm signal withh a 555 timer or something.
555s are also commonly used for the pwm speed control.
I assume you’re talking about thhe POST boot process where it tests the fans on full blast. For a few seconds… you’d hope the fan is controlled by a thermal feedback loop of some kind eh?

>accidentally used GDS footprint for NMOS on PCB
>every SOT23 MOS known to man is GSD
options? i think im gonna have to deadbug it with some enamel wire or something.

>>2855842
Yo there, what would be needed for a non touch lcd monitor
to enable the touch capability?
I have seen there are bulky kits to do so, or some kind of lvds-ttl translator
boards with external power, besides the touch panels with
factory enabled have it integrated in the driver for the 30/40 pins connector.
I have a couple of TSC2046IPW lying over there and have a couple of monitors
of 17 and 19 inches and perhaps these would be a fun toy for my childs to have.
I know there are resistive/capacitive system and inos/raspberrys modules so perhaps it
could be bridged somekind of layer (which) across the screen and update the
eeprom to setup the new x-y coordinates?
Is there a chance to setup a mobile phone flex touch to a monitor?
looked for some service manuals of pos screens but these seems to
have the rs232 and usb at the same time and I don't get how must be connected to
work.


also
how many connectors there are, 30,32,35,40 or 50 pins?
Which term should be searched for the flex flat cables?
Thanks anon!!

File: flipped D for S.jpg (1.41 MB, 2524x2219)

1.41 MB JPG

>>2859729
If you'd picked through-hole MOSFETs, you'd be able to do what I did:

>>2859761
A resistive touch panel would only have like 4 wires, though it would need some sort of analog measurement circuit.

>>2859639
It gets past boot, the problem is the proprietary board software won't run with a fan error.

>>2855842
Can any of you delightful human beings help me find schematics for a good, anon-tested active direct inject box? Thanks

File: 347787AA-2DE8-4A53-A926-8(...).jpg (272 KB, 1592x1360)

272 KB JPG

>>2859861
> active direct inject box?
That’s a simple op-amp usage example.

Use a J-FET input op-amp if you have an insanely high input impedance (or, maybe you’re just an autist)

File: 046E99ED-F94F-455C-9D71-0(...).jpg (43 KB, 620x489)

43 KB JPG

>>2859781
You guys are funny.
I’d just clip them and run some 14 GA to the holes.
All kinds of shit is mounted that way when they discover that the IRZ44N can just be bolted to the case and it will prevent burning it out.

>>2859866
> inb4 b-b-but wires longer than 3mm might cause an extra 5pf of stray capacitance that will bring upon the end of the world!

File: BSS138.jpg (17 KB, 700x557)

17 KB JPG

>>2859866
In my case, it was mechanically better to mount them with bent legs. In this anon's case, he's got a SOT-23, not exactly something you can bolt to the case. If the has extra room, he might be able to bodge two of its pins across other pads, and only run one jumper wire.

How can I reduce BT interference when soldering some wires near the BT antenna of a device?
Would thinner/higher gauge wires help? Or is there something specific I should use?
I am a beginner, I don't know much, and the wires are for audio.

>>2859888
Are you actually getting interference? Or are you asking pre-emptively?
I don't think you're likely to get significant noise, as bluetooth radio frequencies are far higher than you'd be able to hear, and far higher than the cutoff filters in audio circuits too.
If you somehow are getting interference, you'll want to run your signal with shielding around it, likely using some thin coax wire. I'd use something like RG174 or RG213, but there's also dedicated audio coax wire that isn't as finnicky about impedance that might be cheaper. You'd tie the shielding to an appropriate 0V / ground terminal, ideally a low-impedance ground rail relative to the audio signal.

>>2859888
>is there something specific I should use?

any components or wires you add next to an antenna is likely to de-tune it and reduce its range
so the best thing you can add is space, or separation

i need to do low side current sensing with as wide of a dynamic range as possible. sensing down to microvolts seems unreliable due to induced voltages from EMI. my thought was to use an op amp summer configuration and add to the sensed voltage a voltage from a precision reference? is this technique ever seen in professional designs or is this a stupid hack?

File: vs.png (6 KB, 862x354)

6 KB PNG

Say I want to turn off a load. Which approach should I go for? For my current application it's just a string of LED lights, but in general what's the difference and when should I use one over the other?

>>2859970
>when should I use one over the other?

2nd way is more expensive, so you use it only when you have to
still, many scenarios require it
i'll mention 2 out of the over 1 billion that exist
- when you have 2 live wires, like a 220V heater; you have to interrupt both lines so you can work on the load safely
- when you compare a bunch amplifiers by connecting them to one set of speakers; you have to interrupt the ground wire as well as some amps are grounded, others not

>>2859981
I don’t know for sure, but I suspect that the double pole double throw is going to last longer due to to less wear and tear on the contacts (the make/break arcs should be more evenly distributed)
Of course, there’s twice the failure rate because of oxidized contacts, too. But better fail off than on,

File: 20240901_095138.jpg (98 KB, 1179x731)

98 KB JPG

WHY THE FUCK DIDN'T ANYONE TELL ME I NEEDED A BMS AND A GODDAMN BATTERY EQUALIZER WHEN I GOT INTO THIS SHIT.
WHERE WAS I SUPPOSED TO READ IT I JUST WANTED BATTERY POWERED PROPS
REEEEEEEEEEEEEEEEE

>>2859947
Bidirectional or single-directional? I'd generally use something like an INA181, though for bidirectional uses you want to have some intermediate ground rail to reference the output voltage to. You can buffer a voltage divider and use it with a current amp like the INA181, but there are also current amps with two REF inputs, allowing you to tie one high and the other low. Browse TI's website for all of their current sense amplifiers, there are different noise specs and gains and other variances. Read some datasheets, check out some appnotes.

For an extra-high dynamic range, you could use two current amplifiers, maybe with different gains, or maybe with different series sense resistors, and sample the higher gain one whenever it isn't overflowing. It's likely not bad for the higher-gain circuit to hit its output rails, you'd just need to make sure the input voltage never hits its the input rails.

I don't think EMI is generally an issue, you're sensing a really low impedance signal source after all. If noise in the current waveform is an issue, you can add an RC filter between the sense resistor and your current sense amplifier.

>>2860038
Shoulda asked /rcg/.

>>2860038
You don’t need that if you manually charge, or trickle charge and you have access to the individual cells

>>2860038
>I JUST WANTED BATTERY POWERED PROPS
lies, you wanted RECHARGEABLE battery powered props. using AAs or other replaceable batteries is very straightforward.

File: PXL_20231206_095957467.jpg (1.69 MB, 3072x4080)

1.69 MB JPG

>>2860045
True
>>2860043
You KNOW you're never going to have access to those if you want a device with a single access port.

No but seriously for some goddamn reason all the tutorials for using 18650s in props never seemed to mention the quality tradeoffs of different BMSs, let alone what battery balancing was. I've built props for people WRONG because I used a 10¢ BMS with no balancing and I know for a fact they won't get the right battery life out of it.
Worse than that, I bought 18650 cells off Amazon before I knew any better.
Nobody tells you this shit when you're starting out, even when I went asking ChatGPT for pointers it just recommended battery platforms without mentioning any pointers on circuit design.
Most of this is my fault, but I'm still pissed I never came across it in the past. Like if you're trying to make something with integrated rechargeables of course you're going to want to charge all the cells at once, effectively and efficiently.

For my next build I'm going to have batteries in parallel for capacity and I just realized as I was writing this post that I didn't even consider if the active balancer will work across those. FUCK.

>>2860056
NVM I'm retarded the balancing will do fine.

>>2860056
If you're using 18650s, why not just have them removable? That way you can charge them seperately, and even replace them once they've gotten bad. Or put a standard JST balancing plug under a removable cover?

Don't feel too bad however. Milwaukee and DeWalt both don't balance the cells in their 18V tool batteries at all. Milwaukee even have cell balancing circuitry inside their batteries, but it never gets activated.

>>2860080
>Milwaukee and DeWalt both don't balance the cells in their 18V tool batteries at all.

Toy companies desu.

File: 7581CA77-1606-40B2-A7C6-5(...).jpg (251 KB, 1213x634)

251 KB JPG

>>2860045
The same problem affects NiCd, NiMH and alkaline cells. One fucked cell can ruin everything.
In 18650s, due to the higher energy density, it’s more of a critical event though, but the problem is nothing new, even for non-rechargables.

Here’s what our outgoing voltage test board looks like at any particular time… there’s always one or two that aren’t quite up to snuff. Those get sold off in bulk and you buy them from some chinese guy off ebay. The rest are committed to sb&d.

So initially, you’re right, you don’t need the battery balancing, but the cells age unevenly and eventually you will.

If you are making your own packs, provide access to the cells individually and make them removable.
Add some actual value there.

18650s were obviously not designed to be tab welded, they just do that in factories because it’s cheaper.

>>2860056
> parallel
And that’s an even worse idea.
Always do them in series dc/dc convert them and eat the 7% loss.

>>2860084
That argument rings hollow when the majority of working professionals with battery-powered tools use these brands, or equivalent brands like Makita and Bosch. Hilti are meant to be a step above, and maybe some Bosch tools or Metabos or other brands might be similar, but I haven't seen an in-depth teardown.

Though maybe your argument is that corded tools or air-tools are what actual pros use.

>>2860108
Depends on what load you're driving. If it's a single 3V LED, then 1S is probably the way to go. If you have multiple loads like LEDs that you can put in series or parallel, if the power is low enough (less than 10W or so) I'd still go with 1S because the lack of required balancing makes things simpler. But for higher power I'd tend towards 3S, as 9.6-12.6V is great for silicon FET gate driving. Assuming you don't need current balancing circuitry for the LEDs that is. At some point, the current becomes high enough that you want to use more than 3S. Higher power loads are typically designed for more than 12V because of the current issue in the first place, as always you should pick the battery pack voltage that allows you to drive your load directly.

>>2860108
What? Why?
>>2860112
Roughly 2 meters of 144 diodes/meter density 12v ARGB LEDs (WS2815). Ideally it'll run for 1-2 hours? The 3v fog machine I'm running in parallel is negligible in comparison, so a 3S2P or 3S3P configuration is what I was planning on.

>>2860103
Impossible to do removable cells due to space constraints. Once they're sealed in there I need to rely on my circuitry and whatever sensors and gauges I've wired in.

>>2860119
I guess impossible due to space constraints is a bit of a misnomer, but impossible due to form factor all the same. They'll basically be glued in place after I shrink wrap them.
I do have a battery welder and nickel strips ofc.

>>2860119
> why not parallel
For the same kinds of reasons it’s hard to use diodes in parallel. If one cell gets dendrites and partially self-discharges or shorts, it’s a serious fucking problem. It’s also difficult or impossible to charge them unless they are (and remain) perfectly balanced over their lifetime.

Virtually every li-ion battery system out there in the world is a series configuration. It’s that series voltage that drives the design of all downstream components.

Ever wonder why every laptop charger in the world is between 19 and 22 volts? It’s the same cell configuration as all modern cordless power tools—18 V.

File: LIP-500MA-11.1-USB+%2802%(...).jpg (35 KB, 800x800)

35 KB JPG

>>2860119
I'd try to break out a balance charge port if nothing else. 3S balance chargers are pretty common and they use a standard plug, pic related.
Your battery life will depend on the brightness of the LEDs, read the datasheet to check the power consumption for full white, and then calculate the fraction of full white you'll be driving them at on average.

Can you run WS2812s directly off the 3S pack voltage? 9.6V might be a bit too low, let alone 7.5V if you run them that low.

>>2860123
>Virtually every li-ion battery system out there in the world is a series configuration
You're misinformed. You're right that virtually every multi-cell lithium ion battery system has multiple cells in series, but the majority of those also have cells in parallel.

>If one cell gets dendrites and partially self-discharges or shorts, it’s a serious fucking problem.
If you have a protection chip on each cell (e.g. DW01), single cell failures won't cause cascading issues. At least not beyond the thermal cascading failure you'd get with any pack. Any tool battery larger than 2-4Ah will have multiple cells in parallel as well as in series. A lot of series+parallel professional packs don't even bother with individual cell protection.

>Ever wonder why every laptop charger in the world is between 19 and 22 volts?
The key reason they use 18V and 19V and such is because higher voltage means lower current for the same power. Thinner wire, less expensive semiconductors. Same reasons electric cars run at hundreds of volts, and why smartphone fast charging boosts up to 9V or 12V, only for it to be stepped back down again.

>>2860123
> Virtually every li-ion battery system out there in the world is a series configuration

You can basically drop the lithium ion restriction, and just say “every” and leave it at that.
12 V car batteries, for example… a huge proportion of world-wide rechargeable battery capacity of all battery capacity in the world… are six cells of 2.1 V each… In series (obvi)

>>2860123

You are so wrong on every point that it's almost impressive. Literally the only thing you got correct was:

>If one cell gets dendrites and partially self-discharges or shorts, it’s a serious fucking problem.

...with the caveat that it's a serious problem for ANY large pack, regardless of configuration.

>It’s also difficult or impossible to charge them unless they are (and remain) perfectly balanced over their lifetime.
It's literally impossible to have cells in parallel go out of balance. They're required by the laws of physics to all be at the exact same voltage, since their terminals are all connected directly. Series strings of parallel groups can certainly still go out of balance, but that's not an issue for cells within those groups.

>Virtually every li-ion battery system out there in the world is a series configuration.
Both mid-size batteries (where common/inexpensive standard cell formats don't exist) and extremely large batteries (where cells would become to physically large and heavy to move easily) are almost always a series-parallel configuration. Only small and large-ish batteries use single cells in series. Obvious exception here is the dumbfuck Tesla battery with a gorillion 18650s in it.

>Ever wonder why every laptop charger in the world is between 19 and 22 volts? It’s the same cell configuration as all modern cordless power tools—18 V.
No, it's because the 12-30V range is a sweet spot for the power requirements of medium-sized electronics. The power-handling components don't need to be highly rated for either voltage or current, meaning cost-savings on the BOM. I don't even know how you came to the conclusion you did. Why the fuck would you cater your entire electronics assembly to a sub-optimal working voltage when you could just add or subtract a few cells to get the battery into a more useful range?

File: TMC4671 breakout board.png (545 KB, 1450x1370)

545 KB PNG

Finished my first 4-layer PCB. It really should have been at least 6 layers since I was running digital traces on the analog side and vice-versa when I needed to, but I'm trying not to make this a $80 JLC order. Man soldering these will be a pain, the pads are 0.2mm wide. But I already bought the DFN chips, so it's too late to just ask JLC to put them on. We sunken cost fallacy now.

The board is a motor driver test bed with a TMC4671 and TMC6200, designed to be wired to a seperate board with transistors and sense resistors (read: wires) on it. The aim is at least 100A at 24V and 48V, I think it's doable. I've got my king MOSFETs, the GL200N06s, and I'm using ZXGD3006 totem poles after the gate driver stages, which may or may not work.

File: 7F6A3BDB-9021-43C3-BA6B-F(...).jpg (26 KB, 400x299)

26 KB JPG

>>2860133
> since their terminals are all connected directly
Voila. Professor Chargely von Lion’s perfectly balanced cells, ready for customer pick-up. Technically, those are perfectly balanced at 0 V.

>>2859781
Thanks for the resistive panel info.

File: IMG_1606.jpg (68 KB, 1000x822)

68 KB JPG

>>2860123
connecting like-chemistry batteries together in parallel is the exact same as connecting capacitors in parallel: youre effectively just adding the surface areas together.

PSA: Don't mix lead-based solder (even in HASL or trace amounts on tip) with tin-bismuth solder.

So I have been shopping for solder paste, and since it is so much cheaper on aliexpress, I decided to check there. But unfortunately, its becoming more and more difficult to find tin-lead solder there. In the past you could have searched for "sn63" or "181 paste" or similar "kosher pork" and find what you want. But unfortunately it is no longer the case since tin-bismuth alloys are becoming more popular, and there are some alloys that melt at 181C or there abouts just like ordinary tin-lead solder. Great right? RoHS and finally low reflow temperatures with shiny joints...
Not so fast. There is a major flaw with tin-bismuth-whatever stuff. It is not compatible with lead. It forms a tin-bismuth-lead alloy which melts at insanely low temperature (like 98C) and this alloy makes grain structure of solder joint really big, which means shit aint holding shit.
And you don't need much lead for this to happen. Tin-lead HASL PCB surface treatment is enough to cause shit joints.

In other words, if you're gonna use tin-bismuth stuff, you should make sure that all your soldering tips are lead-free (as in not contaminated with lead from ordinary tin solder), and PCBs are lead-free (ENIG, lead-free HASL, etc.).
If your PCB has tin-lead HASL, you should use either tin-lead solder (obviously the best choice), or classic lead-free alloys like SAC305, or SAC0307.
And if you struggle with classical lead-free shit when hand-soldering, I suggest throwing away your old iron and getting something modern like T12 or whatever other TS100 with integrated heater.

Anyway. Fuck EU. Your shit is ruining it for everyone.

>>2860133
> Only small and large-ish batteries use single cells in series
I’m just a regular person, I look around:
My mouse uses 2 AA cells in series.
My flashlight uses 2 C cells in series.
My mag light flashlight uses 3 D cells in series.
My 12 V drill uses 3 18650 cells in series arranged in a little pyramid.
My alarm clock uses a 9 V battery with 6 cells in series in it.
My radio uses 2 AA cells in series.
My emergency motion activated light uses 4 AA cells in series.
My booklight uses two cr2032 cells in series.
Apparently my car battery has cells inside it in series. Makes sense, I see like 6 individual caps on it.

I can’t wait to see your much much bigger list of all the purely parallel cell connections. I’d also like to know what planet you’re from.

>>2856546
>is it possible to retrofit a (tft) lcd panel from cold cathode lamps to led strips?
Yes, there are kits sold on aliexpress.

>>2860165
> exact same
Not even in the same realm.
For one, a battery is what we call an “active” element or component. A capacitor is a “passive” device.
If you were a biologist you’d be making a statement about the plant and animal kingdoms being the exact same, and hilarity would ensue.

>>2860178
> or trace amounts on tip
> like 98C
The 43/43 sn pb bi is a low of 144C so it depends crucially on the proportions. Just flood it with your 60/40 and you’ll probably be fine.
Don’t worry, NASA, the DoD, and Boeing still use 60/40 SnPb to keep the world safe for democracy.

> buy a new soldering iron every time some internet rando encounters lead-free solder
No thanks. But that reminds me, my TS200 needs a firmware update before I can download the HASL DLC from the alibaba play store.

>>2860184
he's not wrong. the line between cap and battery blurs as energy density increases.

>>2860191
> line blurs
Why do I get the feeling I’m about to be hit up with some infinite energy >100% efficiency circuits?

>>2860193
because you don't know how capacitors or batteries work. read a book.
electrolytic caps are just really shitty batteries that are always discharged.

>>2860178
Oh shit, really? I used that bismuth shit for a while, wonder if I'll have any failures down the road. Hope I didn't use it on those boards I sold to one guy, because those boards were definitely 60/40 HASL.

>>2860179
All of those except the car fall under the "small" category. I'd guess anything under 3Ah total capacity, though larger cells like 26650 go higher. Above that, you start to put cells in parallel. Like in an 8Ah tool battery. The car is ~40Ah, which I guess is "large-ish", but car batteries are a bit of a unique case since they're not really used for their energy storage but their power capability. Either way you're trying to use anecdotal evidence to disprove a general trend, not a winning battle, that or you're being a semantic moron by literally interpreting the other anon's "Only" to be an absolute.

The point the anon is making is that you commonly find individual cells less than 3Ah (cylindrical, coin, pouch cells), and between 50Ah and 300Ah (prismatic cells). Anything between those ranges is probably composed of individual cells in parallel (unless it's a lead acid), and anything above those ranges is too large for an individual cell to be made from it, like for EVs and grid storage.

>>2860184
Quit being a contrarian. His point is that the capacities add up, the surface area adds up, everything adds up, whether it's in the same casing or not. If your failure rate due to dendrites are a matter of probability based on surface area, it doesn't matter how that surface area is distributed. Big "if" though.

Let me reiterate:
REAL BATTERY PACKS HAVE PARALLEL CELLS ALL THE FUCKING TIME

>>2860199
Nah this isn't really correct. While some supercapacitors store energy both electrostatically and electrochemically, all conventional electrolytic capacitors store energy solely in their electrostatic field. There is no chemical reaction in them at all, besides the irreversible reaction that grows the aluminium oxide layer in the first place.

File: PXL_20240727_065027421.jpg (2.66 MB, 4080x3072)

2.66 MB JPG

>>2860123
Yeah but don't the parallel cells self balance? Like
>>2860125
>>2860130
>>2860133
>>2860165
All say? I double checked with ChatGPT and it said that it should self balance in parallel as well.
You make a good point mentioning the potential for frays but I'm going to mitigate the possibility as much as possible with liquid E-tape, shrink
wrap, and hot glue as necessary.

Also >>2860103 I'm not supposed to tab weld these? I know any heat in them is bad but I need permanent connections since this project will not only heavily involve motion and sudden shocks (GY-521 inside) but also a very cramped housing

Also >>2860125 why would I need a balance charger if I'm already using a BMS? The nicer DALY BMS I ordered has a charging line and the 18650s are protected button tops that I got from lithium ion wholesale (the store brand high capacity ones)
Historically I've been able to run 12v LEDs on 9v and lower but those weren't ARGB fwiw. And the BMS comes with a load cutoff at 2.7v so I'm hoping I see the drop before then and I can just switch off the load before charging. I'm not entirely sure if I should include a load shutoff circuit when charging but if I've gone this far I probably should just to be safe.

Picrel, my last build (no longer working on batteries, work when plugged in, but it lasted a day). I'm doing cooler stuff with the prop design this time too, my Makerspace just got a belt printer.

>>2860236
Stop asking chatGPT for anything engineering besides programming. Even programming is a bad idea.

>why would I need a balance charger if I'm already using a BMS
Common BMSs protect against undervoltage and overvoltage on a cell-by-cell basis, but they don't act as balance chargers. You can't just feed a constant-current into an imbalanced pack with a BMS and expect it to get fully charged. You won't damage it, but you won't get a good charge out of it. They also protect against overcurrent, but that doesn't mean you can feed a constant voltage and expect a nice charge curve.
Some BMSs do have actual balancing circuits, but AFAIK the cheapy alibay units don't. The Daly ones are fancier so they might, but I was assuming you'd be using the same BMSs as the ones you put in the older units you made. If the Daly BMSs have a charge circuit, that should be fine then. To be nice, I'd mark the input voltage range and current requirement (and plug polarity) somewhere on the unit.

Using a balance plug and a non-charging BMS is cheaper per unit, but more expensive per charger. The advantage of using an external charger is it gives you more freedom in choosing charge current, an internal charger will want no less than 2A @15V for example, making it more expensive to replace.

>>2860382
I think I'm misunderstanding you (or retarded)
I thought the whole point of a BMS, besides protection, was to evenly distribute a charge current. I understand this is imperfect because individual cell state and charge can vary, but I figured the latter would be fixed by an active balancer anyway.
The "charger" I was using was a dirt cheap 3A wall outlet AC-DC 12v barrel jack converter, and I was just piping the output from that straight to P+ and P- on the circuit with the load turned off. Is there a nicer approach I can take? The DALY BMS supports 20A so I was considering a higher amp charge current, but I've never seen mention of a charge circuit separate from the BMS itself.

File: daly_bms_wiring_diagram.jpg (204 KB, 1450x1425)

204 KB JPG

>>2860382
>>2860393
Picrel. I know the active balancer just goes in parallel with this along the batteries, but would the balanced charger go in parallel with that? Also, since there's an active balancer in parallel already, won't that balancer just balance the batteries as they charge?

File: speaker-parts.png (106 KB, 600x239)

106 KB PNG

I recently changed the magnets for a few bucks (from ceramic to Alnico 5) on some garbo chinese $5 guitar pickups, and now they sound as good as Fender pickups that cost thirty times more. This got me curious: what makes a speaker a good speaker? Why do some speakers cost $20 while others are $150? What makes them better? Let’s say it’s a 60W, 8-ohm speaker. Sure, one has a brand name while the other doesn’t, one is made in sweatshops etc. But what parts of the speaker are different? Which part makes the most difference?
One more example. In a microphone, the capsule makes the most significant difference in sound quality. High-end mics that cost several hundred or thousands dollars still use capsules that only cost $20 to $40.

You won't believe but absolutely everything in electronics is a tradeoff. Everything. I have a battery operated uC board in a deep sleep mode that is only drawing 800nA. I also need to power it from a USB, so I was going to install a protection schottky. But then I decided to go a fancy route and minimize the voltage drop and used a cool progressive zero drop diode chip. The caveat? Its quiescent current is a whopping 7uA! Ironically it is actually pretty damn low, but it still 10 times higher that the uC current! Fail. So I went with a jelly bean schottky. And the voltage drop is only like 0.1v since the current is extremely low, and it is quite acceptable. Conclusion: don't complicate things! Every addition is going to affect something else!

File: Dek1845151362073006080.png (770 KB, 1239x771)

770 KB PNG

>>2860216
Depends on the thermal cycling I guess.

>>2860186
>The 43/43 sn pb bi is a low of 144C
It forms different alloy, which does melt at 96C.
> Just flood it with your 60/40 and you’ll probably be fine.
That would work. I think. But problem is that getting 63/37 or whatever other tin-lead-silver solder paste is becoming more and more difficult.
Aliexpress rules (or is it because of RoHS and EU?) don't allow sellers to sell leaded solder alloys, so you have to keep on searching shit that has "Sn63" where "Pb37" part blurred out to get real deal which is annoying. And you might get tin-bismuth-silver alloy instead because apparently China doesn't know how to get rid of bismuth (and its cheaper than tin-lead).
Which is why I decided that fuck that shit, I'd rather get SAC305 or SAC0307 which is compatible with HASL lead and tin-lead solder in general.

Another issue is that a lot of electronics are switching to tin-bismuth shit, and most people don't even bother to check at which temperature solder on the board melts, and then wonder where did all resistors go on the other side of PCB when they replace some BGA IC or something.
>No thanks.
900 style irons suck at classical SAC lead-free as they don't get hot enough, or idk, too slow to react to temperature change, no idea. I had to set it to max temp in order to solder lead-free shit, which leads to lifted pads and burnt tips.
T12 and similar where heater is integrated into the tip are fine, and can solder lead-free shit at reasonable 330-340C. You don't need to buy anything with firmware.


Anyway, now idk what do I do with SnBi solder paste I've ordered. I wanted to solder LEDs with it to HASL lead board, but I had some sort of bad feeling for it, decided to google and realized it won't work (especially since LEDs can get real hot, like 100C hot)
Maybe its finally time to switch to RoHS solder. Bleh. Or keep using it to desolder shit like connectors.
Fucking evil green paste.

>>2860434
a lot of aspect have influence on speaker character. mechanical or magnetic saturation distortion, cone weight, stiffness and suspension, magnet size/type (strength), coil size, parasitic capacitance, cabinet stiffness, cabinet weight, air volume, and a bunch more....

its basically a very complex filter and nonlinear transfer function.

SpectreSoundStudio, Johan Segeborn and Jim Lill are good YuToob channels

File: AE529A32-D831-4226-91F8-8(...).jpg (165 KB, 668x822)

165 KB JPG

>>2859372
> hey man, that’s a DRIVER
>>2860434
> no, it’s not.

>>2860236
> self balance?

Yeah, technically, yes they will self balance.
But a typiclBMS can no longer “see” individual cell voltages and they age unevenly.

Individually protected cells exist but, those are a failure point and end up killing the whole pack (but preventing a potential disaster). Never seen in commercial products.

Imagine you have some regular alkaline cells and you hook a new D cell up to an old AAA cell. The AAA cell will explode. Similar situations occur when Li-Ion cells age unevenly which is why paralleling is generally not a good idea.

>>2860165
And yes, capacitor guy, a bad capacitor can kill other capacitors when it leaks it’s electrolyte as it ages, either by shorting out and passing DC or by going open and unable to quench AC peaks in parallel…and also destroying PCBs when they go batteries do.

>>2860395
Thanks for posting that diagram. As we can see, the BMS senses the voltage on (ideally) each and every cell, and if they’re all in parallel it simply can’t do that. There’s no yeahbuts here.

Of course you could design a BMS that works with parallel cells… each cell protection device using SPI to talk to every other one to balance the delivered current evenly through mosfets, but even small packs have no problem developing 50 A. Even an individual cell 18650 can output 10 A peak initially.

The series configuration is not perfect, but it is widely understood to have a higher degree of intrinsic safety, and this hasn’t changed for over a hundred years.

If you do have a series/parallel connection do it at the *ends* of long series connections… never parallel individual cells so you get the intrinsic safety of the series connections.

E.g. lead-acid cells in submarines which typically use 126 cells in series, but can operate them in parallel in an emergency, but they don’t have to.. they can be using one bank and charging the other if they aren’t in, say pursuit or escape.

>>2860434
>>2860475
My theory is everything sounds the same and the filter is in your brain and you improve the sound by dropping more money at the expensive stuff which makes you feel better. There's no measurable metric that could prove that expensive speakers are "better". Or rather you could measure and compare transfer functions and all of them will look different and then you arbitrarily assign a higher value to one of them based on some arbitrary criteria: what other people say, the cost of the upgrade, or that it sounds similar to something you are used too. Remember those simple tube guitar amplifiers from the 1950's - 1960's with absolutely horrible specs, high distortion, voltage sagging etc. The have a cult like following because someone long time ago decided that's the "proper sound". It is like arguing about what cottage cheese brand is the best.

>>2855842
i feel like a retard but i'm struggling to understand this. i have a +/- 15v power rail. i want to use a potentiometer to output -1v to +5v. conceptually i know how to do this with a combination of fixed resistors and using the potentiometers to act as variable resistors to bias current flow in either direction, but i'm really struggling with finding the correct values such that "minimum" is -1v and "maximum" is +5v.

>>2860495
The whole principle of how resistor dividers work, is in how much voltage each resistor drops. Assuming an arbitrary 1mA through the resistor chain; what fixed resistor would you need to drop 10V? What resistor (the pot) to drop the next 6V? And what resistor to drop the final 14V?
Once you have nominal values, you can scale them to any arbitrary current/resistances you like.

>>2860515
doesn't the bottom resistor have to drop a variable amount of voltage, dependent on how much voltage the potentiometer drops?

>>2860179
>I’m just a regular person

That's probably why your reading comprehension is so bad. You listed a bunch of things that are small batteries that use series strings of cell groups where each cell group is just a single cell. In other words, helping to prove the point that was being made. Honestly, the most out-there thing you listed was:

>My booklight uses two cr2032 cells in series.

...which makes me want to find and strangle the nearest industrial designer.

>>2860484
>But a typiclBMS can no longer “see” individual cell voltages and they age unevenly.

This HAS to be a troll, there's no way someone could be this confident in being wrong, right? I want you to think really hard about this, an explain to me how, hypothetically, two cells connected in parallel can have different voltages. Their terminals are connected. They are the same node. They CANNOT have different voltages, barring some unbelievably small difference arising from the parasitic properties of real materials..

>The series configuration is not perfect, but it is widely understood to have a higher degree of intrinsic safety,

No, it doesn't, how the fuck are you coming up with this shit? Series configurations of cells chemistries that cannot handle "top off" charge cycles are 90% of the reason battery management systems exist in the first place. "Intrinsic safety" has absolutely nothing to do with anything here, because putting cells (or, more fundamentally, voltage sources in general) in series is the ONLY way of getting a higher voltage out of a battery without active electronics.

>If you do have a series/parallel connection do it at the *ends* of long series connections… never parallel individual cells so you get the intrinsic safety of the series connections.

Just stop posting. There is not a single commercial/industrial pack made like this, barring some insane one-off module that I just made up to cover my ass in case there was some weird fringe case where it made sense

File: resistance is futile.png (19 KB, 620x992)

19 KB PNG

>>2860532
Yes, but you're overthinking this, I think. You don't actually need to think about current flow directly, just the relative values of each resistor.

Since your voltage source is 30V rail-to-rail, you want the top resistor to drop 10V (to get 5V at the top end of the potentiometer's wiper range) and the bottom resistor to drop 14V ( to get -1V at the other end). That leaves 6V to be dropped by the potentiometer itself. There are an infinite number of solutions to this problem, so you need to pick at least one value arbitrarily. 10k pots are plenty common, so let's go with that.

You need the ratio between the resistance of your chosen "base" resistor and how much voltage it drops. In this case, 6V is being dropped by 10kΩ. 10k / 6 = 1.667k. You can simply multiply the other required drops by this amount to get the required resistances.

10 * 1.667k = 16.67k for the top resistor.
14 * 1.667k = 23.34k for the bottom one.

You can scale this method up to an arbitrary number of steps in your divider ladder, of course, but these numbers get you what you're looking for here.

>>2860547 (You)
Also remember that this is assuming that whatever the wiper of the potentiometer is connected to has a very large resistance compared to the resistors in the divider network. If it doesn't, the additional current it draws through the network will throw everything off. You can imagine this in the extremes of either infinite output resistance having no effect on the voltage at the wiper terminal, and zero output resistance completely negating the effect of the divider network.

If you need any significant amount of current out of the terminal and cannot afford to be wasting power heating up low-value resistors in the divider network, you'll want to use a buffer at the output of the potentiometer's wiper terminal.

>>2860542
> two cells connected in parallel can have different voltages?
They can’t, but they can have vastly different internal resistances.

>>2860547
>>2860549
THANK YOU, that makes sense and i was definitely overthinking this.

File: B7B9A197-77F9-4A69-93D6-A(...).jpg (802 KB, 1800x1055)

802 KB JPG

>>2860542
> is not a single commercial/industrial pack made like this
Except they are all made like this. What planet did you say you were from? :-) roflmao….

The way you get higher currents is bigger cells.

That’s why things like “D” cells exist, otherwise, everyone would just parallel a bunch of “AAA” cells together.

Guess how many commercial 18650 holders wire them in parallel? Zero. The answer is zero.

File: tegaki.png (8 KB, 400x400)

8 KB PNG

>>2860393
>>2860395
I might be retarded, but I was under the impression the cheapest BMSs didn't do balancing, just out of balance protection.

>>2860495
Pic related. If you give a potentiometer 5V and -1V as its rails, its output will naturally vary between those values. If you don't have access to those voltage rails, then it's a matter of adding resistors above and/or below the potentiometer such that the voltages at either end of the potentiometer are 5V and -1V. You know the voltages, you set some initial resistance for your potentiometer, then you can use the current flowing through the potentiometer to calculate the required resistance for the other one or two resistors.

>>2860564
Any tool battery larger than 3-5Ah uses parallel cells. Any car made with cylindrical cells uses parallel cells. If you don't connect the parallel cells together, then you need a seperate BMS for each one to keep them in balance. This may be done for cars, but I don't believe it's done for tool batteries. Tool batteries don't even have balancing in the first place, lmao.

File: 8DE97515-AA6B-43AB-A68C-A(...).jpg (567 KB, 939x1202)

567 KB JPG

>>2860569
> Any car made with cylindrical cells uses parallel cells
It would be more accurate to say that it uses parallel batteries.
The tesla, for example, uses 6S (6 series) cells per battery, and the BQ76PL536A-Q1 to manage each battery. At no time are individual cell’s voltages masked by parallelizing them at the cell level.

What happens if 3.3v is fed into the output a 3.3v voltage regulator that is turned off (no input voltage)? Do I need a protection diode?

>>2860639
> protection diode?
It’s probably fine, but if you are worried about it, it doesn’t hurt to add one, two or three diodes in the usual places.

CH340 is driving me nuts. It is probably me since it cannot be *that* bad. Just too many RX errors. TX is fine. I was wondering if it is more stable if powered with 5v vs 3.3v. I am also using a prototype board so I am hoping that once I make a PCB it will eliminate some noise.

>>2860630
Yeah I guess that's a good distinction to make. But man it must have a lot of those chips, why wouldn't they use a 12S or 20S BMS chip instead? Or larger cells?
>individual cell’s voltages masked
Two cells in parallel have the same voltage. I'd say that the condition / health / effective capacity is partially masked; it's averaged between all the parallel cells. If your parallel cell count is low, you can infer the condition of the worst cell in a parallel bunch by assuming all the other cells in that lot are as good as the pack average. This isn't easy if you use highly parallelised packs since the difference is going to be small. I don't know if any BMS chips actually do this.

>>2860639
Depends on the regulator. Some definitely don't like it, others are fine (usually because of an internal diode). If the datasheet or appnote doesn't say (it usually does), I'd add a diode.

>>2860661
Are you using a CH340 with a crystal oscillator? If not, there's your problem. If so, it could be that your crystal is shit. I don't know if that's a possible symptom of picking the wrong size of crystal bypass caps though.

>>2860558
>They can’t, but they can have vastly different internal resistances

While true, this is very rarely the case unless something went wrong with one of the cells. Even if it does, it's not necessarily a huge issue. The greater internal resistance induces greater voltage drop under load, which causes the battery to supply less current in direct proportion to how badly it's doing. This can cause heating issues in extreme cases, but, again...that shouldn't regularly happen if the cells were all of the same model and age to begin with.

>>2860564
>Except they are all made like this.

Can you even keep track of what you yourself are saying? Cell capacity has nothing do with whether it's a good idea to make a battery of parallel groups of series cells or series groups of parallel cells.

>Guess how many commercial 18650 holders wire them in parallel? Zero. The answer is zero.

"Large-ish" commercial batteries don't use 18650s in general, because that's retarded (see again: Tesla). I had to think for a minute what you meant by "18650 holder" because my first thought was of the [pic related] type. They don't "wire them in parallel" or at all. You will note in the example, by the way, that the tabs are being welded on such that the cell groups are in parallel. Because that's the non-retarded way to do it. The only reason you would ever use parallel groups of series cells is if you basically needed more than one battery for redundancy, because that's what you end up with at that point.

The tiny little bring-your-own-cell USB charger/flashlight battery packs use series-only batteries because no shit they do. Except, oh, wait, even THEY don't always do that. I actually have a few of something like these:

https://www.ebay.com/itm/354166853534

...that I used to use for my cameras. And, oh, hey, look. All the cells are in parallel because this is one of the one odd cases where the extra voltage of series cells isn't wanted but the extra capacity of multiple cells is.

File: batt.png (664 KB, 598x597)

664 KB PNG

>>2860729
>[pic related]

Maybe I was the retard the whole time...

(j/k it's still you lmao)

>>2860715
> why not 12S or 20S?
Tesla used commodity parts in the early days before they were big enough to demand (or make) proprietary bulk products from suppliers. Using 18650s in the first place was indeed, retarded. In the model Y they are using those enormous 46800 cells and less parallelism. They’re still retarded though.

> you mean condition/health/effective capacity
Exactly. By “voltage” I was thinking of the disconnected open-circuit voltage. Which is obviously completely masked (or, eliminated/impossible) by the parallel connection. BMSs actually do all kinds of crazy things like send in pulses and analyze the voltage decay rate and whatnot, and have under-volt-trickle-charge-with-pulsating-DC-recovery mode, along with memory/history of the individual battery characteristics.

File: 251C3B92-0BA3-4F2C-92B2-C(...).jpg (392 KB, 911x725)

392 KB JPG

>>2860729
> this is very rarely the case
The tesla model X has over 8,000 individual cells.
I think it’s time for a statistics lesson. Or, we could just ask the Samsung Galaxy Note 7 team.

Personally, for DIY purposes, I use individually protected 18650s that have the protection in the top of the cell for some reduced capacity and increased cell size. Cheap insurance.

>>2860730
It only took you 27 hours to find those two counter examples. But they’re good ones!
Everyone knows “ANN POWER” and “Wewel-Tech” — household names. Wewel-tech even sold 6 of them so far. I bet they’re all gonna retire on that income.
And they’re both well-respected, every-day household names.
Just yesterday, I was on a blind date and talking about Wewel-Tech’s innovative parallel cell technology. She maced me, and ran away but I don’t know why. Weird.

File: owl no its fascinating do(...).jpg (42 KB, 500x332)

42 KB JPG

>>2860741
>The tesla model X has over 8,000 individual cells.
>I think it’s time for a statistics lesson

Tesla has sold north of 4,000,000 cars as of LAST year. There have been 232 confirmed fires, and that's with, again, some of them sporting that ridiculous battery. Even assuming 90% of outright cell failures weren't that level catastrophic, that's still pretty rare, even before taking into consideration that these are cells exposed to frequent vibration and high charge/discharge rates.

Beyond that, I mean...it's an electrochemical cell that's known to be volatile when it fails, and manufacturing isn't perfect. Some of them are going to fail hard, IDK what to tell you. That's half the reason a modern BMS keeps such close tabs on battery conditions 24/7. Until something better and less likely to get violent when things go wrong, not much else you can do.

>>2860748
Oh boy, he's resorted to base-level mockery and exaggeration to the point of untruth. The meme macros come next.

My standards for engagement are low, but this was already a waste of time. I cannot even be arsed past this point.

ANN POWER? More like Unpowered.
Wewel-Tech? More like Memel Tech

File: tesla_fuses.jpg (69 KB, 658x288)

69 KB JPG

>TRUST ME BRO EVERY SINGLE COMMERCIAL BATTERY PACK HAS A CELL-LEVEL BMS AND NO CELLS IN PARALLEL

This dumpster fire is worse than Reddit.

ok you're both shouting over each other.
stop claiming absolutes or claiming that the other person said absolutes they never said, you're both guilty of both

i want to buy an electric car with a dead battery and swap it to use (super) capacitors instead, wouldn't that be fun?