/ohm/ - electronics general, salvaged edition

[Anonymous No.1753417]

ye olde thread: >>1748210

>RULES
0. Electrics ≠ electronics. Appliances/mains/sparky stuff to /qtddtot/ or /sqt/. PC assembly >>>/g/
1. Search web first. Re-read all documentation/data-sheets related to your components/circuits. THEN ask. Show your work.
2. Pics > 1000 words. Post relevant schematic/picture/sketch with all part numbers/values/etc when asking for help. Focus/lighting counts.
2.5. State your skill level if asking an open-ended question.
3. Read posts fully. Solve more problems than you create.
4. /ohm/ is an anonymous, non-smoking general.

>I'm new to electronics. Where to get started?
It is an art/science of applying principles to requirements.
Find problem, learn principles, design and verify solution, build, test, post results, repeat

>Project ideas:
http://adafruit.com
http://instructables.com/tag/type-id/category-technology/
http://makezine.com/category/electronics/

>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

>Design/verification tools:
LTSpice
MicroCap
falstad.com/circuit/circuitjs.html
NI Multisim
CircuitLab
iCircuit for Macs
KiCAD (PCB layout software, v5+ recommended)

>Components/equipment:
Mouser, Digi-Key, Arrow, Newark, LCSC (global)
RS Components (Europe)
eBay/AliExpress sellers, for component assortments/sample kits (caveat emptor)
Local independent electronics distributors
ladyada.net/library/procure/hobbyist.html

>Related YouTube channels:
mjlorton
w2aew
jkgamm041
eevblog
EcProjects
greatscottlab
Photonvids
sdgelectronics
BigClive

>Li+/LiPo batteries
Read this first: http://www.elteconline.com/download/pdf/SAFT-RIC-LI-ION-Safety-Recommendations.pdf
>headphone jack noise
Look up "ground loop isolator".
>I have junk, what do?
Get rid of it.

[Anonymous No.1753434]

>>1753097
no, im using a proper 7812, just wanted a security measure if the regulator fails.

[Anonymous No.1753435]

>>1753434
meant safety.

[Anonymous No.1753437]

Can anyone tell me what the resistor color code on these reads? I can't even figure out what colors they are supposed to be.

They were supposed to be 10K but I think they might be 15K. Problem is I can't see well enough to read them and have to trust the shop.

[Anonymous No.1753439]

>>1753437
ive never used these dumb resistors because theyre literally unreadable.

[Anonymous No.1753441]

My masterpiece if finished, time to make a PCB

[Anonymous No.1753442]

>>1753441
What does Q1 do?

[Anonymous No.1753443]

>>1753437
10k with 1% tolerance

[Anonymous No.1753444]

>>1753442
U3 is a smol boy and can't handle all that BBC load it will be switching, so instead it acts as a driver for the Q1 big guy which can handle much more abuse

[Anonymous No.1753452]

Need to replace STK2230 for my amplifier. I was just wondering what is that code on upper right on these IC's. Mine is 7F17 and pictures from sellers have OC14, 2CET, SA03, B9B15 etc. Is this manufacturer related thing or what? Those look all the same but just that code is different.

[Anonymous No.1753459]

>>1753443
Really? Thanks! What colors are the bands?

[Anonymous No.1753461]

>>1753459
brown black black red brown

[Anonymous No.1753474]

I started a new job this week and been getting involved with programmable logic. Our company uses WinCUPL for our IDE but we all think it sucks dick. My boss said I’d be the office hero if I could find a program that’s either better or outputs a logic circuit diagram with a .jed file. Any of you know of anything like this? Would make trouble shooting way easier

[Anonymous No.1753480]

>>1753461
Ah . . . that's the thing. In person the two "black" ones are different colors. :-(

One is much lighter and looks greenish.

[Anonymous No.1753481]

>>1753417
What is the intuition behind the convolution when the output is determined by convolving the input with a transfer function? Why can't we just multiply them element by element beginning from t0 of the input signal. Every textbook just says, ok lets apply convolution. Why? Things like "the tail of the input sequence affects the output the most" say nothing to me.

[Anonymous No.1753534]

How do I get started learning integrated circuits?

I want to start doing more interesting projects without resorting to Arduino.

[Anonymous No.1753552]

>get Harbor Freight multimeter because cheap
>see a how-to-calibrate video on youtube from 2015
>it has a replaceable 10-amp fuse and calibration potentiometer
>open my meter up
>Soldered in 5-amp fuse and a hole where the potentiometer should be
Status: CHINKED

[Anonymous No.1753568]

>>1753534
Inputs, outputs. Pull down inputs to ground. Read the datasheet.
Much like digital solutions, they can be very powerful, accurate and compact, but the tradeoff is 20-200 pages worth of manual to trawl through.
Usually there is a section with recommended layouts, component values, applications and such. Sometimes they work, sometimes they don't.
Use star grounding/star power, especially for amplifiers and digital chips.
Good luck.

[Anonymous No.1753574]

>>1753459
At this point, you should probably look into getting a $5 DMM.

[Anonymous No.1753581]

>>1753534

You have a 555 timer there. Everything is relative. Many of the 555 timer uses can be substituted with better (in some ways) circuits made with one or two transistors.

[Anonymous No.1753584]

>>1753581
>one or two transistors
start with one

[Anonymous No.1753589]

>>1753347
see the datasheet. this reputable manufacturer's products are specified for rms https://www.littelfuse.com/~/media/electronics/datasheets/varistors/littelfuse_varistor_la_datasheet.pdf.pdf

>>1753401
lever-type terminal blocks are based, but not stumble-over easy to find

>>1753434
why would the regulator fail? keep it cool and keep the input voltage within the regulator's spec, and it won't

>>1753437
>1002 1
10k 1%. unless the first one is actually grey in which case it's 18k 1%, in which case I would kek

>>1753474
>using PALs
>2020
y tho, when there are CPLDs and FPGAs that are probably cheaper and faster

>>1753534
read databooks to get a sense of what's out there
read cookbooks to learn how to combine them with other parts, for example, the Op Amp Cookbook and the 555 Timer Cookbook
read random electronics blogs
read some more
then see the OP for instructions under "I'm new to electronics" and repeat until your power level is sufficient

>>1753568
>star grounding
fuck that shit, just use a ground plane unless you're throwing amperes around or counting single microvolts. split grounds deployed carelessly are just asking for trouble

[Anonymous No.1753597]

>>1753589
>just use a ground plane
Fuck you, I would if I could. How else am I supposed to ground these bastards?

[Anonymous No.1753598]

>>1753597
ohhhhh. star ground is good

[Anonymous No.1753655]

This might be a stupid question but it's been bugging me for a while. Can a ldo regulator replace or perform better than a voltage reference?
I'm looking at my parts for a project and my reference voltage has a 48uV rms where as my regulator has a 300nV rms both at around 10hz-10khz.

[Anonymous No.1753682]

>>1753655
check the psrr in addition to noise. also check the temperature and load range over which that's measured. some good ldos probably are better than some shit references.

[Anonymous No.1753687]

Is there any way to connect/make an arduino/raspie powered cassette tape player? Cant find anything on the internet

[Anonymous No.1753689]

>>1753687
>Cant find anything on the internet
anon you have the chance to become a trailblazer, be featured on hackaday, get all the bitches at the next gadgetcon. Think about it.

[Anonymous No.1753694]

>>1753689
Opportunity is wasted on me since I'm a beginner to electronics. I dont know much about this stuff, all I know how to do is what the arduino book tells me

[Anonymous No.1753700]

>>1753694
idk about you, but for me making projects is the best way to learn.
Think about what you want to make: some sort of jukebox where the user can select which tape to play from a screen, or maybe remotely? And then the electronics will operate the actuators to get the selected tape to play?
Break it down in big blocks. Screen, web interface (?), microcontroller/SBC, electronics (actuators, sensors), tape player. Now break it down into smaller blocks, and keep doing it until you hit a small enough block that you can build or at least look up.

[Anonymous No.1753701]

>>1753694
this is a low-shitposting thread. wait for an answer on /mcg/ or ask reddit

[Anonymous No.1753702]

>>1753701
>this is a low-shitposting thread
sneed

[Anonymous No.1753706]

>>1753700
See what I mean by "beginner" is that I dont really understand what you just said. Maybe I need electronics explained to me like a retard

[Anonymous No.1753711]

>>1753417
>>1750645
Got a KD3005D 30V 5A linear DC supply 2 years ago for 85 bucks, replacing a cheaper $55 "Lavolta" which came dead on arrival. Haven't used it often, but it seems to work well and had good forum reviews stating the components are of high quality. If you aren't planning on using it for something demanding like electrolysis/plating, quality isn't as important, it should either work or it doesn't.

[Anonymous No.1753713]

>>1753706
then why tf are you even trying to do

[Anonymous No.1753714]

>>1753713
Why tf are your grammar even bad?

[Anonymous No.1753725]

>>1753002
This seems to be a relevant solution.
https://www.instructables.com/id/My-protection-circuit/

[Anonymous No.1753726]

>>1753700
Well this, except it's all dead-simple except for the tape player part. AFAIK cassette tapes are analogue audio, so there's not really much for an arduino to do anyhow, except for handling remote commands. Just chuck an amplifier circuit, some outputs. maybe a mute button, and some analog switches so you can control it via a remote or BT. Allowing it to broadcast audio thorough BT or even write to a cassette through BT would be neat, but even with an ESP32 that might prove to be a pain to design.

The cassette player itself would basically be a module ripped out of a car stereo, since it's a heavily mechanical device and there's no need to reinvent the wheel.

>>1753706
It sounds like arduinos aren't really the best place to start, you'd be better off learning analog electronics for now.

[Anonymous No.1753731]

>>1753725
A TL431 would waste much less power than that zener, not to mention a dedicated Li-ion protection or power-bank circuit.

[Anonymous No.1753732]

>>1753726
Thank you for explaining the cassette thing. I wanted to make a pip boy/wrist mounted information device thing and include a tape player. In essence, what you're saying is that I'd really only need to connect a speaker and a switch, right? If so, then I'd just connect it to a speaker I'd already have in the pip boy that I use for other programs. Is there anywhere I can get a bare/minimal cassette player module?

[Anonymous No.1753736]

>>1753731
>actually using the correct component
I demand to make everything out of jellybeans reeeeeeeeeeeeeeee
t.the guy you should get a load of

>>1753732
>cassette
y tho, in current century people use solid state storage to record/play back audio. look up ISD ChipCorder on instructables for gorillions of example circuits
if you'd just stated the much more defensible pip-boy part in the first place you would have received a lot less grief

[Anonymous No.1753742]

>>1753732
I assume automotive sets will be some of the smallest ones you can get, but (knockoff) walkmans would probably be even smaller. The automotive insertion ones are probably a fair bit bulkier than the clamshell-style ones, but the clamshell ones require you to work around the particular opening mechanism.

Either way you'll need to put a lot of work into designing the case for it, and it would likely end up too big for a wrist-mounted device. I'd opt for getting a (knockoff) walkman, 3D-printing a new case/decorating the old case such that it fits the atompunk aesthetic, slapping that on your waist then running some nice stiff coax cables from it to your pip-boy.

The other option is to ditch the cassette entirely and go for another method of audio storage. Any sort of memory card would work fine, if you wanted to be a little retro you could go with an older sort (like Compact Flash) and just have the card be an SD-card adapter. There might also be a miniature optical disc format, or some tiny sort of hot-swappable HDD. Data storage tapes might come in small enough sizes, but all tape cassettes require some sort of mechanical system to read them, so it's likely not going to be terribly compact, even if they are cheaper than HDDs (because they have all those mechanical parts built into their case).

Another retro method would be to store audio on CDIP EPROMs and plug them into a ZIF socket on your pip-boy. CDIPs are cool.

[Anonymous No.1753749]

>>1753742
My thinking is that, if the mechanical workings of it can fit into a small, portable box (walkman or other), I can fit it in a pip-boy case, or design the case specifically to fit the components exactly how they would be if I only took the plastic shell off.
Tell me more about the CDIP EPROM thing though. Don't really know what that is or what a ZIF socket is but if the cassette thing doesn't work out I'm all ears for alternatives.
Note, the device probably won't be wrist mounted. I'd like it to be more akin to FO1-2's Pip-Boy 2k, pic related

[Anonymous No.1753753]

>>1753749
This is an EPROM in a CDIP package (ceramic version of PDIP). Note the difference between an EPROM and an EEPROM, the former is an earlier technology that requires UV shone through the window in order to erase the bits (Erasable Programmable Read Only Memory), while an EEPROM has no window as it can be erased just by setting a pin or two high (Electrically Erasable Programmable Read Only Memory).
Note that due to being an old technology their storage spaces have an upper limit around 4Mb, which is smaller than a single song would take up if it were of the somewhat common modern 128kb/s resolution. But you could easily drop an octave or two from the upper end of the audio spectrum, and stick to a lower resolution DAC and hopefully get that down by an order of magnitude.

I see that space isn't as big an issue as I was thinking, and neither are you trying to fit stuff into an existing pip-boy mockup of some kind. Do you have the technology to fabricate a housing that mimics a pip-boy in the first place?

[Anonymous No.1753756]

>>1753753
And this is a ZIF socket. ZIF stands for "zero insertion force", because there's a lever to lock them into place as opposed to having to force the IC into the socket. It's preferable for ICs that you're inserting and removing often, and is used for programming memory ICs or MCUs, and in those component testers. This green color is by far the most common, but I've seen the odd black one or two. They come in different sizes too, so you'd need to ensure that if you were going this way you'd get the one with enough pins and the right distance between the rows to fit your selected IC.

[Anonymous No.1753762]

>>1753481
I think I got a little closer so I now at least understand what we are convolving (not why it works). Suppose we have a "system" for example a simple RC filter, and we know its response to an impulse function, lets call it h(n). Now we can calculate its response to ANY input without using the actual system by convolving the input signal x(n) with h(n). However why the hell do we want to do that since we already have our system AND we have measured its response to the impulse function. So we must have an actual live circuit. Why cant we just measure another response to any other input. What are we, lazy?

[Anonymous No.1753766]

>>1753762
I have no clue what the fuck you're on about, but if it's about representing an arbitrary signal as a sum of simple signals, like some sort of multi-step response of fourier magic, it's often done because the total waveform itself is difficult to throw through the mathematics. Take an RC low-pass filter. A sine wave of any frequency can be calculated because it only has one frequency component, but a square wave with arbitrary duty-cycle has infinite components and thus would result in a numeric/approximate answer. So by beforehand numerically solving the most simple of signals, in this case the step response, we can analyse that and obtain a limit, the inverse exponential. So the square wave with variable duty-cycle is just the repeating sum of these exponentials with a time shift between each one. Which is why we get these tables with functions and their corresponding output functions for these sorts of numerical/infinite sum calculations, like fourier transforms and inverse fourier transforms.

Or did I completely miss the boat? We never used the term "convolution" in the electronics papers I did, only once in astrophysics and that one went right over my head.

[Anonymous No.1753769]

>>1753753
I don't mind audio in lower quality. If anything, it only enhances the retro aesthetic. Overall though I really like the ZIF idea now that I understand it. In regards to the case though, I don't have access to a 3D printer, so I'll probably just have someone make it for me online and ship it out to me. Guess I better start getting handy with CAD now.
Side note, how would I keep the EPROM from being erased? If this is in stead of the cassette idea, it would also have to be modular and easily replaceable as well, which would mean the ZIF socket would be on the outside, easily reached by sunlight. Are there chips with removable shields?

[Anonymous No.1753770]

Where do you guys get your electronic parts? Fry's electronics is a shell of it's former self and I hate having to resort to amazon to get stuff

[Anonymous No.1753773]

>>1753770
Ebay

[Anonymous No.1753775]

>>1753769
3D printing + painting is just the easiest method, and would help for fitting in an existing (cassette) mechanism since it wouldn't take too long to make multiple brackets with different fits for prototyping. If you have access to some hand-tools (a dremel would also help) then actually making the thing out of thin sheet metal over an MDF frame might be a better option. Have a look at construction techniques used for this sort of atompunk (or steampunk I guess) build online. But remember, if he's gluing a watch movement to it, he's full of shit.
Thicker sheet metal that you weld or braze together could be nice too, if you have access to that equipment.
>how would I keep the EPROM from being erased
Usually an EPROM would have a sticker over the window, which is fine, but if you instead have something (orange acrylic?) fold/clip over the window and store the rest in a dark box it won't be necessary.

>>1753770
AliExpress. Or at least I did until all my products stopped arriving at all. And now my country has an extra 15% tax on everything bought from overseas. The golden age of chinkshit is waning. FUCK YANWEN.

[Anonymous No.1753783]

>>1753775
>MDF frame
What is that
>if he's gluing a watch movement to it, he's full of shit
I'm autistic. What does this mean?

[Anonymous No.1753786]

>>1753766
>if it's about representing an arbitrary signal as a sum of simple signals
no, that would be a fourier transform or laplace transform of more general signals.
i was talking about predicting a response to an arbitrary input signal based on the response to a specific signal (the dirac delta function)

>total waveform itself is difficult to throw through the mathematics
thats what i believe is the whole point of the LTI approaches including convolution

the way i understand the idea, every text books says this:
>if you know the response of the system to the impulse function then you can calculate the response of the system to ANY input.
how? using convolution.

so using your example,
>>A sine wave of any frequency can be calculated because it only has one frequency component, but a square wave with arbitrary duty-cycle has infinite components and thus would result in a numeric/approximate answer.

instead of calculating the response to a single frequency, the textbooks talk about obtaining a response to the impulse function.
and then if you need to know the response to a square wave, you apply convolution with the previously obtained response to the impulse function.
there is the whole math theory behind that which i don't quite understand but based on what it looks like it is something about correlation: it is like a weighted moving average or something like that, see pic

[Anonymous No.1753794]

>>1753769
the ISD ChipCorders are low enough quality. larping teh 1980s isn't going to get your dick wet

>>1753770
read the OP, "Components/equipment"
LCSC is my fave

[Anonymous No.1753796]

>>1753794
>the ISD ChipCorders are low enough quality. larping teh 1980s isn't going to get your dick wet
I don't really know what you mean by this or what you're saying. I didn't really say anything about whatever chipcorder you're mentioning either, but if you didn't get the picture at my mention of "Pip-Boy", yes, this device is going to accompany me on hikes and travels so I can feel larpy. It's not anything serious.

[Anonymous No.1753815]

Ok, 1st time trying to solder electronics together.

>I have
1 soldering iron
1 water soluble paste flux
1 roll of silver solder
1 fire extinguisher
1 luxeon rebel star RGB LED
5 feet of 18 awg wire
Wire cutters/strippers

I'm trying to solder the 18 awg wire to the LED so I can hook it into the arduino / breadboard.

Am I safe to do this or am I being a retard?

[Anonymous No.1753821]

>>1753815
>plumbing solder
>plumbing flux
lolno
you need something much smaller in diameter for that properly sized iron to melt it, and you need rosin flux that won't continue eating through the board after you've finished. is there a Pic related near you? they have small containers of flux-cored solder that will work much better

[Anonymous No.1753825]

>>1753783
MDF is medium density fibreboard. A cheap and shitty mock-wood that's often used in cheap furniture underneath some sort of coating. While it looks pretty bad and isn't exactly strong, it's dimensionally stable and doesn't have a grain to worry about, and is relatively easy to "machine". Hence a frame made of this stuff would be more than good enough as the structure for a mostly aesthetic project like this.

A watch movement is the part of a mechanical watch that has all the gears and such in it. While it's a neat piece of tech, some steampunk guys just glue one onto their "art", regardless of whether it still works, which is an insult to the fine workmanship of a purely practical watch movement. The fine details of a watch movement also contrast heavily with the much larger details of basically everything else on the thing.
>I'm autistic
It's like having a mod that adds machines with 64x64 textures even though you're using the default 16x16 textures for everything else, making your machine room look like shit.

>>1753815
Silver solder melts at like 500C, while electrical solder melts at less than 200C. Try again. Gonna need an electrical soldering flux too.

[Anonymous No.1753827]

>>1753821
>>1753825
Thanks anons. Saved me from destroying my stuff.

[Anonymous No.1753829]

>>1753827
no worries m8
this is the stuff you want. okay, leaded solder is actually better for noobs who haven't gotten the quickness yet, or for applications like this where you will be heating all that aluminum core too, but it's the stuff you can put up with and have half a chance to make some joints that won't completely suck
https://www.harborfreight.com/lead-free-rosin-core-solder-69378.html

[Anonymous No.1753833]

>>1753827
another thing that can really help when working with those aluminum core pcbs is a pre-heater. if you have some kind of hot plate or other device that can generate bottom heat, there's that much less heat being sunk to ambient while you're trying to solder that bitch up
oh, and 18awg is really overkill for that application. 24awg doorbell wire would be plenty, AND easier to solder

[Anonymous No.1753838]

>>1753825
>It's like having a mod that adds machines with 64x64 textures even though you're using the default 16x16 textures for everything else, making your machine room look like shit
That made much better sense. You should be a social worker.

[Anonymous No.1753841]

>>1753815
>1 water soluble paste flux
whats that. never heard of that. is it for SMD?

[Anonymous No.1753850]

>>1753841
Read the responses, it's plumbing flux, probably zinc chloride. Zinc chloride is a very active flux that will slowly corrode the copper if not washed away, which is fine for ~1mm thick copper plumbing, but not fine for the fractional mm that you get on a PCB. In general, the more water-soluble a flux is, the more inorganic activators it has, and therefore the more active it is. The common RMA (rosin, mildly activated) has organic activators in it and is pretty-much entirely non-soluble in water. No-clean fluxes are designed to be as non-corrosive as possible while still being good fluxes, but some experienced solderers prefer to wash it off anyhow. IIRC, common no-clean fluxes work better with leaded solder than with lead-free.

[Anonymous No.1753893]

what pcb trace clearance would you recommend for eu 230v mains?

[Anonymous No.1753899]

Hi /ohm/ I've been gifted this small takoyaki grill from japan. The problem is that it's rated 100V (700W) and I'm in europe. I don't want to get a step up converter that would probably cost more than this thing.

So I opened it and looked at the thermostat, which is a ksd301g and it says 250V 10A (pic related are the guts)
Will it work just fine at 220v? Do I just have to replace the plug?

[Anonymous No.1753907]

>>1753899
>Will it work just fine at 220v?

will probably work for a time but die pretty soon. pumping twice it's rated current will likely be traumatic to the element. sure the thermostat will switch off, but only after a delay.

your two options are to get a 700W+ dimmer, or find a load to put in series, like maybe a space heater.

[Anonymous No.1753910]

>>1753899
If you connect it to the 230V mains you get a current of about 16A, the typical current of a standard B16 circuit breaker. The power would be about 3700W instead of 700W. It would destroy your device.

What you can do is find a 1kW (min. 700W) step-down transformer which is both large and heavy, but such transformers do exist to connect american 110V appliances to the european mains supply.

[Anonymous No.1753912]

>>1753899
Get a big transformer.

[Anonymous No.1753918]

>>1753899
>step up converter
to convert 220V UP to 100V?

>>1753899
>250V 10A (pic related are the guts)
>Will it work just fine at 220v? Do I just have to replace the plug?

just because one component is rated for 220V doesn't mean the device will work with it.
I can put a capacitor rated at 1KV into my fleshlight and that doesn't mean i can suddenly connect it to the trolley wires outside
Go to a chink eshop and buy a mains converting wall wart

[Anonymous No.1753919]

>>1753910
>step-down transformer
Amazon has such a transformer that isn't that large, €60, free shipping.

[Anonymous No.1753944]

>>1753534
Patrickhooddaniel on YouTube has a good series on AVR chips

[Anonymous No.1753952]

>>1753899
Don't plug it directly in, you'll trip a breaker, and the thermal resistance between the element and the bimetallic strip is enough that your element will likely kill itself pretty quickly. Your octopus balls will also be in a sad shape.

Put a capacitor in series with it, a big motor-start capacitor. If you need 110V across it and you have 220V mains, then the capacitor needs to drop 190V (blame pythagoras). At 10A, 190V means X = 19Ω. f = 50 or 60Hz, say 50Hz for starters. X = 1/(2πfC), or C = 1/(2πfX) = 1/(100π*19) = 170µF. Go for a value around about there, ±10% won't matter since it's running with feedback anyhow. But check to see if such a capacitor is any cheaper than a proper step-down transformer.

Or a switcher, I guess. Be pretty easy to just rectify mains, PWM it at 25% (if 50% would work then a diode would be perfect here) with a slow-ass frequency since it doesn't matter (slow enough to use an SSR, though maybe a single MOSFET would be preferable), and feed that choppy shitty DC into the element. Could even use a simple RC phase delay circuit in order to get the timing of the switching right, no need for any fancy timer or micro or switching controller.
Or if you want optimal performance, clamp a thermocouple right on the element and PID that bitch.

[Anonymous No.1753975]

>>1753441
single layer pcbs are hell
now i will have to resort to jumpers which means more soldering work

[Anonymous No.1754037]

Not sure if I should ask here or the uC thread, but I have a raspberry pi that I'm using to turn an old HiFi speaker into a HiFi WiFi speaker and it sounds kinda soft. I know Rpi isn't know for its audio (that's why there are DACs for it), but does the Rpi need a pre amp as well? I've already jacked up the settings as high as I could on my stereo and that's clearly not making a difference.

[Anonymous No.1754041]

>>1753893
Pic related, the columns you are probably interested in are B4 and A6. the > 500V line is actually showing calculated clearances for 1500V

>>1753975
nah they're just not easy mode. component placement and orientation matters a lot more
move/rotate RV2 and R4 to make it to the upper opto
move the wiper of the pot to the outside of U1 to make the lower opto connection
run ground from pin 4 to the power supply around the outside, then to the top opto. then break the connection between the two optos and escape your way back to pin 6 between the pot legs

>>1754037
yes, the pi needs a preamp, just like any other headphone or line-level output

[Anonymous No.1754045]

>>1754041
>>1753893
>Pic related
>doesnt Pic related

[Anonymous No.1754047]

>>1753893
and, to be clear, you want the 301...500V row of the table because peak voltage

[Anonymous No.1754073]

red pill me on car alternators /ohm/.
How, why.

[Anonymous No.1754077]

>>1754073

i use two dynamos hooked up to back wheels. it's good enough for my Kia.

[Anonymous No.1754090]

>>1754045
is pcb trace internal or external condutor?

[Anonymous No.1754091]

>>1754077
>>1754073
I meant more in the sense of using second hand units for diy projects

[Anonymous No.1754092]

>>1754090
external, unless it's on an inner layer

[Anonymous No.1754094]

>>1754092
fuck, i don't have that kind of space, 1mm will have to do, i bet the sheet values are for like wet air and shit like that anyway

[Anonymous No.1754106]

Input is a 60Hz sine wave. I understand that the high frequency noise gets coupled back to the "ground" but where does the high frequency content rejected by the inductor go ? I have heard people saying it gets "reflected".
Also, any reason to choose one circuit over the other ?

[Anonymous No.1754115]

>>1754106
I don't think a 60Hz sine wave has high frequency content.

[Anonymous No.1754117]

>>1754115
I'm studying power line (mains) filters, so there is a lot of high frequency noise on them.

[Anonymous No.1754127]

>>1754106
>"reflected".
i think that means it stores the energy then releases it back to the mains. reactive power is a similar concept.

[Anonymous No.1754129]

>>1754106
>Also, any reason to choose one circuit over the other ?
this is a different context, but look up power correction. that depends if your load leads or lags. same idea.

[Anonymous No.1754131]

>>1754106
Inductors act as high impedences at high frequencies. AFAIK, if it's a choke, it gets lost as heat. If it's a toroid, also lost as heat. If it's a straight coil, I believe it acts as an antenna.
Reflection happens if the load+inductor impedence isn't matched to the source impedence.

[Anonymous No.1754138]

>>1754131
How can reactive power be lost? only a tiny percentage of I^2*R will be lost as heat since R is very low

[Anonymous No.1754143]

>>1754106
>choose one circuit over the other ?
Audio fans seem to use RLC filters. Pic from
http://www.audiomisc.co.uk/mains/filters1.html

[Anonymous No.1754145]

>>1754094
A is for terminals, B is for traces. if you apply conformal coating you can use A7 clearances between component terminals and not worry about it. otherwise post pics of house fire

>>1754106
reactance: X(L)=2*pi*f*L
inductors resist changes in current. since it is probably voltage noise, it just stays on the mains and doesn't reach the inside of the filter. or, the converse, if the noise is being generated inside your equipment. "reflected" isn't a bad term for intuition
>one over the other
power factor, as other anon mentioned. but why not both?

>>1754143
audiophools are cargo-cultists

[Anonymous No.1754148]

>>1754138
>How can reactive power be lost?
If it's coupled out-of-phase with itself, or reflected (out of phase) back to the source, or blasted out into space. Respectively.

>>1754106
>Also, any reason to choose one circuit over the other ?
Depends on the frequencies of the noise. Capacitors are usually easier: smaller, cheaper, more intuitive, and less likely to produce voltage spikes.

[Anonymous No.1754151]

>>1754148
>back to the source
So not really lost, as opposed to heat loss or
>blasted out into space
i.e. converted to a different form of energy
not sure if that distinction is that important for practical purposes though

[Anonymous No.1754201]

>>1753687
>>1753689

I'd rather buy some cheap cassette player with digital controls and slap an arduino over there.

also, I remember some dude buying an electrical typewriter from 1980's featuring a daisy printing wheel, wired cables to a Z80 interface and used it as a text-only printer on his vintage CP/M machine. I guess arduino could do way better.

[Anonymous No.1754207]

>>1754106
Not op but, if those two circuits are basically equivalent, and electrolytic capacitors end up leaking, then why don't we see more psu's using inductors?

I think I have seen an inductor in series in some psu's schematics, but never without a capacitor as well.

Your PSU or amplifier would last forever. No?

[Anonymous No.1754213]

is there some sort of list of stuff you can take off of old electronic equipment to re use?

[Anonymous No.1754250]

>>1754151
blasted out into space is a category that also includes being incorporated into the magnetic field around the inductor, where it can be blasted right back into the wire when the field partially implodes

>>1754207
inductors resist changes in current. that's exactly the opposite of what is desired when a bursty load like a CPU gulps down a massive quantity of charge two or more times every nanosecond in order to switch millions of its internal transistors from one state to the other. but that is exactly what is needed so that e.g. a 12V source can supply charge in a controlled (and controllable) manner to feed those caps, in order to maintain the required voltage expected by those transistors. the same is true in turn on a switched-mode converter which needs a cap on the input to store enough charge to push through the inductor at its own rate of roughly once every microsecond. and so on all the way to the wall plug. many non-switched-mode supplies enjoy the additional complication of not being stable without some load on the output, in order that they can maintain good load regulation by feeding/starving sudden changes in output demand
>last forever
probably not, just about anything through which current flows degrades because of it. entropy never sleeps

>>1754213
no, half of them aren't even marked in any way that makes sense. better off buying somewhat known quantities off of alibay

[Anonymous No.1754252]

>>1754213
Depends on the equipment, doesn't it? Generally you can even de solder all the components and re use those.

[Anonymous No.1754257]

>>1754145
>audiophools are cargo-cultists
Yes, but I doubt that makes that filter less efficient than other filters of the same order, just overkill for their particular use-case. For low-noise scientific equipment, I imagine those sorts of filters are somewhat common, but of course I'd have to check this.

>>1754213
>>1754252
I think he means more a priority list. Things that are actually valuable, like old (non electrolytic) capacitors, big coils and transformers, tubes, metal can novelty packages, obscure buttons and switches, that sort of thing. But they're pretty self explanatory, that anyone with a year or two of electronics knowledge would know what to look for. Something that isn't too specialised or common, something that would be easy to incorporate into a circuit.

But personally I think old equipment like that could often be more valuable if it was repaired and polished up, then flicked on etsy or wherever.

[Anonymous No.1754262]

How hard would it be to make a magnetic levitation stand by hand? I've seen a bunch recently and want to try it out, but they all are either over $100 or will take a month and a half to ship. I haven't been able to find much info about how they work, since Chinese postings don't really have much content. Any ideas?

[Anonymous No.1754265]

>>1754257
yeah thats what i meant, though it doesn't really have to be old stuff, are there things you can salvage from even newish stuff?

[Anonymous No.1754270]

>>1754265
y tho

[Anonymous No.1754290]

>>1754262
I imagine the main price will be in winding those solenoids and buying the rare-earth magnets. But you should be able to find an open-source design and make it without too much trouble. The most difficult part circuit-wise is the feedback loop, inferring the position of the hovering piece via way of back-EMF in the main coils. And then you need a PID control scheme to feed back into the coils. With any luck all the specifics of that have already been worked out for you, but you'll still need possibly 4 full H-bridges. It's not that cheap of a project in the first place. What do you want to hover anyhow? Could it be done with diamagnetic levitation?

[Anonymous No.1754296]

>>1754290
>open-source design
That's the problem, apparently, most of the designs I've seen are where the magnets are on top and it stops things from falling down, rather than keeping them floating above a surface.
>not that cheap of a project
I wasn't expecting to be able to buy a kit for a fiver; the ~30-40 I've seen sounds reasonable, but the real killer for me is the shipping time. A month and a half and I'll have other problems to deal with.
>what do you want to hover
Nothing too heavy, just some decorative pieces.
>Could it be done with diamagnetic levitation
Maybe, but I'd also like for it to rotate and to be able to vary the speed of rotation.

[Anonymous No.1754306]

>>1754262

Just forget about the coils (or use something that might look like them) and suspend your object above it with a magician's "invisible thread"

It won't even consume any power.

[Anonymous No.1754317]

>>1754213
i feel like it is mostly not worth the trouble. lots of specialized parts and the general parts are not that valuable since you can buy them new dirt cheap. the worst mistake is to buy some tv at a thrift store for like $5. then good luck getting rid of it once you desoldered a dozen capacitors and some odd transistors and diodes. not worth it at all.

[Anonymous No.1754343]

>>1754296
um, do you have a magnetics simulator? i think there was a foss version

[Anonymous No.1754350]

>>1754343
I don't, haven't done work with magnets in the past.

[Anonymous No.1754369]

>>1754350
Yeah, neither. But I do want to get into designing brushless motors.
The thing about levitating from beneath the object is that the thing being levitated must be a permanent magnet, and so will often want to flip over even if it's sitting in an actively stabilised localised trough in the magnetic field (which is sufficient for a lump of iron suspended below a hovering device). So not only do you have to balance the translational forces but the rotational ones too. I'm not sure, but I think the commercial ones have 3-axis gyros (maybe accelerometers too) inside the hovering part, and communicate this to the powered base via electromagnetic coupling like some sort of NFC. Otherwise I can't see how you'd monitor rotation and translation separately without some real complex algorithms. So you may need to sort this out.
Another issue is the reading of the magnetic field by the base station in the first place. As the thing hovering is a magnet, as it moves it will distort the magnetic field created by the solenoids, so you should be able to read this with a few hall-effect magnetometers. In other words, it won't be necessary to read the back-EMF, which simplifies things greatly.
To avoid needing a bunch of H-bridges, you could have the inductors constantly on at 50% duty-cycle, and just increase or decrease that duty-cycle to vary their magnetic fields. Might not even need permanent magnets on the base like this, but I'm not sure. If you want to use solenoids more-or-less vertical for translation, you could then possibly use (horizontal?) solenoids for rotation, and hence have seperate feedback loops for translation and rotation. But that's just speculation until you mess about with a simulator.

I'd be tempted to make my feedback loops entirely analog for simplicity's sake, but I think PID control is necessary, so I'd want a dedicated PID IC, assuming they exist. Digital would better allow me to mess about with algorithms.

[Anonymous No.1754380]

>>1754369
cont.
Chances are you'd need to try and combine the rotation and translation information together in the same feedback loop, and you'd want to simulate the magnetics in order to determine how the magnetic field will change as a function of translation in all three dimensions, and rotation in two of them.

I don't think it would be very easy to induce rotation about its vertical axis without something along the lines of an induction motor core, for which you could modulate the duty-cycle of the hovering coils in order to create a time-rotating magnetic field with a constant hovering force. But whether that modulating magnetic field would be enough to overcome friction to a significant extent is another issue altogether.

The only experience with magnetics I have is looking at a friend's project in a maker space. He had obtained one of those plastic rubbish bins with a swinging pivoting lid, fixed a magnet to it and a solenoid to the base, and used an arduino controlling the solenoid to feed energy back into the swinging system to make it continue swinging forever. The only problem was that the 12V supply he used for the solenoid was being fed directly into the arduino's ADC that he used for back-EMF detection. Which I told him. He didn't fix it, and I was amazed it kept working. I suspect most of the voltage was being shunted away by the input-protection diodes of the ATMega, meaning the voltage across the solenoid would be significantly limited.
Similarly, another guy at that maker space made a case to show a bottle of tonic water full of strobing UV LEDs (quinine is fluorescent), but when I asked him where the current limiting was he responded "The LEDs were far too dim with a 100Ω resistor in series with them". I explained that you're supposed to have one such resistor for each LED, or a smaller (higher-power) resistor for multiple LEDs in parallel. They were also being directly powered from an arduino, causing it to get hot. The LEDs gradually dimmed.

[Anonymous No.1754382]

>>1754380
The moral of the story is to learn electronics before you learn arduino.

[Anonymous No.1754400]

what's the absolute cheapest way to bring 12v dc to 5v dc?
what's the most energy efficient way?

[Anonymous No.1754412]

>>1754400
Cheapest, either a resistor and 5V zener or a tiny linear regulator. But the linear regulator has a far better result, so I'd certainly go for it instead of the zener.
Most energy efficient way, either a buck converter or capacitive switcher. Or a zener or reference diode I guess, if your load is really fucking low-power.

In general, a linear regulator is best for cases where you need a highly stable and noise-free power supply but don't need that much power (e.g. powering ICs, discrete analog circuits, etc.), while anything using a few watts of power or more (a fan, heating element, LEDs, H-bridge, etc.) would be better off using a buck converter. Capacitive switching converters are as efficient if not more so than a buck converter, but are typically lower power and fit in an overlap region between the linear and buck designs. One use for them is for lower-power circuits where power consumption still really matters, like on battery-operated devices.
For devices where high-power and high ripple-rejection matter, a buck converter will be used in conjunction with a linear regulator circuit of some kind on the output.
Voltage reference diodes are not meant to power anything, rather they're just meant to provide a stable voltage with which to feed into a comparator or other circuit.

But also note that if ever you need the enough power that a linear regulator is out of the question, it may be possible to improve your circuit by changing your load. Say you need to step 12V down to 5V for a 5V fan. It may be more efficient to simply use a 12V fan instead. This change will be space efficient and more electrically efficient than using a voltage converter.

And worthy of mentioning is the MOSFET logic-level shifter, which (when provided with 12V and 5V power rails) will convert a 2-way logic signal from one level to the other. Just in case you were about to use a linear regulator to decode RS232 or something.

[Anonymous No.1754416]

GOAT flux remover

[Anonymous No.1754417]

>>1754416
Rosin flux? I didn't think sodium carbonate would do much to organics like rosin.

[Anonymous No.1754420]

>>1754417
rosin is just an assortment of organic acids and maybe an activator. kinda like food, if you think about it. hot water + high pressure + medium-strength base should be enough to mobilize it. don't be too surprised if you see Pic related in a rework shop

[Anonymous No.1754459]

Nobody told me STMicroelectronics is yuropeen

[Anonymous No.1754460]

>>1753736
>y tho,
i assume for MUH RETRO FEELS
same reason why some bars still use juke boxes with vinil records even though they could be replaced by a $50 mp3 player

[Anonymous No.1754465]

>>1754460
>$50
More like $5

[Anonymous No.1754500]

>>1754459
now you know. /ohm/ does not endorse purchasing from ST, NXP, wurth, infineon, or any other eur*pean manufacturer.

[Anonymous No.1754501]

>>1754500
why not? I was actually going to start ordering TI stuff only, but now I'm gonna look for STM parts first.

[Anonymous No.1754508]

>>1754412
not that anon, just wondering. what's the difference between a simple zener and a proper regulator for _small_ currents and high input resistance specifically?

[Anonymous No.1754510]

>>1754508
I don't understand why you'd use a zener as anything other than a reference. 7805-type regulators are dirt cheap, give beautiful output, handle high currents, and are easy to set up.
Also, reverse diode breakdowns are noisy as hell, and literally used to make analog white noise generators

[Anonymous No.1754518]

>>1754510
don't have to use them, just wondering why they are bad. if they are noisy, why use them for anything at all? a reference or whatever.

[Anonymous No.1754527]

>>1754518

1) in many cases, zeners are used for over-voltage protection. so they spend nearly their entire life turned off except when a spike shows up. a regulator is active all the time.

2) zeners have no upper voltage limit. you could hook one up to 1 million volts and it would put out 5.1V happily and forever.

[Anonymous No.1754537]

how do big chink factories make pcb traces? do the print the schematic with an ink and then etch them, or do they use some super fast mill?

[Anonymous No.1754574]

>>1754537
Lithography and cyanide.

[Anonymous No.1754601]

>>1753975
>tfw will have to switch the endmill 4 times for this one smol pcb
fug, i wish i wasn't poor and could afford ATC

[Anonymous No.1754604]

>>1754510
quiescent current on 78xx and many other jellybean 3-terminal regs is shite, sometimes more than a reference load requires. adjustable regs add the programming divider on the output on top of that
>zeners are noisy
so use a band gap reference instead

>>1754601
>CNC routing a high voltage board
dis gon b gud

[Anonymous No.1754608]

>>1754604
don't worry fren, i have the safety hole and everything, i will be milling 1mm spaces around the traces and completely mill of the copper on the ends of the safety hole so the lv and hv parts will be completely separated

[Anonymous No.1754635]

>>1754608
cutouts increase creepage distance. I don't believe tracking is your main concern here as long as you clean the board well and make sure you don't get any contaminants on it once cleaned
here's the IEC/UL standard just for giggles. maybe you can relax a little bit in some places
think very hard about clear coating it for anything beyond bench prototype duty

[Anonymous No.1754698]

If I use conductive ink for traces instead of having a pcb made will it mess up the analog devices and adc readings of my circuit?

[Anonymous No.1754711]

Can someone identify this chip? It's on my pellet stove's controller.

[Anonymous No.1754725]

>>1754711
are you incapable of google
https://www.st.com/en/microcontrollers-microprocessors/st72325j6.html
https://www.st.com/en/memories/m95040-w.html

[Anonymous No.1754760]

>>1754501
What was wrong with TI? Not a huge fan of STM.
NXP absorbed the Freescale/Motorola stuff, so there is some legitimacy in there somewhere...
Also ON Semi bought some of the old Motorola fabs that made discretes (e.g. 2n2222 / 2n3055) so that might be as close as you can come to getting something modern with old school lineage

[Anonymous No.1754762]

>>1754760
>What was wrong with TI?
nothing, I'm just trying to be an ethical shopper, so no (or as little as possible) chinkshit. Then i was gonna buy TI stuff cause big, well-known western manufacturer, but being a eurofag myself now that I know I can support a european company I'll look through their catalog first.

[Anonymous No.1754820]

If I have a 3 phase full wave rectifier, what can I expect the DC output current to be for a given input AC phase current? Say a load is connected to the DC side and it is causing a known current to be drawn from the supply in each phase. I'm having trouble figuring out the mathematical relationship between the DC side current and the AC phase current. I can measure both sides using AC/DC current clamps but I feel like my measurements are not as they should be, I just want to know my SCR rectifier is working properly.

My colleague says he thinks they are related by a factor of sqrt(3) but my measurements don't quite reflect that and I can't find much information online.

I don't know if this is the right place to ask about bigger power related stuff.

[Anonymous No.1754831]

>>1754820
Well the DC voltage will still have some ripple in it, so it depends on whether you're using a resistive load or a current sink. Pretty sure you could calculate an integral using ideal diodes, if you really want to. Otherwise I'd just shove it through spice and see if it matches up with any particular theory.

[Anonymous No.1754861]

>>1754820
Replace SCR with an ordinary rectifier and re-measure?

[Anonymous No.1754868]

Anyone have recommended AGC circuits to shove in analog circuits? I'm open to all topologies, be they using OTCAs, JFETs, LDRs, or something else entirely.

[Anonymous No.1754881]

>>1754831
I didn't think the load mattered since I'm just interested in the input/output current relationship of only the rectifier. How much current is drawn from the source and the input/output relationship of the system as a whole will depend on the load, yes, but the 'transfer function' of the rectifier alone should be the same regardless shouldn't it?

>>1754861
What do you mean by ordinary rectifier? You mean with just regular diodes? We're using SCRs for controllability.

[Anonymous No.1754892]

>>1754881
Well depending on the type of load you're using, it will pull a different current when the voltage level is different. Unless you're using sufficient filtration capacitors, but you never mentioned any.

[Anonymous No.1754893]

>>1754892
either way you're going to get a non-sinusoidal input current waveform

[Anonymous No.1754898]

>>1754881
>regular diodes
Yes

>using SCRs for controllability
I understood that, however just replace them temporarily and it will look like the SCRs in the "on" state and measure that, then put the SCRs back.

How much current are we talking about here, anyway, like 100s of amps?

[Anonymous No.1754903]

>>1754892
>>1754893
So the output current of the rectifier can't be expressed with just one expression in terms of the input current?
We aren't using any kind of filtration yet, and yes the input current waveform has a LOT of harmonic distortion. Is it the lack of filtration that is causing the DC output current to be less than the input phase current? Because that's what I'm measuring, and it's not something I expected.

>>1754898
I see, maybe we'll try that. We'd just have to get some high power diodes cause yeah we're looking to put out a few hundred amps.

[Anonymous No.1754905]

>>1754903
>put out a few hundred amps
Just be aware that SCRs have a slightly higher voltage drop due to the extra junction.

[Anonymous No.1754907]

>>1754903
Well, the RMS input current will be equal to the RMS output current in any system thanks to Kirchoff's current law / the conservation of charge. But RMS is only useful if you have a resistive load. Other types of loads will give you other effective current values, which will also all sum to be equal to one another. But if you've got a non-resistive load and are measuring the input and output RMS currents, it may not be indicative of the power going into the load. I think.

[Anonymous No.1754927]

>>1754907
Wait, maybe RMS doesn't necessarily make sense through a system with a filtration capacitor AND a non-ohmic load.
I'm trying to picture a 50/50 square wave being input to a perfect diode, then with a finite capacitance and a current sink as a load. The capacitor voltage would sit constant for half the time, swapping to a downward slope for the other half of the load. The input current would have a spike (dirac delta?) at the start, a constant in the first half, and zero in the rest. I think integrating the input current as a simple average would give the same current as the load (because current integrated over time is charge, which cannot have a net change), but doing the RMS of the input current would probably give a different answer to the average. This might even be true for an ohmic load too.

[Anonymous No.1754928]

>>1754905
Noted, thanks

>>1754907
>the RMS input current will be equal to the RMS output current in any system
With a 3 phase input, is the RMS input current just the RMS of any of the phase currents? Why isn't sqrt(3) factored into that in some way?
Because my current clamps are not measuring the same current on both sides. Unless I'm using my clamps wrong, should I be measuring the DC side using AC mode on the clamps?
To be honest, our load (capacitors) seems to be getting the power we need it to be getting, I'm mostly just trying to wrap my head around the theory and measurements I'm getting.

[Anonymous No.1754930]

>>1754928
See >>1754927, the RMS current probably doesn't make sense for your application. I'd just do everything in terms of power instead. As long as you know the voltage on the output you can know the current that will be going into a load. Do you really need to know the factor that relates RMS current input to RMS current output?

[Anonymous No.1754935]

>>1754930
>Do you really need to know the factor that relates RMS current input to RMS current output?
Not really, mostly for my understanding for how the hell a 3 phase current relates to the DC output, because it's been annoying me for a while. It would be much more straight forward to do everything in terms of power but It's also just nice to know that the measurements make sense. I maybe should scope the input for average rather than RMS to see if they line up more with the output.

[Anonymous No.1754988]

>built lamp from old laptop screen
>accidentally tip it over
>rip off gold contacts i had wire soldered to
how do i repair this? I traced the pwm-contact to a mc, can't seem to find the backlight_en tho.

[Anonymous No.1754993]

>>1754988
>I traced the pwm-contact to a mc
What's a mc? Post a pic of the torn-off pads.

[Anonymous No.1755003]

>>1754465
not really if you want some decent speakers with it

[Anonymous No.1755012]

>>1754993
microcontroller

[Anonymous No.1755013]

>>1755003
I was assuming the speakers weren't part of it, since when one says "MP3 player" they usually think of a PMP, not a home sound system.

[Anonymous No.1755044]

So, "varistor voltage" is the voltage where the varistor starts quickly dropping in resistance and "max operation voltage" is the maximum voltage the varistor can handle on it without exploding?
So if i buy the highlighted one is it the correct one? I want to put it between live and neutral in a 230V outlet

[Anonymous No.1755049]

>>1755044
>I want to put it between live and neutral in a 230V outlet
Then just buy a standard mains MOV. The kind of things you find in power strips and protection circuits. Pretty sure MOVs have better forward characteristics than SIOVs. I can't see a clamping voltage on what you posted. My ones are rated at 275VAC with a clamping voltage of 710V, if that gives you a ballpark.

[Anonymous No.1755051]

>>1755049
But it is metal oxide
And they don't really have a category called "varistors for mains"
https://www.tme.eu/cz/en/katalog/tht-varistors_100397/?s_field=1000011&s_order=desc&search=varistor&visible_params=283%2C45%2C2%2C2167%2C10%2C31%2C288%2C284%2C193%2C127%2C43%2C32&mapped_params=283%3A1441875%3B284%3A1442176%3B

What is the name of the ones you are using? basically i want a generic mains protecting MOC i can include in all my circuits that use mains

[Anonymous No.1755182]

>>1755051
that looks like it'll do
they're not just magical amulets tho. every part of your input network needs to be evaluated for cause and effect. usually, you put them *after* the fuse/circuit breaker so that your circuit is disconnected from a large surge or overvoltage condition, rather than the MOV taking its limit and then blowing up. so you would rate it to be sufficient to blow the fuse

[Anonymous No.1755193]

>>1755182
yeah i know, the whole thing will look like this

[Anonymous No.1755213]

>>1755193
EN 60127 slow glass fuse:
1 A forever
1.5 A at least 1 hour
2.1 A max. 2 minutes
4 A max. 3 seconds
10 A max. 0.3 seconds

[Anonymous No.1755266]

i've been searching online for an answer but can't find any that actually help. i have some handlebar switches that i need to map out the leads. it's going to be a simple 12v circuit that it will be controlling, but i can't figure out how to map the wires to the switches. i've got a multi-meter

[Anonymous No.1755274]

Are there any common D-flop ICs in DIP variant?

[Anonymous No.1755283]

>>1755266
do you really? maybe you could find wiring diagrams in your vehicle's service manual. if you're really lucky you can find switch pinouts and throw diagrams
if that's not enough, you need the switch itself to reverse-engineer. number/identify the pins, draw a grid, set continuity mode, check each pair of pins and mark box at row/column if continuity established, set next position, repeat...

[Anonymous No.1755301]

>>1755044

you dont hook up a 230V varistor to a 230V line. if you do, then it'll be triggering all the time. in north amerikka you typically see 160V MOVs for 120V lines. THAT makes sense.

[Anonymous No.1755315]

>>1755283
it's an aftermarket part i bought on ali, i messaged the seller if they could provide a diagram. if they don't it looks like i'll be having fun with continuity

[Anonymous No.1755327]

If I needed to make a pcb for 0.3mm traces, how would I go about it.

[Anonymous No.1755380]

>>1755327
>pirate or download PCB design software
>learn how to use it (define components and footprints, create a schematic, route a board)
>export gerber and drill files
>upload the files to one of a dozen cheap pcb manufacturers
>pay <$10

[Anonymous No.1755391]

>>1755327
just jumper them 30 AWG wire-wrap/kynar wire for areas where you were considering using traces. That's .255 mm.

[Anonymous No.1755394]

>>1755182
Not him, but if you have a pair of chokes on L and N before any X capacitors, will voltage spikes get past them at all?

>>1755266
>handlebar switches
Are they variable speed controllers for an e-bike, or just on/off switches? If they're speed controllers, there will be at least one MOSFET in them, not something you'd necessarily get good results from with a continuity test. If it directly connects to a 3-phase motor then it will have a bunch of MOSFETs in there regardless of whether it's single-speed or not.

[Anonymous No.1755404]

>>1755394
they're for a motorcycle but i'm using the on an ebike. they're all simple switches that just have multiple wires coming out. i'm assuming it's just a case of putting 12v to the light i'm going to use with the wires to the switch attached. the wires from the light have a diagram so i know what to do with those, i just need to pick up a 12v power supply tomorrow

[Anonymous No.1755422]

>>1755404
How many wires are there? Link to listing?

[Anonymous No.1755440]

>>1755422
9 wires
https://www.aliexpress.com/item/4000250854994.html

[Anonymous No.1755449]

>>1755440
Oh, yeah for that sort of thing it's best to just mess about with continuity on your DMM. Or take it apart, but if a spring goes flying out it might be hard to get everything clamped together to reassemble it.

[Anonymous No.1755457]

>>1753437
can always check it with a volt/ohm meter.

[Anonymous No.1755458]

>>1755457
that was like 4 days ago

[Anonymous No.1755460]

>>1753437
these are horrible. can never figure out if it is red orange or brown. the background blue varies from light blue to dark blue from part to part even in the same batch and affects the bands color.

[Anonymous No.1755481]

>>1753417
I wanted to make a toy lightsaber because I dont want to spend $200 on one at disneyland, so I hooked up this LED to my arduino, but it's not nearly bright enough. Does anyone know what kind of lights they use in the expensive lightsabers Disneyland sells?

This is the first time I've ever messed with circuitry/electronics.

[Anonymous No.1755486]

>>1755481
Why are you using an arduino? With no switching transistors no-less. Power it directly off a good strong power supply (2A USB should be sufficient) with the correct current-limiting in place. Check the LED's datasheet for what kind of current it wants to run off, because I'm pretty sure an arduino won't be able to give it all it wants.

[Anonymous No.1755487]

>>1755486
>why arduino
I want to be able to change the light to pre-set colors using a rotary encoder.

>switching transistors
I'll Google that.

>2A USB
I'll Google 2A

To be honest there's probably even more I dont even know that I don't know.

[Anonymous No.1755520]

>>1755380
Id need to prototype several boards and the cost would add up plus the wait.
>>1755391
Its a mixed signal design so Im a little worried of the wire acting like an antenna.

[Anonymous No.1755534]

>>1755394
inductors in series add unless magnetically coupled
it depends which end of the filter you are looking into. the usual line-filter-in-IEC-inlet arrangements have caps on both the line side and the load side of a common-mode choke

>>1755487
you're gonna need a bigger boat
http://henrysbench.capnfatz.com/henrys-bench/arduino-output-devices/10-watt-high-power-led-with-an-arduino/

>>1753437
ask me again why I love SMD

[Anonymous No.1755565]

>>1755487
>I want to be able to change the light to pre-set colors using a rotary encoder
Well, I'd do that with three potentiometers and a LM339 quad comparator, but that's just me. A single rotary encoder is a pretty elegant way to go too, so long as you're not lugging about an entire arduino dev board. Consider an ATtiny85 plugged into one of those digispark guys.
>2A
That means two amperes of current, the maximum current allowed by USB2.0's standard. The arduino itself can only source 50mA of current or so. Maybe 100mA if you're using the right combination of pins. What you need to control the PWM (look up PWM it's what you need to know) with the arduino or comparator circuit is to use transistors (some low voltage MOSFETs should be what you're looking for, perhaps FQP30N06), which can handle far more than 50mA.

Then again, since these are apparently 10W LEDs, just using PWM and resistors is probably insufficient. Requiring 1W resistors and all that. So while it is a more advanced type of circuit, I'd suggest looking into constant-current drivers.

[Anonymous No.1755578]

>>1753437
use a multimeter you cuck
honestly i don't understand why they just can't print a number on them like they do with smds

[Anonymous No.1755580]

>>1755578
IT WAS 4 DAYS AGO AND HE ALREADY GOT ANSWERS

[Anonymous No.1755585]

I'm trying to connect a radio to my arduino via serial. The radio has a single pin for both RX and TX, arduino has them separated into two pins. If I just short RX, TX and the radio together it doesn't work, but it does work if I connect only the RX or the TX to the radio. Am I correct in thinking that all I need to do is put diodes between the arduino and the radio to stop the radio's signal from going into the low resistance TX rather than the high resistance RX, or am I going to need to do something more advanced? I do have an attiny that I might be able to get to work as a "translator" by bitbanging or something.

[Anonymous No.1755593]

>>1755585
you need to detect when the radio is transmitting or receiving and redirect that traffic to the correct arduino pin
having both rx and tx on the same pin is retarded as fuck
for example you can use a tranny to block the rx pin when you want to transmit via tx on arduino and then when you are done transmitting you use another tranny to block the tx and open the rx line
so basically any time you aren't transmitting via tx it will be always blocked and rx open

[Anonymous No.1755597]

>>1755585
RX and TX on the same pin? Are you sure it's still sending information via I2C and not some other standard? If it is I2C, and there aren't any RS232-style status pins (ready to send, etc.), it should still be possible to just use 1 wire so long as you can change a pin from digital-input to digital-output on interrupt. But it might get a little complicated, such that you may want to use three wires on the arduino: one as the input, one as the output, and a third to control the logic/transistors saying that the arduino is transmitting.

[Anonymous No.1755615]

>>1755585
What radio is it?

[Anonymous No.1755619]

>>1755585
>>1755615
Oh yeah I was gonna ask this too. I've got a radio IC lying about somewhere myself but I think it uses normal I2C. It's mostly for telling it to change up or down frequencies, but it can also read the radio station metadata and tell that to the MCU IIRC.

[Anonymous No.1755627]

>>1755593
I'm thinking it might be easier to do something like that, yeah. I can put it on an interrupt pin and just read the serial string in software, then switch to output to transmit a reply. But if I could do this with the actual hardware serial interface obviously I would prefer that. I've seen on some forums discussions about putting a resistor between the arduino's TX and RX pins to somehow black magic this thing into working. Any string the arduino transmits would be duplicated into its input, but I could filter that out easily enough.
>>1755597
I'm pretty sure it isn't I2C, but I can't be sure. https://github.com/SpektrumRC/SRXL2/blob/master/Docs/SRXL2%20Specification.pdf is the datasheet for the protocol, as far as I can tell the relevant bit for this is page 5, it seems to state pretty clearly that this is UART.
>>1755615
It's a Spektrum SPM4650, it's a 2.4GHz serial radio receiver with telemetry transmitter, the sort of thing used in RC for drones.
The use case for me is that the radio sends a handshake signal, my arduino replies with its handshake signal, then the radio starts feeding information in six channels to the arduino. The arduino at this point has to send a "keep alive" signal or something like that every 50ms or something, I haven't really gotten that far yet. The arduino can also send strings to the receiver, these are broadcast to the other radio, a function called "telemetry" that I plan to use to show battery voltage and current draw.
What I'm doing right now is figuring out the handshake. I'm able to read the signal clearly, an example line looks like:
A6 21 E 10 30 A 1 1 AA C4 52 E3 6C A4
But I need to be able to reply with my own:
A6 21 E 30 10 A 1 1 //UID&CRC//
If I don't reply it will never switch to giving me channel info, it'll just keep spamming its handshake.

[Anonymous No.1755628]

>>1755627
>putting a resistor between the arduino's TX and RX pins to somehow black magic this thing into working
https://electronics.stackexchange.com/a/183431
Is an example of this, by the way.

[Anonymous No.1755630]

>>1755627
It says it's a UART, I believe it.

>>1755628
Yeah this will work, so long as the resistance is significantly larger than the UART's output impedance but significantly smaller than the UART's input impedance. Or vice-versa, thinking is hard. I can't see why we can't just avoid this and use common-collectors on both ends with pullups so either end can't fight one another.

[Anonymous No.1755660]

>>1755630
I don't know what the internal pullup on the radio is. The example on stackexchange is for a 10k internal, which sounds plausible, but if I were to figure this out on my own would it be safe to use a potentiometer and simply turn up the resistance until I hit the point where the RX starts receiving?

[Anonymous No.1755803]

I made a shitty inverting delay circuit. It turns the load (the LED) off for a little bit when the button is pressed.
It draws more current than the LED when it's active though.
I think having the inverting transistor on the emitter of the other transistor is a no-no. I can't remember.

[Anonymous No.1755906]

I'm the guy with the parallel battery cells from the other thread, but I have a few follow-up questions.

For the first prototype I'm probably gonna use a 1-cell 18650 power bank circuit to save space and combine the boost and charger circuit. That way I won't fry anything, but I will be lucky to get two hours from a mid-range SBC with a screen. I'm new to battery circuits, so let me get this straight -- if I have n batteries that are perfectly identical, then I can put them in parallel without danger. The danger comes from when one battery shorts and the other cells dump into it, so one anon said to get protected batteries, but they aren't rated for that kind of current and it might

Could I just put fuses inline with each battery to avoid that? If the protection circuit fails then the current will probably exceed 3A since the charger is only rated for 2A anyway? What are the other strategies? Provided I match the IC on the power bank circuit to my particular battery chemistry the main problem I have is to ensure the batteries don't fry one another from overcurrent/overdischarge?

[Anonymous No.1755952]

>>1755906
> 18650
consider the 21700, some power tool manufacturers are moving to it

> parallel problems
consider the problem with running transistors in parallel. You can match as best you can but it won't be perfect, so we limit them (in this case, with .33Ω resistors) to balance out their relative contributions. Otherwise, a minor difference will start shunting more current through one which will runaway and burn out.

Well-matched cells work fine at the start (it's just generally a good idea) but over time they deviate from one another, and those deviations can accelerate more deviation! It's not (by any means) proof against catastrophic/cascade failure.

You need to protect the cells from over discharge anyway, and you can't really do that when they are hooked up in parallel. If one of the cells develops a higher internal resistance than the others, the other cells will mask it's individual voltage.

[Anonymous No.1755958]

>>1755803
> draws more current when active than LED

big clive did a teardown of a bunch of "night lights" and found that a lot of them have a very inefficient circuit that simply shorts out the LED with a transistor to turn it off so it uses more energy during the daylight!

Not a lot of power though, but against most people's expectations.

[Anonymous No.1755972]

>>1755906
>if I have n batteries that are perfectly identical, then I can put them in parallel without danger
relatively speaking, no more danger than any other volatile chemical reaction you might run
>protected batteries
protection circuits see overcurrent situations and shut down when they see them. in the case of the ubiquitous DW01 there's a 10-20ms overcurrent grace period for whatever level you set and a 50µs grace period for 9x that threshold current. that's not a bad idea overall
>2A
each battery would be seeing only 500mA during charge. but I think a fuse would be too slow to avoid damaging the other batteries, see >>1755213

>>1755952
>If one of the cells develops a higher internal resistance than the others, the other cells will mask it's individual voltage.
yes, but the voltage will equalize through the few milliohms that is the parallelling strap. the net effect will appear as a lower pack capacity. it is true that the troubled cell won't contribute its full part to the output current and the others will have to work harder. what you are saying is 100% true in the case of a series connection without balancing

>>1755660
you only need one pullup on the line. check the datasheet for whatever MCU you're driving it with to see if its USART has an internal half-duplex open-drain function. if not, just use a diode on the Tx pin of the MCU so it acts like it's open-drain

[Anonymous No.1755991]

sup /ohm/
I'm having trouble with some chinese audio module with usb, bluetooth capabilities, it also has a tiny 7 segment display, it supposedly takes 12V (Which is alright I think, it has a 5v linear voltage regulator), however every single one of these modules has smoked for some reason when using it along a 12V 3A, I'm thinking about using a buck converter to reduce heat dissipation to drop the voltage down to 5V and a linear current limiter to avoid any more issues, is my approach alright? I'm also worried about the output impedance of my current limiter circuit (I'm going to be using an LM317).
I should add though, the first module failed on the voltage regulator, which smoked, the second one failed on what I think is a transistor, which sparked and smoked.

[Anonymous No.1755995]

>>1755991
> 12 -> 5v regulator

Bad idea, the linear regulator has to burn off the difference, and that's a lot. Buck converter has merit (if, for no other reason that efficiency)

[Anonymous No.1756004]

>>1755991
>it has a 5v linear voltage regulator
>I'm thinking about using a buck converter to reduce heat dissipation to drop the voltage down to 5V
But the built-in 5V regulator won't work off just 5V as its input, you'd need to use a 7V power supply or thereabouts. Still, if the chinky module is only drawing a few mA like it should, the 5V regulator inside shouldn't let the smoke out. I'd do some more advanced troubleshooting first off. Were your wires the right way around?

[Anonymous No.1756010]

>>1756004
Im thinking of bypassing the linear voltage regulator altogether, and yes, I checked many times and the wires where right.

[Anonymous No.1756012]

>>1756010
>>1756004
Well, also adding an important detail, while I checked if the wires where right, this was not after many people tampered and tried to "fix" it (It's not really my own personal project, just helping out someone for fun), and believe me, it's a fucking mess, so maybe some IC is already faulty or something.

[Anonymous No.1756013]

>>1756010
>>1756004
>>1756010
>>1756004
Well, also adding an important detail, while I checked if the wires where right, this was not before many people tampered and tried to "fix" it (It's not really my own personal project, just helping out someone for fun), and believe me, the whole thing its a fucking mess of wiring and solder, so maybe some IC was already faulty or something before I could even get my hands on the module.

[Anonymous No.1756017]

>>1756013
Never trust anyone's previous work, especially not your own. Get a new one, and boot it up solo with a current-limited benchtop power supply

[Anonymous No.1756020]

>>1755991
look up the regulator in question. maybe the Vin max is something ridiculous like 7V. yes, there are uses for them, no, taking 12V down to 5V isn't one of them
>LM317
or just choose a buck with a current limit appropriate to the application. they do come in all shapes and sizes

[Anonymous No.1756021]

Could someone tell me why Ib is equal to 3.3 mA?

[Anonymous No.1756024]

>>1756021
but Ib is 162mA
and the 560Ω resistor isn't doing anything

[Anonymous No.1756054]

why are you still using LTspice, anon?

[Anonymous No.1756055]

>>1756054
What is that?

[Anonymous No.1756058]

>>1756055
Micro-Cap 12.2.0.3

[Anonymous No.1756059]

>>1756054
no, really, anon, why are you?

[Anonymous No.1756060]

>>1753441
Is this an AC chopper circuit? It looks oddly familiar to one I made about 3 years ago

[Anonymous No.1756067]

so common mode chokes (like a transformer with identical winding counts) only remove high frequency noise from lines or do they do also other stuff?

[Anonymous No.1756073]

>>1756054
>>1756059
answer me senpai

>>1756067
they also help keep noise, generated by your circuit, inside your circuit. combine with caps on both sides for best results
they're also inductors, with all the consequences that entails

[Anonymous No.1756079]

>windows only
>not using based KiCAD

[Anonymous No.1756105]

>>1756060
it's a led dimmer, it works by slicing up the sine

[Anonymous No.1756107]

>>1754993
this is what it looks like now
and what the other anon said, MC is microcontroller

[Anonymous No.1756117]

>>1756107
oof
For starters that solder joint on the big rectangular pad looks too cold and too dry, you should use more flux and up the temp for a short while.
As for the torn-off pads, you could either scratch away the solder-mask on that side (or the other side) and solder directly onto the traces, or you could solder a wire to the µc pins themselves. With any luck there's a nice big passive on the same node to jam a wire against instead of that piddly QFP, but you take what you can get. Technically you could drill/poke out a via and jam a wire in there and solder it to both sides, but that sounds dodgy and I've never heard anyone do it before.
As for materials, I'd use a thinner solid-core wire (28awg?), preferably with heat-resistant silicone/ptfe insulation, and secure it with some silastic or hot-snot as a strain relief. You know, so the wire doesn't bounce about and snap or delam.

[Anonymous No.1756120]

>>1756021

your circuit is completely fuckered, so dont expect any rational explanation for any currents. if you built it in real life, it would burn out your transistor.

you need a resistor from base to ground (say 1k-10K) to keep the base current in a sane range.

[Anonymous No.1756130]

>>1755481
Really should be using power seperate from the arduino, and use the arduino to switch a mosphet on. Mind the max rated voltage and current for the LED too.

[Anonymous No.1756185]

chemically speaking, what's the difference between tin plating and tin wetting?

[Anonymous No.1756207]

>>1756079
>not using Wine
>not enjoying a massive library of models of real parts from multiple manufacturers, from valves to varistors
I love KiCAD but the workflow ain't there for simulation yet. maybe in 6

>>1756107
>anything in a QFN package is a microcontroller
it's not a tumor https://www.analog.com/media/en/technical-documentation/data-sheets/ADD5201.pdf
how's that diode to the left look, the one with the arrow pointing to it on the silkscreen? is the big coil marked 100 continuous? how about the little fuse on the board marked F2 by your not-purple lead? did you short your solder blob to the other side of the output cap?

[Anonymous No.1756237]

>>1756120
>assuming it isn't a power transistor capable of multiple amps through CE

>>1756021
Read this over, it certainly helped me better understand the process of selecting resistor values.
https://www.nutsvolts.com/magazine/article/may2015_Secura

[Anonymous No.1756252]

>>1755952
>consider the 21700 cells
I read some about them and it seems like these would be pretty ideal. I can't find any cells on digikey but ebay is flush with them and they are cheap. ~$15 for a 5Ah cell seems pretty cheap when Adafruit/Sparkfun are selling 2.5Ah 18650 cells with protection circuits for ~$20. Sure they don't have the protection circuit but they aren't that expensive... Smells like a scam to me...

So I don't understand, then. Power banks that do just this exist, and millions of laptop batteries for example have parallel configurations. Am I missing something obvious? I considered a BMS PCB but I couldn't find any that weren't targeted at series configurations. I'm willing to use excessive force to make this work and figure out an elegant solution later, anyway. It's all a learning process for me. I also found some ridiculous 13Ah cell online that I could maybe buy and forget all this parallel bank nonsense.

>>1755972
>500mA charge
I hadn't yet thought of this. The charger definitely cannot supply the current required for the 0.5C for more than two cells, these are all 2A PCBs. How do laptop batteries work, then?
>battery protection circuit
If they work as advertised then it should ideally work, otherwise I could get a overprotection circuit off ebay or elsewhere.

[Anonymous No.1756274]

>>1756073
the spice must flow

[Anonymous No.1756333]

https://hackaday.com/2020/01/23/the-truth-is-in-there-the-art-of-electronics-the-x-chapters/

So, /ohm/, will you be getting the latest art of electronics?

[Anonymous No.1756357]

>>1756252
in my experience, laptop batteries are usually 3S2P arrangements, using a 3S-specific protection/balancing/metering circuit. they are of course charged at an appropriate rate which is on the order of C/2 or so, generated by the laptop, in close coordination with the system power management circuitry. you don't need to charge at 0.5C but you will need a charger that won't time out on you before you've reached a full charge, and ideally you would like to charge them as quickly as you can without exceeding any electrical or temperature limits
anyway there are other charging ICs which can handle 6A or more going into the pack. it is worth perusing parametric selector guides to find one that's right for you. or Aliexpress, if you prefer modules

>>1756274
but Micro-Cap *is* Spice under the hood. supports like four different non-MC Spice backends plus its own internal Spice engine. and look at all those models

[Anonymous No.1756374]

>>1756333
Looks pretty nice actually. I've certainly delved into a few too many of those niche subjects.

[Anonymous No.1756394]

>order parts from digikey
>they send me something completely different from what I ordered
>bitch about it to customer service
>now they want me to drive a half hour to the UPS store to mail back $2 worth of parts
I mean they did eventually send me the right ones, but I still feel like I'm getting fucked

[Anonymous No.1756539]

>>1756252
>>1756357
Okay I'm on to a different part. I've been taking notes all day and now I'm thinking about when the batteries run out. Be gentle with me, I'm new to this. If I have a power supply, what circuit do I wire up so that the supply will run the device and charge the batteries at the same time? Clearly I need the two loads in parallel. Then, if I disconnect the power supply, the batteries will resume powering the device, right? There's one more hiccup that I plan on using a charge controller and I'd like it to be plugged in with the power supply but that's a more incidental problem I can figure out.

[Anonymous No.1756551]

>>1756539
> what circuit do I wire up
what are you powering? A flashlight is one thing, but some computers can't tolerate missing more than one cycle at 60Hz. HP's servers had this weird switch-over circuit in a clear plastic box that could maintain the power should one of the dual supplies die.

But, ideally, no circuit - the most reliable solution - just always power it from the batteries. Of course, you are trickle charging them at the same time and the batteries will bring down the voltage to somewhere under 4.2v.

The exact voltage doesn't matter because you're using a buck or boost converter.

Unfortunately placing a scope on charging controller outputs sometimes shows all kinds of weird pulsating DC bullshit which isn't good if you're injecting some of that into your load. Obviously you want some kind of charge controller in there though.

[Anonymous No.1756558]

>>1756539
you're looking for a feature called "power path management" or "load sharing" which is slightly unusual but usually part of the charging controller. don't try to power your device from the battery through the charge controller or your battery pack's life will be shortened

>>1756551
>most reliable solution
lolno

[Anonymous No.1756563]

>>1756539
>>1756558
by which I mean don't try to power your device through a non-load-sharing charge controller then through some separate converter. non-load-sharing charging ICs often have battery protection timeouts which will get in your way if you try
bq24193 is used by the Nintendo Switch for power path management, but requires a MCU to set the charging parameters. the MCP73871 is a linear (inefficient) charger and a little bit short on current, but stands alone, does what you want and is available in module form https://www.aliexpress.com/item/32779483651.html

[Anonymous No.1756600]

>>1756394
Wow you actually exist.
Every time i go to that website i wonder how they can possibly stay in business by charging 5 times more for everything they sell.
Nobody would be retarded enough to buy from them right?
How wrong I was, you people are actually real, holy shit.
How can you even function in day to day life?
Like for example for the price of one arduino uno there you can bay 7 (SEVEN!!) exactly the same ones from chinks.
There is no way you aren't a burger.

[Anonymous No.1756607]

>>1756600
basically people at arduino, digikey, sparkfun etc. are from western countries and need a lot of money to survive. They also make new stuff (new board designs, libraries, new tutorials/guides, etc.).
On the other hand chinks survive on a bowl of rice a day and just copy the designs. This allows them to sell stuff for way cheaper obviously.
Buying chinkshit is the way to go if you literally don't give a single fuck about society. If you feel like doing the right thing however, spending a few bucks extra might not be such a tall order.

[Anonymous No.1756609]

>>1756607
>If you feel like doing the right thing however, spending a few bucks extra might not be such a tall order.
I'm good, thanks fren.

[Anonymous No.1756631]

>>1756607
Incorrect. If you need to buy an arduino, you get one from the chinks because arduino boards are basically all open-source. Same applies to high-tolerance passives and the like. If you care about quality and repeatability in something like a higher-end ADC, then you're better off buying from a more reputable seller in the western world. LCSC is apparently a good go-between.

The free market should dictate which businesses can survive, not shallow feel-good social movements. Taxation is theft and columbine did nothing wrong.

[Anonymous No.1756640]

>>1756631
the free market is a race to the bottom and consumers don't know what's best for them

[Anonymous No.1756642]

Hola

[Anonymous No.1756652]

fuck i need to make some inventory system, i just ordered a bunch of pots and when they arrived i found out i already had 20 of them in stock in my parts box

[Anonymous No.1756671]

>>1756600
i wonder if posters like this even do anything with electronics other than arduino. SMD ICs are cheaper on digikey than ali unless you're buying a very limited set of them, and while you're there you can get all of the specific passives you need for an extra $1 shipped in 1/10th the time. i feel like you'd have to be really poor to take issue with that.

[Anonymous No.1756673]

mcg is dead right now or I'd ask there. My batch of stm32 clones finally came in from China but due to a miscommunication between my friend and I, I didn't think I needed a USB<->Serial breakout.

From what I'm seeing on amazon, it's a bit of a crapshoot. Some of them are good quality but ludicrously expensive.
Others aren't even actual usb-serial converters, but microcontrollers that are trying to bitbang it and wind up fucking up the transfer at best.
How do I know what to choose?

[Anonymous No.1756675]

>>1756673
I do have an arduino uno, would it be possible to jury rig it into a usb-serial converter?

[Anonymous No.1756681]

>>1756675
can't you use generic ftdi like for programming an arduini mini for example?

[Anonymous No.1756683]

>>1756551
I'm building the NODE handheld computer: https://n-o-d-e.net/terminal_3.html
But with my own twists to make it harder.

So by trickle-charging you mean a circuit like what I've described?

>>1756563
I see. I just checked against the Adafruit part the article specifies and I see it's got a load-sharing chip on it. To do this incrementally I'll probably suck it up and buy the $20 chip and then try to replace it later.

I've been reading that the 0.5C rule of thumb is not actually that hard-and-fast and e-bike guys say that charging at 0.15C is okay and even beneficial for the pack. If I shoot for 8 or 9Ah then the board can supply that with its 1A current output. Later on I can try my hand at making a power circuit myself. It'd be really cool if I could power the whole thing from USB-C so I want to look into that.

By the way, thanks for the help, you guys are very informative and I appreciate you tolerating the asinine questions.

[Anonymous No.1756689]

>>1756681
That's what I mean, I can't tell which ftdi breakouts are good and which are bullshit. I've read reviews on some that said they weren't FTDIs, but microcontrollers trying to emulate them. They fuck up the clock/timing and wind up just corrupting the data.

[Anonymous No.1756696]

>>1756689
interesting
when i was buying ftdi dev board i just typed ftdi into ali search and got the first one that had lots sells since if the product is a scam then it won't have very high sold count and you can easily see in the user reviews
anyway even if the one i have is a cheap fake it's been working fine for over a year and can program my mini pros no problem

[Anonymous No.1756720]

>>1756683
Turns out I don't have to buy the expensive Adafruit board because ebay has clones for $3 that use the same MCP73871 chip. That's pretty ideal, all I need to do is put a boost converter on the output to guarantee enough voltage and I'm good to do. It will charge the batteries very slowly, though. Is there anywhere I can find ICs like this? If it's a whole IC I can rig up some perfboard and do it myself, I'd like to use USB-C and leverage that ~4A power input to hopefully charge the batteries faster.

[Anonymous No.1756741]

>>1756683
>trickle-charging rasberry pi

Yes, similar to what you described. I have set up rasberry pi's to act as backup web servers from deep cycle lead acid cells (they will run for a few days serving "static" content to make it look like we're not down :-) When the AC power is cut, the batteries just keep outputting, and everything stays up. We put these in the telecom rooms because they're so small and the pbx and networking already had the batteries.

The system is inherently reliable with no switching over components, as billions of similar configurations has been used for many years — in cars.

Obviously iPhone (or whatever) can also do this, but it's a lot more sophisticated, as you've seen with a $20 chip or other hard-to-find components (or will be hard-to-find in the future). I realize that l-ion cells have different charging/discharging needs, but as a general concept it's fine. You are right to look at the e-bike technology, that's very similar.

It's easier on the batteries to charge them at lower rates... why stress them out if you're not in a hurry?

If it were me, I'd get an 18650 battery holder and manage the cells manually like AA's. This is also becoming popular, then you have the option to switch them out manually with fully charged backups if you really need to.

The key is the flexibility of the dc/dc converter — these run the world. My graphics card is basically the GPU and about 10 dc/dc converters that take 12V inputs and turn it into 1.1V at on the order of 100A (in total).

[Anonymous No.1756852]

>>1756600
yet convenience stores still exist and do plenty of business. sometimes you need $10 worth of components right away and don't want to pay $20 to get it here via DHL
>buying arduino boards from an authorized distributor
3/10 made me reply. post again when you make your own MCU boards

>>1756671
>SMD ICs are cheaper on digikey than ali
5 cents for (100) 0603 resistors at LCSC. 10x that at digikey

>>1756652
I've never had that happen, nope
>looks at 3 bags of 100 10k 0603 resistors
there's a few inventory manager webapps out there already, both hosted and locally-installable, if you aren't excited about the process of writing one

>>1756673
you'll be much happier with an ST-Link v2 dongle so that you can debug your programs with gdb when you grow out of duino

>>1756720
the MCP73871 is available on ali, fwtw
>perfboard
they're QFN or WLCSP packages and some of them are especially layout-sensitive
in general, chipmakers probably won't have much in the way of linear chargers because there's little/no commercial demand for such a thing
EEVblog's been going down the USB-PD rabbit hole lately for his µSupply project, and it is a great big rabbit hole. often you end up with a separate MCU just managing the power delivery communications. benchmarq (now TI) has a lot of nice battery management ICs in the switched-mode load-sharing charge controller space, which you can look up at TI's website using their parametric selection tools. start here and check/uncheck boxes as you desire
http://www.ti.com/power-management/battery-management/charger-ics/products.html#p1152=1;1&~p2192=Power%20Path;USB%20C/PD

[Anonymous No.1756861]

how long did it take before everything just "clicked" for you guys and electronics
I feel like I'm constantly, consistently missing some key intuition that makes most everything incomprehensible

[Anonymous No.1756901]

I'm reading about wiring in houses and want to check if I'm understanding this right. There are a bunch of dedicated circuits in a house, and each dedicated circuit has a few outlets on it. If a circuit is 120 volts and 15 amps, then the sum of power draw (in watts) of all the devices on that circuit cannot exceed 120*15 = 1800 watts. Is that close to right?

I found out I can use a "kill-a-watt" to measure power draw on one outlet, can I somehow measure how much is being used on all the outlets on a circuit?

[Anonymous No.1756917]

>>1756861
it's a big field
it took a year or two and a decent mentor for it to click for me. key insights that made me much more effective:
>if the basic laws don't appear to hold, it's because you're not accounting for something in your model
>bipolar transistors: in active mode, one gulp of current flowing between b and e causes beta gulps of current to flow between c and e. if beta gulps of current are not available to flow between c and e, the transistor is in saturation mode

>>1756901
1. yes
2. you need a clamp-on current meter at the distribution panel
but see /ohm/ RULE 0

[Anonymous No.1757021]

>>1756607
If you buy stuff from western countries you are supporting theft and padding of exec salaries for hookers and blow parties. The real cost is closer to what a chinese company will sell them to you for.

The labor cost idea is bullshit. On both sides, factories are just about automated to the point of a "lights out" line — there's hardly anyone around — and it's certainly true of pcb "products" like arduino where they took a couple of application note circuits and autorouted a pcb for it. Big whoop.

Besides, people have the whole thing backwards... at what point do you think it will make sense (cost-wise) for a huge market like China to start buying U.S. made products? I don't mean "assembled in the U.S. from global parts", or "designed in the U.S.A" <fine print: made in china>. Even if we do 10 times better, we still won't get there.

The chinese think GENERATIONS ahead. Western companies only think as far as the current 2-year tenured CEO's golden parachute clause or government bailout. Wake up people, and start to compete.

[Anonymous No.1757026]

>>1756861
Just one physics paper before I felt comfortable with it, but I'd had prior exposure (mainly from /ohm/). The second paper helped a bunch too.

I (and others too in all likelihood) look at a circuit as a series of building-blocks chained together. The actions of any block should be separable from the rest and describable with a fairly simple algorithm or equation. For cases where input impedances are trivially high and output impedances are trivially low (digital or op-amp circuits) you usually only need one equation. But for transistor circuits you often need both a small signal AC and a DC model. There was a YouTube channel that did AC and DC analyses of common transistor amplifier topologies, which might be helpful for you, though I don't dabble much in transistor circuits.
Understanding feedback loops is one of the most important small thing to learn in my opinion, it covers things like PIDs and SMPSs and oscillators, filters too.

[Anonymous No.1757034]

>>1753417
I was looking at the fluke 117 as a good multimeter for small circuits, would that be a good choice for electronics and general house electrical work?

[Anonymous No.1757041]

>>1757034
> fluke 117.
Yep, it's got more features than my 8020b that I've been using for decades for both electronics and electrical. Back then, there wasn't a lot of choice though.

[Anonymous No.1757043]

>>1757034
It's a solid choice, but check out the eevblog forum multimeter spreadsheet
>inb4 autoranging is a bad thing

[Anonymous No.1757061]

>>1757043
>autoranging
it's terrible, Muriel

>>1757021
this anon has a clue

[Anonymous No.1757141]

>>1757021
>he buys his components directly from chinks
enjoy your Coronavirus m8
cough cough, amirite?

[Anonymous No.1757143]

>>1757141
No it's called kung flu

[Anonymous No.1757144]

If I want to specify the LUT mask and the interconnections of the LUTs in an FPGA can I do so with UDPs in verilog? In other words, does the truth table and I/O defined a verilog UDP have a one-to-one correspondence with the physical device?

[Anonymous No.1757150]

>>1757021
>If you buy stuff from western countries you are supporting theft and padding of exec salaries for hookers and blow parties.
That's a separate issue, and as an aside I'm not sure digikey CEO has a yacht to fuck whores

>pcb "products" like arduino where they took a couple of application note circuits and autorouted a pcb for it. Big whoop.
yeah, also Columbus just took a bunch of ships and sailed straight west. big whoop

>at what point do you think it will make sense (cost-wise) for a huge market like China to start buying U.S. made products?
never, because the party will probably block imports or somehow fuck with the free market to give chinese companies an unfair competitive advantage. Like they've been doing since forever, while the WTO looks the other way.

>Western companies only think as far as the current 2-year tenured CEO's
same for western consumers apparently. Money you spend buying chinese products is money that leaves your community possibly forever, or until the chinese guy you gave it to returns as a tourist with his family and his kid shits in your city's public fountains.

>Wake up people, and start to compete.
indeed my friend, wake up to the reality that western companies will always struggle to compete with chinese ones that are partly funded by their government to lower prices even further and flood western markets, destroying western companies (see bicycle and solar panel industries of europe), and that have negligible salary and safety costs.

How you guys think you're the smart ones for giving more power to antagonistic entities baffles me. And for what, a few dollars more?
"Fuck other people, I got mine" I guess

[Anonymous No.1757151]

>flatmate asks me if I'm otp
>ask if he means "one time programmable"
apparently it means "on the piss" haha isn't that funny ha

[Anonymous No.1757154]

>>1757151
haha you are such a geek i hope coronachan comes for you soon

[Anonymous No.1757156]

Is there a recommended way to make a raspberry pi solar online? (zero w) or is solar panel hooked up to usb battery pack the way to go

[Anonymous No.1757161]

>>1757156
There are solar charge controller ICs, and likely ones small enough for this sort of use. Even without MPPT (maximum power-point tracking) you need a buck or boost (or both) converter, and preferably also a "losing power" instruction sent to one of the GPIO pins when the solar cell voltage gets too low. If you also need a battery then those USB power banks with solar panels on top could do the job, but last I checked those circuits are rather rudimentary. I'd at least look on mouser/digikey/arrow/farnell/lcsc for MPPT switching controllers, either ones that are specifically made to charge 1S lithium-ion (in conjunction with a normal boost converter), or ones that are made to output 5V, depending on whether you want a battery or not.
Last I checked TI was a leader in solar charging controllers and the like.

[Anonymous No.1757203]

Is there anywhere I could look for confidential datasheets that can't be found on google?
I'm looking for the GPEL31XXA Programming Guide but at this point I'd be fine with having a programming guide from any of GeneralPlus' 32-bit MCUs.

[Anonymous No.1757214]

I'm trying to find the name of a type of switch but I can't seem to figure it out. I'm looking for a type of switch that clicks on for just a moment when pressed then turns off even if you keep holding the button. I've seen them before but describing it on google doesn't seem to be helping. I know I could make a very simple circuit to do this with a normal push button, but I'd rather have a mechanical action for it and I'm trying to avoid any unnecessary electronics.

[Anonymous No.1757218]

>>1757161
I underestimated how difficult electronics was I figured positive from solar panel to negative of an 18650 then negative of panel to positive of 18650 then 18650 to usb or panel to usb to battery pack to pi

[Anonymous No.1757277]

>>1757214
> momentary switch
In all likelihood you would still have to debounce it
Inb4: monostable mulivibrator

[Anonymous No.1757283]

>>1757156
I use off the shelf car battery solar cells used to keep your battery topped up. They come in a variety of sizes. Then I remove the little circuit board from a 12v USB adapter and plug it into the pi. It uses a surprising amount of juice though.

[Anonymous No.1757294]

I have a 5v and 3.3v (probably just 5v run through a regulator) source. Is there a way to combine them to get 6+ volts. They are tied to the same ground.

[Anonymous No.1757298]

>>1757294
What is the nature of the source? If you can tap off the output *before* the regulator, that's gonna be 6 or 7 volts probably.

[Anonymous No.1757360]

>>1757294
>They are tied to the same ground.

voltages cant add or subtract if they share a ground. if you can get to the input of the +5 reg, then connect it to a 7806 or 78L06 to generate 6V.

[Anonymous No.1757363]

>>1757294
>They are tied to the same ground.

voltages cant add or if they share a ground, only subtract. if you can get to the input of the +5 reg, then connect it to a 7806 or 78L06 to generate 6V.

[Anonymous No.1757364]

>>1757294
>They are tied to the same ground.

voltages cant add if they share a ground, only subtract. if you can get to the input of the +5 reg, then connect it to a 7806 or 78L06 to generate 6V.

[Anonymous No.1757365]

Let's say i have 3 traces on a pcb.
Trace 1 and 3 have to be at lest 1cm away from each other because they are high voltage around 300 and they will arc.
Does that mean the trace (copper) 2 between them (not connected to anything) has to be removed? or can it be left there?

[Anonymous No.1757368]

>>1757365
Distance between 1 and 2 in the picture is 1cm

[Anonymous No.1757378]

>>1757365

you've created a distance for isolation, and then nullified it by putting an electrical path in between. not very clever. remove it!

[Anonymous No.1757381]

Oh my god i'm such a retard!
I just snapped two end mills plunging them into the PCB with a probe command, and it took my two fucking tries before it clicked in my brain
Of course the fucking probe won't connect when i am probing inside of the isolated area, for FUCKS sake
Thank god chink endmills are cheap

[Anonymous No.1757400]

>>1757381
turned out pretty good
i fucked up at the end a bit when i was cutting the board shape out of the larger pcb and managed to break off a piece of atrace, but that can be easily fixed, but otherwise really good.
Even the area on the right for mains power where i removed copper with a bigger tool for faster speed ended up pretty decent. Not as good as a chink fab house obviously for i have a fully functional prototype in just few minutes of milling and all i needed was one endmill and two drill bits

[Anonymous No.1757421]

>>1757144
>primitive
not an __asm directive, no, it's just another way to specify logic, which can be optimized and synthesized like everything else. maybe you would like to write your own IP in EDIF instead, if placement is that important to you

>>1757218
lol no

>>1757400
berry gud

[Anonymous No.1757508]

>>1757218
Lithium ion cells require specialised charging circuits. The TP4056 is a very common circuit for this and if you look at the datasheet you should find the minimum and maximum voltage it will run on. If the voltage range is wide enough, you'll be able to just chuck the solar panels into that. But it won't be very energy efficient compared to a proper buck or boost circuit, and a buck or boost circuit is less efficient than a MPPT circuit.

>>1757381
Aren't you supposed to probe the board before milling it?

[Anonymous No.1757517]

>>1757400
like, it looks like you milled off a bunch of the copper? Why bother??

[Anonymous No.1757524]

>>1757517
mains voltage, duh

[Anonymous No.1757529]

>>1757524
Well, you got 20,000 volts worth of clearance in some places... a bit excessive. Just mill out slightly wider gaps and ground the excess copper. That's safer, too.

[Anonymous No.1757551]

hey my weed pen battery and tank are too far away from each other because of apparently a bad design. I am wondering how to make a better connection. It is a magnetic system. You drop the cartridge in and it stays. (indigo pro is the brand).

I was wondering on any ideas on how to fix this? thanks

[Anonymous No.1757571]

>>1757529
not mine, but while you are correct, observe the imperfect removal. when it comes to safety, too much beats answering very difficult questions from insurance jews

>>1757551
sounds like a mechanical problem. contact the maker

[Anonymous No.1757581]

>>1757218
if you need load sharing i.e. to use the device while you are charging the battery, the TP4056 is not the part for you. see the MCP73871 instead, which fortunately is also available on chinkboards
other anon is right in that you will probably need a boost on the 73871's output to bring it up to the level the pi probably demands
or you could use one of the many raspi "UPS" boards already out there and let someone else handle the complexity for you

[Anonymous No.1757589]

>>1757571
> safety
all according to design

[Anonymous No.1757625]

>>1757589
Based.
Also isn't that FR-2? No wonder it looks like shit.

[Anonymous No.1757657]

>>1757589

[Anonymous No.1757663]

>>1757421
>write your own IP in EDIF instead
Right, so say I'm autistic and want to specify the LUT mask and the interconnections of the LUTs in an FPGA. Now, you're telling me I can get that done with an assembly language? Great, what kind?

[Anonymous No.1757672]

>>1757663
non-portable, existence and format depends on your FPGA and synthesis tools. distributed RAM configures a LUT as a (typically) 16x1 RAM and some synthesis tools/engines may expose it as a primitive, with which you can do the obvious thing of driving the address lines explicitly and doing what you will with the data out, or can infer distributed RAM from behavioral Verilog. not that you will necessarily have any documented way of manipulating the interconnect at this level and if you do it sure as fuck won't be portable

[Anonymous No.1757682]

>>1757508.
>Aren't you supposed to probe the board before milling it?

You need to probe when you change bits to set z0

[Anonymous No.1757685]

>>1757682
>change bits
Oh, I thought you'd have those little collars on each bit or some other way to know the z-offset of each bit. That way you could use an actual probing tool for x, y, and z probing in case you need to do some 3D milling.

[Anonymous No.1757718]

>>1757517
I have to do it since that side is mains voltage
>>1757529
>Well, you got 20,000 volts worth of clearance in some places
The pcb looks big in the photo but it's not, in fact the spaces between the mains traces are so small i wouldn't even pass safety ratings, the are supposed to be like 2mm but are barely 1mm
>>1757625
unless i got scammed by chinks it's fr4
also the copper removal was done with a large end mill for speed, otherwise it would take forever that is why the traces aren't as pretty as the vbit ones on the low voltage side
>>1757685
usually chinks only put those on drill bits for some reason

but either way, i don't have to z level for drilling and board cutouts but i have to always level for anything that does copper removal because i need to have a precision within 0.03mm tolerance,

[Anonymous No.1757719]

>>1757718
>unless i got scammed by chinks it's fr4
Looks too brown to be fibreglass, but maybe that's just all the burrs. Gonna solder-mask it?
>usually chinks only put those on drill bits for some reason
You could also put those stop-collars with set-screws in them on each bit, picrel

[Anonymous No.1757720]

>>1757589
please don't bully my smol desktop mill, it's not a $5 000 000 Haas monstrosity, but it does the best it can!
I will take ragged edges over shitty fecl3 and toner transfer anyday

[Anonymous No.1757723]

>>1757719
>You could also put those stop-collars with set-screws in them on each bit, picrel
that would cause too much vibration and it's not like probing is ton's of work, i only put a clamp on the bit and hit a button that runs a probing macro and as a bonus get a much more precise result that with a ring

I don't really bother with solder masking.
Basically the first board i make is a prototype to test if it works and if it does i order 10 of them from chinks for $2, that are flawlessly made, with silk screen and vias and everything and call it a day, making anything more than prototypes at home is a waste of time when chinks exist in my opinion
but that being said, when i need only one copy a pcb and i need it NOW, i do want to try doing my own solder paste. I built this ghetto uv exposure chamber and i have a tube of solder mask lying around. I never actually tried using it yet, but i want to since I want to try a new idea where i apply solder mask to the pcb BEFORE i cut any traces into it and then mill it and expose the pads by dragging, so that way i will get perfect solder mask coverage with almost zero work needed

[Anonymous No.1757725]

>>1757723
>diy for prototypes, chinks for final products
Ah, I can agree with you on all that.
>1W LEDs
Maybe I should have gone with those instead of a 100-pack of 5mm ~60mW LEDs.
>i want to since I want to try a new idea where i apply solder mask to the pcb BEFORE i cut any traces into it
But it's nicer to have solder-mask on the bare fibreglass, not just over the copper. Do you have one of those spring-loaded end-mills for eating off solder-mask from copper?

[Anonymous No.1757726]

I'm working on giving my cars stock stereo bluetooth so I dont have to use a shitty gross aftermarket deck.
I just soldered on an aux cable to the tape deck which works and sounds great.
Now I need to add something that can recieve bluetooth input, output via 3.5mm, and be powered by the cars 12v DC.
I've done some shopping around and there seems to be a large selection of chinese blue tooth transmitters that are just simply red black for power, and a 3.5mm cable.
I could take my chances with one of those, and not be able to control the music without using my phone.

Could I make something better with a raspberry pi or an arduino?
Id also want to use pic related as its display.
The car is from the 90s so I don't want a big touchscreen.

[Anonymous No.1757727]

>>1757725
Solder mask is there to protect the copper, and to help with soldering, because if you have only 0.2mm spaces around pads with no solder mask you get a lot of annoying solder bridging
Having solder mask on the fiber glass looks nice visually but serves no practical purpose as far as i know and it's not like will be entering your pcbs in art competitions

>Do you have one of those spring-loaded end-mills for eating off solder-mask from copper?
Yes and the fucker wasn't cheap or easy to get, to get good quality i had to order it from a smol machinist in Austria for $50 (and that is actually very cheap as far as these tools go)

[Anonymous No.1757729]

>>1757727
Well when it comes to milled boards I'd put solder mask on the fibreglass to prevent all those fibres from water ingress, as well as the looks.

[Anonymous No.1757730]

>>1757726
This one looks like it would work.
It doesnt have a display but it does have buttons.
Which I could probably wire to my broken cruise control buttons on my steering wheel.

It would be ideal though if I could get this with a 20x4 display that read:
Artist
Album
Song
Time elapsed - Song length

I dont know the first thing about arduinos or raspberry pi's though...

[Anonymous No.1757731]

>>1757726
you are overcomplicating it, just get a car stereo that supports bluetooth, the selection on the market is massive

Or if you absolutely must be autistic, you can get the chink bluetooth -> 3.5mm module and then to play music you connect you phone to it, but also you buy a remove music controller that attaches for example to the stairing wheel which will connect to your phone (like handsfree for example) and then you control the volume and shit with that

[Anonymous No.1757732]

>>1757731
>remove
*remote

[Anonymous No.1757733]

>>1757730
I don't know, I've heard bad things about the built-in BT amps, and that it's better to go for an external BT receiver. But that one does say 4.0 so it might be fine.

[Anonymous No.1757734]

>>1757731
I already have a bluetooth stereo that I bought a few months ago to replace the stock one.
But its so ugly, fails to connect frequently, and doesnt agree with my cars amp so there is lots of background noise hiss.
I 100% know I'm over complicating it. I could go buy another deck that will look a different flavor of ugly in this old interior, and It will probably have similar issues.
But Id much rather work on a perfect "stock look, aftermarket functionality" solution to my tism.

[Anonymous No.1757735]

>>1757734
also don't forget the sound quality from the $2 chink bt adapters will not be exactly audiophillic

[Anonymous No.1757737]

>>1757735
That is true...
Im not too much of an audiophile.
but if it of the same quality as the aux input straight into my phone then I'll be happy.

The deck I have right now is actually the second one I bought. pic related
I first bought a touchscreen one also by kenwood, but it was so bad that I had to return it.

[Anonymous No.1757749]

>>1757672
I think I can work with that. Thanks.

[Anonymous No.1757862]

Hey /ohm/
I think my chink arduino uno clone has the 3.3v and 5v output switched, however since I have no multimeter at hand, I only have the arduino itself to measure the voltage.
The measured voltage values are:
>0.00v for GND
>3.10v for 5V
>4.96v for 3V3

Should I trust these measurements?

[Anonymous No.1757866]

thinking of pursuing an Electronics Mechanic trade degree. What kind of demand is there for Electronic Mechanics?

[Anonymous No.1757915]

How does one estimate the battery life of a circuit? I made a simple LED indicator and used a 2xAAA battery holder because it's what I had available, but I have no idea how long it will actually last or what size battery I should actually be using.

I have a multimeter. Is there a simple process and equation to measure how much the circuit draws while closed and how long it will take to drain the batteries?

This is probably really dumb to ask but I'm new to electronics.

[Anonymous No.1757929]

>>1757862
that is impossible
fartduino has a simple linear regulator that takes 5v, heats away 1.7V and the rest is the 3.3V supply.
It cannot raise voltage and it does not work in reverse
I bet you are using analog read, ?
1023 is NOT 5V it means input voltage.
You would have to switch to the internal 1.1V reference and then measure the voltages gain through a voltage divider
just get a fucking Fluke you poorfag

[Anonymous No.1757931]

>>1757915
Measure the current it drawing and then
2000mah/current you measured in ma = hours of how long will the thing run

So if the led is drawing 20ma it will run for 100 hours

[Anonymous No.1757936]

>>1757931
and obviously if the batteries are in series it will be half the hours

[Anonymous No.1757937]

>>1757931
Thanks, that makes it really simple for me. I appreciate it.

[Anonymous No.1757952]

Just remembered I already have a bluetooth reciever seen here >>1757737
Unscrewed the case and it looks like it should work.
3 wires for what I assume is + - ground (will verify before I do anything)
Only problem is the screen. From what I remember, all it did was show the company logo, and maybe a phone icon if I got a phone call. Wonder if there is a way to reprogram the board?

[Anonymous No.1757957]

>>1757915
once you have determined the current draw of your load with an ammeter, look up the datasheet for your battery(ies), which will contain a graph of load current vs. capacity. divide capacity (mAh) by load (mA) to get hours (h). run time will vary by battery type of course. batteries may exhibit significant internal resistance that could greatly change the current draw as well

>>1757952
maybe. go look up datasheets. doubt you have the stamina to even put rockbox on it

[Anonymous No.1757962]

>>1757952
Almost forgot!
I want to relocate those buttons. Was thinkin of replacing the buttons with some wires that lead to my broken cruise control buttons on my steering wheel. (pic related) Is that possible? Wires would probably be about 6' long tops.

[Anonymous No.1757971]

>>1757929
>You would have to switch to the internal 1.1V reference and then measure the voltages gain through a voltage divider
Just did that, turns out everything was fine.
>poorfag
guilty as charged

[Anonymous No.1757976]

>>1757962
kek, aren't you a lucky cunt this was uploaded literally 5 hours ago
https://www.youtube.com/watch?v=n_b_0BWP0mI

[Anonymous No.1758003]

This is from an antisurveillance device.
Look at those beautiful 90 degree traces, you don't see such high level art on the modern pcbs anymore.
And this actually detects radio signals and shit which makes those right angle traces even better

[Anonymous No.1758007]

how bad of an idea is it to link 18650s together with copper tape rather than spot welding them

[Anonymous No.1758011]

>>1758007
from what I can read, the adhesive can't handle the current that can be demanded from the battery bank. Is that about it?

[Anonymous No.1758067]

>>1757976
I went and installed the stock stereo and the bluetooth reciever and got this https://streamable.com/bhp7u
I assume its the same cause as the video you linked.
Is there a way to verify without an oscilloscope?

[Anonymous No.1758113]

>>1757971
Get an ANENG 8009, it's pretty damn good for the price.

[Anonymous No.1758114]

>>1758067
if it's all sub-20kHz you could record it with a sound-card

[Anonymous No.1758115]

>>1757971
are there any electronics projects that will let me fuck reisen?

[Anonymous No.1758117]

>>1758007
figure the cross sectional area and check an ampacity table for a wire of equivalent cross sectional area

>>1757976
based and letsgetstartedpilled. that belongs in the next OP

[Anonymous No.1758119]

>>1757862
Put them under load. If they dip, throw it out. Otherwise, they were probably just swapped

[Anonymous No.1758120]

Does anyone know if there's a proper name for this kind of connection/pin? I need to buy a short extension wire with the same connector on both sides but i'm not sure what to search for.

[Anonymous No.1758127]

>>1758120
post banana for scale
also post more information about the application

[Anonymous No.1758140]

>>1758011
Is the adhesive conductive? How would you connect the copper itself to the cell ends?
Either get a case with spring-loaded battery contacts, or buy cells with tabs already welded onto them. Or go for lipos instead.

[Anonymous No.1758148]

>>1758127
Button cell battery and 50p piece for scale.

As for the application its a pretty simple circuit with an LED light, a battery pack and a button, encased in a metal tube like a flashlight. Only issue is that the other wire that's supposed to connect to the battery is too short to pull out of the metal casing, so I need some kind of extension that I can connect to that wire, and then connect to the battery.

[Anonymous No.1758193]

>>1758148
show the connector face-on. is that the battery end, or the lamp end?

[Anonymous No.1758207]

>>1758148

looks proprietary. saw something similar on an Ikea power supply, but not similar enough. hermaphrodite connectors arent very common, so you're better off extending the cable using solder, instead of making an extension cord.

[Anonymous No.1758249]

>>1758067
>I assume its the same cause as the video you linked.
How about you stop pressing the horn? that would help

[Anonymous No.1758250]

What's your favorite type of centipede? I like this one.

[Anonymous No.1758251]

>>1758067
easy way to check, try to power the the bt adapter from a 5v battery

[Anonymous No.1758252]

>>1758251
but the website says its 12V DC
does it not matter?
I'm new to electronics

[Anonymous No.1758258]

>>1758252
>does it not matter?
it does, i assumed it was usb dongle, they are always 5V, you have to use 12V battery then. If it was made specifically for use in cars it makes sense since car batteries are 13V

[Anonymous No.1758262]

>>1758258
Yeah its one of these cigarette lighter fm transmitters.
Its got an aux port for input/output so I can use it as only a bluetooth reciever and and plugging it into my tape deck mod.

Anyway, if I setup a battery for it and it fixes the interference, wouldn't I need frequently replace that battery?
Id prefer a more invisible no maintenance solution.
That video talked about a DC/DC converter to isolate the grounds.
Would that be an easy inline fix? or do I need to make an actual board for it?

[Anonymous No.1758264]

>>1758250
CDIP EPROM? I'd tend to agree with you. Has anyone put that in a circuit with UV LEDs for self-erasure?

Also where can I buy a non-volatile D or T or JK flip-flop? So I can use 5c pushbuttons instead of 50c toggle switches.

[Anonymous No.1758266]

>>1758262
the answer is in the video. watch it again and again until you have the answer

[Anonymous No.1758267]

>>1758264
you make your own with a microcontroller because of all the tradeoffs involved in non-volatility

[Anonymous No.1758269]

>>1758262
the battery is a cheap way to test if it fixes the problem, remember you don't know if the common ground is causing the problem yet
if it works then you can buy an isolated power source

[Anonymous No.1758318]

I purchased this smol soldering tip for soldering smds but every time i put solder on the tip it magicaly sprints up to the wider section of the cone and none is left on the tip and when i add more i end up covering the smd with a solder blob
so wat do?

[Anonymous No.1758353]

>>1758193
It's the lamp end. The battery pack has the same connectors, but there's no length to them so I can't reach them into the case to plug in the other wire.
>>1758207
I don't have easy access to a soldering iron and even if I did, the wire is too far in for me to effectively do that. There's about enough space for me to reach in with a couple of fingers to attach a wire, but not to do any kind of work with a soldering iron.

[Anonymous No.1758364]

>>1758363
>>1758363
>>1758363
>>1758363
>>1758363
>>1758363

[Anonymous No.1758370]

>>1758264
>a non-volatile D or T or JK flip-flop?

any CMOS chip can be made non-volatile if you add a coin cell, diode and resistor, like in the top RHS of pic related.
the 100K resistor makes it so that the diode drop is mostly eliminated.
battery will last 1000 years. give or take.

[Anonymous No.1758613]

>>1758267
>>1758370
no fuck you I'm using magnetic cores

[Anonymous No.1758653]

>>1758318
if you can still get small drops on the iron, don't worry too much about it. given flux, molten solder will flow to where it's warm enough. otherwise use smaller solder wire

>>1758353
looks pretty uncommon. you might be able to afro-engineer it with 2.54mm-pitch "Dupont" pins and some self-fusing tape for mechanical support. or you can splice some more wire into the end you have coming out. that heat shrink seems like a bit of a clue

[Anonymous No.1759321]

>>1757021
>The chinese think GENERATIONS ahead

there is literally no proof of this anywhere thanks for the laugh

[Anonymous No.1759654]

EXIT
>>1759651
THIS
>>1759651
WAY
>>1759651
PLEASE
>>1759651
EXIT