/ohm/ - Electronics General: higher output current edition

[Anonymous No.2418059]

Thread oxidised:>>2410396

>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.

>Incredibly comprehensive list of electronics resources:
https://github.com/kitspace/awesome-electronics
Additional resources below:

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

>Don't ask, roll:
https://github.com/Rocheez/4chan-electronics-challenges/blob/master/list-of-challenges.png

>Archive of Popular Electronics magazines (1954-2003):
https://worldradiohistory.com/Popular-Electronics-Guide.htm
>Microchip Tips and Tricks PDF:
https://ww1.microchip.com/downloads/en/devicedoc/01146b.pdf
>Li+/LiPo batteries required reading:
https://www.elteconline.com/download/pdf/SAFT-RIC-LI-ION-Safety-Recommendations.pdf

>Books:
https://libgen.rs/

>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 (arguably has minor issues with mains grounding)
Horowitz and Hill, The Art of Electronics

>Recommended Design/verification tools:
KiCAD 6+
Circuitmaker
Logisim Evolution

>Recommended Components/equipment:
Octopart
eBay/AliExpress sellers, for component assortments/sample kits (caveat emptor)
Local independent electronics distributors
ladyada.net/library/procure/hobbyist.html

>More related YouTube channels:
mjlorton
jkgamm041
EcProjects
Photonvids
sdgelectronics
paceworldwide

>microcontroller specific problems?
>>>/diy/mcg
>I have junk, what do?
Shitcan it
>consumer product support or PC building?
>>>/g/
>household/premises wiring?
More rules-driven than engineering, try /qtddtot/ or sparky general first
>antigravity and/or overunity?
Go away

[Anonymous No.2418070]

>>2418059
I know devices can get hot from wireless charging like a Qi pad. Would it be possible to put a coil inside a mug/thermos/cup and instead of charging a device use it to heat up, for example, a cup of coffee by setting it on the pad?

[Anonymous No.2418080]

I don't quite understand what this ic's supposed to do

[Anonymous No.2418090]

this is a working ar contact lens
breddy gool

[Anonymous No.2418097]

>>2418080
for something like a VU meter, says it comes in linear or log

[Anonymous No.2418161]

How do I build a constant current LED array to allow for use of a PWM controller?
Should I bundle the cobs together in pairs of 7 until I reach 24V? what resistor and other circutry should I choose?

[Anonymous No.2418162]

>>2418161
meant constant voltage

[Anonymous No.2418164]

>>2418070
That's similar to how an induction heater works, they just heat up the pot/crucible with eddy-currents. But if you want yo use Qi charging, you'll likely need to do some sort of impedance matching to keep it happy and not overloaded. This either means fine-tuning an eddy-current disc (or coils) to have the right amount of impedance, or just using a normal Qi charging coil connected to one of those flexible heating elements.

>>2418090
How do those actually work? Can they display images on your field of view that are in focus? Getting any kind of pixel-matrix projector in there would likely be a nightmare, and IIRC you can't just put a thin film transparent display and expect it to be in focus without REALLY getting into fresnel optics if not metamaterials. But maybe a laser diode and 2-axis MEMS gimbal could fit. I'd be down with a vector display.

[Anonymous No.2418165]

>>2418164
>How do those actually work?
the thing on top is a tiny micro oled display
they probably then do some laser shit to get the lens into focus

[Anonymous No.2418168]

how to make a buck convertor from discrete components salvaged from waste electronics?

[Anonymous No.2418169]

>>2418161
So you want an LED array that you can put a fixed-voltage variable-duty-cycle supply to? In general, resistors will be more likely to work with any frequency you can think of (some constant-current drivers may get strange being PWM'd) but the decision more comes from what kind of power the LEDs are. Below 5-10W or so resistors make more sense, above it constant-current drivers make more sense. With some give for how distributed the setup is, LED strips go up to 20-40W just on resistors since the heat is well distributed.
(possibly a wholly useless paragraph)

While anything within spec goes with constant-current drivers, usually you'll want at least ~15% of your total voltage drop for resistors if you go for resistive current limiting. So at 24V you'll want at least ~3.6V to be dropped across a resistor. You could round that down to 3V and go for 7 LEDs like you thought and it's probably fine, but I'd maybe err on the side of caution with 6 LEDs, or at least do a test run to see if it enters thermal runaway with a bit of hot air gun coercion. I don't know if there's an easy way to predict that, but I guess the schockley diode equation with correct coefficients plus some thermal bits and bobs would lead you to a usable differential equation if you're really interested.
15% is just a number off the top of my head since 10% sounds a bit low, but I'm hardly an expert.

Calculating a resistor is simple, and there's probably dozens of online calculators for that, though ones for multiple series LEDs are a little more elusive. The equation is as follows:
>R = (V_source - (N_led * V_led)) / I_led
For each string.

Proper constant current is still worth considering if you don't have a large amount of series strings. Especially if you just chuck an LC filter on the PWM, or have a solid power rail to the CC drivers and send the PWM signals to the reference bit. Shouldn't be too tough using TL431s or LM317s.

[Anonymous No.2418170]

>>2418070
>heat up, for example, a cup of coffee by setting it on the pad?

this would be like those USB coffee warmers: completely worthless coz vastly under-powered.
same thing for peltier-based warmers.
what i did was get a $3 braun coffee maker from the thrift store, and cut off everything except the ''keeps-hot'' plate at the bottom.
then i replaced it with a heated milk-frother machine for the same price which looks better coz didnt have to take a hacksaw to it.

[Anonymous No.2418175]

>>2418168
>>2418168
Assuming you can't salvage a switching regulator or controller IC, a dual comparator (or maybe op-amp) IC is where I'd start. Can do self-oscillating really easily and effectively, while fixed duty-cycle fixed-frequency is also very doable. 555 timers work too I think, but feedback might be a bit strange.
If you only have access to transistors or logic ICs you'll have a somewhat worse time, especially with feedback, but it's not impossible. A long-tailed pair would get you a decent enough comparator-like arrangement for feedback, assuming the threshold voltage of a MOSFET or logic threshold isn't reliable enough for you. Of course, if you can salvage a TL431 or equivalent (say, from an old SMPS) that would be even better.

This should go without saying, but you'll want a dedicated switching transistor, probably a PNP or P-channel for a buck, if you're not using an IC designed to source significant amounts of current. And naturally you will need a solid understanding of circuits for this.

Also make sure you don't mistake a choke for an inductor, they're not the same thing.

[Anonymous No.2418178]

>>2418168
>salvaged from waste electronics?

just find 1 transistor and 1 zener.
if you cant find the right zener, just add diodes in series until you reach your goal.
but a lot of older stuff has LM317s in it, and that's all you need.
e.g. a Cisco Explorer cable TV decoder i took apart recently had 3 LM317s.

[Anonymous No.2418182]

>>2418178
I have a bunch of LM317s collected but don't they waste huge amounts of power as heat?

[Anonymous No.2418187]

>>2418097
Thank (You) for replying
>VU meter
So I could connect it to an amplifier output and have it indicate the volume or something?

[Anonymous No.2418189]

>>2418169
So my idea to cheap out and only use one driver for my aquarium lights is less reccomended? I get the impression that most aquarium lights(what I want to build) use CV at 24V.

What is your reccomended way to make a 2 channel dimmable light with an esp32 as the controller?

[Anonymous No.2418194]

>>2418189
>>2418175
Admendum

Would this work?
A CV power supply->Controller->LDD CC power supply->LED series

[Anonymous No.2418200]

>>2418182
>don't they waste huge amounts of power as heat?

depends.
power wasted as heat is calculated as (Vin - Vout) * load current
if current is small, or the differential voltage is low, then it shouldnt matter one bit unless you're on batteries.

[Anonymous No.2418205]

>>2418187
>>2418097
But it doesn't werk :/

[Anonymous No.2418207]

>>2418187
>connect it to an amplifier

if you do, it'll look like crap.
you really have to condition the signal with a precision rectifier and peak detector for it to look good.
see the 2 op-amps in this schematic for example.

[Anonymous No.2418209]

>>2418207
Shit that looks complicated. Right now I can't even get it to work with a regular voltage supply.
Regardless of the signal applied all LEDs light up.

[Anonymous No.2418246]

found another obscure chip. don't really know what this could be used for either.
It's german made too, guess it's quite old.
I got a whole stack of them still in plastic packaging.

[Anonymous No.2418249]

>>2418246
Ah, I get it. It's a doorbell IC
https://www.youtube.com/watch?v=RF9SGqEax8o

[Anonymous No.2418271]

>>2418059
Noob here. I have some general understanding of electronics from high school and I understand the theory but I seem to be let down in practice. I will build a basic circuit that works, but as soon as I go try add anything to it, my voltage is getting eaten by diodes etc and I need to add more power. Very soon my power requirements become unweildly for a battery based setup. Am I just going about the whole thing wrong? Do you need to design the whole circuit in one go? How can you build a circuit without testing the individual parts work seperately?

[Anonymous No.2418285]

so just to confirm i do need a fly back diode even for a small fucker motor like this right?
i will not be using a breakout board
that seems like a diode to me in the pic

[Anonymous No.2418286]

No diodes for battery shit
You are probably doing some arduino shit right? basically you want to direct drive that shit directly from a batter. if you have a 5V tolerant chip you can just hook it up to a lipo directly
if not use buck converter to lower the voltage without wasting power on diodes and shit

[Anonymous No.2418288]

>>2418286
>>2418271

[Anonymous No.2418289]

>>2418285
>that seems like a diode to me in the pic

of course.
they're not gonna make a board that's gonna die after 1 use.
so diode is there coz it's necessary.

[Anonymous No.2418290]

>>2418289
the thing is i have the motor on my breadboard now without any diode present and it works fine hence my confusion

[Anonymous No.2418299]

>>2418286
>No diodes for battery shit
You still need diodes for things and in many cases it's transistors that steal all my voltage. In my last project I was trying to build a H-Bridge to get an RC car to go forward and reverse. When I added the transistors the voltage dropped so low it became a problem.
What i've been doing is getting like 4.7v from a lipo cell and using an esp to control transistors to deliver the power, but after adding more stuff I end up needing more than 4.7 and my design with a buck converter turns into a design with a boost converter. This just seems fucking stupid to me and I'd be lost as fuck if I didnt have a good power supply to smooth out the workflow a bit.

[Anonymous No.2418316]

>>2418290
>it works fine

it may work now but, if you're using a semiconductor to switch it on/off, then you're hitting it negative voltages which will likely lead to an early demise.
(depending on its ability to withstand abuse.)

[Anonymous No.2418340]

if i am feeding pwm into mosfet gate from a micro, 1KHZ, do i need a resistor between the micro and gate?
surely the gate capacitance is so tiny it won't damage anything right?

[Anonymous No.2418459]

which one of the book do you recommend reading if im scared to touch a lightbulb or car battery when changing it?

[Anonymous No.2418461]

>>2418340
>need a resistor between the micro and gate?

you need 2 resistors.
always two.
never one or three.
one in series in case the FET shorts and sends 30V into your micro.
one from gate to ground coz when you turn it on, the micro output may be high-Z for a time.

>>2418209
>I can't even get it to work with a regular voltage supply.

did you read the part where the supply voltage must be between 12 and 25V?

[Anonymous No.2418463]

I am trying to rebuild a battery pack for as cheap as possible for a moped turn signal device. The battery pack consists of 5x Sub-C size NiCd 1.2v 1500mAh 22.2m wide by 42.9mm tall rechargeable batteries wired in series with two 0.5mm thick spade connectors coming off of them for + and - connection. They sit in a box on the luggage rack and are subject to up and down motions from the suspension. I am trying to rebuild the battery pack but I don't have a spot welder to spotweld nickel strips to them I am considering make a diy spot welder but don't know how and don't know what components to buy or how many amps the battery should have to be able to weld the thicker battery tab in this pic. I have seen guides of how to make them using a battery, capacitors, momentary switch and a lawnmower solenoid but want to make sure I get the right components to make good spot welds and weld something thick. Would solder damage the battery? What's the best way to rebuild this battery pack or am I over thinking this.

[Anonymous No.2418473]

>>2418459
You should apply for a position at hackadong.

[Anonymous No.2418490]

I got inspired by Eric Lundgren about e waste recycling. It seems like disposable vapes are here to stay and I have quite a few of them already laying around. So far I've only gotten 3,7V 400mAh cylindrical LiPo batteries. I was wondering what kind of neat projects could I do with them? Something that could be sold would be preferable. So far I've only came up with gadgets and toys such as :a flashlight,a small fan, a toy drone and some decorative light. I'm finishing my 2nd year of mechatronics, so most projects wouldn't intimidate me.
>Inb4 LiPo toys, dangerous yada yada

[Anonymous No.2418492]

>>2418463
>Would solder damage the battery?

i've dont it dozens of times without a hitch. nicads dont blow up like LiPos do.
use file or dremel to remove a small area of zinc coating.
load up a 35W iron with some solder to increase thermal mass - touch area with heat and fresh solder (also extra flux if you're feeling fancy)
it should wet in 3-5 seconds, so remove heat - tin wire - reheat both at same time and join.
there's a trick for the negative terminal: dont solder in the middle of the circle; solder on the lip (where the horizontal turns vertical)
it takes less heat there, for some reason.

[Anonymous No.2418493]

>>2418490
https://www.youtube.com/watch?v=veMiNQifZcM

[Anonymous No.2418509]

>>2418492
Cool, ok ill give it a shot.

[Anonymous No.2418518]

>>2418509

practice on the old ones.
technique counts for a lot.

[Anonymous No.2418528]

can i use linear regulators in paralel?
i need to pump 8A but the regs i have are 1A to 2A tops

[Anonymous No.2418561]

Isn't this kind of bad way of driving parallel LEDs? One of the LEDs will be stressed more than the others.

[Anonymous No.2418707]

To create a crystal oscillator circuit all I need is a crystal oscillator to oscillate and an amplifier to get rid of dampening?

[Anonymous No.2418708]

>>2418707
Why are there always capacitors in these types of circuits? Is apart of the amplification process?

[Anonymous No.2418741]

i want to trigger a relay only when there is something else drawing a big current. is there a name for using a signal wire to trigger the relay that is shaped and placed next to a high current wire to induce the relay signal?

[Anonymous No.2418751]

>>2418080
>tranny's off, led's on
>tranny's on, led turns off
what's there to get?

[Anonymous No.2418871]

>>2418316
ignore this moron, ive had a similar setup running for years without a problem. when dealing with such low currents and voltages, a tiny little motor like that means nothing. your switching transients are also negligiable. once you move beyond 3.3v electronics, then you need protection for your circuits, but tarduino shit like this doesnt matter

[Anonymous No.2418883]

>>2418189
Just one CC driver to multiple series strings is not advised. One resistor or CC driver per LED string is recommended.

>>2418194
By controller you mean PWM controller? If the CC power supplies are fine with it then sure. Again though, only one series string of LEDs per current regulator.
Draw a diagram.

>>2418207
Any idea what kind of ideal rectifier that's meant to be? It reminds me a bit of the full-wave rectification circuits I have seen, but those just have something like an all-pass filter after the first 2-diode ideal rectifier, not a second ideal rectifier. Also that 2-diode arrangement with the resistor is also a bit alien to me.

>>2418340
Depends on the MOSFET. If it's a power FET then yes, if it's a small-signal FET then no. Small-signal FETs these days does encompass FETs that can handle multiple amps of current, and since they're so easy to drive I've been taking to using them for driving nontrivial loads.

>>2418490
How small is the cell? I've always wanted to make really tiny FM radios. When you use the headphones for an antenna, all you really need is a cubic centimetre or so of electronics, plus a battery.

>>2418561
It's usually fine if they're well-matched, and even more likely to be fine if they're thermally connected to one another. You'll find parallel arrangements on COB LEDs, which fit both of these prerequisites.

>>2418707
>amplifier
Maybe if you want an analogue (i.e. sine wave) oscillator, but for most cases a square wave is fine. Enter: the schmitt inverter.

>>2418741
Clip-on current transformer, but note it outputs an AC waveform. Also may not be able to power a relay by itself without some sort of comparator circuit.

[Anonymous No.2418917]

>>2418528
>can i use linear regulators in paralel?

traditional way to do this is to use 1 regulator and 1 or more pass transistors with small emitter resistors to equalize load.
PNPs shown here, but NPNs also doable.

[Anonymous No.2418924]

>>2418917
How does that circuit work? I don't get how the BJTs are under any form of negative feedback. Does the voltage at pin 1 of the 7812 really change enough? And if so shouldn't R1 be a bit larger?

[Anonymous No.2418944]

>>2418924
>How does that circuit work?

pretty simple.
as the load current increases, voltage across R1 rises, turning on the PNPs more.
as the PNPs deliver more current to the output, less comes from the regulator, so R1 voltage decreases, turning PNPs less on.
so, it turns into a balancing act, a search for equilibrium, which is what feedback is all about.

[Anonymous No.2418946]

>>2418883
>>2418194
Plan on using LED Strips to wire them up in series. Isnt it supposed to work like that?

[Anonymous No.2418952]

>>2418946
>work like that?

sure it works electrically, but when you lay it out physically, it turns out you have to be runin' wires all o'er the damn place.
so it's easier to go parallel when you can.

[Anonymous No.2418990]

>>2418751
I don't get it

[Anonymous No.2419023]

i have battery hooked up to esp32 adc and the readout keeps fluctuating randomly by about a range of 0.05V in either direction. Any idea how to make the readout more stable?
I don't use any voltage dividers the battery is hooked up directly so there can't be any noise caused by high value resistors and stuff

[Anonymous No.2419031]

>>2419023
>Any idea how to make the readout more stable?

that's caused by quantization noise (aka rounding) and it cant be eliminated ever.
you can reduce it by averaging a number of readings, but then that slows response time.

[Anonymous No.2419038]

>>2419031
what is a good sample size to average? 100?

[Anonymous No.2419041]

>>2419038
also should i add any delays between samples to reduce the noise or can i just rapid fire them at max speed?

[Anonymous No.2419046]

>>2419038
>good sample size

i'd do a running average of 10.
but i enjoy bouncy numbers, so you may want more, if that's your style.
no delays needed.

[Anonymous No.2419111]

pic related is what i want to modify
The inner part of the headlight houses a H7 bulb that is used for the high beam. I would like to have it on at around 30% power to use it as a daytime running light. Is there any more or less off the shelf solution thats cheap enough to bother doing this?

[Anonymous No.2419115]

so normally for a dc motor you want to have a small cap on it to prevent emi, but what if this cap is on a cable about 70cm away from the motor, will that still work?

[Anonymous No.2419126]

>>2419111

adjustable buck converter good for 5 amps or better. cheap at Ali if you're willing to risk being chinked.

>>2419115

any distance reduces effectiveness. but 70cm sounds alright by me.
in fact, if you leave the cap out altogether, who's gonna know?

[Anonymous No.2419196]

>>2419126
Thanks, nice. Since there are 2 bulbs im looking at the 10A version, but that has adjustable voltage instead of adjustable current like the 5A in the pic you posted. That shouldnt make a difference right?
Also, i need the system to fulfill a IF function, meaning that the DRL switches off once the headlight switch is used. Whats the best way to realize that? Some kind of relay i suppose?
This electronics stuff is kind of intriguing ngl

[Anonymous No.2419233]

>>2419196
>has adjustable voltage instead of adjustable current like the 5A in the pic

the pic says ''CC/CV'' which is Shanghainese code for ''you can adjust current, voltage or both''
once you locate a 10A version, you can get lots of info on setting it up on Youtube by searching the part number.
e.g. searching for ''XL4015'' leads to tons of videos on the model i put in the pic.

>fulfill a IF function, meaning that the DRL switches

this is gibberish to me.
i ride a tricycle.
my mom wont even let me motorized it.

[Anonymous No.2419236]

The datasheet recommends 10uf and .1uf caps on the power rail.
What happens if i have several of the same micros on the same power rail?
Should i include those two caps for all of them or only one of them?
because if i include them for all of them they will be in parallel and basically just become one big 30.3uf cap

[Anonymous No.2419251]

>>2419236
>they will be in parallel and basically just become one big 30.3uf cap

that would be true if PCB traces were ideal conductors, which they're not.
not only do you ''need'' one set per micro, they should be really really close to the power pins on each micro.

[Anonymous No.2419254]

>>2419236
Each micro requires its own caps that are as close as physically possible to 3V3 pin. The reason why we use different values is hidden parasitic inductance of capacitor. A capacitor will only behave like a capacitor up to a certain frequency, after which it will behave like a inductor.

[Anonymous No.2419256]

>>2419233
what i mean is when i manually switch on the lights the power to the buck converter should be cut. I suppose a simple normally closed relay should be up to the task.
>adjust current, voltage or both
so the bulb will get dimmer right
>this is gibberish to me
so is this for me man

[Anonymous No.2419257]

>>2419254
>A capacitor will only behave like a capacitor up to a certain frequency, after which it will behave like a inductor.
i don't understand why the aliens who designed our universe simulation came up with such retarded redundant laws of physics.
it's so fucking annoying having to constantly think about shit like emi and wire impedance etc etc when designing circuits
and if i want to have wireless shit it's downright suicide fuel

[Anonymous No.2419259]

>>2419257
>i don't understand why the aliens who designed our universe simulation came up with such retarded redundant laws of physics.
It's there to filter out brainlets.

[Anonymous No.2419274]

>>2419256
>a simple normally closed relay should be up to the task.

yeah.
you'd put a 12V 10+ amp NC relay on the positive OUT terminal to make sure 12V isnt fed back into the buck.
actually, it would be simpler to just just use a diode to ensure current only flows out, never in, but you may screw up the direction of the diode, and the power rating for it.
i'm guessing you have a grasp of relays and wont screw that up.

>so the bulb will get dimmer right

uh, yeah, that's been the whole point of this exchange.
both the CC and CV pots will dim that f*cker down however much you want.
you can make it so dark it starts picking cotton.

[Anonymous No.2419294]

Henlo,
I am making 3D printed parabolic speakers/mics. Does anyone what is the best way to glue/fasten these things? By the rim? Should their backs be touching something or I should avoid that?

[Anonymous No.2419310]

>>2418917
does that also work with fets and any ldo?

[Anonymous No.2419312]

>>2419310
>does that also work with fets and any ldo?

fets: no
ldo: yes

[Anonymous No.2419313]

>>2418946
That's just one series string. So it only needs one constant current driver.
If those are actually "LED strips", the ones that come with resistors, then you don't even need the constant-current driver and can power them with CV.

>>2419023
Have you separated your analogue rails from your logic power rails? A star ground and some chokes may make your measurements a bit better.
Aside from that, as the other anons say averaging is likely necessary. I'd average across 8 or 16 samples, because the math is likely a tad easier; you can just do bit-shifts instead of division.

>>2419196
You just wire the normal night-time lights to bypass the buck converter and supply the full battery voltage. Basically a switch in parallel with the buck converter. Not sure how cars are wired, that will either be through a relay or directly through the light switch.

>>2419294
By the rim. I'd use something a bit pliable like silastic or silicone. Best if the back surface isn't touching anything, as it will rattle when it moves.

>>2419310
Selecting the right resistors would be tough, threshold voltages for FETs aren't reliable. May want to make a current mirror instead.
Well it's not like FETs don't have the possibility of thermal runaway when being used in the linear region.

[Anonymous No.2419315]

>>2419312
shit, welp i probably have some trannies around here
so does this bypass the dropout voltage? normally if i drop 1V at 2A at a ldo it gets hot as fuck
but since the trannies do the power transfer that means i don't lose almost anything right?

[Anonymous No.2419320]

>>2419315
>that means i don't lose almost anything right?

you lose whatever the LDO usually loses.
it'll just run a bit cooler.

[Anonymous No.2419331]

>>2419294
Give it the hot glue treatment. Or jizz on it.

[Anonymous No.2419554]

I'm a senior EE undergrad. They've taught us basic mesh and nodal analysis. However whenever I come up with "interesting" circuits with my intuition, there's not a single tool in the arsenal they've given me to let me begin analyzing those circuits.

this circuit acts a bit like an automatic-gain compressor. I came up with it by imagining the BJT as a voltage-controlled resistor. When the output signal is below the turn-on threshold, the feedback current flows entirely through Rf as normal. However, once the output signal exceeds the threshold, it begins to turn the transistor on, providing an alternative low-resistance path for the feedback current. This acts as a resistor divider, greatly diminishing the effect of Rf, and thus the overall gain of the amplifier.

even though I can qualitatively explain how it works quite well, I can't for the life of me figure out how to analyze the circuit so I can improve it or see its deficiencies. And this is just a single quick example circuit where I experience this problem.

how did/do The Greats figure this shit out? What advanced techniques have I yet to hear about that lets you tackle these nonlinear circuits?

[Anonymous No.2419611]

>>2419554
Write out a bunch of equations relating current to voltage and hope there's an analytical solution.

Also that circuit needs a split-supply for the inverting amp to work, but the BJT only sinks current in one direction. Enter: the reason people use JFETs.

Pic related is a really interesting but simple AGC circuit, just for interest's sake.

[Anonymous No.2419612]

>>2419320
how much is it tho? what is the split between the ldo and trannies? if the trannies trandsmit like 90% of the power then thats a ton of power savings

[Anonymous No.2419623]

god damn lying chinks
i got this buck because it's very small and can deliver 1A according to the seller
Except the voltage folds nowhere near the 1A
chinked again DAMN IT

[Anonymous No.2419626]

>>2419623
those are chinese amps sillyhead
you got exactly what you didn't pay for

[Anonymous No.2419628]

>>2419554
Your transistor is acting like a diode, it will start turning on when emitter base voltage exceeds threshold of typically 0.6V. You can model this by using shockley diode equation.
>how did/do The Greats figure this shit out? What advanced techniques have I yet to hear about that lets you tackle these nonlinear circuits?
There are a lot more analytical techniques for linear circuits, so one way is to linear your circuit by replacing transistor with ideal diode with series resistance and voltage drop. If you already understand your circuit there is no point in messing with analytical expressions, just put everything into spice and simulate. In real world component variations, temperature and tolerances will cause variation in circuit behavior which if you want to model with analytical methods will produce unreadable garbage. Remember the fundamental theorem of engineering: an approximate expression that is easy to understand is better that exact expression that is complex and hard or impossible to understand. Always think what a boomer engineer would do: punch in some numbers in to excel, approximate Pi=3, draw over some plots in a datasheet and produce an answer within 5% of actual result from spice simulation. Stop thinking like a scientist and think more like an engineer.

[Anonymous No.2419647]

i often need to test if my power sources (buck, ldo, wall wart..) can handle outputting enough current to power my designs.

Is there any way to make a simple adjustable load? Basically a load i can connect to any DC powersource between 2V - 12V and set it to specific current like 100ma, 500ma, 1A, 2A, 4A (4A is max i need) and it will draw that amount from the power source?

I was thinking about making a design where is use LM317 in constant current mode and short the output. I will then have several resistor that can be turned on or off by jumpers hooked to the output of the LM317 to switch between various currents i want to test.
It's not perfect idea since i will have hardwired 4 current options and no more granularity than that.
Since lm317 is 1.7A tops i would build say 3 of those in paralel and i can turn each other on or of and each lm317 would have jumpers for 100ma, 500ma, 1A, 2A, so if i have 3 of them i then have 3*4 current draw options to choose from.
The jumpers will actually be mosfets so they can be controller by a micro automatically, so i can just select current i want and the micro will go
"Ok he wants 6A, i will turn on two lm317, set one to 2A, for the other one i will turn on two of it's resistors in paralel to get 4A"

Before i waste my time building is, is there some easier way to do this?

[Anonymous No.2419655]

>>2419612
>what is the split between the ldo and trannies
Say the voltage after the bridge rectifier is [Vcc]. The voltage at the input of the LDO is [Vcc - I_ldo * R1]. The voltage at the emitter of the transistor is then [0.7 + Vcc - I_LDO * R1]. The current through a given BJT is then calculable since [I_BJT = V_R2/R2 = (0.7 - I_LDO * R1)/R2 = 0.7/R2 -I_LDO * R1/R2]. If 0.7V is small compared to the voltage drop across the resistors, that simplifies to [I_BJT = I_LDO * R1/R2]. Since R1/R2 is 1 in this instance, that simplifies to the current through each BJT equalling the current through the LDO. But of course you can change that.

[Anonymous No.2419660]

I don't know if this is the best place to ask but is there any specific good brand of these type of dc power supplies to buy?

I'm trying to organize the outlets behind my tv stand by replacing all the power supplies for my shit but every website just has the same 10 indistinguishable chink brands so I have no fucking clue what to buy. I have like 5 different devices being powered by 5 different 5v and 12v adapters that don't fit on a power strip together so I want to try and make things uniform.

[Anonymous No.2419669]

>>2419647
ha! i fucking did it
i made it so simple it's retarded
i am using a mosfet as a load and then manipulating the mosfet gate with an opamp controlled by voltage divider pot to draw exactly the current i want
look how simple this shit is, just 4 parts
i am such a smart boy

[Anonymous No.2419670]

>>2419669
That's one way to do it. Just be careful that you don't melt your breadboard.

[Anonymous No.2419674]

>>2418883
>How small is the cell?
The battery is 1.5inch long and 1inch in diameter. I'll be honest I don't have hands on experiece yet with radio waves and antennae , but it sounds pretty cool. Apart from car owners, who even has or wants a radio nowadays

[Anonymous No.2419675]

>>2419612
>if the trannies trandsmit like 90% of the power then thats a ton of power savings
No it's a linear regulator. If you have a 15V power rail and you're dropping that down to 12V, then you're running at exactly 80% efficiency regardless of what the voltage is being dropped by.

>>2419623
Design your own switching converters, it's fun I promise.

>>2419647
I use a length of nichrome wire. It's 60Ω from end to end, so I just clip alligator clips to a desirable length. I made a cradle for it, which inadvertently looks like a wrestling arena.
>lm317 is 1.7A tops
That's only at minimum dropout. The actual limit is on temperature, and hence on power. If you're dropping 5V across it, 1.7A is going to mean 8.5W, which requires a nontrivial heat-sink if you don't want to fry your LM317. Well LM317s have thermal protection so they won't just kill themselves, but it means you should measure what current they're drawing before writing down results.

Also measure the amount of ripple they produce with a scope while they're loaded down, in case that matters to you. I'm currently doing a series of experiments, trying to figure out how best to reduce output ripple of SMPSs under a variety of different load conditions. I'll make a blog post about it when I finish (idk in like a month) if you're interested.

A better situation would be to use a power BJT clamped to a heat-sink, controlled as a darlington pair by an op-amp with a feedback loop set to measure current. Use a pot or DAC or whatever to control that. A MOSFET may be easier to use, but they have thermal runaway issues when used in the linear region, so I'd avoid them. Using load resistors as well as transistors complicates the design a bit (need more resistors and transistors/relays to get a wide range without trivialising the voltage drop across the resistor) but it may be simpler thermally. Could even use a switching converter to power a fixed load, which would have the advantage of not dumping heat in a tranny.

[Anonymous No.2419676]

>>2419674
>The battery is 1.5inch long and 1inch in diameter.
Not hugely small then.

FM radios are easy because A: you can just buy an RDA5807M (or its module) and hook it up to a (small SMT 6/8-pin) MCU, and B: the wavelength is long enough that you don't need to give a shit about transmission lines and impedance matching and coax.
The tough part isn't the electronics, it's getting the mechanics in a good package with a convenient user interface. My thinking is to have a ring or two around a cylindrical body, both of which are rotary encoders thanks to a couple of holes in the frame and white+black marks on the inside of the rings acting as optical encoders. That handles volume and frequency. But there's a ton of other commands you may want to send to the module (read the RDA5807M datasheet) so figuring out what to cut and what to keep is difficult. Integrating control buttons already on headphones/IEMs is a good idea if available, since they're pretty easy to read.
A 3D printer (or machine shop) would be recommended for small form-factor electronics of any kind, especially one with a complicated UI like this.

Other battery projects I could suggest would be datalogging, long-duration MCU projects where the challenge is getting the quiescent current as low as possible. Also energy harvesting RF, because that sounds fun.

[Anonymous No.2419681]

is this tip usable?
can I save it by retinning?

[Anonymous No.2419684]

>>2419647
>Is there any way to make a simple adjustable load?

like the other anon said, a nichrome rheostat is the simplest cheapest option.
you can buy one already made from China, or make your own from a spool of wire and a wood dildo.

[Anonymous No.2419687]

i am getting really fucking tired of googling datasheets for my components and seeing this shit constantly

[Anonymous No.2419688]

>>2419681
>can I save it by retinning?

try it and see. i've brought some Weller tips back from the dead that looked worse.
while it's hot, scrape off the crud with the back side of a box cutter, and apply solder.
keep doing this for a while until it's all shiny and chrome.

[Anonymous No.2419689]

>>2419687

Progress. It's a b*tch.

[Anonymous No.2419701]

>>2419687
Mostly a non-issue for hobbyists. Obsolete parts make actual companies want to use the newer one so they don't have to revise the design as soon when the parts get out of stock. For hobbyists it doesn't matter unless the part is actually getting out of stock soon/now. Even then, there's plenty of drop-in replacements for common parts, or even other manufacturers still making them. Or chinks relabelling newer parts as them.

I take what I can get so long as it's got a datasheet. On LCSC there's all sorts of chinese parts that are cheap and effective, I recommend hunting through their different categories. The search isn't properly parametric though, and Octopart's search only covers LCSC's non-chink parts, so looking for something like an analog multiplier is going to leave you very frustrated. Their LED driver section is great though.

[Anonymous No.2419735]

So does this mean input GND and VCC GND have to be seperate?

[Anonymous No.2419739]

>>2419735
No, the grounds are the same.

[Anonymous No.2419759]

anyone have an example of how an high voltage industrial e-fuse is designed? like 600 V 10 A, or 60 V 100 A? seems most off the shelf solutions are designed for much lower voltages and current.

[Anonymous No.2419783]

>>2419739
Why is it drawn as seperate like that then?

[Anonymous No.2419791]

>>2419783
Probably because of the capacitor rating label (220uF 6.3V). Easier to read.

[Anonymous No.2419801]

>>2419791
Hmm. I can't get this amplifier working on breadboard. I guess I have to solder it. The capacitances and shit are propably too high.

[Anonymous No.2419803]

>>2419791
But I'll try this circuit diagram. Thanks
Also why isn't there stated what kind of amplifier it is?

[Anonymous No.2419815]

>>2419803
>what kind of amplifier it is?

it says Total Harmonic Distortion = 10%
so it's the kind of amp that causes people to think they have wasps eating their eardrums.

[Anonymous No.2419818]

>>2419803
It's used for CB radios. Mono output at max 5W, through a 4-Ohm mid/tweeter.

[Anonymous No.2419820]

>>2419801
If you like shitty, cheap, class-d amps then check out PAM8403 & PAM8610.

[Anonymous No.2419923]

Does anyone else see the retarded dog in the thumbnail?

[Anonymous No.2419926]

>>2419923
no

[Anonymous No.2419928]

>>2419926

[Anonymous No.2419933]

>>2419928
I dunno, man. Looks like a beaver to me. Then again, everything looks like beavers.

[Anonymous No.2419988]

>>2419815
I'm used to chinky class d amps. I wonder what's worse - cheap analog or cheap digital amps.
https://youtu.be/ZfIEkmnY5xY
>.1 is pretty high
lol.
With shitty car radios it doesn't matter, I suppose.
Guess I'll look for a better amp IC.

[Anonymous No.2419989]

>>2419820
I've had enough of shitty class D. I wanna try something else.

[Anonymous No.2419991]

>>2419988
>>2419815
*he's talking about 1/10th of a percent
I can only guess what 10% sound like...

[Anonymous No.2420019]

>>2418059
k thx roll

[Anonymous No.2420034]

>>2420019
I'll kill myself reroll

[Anonymous No.2420037]

>>2420019
Do you have the tools? Do you have the talent?

[Anonymous No.2420040]

>>2420034
I'll die of boredom reroll

[Anonymous No.2420068]

>>2418059
roll

[Anonymous No.2420104]

>>2418059
N555 chip are so easy to use

[Anonymous No.2420116]

>>2420104
>N555

[Anonymous No.2420146]

>>2420104
The pinout is fucking awful though. And the output doesn't go higher than Vcc - 1.4V. And the quiescent current is really high. And they make both really shitty VCOs and really shitty voltage controlled PWM controllers, since they do both with the same pin.

[Anonymous No.2420210]

So I picked up a power bank the other day to use to charge my phone and headphones while I'm out on multi day camping trips. It turns out that most modern power banks have some auto shut-off crap for low draw charging like Bluetooth headphones so I literally cannot charge them from it. Now I have to cut open a cord and solder in a resistor to get it to match the required current draw. Holy FUCK I hate technology sometimes.

[Anonymous No.2420237]

Anyone ever mess with increasing the voltage of an atx power supply? How high do you think you can get the 5v or 12v rail on them?

[Anonymous No.2420267]

>>2420210
Take it apart and post IC name. Small chance that by changing a programming resistor or capacitor you can alter that behaviour.

>>2420237
Never tried, and while it’s probably quite possible from a schematic perspective, trying to implement that on an existing board may not be easy or even feasible. Best case scenario it’s just some programming resistors. Worst case it’s a whole cutting of traces and introduction of new components, altering RC networks, etc.

Then again, considering 12V is often used as a MOSFET gate voltage, you may find it impossible to change it beyond 10-15V without even more major modification. Changing the output rails may also change any high-side LVDC supply rails in the same way, if they have insufficient/no feedback.

[Anonymous No.2420273]

>>2420267
It's an anker powercore select 20000, I'm not really confident in my ability to change shit on the IC without fucking it up and bricking it. Seems like it would be much easier to just wire a small resistor into a USB cable.

[Anonymous No.2420274]

>>2420273
Also adding to this I don't even see how to take the fucking thing apart. It's some molded plastic bullshit and I'm pretty sure I will trash the top and all the stupid plastic retaining clips trying to pry it off with a razor blade or whatever. Why does it have to be like this, I remember when shit used to have screws and rubber gaskets HOLY FUCK REEEEEEEEEEEEEEEEE

[Anonymous No.2420279]

>>2420273
>Seems like it would be much easier to just wire a small resistor into a USB cable.
Sure, if you don't care about the shitty efficiency that comes with a parallel resistor. Try to buy one of those "charging only" cables with only power and ground conductors in them, much easier to mess with. And don't heat-shrink it back up again before you ensure that whatever resistance you pick will absolutely keep it awake. With any luck it won't require you to dump more than half a watt, but I wouldn't be too hopeful.

[Anonymous No.2420291]

>buy OneWire probe
>look at it
>3 wires
damn, chinked again

[Anonymous No.2420294]

>>2420291
This bait or something? 1-wire stuff has 3 wires because 1-wire is short for 1 (data) wire.

[Anonymous No.2420297]

>>2420294
Wasn’t there a protocol that used the power wire for data, with a diode and cap on the slave device to keep itself powered while the line is pulled down? IIRC one-wire could be used like that.

[Anonymous No.2420323]

What use would you have for a transistor with an insanely high on/off current ratio?

[Anonymous No.2420381]

>>2420323
How insanely are we talking here?

[Anonymous No.2420430]

Is it better to get chicks to make you PCBs or to just etch it yourself?
I have a plotter so etching is really easy for me

[Anonymous No.2420433]

>>2420430
You mean like baby chickens?

[Anonymous No.2420434]

>>2420433
yeah cheap labor

[Anonymous No.2420439]

>>2420434
Are you planning to produce a batch of boards? Do you need multiple layers and fine-pitch features?

[Anonymous No.2420440]

>>2420439
I just want a 1 layer board to save myself from having to do a ton of hand wiring

[Anonymous No.2420442]

>>2420440
I think you already know the answer.

[Anonymous No.2420444]

>>2418164
they are using very small oled screen with microlens array, cool stuff

[Anonymous No.2420445]

>>2420442
Do you know where to buy big 7" x 7" copper clad boards?

[Anonymous No.2420446]

>>2420445
Marlin P. Jones is nice if you're in the USA.
https://www.mpja.com/search_results.asp?q=copper%20clad%20board

[Anonymous No.2420447]

>>2420446
thanks bro

[Anonymous No.2420449]

>>2420447
yw

[Anonymous No.2420463]

>>2420323
So just a normal tranny?

[Anonymous No.2420491]

>>2420463
>normal tranny

[Anonymous No.2420506]

quick question. i have pc fan powered by 12v with 3 wires.

What does the tachometer wire output? is it 12V pwm?

[Anonymous No.2420510]

>>2420506
i am only getting garbage like this out of the tacho wire, that doesn't change in any way if i speed up or lower the fan speed

[Anonymous No.2420513]

>>2420510
figured that shit out, turns out you have to pull that shit up, which means the peak of the pwm = input voltage so i need a voltage divider, and then the frequency of the square wave is the speed
this is some caveman shit, but i guess i can work with it at least

[Anonymous No.2420580]

>>2420513
a pull up resistor is caveman shit? huh.

[Anonymous No.2420644]

>>2418528
Here's another way of paralleling linear regulators. From:
https://www.maximintegrated.com/en/design/technical-documents/app-notes/7/7296.html

[Anonymous No.2420646]

How are battery chargers capable of charging multiple cells in parallel with adjustable current limiting typically made? Something like pic related, which can switch between rapid charging at 1C and slow charging at 1/2C. For a 2000 mah AA cell, this is 2A or 1A per cell, so it can't be something like a resistor, which would dissipate far too much heat.

[Anonymous No.2420703]

>>2420646
>it can't be something like a resistor,

of course not. it's uses biodes.

[Anonymous No.2420706]

>>2420646
PWM Fets.

[Anonymous No.2420711]

>>2420703
>biodes
We call them "wires" in Tanzania.

[Anonymous No.2420730]

>>2420706
straight into the battery with only its internal resistance slowing the current flow? or do you mean PWM FETs in something like a current controlled buck converter?

[Anonymous No.2420752]

>>2420730
I'd hope there's some kind of filtering and polarity checking after the fet but that'd be up to the whoever designed the charger.

[Anonymous No.2420764]

>>2420646
Low-power chargers (like common AA NiMH chargers, or <1A total lithium ion chargers) will almost certainly just use resistors. For higher currents, yeah buck converters. Need a feedback loop to get the CC and CV curves anyhow, so just feeding those into a standard buck IC instead of an op-amp + pass transistor isn't much more difficult.

[Anonymous No.2420765]

>>2420764
>Low-power chargers ... will almost certainly just use resistors
Meant to say linear regulators. Dumb resistors make for very bad chargers, unless you just need a trickle.

[Anonymous No.2420773]

>buy my first multi-drawer unit for components
>the drawers are really difficult to remove
>don't sit well inside
>flashing everywhere
>entire plastic casting must be made from a decades-old mould
>spend an hour and a half making it usable
chinked again

[Anonymous No.2420776]

Is there something simple I can practice soldering with that won't be a big money sink? I see these practice kits on Amazon for ~$13.

[Anonymous No.2420777]

>>2420776
>amazon
aliexpress kits are like ~$2 for one of those LM317 boards, I got an SMD soldering practice board (didn't actually do anything, just practice) for under a dollar.

or just get some perfboard and make an LED grid or whatever.

[Anonymous No.2420882]

>>2418059
Hey, i want to swap the rumble motors in my USB xbox360 controller to something more powerful, but the ones i tired, despite being 5v ones too, barely moved (and/or bluescreened my pc) so i assume they want more current.
After reading up on it, i found that i could use relays or mosfets (i think) to turn a separate circuit on/off, controlled by the old motors leads, but that would mean i lose the variability of the speed (so instead of it turning on at 40%, its either 100% or nothing)
And this is where my google-fu fails me: i mostly find amplifiers that i assume are mainly for ac/signals, and/or wouldn't properly electronically "decouple" my would-be motor form the controller (so i would still fry my pc), or it relies on a potentiometer to regulate the output instead of relying on the input voltage.

I'm sure there's a dead simple solution, or a transistor/ic made just for this, but for the life of me i cant find it.

[Anonymous No.2420883]

Is it just me, but most PVC insulations are shit.
I've used a lot of pvc insulated wires at various places. and now I'm learning the hard way, that the insulation gets brittle over time and thus, any vibration or movement will break them.

Please recommend alternative that ain't as expensive as silicone.

[Anonymous No.2420884]

>>2420882
What format is the variable-speed signal in originally? Chances are it's PWM anyhow, so feeding it to a MOSFET would work just fine.

[Anonymous No.2420887]

I have this PSU that's rated for 12v@18a 220w. I've already modified it to run at 24v. My question is would the 220w rating be affected? The components are all good for 18a but not sure about the actual transformer. Would pulling over 220w overheat it?

[Anonymous No.2420892]

>>2420884
Thanks, can i tell the format only using a multi meter? i don't have an oscilloscope
Also, how can i tell which mosfet i need, what keyword should i use for further reading?

[Anonymous No.2420901]

>>2420887
yeah. If you need more watts go to the thrift store and buy 2 car battery chargers for cheap because you can take the transformers out and put them in series.

[Anonymous No.2420913]

What does this PCB do?

[Anonymous No.2420938]

>>2420892
>Thanks, can i tell the format only using a multi meter?
Maybe if it can measure kHz frequencies. Got a logic analyser?
>how can i tell which mosfet i need
Depends mainly on motor current and gate voltage characteristic. You need a FET that's sufficiently turned on with 5V at its gate, to pass the motor's current without significant voltage drop or heat loss. A small-signal FET like an AO3400 might do the trick, but chances are you'll want a "logic-level" power FET like the IRLZ44N. There are more modern solutions, but the Z44N is common enough and should work fine.
>keyword
You want an N-channel MOSFET. The primary spec on the datasheet you want will be labelled: "Drain-to-Source On-Resistance" or something similar, and you're looking for the one rated at or near 5V. For example, the Z44N is rated at 0.025Ω, meaning it will drop 0.05V across it at 2A. Kinda overkill to be honest. Then the AO3400 is rated at 0.019Ω at 4.5V. The 3400 is an SMD FET only though, not sure if there are common THT equivalents.

Also don't forget the freewheel diode.

>>2420913
Voltage or current negotiation maybe? Could also be data buffering/amplification.

[Anonymous No.2420983]

>>2420913

all them there lines marked Rx and Tx suggest a USB to cereal adapter.
probably Corn Flakes.

[Anonymous No.2420989]

>>2420913
It copies pics of your dong and sends them to the CCP who put them in your personal blackmail folder/FBI file. Stop taking pics of your wang, or any wangs for that matter.

[Anonymous No.2420992]

the time has come to mill another pcb
with chinks pcbs being so cheap i haven't done this in a while so i hope i remember how to do this shit still
the 2 sided alignment is always the most annoying
thing

[Anonymous No.2421002]

>>2420989
>It copies pics of your dong
Hunter B., ladies & gents.

[Anonymous No.2421067]

>>2420989
but Wang is my best friend, we eat kungpao together ;_;

[Anonymous No.2421120]

>>2421067
I overheard him calling you a stupid gweilo to his cousin Huang. Then he talked about peeing in your coke. You need better frens, Anon.

[Anonymous No.2421152]

maybe the wrong place to ask, but can anyone good with DSP explain how you create a Downsampler that accepts arbitrary sample rates, instead of just whole ratios (2x, 4x, 8x etc)?

I guess one way to do it would be to upsample first, and then downsample afterwards. But this seems really impractical, for example if my target Hz is 9% of my original rate, then I would need to upsample by 9x and then decimate by 100x. Decimating by that large a number won't work because of block sizes; what am I missing here?

[Anonymous No.2421155]

>>2421152
ok, seems like I need to do resampling (interpolation + decimation) using FIR implementations for efficiency...
i'm not sure this can work really well though in realtime where you can move a knob back and forth

[Anonymous No.2421180]

I don't know if this goes in /ohm/ or /stupidquestions/ or what ... but hoping someone more knowledgeable about DC wiring can give me some advice.

I've got a LED light strip that's going around the perimeter of a room.
But I'm not sure how to wire it ... as I see it, there are four options (pic related):

-ONE- Just wire up one end and run it around the room in one direction
-TWO- Split the light strips in two, and run it in two directions
-THREE- Make a full loop and wire both ends with both positive and negative
-FOUR- Make a full loop and wire both ends, positive to one end and negative to the other

Assume my DC PSU is has the appropriate current for the job.
With -ONE- especially, there would be noticeable voltage drop by the end of the run, and the LEDs at the "end" will be much dimmer than the LEDs at the beginning. I think -THREE- and -FOUR- mitigate that the most, but I don't know if I'm supposed to do it that way.

[Anonymous No.2421187]

>>2421180
Run a pair of stranded 10AWG wires parallel to the LED strips and tap off of it every 1 or 2 meters. If the total strip length isn't too long then run wires to both ends of the strips and possibly the middle.

[Anonymous No.2421205]

>>2421180

1 is simplest. try it first to see if there is actually a problem that needs solving.
2 and 3 are functionally equivalent and MAY be an improvement over 1
4 is worst choice possible

[Anonymous No.2421207]

>>2421180
I ran 38 feet around a room. At each end, and every 13 feet or so, I ran a set of wires back to the driver. These were 24 VDC LED strips, and when the switch is on, I get about 22 VDC everywhere. I was concerned with voltage drop as well because I was doing cove lighting, and I really wanted equal brightness throughout. The hell was finding a dimmer that would truly give me 1% brightness without any flickering. I don't know if it's because I am really pushing the limits of my driver or not. I couldn't really find a design guide that would walk me through anything.

[Anonymous No.2421232]

>>2421187
>>2421205
>>2421207
Thank you all for the responses.

>4 is worst choice possible
got it.

[Anonymous No.2421256]

>>2421232
No problem, Trent.

[Anonymous No.2421495]

>>2420992
Looks fine, though there are a few minor aesthetic changes I'd make. Like swapping those 1mm traces down to 0.8mm or so. Getting rid of superfluous wiggles.

>that DIP-8 top right
>some pins have connections on the top layer
>presumably other connections on the bottom
You got plated through-holes? Or rivets? Otherwise you'll need to solder on the top and bottom, which isn't possible if you're using sockets or with some other footprints. I run a python script on KiCAD that removes any THT pads that are on the same side as the component itself, so I can only make connections to it via its other side. Gets rid of the pads that would otherwise be unconnected and clog up that copper pour (and get in the way of traces going between pads).

[Anonymous No.2421532]

>>2421495
i tried rivets once but they were a pain in the ass so now i just use wire and solder it on both sides of the via. the pads on the other side i just leave there since i don't know how to remove them, is there any easy way to do it in kicad 6?
also since i haven't done pcb milling in a while i of course fucked up the design and forgot to flip the tht components around, which normally i don't need to do since chinks pcbs have plated holes, but for milled pcbs they have to be mounted from the other side

[Anonymous No.2421536]

>>2421495
>Like swapping those 1mm traces down to 0.8mm or so.
they already are 0.8mm traces, i always use 0.8mm for milling to offset for any runout on the cnc, i would rather have a fatter trace rather than a too thin or broken trace

[Anonymous No.2421601]

>>2421532
>i tried rivets once but they were a pain in the ass
Agreed. Especially if you don't have the dedicated tool.
>now i just use wire and solder it on both sides of the via
Yeah I also use solid core wire vias, but I hammer them down so I can cover them with solder mask.
>the pads on the other side i just leave there since i don't know how to remove them, is there any easy way to do it in kicad 6
I wrote my own script for it (and by that I mean a python program I put into the directory instead of an actual plugin), maybe someone's already made a plugin that does it, but I couldn't find one from a quick search.
I can post mine if you want, though I do plan on going back to figure out how to get it into a plugin format.

>they already are 0.8mm traces
Oh ok, carry on then. I use 0.5mm because that's how small I can trust myself to paint.

[Anonymous No.2421678]

>>2419815
>what kind of music do you like?
>DIRECT CURRENT

[Anonymous No.2421684]

Bought two Razer core x chroma boxes off Amazon, the one on the right has what looks like oxidation on the pads but I really don't know...
Does it look bad to you guys?

[Anonymous No.2421694]

>>2421684
Here's a better pic

[Anonymous No.2421698]

>>2421684
It's probably just shit lead free solder that always looks like a cold solder joint. Did you test them yet?

[Anonymous No.2421720]

>>2421684
Also try scratching one of the green pads to see if it's just on the surface.

[Anonymous No.2421723]

>>2421694
It's corrosion. Looks like moisture.

[Anonymous No.2421795]

>>2421601
i actually have the special rivet tool for pcb rivets but i refuse to buy the overpriced as fuck copper rivets from the manufacturer and chinks don't sell the correct size ones

[Anonymous No.2421855]

my solder tip broke in half

[Anonymous No.2421857]

>>2421855
You get what you pay for, mein neger. Shop smart- shop S-Mart.

[Anonymous No.2421858]

is there any obvious way to reduce the number of components needed to create this pseudo-randomly LED fade in/out circuit?
https://electronics.stackexchange.com/a/137627
in particular the PWM generation aspect

[Anonymous No.2421860]

>>2421857
it still works

[Anonymous No.2421863]

>>2421860
Noice. How much longer until it dissolves?

[Anonymous No.2421868]

>>2421863
I'm just gonna keep using it until it stops working

[Anonymous No.2421882]

>>2421868
Post an update when it takes a shit. The tip liberated itself from the shaft on July 4, 2022.

[Anonymous No.2421918]

>>2421858
Can do PWM to a 555 just by feeding a voltage into the CV pin. Not very well mind you, but it kinda works. That 2-comparator method is what I'd usually use, but I think you may be able to just use one comparator oscillator with a diode so it's a triangle wave, then use a transistor to change the current going from the timing cap. You'd get a free comparator from the deal.

Also use a LFSR for randomness.

[Anonymous No.2421919]

>>2421858
>obvious way to reduce the number of components

dontcha' know the bitches love a guy with a big BOM.

[Anonymous No.2421924]

>>2421918
i'll keep this in mind thank you
>LFSR
do you know any comparable circuits? i've come across this in my searches, but haven't understood how i could use the idea since i'm a beginner.

>>2421919
unfortunately my constraint is how to translate a circuit into a relatively small PCB with smd components, so i've just been trying to think through how to do it with the fewest possible components as a starting position. and for purely educational purposes, i'm avoiding using a microcontroller.

[Anonymous No.2421926]

>>2421924
>relatively small PCB with smd components

maybe you can replace the 555 with a single schmitt trigger - you save 2 pins and 1 resistor.

[Anonymous No.2421944]

>>2421926
true, i have run into that possibility. i actually found a circuit using a single 7414 for doing close to what i need, but it also adds so many capacitors/resistors/transistors that i wasn't sure if the reduction in ICs was worth it.
https://blog.zakkemble.net/leds-looking-pretty/

but just replacing the 555 with a single schmitt trigger i'll look into as well.

[Anonymous No.2421946]

>>2421944
also forgot to mention, this linked circuit also loses 2 LEDs compared to the 4026 which could be fine but would require physical testing/brightness checking.

[Anonymous No.2421949]

Why are there so many fucking variants of the same connector

[Anonymous No.2421951]

>>2421949
Because if there were only one then everyone would complain about Big Connector tycoons who wear stovepipe hats and drive steam locomotives powered by burning peasant-class children to and from work every day.

[Anonymous No.2421952]

>>2421926

crap, i posted a non-inverting gate.
should be inverting, like the SN74AHCT1G14

[Anonymous No.2421955]

>>2421949

if you use a standard design, someone else will make cheaper replacement parts.
but you can charge monopoly prices if you're the sole source, esp if you patented it.
see Apple Inc, and Jews Inc.

[Anonymous No.2421958]

>>2421944
Wild, I was just messing around with a similar circuit. What are you using it for?

[Anonymous No.2421966]

>>2421958
>What are you using it for?
Hope to hypnotize the neighbors cat so i can finally lose my virginity.

[Anonymous No.2421967]

>>2421952
would the SN74AUP1T14 work as well? it has a lower supply voltage which works better for me (also the reason i was going to use a CMOS 555). and would an inverting gate like this just be used directly in place of the 555 + additional resistor directly given the circuit i posted or should i be rethinking the dual OA and/or 4026 as well?

>>2421958
first reason is strictly educational. i'm a programmer and interested and learning more about this side of things because my interests may have me interacting with hardware occasionally. i want to buy components so i can learn how to put the circuit together and more theory on what they do individually, but since it takes a while for parts to be delivered i'm trying to figure out at least what path i should be following.

but ultimately, i'm making an add-on board for electronic conference badges.

[Anonymous No.2421973]

>>2421967
>electronic conference
Do they have those tiny boxes of cereal there?

[Anonymous No.2421992]

What kind of DE are typically studied in EE?
I know there are millions of different types of DE in general math, but sounds like in EE it is mostly 1st order RC / LC networks and 2nd order RLC? Does that also include non-homo DE with various forcing functions? Is that knowledge even need in real life? since there are probably ready made solutions for all kinds of patterns and very few real life DEs have solutions. What about the Laplace transform?

[Anonymous No.2422023]

>>2421924
An LFSR wouldn't be too hard to implement, just a 74HC164 and some XOR gates, with the input pin going to a low-pass filter that feeds the PWM circuit.
There may be some analogue methods like random noise generators seen in synths and such.

>>2421992
Anything faintly related to fourier/laplace transforms, and the kinda stuff you see in control theory.

[Anonymous No.2422080]

>>2422023
I'll look into that then, thanks.
it's a bit expensive but what do you think about using a LTC6992 PWM generator with the original circuit? could output into the 4026 and get input from the 555 in the same way the dual opamp circuit is?

[Anonymous No.2422119]

>>2422080
Do you want it to be random? Because that original 555 circuit just looks like a consistent fade-in/out effect.

If that kind of breathing is fine, I think the most compact solution will be to try to get it working with a dual op-amp as both the low-frequency-oscillator and the PWM circuit. Pic related kinda works, but you'll want to test it a lot first.
The cheese method would be to use a comparator with negative feedback to output PWM.

If you want something that actually looks random then you'll have to go somewhere else. Picking up noise with an open FET gate or CMOS logic input would be a fairly simple way, but that would take even more trial and error. Could definitely be as small though.

[Anonymous No.2422124]

>>2422119
the 555 circuit goes to a 4026 counter which each LED would be connected to and have 10 different patterns (pic related).
here's a video of the circuit excluding the PWM fading.
https://www.youtube.com/watch?v=c_97iU6JFX4
actual randomness is really not important. it's more about not being an evident loop of a single pattern. the 74c14 circuit i linked earlier is a good example i think.
https://www.youtube.com/watch?v=U9iPE8rAUkQ

[Anonymous No.2422157]

>>2422124
>actual randomness is really not important.
Oh ok.
Still, give the first 6 minutes or so of this a watch:
https://youtu.be/0yB_h_wFkh4
Using a transistor base-emitter junction you can make a less predictable oscillator than the 555. Just add a capacitor across the noise generating bit and you should reduce its effective frequency. Increasing the resistance and decreasing the capacitance should result in less predictability compared to the other way around.
Also you know, a BJT is smaller than a 555.

[Anonymous No.2422158]

>>2421684
>>2421694
Holy shit I flipped the board and there's very obvious residue, return initiated thanks guys

[Anonymous No.2422184]

>pinephone arrives
>it's a buggy mess
Well at least I'll have fun making custom I2C peripherals for it. Any interesting ideas? Might be possible to give it the capability of doing payments, though that would likely be a nightmare on the software end. Pretty sure I'll put a LoRa transceiver on there since that sounds extra fun. If I can get a thermal camera into a custom case I'll give it a shot too.

A meme case with some test equipment in it could be fun.
Or a fully waterproof case.

[Anonymous No.2422196]

>>2422124
if you want to use a 555 then use 2 and logic gate "and" or "nand". this will only change the frequency to be when the 2 are both sending a pulse. this isnt random by itself. you can however add something. you see the 555 timer will want at least 2 cap in any circuit its used . you can add a transistor to connect and disconnect a cap thats run in parallel to the first. this transistor can if mosfet small signal type take a reading from another line like the output signal or what ever. time is a factor but you can do slow turn on. super simple resitsor cap bjt instead for better result. it staggers the on and the off can be a alternate with smaller cap slow turn on to actually shut it off. these can be refuted with another set with a serial register given signal by the output final from nand gate and every lets say 3 times the serial in serial and parralel out it tgrips another transistor to switch on from the slow turn on 1 to off or from pin 8 off on slow turn on 2 to back on . this doesnt change if the 555 said its going at frequency 1 or 2 its the interrupt... with a interrupt

the maze of madness only ends when you give up on the size being to big and wont stack with female / male header even more. you can add in relays to to latching relay logic gate calucator circuit shit to change a binary input to a binary output read by a parallel in parallel out eeprom

you can also bypass some of the stated and shoot for the eeprom with multiple 555 timers with different caps and another for the eeprom clock speed pulse if you want to be lazy . sometimes it reads pin 3 as on and others off but not at the frequency its making because of the clock frequency of the eeprom clock pulse 555 timer

[Anonymous No.2422282]

any cheap impedance analyzer you recommend? or I'm cursed to waste >20k on some keysight shit?

[Anonymous No.2422303]

>>2422157
this video was fascinating thank you. i'd definitely like to learn more about this. let's say i used a noise generation oscillator, do you have a suggestion for how to use it to drive the LEDs, or should i still use the 4026 and run the reduced BJT into its clock pin?
ideally, i'd like to find an alternative for it since its current output is really low and the old 4000 series is pretty weak at low voltages, but i don't know what else to use.

>>2422196
it's not so much that I want to use a 555, but rather I'm a beginner and almost all my search results return circuits that use it.
i don't fully understand your suggestion and it sounds like it uses more components(?), but i've copied it down and will research everything you said, thank you

[Anonymous No.2422348]

>masters in RF electronics/EMC
>masters in computer vision
I live in the third world(south america lel) if I get in the AI meme I will just be another brown person writing shitty medium tutorials. Meanwhile EMC is literally wizardry. help sirs

[Anonymous No.2422352]

>>2422348
Stay out of direct sunlight for the next 7 years. That should lighten up your complexion up a bit. P.S. Don't eat bananas in front of people because they might mistake you for a loose chimpanzee, tie you up and drop you off at the zoo.
Good luck, Anon.

[Anonymous No.2422356]

>>2422352
you are implying I can emigrate out of here
>EU needs a company to sponsor me and has a housing shortage
>USA same thing with the sponsors
covid killed my chance at making contacts while I was in grad, I know no one in the industry. Almost biting the javascript web development bullet

[Anonymous No.2422359]

>>2422356
You could blend in with the Central American horde of faggots and just walk right into Texas. Joe Biden will send a private jet to fly you anywhere you wanna go, and you can loot high-end department stores without worry of getting arrested. You could go to Los Angeles and make a living by looting freight trains for Amazon packages, and shooting random passers-by. Don't forget to apply for your free* drug paraphernalia care package and free* bus pass. Viva Murica.

*Paid for by people who actually have jobs.

[Anonymous No.2422405]

I want to get into electronics.
I know there's a guide in the sticky and I haven't read it yet, but is there something that I should buy as a beginner? I probably don't have anything. Like should I get a breadboard or a multimeter? And some basic components like resistors etc.? Like what's the minimum that I will use for sure, to just learn and not needing for something specific in particular

[Anonymous No.2422425]

>>2422359
You sound like the guy they apprehended yesterday

[Anonymous No.2422438]

>>2422303
By itself it's just a replacement for the 555 oscillator. But you may be able to use it to replace the 4026 as well with the right surrounding circuitry. It's a white noise generator so my thinking is to get rid of the high-frequency noise with a filter, schmitt trigger it to a moderate frequency square wave, then feed into some clock dividing ICs. Twisted ring counters are decent for this. But that takes a lot of ICs. That without the clock dividers would make a particularly random oscillator for the 4026 though.

>>2422405
There are decent starter packs on amazon or aliexpress or ebay. The ones with an assortment of resistors and capacitors, a breadboard, jumper wires, plus some diodes, transistors, op-amps, etc.
But a lot of those kits have an arduino instead of simple ICs, while I'd personally recommend getting into simple digital and analogue circuits first (understanding ohm's and kirchhoff's laws) before messing with microcontrollers.

[Anonymous No.2422441]

>>2422405
>something that I should buy as a beginner?

you can find shit like this at Amazon, Ali, Ebay, etc, if you search electronics kits.
choose one that comes with a manual (projects to build)
if you want to go for computers, similar kits exist for Arduino and Raspberry Pi

[Anonymous No.2422442]

>>2422425
The antifa tranny discord shooter? lmao

[Anonymous No.2422501]

>>2422438
thanks i really appreciate it, this will be what i pursue as of now. maybe i can use the newer CD74HC4017 instead of the 4026 because it has a lower supply voltage range, and greater output current.
>Counter ICs Decade Cntr/Div w/10 Decoded Output

[Anonymous No.2422642]

Since the MCU thread is dead, I'm asking here.
Which would be the best digital protocol to use to communicate two microcontrollers through a 433 MHz wireless link?
I'm leaning to use UART with a certain signature but I want to read other opinions.

[Anonymous No.2422670]

>>2422642
I'll remake the mcu general later. Do you know what kind of keying those modukes use? If it's on-off-keying I'd lean towards manchester rather than async. Too easy to get lost in a sea of zeroes or ones, even with UART's start and end bits, because there is no "asserting low" like there is with RS232. That said, there is definitely a draw to using a protocol supported by your MCU's native hardware, and now that I think about it the end-bit isn't needed anyhow. Maybe ask the ham general. And watch Ben Eater's series on data transmission and error correction.

[Anonymous No.2422680]

>>2422642
i2c . spi could be used but just i2c it. 2 line connection. if its a radio transmitter and receiver 2 lines should be all you need to send a signal to be carried by the carrier signal

[Anonymous No.2422763]

>>2422680
I2C is synchronous though, it needs a clock signal. Do you mean to mix it with the clock as manchester? Guess you could do that in external hardware (IIRC just an XOR, maybe you'd need extra stuff) so you can still use the integrated TWI hardware on MCUs and sensors.

[Anonymous No.2422768]

New /mcg/ (microcontroller general) for anyone waiting: >>2422767

[Anonymous No.2422862]

so i can drive a dc motor with pwm.
but i was wondering what is the minimum frequency? because of you use like 0.001HZ then it's not really pwm anymore, you are just turning it on and off

[Anonymous No.2422864]

>>2422862
>when does a bunch of sand grains start becoming a bunch of sand
really makes you think huh

[Anonymous No.2422871]

>>2422862
>then it's not really pwm anymore, you are just turning it on and off

if it's turning on/off then it's PWM even if the period is billions of years.

[Anonymous No.2422885]

Hi Guys, I'm wondering if someone can help me? I'm trying to find out how to link up these in a circuit. Basically I have a quartz clock mechanism that runs on a single 1.5v AA battery, 4 Micro LED's that are 3-3.3V DC, and a 9V battery all operated in one circuit by this small switch I cant find out if this single 9v battery would be enough or what capacitor I would need. I have been trying to understand simple electronic circuits and it seems there was something about voltage drop in relation to the LED's, and if each LED is 3v, plus the clock mechanism then 9v would not be enough? Thanks if you could advise me

[Anonymous No.2422894]

>>2422885
>Thanks if you could advise me

thanks if you could clearly explain what the fuck you are trying to accomplish.
nobody's gonna show you the way if they dont have the destination.

[Anonymous No.2422974]

>>2422885
Yo, clock boy. Where do you live now?

[Anonymous No.2422976]

>>2422894
My bad, I will try to explain a little better. Here I have made what I think the circuit would look like? Basically I want to make it so when I engage the switch the 4 small LED's light up and the clock mechanism begins to turn. Normally the clock is operated by an AA battery inside it, but I'm hoping I can just solder wires to the terminals and it will still work that way.

So yeah just so that the lights and clock are both operated by one switch and one battery, or 2 batteries if needed. I suppose my main questions are,

1. What Ohm resistor I would need for this circuit

2. How would I wire the circuit correctly i.e. in series or parallel.

I really don't know much about circuits to be honest, I have done some research on simple LED circuits on YouTube and got them to work, however integrating the clock mechanism, and working out what voltage battery and what ohm resistor I would need I'm struggling with.

[Anonymous No.2422991]

how the fuck do i turn 10k thermistor resistance into temperature? it thought it was just some simple calculation but everyone seems to use Some gay ass lookup tables i don't have space for

[Anonymous No.2422993]

>>2422976
>How would I wire the circuit correctly i.e. in series or parallel.
Parallel. LEDs don't need any current limiting, but the clock does.
As for the resistor value, you first need to know the current draw of the clock. For example, assuming a 10mA draw you'd need 7.5V / 0.010A = 750 Ohm.
That being said I'd change the circuit a little. Typical voltage drop of a white LED is around 3V so you could light up each LED in parallel with 2 AA batteries in series. You'd still need a resistor for the clock but now you need to lose only 1.5V instead of the 7.5 from the 9V block so it's more efficient.

[Anonymous No.2422999]

>>2422993
>2 AA batteries in series. You'd still need a resistor for the clock
He could just tap leads of one battery.

[Anonymous No.2423011]

>>2422993
Thanks for this. I’m struggling to understand, if each LED has a voltage drop of 3v does it not stack to the point I need like more than 12v to even get them to light up? How would 2 AA batteries be enough? Also is there a way to workout the draw of the clock without one of these current measuring devices?

[Anonymous No.2423021]

>>2423011
If identical LEDs are wired in parallel then the voltage drop is the same as one LED. If they're wired in series then it adds up.

[Anonymous No.2423030]

>>2423011

[Anonymous No.2423032]

>>2422991
>how the fuck do i turn 10k thermistor resistance into temperature?

I'm using a 100k in a voltage divider, with arduino nano to read the voltage and convert to temperature. Per the attached image you can see it's somewhat linear. That curve is from the formula that is used to generate the lookup tables. I can't find my link, but it's online somewhere.

[Anonymous No.2423037]

>>2423030
thank you, I will try this!

[Anonymous No.2423040]

>>2423037
3 AAs in series gives you 4.5V. If your LEDs have a forward voltage drop of 2V, then you want a 33-Ohm resistor in series with each LED.

[Anonymous No.2423053]

>>2422999
>He could just tap leads of one battery.
That would mean one battery dying early.

>>2423011
>I’m struggling to understand, if each LED has a voltage drop of 3v does it not stack to the point I need like more than 12v to even get them to light up?
If you connect the LEDs in series then yes, the required voltage stacks and each additional LED needs another 3V. If you don't have enough voltage then none will light up.
If you connect them in parallel then each LED gets the required 3V, but the total current drawn from the battery goes up. Instead of the voltage requirements stacking, it's the current that stacks now. So if an LED draws 4mA then each LED connected in parallel draws an additional 4mA.
>inb4 does this mean you're using less energy when connecting in series?
No, in fact it's exactly the same if you ignore any losses from wire resistance and such. 3 LEDs in series : 9V 4mA, 3 LEDs in parallel: 3V 12mA. Total power consumed is voltage multiplied by current so it's 36mW in either case.

[Anonymous No.2423057]

>>2423053
>one battery dying early.
The clock doesn't draw much current (190μA).

[Anonymous No.2423246]

>>2422862
PWM usually implies that it's being switched at a frequency high enough not to influence other time-based functions of the load. Say for a fan motor, it's effectively PWM when the frequency is high enough that the fan's own rotational inertia buffers the changes from high-to low; acting like a low-pass filter effectively, while 0.1Hz definitely isn't acting like PWM because the fan noticeably speeds up and slows down during that time. And for LEDs, it's effectively PWM when persistence of vision makes it look continuously on, and not when you can see the flicker. At the extremes, you could switch a heating element on and off with sub-1Hz and still be effectively PWMing it if the thermal mass is large enough.
Nothing rigorous, but that's just my rule of thumb for what it means for something to be usefully PWM'd.

[Anonymous No.2423258]

hye-o. I'm really bad with electricity. I want to understand capacitors.

are capacitors like this? a resistor will take a large amount of energy: the fairly infiniite amount of energy from the power station, and reduce it down to a very small size. A capacitor will do the opposite. Since because it collects energy and the energy flowing into a system from the power station is generally speaking infinite, then I imagine it collects enough energy to reinstate the energy reduced from a resistor even if the resistor is placed before it in the series.

[Anonymous No.2423269]

>see 120V to 240V switch on SMPS
>trace circuit
>it turns the bridge rectifier into somewhat of a voltage doubler
Oh that's kinda clever. Means you don't have to spec the bootstrapping circuit to cover a wide input voltage range.

[Anonymous No.2423279]

>>2423258
No. Capacitors and inductors are opposites of one another. It's easier to understand impedance in the framework of complex numbers, which I'd recommend looking into. Or calculus at the very least.
Resistors serve to limit current in voltage-based situations, and limit voltage in current-based situations, through I = V/R and V = I*R respectively. They dissipate power through P = V*I = I^2*R = V^2/R.
Capacitors buffer voltage through changing currents, through I = C*dV/dt. As you can see, if voltage across a capacitor remains constant, current through that capacitor is zero. They do not create or destroy power, just input and output it over time. They store energy through E = 0.5*C*V^2.
Inductors buffer current through changing voltages, through V = L*dI/dt. As you can see, if current through an inductor remains constant, voltage across that inductor is zero. They do not create or destroy power, just input and output it over time. They store energy through E = 0.5*L*I^2.

FYI: P = dE/dt; power is energy over time.

[Anonymous No.2423289]

>>2423279
I still understand very little about electricity

I think I understand a little. I've had trouble developing a stupid experimental means by which to measure the generation of electricity but I would imagine that whatever means I eventually develop to produce electricity would fluxuate in strength

probably also like real electricity from the power company. so I think that capacitors regulate the electric current by storing small amounts of electrical charge to help in small periods of electrical fluxuation

I haven't integrated that into my understanding of electricity yet. from what I understand, the power company always makes a lot fo electricity, so I don't really see how having a capacitor will help keep the machine running whenever power fluxuates briefly, but I guess that I think I understand that this is how that works for right now

[Anonymous No.2423346]

Why does the oscillation frequency of a step response of an underdamped series RLC circuit equal the resonance frequency? It kinda makes sense logically but what's the theory behind it? AC Sweep generates a bode plot with the peak at the resonance frequency, and the transient analysis agrees with that.

[Anonymous No.2423482]

>>2423289
>the power company always makes a lot fo electricity
>I don't really see how having a capacitor will help keep the machine running whenever power fluxuates briefly
Sudden loads (like switching a motor on) can easily draw dozens of amps instantaneously. If you have a 5A power supply but a motor that needs 20A in short bursts, then you'll need some capacitors to supply that current. Or maybe you do have a 20A power supply, but there's 1Ω of resistance between the power supply and the load, which will limit the instantaneous current significantly.
In the case of wall power, it's AC so capacitors directly across it are generally reserved for noise suppression, which is a somewhat different kettle of fish.
There's also the equation for post-rectification ripple as a function of capacitance:
>I = C*dV/dt
>I = C*∆V/∆t
>I = C*∆V*f
>∆V = I/(C*f)
Where f is frequency, C is capacitance, I is current, and ∆V is the peak-to-peak ripple voltage after rectification. Assumes a constant current load and more of a sawtooth wave, but it still holds up fine in most cases. Double the frequency for a full-bridge rectifier.

Are you the generator designing anon?

>>2423346
Intuitively, it's just the energy stored in the capacitor moving to the inductor and back again, and it does so with the natural resonant frequency because it's a simple harmonic oscillator. If you want an analytical expression you could just construct the differential equation for the system, assuming that you set the initial conditions to be some non-equilibrium state. The angular frequency should fall out pretty easily from the exponential, assuming your test function has both a real and imaginary component to describe the finite/nonzero Q factor.

From the Fourier perspective, the step function contains all frequencies at nonzero values due to its lack of periodicity and infinite steepness, so naturally some of that is going to resonate with the tank circuit.
>inb4 fourier transform of step is undefined

[Anonymous No.2423485]

>>2418059
Hi

May I join with a question, since I don't want to make an entirely new thread for this and this one seems fitting for my question.

Basically im a looking for a USB C power board that can output 6.5-7V at 1000MA or preferably a bit more.
I want to power a device with a USB power brick with QC and or PD.
All boards I found only seem to be 5/9/12/15/20V.
Is this a useless endeavour and I will have to use and additional step down or step up converter in-between?

[Anonymous No.2423495]

>>2418059
I want to embed magnets into a high-permeability material to multiply the repel-forces between them.

Permalloy & metglass is expensive. so I'll use something cheap.......
How many times stronger is the repel-force if I use Grade-AA MILD-STEEL?
material = Grade-AA MILD-STEEL
permeability = 1000

Another problem is blocking the Magnetic-Pole-return-path...
Do any of the configs in pic solve the problem?

[Anonymous No.2423553]

hi /ohm/

i have a variac that i wanna build a ac/dc power supply from

is it really as simple as it seems?
>FUSE
>isolation transformer at mains input
>variac hooked up to it
>spst switch at variac output
>ac toggle = pass through
>dc toggle = full wave bridge rectifier
>in rectifier: some big ripple filtering caps and a bleed resistor
>ideally a voltmeter+ammeter
am i missing anything?

[Anonymous No.2423558]

>>2423553
Perhaps better to put an isolation transformer or step down xfmr (if needed) after the variac/autotransformer.

Be sure the output of the autotransformer has a fuse in the secondary because it is easy to fry a variac output winding if the output current is too high. An input fuse can be quite happy while you are frying that part of the output winding that supplies the output current. Refer to specs for fuse value. Many have a fuse built in, but might require disassembly to replace it.

[Anonymous No.2423564]

>>2423558
Correction: ...it is easy to fry a variac output winding...

Should read: ...it is easy to fry a variac...

This was wrong as a variac has only one winding.

[Anonymous No.2423595]

>>2423482
>inb4 fourier transform of step is undefined
is it not? this is confusing because on the one hand there is LT for cases like that but on the other hand it may turn out that some of those "bad" functions are FTtransformable after all. I remember some examples of that but it is all in the details which I don't remember. I know there are some tricks to evaluate the FT of the step function but it can be done (sgn). Same with dirac gelta. It is all about dirty tricks. Math is dirty.

[Anonymous No.2423614]

>>2423553
>am i missing anything?

a sanity check.
i once had a DC supply that could output 75Vdc but was leery of using it coz i knew that a quick twist of the wrong knob could destroy my project in a microsecond.
your idea would create something like 160Vdc at max, which should be terrifying to use on anything expensive, or on any animals or humans you might care for.

[Anonymous No.2423623]

>>2423614
Shouldn't the triangle have 2 bolts of lightning inside of it?

[Anonymous No.2423652]

>>2423614
so would it be just better to switch between single 3v, 5v, 12v supplies and bypass the variac for dc entirely?

[Anonymous No.2423655]

>>2423623

mind your own business punk

[Anonymous No.2423663]

>>2423655
Safety is my business.

[Anonymous No.2423729]

I have these pins coming through the board, and I need to attach wires to a few of them. The problem is, there is only 2-4mm of clearance between the pins and the bottom casing. Its 24awg wire, but i feel that bend will put pressure on the solder joints and they will break.

what am i best off using if i need a 'wire' at an exact right angle that will fit into this tiny space?

[Anonymous No.2423730]

>>2423729
magnet wire

[Anonymous No.2423758]

>>2423729
>only 2-4mm of clearance between the pins and the bottom casing

-- solder the wires FLAT so there is no bend.
-- place a sheet of bakelite or similar (e.x. PCB without copper) at bottom of cabinet

[Anonymous No.2423761]

>>2423730
>>2423758
thanks! i felt like a stupid question - but that also usually means a quick answer

[Anonymous No.2423788]

>>2418271
Use op amps?

[Anonymous No.2423844]

>>2423495
Take apart an old transformer and use its silicon steel laminations. Pretty sure you can buy that stuff as-is from alibaba and such, but it may cost a bit. Or just use the good old Halbach array, which is (A). Note that the smaller the side magnets are, the less of an impact they'll have, which is why most Halbach arrays I see use similarly sized magnets for the forward poles and the sideways poles.
(C) and (E) are a non-improvements over (A) with virtually no difference. (B) is effectively the same as no extra magnets at all, same for (D) and (F).
The problem with (G) is that the side iron will drag the north-facing magnetic field wider, I suspect it would be better to just have the permeable core on the back of the magnet. Or at least no further than halfway up its length. Having a permeable core at the back of a Halbach array would be ideal.

You could also double up the magnets to complete the magnetic circuit, by having a N and S pole next to one another. It may be helpful to watch Applied Science's video on magnetic circuits.

>>2423553
As the other anon said, isolate it. Don't want a power supply referenced to mains via a bridge rectifier. Best case any electronics gets a shitton of 120Hz noise. Worst case you get a ground loop and trip an RCD.
Instead of a fuse I'd put a fast-blow breaker, and use a crowbar as a part of current-mode feedback.

>>2423652
Yes, especially if you just need a fixed DC voltage. Research what variable DC power supplies do.

[Anonymous No.2423848]

>>2423844
okay thanks i'll look into mounting 3.3/5/12v dc units. that's really what i need, didn't know if i could remove some circuits/fiddles. but diy power is kinda dumb

[Anonymous No.2423852]

>>2423848
>but diy power is kinda dumb
Not unless you make it smart!

Whatever you do, you'll likely want some form of current limiting protection, so look into a pass transistor and op-amp or an overcurrent latch that you can add to a CV power supply.

[Anonymous No.2423895]

I have a couple of old radios and when I press a button, sometimes it'll behave as if I pressed a different button, or both in quick succession. Is this just an issue with dirt/dust? Or deeper rooted? Function is fine otherwise.

[Anonymous No.2423902]

>>2423895
Some devices will multiplex their inputs, sharing multiple buttons along one I/O line by attaching different resistors to each switch. If a button contact is getting dirty, then it could cause misfires as the resistance of the contact would start contributing to the circuit.

[Anonymous No.2423903]

>>2423902
Makes sense, thanks!

[Anonymous No.2423907]

I'm trying to repair my adapter to learn more about electronics. It turns out the resistor fuse is broken. The PCB says (RF1 3.3 ohm 1 Watt) or F2 T1A.
Can I replace the resistor fuse with regular 1A fuse ? It's hard to find resistor fuse here. Thanks in advance.

[Anonymous No.2423909]

>>2423907
>Can I replace the resistor fuse with regular 1A fuse ?
I think that's what it's implying, yes. Not sure how you calculate current rating of fuseable resistors, maybe there's a calculator for those. I imagine that's an input fuse, so 1A is plenty of room for anything below a hundred watts or so, maybe you could go lower (especially if it's a little USB adapter) but keep inrush current in mind.

Also look for any possible internal cause of the blowing of the fuseable resistor. It's just a symptom, the cause may well still be present. So get more than one fuse, assuming you can even get ones that fit. I'm guessing axial lead M205 fuses mounted vertically.

[Anonymous No.2423910]

>>2423758
How do you get the solder to melt properly?

When I work on an of the shelf device I never get the factory solder to melt, not even with my bigger stronger iron that is used for glas work.

On the hand I can apply solder to my wire and the contact point, but even on the highest setting and long preheat I have trouble connecting the wire to the contact as the copper seemingly just soaks up the heat too fast.
This was a huge problem when I tried to solder some wire to a UVC LED chip that is only 3.5x3.5mm and has 4 contact points.
Each point being less than 1/6 of the area.
Individually applying the solder worked easily, but the issue always appeared when actually making the connection.

[Anonymous No.2423912]

>>2423895
goddam I cannot figure out how to get this AV receiver apart

[Anonymous No.2423914]

>>2423910
You need:
a higher power iron for lead-free work on boards with a heavy ground plane
a larger tip, not conical
good rosin flux
good LEAD solder

[Anonymous No.2423920]

>>2423912
The offending switches. They look pretty clean.

[Anonymous No.2423931]

>>2423920
Check the solder joints on the other side of the board.

[Anonymous No.2423939]

I hate the antichrist.
What the fuck does COM+5V mean on a fucking casio cash register? Is it saying that it's rs232 with a pin at 5v for power on the unused pins in the db9 connector? Why then, when I probed them, did I not find a fucking 5v

[Anonymous No.2423950]

>>2423931
I've already got it back together and the buttons are behaving. For now, of course. If I have cause to take it apart again later, I'll remember to check it out but it all looked kosher to me at a casual glance.

[Anonymous No.2423951]

>>2423939
RS232 or COM ports are old-style PC serial ports. It's the function USB was created to replace.

It probably doesn't send any voltage unless it has something to say, but I'm no pro. Google might have plenty of documentation on RS232 beahvior.

[Anonymous No.2424010]

>>2418059
not sure if this is going to die or not before I get some help, will re ask in a new thread if it does.

so i'm trying to make a mechanical fretboard for guitar hero controllers, I don't want to pay 30 fucking dollars for a premade one to show up, and if I can avoid it, I would rather not etch pcb. I have some idea on what I will do/what I can try, and from the few I have taken apart, it seems like they are all supported by plastic supports on the back side. I think that if I had a material that was about as thick as the pcb and rigid, I could get away with it, and line it up based on, but I need something that I can mount the switches to...

any idea on a good cheap material that if it did flex won't snap, preferably something that I don't have to be concerned about dust (I was considering a carbon fiber sheet for a moment)?

[Anonymous No.2424023]

>>2423844
Excellent summary, anon.

RECAP:
(A) For small magnet size.
(C) For magnet-size larger than is manufactured.
(E) For when its (C), but an isolated rotating shaft must pass thru its center.
(G) is same as (E), but this is a more powerful version.

Is (B), (D), (F), (H) needed for it to work?
Is (B), (D), (F), (H) an improvement over their counterparts?

Again, thanks for the assistance anon.
I consider this excellent progress.

[Anonymous No.2424050]

>>2423914
I thought 58W and especially 80W would be overkill, especially on those small SMD LEDs.
Won't I kill the LED with a bigger tip?

Thanks for the advice.

[Anonymous No.2424078]

>>2423920
May be cracked solder joints, which can be thoroughly invisible and really hard to troubleshoot. I'd consider reflowing everything on the back of the button PCB, the force required to push a button could lead to incremental fractures on the stressed board.

>>2424010
Carbon fibre would work great, due to its high rigidity. Fibreglass would be fine, with or without the copper. So would sheet plastic like polycarbonate or maybe even acrylic. Acrylic is a tad brittle though, so I'd try polycarbonate. If you're able to, I'd try to use a heat gun to bend it into an extrusion-like shape to give it strength in multiple directions. Might even be able to buy material like that.

>>2424023
I don't think (G) will be significantly better than (E).
(B), (D) etc. isn't a requirement, but it will be an improvement. Again though, the thickness of the sideways magnets and the thickness of the silicon steel material does matter. The thinner it is, the less effective it will be compared to not having it.

[Anonymous No.2424079]

>>2424050
>Won't I kill the LED with a bigger tip?
By bigger tip he means larger tip surface area. In general, you want to maximise the contact area of the tip, because doing so maximises the heat-flow into the workpiece, and minimises the time required to flow the solder. The shorter the time to flow the solder, the less heat makes it into the core of sensitive semiconductor components. It also means less of the flux will be burnt away, making your joints shinier and cleaner. Getting in and out fast is pivotal to good soldering for these reasons. Of course, you can't use an 8mm chisel tip for everything, you size your tip for the workpiece. In general, I'd recommend a screwdriver tip that's about as wide as the work you're handling. This likely means a 1.2mm or 1.6mm tip. The edge of the screwdriver is excellent at making solid contact along the seam of the joint, and making this contact with both the board and the part. The same applies to knife edge tips.
Conicals are just bad at everything.

The same desire to get in and out fast is why a higher power rating matters. It lets you heat up the joint quicker. I'd suggest that for an SMD board without a big ground plane you'll be fine with maybe 40W, but considering most cheap import irons are 70W baseline I see no reason to settle for less. Higher power also means you rely less on tips with high heat capacity, and so can respond to changes in thermal resistance and set temperature much quicker.

[Anonymous No.2424091]

>zener diode pass-transistor linear regulators with a cap on their zener are just improved capacitance multipliers
woah

[Anonymous No.2424093]

>>2424078
>I don't think (G) will be significantly better than (E).
(E) has the N-POLE field return path thru the center air-gap.
I thought Halbach array was so great "specifically" because it hindered the magnets field-return path back to its own counter-POLE?
Wouldn't (G) therefore be "halbach" better?

Also, volume(magnet) -vs- volume(lamination) would ultimately peak at some consistent ratio?
Having a tiny-magnet & meters of laminations is going to give a super-magnet?
What rule-of-thumb ratio seems industry consistent?

[Anonymous No.2424094]

>>2424093
>(E) has the N-POLE field return path thru the center air-gap.
But it doesn't significantly return through the middle anyhow, at least not if the magnets are significantly wider than the internal hole. Most of the returning flux will go by way of the outside, not the inside.
(G) will be better, but I'm suspecting the difference will be marginal in most situations. The benefit of a Halbach is getting the flux pointed more directly outwards to stop it from looping back around, but in the middle of your circular magnet it should already be pointing outwards, Halbach or not.

>Also, volume(magnet) -vs- volume(lamination) would ultimately peak at some consistent ratio?
Very geometry dependant I suspect, change the width or shape and the ratio would also change. If you're worried I'd install a magnetics simulator, there are a few FOSS ones out there.
>Having a tiny-magnet & meters of laminations is going to give a super-magnet?
Definitely not, it's much closer to the other end. If anything, laminations are used to replace otherwise structural pieces of volume, not to replace magnet volume. Motor stators tend to have the bulk of their structure made from silicon steel laminations expressly to guide the magnetic flux. While the same also applies to permanent magnet rotors to some extent, it's less of a big deal. Even less important when using a Halbach array, which takes away some of that magnetic flux that would be in the lamination. Also on the rotor of a synchronous motor there is no rotating magnetic field, so the laminations don't actually have to be laminated; there are no eddy currents to worry about.
>What rule-of-thumb ratio seems industry consistent?
I've never even seen the magnetics industry I'm afraid, I'm just saying what I've learnt as a hobbyist trying to design a brushless DC motor, and as someone who studied EM for a few semesters.

[Anonymous No.2424142]

>>2418070
I literally have one right here with me. It even has a Chinese knock off "Starr Whars" graphic on it

[Anonymous No.2424143]

>generic blue lipo batteries

can any autists here give me a quick rundown on these. why do they often come with a 12v supply, despite lipo charging voltage should be over 14v ? do they have a charging module in them?

[Anonymous No.2424144]

>>2418182
Yes, but any linear regulator does.

[Anonymous No.2424145]

>>2418246
Some cool synth voice unit, as it generates multiple harmonically related tones. Idk if your in to synth stuff.

[Anonymous No.2424147]

>>2418286
You dumb nigg3r, its a FLYBACK diode. It protects the switching device from inductive spikes.

[Anonymous No.2424166]

Posted in sqtddtot but thought I’d ask here before the thread dies. Does anyone know what the hell these iPhone battery leads are made out of and if it’s possible to solder/spot weld a new connection without sustaining grievous bodily harm? know I should probably shitcan it, but I hate wasting a good condition battery.

[Anonymous No.2424172]

>>2424166
It's hazardous waste. Do you like house fires?

[Anonymous No.2424209]

i don't want a cool post to go to waste so i will wait for a new thread.

[Anonymous No.2424218]

>>2424172
Only when they aren’t mine. I knew in my heart I had to dispose of it, but I figured I’d get a second opinion.

[Anonymous No.2424352]

>>2424209
same

[Anonymous No.2424364]

>>2424218
There are certain times when you just gotta throw shit out and this is one of those times. It sucks, but if that battery shorts out it could cost you dearly.

[Anonymous No.2424396]

>>2424094
>But it doesn't significantly return through the middle anyhow, at least not if the magnets are significantly wider than the internal hole. Most of the returning flux will go by way of the outside, not the inside.
Draw a new circumference at a radius that divides ring's area in half.
Top area half is halbach blocked, bottom area half isn't.
>(G) will be better, but I'm suspecting the difference will be marginal in most situations. The benefit of a Halbach is getting the flux pointed more directly outwards to stop it from looping back around, but in the middle of your circular magnet it should already be pointing outwards, Halbach or not.
true, but...
>getting the flux pointed more directly outwards to stop it from looping back around
Doesn't that "halbach" effect double its force?
Isn't that the hole point of using them?

[Anonymous No.2424397]

end goal is being able to make synths and musical instruments, not purely from components but also probably some teensy/arduino stuff down the road. but initially just components.

tried this a couple years ago but my fuck ups ran the budget up too high. have soldering iron, money and time again so wanted to try again. should i start with Ben Eater's 8-bit PC breadboard kit, so I'm less retarded this time? it's like $300 but it's essentially a course on low level computing

[Anonymous No.2424399]

>>2424397
here it is: https://eater.net/8bit/kits

would this give me a good foundation for proceeding to make some modules? i was making very dumb errors like buying the wrong IC size, wasting time with broken bread boards, breaking my soldering iron, etc. i've wanted to do this for awhile anyway but not sure if it's a huge unproductive detour from building synths/modules

[Anonymous No.2424403]

>>2424399
>>2424397
and i want to learn to diy pedals just get me fast tracked there bros

[Anonymous No.2424404]

>>2424397
>should i start with Ben Eater's 8-bit PC breadboard kit, so I'm less retarded this time?

no, this would be like doing a lot of brickwork in order to learn carpentry.
if yow wanna learn synths, stay with synths, not vintage computers.

[Anonymous No.2424405]

>>2424404
any good kit you recommend? I got really bogged down last time trying to order exactly the right component. i know if i can just get the hang of it that'll be behind me shortly.

[Anonymous No.2424408]

>>2424404
last time i was trying to go with no kit and just this tutorial series https://www.youtube.com/watch?v=5DKU1m5_b_8&list=PLyE56WXw0_5Q5QGMEXWmskuhojKyRdA3T

but it was hard for me to read the schematics sometimes or source the exact component

[Anonymous No.2424409]

>>2424094
>>Also, volume(magnet) -vs- volume(lamination) would ultimately peak at some consistent ratio?
>Very geometry dependant I suspect, change the width or shape and the ratio would also change. If you're worried I'd install a magnetics simulator, there are a few FOSS ones out there.
Decade or so ago, I compared what "science" had to offer -vs- "measured" results.
https://www.kjmagnetics.com/calculator.asp
"science" is below par over the full dynamic.

[Anonymous No.2424411]

>>2424094
>hobbyist trying to design a brushless DC motor,
I'd be interested in seeing its design, I too am faced with these issues.

[Anonymous No.2424417]

>>2424408
>just this tutorial series

you need to find a better teacher who is
- organized
- explains well
- draws well
- has a plan, and stays on topic

instead you chose of the many many Youtube idiots who just turns on the camera and blabs for 30 minutes with zero preparation.
i cant recommend anyone coz i aint into any of that jazz.
but a model teacher would be the EEVblog guy on Youtube.
also, i'd start with guitar pedals coz they can vary from the supremely simple to very advanced.

[Anonymous No.2424418]

>>2424094
>already be pointing outwards, Halbach or not.
>to stop it from looping back around
Pointing it forward & preventing it from looping back is what makes it stronger.
Imagine that entire outwards + looped-back field-return-line compressed into outwards and back to closer adjacent POLE, instead.

[Anonymous No.2424419]

>>2424417
>EEVblog guy
I'm not into disorganization either of course, it was just the most straightforward series I found. A blog or so would be preferred to video (ideally with videos of the actual construction/soldering), but I couldn't find shit.

Maybe I'll try pedals first, just more enthused for synths really. But you're right in that it'll be easier finding a good resource for pedals..

[Anonymous No.2424424]

>>2424419
https://sound-au.com/p-cat.htm
https://www.diystompboxes.com
https://www.freestompboxes.org
http://www.generalguitargadgets.com

Something to get you started.

[Anonymous No.2424425]

Is it feasible to DIY something like this: https://www.sweetwater.com/store/detail/REV2mod-16--sequential-prophet-rev2-16-voice-polyphonic-analog-synthesizer-module?main_web_category_rollup=498/510&mrkgadid=3281233063&mrkgcl=28&mrkgen=gpla&mrkgbflag=0&mrkgcat=keyboards&synthesizers&acctid=21700000001645388&dskeywordid=92700046938733690&lid=92700046938733690&ds_s_kwgid=58700005287622136&ds_s_inventory_feed_id=97700000007215323&dsproductgroupid=428799361838&product_id=REV2mod-16&prodctry=US&prodlang=en&channel=online&storeid=&device=c&network=g&matchtype=&adpos=largenumber&locationid=9032875&creative=217452498644&targetid=pla-428799361838&campaignid=916403487&awsearchcpc=1&gclid=CjwKCAjwq5-WBhB7EiwAl-HEkoP2nEZ_J-jjtQcj9I2Y4LYVXJIuIuGmvcDWnLsSCRS1nrHDnFD6LxoC6m0QAvD_BwE&gclsrc=aw.ds

A DSI Prophet Rev2, a 16-voice analog synth. If not, is it possible to DIY clones of classic monosynths like Roland SH-101? just throwing out my ?'s before the thread dies then i'll leave u guys alone

[Anonymous No.2424426]

>>2424424
Cool, so you think if i do this I might be able to just figure out the synth thing on my own? e.g, just read the schematic for a VCO and know what I need to order, rinse and repeat until I have all the modules I want? seems like an extremely rewarding hobby once u get past the language and skill barrier

[Anonymous No.2424428]

>>2424425
Mother of pearl, that's a lot of url. Here, take this.

https://www.sweetwater.com/store/detail/REV2mod-16--sequential-prophet-rev2-16-voice-polyphonic-analog-synthesizer-module

[Anonymous No.2424432]

>>2424428
>that's a lot of url
Yeah, sorry. But do you think it's something I could accomplish after a few years, if I was given a schematic? Or is there an innate magic to the manufacturing like with CPUs?

That would be amazing to have a studio full of cheap, quality DIY gear.

[Anonymous No.2424433]

>>2424424
Do I need to read 'Art of Electronics' or some other circuit design book if I want to get good at this?

[Anonymous No.2424434]

>>2424426
You will figure it out, but you seem impatient. Start small and take your time. Ask questions if you get stuck. Once you get a few circuits under your belt (i.e. after you fail several times) you'll gain a better understanding of more complex circuits and how to select components for your own modifications and designs. Just keep at it and go easy.
>>2424432
>after a few years
It all depends on how much time you have to spare and if you have other responsibilities/distractions that regularly require your attention.
>>2424433
>Do I need to read 'Art of Electronics'
You should eventually but it covers more advanced topics and circuits.

[Anonymous No.2424439]

>>2424434
>you seem impatient
Good call, you're right. I'm angry as well. I spent way too much last time and have nothing to show for it. I will relax and embrace the process rather than smashing my keyboard hoping for success.

>It all depends on how much time you have to spare and if you have other responsibilities/distractions that regularly require your attention.
Sounds like it's not unreasonable then, but is this still true if I was given a schematic or the instrument first hand (from which I'd derive the schematic)? I can see how reverse engineering without a physical copy would be challenging, but if you have a circuit in front of you that you can map out, shouldn't it be, for lack of a better word, trivial? Assuming you can appropriately map the circuit to a schematic and implement said schematic. Am I massively underestimating or misunderstanding something here?

I can dedicate a lot of time, and I know I'm being a shit ITT but sustained focus over time is one of my strengths. I'd kill for an IRL mentor though as navigating this shit is cumbersome, but pedals will probably be easier.


>You should eventually
Is that the best book for circuit design? Basically, I'd eventually like the skill to DIY overpriced music equipment rather than pay the ridiculous prices for them.

[Anonymous No.2424442]

>>2424439
It covers a ton of ground. I think the best approach is to pick something relatively simple and build it. Then modify it until you break it. Read the datasheets of the components and ICs you'll be using. Don't limit yourself to a single book either.

[Anonymous No.2424450]

>>2424439
>if you have a circuit in front of you that you can map out, shouldn't it be, for lack of a better word, trivial?
It would seem so. Then you power it up and it either gasses you with the magic fentanyl cloud or does nothing at all. Troubleshooting time, YAY.

[Anonymous No.2424452]

>>2424450
lol fair enough, at least components are cheap

[Anonymous No.2424458]

>>2424442
If it's that easy why don't you see people selling kits for all popular pedals and classic $1k+ synths?

[Anonymous No.2424465]

>>2424458
Never said it was easy. Stomp boxes are pretty straightforward for the most part, but synths and effects are a different animal. You start getting into FPGAs and Verilog or VHDL for realtime digital effects, DSP and all the math that goes with it, filters and pre-amp stages, MIDI, etc. Analog shit is even less forgiving as far as consistency between boards.

[Anonymous No.2424470]

>>2424465
Well if it's not infeasible to accomplish a clone at home after enough experience and study, then why don't we see DIY kits being offered for nearly every popular synth or pedal? There'd be huge economic incentive to offer literal clones for $2k+ pieces of equipment, given that components would cost probably less than $100.

Just curious, this has always made me feel it's unlikely to be achievable.

[Anonymous No.2424471]

>>2424470
It's impossible to compete with a slave colony of 1.5 billion jaundiced bugs. Economy of scale, or something- and Joe Biden being a CCP agent and colossal fuck up.

[Anonymous No.2424474]

>>2424470
You also have to write the code for all the micros and FPGA unless there's some FOSS shit on github that fits your requirements.

[Anonymous No.2424477]

>>2424471
>>2424474
Why aren't the slave colonies producing bootlegs that undercut the price though, like they do for everything else? I understand you'd have to write the code too, and believe that's arguably the bigger obstacle, but that doesn't apply to purely analog devices, which tend to fetch extreme prices for the classics.

They'll bootleg entire windmills for example, why is a Rev2 or SH-101 off limits?

[Anonymous No.2424478]

>>2424471
And regarding competition, these prices are often high because it's 'boutique', not because it's uniquely hard or expensive to manufacture. Take the Juno-6 or Juno-60 for example, a very small synth that fetches extremely high prices for what seems could be chinked out and bootlegged for under $500 no problem, including the keyboard and housing. I just don't get it?

[Anonymous No.2424480]

So anyway it seems like there's an extreme opportunity for someone who can make bootleg classics, not that that's why I'm here, but it's what makes me skeptical that a person can realistically reproduce a circuit like that. But I also don't see why it wouldn't be `trivial`, once you can transcribe diagrams and build accordingly. It's just a weird thing to me, doesn't make any sense given that the chinks exploit everything like this.

[Anonymous No.2424482]

>>2424470
>why don't we see DIY kits being offered

imagine having to deal with druggie musicians who are very technically incompetent, and very eager to thrash your product coz they dont know which way a polarized resistor goes in.

[Anonymous No.2424484]

>>2424482
Imagine selling on alibaba where it doesn't really matter though? i guess that's maybe a problem for american sellers though, but it still seems odd. if you're given an pamphlet on what to solder where, it's just a mechanical process after that

[Anonymous No.2424485]

>>2424480
>what makes me skeptical that a person can realistically reproduce a circuit like that.
The most difficult part is producing a multi-layer PCB with fine-pitch traces and pads. Second is designing the circuit/board, writing the code and debugging. The easiest part is assembling the board.
IN MY OPINION

[Anonymous No.2424487]

>>2424485
again if it's purely analog like many famous synths you don't have this coding and debugging phase right? i don't know what a 'fine pitch trace' or 'pad' is but if it's related to digital signal processing it can likely also be thrown out in place of analog potentiometers, right (which are, presumably, much simpler)?

[Anonymous No.2424489]

>>2420297
Not h but do you mean something like this?
https://hackaday.com/2019/04/27/driving-a-16x2-lcd-with-voltage-modulation/

[Anonymous No.2424490]

i.e, i get that digital is probably hard, but analog - hypothetically - seems as easy as copying and pasting a design. there aren't 'closed source' analog devices for example. you can look at a circuit and presumably tell, after enough debugging, what's going on, without being an artist or a sound designer or anything like that

[Anonymous No.2424491]

>>2424489
*him
fucking mistyped again

[Anonymous No.2424496]

>>2424490
Why do people spend thousands of dollars to buy an old video game still in the shrink wrap? The chinks could clone it for $.0001. Why don't they?

[Anonymous No.2424497]

>>2424496
These aren't collectors items still in the shrink wrap, they're unique circuits that produce unique sounds. I don't care what it is, if I could buy it for 10% of the price I would. But at current prices, I simply will not fork over $30k for a studio when I could just try hard for 2 years and do it for $1k. Compounding on this is the fact that I never intend to be pro or publish, just wanna have fun.

[Anonymous No.2424499]

>>2424496
And I'm not a chink, I'm Taiwanese retard.

[Anonymous No.2424506]

>>2424497
There was an Anon here a while back who posted his diy synth / effects rack module setup and it was pretty extensive. There were patch cables all over the place. Pretty fukken kewl.
>>2424499
I hope you don't get Sheenpeen'd, Anon.

[Anonymous No.2424508]

>>2424408
Lantertronics is a goldmine. There’s also this guy who just made a video about a white, blue, and pink noise generator, he does good stuff too.
No clue about kits, but you should aim to get yourself to a point where you have enough knowledge to not have to rely on a full parts kit. Also make an elektrosluch, a simple audio circuit.

>>2424425
It’s feasible to DIY anything that doesn’t require proprietary components. But it’s not necessarily easy. Any complex circuit can be broken down into numerous smaller sub-circuits. Keep deconstructing until you understand a low layer, then start building up again to see what the big picture says. There’s some schematics online for old synths that I recommend diving into for half an hour or so.

>>2424426
It’s not that rewarding to just copy someone else’s design. What’s really rewarding is to come up with your own design that actually works. You won’t be able to do that initially, but treat all these existing circuits as just means to that end. Try to learn exactly how every circuit you make works.

>>2424485
Just order boards from the chinks.

[Anonymous No.2424515]

>>2424508
>There’s some schematics online for old synths that I recommend diving into for half an hour or so.
Any model recommendations?

[Anonymous No.2424520]

i finally got my buck/boost chips in the mail today, now i can build my power supply the right way

[Anonymous No.2424539]

>>2424520
>the right way

hardly.
switching supplies are dirtier than my mom.
gonna waste a lot of time troubleshooting why certain sensitive circuits wont work before realizing that.

[Anonymous No.2424549]

>>2424515
TB303 was good IIRC.

>>2424520
Great! You got variable frequency ICs, right? With external switches? Using current-mode boost? With synchronous rectification? With high-current totem-pole gate drives?

[Anonymous No.2424551]

>>2424539
>not using a capacitance multiplying linear final stage
NGMI

[Anonymous No.2424584]

>>2424143
They look awful. Wouldn't touch with a barge pole, but if you're willing to slice off the blue shrinkwrap and check for protection/charge circuits that could be an informative process.

>>2424166
I soldered a wires to the end of the connector of one of those, which worked long enough without anything exploding. Not a very secure connection though.

>>2424409
That just means your characterising variables aren't accurately defined. Unless you have something to say to Mr Maxwell over here?

>>2424411
Pic related. It's based off an existing stator (no time for making a stator from scratch), but I'm turning a single-sided outrunner drone motor into a 2-sided wheel hub motor. Roughly following the idea from https://www.instructables.com/id/Make-Your-Own-Miniature-Electric-Hub-Motor/
There may be some changes I'd make if I weren't constrained to this stator, but for the most part I'm happy with it.

>>2424418
Yeah but how much flux would actually flow back through the hole in the first place? I'd guess less than 10% of the total forward-facing flux.

>>2424489
I guess so, but in the case of something so complex I'd rather see him using crossover circuits to combine an AC signal with DC power. I was talking about this:
https://www.maximintegrated.com/en/design/technical-documents/tutorials/1/1796.html
Don't gaslight me into thinking 1-wire uses 3 wires, obviously it just uses 2 wires.

[Anonymous No.2424599]

>>2424584
>halbach doubles power
>10% specs
k

[Anonymous No.2424601]

>>2424599
Halbach doesn't intrinsically double power, what matters is the geometry of the magnets and magnetic flux. Putting a tiny 0.1T magnet next to a 1T magnet to make it a halbach won't suddenly double your power density.

[Anonymous No.2424602]

>>2424584
>kjmagnetics.com lies
>kjmagnetics.com is incompetent
>dismissed
k

[Anonymous No.2424606]

>>2424601
>hallbach aint "double"
>I say different
k

then, its a halfarse usage only?
then complain it aint double....
k

[Anonymous No.2424612]

>>2424584
>Pic related. It's based off..
Yes indeed indeed,
"le same ol same ol".

[Anonymous No.2424651]

>>2424606
From Wikipedia, likely the source that you're getting "muh double" from:
>The field is twice as large on the side on which the flux is confined (in the idealized case).
>in the idealized case
The idealized case is approached fairly closely when the horizontal magnets and the vertical magnets are both very similar sizes. Even better would be to have 45° angled magnets between the vertical and horizontal magnets. The ideal case is only actually reached when the magnetic field continually rotates within the magnetic mass. When you have only some small magnets next to the larger ones, the small magnets' weak magnetic field strength isn't sufficient to fully divert all the magnetic flux away from the back side and towards the forwards side.

>>2424612
>but of course I know better than industry leaders
The burden of proof lies in the hands of the maker of extraordinary claims. Until you can actually prove to me that a radical change in geometry will lead to significant increases in power density without significant compromise, all I hear is "muh big motor" schizoposting and samefagging. As I have done for the last month. The only issue you're faced with is reality.

[Anonymous No.2424654]

>>2424651
>The idealized case is approached fairly closely when........
"no shit sherlock"
then use it that way, anon.
"its the point of using it"

[Anonymous No.2424656]

>>2424651
>As I have done for the last month.
Best leave you there in your own made-up adventure telling, anon.

Pull up ya big-boy pants n go get em, spazzbot!!

[Anonymous No.2424668]

What ohm potentiometer should I use for a headphone volume control? Google is returning conflicting results. Some say 10k is fine, while other say it needs to be more like 500 ohm if you don't want most of the range to be wasted.

[Anonymous No.2424674]

>>2424668
>range to be wasted.
use a logarithmic pot

[Anonymous No.2424676]

>>2424668
On the output of an amplifier? You can calculate what kind of dB curve you’ll get for a given potentiometer size by knowing the impedance of the headphones. That kind of logarithmic scale should tell you whether or not it’s a good fit. I think ideally you’ll see a relatively straight line in the log scale as the pot is rotated, and while this won’t be perfect due to the end point being literally negative infinite dB, you can modify this with a static resistor or two.

I’d suggest 1k would be sensible for 32 ohm headphones, but that’s just a guess.

[Anonymous No.2424677]

>>2424676
>On the output of an amplifier?
It's an inline volume control, mainly for plugging into my phone.

>I’d suggest 1k would be sensible for 32 ohm headphones
I'll mostly be using it with 16 ohm IEMs.

[Anonymous No.2424684]

>>2424676
>>2424677
Here’s some graphs, red is with 1k linear pot, blue with 100R, green with 16R. Y axis is in dBV. Basically you want to get it such that the curve is pretty linear, but the dropoff past x=0.1 is unnoticeable, so the loudness at x=0.1 must be sufficiently low that it’s inaudible. I’d suggest that 40dB isn’t enough for that, and considering that the 1k curve is already rather nonlinear, you’ll likely need either a log pot or a modified curve using extra resistors. Mess about with that mathematically, chances are there’s a decent result that can be arrived at without too much effort.

[Anonymous No.2424721]

what is the best way to clamp lipo battery to 3.3V output?
Basically i need to power some 3v3 stuff but i can't use a linear regulator since those usually drop 1V or more so when the lipo is at like 3.5V it would already crap out. I also cannot use ldo because i need 3A. Not to mention linear regs waste a lot of power.
Then there are bucks but those also usually need like 1V more on input than output and i tried some chinkshit ones i have and they shit themselves at about above 500ma current draw.
So wat do?

The most obviously solution would be to use 2S lipo with a bigger buck or linear regulator but i just don't like that idea much because i will be using twice as big battery with twice as big voltage when i need just 3V

[Anonymous No.2424726]

>>2424656
lol, it wasn't even a motor.
I kept telling him its not a farken motor, stupid!!
Fuck knows where he got motor from.
I'm sure he's a tard.

[Anonymous No.2424812]

>>2424721
You can use buck-boost converter which will work even if voltage is equal, below or above desired output voltage.
https://www.analog.com/en/product-category/internal-power-switch-buck-boost.html
I'm pretty sure there also exist some chink buck-boost modules, but performance will be a bit of a gamble.

[Anonymous No.2424825]

>>2424520
congrats you succumbed to peer pressure. there is no "right way". it is whatever is optimal for you. switching power supplies have been overhyped due their high efficiency. but they are noisy. what is more important to you?

[Anonymous No.2424831]

>>2424721
so you are basically saying, if it is 4.2v, give me 3.3v. but if it is 3.3v, still give me 3.3v. that is hard to achieve.

[Anonymous No.2424837]

>>2424721
Use two rechargeable NiMH/NiCad batteries in series and boost them to 3.3V.

[Anonymous No.2424842]

>>2424837
Add more in parallel for more current.

[Anonymous No.2424860]

>>2424837
Bucking is better than boosting imho
I would use 2s and a buck. Unlike ldo buck will waste very little power and the 2s bat will last same amount as 1s 2x capacity bat

[Anonymous No.2424864]

>replaced broken USB connector with new one
>fixed a broken trace

[Anonymous No.2424875]

>>2424860
Buck converters are definitely more efficient than boost converters, given the same source and load. Maybe Anon should use separate battery packs, one for the micro and the other for whatever load it's controlling.

[Anonymous No.2424878]

>>2424864
Did you use lead-free solder, Anon?
RIP, Paulie Walnuts.

[Anonymous No.2424879]

>>2424878
no

[Anonymous No.2424880]

>>2424879
Good. Does it work? Are the anchor points soldered well? Is it ewe-ess beecee?

[Anonymous No.2424883]

>>2424880
Yeah it does work, one of the anchor points is a lifted pad so I will have to be careful with this port

[Anonymous No.2424884]

>>2424883
Well it works so good job, Anon.

[Anonymous No.2424890]

Is this a good fume extractor?

[Anonymous No.2424899]

if I have a little toy music keyboard that uses rubber membrane switches, and I want to convert it to being digitally controlled with a microcontroller, how would that work? Just swap out the rubber membranes for transistors and turn them on to turn a key on?

[Anonymous No.2424900]

>>2424890
Fart into it and then run to the other side and take a whiff.

[Anonymous No.2424901]

>>2424900
It's fine I solder outside

[Anonymous No.2424903]

>>2424901
I solder in the shower. For safety.

[Anonymous No.2424907]

>>2424899
https://www.instructables.com/Add-MIDI-port-to-Keyboard/

[Anonymous No.2424908]

>>2424890
unironically much safer than the shitty "fume extractors" with filters that don't even work
I use pic related

[Anonymous No.2424935]

what happens if you keep turning this forever?
does it go back to 0?

[Anonymous No.2424942]

>>2424935
Tried it just now.
Turning it pass its limit clockwise results in a small click per turn.
Turning pass its limit counter-clockwise results in no click.

Both cases I can just keep turning.
Both cases the resistance between the wiper pin and one of the other pins either measures max (1k ohms for my pot) or zero.
And in both cases the pot survives and still works.

[Anonymous No.2424946]

>>2424942
thanks fren, that is what i needed to know
one more thing, how many full turns does it take to go from 0 to max?

[Anonymous No.2424950]

>>2424946
Uh, 10 turns? These thing are called 10-turn trimmers after all.

[Anonymous No.2424951]

>>2419660
the back of your existing wall warts will have:
1. Voltage rating
2. Current rating
bettter quality switching supplies will also have a power efficiency, Classes 1-6 the higher the numbr, the more efficient the supply. Just match what you have with what you want.

DIY CHAD MODE:
If you're sure your supplies are 5 and 12 volts respectively, calculate the total current draw for each voltage, add alittle extra, and replace all the warts with a single dual voltage supply like picrelated.

DIY CHAD MODE 2:
If you have some old high curent 12v adapters lying around, say for example, 12@3A, and you have two 12v@1A wall warts, you can power the two 1A devices from the single 12@3A supply. Same principal with the 5v devices. Many ways to deal with this, especially if you save old switching supplies.

[Anonymous No.2424953]

>>2424946
Pretty sure they’re 10-turn pots. Also they’re trimpots so they’re not rated for more that a few hundred motions.

[Anonymous No.2425016]

Are there any PCB CAD software that let me import an image or sketch so I can easily line up my components with the 3d printed housing?

[Anonymous No.2425086]

retard tier question. what is the S R Q thing in this block diagram?

[Anonymous No.2425113]

>>2425086
>S R Q thing

it's the simplest flip-flop what can be made.
https://en.wikipedia.org/wiki/Flip-flop_(electronics)

[Anonymous No.2425139]

>>2425086
An SR latch (SR stands for Set Reset), basically a single bit volatile memory cell. Send a pulse to S and Q goes high, send a pulse to R and Q goes low again. While S and Q are both low, the latch will retain its former state.

>>2425113
It’s not an SR flip-flop, it’s an SR latch. The difference is that a latch is purely asynchronous, while a flip-flop is a clock-triggered gated variant of a latch. Asynchronous gated latches also exist, for some reason.

[Anonymous No.2425151]

>>2424825
this is going to be used as a way to test battery operated shit before i buy it from goodwill/thrift store/etc. i do intend to put a couple of filter caps on the output to (hopefully) clean up the current a little bit, but i'm not all that concerned about noise.
>>2424549
they were $5 on ebay, that's all i care about. they can literally blow up after powering a small lantern for 5 seconds. so long as i get confirmation that my soldering and basic idea was solid, that's a win for me.

[Anonymous No.2425158]

r8 my first perf-board soldering job. it's 2 potentiometers to replace a trimpot

[Anonymous No.2425197]

>>2425151
I get you, but at least picking something with variable frequency gives you more choice when selecting inductors.

>>2425158
Looks fine to me. Post circuit!

[Anonymous No.2425202]

>>2425197
https://www.technotaj.com/2020/12/11/replace-trimpot-with-regular-potentiometer-on-buck-converter/

stolen from this kid

[Anonymous No.2425205]

So I got a little bit overzealous at work and volunteered myself for a project before figuring out all the details. How do I model an RC circuit that includes transistor (MOSFET, I guess) elements as an ODE so that I can simulate it numerically? I have no idea how to model the transistor part.

t. desperate statistician

[Anonymous No.2425206]

>>2425205
>an RC circuit that includes transistor
Take a look at the datasheet of a transistor. MOSFETs can be modelled somewhat simply, as the gate-source voltage increases then the source current increases linearly, until the source current is saturated (limited by something other than the transistor) at which point the effective resistance of the MOSFET starts decreasing as a function of gate-source voltage. The same applies to BJTs, except they're controlled by current not voltage, and their collector-emitter voltage decreases instead of their resistance.

You can find both of those relationships in a transistor's datasheet, and using them would make a half-decent representation of a transistor. If you want to look deeper, I'd recommend researching how spice simulations model transistors. Spice files are available for all sorts of transistors out there, so you could likely construct something where you can swap spice models to see how the transistor behaves. And also compare your sim to an actual LTspice sim.

[Anonymous No.2425209]

>>2425202
I prefer the symmetric version. At least I think the outer resistors are 10k, they may be larger.

[Anonymous No.2425213]

>>2425209
there aren't any resistors. you're just matching up the coarse pot with the trim pot's value, then adding a second fine pot that's a fraction of the original.

[Anonymous No.2425218]

>>2425213
I just prefer the symmetry, makes things arguably more linear.

[Anonymous No.2425239]

Confused about the voltage polarity over the inductor. When the switch is turned on, the voltage drop over the inductor is positive which agrees with the current flow from + -. But what about the back EMF due to the Lenz law: V = -LdI/dt. How does the negative assign agree with the voltage polarity across the inductor? Sounds like a contradiction. Or the back Emf is what opposes the current? Since the back Emf produces another current (self induction) in the opposite direction? Are there two Emfs involved or am I confused?

[Anonymous No.2425281]

>>2425239
Current wants to flow in the same direction, downwards. In order to force this current, the top of the inductor gets more and more negative to try to break down the switch and cause current to flow. Might be easier to look at a boost converter, which actually does have current keep flowing inside the inductor instead of hitting you with that “haha current = 0” condition. In the case of a boost converter, the voltage at the bottom of the inductor definitely gets higher than the voltage at the top of the inductor.

An EMF is just a voltage induced into a coil, usually in order to oppose some other change (changing current or changing magnetic field), like how it limits current as a motor gets up to speed.

[Anonymous No.2425298]

would this work?
two lipo batteries connected in series to double their voltage but then also each one is connected to its own tp4056 charging module and then they are both charged at the same time from a single 5v power source

[Anonymous No.2425301]

>>2425298
No. Because, for the TP4056 modules, the 0V line on the input is connected to the 0V line on the output. By connecting the right's B- to the left's B+ you're shorting the left module's pins together. To charge both lipos at the same time while keeping them connected in series to a load you'd need a DC isolator. But more sensible would be to use a boost converter + balancing BMS.

[Anonymous No.2425308]

>>2425301
so if i balance the cells beforehand and then charge them in paralel they should stay balanced correct?
if i use single tp4056 and boost converter, connected to the b+ and b- pins of the 4056, and then from the boost converter i connect to the two lipo cells which are in paralel, won¨t that fuck up the tp4056? will it still work correctly?

[Anonymous No.2425313]

>>2425308
>so if i balance the cells beforehand and then charge them in paralel they should stay balanced correct?
To run in series? They will initially, and if you always detach them to charge separately that should be fine. I'd consider putting an undervoltage limiting protection circuit on them, but IIRC those lipos come with them. Just it's kinda tedious to remove and replace batteries constantly, and use a resistor to balance them before connecting them together if you intend on charging them with the same TP4056.

A TP4056 can charge cells in parallel without issue (so long as they're balanced BEFORE you connect them together), but it can't charge them in series. It can only charge 1S arrangements, you know, 3.0-4.2V or so. DC input 5V max. You'd be better off buying a 2S charge IC/module.

[Anonymous No.2425316]

>>2425313
well i will be using the battery in a project where i want to have a simple usb c port to hook usb cable to and it charges both lipos, they will not be removable and i don't want to use an external charger since that is too bothersome, i want it to work with any 5v DC power source, so i can use generic phone wall warts too
is there something like the tp4056 module for 2s batteries?
fuck, i just want some very simple solution i can slap on my pcb. until now i only used 1s batteries so it was easy i juse soldered the tp4056 module directly onto the pcb and boom done

[Anonymous No.2425323]

>>2425316
wait i just got a big brain idea

5V --> Boost converter to 12V -> LM317 CC --> LM317 CV --> Two lipo cells in paralel

this way i don't even need tp TP4056 and i can charge both of the lipos in parelel
5V wall wart feeds into boost conv that pumps the voltage to 12V then the first 317 in CC mode ensure the max current is not above 1C of the two lipos, and the second 317 in CV mode ensude the voltage never goes above 8.4V. And this way i can even charge faster than 1A which is maximum tp4056 can do, with 317 i can charge up tp 1.8A.
As long i mage sure the two lipo cells are same voltage before hooking them in paralel to this it should work great.
This is so simple and smart i can't believe nobody has thought of doing this before.

[Anonymous No.2425327]

>>2425323
>Two lipo cells in paralel
Mate it's series that stacks voltage. Stop getting it confused with parallel, because it's confusing me. Also parallel has two "l"s in the middle.
>LM317
Literally can't even handle low enough dropout for your plan to work. Pic related.
>boost converter
You'd have a better time just shoving in an isolated 5V-to-5V switching converter so you can use two TP4056s. A CC/CV LM317 setup doesn't have any hysteresis, nor any other protection features. If you do use a simple CC/CV charger, do it with a less ancient regulator like an AMS1117-adj, and ensure there's battery protection on the cells. A dedicated 2S BMS may be better for balancing purposes than relying on individual protection circuits, but I'm no expert.

Also you can go for the smaller TP4056s, the ones without the battery protection circuits, because I'm 90% sure your lipos already have those.

[Anonymous No.2425328]

>>2425323
>>2425327
Oh and there are products like this:
https://www.aliexpress.com/item/1005004346215827.html

[Anonymous No.2425330]

>>2425327
yeah i meant in series, i'm just a broke ass hood rat with a soldering iron trying to make this shit work, these fancy ee terms confuse and anger me
>>2425328
shieeet that is pretty cheap and looks like what i need.. i'll order some i guess i can manually charge the batteries before those boards come since the shipping takes a month from chinks
looks like that thing only chargers lipos right? so it is up to me to make sure they don¨t get undervolted

[Anonymous No.2425332]

>>2425330
>looks like that thing only chargers lipos right?
Well both lipos and standard lithium ion cells. And you have lipos.
>so it is up to me to make sure they don¨t get undervolted
Well as I said, your lipos probably already have protection circuits inside them, so they wouldn't get undervolted.
May still need to put some thought into balancing, not sure if those protection circuits will allow the charger to balance cells or if you need a dedicated 2S balancing BMS.

>don¨t
What kind of keyboard are you using that you got an umlaut instead of an apostrophe?

[Anonymous No.2425338]

Time to design another PCB yay.
This will the biggest one i made so far. I sure as fuck don't look forward to soldering then millions fets

[Anonymous No.2425339]

>>2425332
>May still need to put some thought into balancing
meh, i'll just make it simple, i will hook the two cells + wires to the adc on my micro and it will be able to see that way if both cells have same voltage, and if not i will just manually re balance them once in a while

[Anonymous No.2425414]

>>2424935
It goes to eleven.

[Anonymous No.2425420]

>>2425281
Turns out I was overthinking it. I am talking about a simple circuit with LR in series. When the switch is closed and the inductor starts "charging", the back Emf =-LdI/dt. But the voltage drop across L is positive: LdI/dt. So it is all about treating the inductor as an active or a passive component. If it is a voltage drop, it is positive. If it is treated as a "battery" then the sign is negative and that makes sense since that voltage is subtracted from the source voltage.

[Anonymous No.2425444]

>>2419681
I'd try some rough grit sandpaper, you can usually extend the life quite a few times.

[Anonymous No.2425445]

>>2419783
The ideal schematic is as clean and well-organized as possible. Not having all the grounds explicitly connected is usually one of the best ways to keep it clean looking.

[Anonymous No.2425598]

>>2425445
Yeah but that one is a strange hybrid of both. The schematic posted in response does it properly by separating them all.

[Anonymous No.2425659]

Has an /ohm/ thread ever broken 500 posts?

[Anonymous No.2425742]

>>2425598
I wouldn't necessarily say that the most readable layout separates every single ground connection. I find the original more readable, given that the input side and output side of the IC are able to have all of the circuit paths followed without having to jump between disconnected lines, whereas the two sides are still separated down the middle since the only active point of connection between them is the IC itself.

[Anonymous No.2425743]

>>2425659
>500 posts?

nah, once bump limit hits, the /ohm/osexuals pull back so as not to waste their seed upon the ground.

[Anonymous No.2425746]

>>2425659
>500 posts?

nah, once bump limit hits, the /ohm/osexuals pull back so as not to waste their seed upon the ground.

>>2425742
>the original more readable

the inconsistency can cause confusion.
as evidenced by the fact that it caused confusion.

[Anonymous No.2425754]

>>2425746
>the /ohm/osexuals pull back so as not to waste their seed upon the ground
Somebody better call janny. I just made a mess.

[Anonymous No.2425799]

Use this pic for the next thread unless you have the gay, baker of /ohm/.

[Anonymous No.2425806]

>>2425799
Eh, it’s not all too clear that they’re LEDs. I’d still consider it if you can’t find a better clown picture.

Any other OP picture suggestions are welcome for my collection.

[Anonymous No.2425821]

>>2425806
What you're looking at is a light box covered with a sheet of plastic full of evenly spaced holes. On top of that is a sheet of black construction paper. Each "LED" is a tinted acrylic peg pushed through the construction paper which acts as a light pipe.
It's called LiteBrite. No LEDs present.

[Anonymous No.2425829]

>>2425754
Then Judah said to Anon, “Go in to your brother’s wife and perform the duty of a brother-in-law to her; raise up offspring for your brother.”

But since Anon knew that the offspring would not be his, he spilled his semen on the ground whenever he went in to his brother’s wife, so that he would not give offspring to his brother.

[Anonymous No.2425836]

>>2425829
Judah was a sneaky desert nigger.
>Just fuck your brother's wife, bro.
>Raise up some Sandsquatches for your bro, bro.

[Anonymous No.2425846]

>>2425821
I’m 95% sure I can see the parallax of the LED chips inside those dots. If it were cellophane then I’d expect to see less of a repetitive pattern to the insides of those dots, if any pattern at all.

[Anonymous No.2425850]

>>2425746
The original confusion stemmed from someone who clearly was inexperienced with reading schematics, no offense to them intended. There's no reason that they wouldn't have had the same confusion with both styles of schematic. Again, it's all subjective I'm sure.

[Anonymous No.2425851]

>>2425846
The original was a low power incandescent bulb.

[Anonymous No.2425863]

>>2425851
Oh so it's images of those coming through the pegs? Interesting.

Anyhow I'm making the OP with a non-clown pic now, with a bunch of JFETs that you can't actually buy anywhere.

[Anonymous No.2425866]

>>2425865
>>2425865
>>2425865
NEW THREAD

[Anonymous No.2425869]

>>2425863
>Oh so it's images of those coming through the pegs?
The facing end of the peg is convex and the stabby side is shaped like an elongated deck prism.

[Anonymous No.2425918]

>>2425158
looks like shit, start again

[Anonymous No.2425932]

Investigating alternate stator/rotor geometries.
This arrangement will work:
Passing a magnet perpendicular past a parallel wire...
Lots of series N-S induced voltages half the time,
lots of series S-N induced voltages the other half..

In direct comparison, this arrangement produces lower voltage + higher current.
Increasing number of magnet arc poles increases voltage at the expense of current.
Its already being surpassed by a superior alternate.

[Anonymous No.2426156]

Can anyone explain why transistors, when used in a self-oscillating circuit can handle much higher frequencies than when they're used for switching?

[Anonymous No.2426394]

>>2424417
when it comes to pedals the queen has to be good 'ld Fran Blanche, her guitar pedals are generally highly regarded, here is her YT:
https://www.youtube.com/channel/UCMLgHbpJ8qYqj3CkdbvC0Ww
her pedals:
http://www.frantone.com/
she has a shit ton of vids now but overall gr8 content, if u come from a rather fast language background I'd recommend watching at 1.5x speed cus she talks kinda slow.