/ohm/ - Electronics General: Solderstock Edition

[Anonymous No.2773862]

Thread cooked:>>2767459

>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://buster-spb.ru/files/SAFT/li-ion_user_manua.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

bake at page 8-10, post in old thread

[Bepis !n/ZnkB9jUw No.2773866]

>>2773862
Luv me the smell of burning skin

[Anonymous No.2773876]

how can I open and repair this piece of shit? lx-3000 headset is out of warranty, audio (and maybe mic) keeps failing
I've read that it's sonic welded
I have a soldering iron and I repaired stuff like mouses, regular earphones, but don't know much about /ohm/ stuff

[Anonymous No.2773888]

>>2773876
Hit it with a rock until it opens

[Anonymous No.2773891]

>>2773888
that could damage the rock

[Anonymous No.2773895]

>>2773876
>open and repair

easiest way to open it is to not open it
there's two more likely sources of potential intermittent breaks
that's where the wire gets bent the most: at the jack and at the speakers
so first thing is isolate the problem area
FIRMLY tape down the cable about 10 inches from both ends, and wiggle the extremities only
again, you need to tape it firmly so that the wiggling doesnt affect the middle of the cable
if that didnt do the trick, then try other short sections

if you must open it, see if you can find a seam, and use an acrylic scratcher and ruler.

[Anonymous No.2773928]

>>2773866
What's does a pack of smokes cost these days?

[Bepis !n/ZnkB9jUw No.2773935]

>>2773928
Can’t put a price tag on lookin cool

[Anonymous No.2773962]

I dont know much about electronics but I need some advice for powering a project. Is it fine to have a li-ion(3.7) attached to a TP4056(5V 1A) charge board, with the power wire then split in two - one side going to power a microcontroller - the otherside to a positive voltage regulator like the L7812 (12V) to power a 12V motor (Below 1A). Or should I be going about this some other way?

[Anonymous No.2773964]

>>2773962
edit - not a L7812 but a buck boost converter to get the voltage up to 12v

[Anonymous No.2773970]

>>2773962
>>2773964
If you draw power from a source and do not exceed the recommended rates it should be ok. A buck boost converter should be seen as a load in your example. If it does not draw excessive current then it should be ok. The fact that the buck boost delivers more volts is irrelevant to your question; it's simply what it does with energy it draws from the battery. All the battery cares about is how much current is being drawn.

[Anonymous No.2773971]

>>2773962
>one side going to power a microcontroller
Should be fine if your micro runs on 3.3V and has an integrated regulator.
If it runs on 5V you need a boost converter.

[Anonymous No.2773985]

>>2773962
12V battery pack straight to the motors. Buck the same 12V down to 5V and feed it to your TP4056 and uC.

[Anonymous No.2773991]

>>2773862
>tfw some hairy lesbians ask you to model for their "women in stem" campaign
>tfw you need the money
>tfw you don't want to touch that gross handle (eww I just had my nails done)
>tfw you're a meme forever

[Anonymous No.2774002]

>>2773991
>3 months later...
>President of Hewlett-Packard
>$50M yearly bonus
>still doesn't know anything about technology
>cuts worker pay, doubles ink subscription price
>EQUAL PAY FOR EQUAL PUSSYHATS
>company crumbles
>blames Russia and/or NAZIS!!!!!!!!!!!!!!!
>???
>bankruptcy

[Anonymous No.2774006]

I'm having a bit of a conundrum. I have these 20kv rated diodes and was doing some experimenting with a hv circuit. I had a 10kv transformer with one 20kv diode as a half bridge rectifier to an air gap to strike an arc. I then took another 20kv diode and shorted it across the gap with a screwdriver, but weirdly enough an arc actually jumped to the diode even though it was in the reverse direction. On a dc supply the forward voltage drop seemed fine at 14v. Is there some sort of high voltage interaction with diodes I'm not aware of?

[Anonymous No.2774007]

God I am so sick of voltage regulators always being so off but still within the spec. I don't want a 3.3V reg to be 3.27V ffs. I want it to be 3.3v. Is that too much to ask??? I am going to build my own I swear. Even a 16 bit DAC should be do the trick.

[Anonymous No.2774009]

>>2773962
Usually you want the highest power draining element in a circuit to run directly off the battery voltage. So that would mean using a 3S lithium battery with BMS and charge regulator, and a step-down converter to power the micro. Or something similar, maybe you’d rather use an existing deep-cycle battery of some sort.

>>2773971
No common microcontrollers have that kind of ultra-low-dropout regulator (e.g. HT7833), but some can run on a wide enough range that they do not need a regulator, which is probably what I’d use. Just keep in mind that the clock frequency at lower voltages may need to be slower, and ADC measurements should use sufficiently low fixed voltage reference.

>>2774006
If it’s high-frequency those diodes may have a non-trivial capacitance, allowing current to flow. There might also be leakage via skin secretions or other gunk on their surface.

[Anonymous No.2774010]

>>2774009
>some can run on a wide enough range that they do not need a regulator, which is probably what I’d use
I expect micros to be fine with undervoltage, but a 3.3V micro handling 4.2V? Sounds a bit risky.
The max charging voltage of those tp4056s is set internally so it's not like you can change a resistor on them and limit the max voltage to something lower.

[Anonymous No.2774016]

>>2773970
>>2773971
>>2773985
>>2774009
>>2774010
Thanks for the replies. It's a handheld device low on space which is why I wanted to go for a single 18650 rather than a 12v battery pack. The microcontroller im using is a RP2040 board with an input v range of 1.8-5.5v.

[Anonymous No.2774024]

>>2774016
Use pic related to charge your battery and boost it to 12V. Buck that down to 5V for your microcontroller (a regulator will be much less efficient and generate heat). You can use a switch or PFET to keep the battery from draining while the device is off.

[Anonymous No.2774025]

>>2774024
>Buck that down to 5V
You could just tap the battery before the boost converter for 4.2V (full charge) to 3.0V (safe discharge limit). Then you won't need another converter or regulator.

[Anonymous No.2774032]

I'll try using graphite felt as desoldering wick tomorrow. Any bets against it becoming a game changer for the purpose?

[Anonymous No.2774036]

>>2774032
Graphite is conductive. I imagine a bunch of tiny threads getting snagged and shorting out random shit all over the board.

[Anonymous No.2774046]

I need help identifying this transformer, I thought it was a center tapped one but why is there 1Ohm between pin 2 and 5? How would a symbol for it look like? Or is it just shorted?

[Anonymous No.2774049]

>>2774036
>Graphite is conductive
Less so than the copper strands in conventional and expensive desoldering wick. I'll just cut or pull a tiny pinch from the carbon welding mat, brush on a little liquid flux and hopefully get much success.
Hopefully it will turn out reusable if it doesn't "wet" like metals. Unless it's totally insulating like the welding mat is intended to be.

[Anonymous No.2774078]

>>2774010
You misread me. Most ATmega can run from 1.8V to 5.5V. So can the CH32V003 if I recall.

>>2774032
Desoldering wick needs to wet to solder to work, otherwise the solder will just stay on the board. Graphite does not wet to solder. Would be like trying to use aluminium.

>>2774046
Chances are one of those 360R coils doesn’t exist. If you delete L2, measuring from J1 to J5 will get 360, and J1 to J2 will give 361. If both did exist, you’d be measuring 360 in parallel with 361, resulting in a resistance of 180.5, so either way your diagram isn’t accurate. There could be a short, or there could be one big and one small winding, hard to say. I’d either use a milliohm meter to map the resistances more accurately, or feed some AC into a coil and see what voltages you measure. I doubt it actually has three windings in a triangle like that.

[Anonymous No.2774085]

>>2774078
>Desoldering wick needs to wet to solder to work
Capillary action and other such properties of fluid dynamics has nothing to do with wetting which is the mixing of materials in the surface layers to create a strong bond.

[Anonymous No.2774135]

>>2774007
It never matters. If it does, you should be using a precision voltage reference instead. Some voltage references like the REF3333 can be used as voltage regulators for low power analogue ICs. They can also push and pull current IIRC.
Your post number is a 1000V diode.

>>2774046
Since L4 and L3 are both really small, I'd guess you're just not able to measure the precise difference between the winding resistance. Build a current source from an LM317 and a 9V battery, and use it as a milliohm meter.
Your post number is that of the CD4046 based PLL IC, even better.

>>2774085
Idk, ever tried getting solder to stick to fine stranded aluminium wire? It just beads right off, even with plenty of flux. Can't see why graphite would be any different.
4085, what the hell is an "and or invert" gate?

[Anonymous No.2774145]

>>2773935
>cool
only retards and lowlifers smoke these days
which one of those are you?

[Anonymous No.2774220]

One of my pc speakers sometimes cuts out and the only way to fix it seems to be to "flick" the speaker cone. What's weird is that this problem has persisted even after I bought new speakers, a new sound card, and new speaker wire. The only constant has been the subwoofer and its integrated amplifier. I'm assuming there's a faulty component in the amp and sending it a voltage spike through the speaker cone is temporarily defibrillating it back to life. Anyone know if my theory makes sense and if it'd be fixable with a soldering iron and a trip to Aliexpress?

[Anonymous No.2774221]

>>2773935
what's looking cool got to do with smoking

[Anonymous No.2774262]

>>2774220
>a trip

you're obviously hallucinating
take your medications
or take your spooky story to /x/

[Anonymous No.2774265]

mcg is dead so i post here
im a noob
i have a 18v li ion battery connected to a buck converter. converter shits out 12v. That 12v is to be fed into an arduino nano.
Can i use the IO ports of the nano as sink for 12v signals? Because ideally i want the nano to switch a 12v led and a 12v automotive relay without ordering additional transistors

[Anonymous No.2774269]

>>2774265
>sink for 12v signals?

see those internal protection diodes?
if you apply a voltage greater than Vcc+0.5V one of them will burn
if you apply a neg voltage, -0.5V or more, the other will burn.
if you kill a nano, god will make you burn.

[Anonymous No.2774270]

>>2774265
No you can't. The ports are not open collector outputs.

[Anonymous No.2774278]

>>2774270
Adding to this: just use a 5V-gate mosfet to make an open drain output. It's just one extra component.

[Anonymous No.2774279]

>>2774269
>>2774270
well, off to the parts drawer

[Anonymous No.2774310]

>>2773962
Another question regarding this. After the TP4056 and 12v buck boost its recommended to go to the 12v motor first? So I should go from the motor to a 5v buck in series, rather than splitting directly from the 12v buck boost and having the microcontroller and motor in series?

[Anonymous No.2774312]

>>2774310
the last word should be parallel not series*

[Anonymous No.2774324]

>>2774262
Couldn't make the shit up man. And I obviously meant a figurative trip. It's a website.
https://streamable.com/xic1v2

[Anonymous No.2774327]

>>2774310
The 12V should only go to the motor, which also needs a flyback diode in anti-parallel (to the motor). The 5V for the micro should come from either the TP4056 battery output (4.2V) or the 5V PSU that powers it and the boost converter. You shouldn't charge the battery and power the circuit simultaneously, so switch between the TP4056 and 5V PSU using whatever method suits you.

[Anonymous No.2774329]

>>2774220
It may be a dodgy crossover cap, or the series capacitor of a single-ended single-supply amplifier. Take it apart and see if any caps look suspicious, measure them with a DMM if possible.

>>2774265
Do you need to measure this voltage with an ADC, or just tell when it's present? Using a FET to make an open-drain is a decent way to go since you can use an internal pullup resistor, though I'd note that if the input will be left disconnected from anything you may also need a pulldown resistor on the mosfet's gate. a 2n7000 is what i use for this, if not a bss138.

You can technically just use a single series resistor and rely on these diodes >>2774269 to save your chip from more than 5.7v, but it's not recommended at all.

>>2774310
What microcontroller?

[Anonymous No.2774335]

>>2774324
Voice coil is deformed so it gets stuck
OR
The coil connections inside the speaker housing aren't great.
That's all I've got.

[Anonymous No.2774339]

>>2774335
That was my though but if that were the case it wouldn't persist if I bought new speakers. And the symptom would move to the other side if I swapped the speakers left to right. But neither of those happened.

[Anonymous No.2774340]

>>2774339
Did you try another audio source? Swap the cables (speaker wire and 3.5mm input etc.)? Maybe your amp is the problem.

[Anonymous No.2774355]

>>2774007
You can adjust them slightly by offsetting their ground pin (usually just tied to ground) with a voltage divider and/or trimpot and/or zener diode and/or old green LED.

Want to know what’s fun? Our current sense shunts are just bare copper wire but over-valued, so we sit there with a soldering iron and dab on blobs of solder until it comes into tolerance. If you go too far, you just take a knife and carve some off.

[Anonymous No.2774362]

The thread counter no longer show the number of posters itt. Did some political niggers pay gookmoot to obscure their gaming of this prestigious underwater welding board again?

[Anonymous No.2774374]

>>2774327
its a small handheld project. Powered with just one 18650. I was advised to go 12v first so I ordered an all in one TP4056 and 12v buck board. I just need to wire it together now, should I split off the TP4056 straight away and have one wire go to my motor and another to my microcontroller, or have it run in series with the motor wire then going to the 5v buck into my mc?
>>2774329
rp2040 board, -1.8v to 5.5v input

[Anonymous No.2774384]

>>2774374
Split the TP4056 output to:
1. the microcontroller
2. the boost converter
FYI, a buck converter drops voltage and a boost converter increases voltage. You want to boost the output from the TP4056 to 12V and send that to your motor. The only thing connected to the boost converter output should be the motor.
The micro runs straight off of the battery (TP4056) output. No buck or boost necessary.

[Anonymous No.2774399]

>>2774374
Assuming the TP4056 board also has a protection chip on it (e.g. DW01), there should be a place to connect the RP2040's power input after that but before the 12V boost converter. If it doesn't have a protection chip (e.g. pic related) and rather goes straight to the booster, then you'd connect the RP2040 directly to the B+ and B-.

The boost converter itself may have a minimum shutoff voltage between 2.5V and 3.2V, but even if it does your takeoff to the MCU won't. So I'd either write a low-power power-checking routine on the MCU to read the battery voltage via ADC and prevent further battery drain, or by including a conventional protection circuit. The former method is more flexible for thresholds and such, but the standby power draw of the boost converter may be too high without fucking about with it to add a mosfet between the battery and the boost converter. Well, that might not be an issue depending on the boost converter IC.

It would even be an option to use the RP2040 as a battery charge controller and/or as a boost control IC, though I probably wouldn't recommend it. At least it would make the charging and discharging voltage thresholds easily programmable.

[Anonymous No.2774401]

>>2774340
>Did you try another audio source?
Yeah, same issue.
>Swap the cables (speaker wire and 3.5mm input etc.)?
Yeah, issue persisted.
>Maybe your amp is the problem.
Yeah that's what I suspect, but I wanna know if it's possible to identify which component is giving out, and if so, how

[Anonymous No.2774403]

I have burnt through 103 mosfets or 16 boards.
All because I have no idea how to drive mosfets in parallel.
And it is getting to a point where even I get tired of it.

Has one of you an answer to the FET-Q and how to drive them?

Should I use a FET driver to drive a smaller FET that in turn drives these massive FETs?

I have six MOSFETS in parallel and I don't want to use one FET driver for each.

[Anonymous No.2774408]

>>2774403
Highly depends on what you're doing with those.
General switching?
High frequency stuff?
High side or low side?

[Anonymous No.2774409]

>>2774403
Post your gate drive schematic.

Also state your circuit's switching frequency, switching current, voltage rails, and describe the load. Additionally, do you have an oscilloscope, or alternatively a multimeter that can measure sufficiently fast transients in min/max mode?

[Anonymous No.2774410]

>>2774408
I am using them to switch a coil ON/OFF for my coilgun.
The FETs should be able to switch as fast as possible.

[Anonymous No.2774412]

>>2774399
I forgot Anon is using a J5019.
>>2774374
Connect the battery and the RP2040 to BAT+ & BAT -. Connect the motor to the boost output.
>>2774401
>identify which component is giving out
I assume that it's in the amplified output stage (MOSFETs or BJTs). If you swap speakers (and cables) and the issue persists on the same channel, it's the amp. You'll need to remove the transistor(s) to test them.
>>2774403
You need to mitigate the inrush current to the gates of all the transistors with appropriately sized resistors.

[Anonymous No.2774413]

>>2774410
So basically general switching, likely low side.
No need for gate drive ICs, fukken totem pole that shit with a pair of BJTs.
The other thing of note is the gate resistor value, it needs to be high enough so you don't fry the gates with too much current.

[Anonymous No.2774418]

>>2774410
What kind of energy cycle do you have? Are you trying to harvest the extra energy out of the coil and back into your caps (requires an H-bridge)? Dissipating it in a snubber network? Simply clamping its decay with a diode? Or ignoring it and assume you're not getting 500V spikes when you turn off the FET? I don't see any diodes on your driver board, but maybe they're just right next to the inductor. Assuming you are getting rid of most of the energy in a sensible method, the other thing to do is add TVS diodes to protect the FETs.

Have you considered using SCRs?

>>2774412
I don't think gate drive current is usually an issue. If they actually sell 10A gate drivers, I doubt any old BJT circuit he cooked up is going to be putting too much current into the gate. I often see ~10Ω resistors with diodes in parallel with them. More likely he isn't sinking current out of them fast enough to turn them off without blowing them from switch-off heating.
Well more likely than that is voltage spikes frying them.

[Anonymous No.2774422]

>>2774408
>>2774409
>>2774412
>>2774413
I am driving a coil with 1.4 Ohm resistance at 30V.

This is my current circuit. I am currently using 470ohm resistors between the driver and the MOSFET gates to limit the inrush current.
I am already planning on adding zener diodes and pull-down resistors to every MOSFET between gate and source.

Switching frequency: Irregular but 1000HZ should be good enough.
Switching current: 1000A max
Load: Coil

I have an oscilloscope that I intent on using to find out how long it takes for the MOSFETS to switch on and off.

[Anonymous No.2774427]

>>2774418
I am using freeweheeling diodes to protect the FETs from flyback voltage spikes.

[Anonymous No.2774429]

>>2774412
>You'll need to remove the transistor(s) to test them
At that point I reckon it'd probably be easiest to just replace them without checking if they're bad, especially since the issue is intermittent.

[Anonymous No.2774443]

Do you really need a tip thermometer? Getting solder hot enough so that it wicks is fine enough, right?

[Anonymous No.2774444]

>>2774422
What FETs are you using? Rds(on)?
>>2774429
They might still be good though. Won't hurt to test them, and it only takes a second.

[Anonymous No.2774445]

>>2774443
It's for calibrating your temp controlled iron so you aren't 100 degrees over or under temp.

[Anonymous No.2774450]

>>2774444
Quality digits. My thinking is in order to test them I have to desolder them, and if they're already out then it's zero additional effort to put new ones in rather than reinstalling the old ones.

[Anonymous No.2774462]

>>2774450
That's perfectly fine, but remember that I'm only guessing at the problem. Take a good look at both sides of the board before you make a decision so you aren't wasting your time.

[Anonymous No.2774465]

>>2774009
>If it’s high-frequency those diodes may have a non-trivial capacitance, allowing current to flow
I looked up the specs for the transformer and the output frequency is much higher than I anticipated at 20-40 KHz. Either my diodes are capacitatively coupling or the frequency exceeds my rectifier's switching speed. What would be the solution? Would faster switching diodes for the rectifier solve this?

[Anonymous No.2774469]

>>2774422
>470Ω
Kinda high, that thing outputs 6A so you should split it up into 1A each. That works out to ~10-18Ω per FET. You can probably go lower than that, since the MIC4429's datasheet says its current is limited by its internal resistance, so it's probably safe to go as low as 1Ω. I'd still use resistors to keep the current spread-out between resistors. Naturally resistance in the connectors and wires is going to be significant, even inductance. When charging the gates up it doesn't really matter, so long as the time it takes to fill up the gate charge is significantly shorter than the time it takes for current to get significant in the inductor (V = L*dI/dt, ∆t = L*∆I/Vcc). But discharging speed matters a lot. You're trying to turn the FET off while current is at its maximum, so avoiding switching losses is key.

More importantly, you want your gate drive loop area small. It's entirely possible to give your FETs deadly voltage spikes thanks to the inductance of the wiring going to their gate. You should put that gate driver on the same board as the FET board, or in the name of modularity, on a socketed daughter-board.

Also your diodes are backwards, lmao.

[Anonymous No.2774471]

>>2774443
> need a tip thermometer
No, as long as the solder wicks you’re fine.
You don’t want to get too hot though, or you’ll get excessive oxidization, solder breakdown, and flux burn.
It’s also a function of time, so don’t dwell on the joint for too long.

[Anonymous No.2774472]

>>2774465
You don't usually get high-frequency diodes that can handle those voltages. Check their datasheet, check datasheets on digi-key, see what you can find I guess.

[Anonymous No.2774474]

>>2774427
Also make sure those diodes didn’t sacrifice themselves on the first cycle. Mosfets are dead by the second cycle. :-)

Is there any way to slowly ramp-up the voltage, frequency, and current to see what the problem might be, and where it occurrs?

[Anonymous No.2774493]

>>2774422
>>2774469
>diodes are backwards
oof

[Anonymous No.2774504]

>>2774493

that's not how you wire a snubber diode
it goes in anti-parallel to the load
pls do a proper schematic that includes the load
so we can laugh at you even harder

[Anonymous No.2774510]

What is more important for precision opamps, low Ibias or low Voffset? For example if I want to measure tiny currents.

[Anonymous No.2774515]

>>2773928
>buying packs of smokes and not packs of loose tobacco

It's like you hate high quality tobacco and money or something

[Anonymous No.2774516]

>>2774504
> proper schematic

[Anonymous No.2774518]

>>2774515
I quit smoking 15 years ago, after 20 years of 1.5-2 packs per day.

[Anonymous No.2774534]

>>2774510
Bias current should be significantly less than the current you're trying to measure, and voltage offset should be significantly less than the voltage drop across the sense resistor. Make of that what you will. You're often better off buying monolithic current sense amplifiers or instrumentation amplifiers, in this case maybe a monolithic coulomb counter.

[Anonymous No.2774563]

Wow the bitmap importing feature is actually really handy.

[Anonymous No.2774578]

>>2774563
> silks component values when it’s already on the component
> power but no + or - or what the voltage is, nor any test voltages/points anywhere

Even Zhang puts those on an average TV board

[Anonymous No.2774585]

Literally every single fucking soldering iron tip I used ended up like this after 10 minutes of use. I don't use corrosive flux but the one available near me that everyone uses. I don't scrape it against hard surfaces and only use sponge. I use solder that everyone else sells and buys. Why the hell does this only apparently happen with my irons, cheap or expensive while I do exactly like I'm supposed to?

[Anonymous No.2774586]

>>2774578
Still have to rearrange some parts, then I'll melt the angular traces into curvy traces, I'll look the silkscreen after all that. There's definitely a need for labelling nodes AND wire colours, and now I see I should have plenty of room for nice big labels. Hope JLC will do soldermask that isn't green for the same price some time this decade.
Grey lettering is front-fab, not front-silk. Front silk is cyan, back silk is magenta. The fella in the corner is a corruption of the logo of Actuonix, picrel. They sell actuators and such.

Wonder if there's a downside to adding ground and power fills? It's just a relay board for reversing a motor from button pushes.

[Anonymous No.2774588]

>>2774585
some things that might make soldering iron tip platings last shorter
>lead free solder
>being at high temperatures for long times
>thermal shock
>corrosive flux
>not tinning your iron before you let it cool down

some things that might make soldering iron tips last longer:
>tips designed to use with lead free solder
>temperature regulation to prevent overshoot (lol, lmao)
>brass wool instead of wet sponge (idk maybe)
>rosin flux with minimal or no activation
>a brain

[Anonymous No.2774589]

>>2774588
I have all tin solder which is the only one available and I'm not using corrosive flux. The same things everyone else does around me as is available

[Anonymous No.2774594]

>>2774585
That's shit tips for you, my dude.
You're not doing anything wrong aside from not having a source of good tips.
I have the same iron as in your pic and I can tell you that the tip it comes with as well as the replacements are absolute dogshit.
If there's anything I've noticed about good tips it's that usually they have 2 platings, the standard nickel stuff that coats it all the way around as well as some other plating that's only on the tip.

[Anonymous No.2774597]

>>2774585
>Why the hell

your iron is too hot. prob 60W. you can fix it using a light dimmer.
your tips are garbage tier. only use brand name ones, Weller or Hakko. if they dont fit, file 'em down.

>everyone else sells and buys

very odd thing to say. doesnt matter what your furry friends buy.
all your supplies should come from amazon, and be brand name.
solder: Kester 44 rosin core, 63/37 lead/tin
flux: Kester or MG Chemicals 8341
flux is actually optional. the flux inside the solder is good enough 97.5% of the time

[Anonymous No.2774603]

>>2774594
Does dremel versatip suck as far as tips go? I didn't use it for soldering for this very reason of the risk of screwing up the tips. I don't even do a lot of electronics work, just need to stick some wires together and to pins in a connector

[Anonymous No.2774607]

>>2774516
In my defense, the TC4429 has two output pins and I thought it sensible to connect them for maximum current.

[Anonymous No.2774610]

>>2774493
>>2774504

But I thought that when the magnetic field collapses in a coil it produces a reverse voltage spike which would be blocked by the diodes.

[Anonymous No.2774617]

>>2774610
lol there you have your problem. They should short a spike.

[Anonymous No.2774624]

>>2774610
>which would be blocked by the diodes

nope.
the flyback diodes arent there to block anything, they're supposed to eat the spike.
i.e. short it out of existence.

[Anonymous No.2774626]

>>2774469
Thank you for the helpful advice. I will do as you said and make the leads from the driver to the gates as short as possible by stacking the driver PCB to the mosfet PCB.
I rechecked and the resistors in front of the gates are actually 100ohm. But yeah I will reduce them to 13 ohm as I am powering the gates with 12V and that gives me 0.923A per gate.

[Anonymous No.2774629]

>>2774624
>>2774617
damn alright I will move the diodes in anti parallel to the coil then. Thanks for the help

[Anonymous No.2774659]

Is this a good solder for wires and pins? I've heard 100% tin takes more heat to melt so it's probably bad for small components

[Anonymous No.2774660]

>>2774659
I recommend having a fren outside of your country buy several spools of 60/40, relabel them as lead-free, and ship them to you.

[Anonymous No.2774676]

I'm trying to fix an old crt monitor and I need to measure some signals in the horizontal deflection circuit but for some reason I'm getting this on every pin on the deflection processor IC. Does anyone know what this is and how I can get an actual measurement of the signal I need? I'm pretty much a beginner when it comes to using a scope.

[Anonymous No.2774677]

>>2774660
Is that for the freedom country? Even in this eurofaggotry we can just order leaded ones

[Anonymous No.2774686]

>>2774676
>how I can get an actual measurement of the signal I need?

you have to connect the scope ground to the chip's ground.
but before you do that, use DMM to make sure there's no voltage difference between scope ground and the chip's ground.
AC or DC.
because the monitor use a VGA port, it should be safe to connect grounds together.
after all, computers are grounded.
but it's a complex subject, and i'm only covering it superficially.
proceed at your own risk.

[Anonymous No.2774691]

>>2774686
I'm connecting the ground to this metal shield here. Does it matter where exactly I'm connecting the probe to when it's all the same ground?

[Anonymous No.2774692]

>>2774677
63/37 is a better blend but not quite common

[Anonymous No.2774693]

>>2774691

there may be several diff grounds on one circuit.
if you can google the datasheet for the chip, it should tell you the ground pin
might be called something weird, like Vss
then you can confirm if it's the same as the shield

just in case, wear eye protection, like in the pic

[Anonymous No.2774694]

>>2774677
>freedom country
Only freedom in this country is for criminals. Everyone else is a slave. That's what happens when everyone pretends that niggers are human and should hold elected office.

[Anonymous No.2774695]

>>2774693
I have the schematics for the monitor and the data sheet for the chip and it seems like it's all one ground. I tested with a multimeter and all 3 ground pins on the chip have continuity with that metal shield.

>https://www.alldatasheet.com/datasheet-pdf/pdf/25272/STMICROELECTRONICS/TDA9103.html

[Anonymous No.2774697]

12V 2A power supply --> L7805 Vregulator 5V 1.2A --> 5V 2A solenoid.

This doesnt work for me. The voltage just drops to 0 and the solenoid doesnt actuate. Is this what happens (im a noob) I was hoping it would still actuate but just not as powerfully.

[Anonymous No.2774698]

>>2774697
Is it an AC solenoid?

[Anonymous No.2774702]

>>2774698
no all DC. I got it working with a 100uF cap over the 12v + and -. Can anyone tell me why this works lol

[Anonymous No.2774703]

>>2774702

regulator manufacturers insist all regulators should have caps both in and out
maybe they werent lying all this time like we thought

[Anonymous No.2774704]

>>2774703
example

[Anonymous No.2774706]

>>2774704
Anon needs a flyback diode too.

[Anonymous No.2774710]

>>2774704
>>2774706
second cap and flyback diode added (hopefully in the right place)

[Anonymous No.2774712]

>>2774710

[Anonymous No.2774714]

>>2774710
Excuse me, sir. That little guy is called Pepo. Picrel is Pepe.

[Anonymous No.2774715]

>>2774714
*Peepo
Picrel is Apu Apustaja.

[Anonymous No.2774716]

>>2774712

looks good
you wont be giving Mr 103-Dead-Fets a run for his money

[Anonymous No.2774717]

>>2774716
>Mr 103-Dead-Fets
lmao

[Anonymous No.2774720]

>>2774659
Yeah, I use that all the time, pretty damn good.
A bit tricky to do solder bridges with because it's got some really good flux in it, but for that I have fluxless solder.
t. fellow hornbach enjoyer

[Anonymous No.2774728]

>>2774704
>>2774706
>>2774712
>>2774714
we now have a mosfet controlling the solenoid and its working well

[Anonymous No.2774731]

>>2774728
Good job, Anon. On to the next project.

[Anonymous No.2774735]

>>2774731
ohh dear the fcc isn't gonna like this

[Anonymous No.2774770]

>>2774586
Get your free copper.
If you’re etching it at home, preserve your etchant.
Ground surround shielding, moar current carrying capacity, etc.

[Anonymous No.2774775]

>>2774585
All flux is corrosive, rosin is just corrosive at higher temperature.
You could use plumber’s flux to clean your tip (e.g. zinc chloride) but make sure you remove all the flux when you’re done. And like, technically you could use it to do your soldering, too, but again you don’t want ant lying around or it’s going to look like an alkaline battery leaked onto your PCB.
I clean up rosin solder residue with alcohol, too… it’s probably a good idea, in general.

[Anonymous No.2774779]

>>2774607
I suspect that is more for heat sinking.
In which case, you don’t really want a socket either.
And change your PCB under those pins to have more copper.
You can also get/make little heatsinks which may help. Get the black anodized ones though. They can come with their own adhesive, or you can use heat conductive epoxy.
I heard there is a middle ground recently—a kind of hear sink compound that is tacky enough to work as a permanent connection, but still removable. That would be ideal.

[Anonymous No.2774829]

>>2774589
>all tin solder
Yeah that's gonna be your problem. Get better tips or change to leaded solder. Electroplate your tips if you're feeling funky.

>>2774610
The diodes aren't meant to conduct while the transistors are on. They're meant to conduct when the transistors turn off. But having antiparallel diodes across your coil directly will cause the magnetic field to collapse slowly. In the case this is too slow for your coilgun, you may need to add a resistor in series. Maybe an entire snubber network would be a good idea. You'd design such a network in order that the peak negative voltage never goes above the rating of the transistors. If that's 50V, then the coil field will decay a lot faster than clamping it to only 0.7V through the diode.

>>2774659
Never seen solder with nickel in it before. Maybe it's better for tips than the tin+copper stuff, or stronger, or whatever. I had a roll tin+bismuth solder once, flows really nicely like leaded stuff, melts at an even lower temperature, but isn't as strong. It's nice to have around for thermally delicate parts.

>>2774691
>>2774695
Is that shielding tied to the ground pin of the wall plug?

>>2774697
For a dumb load like a solenoid, why not just use a series resistor instead of a voltage regulator? 3.5Ω or thereabouts, 14W if it's on continuously but less if it's just being pulsed. A low current regulator and an output cap works too, I guess.

>>2774731
>billions
check those clearances bro

[Anonymous No.2774831]

>>2774829
>Is that shielding tied to the ground pin of the wall plug?
Yes. does that matter?

[Anonymous No.2774832]

>>2774829
>check those clearances bro
lol
https://www.youtube.com/watch?v=_7pWGYcRvj4

[Anonymous No.2774836]

>>2774831
No that's a good thing, assuming there isn't substantial voltage drop between the ground pins of each plug. Put the scope and device under test on the same power strip to see if that makes a difference, or tape off the ground pin of the scope. It's probably fine. What happens if you probe a ground trace instead of a signal? Or just clip the probe onto the shielding? Both with or without the ground clip of the scope.

>>2774832
based albania

[Anonymous No.2774848]

Let us say that I know zero about electronics. Where do I start?

https://imgur.com/gallery/y3N7qGC

[Anonymous No.2774849]

>>2774728
oops. too late i guess. here's a similar circuit that uses ir to turn it on/off

https://youtu.be/ZSJMUgLWbH0

[Anonymous No.2774864]

>>2774848
Did you read the OP?

[Anonymous No.2774865]

>>2774864
Nope. Thanks! Reading...

[Anonymous No.2774874]

RANDOM NOSTALGIA POST
>RANDOM NOSTALGIA POST
RANDOM NOSTALGIA POST
>RANDOM NOSTALGIA POST
https://www.youtube.com/watch?v=agS6ZXBrcng
https://www.youtube.com/watch?v=D_1koH1cKG0
NEVER FORGET WHAT THEY TOOK FROM YOU

[Anonymous No.2774886]

>>2774836
I tried probing a ground trace on the pcb with the ground clip on the metal shield and I got the exact same signal but inverted. I don't really understand why, but I don't know all that much about electronics yet so maybe it's something obvious and I'm just dumb.

But either way, I was able to get it working by clipping the ground connector to one of the ground pins on another IC closer to this chip and I'm not seeing that signal anymore.

[Anonymous No.2774911]

>>2774717
>gif
I do this to boost performance.

[Anonymous No.2774958]

>>2774597
Flux in solder goes bad with age. Good idea to have it around just in case or if your a cheap fuck and buy old solder.

[Anonymous No.2774963]

>>2774874
Why would you do this?

[Anonymous No.2774966]

Ok curvy traces is one thing. But how about having the traces vary in width as they have more or less room? I think it would look disgusting, but kinda funny at the same time.

[Anonymous No.2774967]

>>2774585
>only use sponge. I
That's why. Sponges suck. Use brass wool and toilet paper
>>2774410
I really wanna try making one with couple AED capacitors and SiC fets.
>>2774958
it doesn't. You can use solder from 1950s just fine. Its pine rosin. Inside of impermeable metal tube.

[Anonymous No.2774970]

>>2774659
>leadfree
It is "for solder" not "solder" my friend.

[Anonymous No.2774975]

https://www.lcsc.com/datasheet/lcsc_datasheet_2005251035_ShangHai-Consonance-Elec-CN3158_C559032.pdf
it's an adjustable voltage tp4056
hope you can read escalator runes

[Anonymous No.2775002]

in circuit B, how does R3 keep Q3 off when Q2 is off?

[Anonymous No.2775005]

>>2775002
Q3 base has to be pulled low to turn Q3 on. If Q2 is off, R3 pulls Q3 base high.

[Anonymous No.2775006]

>>2775002
Q3 is a PNP. It's off when Q2 is on. R2 and R3 form a voltage divider relative to Q3's base, and limit collector current on Q2.

[Anonymous No.2775007]

>>2775005
thank you

[Anonymous No.2775008]

>>2774963
Why would I not?

[Anonymous No.2775010]

>>2775005
>>2775006
thank you

[Anonymous No.2775011]

>>2775010
You're welcome.

[Anonymous No.2775036]

>>2775011

they're real
i called it

[Anonymous No.2775038]

>>2775036
>no relation

[Anonymous No.2775039]

>>2775006
>>2775006
>R2 and R3 form a voltage divider relative to Q3's base
Sorry, I fucked up. This is wrong because when Q2's base is high, the base of Q3 is pulled low and the resistors are "invisible" to Q3.

[Anonymous No.2775041]

>>2775036
If the chinks knew what was good for them, they'd do a real hall of costs on the nose tribe, and enslave the monkey tribe again.

[Anonymous No.2775042]

>>2774966
No, it looks awesome.
Check out some old Sony.
That would be optimum.
Get your free copper that you paid for.
It’s kind of funny that CNC routed PCBs kind of do this naturally.

[Anonymous No.2775047]

>>2775039
>This is wrong because

your corrected explanation is still wrong
R3 is always doing something
R2 and Q2 disapear when input is low
but i'll let you figure that out yourself

for the benefit of the original noob, maybe i can come with a mnemonic for when BJTs turn on that makes sense to zoomers
- NPNs are cis straight: you stick it in their ass/base and splooge current into it to turn it on
- PNPs are gay/trans: they get turned on when you felch the splooge out of their ass/base

[Anonymous No.2775048]

>>2775047

yeah, that doesnt even make sense to me
i sound like a MAGA drone

[Anonymous No.2775050]

>>2775048
I think you're just full of ego for no discernible reason, but you do you.

[Anonymous No.2775058]

How common is it to create a spilt power supply for a battery powered opamp by setting a virtual ground to VCC/2? Sounds like cheating and there must be some issues with that. Because if something sounds so simple, there are always some issues with simple solutions.

[Anonymous No.2775061]

>>2775042
how fashions change. all these random curves and 90 degree angles would be totally ridiculed these days. now the fashion is straight lines with 45 degree turns and parallel to each other.

[Anonymous No.2775062]

>>2775061
>all these random curves and 90 degree angles would be totally ridiculed these days
Ridiculed by people who don't understand why, maybe.

[Anonymous No.2775063]

>>2775002
Q3 only opens if its Vb is < Vcc-0.7v so about 14V
When Q2 is off, Vb of of Q3 = VCC and it is closed. when Q2 opens, it pulls Q3 Vb down to about 10V (since R3:R2 ratio is about 1/3) and it will turn it on. Not 100% about 10V though it is just a quick guess, since some current is flowing into Q3 and the load.

[Anonymous No.2775065]

>>2775062
Probably. Especially when youtubers like Feranec and various gurus like Bogatin and that grounding guru Rick Heartley(?) constantly scare people with poor signal integrity and other other issues that are only important for extremely high frequencies and are not an issue at all for 99% of the hobby projects.

[Anonymous No.2775066]

>>2775065
Oh and avoiding 90 degree angles is also a scare story about how high currents can supposedly cause heating issues at the corners but they are really talking like really high currents, 10-50A. Straight angles are perfectly fine but people avoid them because aesthetics.

[Anonymous No.2775067]

Ok, from what I've managed to find out about soldering as a complete noob and retard:
Don't use pure tin solder because it destroys tips regardless of quality and sucks when melting, use only name brand 60/40
Irons don't matter but clean them with brass only or a damp rag
If rosin is not available, use only non-acidic flux like something vaseline based
Acid/plumbing flux is only good for cleaning corroded or extremely dirty tips
And finally don't bother soldering on PCBs because it won't work no matter how much care I take, stick to wires and plugs

[Anonymous No.2775068]

>>2775065
>only important for extremely high frequencies and are not an issue at all for 99% of the hobby projects.
That's correct.
>>2775066
>Straight angles are perfectly fine but people avoid them because aesthetics.
Yes. There are plenty of consumer grade electronics devices that implement right-angle traces without issue. The PC Engine/TG-16 are good examples from not that long ago. Not to mention hand-drawn circuits.

[Anonymous No.2775069]

>>2775058
>How common is it

it's practically a law that you have one in your analog circuit

>there are always some issues with simple solutions

none worth mentioning
source impedance is around 100 ohms, not 1 or 10, but usu no big deal

[Anonymous No.2775073]

>>2775067
>use only name brand 60/40
agree
>Irons don't matter but clean them with brass only or a damp rag
Irons matter. Tip metallurgy matters. How you clean them is up to you.
>If rosin is not available, use only non-acidic flux like something vaseline based
No vaseline. Use RMA or pure rosin flux. RA flux is more aggressive and will corrode things if you don't clean it up.
>Acid/plumbing flux is only good for cleaning corroded or extremely dirty tips
Meh. Just buy new (official brand name) tips. If you take care of them they'll last years.
>And finally don't bother soldering on PCBs because it won't work no matter how much care I take, stick to wires and plugs
Use quality tools and consumables. Keep practicing.

[Anonymous No.2775074]

>>2775067
>don't bother soldering on PCBs because it won't work

PCBs are the easiest thing ever
gotta be something wrong in your world
prob a decade's worth of oxidation

[Anonymous No.2775075]

>>2775061
It’s not a fashion. Computers are now starting to be able to create optimal designs for structures, and it usually looks curvy. The jagged square corners, right angles, and thin traces were limitations of autorouting programmes. We can do a lot better.

[Anonymous No.2775078]

>>2775075
>The jagged square corners, right angles, and thin traces were limitations of autorouting programmes. We can do a lot better.
Autorouting always sucked. That's why many things were done by hand.

[Anonymous No.2775080]

>>2775066
There is no excuse for shit like picrel.
It’s because they are too lazy, stupid, or ignorant to correct the auto router and hand place some 30° straight lines. A child could do it.
It’s true, there are effects with sharp corners, or, god forbid acute angles and dead-ends but those are more prevalent at high currents and high frequencies.
There are also high frequencies present in “low frequency” switching which everybody forgets. Like when your voltage regulator oscillates only on the breadboard but on the PCB it’s fine. A breadboard is shitty in the same way a 100% auto routed board is shitty when compared with a well designed one.
And there is no good reason, if you’re DIYing something, that you can’t have a more optimal, well designed pcb layout.
Don’t necessarily copy industry made techniques, most of them are to cut costs. Even the prevalence of SMDs and flip chips—cost cutting.
One day we might be laying down glass traces for optical PCBs and they’re not gonna be using hard right angles there, I can assure you.

[Anonymous No.2775081]

>>2775074
No, I only get blobs of solder when I try to make traces, regardless of using flux

[Anonymous No.2775082]

>>2775080
>There is no excuse for shit like picrel.
If it functioned there's an excuse.
>It’s because they are too lazy, stupid, or ignorant to correct the auto router and hand place some 30° straight lines. A child could do it.
What's the purpose of autorouting if you have to do it manually after the fact? lmao

[Anonymous No.2775083]

>>2775081
>when I try to make traces,
kek. we are not that easily trolled. try literally any other thread on this board.

[Anonymous No.2775095]

>>2775083
Ok
When I use perfboard it's next to impossible to make traces between distant components or ICs because the solder is hard to distribute evenly

[Anonymous No.2775099]

>>2775095
>When I use perfboard it's next to impossible to make traces between distant components

ok, we actually have some cold-solder experts who think that's a wonderful way to connect components, but I'd respectfully suggest you consider using wires.

[Anonymous No.2775109]

>>2775063
The R3 R2 voltage divider is irrelevant in this case. When Q3 is open, there will be a few mV voltage drop across it, and the rest will be dropped on the load. So the voltage on the base will be Vload - Vbc. Or alternatively, it is VCC- Vbe.

[Anonymous No.2775123]

I'm curious about learning how to make USB button boards, mostly because I'd like to make PC peripherals and fightsticks; where do I start learning how to do that?

From what I understand I'd be using Teensy boards, but what language do I learn/what do I need to learn in order to avoid just being someone who copy/pastes code from others? Python? Should I just buy an Arduino kit and follow some tutorials?

[Anonymous No.2775130]

>>2775123
>>>/diy/mcg

[Anonymous No.2775134]

>>2775130
My bad, thank you.

[Anonymous No.2775149]

What's the deal with Realtek ICs? From a consumer perspective they are ubiquitous, tons of hardware uses them, so you'd think they would be readily available. Yet I can't find datasheets for a component I'm interested in anywhere, and also none of the big electronics sites like Digikey or Mouser or Arrow has them. Is it not recommended to use this brand on ICs in my design?

[Anonymous No.2775165]

can any big brains tell me the output current of this board, running the output at 12v?
https://www.aliexpress.com/item/1005004957459719.html?spm=a2g0o.order_list.order_list_main.28.b0bb1802kr8eVk

[Anonymous No.2775173]

>>2775123
But yeah a good way to get started understanding mcs and their various inputs is to buy a kit. Buy one with from a decent company and one that has a book of things to make with instructions and code. I like kitronic but they may just be UK based.

[Anonymous No.2775186]

>>2775165

it says 5 watts max / 12 = 0.416 amps. But be sure you abide by this advice as well: "Change the resistance value of the resistance in the circle on the left to change the charging current. The resistance package is 0603. Note that there is a risk in modification. If there is a problem, our shop is not responsible. "

[Anonymous No.2775209]

>>2775058
It’s fine. PSRR might be slightly worse than a proper split-rail setup, but it’s better than just making each section of your circuit have its own DC blocking capacitor, especially when you’re using op-amps and you don’t need to worry about bias voltage. You’ll still want some high-pass caps if you need to use the switch built into a barrel jack or audio jack to turn on/off the battery power. If you have a spare op-amp, use it to buffer your divider. Add a filter cap on the divider too, a big one if you’re not buffering it in order to drop the high frequency impedance.

Can we get centre-tapped 9V batteries?

>>2775081
By PCB do you mean dot board? Try strip board instead.

>>2775165
Look up the datasheet of the 6-pin IC on the board. It may have a maximum input or output current that’s more limiting than it’s supposed maximum power rating. You may also have to calculate its power dissipation based on its efficiency. You’re probably better off buying a seperate protected TP4056 board and boost converter board.

[Anonymous No.2775222]

>>2775186
thats for the input charging (i think) as its a battery management and buck all in one
>>2775209
okay guess ill just have to wait till it gets here

[Anonymous No.2775223]

>>2775222
boost not buck*

[Anonymous No.2775233]

>>2775209
>>2775209
I want a center tapped car battery!
Imagine someone decided that cars needed a dual rail. But regardless I just thought that a car battery would be a cool addition to the work bench to power high current projects. I have a spare little JD battery and it never occurred to me to use it for that.

[Anonymous No.2775247]

>>2775209
>Can we get centre-tapped 9V batteries?
Open the case and center tap it yourself. If it's a shit battery it'll be a voltaic pile. Otherwise it's 6 AAAA cells.

[Anonymous No.2775351]

>>2775073
That's an equivalent of a mechanic saying don't use a 1 dollar screwdriver, use a name brand 100 dollar screwdriver, those cheap screwdrivers snap in half the first time you try to turn something with them.
I've seen repair shops using cheapest irons without temp control and unknown solder. There's something wrong with the way he's using the iron, not the equipment's fault. The cheap stuff sucks because it will fail after some use, not immediately

[Anonymous No.2775358]

>>2775351
Good advice, ching tran chong. Keep using shit tools since that's all you know. Beginners need to learn proper technique, and it's next to impossible using garbage. I'm sure you'll have a list of anecdotes to shit into the thread.

Faggot.

[Anonymous No.2775360]

>>2775358
Here's your chingchong tool, fag.
It's for wires and larger electronics. Will still get fucked up by thermal shock and abrasion if used incorrectly. If used correctly one tip will last for years like mine did

[Anonymous No.2775361]

>>2775360
What the fuck? Are you retarded?
>yes
Nevermind.

[Anonymous No.2775362]

>>2775361
If you want to be taken seriously while calling others retards, provide intelligent comments and advice since the retards don't.

[Anonymous No.2775365]

>>2775362
Refer to the first reply >>2775073 which you took issue with for some unknown reason.
Anything else I can help you with, retard?

[Anonymous No.2775366]

>>2775365
> retard
Got the wrong guy, my medical says otherwise
You'll leave this thread and still have an inferiority complex

[Anonymous No.2775368]

>>2775366
Ok, Mr. Dremel butane soldering iron. lmao

[Anonymous No.2775375]

>>2775058
>How common is it
very
>Sounds like cheating
lol
> there must be some issues with that.
need to buffer it sometimes if using it as a low impedence "ground"

[Anonymous No.2775491]

Where do I buy these MOSFETs?:

1E34BN
30N60A4D

1E34AT
30N60A4D

I'm trying to fix a Crown DSi 2000 amplifier. I opened it and noticed a dead short across the output of the rectifier and the two MOSFETs next to it. The short resistance was lowest at the big transistors so I pulled them out of the circuit board and tested them in isolation - they are both shorted without any ambiguity: all 3 pins are a straight short to all other. When I went to DigiKey it couldn't find an equivalent MOSFET by the codes alone. What do you guys suggest I should buy as a substitute component?

[Anonymous No.2775494]

>>2775491
I searched "30N60A4D" and this was the first result:
https://www.mouser.com/datasheet/2/149/HGTG30N60A4D-89590.pdf

[Anonymous No.2775499]

>>2775494
I just need 2 not 450

[Anonymous No.2775501]

>>2775499
They're EOL parts.
https://www.amazon.com/s?k=HGTG30N60A4D&ref=nb_sb_noss

[Anonymous No.2775504]

>>2775494
>this was the first result

Machete dont google either

[Anonymous No.2775506]

>>2775501
Do they not make a modern IGBT MOSFET with the similar specs?
>>2775504
I did Google.
It sounds like if I order the specific part then it will be out of stock or overpriced, but if I find a modern IGBT 600V MOSFET then it will be cheap.

[Anonymous No.2775508]

>>2775501
Just in case you hate amazon:
https://www.ebay.com/sch/i.html?kw=HGTG30N60A4D&toolid=20004
>>2775504
lol
>>2775506
>IGBT MOSFET with the similar specs?
Search "IGBT 600V 75A TO-247-3"

[Anonymous No.2775510]

>>2775506
https://alltransistors.com/igbt/transistor.php?transistor=380

[Anonymous No.2775518]

If I have a 10Vpp (±5V) audio signal that I want to route to my headphones through a suitable op amp (currently looking at the NJM4556), do I lower the input voltage before the op amp, or do I set it up so that the gain is <1?

[Anonymous No.2775529]

>>2775518
Attenuate the signal first (passive or active), then send it to your headphone amp. You could also use an audio transformer.

[Anonymous No.2775533]

>>2775518
Can you tap the signal prior to the amp stage (switched 3.5mm jack)?

[Anonymous No.2775555]

>characterizing a NMOS for a lab (ZVN2106A model, we were trying to find lambda value)
>noticed when gate voltage was near threshold, drain to source current dropped significantly
>hmm
>suspect gate was leaking some current, so placed a 1Mohm res between gate and source
>current now works as predicted
So there was current through the gate. But why? I'm totally stumped.

[Anonymous No.2775557]

>>2775555
*Meant to say "between gate and the gate's voltage source"

[Anonymous No.2775558]

>>2775518
Do volume controls raise/lower the signal voltage, or raise/lower(limit?) the current?

[Anonymous No.2775560]

>>2775501
> things should be designed with the IRFZ44N, which is, I think, the most popular mosfet in the world. It you don’t want it to be EOLed, that is.
No idea if it’s any good for audio though, it needs a large linear region for that so it might be tricky.

[Anonymous No.2775617]

>>2775518
>10Vpp (±5V) audio signal ... to headphones

headphones are loud at 1/2 a volt
you wanna drop the signal 20 fold
so all you need is a 640 ohm resistor in series
no batteries required

use a pot if you're super rich and wear gold lamé underwear

[Anonymous No.2775667]

>>2775560
Well considering it originally had two 390V 96A IGBTs, I suspect IRLF44Ns wouldn’t work. The transistors probably make up a push-pull transformer driver for a low-noise SMPS, as opposed to anything audio-side. Because 390 is a tad more than rectified 240V mains will ever be. Though maybe the tube guys’ thing for high voltage amplifiers feeding transformers is contagious.

>>2775617
Different headphones have different sensitivities and impedances, there is no one-size-fits-all approach here. A pair of resistors, a dual-gang pot, and some clipping diodes for protection is what I’d use.

[Anonymous No.2775683]

>>2775667
> headphones different sensitivities and dependencies
I heard the “standard” for headphones nowadays is 32 ohms. Can anyone confirm/deny that?
So if you were gonna design a generic headphone thing, that’s probably the target you want to hit.
I’ve never seen a headphone amp with an hms switch though.
I used to have a pair of radio shack outdoor speakers with a 4/8 ohm terminal.

[Anonymous No.2775699]

>>2775366
Hey Anon. I apologize for calling you a retard. It was uncalled for.

[Anonymous No.2775710]

Do you 3D print your project boxes or buy ready made ones and adjust them to your needs? I don't want to buy my own 3D printer and would rather order on jlc3dp. However I am not sure if it is worth the learning curve. 3D modeling software is much more complicated than the average EDA package. I watched some videos and I am not really sure I want to spend that much time learning how to use it just to make a custom project box.

[Anonymous No.2775714]

>>2775710
Sounds to me like you've already decided. If you have some other uses in mind then I'd say buy a printer, but for a one-off it's probably not worth it. You can always craft your own out of materials at hand for a relatively low cost.

[Anonymous No.2775730]

>>2775710
I 3D print enclosures I make using Sketchup Make 2017. It’s easy to use and powerful enough to do basically anything I want, but it lacks native parametric modelling or the ability to add instant chamfers. So more complicated stuff can be time consuming. Also it crashes once per few hours or so, so save often. For parametric stuff I use OpenSCAD, it’s easy if you’ve programmed before, still doesn’t do chamfers though. But I wouldn’t recommend ordering 3D printed parts from the chinks. I often do test prints and design revisions to make sure everything fits perfectly, waiting a week between each print would be hell.

[Anonymous No.2775734]

>>2775710
>3D modeling software is much more complicated than the average EDA package
I would say it's the opposite, but okay.
The workflow is actually quite simple. You make a 2D drawing of a part then extrude it into the 3rd dimension.
For a simple box, it would be something like
>draw shape of base
>extrude 2mm or however thick you want it to be
>sketch outer and inner walls
>extrude to desired height
>sketch holes
>extrude cut
>new part, sketch the top cover, extrude, add holes
Of course, this is all if you choose a user friendly CAD like Solidworks and not gentooman specials like OpenSCAD or Blender.

[Anonymous No.2775736]

>>2775734
We’re all gonna be using freecad forks by 2030, I promise.

[Anonymous No.2775737]

>>2775710
FreeCAD is reasonably easy to use and it's free.
The problem is gonna be fixing up your prototypes and print settings. I don't know how those 3d printing services handle that, I imagine they'd have some quality control to ensure the printed part is in spec to the drawing but xing's gonna chang so who knows...

[Anonymous No.2775738]

>>2775736
lolno
I forgot to add Freecad to the list of gentooman specials. It's a piece of shit developed in typical FOSS style where the devs live in a command line and can't pull their heads out of their asses for long enough to develop a proper UI.
But at least it has some UI, not like OpenSCAD.

[Anonymous No.2775744]

>>2775710
I assume you’re talking about a plastic box.
I just cut PVC or plexiglas panels and assemble them and use solvent welding at the seams.
It takes 20 or 30 minutes.
Most projects should be in metal enclosures. You can get a large aluminum hollow extrusion or a big 1U rack-depth case and cut it with woodworking tools, and two end-plates from stolen street signs.
Or, two U-channels bolted together.
Audio projects should be in ferro-magnetic enclosures, so usually I just put the electrical components in a metal electrical box (they are like $1) and put *that* inside the for-show case.
Steel cases are a bit of a bitch, but you can get that 70’s look by taking a metal sheet, bending it, and then cut a matching groove in two side panels made of wood.
Wood itself is a pretty nice case.
3D printed cases are expensive in PLA alone, covered in lines and other artifacts, and take a long time to print and have terrible thermal properties (they melt).

[Anonymous No.2775746]

>>2775744
…cont
Here’s an example with the wood sides I was talking about. Now, the top is also wood, which I don’t like, but if you make the front panel and the top one piece of sheet metal with a single bend in it, and cut matching groves in the wood so the panel slots in. Also you can make the top panel bend an obtuse angle, cut the sides to match, and have a sloped panel where you put all your mixer controls. Or whatever you’re making.

[Anonymous No.2775751]

>>2775683
32Ω is pretty common. But sensitivities change too. I've got 32Ω IEMs where 1/16 bars of volume on my phone is too loud, and other IEMs where 4/16 bars is perfectly fine. There is no one perfect voltage level, so unless you only ever use one model of earphones you should always make volume adjustable.

>>2775738
Yeah it's shit right now. Ondsel and Realthunder are barely any better. But Ondsel is potentially putting real money into the project, and making pull requests to the original repo as they go. So give it 5 years and hopefully it will be as usable as KiCAD. Maybe. SolveSpace anyone?

>>2775744
Anything that large you're better off using metal or wood enclosures I agree. Aluminium is good enough for most cases, it stops electric field noise just fine. Magnetic field noise is much less of an issue, but when it is a problem yeah you'll want to use an iron case. But before that I'd put some effort into reducing your loop area with proper PCB design and twisting your wire pairs together. Point-to-point can eat my ass. Either way, use whatever is cheaper and has easy grounding.

I 3D print cases for smaller things. Like 3cm circuit boards, a housing to hold a limit switch and RJ45 socket, motor mount, that sorta stuff. Build PCB standoffs into the case at whatever angle or position you want. Build a battery enclosure with slots for spring-loaded contacts into the case too. The cases I make are like a dollar of PLA, and at that scale it's more than strong enough because you can fill a bunch of otherwise empty space with structural infill. As for looks, if I cared I'd get a PCB made (or etch my own) to be a faceplate.

[Anonymous No.2775784]

>>2774422
>>2774427
So, wait a second. Are you saying you *can* parallel diodes to get double the current capacity?
I was told “no” previously.
Or are these things supposed to be half a bridge rectifier?

[Anonymous No.2775785]

Im repairing (torturing) poor little 2012 macbook.
It didn't have backlight, and I've replaced the BGA LED step-up converter chip (LP8550) and there were still no backlight...
I checked schematics and apparently backlight is turned on via P-MOS tube, which was dead.
After replacing the MOS-tube I had the backlight, but its behavior was weird since it turns on like a fluorescent tube, blinking and making clicking noises....
And it shouldn't do this, because shit's solid state and uses LEDs. What can cause this? Flux under the BGA? How do I get that flux out from under there? How long it takes for IPA to evaporate from under the BGA (think big BGA like CPU, because I spilled a lot of it on the board)?
Why did Apple decided to use contraption with 3 MOS tubes instead of just sending enable signal straight to the driver (and there is some reset signal too, which can just idk, pull that 3.3V logic down)?
And in general, why did apple add P-MOS tube on every fucking device in this laptop, audio, ethernet controller, idk what else? If you check schematics for any other laptop, only shit you gonna find is a fuse or jumper. Power consumption? Really low power sleep mode?
>>2775358
Pic is a lie. Nothing did happen. -15 aliexpress points
>>2775067
>Don't use pure tin solder because it destroys tips regardless of quality and sucks when melting
Lead-free just sucks.
>use only name brand 60/40
You can try noname, but it would cost about as much as brandname because lead is lead, and tin is tin.
>Irons don't matter but clean them with brass only or a damp rag
Don't use damp sponge. It will fuck your tip, because I know you will use tap water and not reagent grade distilled water.
>If rosin is not available, use only non-acidic flux like something vaseline based
Vaseline isn't flux. Read label for the flux.
Rosin is good for big stuff, but when you solder small BGAs and idk, resistors, factory made flux is better since it stays clear for much longer and doesnt bubble.

[Anonymous No.2775787]

>>2774422
>>2774427
So, wait a second. Are you saying you *can* parallel diodes to get double the current capacity?
I was told “no” previously.
Or are these things supposed to be half a bridge rectifier?

[Anonymous No.2775788]

>>2775785
oh yes, i also replaced polyfuse (since it was suspicious) with a link.

[Anonymous No.2775789]

>>2775787
You can parallel diodes, but they're not necessarily going to share current evenly unless they're thermally bonded. Which two diodes in the same package are. Worst case scenario parallel diodes undergo thermal runaway from uneven current sharing, but for short pulses of high current it probably doesn't matter.

[Anonymous No.2775797]

>>2775789
Thanks, that makes sense. I keep forgetting it’s *thermal* runaway. I guess you could theoretically do the same thing with a lighter underneath your diode.
Also, I thought of another thing with that 2-in-one package—they’re both from the exact same batch, not two random diodes made 30 years apart from the parts bin. So they die size/doping/oxide layer is gonna be pretty much identical too.

[Anonymous No.2775845]

>>2775751
>Build PCB standoffs into the case at whatever angle or position you want. Build a battery enclosure with slots for spring-loaded contacts into the case too

That's the main reasons I'd want to 3D print a custom enclosure. Regarding the standoffs, the alternative would be to buy a generic project box with pre-made standoffs and then design your PCB board around that. But the battery placement is always an issue. How cool would it be to make it look professional and have a battery compartment on the back side.

[Anonymous No.2775847]

>>2775787
From what I understand it is a big no-no in theory because of the uneven current and all. but in practice I've seen multiple parallel rectifiers in the schematics of commercial products.

[Anonymous No.2775881]

>>2775710
get a 3d printer its 2024

[Anonymous No.2775892]

Sounds like a dream job.
>Get working gerber files
>Add a dummy chip
>Hide the main chip under the heatsink
>Done
How can China even compete with this?

[Anonymous No.2775917]

I got a 12v Victron smart battery charger. The charger came with two wires. Ive connected one part to the battery permanently and it has a socket to connect the second wire into which goes into AC. In the booklet it says you must always first connect the two wires and then AC and to disconnect first disconnect AC and then the wires.

What is the point of this? Sometimes its so annoying and I just turn all of it on by pressing the power button on the AC socket.

[Anonymous No.2775921]

>>2775847
Reminds me of the old saying
“In theory there is no difference between theory and practice, but, in practice, there is.”
I guess this is another one of those cases.

[Anonymous No.2775922]

>>2775881
> 2024 and no 3d printer
I can’t buy one, last year you told me to put all my money in NFTs because it was 2023.

[Anonymous No.2775927]

>>2775892
China invented it. Remember “the big hack” article? It’s easy to hide a padauk-sized microcontroller inside the PCB or inside the ethernet pulse transformer as the U.S. does.
Obviously, in this case, someone is just trying to get around sanctions (and it’s probably not realtek)

[Anonymous No.2775930]

Bought a cnc recently and first time using it and it was a failure. First I routed backside (left image), I tried several times to get the Z right but somehow fucked it up and drilled too deep which removed a lot of copper. I went ahead and did the other side, which was slightly better.

While drilling I noticed it lowers Z too fast which fucked up some drills. And then after flipping it over, I noticed I fucked up the alignment by 5mm, heh

Everything looks so much bigger on kicad & flatcam, when I start drilling I noticed there is not much room for error. Will try again tomorrow

[Anonymous No.2775931]

>>2775930
now I noticed that Y alignment is a bit off as well, so I also nudged things while flipping. I should try using alignment holes instead of hoping that it will flip perfectly

[Anonymous No.2775932]

>>2775930
I’d look into why those L shaped borders look so laser perfect, different tip? Higher rpms? More flutes? Deeper cuts? Different pass?

[Anonymous No.2775933]

>>2775931
Maybe you should do two thinner single sided pcbs and epoxy the two non-copper sides together aligning the holes.

[Anonymous No.2775942]

>>2775932
They are cutout lines. It is the same tip but they are done with 0.15mm steps with 5 passes.

They probably look better because it wasn't cutting too deep unlike other cnc jobs

>>2775933
Yeah, I can try that in future but I don't have single sided PCBs at hand right now. But still I should figure out how to do 2 sided PCBs. I imagine it will be a lot easier if I just use alignment holes. I know where I fucked so next one will be a success I hope

[Anonymous No.2775947]

im happy with jlcpcb for pcbs but i've wondered, has anyone used jlcpcb for cnc? do they actually make a quality product (follow the drawings, meet tolerance, etc.) or is it a total joke?

[Anonymous No.2775955]

>>2775847
If they're reasonably well-matched (from the same batch or even just the same factory may be good enough) and you derate the total current rating by ~20%, I doubt you'd run into trouble. The problem arises when you want to get 50A so you put 10 5A diodes in parallel. Current sharing is a much bigger issue at higher numbers of parallel components.

>>2775917
Power cycling order, maybe the AC power supply stage needs to be bootstrapped by the 12V battery. It would be a very strange design though, since off-line converters that just work without bootstrapping are very common.
The naive alternative would be that it needs to be connected to mains before a battery can be connected, which would run into trouble in event of a power cut. In reality, it should be trivial to have the control circuitry run on either source of power.

I'd take it apart and reverse engineer the relevant section of the schematic to see for sure, and if you can modify it to be fine with either. But I've never seen a battery charger break because of incorrect power cycling order.

>>2775933
That sounds awful. Aligning front and back sides is just a matter of getting a clamping setup that's consistent. For my machine I 3D printed a corner clamp that always holds the boards hard against the front-left corner, then I flip them diagonally about that corner. Alignment holes work well if you can properly mount pins for them to slot into.

>>2775947
I've seen bad stories from outsourced CNC work (Brandon Hererra's AK-50), I think the issue with that is workholding being changed and losing reference. Anything automatic via CNC should be pretty accurate, but when a human comes to take the part off the vice and flip it about, then you get issues. So I'd only rely on JLC for parts that can be made in a single operation.

[Anonymous No.2775979]

I want to buy a type 3 surge protector for my appliances because I have lost shit in thunderstorms before.
I've found 2 models here. They differ only in 2 attributes:

model 1:
Up: 1kV
Response time: 25ns

model 2:
Up: 1.3kV
Response time: 1ps

I'm interested in model 2 because of the fast response. Do you guys have any advice?

[Anonymous No.2775990]

>>2775955
Thanks I reread the manual it is weirdly structured but I read between the lines that this is probably only necessary for first time setup.

Im too much of a noob to take it apart but im getting somewhere with my 12 circuit. Connecting computer fans, leds etc ill get there one day

[Anonymous No.2775994]

>>2775979
Isn't the joule rating more important?

[Anonymous No.2776001]

>>2775917
It may be a dual voltage (12V and 24V) charger that decides which voltage you have on startup. Victron solar chargers do this too.

[Anonymous No.2776006]

>>2775994
all the other parameters I've found are the same between the two models, but the specs don't have "joule rating".

[Anonymous No.2776021]

>>2775955
>That sounds awful
Shhh… I’m gearing him up to make multilayer PCBs

[Anonymous No.2776026]

Why don't MMs auto correct for the burden voltage? Seems trivial. If the shunt resistor value is known, and the voltage drop is known, why not do the math and re-calculate the current?

[Anonymous No.2776062]

>>2775785
Okay it was flux or alc residue, in the morning it did start up normally

[Anonymous No.2776077]

>>2775979
>1ps
Haha yeah right. The stray inductance and capacitance alone in that thing is going to have a higher response time than that. Find the specs of the MOV it uses to actually see it’s response time, but also the absorption energy. You want that to be high, at least 250J. Time doesn’t really matter, a decent in-line power filter (X2 cap, normal mode chokes, X2 cap, Y caps) should slow down an impulse to the millisecond domain. Common-mode chokes might help, not too sure. Also put a fast blow fuse before the MOV(s) which should be before the filter. Ideally all this is built into your appliance anyhow, but replacing dead MOVs from ATX supplies is a pain.

>>2776026
Depends on the current source. A 0.1V burden voltage on a 10V source is going to effect its output current by 1%, but on a 1V source it would effect its output current by 10%. If burden voltage is significantly impacting your current measurements, you’re using the wrong tool, bolt the wires on a low-value inline shunt instead.

>>2776062
Gratz

[Anonymous No.2776094]

>>2776077
>Find the specs of the MOV it uses to actually see it’s response time
the response time is likely due to the GDT or the SAD, not the MOV. also, can't know the MOV they use without buying one and opening it.
>Also put a fast blow fuse before the MOV(s) which should be before the filter.
if I want to buy one, it is obviously because I don't want to do it myself.

[Anonymous No.2776099]

>>2773862
I want to make a watch and i have been googling information about the crystal oscillators. Is there such thing has higher quality crystals.
https://www.river-ele.co.jp/ev/products_post/tfx-05x/
shit like this is 8$ a pop why would you ever buy it? looking at the technical data it doesn't seem like there is much improvement in error. To get the most accurate crystal should you just look for the lowest ppm and that's it?

[Anonymous No.2776103]

>>2775979
i guess i've found my answer:
>All SPDs have sufficient response time to “turn on” and shunt surges. The response time of an MOV is 1000 times faster than the time it takes for a surge to reach full current (i.e., 8 microseconds).
>The commonly specified <1 nanosecond response time for a complete SPD is not realistic. [...] Instead, it is important to focus on the measured let-through voltage, or UL’s Voltage Protection Rating (VPR) that takes into account many variables including the initial response of the components and the added inductance of the system.
so i'll just get the one with lower Up. I wonder why the other one exists though.

[Anonymous No.2776114]

What makes an OpAmp and audio opamp? There are so many precision OAs, ultra low noise, high bandwidth, ultra low harmonic distortion, high input impedance, low offset current, high voltage VCC and input, etc. but their typical applications are measurement and instrumentation and not audio.

[Anonymous No.2776117]

What's the purpose of the bypass diodes?

[Anonymous No.2776118]

>>2776114
audio op amps have ultra low noise, high bandwidth, ultra low harmonic distortion, high input impedance, low offset current, high voltage VCC and input, etc for the frequency range of audio. 20 to 20khz. You can use high percision OP amps that also have all these features and work over the range too. Calling it an audio opamp just lets you know it has these features.

[Anonymous No.2776119]

>>2776117
So you don't backfeed the panels with your batteries.

[Anonymous No.2776120]

>>2776118

Ha! I was just typing this. This OA is advertised as 0.00001% THD but look at frequency response.

[Anonymous No.2776121]

>>2776117
>>2776119
Oops. It's so the panels don't overpower the weakest link, in the event that one or more panels are uneven output.

[Anonymous No.2776136]

>>2776120
Yeah its why when you search for opamps for audio shit you search for audio specific opamps it removes alot of garbage that will have sneaky shit that will make your stuff sound like trash if you dont read the whole data sheet.

[Anonymous No.2776144]

>>2776121
How does it work? Let's say the panels produce 7V each (14V in series). And one is producing just 4V. What will happen then?

[Anonymous No.2776146]

>>2776136
Plus I figure certain parameters don't really matter. For example Voff and Ibias, and RRIO may be critical for instrumentation but are probably irrelevant for audio.

[Anonymous No.2776149]

>>2776144
The diode will keep the weak panel in a forward discharge state, but the other panels will essentially cancel the weak panel out unless they're shaded and put out less or equal voltage.

[Anonymous No.2776155]

>>2776117
I guess it is to limit the maximum generated voltage from each panel, so you can charge your battery at/until a safe/desired voltage.
E.g.: if you use 7.5V bypass diodes and the 1V blocking diodes, your maximum generated voltage will be 14V. Which means your battery will be recharged until 14V.
I've never worked with solar power but I wonder if it would be better to limit the total current instead, because in the end this is what really matters.
The battery probably has an specification of maximum input current. Without current control, the maximum current generated by the panels shouldn't surpass the maximum input current of the battery.

[Anonymous No.2776159]

>>2776117
When a panel is shaded and in series with an unshaded panel, the unshaded panel can force current through, heating the shaded panel up from its substantial voltage drop. The diode bypasses the shaded panel with a much lower voltage drop.

>>2776155
They’re Schottky diodes, not zener diodes. The diagram of using them without a charge controller isn’t ideal, that setup could eventually boil the batteries dry if the open-circuit voltage is above 14V or so.

[Anonymous No.2776161]

>>2776159
>They’re Schottky diodes, not zener diodes.
my bad

[Anonymous No.2776176]

What's the point of high current buffers for headphone amps? If they are going to draw 200mA won't your ears explode?

[Anonymous No.2776178]

>>2776176
aren't them used to guarantee that your phone will get the current no matter the impedance? like, different phones have different impedance etc

[Anonymous No.2776179]

>>2776176
Bass tones can draw a lot of current, if the headphone can play that frequency range.

[Anonymous No.2776183]

>got a DMM lead without continuity through it
>didn't know which end was broken
>connected either end up to my function generator at 50kHz
>negative on spike, positive on banana
>ran scope probe along wire from end to end
>50kHz sine capacitively couples to my scope probe
>signal gets weak only near the spike
>flip it around
>signal only gets strong near the spike
As expected, the bonding of the wire to the spike was faulty. Gonna chop it off and replace it with another spike.

[Anonymous No.2776187]

>>2776183
You should just buy new probes even if they're shitty. Otherwise, mark the probes so you don't use them on anything that could kill you.

[Anonymous No.2776205]

>>2776187
This will be my backup meter, or more likely a meter for a friend. I have a pair of really shitty probes, but I'd trust them less than buying replacement DMM probe ends and fixing one on the end of this wire. Be even easier if it were the banana that needed replacement, but what can you do.

[Anonymous No.2776236]

>>2776176
there is none, it's just audiofraud bullshit

[Anonymous No.2776240]

>>2776176
Some headphones are less sensitive than others, and hence need higher drive power than others for the same loudness. Headphones with higher impedance run on higher voltages and lower currents, while headphones with lower impedances run on higher currents and low voltages. So conceivably, a low-sensitivity low-impedance pair of headphones could require a high-current headphone amplifier to run.
As the other anon alluded to, the sensitivity and impedance at certain frequencies may differ, meaning you may need to pour a different amount of current or power to get the same loudness. If the sensitivity is different you'd need to fix it in an EQ, if the impedance is different but the sensitivity is the same, you may need an amplifier that can provide more current or more voltage than you would otherwise. A common case might be impedances being lower at low frequencies, because there's less inductive reactance. A dynamic driver's pressure is a function of force and hence current, so in that case you might actually find that a constant-voltage driver results in louder bass than mids, which would need to be tuned back with an EQ, else it would draw more current for the same voltage. Or you could run it on a constant-current headphone amp. Electrostatic drivers don't have this problem.

[Anonymous No.2776242]

>>2776099
You’re using the wrong technology.
You need an antenna to listen to WWVB instead of a super-accurate crystal.
Also, even if you did get a bunch of crystals to characterize them, the only way the likes of you would be able to do that is with WWVB and let it oscillate for a month to see how far it was off.

[Anonymous No.2776244]

>>2776205
Here, bro… this is your controller.

[Anonymous No.2776252]

>>2776242
GPS is even more accurate, if you properly adjust for offset:
https://mitxela.com/projects/precision_clock_mk_iii

[Anonymous No.2776283]

This >>2774032 was a terrible idea!
The graphite fibers breaks down to millimeter sized little frickers and it does NOT wick anything. It's all over in my open jar with flux paste now!
It's not even worth cleaning the tip with.

Graphite felt is outright something to BAN in the vicinity of electronics production along with helium. Beware! It's the Devil's spawn!

[Anonymous No.2776284]

>>2774135
>what the hell is an "and or invert" gate?
It's a single NPN transistor as a NOT gate.

[Anonymous No.2776294]

>>2776283
turns out the ability to wet to solder, and the ability to wick solder when in fine strands, are the same thing
who could have guessed

>>2776284
>It's a single NPN transistor as a NOT gate
gpt-tier answer
i googled it, it isn't "AND or INVERT", rather it's "AND, OR, INVERT", or in other words "AND, NOR". don't think i'd ever use one, for combinatorial logic i'd make a karnaugh map, which usually produces logic that's best constructed out of nand or nor gates alone.

[Anonymous No.2776305]

>>2776294
>who could have guessed

is that it?
you owned the fool, coulda gloated a lil more, done a dance, posted a meme

[Anonymous No.2776375]

>>2776240
>So conceivably, a low-sensitivity low-impedance pair of headphones could require a high-current headphone amplifier to run.
In theory yes, but only in an extreme case where you have extremely low sensitivity headphones like 85-90dB/mW at 32ohm and you want to destroy your years and crank the volume up to >110dB. But in practice most headphones have sensitivity >100db/mW. So the most power you'd ever need out of them is 1mW. At 32 ohm it is about 5mA. Also, the extremely low sensitivity headphones usually have high impedance. And even 100dB is too fucking loud.

[Anonymous No.2776390]

Horrible job but it works. I cut too deep again. Especially front/left pic. This time it is me being dumb and not trusting the tool. It felt like it is not cutting enough and I bumped the cut height and then it was cutting too deep.

I also broke a drill because I forgot to zero my Z. I shut it down immediately but then my cnc controller coordinates went out of sync. I was able to align it again though

[Anonymous No.2776391]

>>2776390
And here it is in its glory. One led flip flops, the other dims using potentiometer& pwm and the last one is controlled by the button. It is attiny13a

[Anonymous No.2776400]

Is it possible to "divide" a PWM timer in a microcontroller into several output channels? I'd like to drive four fans with a single 16-bit PWM timer

[Anonymous No.2776423]

>>2776400
Yes but check the power on that pwm signal, you might have to amplify it.

[Anonymous No.2776435]

>>2776400
> posts screenshot png of ascii art…

[Anonymous No.2776488]

>>2776423
Thanks, maybe i could use a ULN2003A to switch the fans? They're rated 120 mA

Any recommendations how to pwm the fans separately (individual control)? If I set up the 16-bit timer for e.g. 100 Hz pwm period, can each of the four channels output an individually controlled 25 Hz pulse width modulated signal?

[Anonymous No.2776489]

>>2776391
>>2776390
And that seems like a very cool project, nice job. Did you use the 16-bit timer for your PWM?

[Anonymous No.2776493]

>>2776488
>ULN2003A to switch the fans? They're rated 120 mA
I don't know which fans you have. If they take less power than your transistor can deliver then you're golden.
>how to pwm the fans separately (individual control)?
Easiest would be what you're doing in your drawing: have 4 outputs coming from the MCU.
>25 Hz
That's probably too low and you'll see the fan "flickering" at that frequency.

[Anonymous No.2776494]

>>2776391
So, what exactly is the purpose of this?
And why a micro in the first place? You can blink an LED with a 555, dim it with a resistor and a pot, and the push button one is self explanatory.

[Anonymous No.2776496]

Does it matter which type of inductor I use for a project? I can't find toroidal ones, only axial ones.

[Anonymous No.2776512]

>>2776496
Yes. Toroidal inductance is specified as TL, while axial inductance is always AL. However, in practice, we usually just erase the T and A parts with little consequence.

[Anonymous No.2776513]

>>2776512
So I don't even need to adjust the value, just use the same value I need as a toroidal (200uH) but with an axial instead?

[Anonymous No.2776530]

Im doing a pcb project, its a stepper motor cnc controller. Its a very simple board, mostly headers, a resistor, and 3 capacitors. the rest of the parts are things i dont make, like the ethernet adapter, mcu, stepper drivers, etc

if i want to distribute the board, how do i minimize/eliminate personal liability (for something like fire) ?

How can I test fire-ability?

[Anonymous No.2776536]

>>2776489
Yeah it is controlled by the timer

>>2776494
There is no purpose. I bought a cnc recently and made this as an exercise

[Anonymous No.2776545]

>>2776400
Depends on the MCU. Ones I know AVRs) have multiple pins that can be used for any one PWM channel, but only one of those can be active at a time. They usually have multiple PWM channels per timer, two is common, I think I’ve seen 4 in the newer AVRs, but only when that 16-bit timer is split into two seperate 8-bit timers. Well most MCUs have multiple timers anyhow. Do your fans have PWM inputs, or will you just be switching power to them directly?

>>2776489
I didn’t think the 13 had a 16-bit timer.

>>2776496
Axial inductors are more prone to radiating and receiving noise.

>>2776530
Use good board-wire connectors, screw terminals are asking for trouble. Add thermistors to key parts of the board (some ICs may be able to measure their own temperature). Code in thermal runaway prevention, when you do something that should produce heat but you’re not reading an increase of temperature, assume that the thermistor is broken first and don’t run until it’s fixed. Fuse on the input, current monitoring on the loads. Airflow and ambient temperature recommendations. PSU recommendations. Buzzer/alert output. Ensure the code and other on-board logic is failsafe.

What’s it for? Why not just use a cheap Marlin board?

[Anonymous No.2776546]

>>2776545
>I didn’t think the 13 had a 16-bit timer
It does not, it is 8 bit

[Anonymous No.2776553]

How long do capacitors hold a charge after a device has been turned off and is there any way to measure how much they're still charged?

[Anonymous No.2776557]

>>2776553
>is there any way to measure how much they're still charged?
Multimeter.
They will discharge through it as you measure, though.

[Anonymous No.2776559]

>>2776545
the board was made to promote the firmware project I work on, but i dont know how to code. so the board was made to be super inexpensive, using the most common shit around, so basically every country can access the parts needed. Its basically one of those red arduino cnc shields (like identical, i copied the sch mostly) with an stm blackpill attached.

>What’s it for?
trick people into becoming firmware developers for the project, because i dont know how.


>Why not just use a cheap Marlin board?

subsidizing the cost of the marlin boards was a lot more money. by making my basic ass board, its like 1/4 the cost. its mostly just 0.1' headers, , not hotends or mosfets or anything like that.

im gonna do some tests this week, and i dont think it will catch fire, but i also dont want to be completely liable if something happens either....

[Anonymous No.2776569]

>>2776557
Won't that fry the meter though if it still has a lot of charge on it?

[Anonymous No.2776570]

>>2776569
Are we talking about microwave capacitors here?

Don't anon, just don't. You sound like you are looking for next darwin award

[Anonymous No.2776576]

So I've got this police ballistic shield with light, and I REALLY want to have it do the strobe from the "parties too hard and dies" memes with that song from caramelldansen (just the flashing colored lights though, no speaker involved)

Currently the lights are LED, at 1200 lumen, feeds from 3x or 6x CR123's and are white only.
(this is the FoxFury B70 Ballistic LED Light)

Can anyone recommend me some RGB LED's and a means of programming them? The strips on amazon seem too weak and not very programmable (they only have preset modes it seems)

I think to do this I will either need to make a custom front cover (which is doable, I know CAD and 3d printing) to fit new lights in, or replace the entire light + handle + switch assembly. Which is also fine, but I don't know much about arduino vs raspberry pi, nor do I know much about LED options

Would really appreciate any product recommendations for LED's or programming boards/languages, or ideas on this in general. Not very experienced with electronics.

[Anonymous No.2776579]

There are schematics of regulated linear power supplies with lots of expensive filter capacitors. Why spend so much money to reduce the ripple if that could be done by the regulator especially if it is a discreet (opamp) regulator. It seems cheaper to pay $3-$5 for a a good opamp with PSRR of 120-140 db up to a few KHz, rather than spending close to $100 on big caps?? I thought big caps are only needed for unregulated PS.

[Anonymous No.2776603]

Hello guys, I have tried replacing a new power jack in the charging port but it seems like I haven't soldered it properly or even damaged? I don't know I would really like some advice.
>>>/g/99652821

[Anonymous No.2776608]

>>2776603
You seem to have stripped a screw head (upper part of the pic).
I don't know why there's a piece of solder(?) sticking out from the jack.
And few hookers get their holes as fucked as those in the board, specially the second from top to bottom. What did you do there?
Anyway, if the board is multilayered and you shorted out those layers, they you fucked up big time. If it isn't, then check for continuity between the contact pads and leads and if you don't have that then clean this mess and solder it correctly, then check for continuity again on those joints to make sure it is as it's supposed to be.
From skimming over your thread, it seems you used tin solder, which is harder to use and requires more heat. You don't really need flux, what's on the solder is generally enough but might help you if the solder just don't want to stick to the pads (cause pure tin is a harder to use).

[Anonymous No.2776612]

>>2776608
Lenovo 3-15ohu6 , I'm not sure it is multilayered

[Anonymous No.2776613]

>>2776612
Neither am I, mate.
Try working on the stuff you can and worry about it later. Resoldering if needed would take less time than researching that.

[Anonymous No.2776614]

>>2776613
I removed that middle piece, not sure it did anything anyways...

But other than that, does it makes sense that in general the motherboard and the laptop itself should be working? Since other than charging I haven't tackled anything else

[Anonymous No.2776617]

>>2776614
If it is multilayered (which it probably is) and you shorted those layers (maybe you didn't, I dunno) when fucking the holes, then you shorted everything in the board. In that case it's fucked.
Get a multimeter a check continuity between the leads of the new part and wherever they're meant to go to. If you have that then your soldering did it's job.
If you have continuity between those leads and everything else everywhere on the board then you fucked up.

[Anonymous No.2776619]

>>2776617
Uhmmm I also did a bit the other two sides of the socket opposite to each other

[Anonymous No.2776621]

>>2776619
What? You mean you soldered the other side of the connector lead on the other side of the board? If it was like that originally then that's what you're supposed to do.

[Anonymous No.2776623]

>>2776621
Yes, I have soldered the opposite side of the board which is also covered in aluminium looking cover (idk what is it for)
But the holes are of course fucked from either side

[Anonymous No.2776624]

>>2776559
>the firmware project I work on
>i dont know how to code
Kek, welcome to the club

>>2776569
If it’s within the voltage limit of the meter, there’s no problem.

>>2776579
If it’s a high-current supply, then you fundamentally need a lot of filter caps to prevent the filtered voltage dipping below the output voltage. Much less of an issue with 3-phase. The more ripple you have, the more headroom you need to get above your output voltage, and hence the more power your linear stage will dissipate. Chads would use a quasi-resonant switching stage with a linear-post regulator so the headroom wouldn’t matter. Your op-amps are probably being run off a different winding anyhow, so PSRR doesn’t really enter into it.

[Anonymous No.2776626]

>>2776623
If it was like that originally, then that's how's meant to be.
Again, get a multimeter and check continuity.

[Anonymous No.2776628]

>>2776626
Okay, I'll get one and return with results tomorrow or after.
Or earlier if there is something else wrong

[Anonymous No.2776637]

>>2776626
Other that that there is no sign of burns or melted or a smell of burnt components or black/grey shades anywhere

[Anonymous No.2776671]

>>2776576
if you just want to pre-program some blinking modes, and change the mode by pressing a button, I think the easiest option is to use a cheap chinese arduino nano.
if the arduino nano is too big, you could use an attiny85 board, but you would be more limited in what you can do (attiny has very few i/o terminals).
those leds draw a lot of current, so you will need extra circuitry to power them on. if you try to connect one of those leds directly to the arduino, you'll likely fry the board.
there are more things to consider, but you will likely have batteries + arduino + power circuitry + leds.
>I don't know much about arduino
>Not very experienced with electronics
before you do the real thing, maybe it's better to buy one of those cheap arduino kits and play with it a little. the first tutorial people always do is how to blink a led. you can implement all the functions you want on a breadboard, and later transfer it to the shield.

[Anonymous No.2776676]

>>2776624
>Your op-amps are probably being run off a different winding anyhow, so PSRR doesn’t really enter into it.
But what would be PSRR of the complete power supply ? Wouldn't it reject most of the input ripple thanks to the error amplifier and the feedback? Do the specs of the error OA have anything to do with that? For example PSRR of LM317 is about 60db. That sounds terrible. What would it take to make it say 100db? Because if it is 100db, there would be no need to reduce the input ripple too much which would lower the cost significantly.

[Anonymous No.2776686]

>>2776559
If you use an ESP32 instead of an STM32, you can use its native MAC capability with a MAC-to-ethernet IC, instead of relying on a more limited UART-to-ethernet IC. There are devboards with an ESP32 and this IC and the port itself onboard, but they don't support PoE. PoE is a cool feature to add.

Also I can't see a voltage regulator there. Can the MCU board handle 24V? Even if it can, the LDO might get kinda toasty.

>>2776576
I've heard of libraries like wled that are designed to control adressable LED strips, they might have good routines for controlling a few RGB LEDs but I'd probably just write the routine from scratch. Not sure the ideal way to drive the LEDs, for battery life I'd want current-regulated switching regulators, but you would need three of them to power the different colours, since they have different voltage drops. You'd want to pick LED drivers with PWM inputs for a micro.

As for the LEDs, I guess look on DigiKey or LCSC or wherever for LEDs in a similar form-factor that can handle high power. Ideally I'd make PCBs in the same shape as the one that came with the thing, that way you can reuse the mounts and the lenses.

>>2776676
>But what would be PSRR of the complete power supply
You can make it arbitrarily good by just having good filtration on the PSU itself. Put 78/79xx series regulators on your analogue power rails and you've already got 65dB on the filter caps for those rails, any halfway modern op-amp circuit running on that will add another 60dB at least. If the ripple on your filter caps is 2V, then the output ripple from that op-amp is like 2µV. It's not something you'd care about. The ripple rejection through the pass transistors is going to be far worse than that, the op-amps PSRR basically doesn't make a difference here. What would make a difference is going for op-amps with a high slew-rate and high gain, in order to make the negative feedback loop with the pass transistor as fast as possible without instability.

[Anonymous No.2776692]

>>2776579
> need big caps?
Yes, what you are saying is true to some degree, but if you increase the unloaded voltage of your unregulated output, the voltage will sag under load you need enough caps to ensure it the ripple doesn’t go under the regulator’s input threshold.
Also, increasing that input voltage requires the regulator burn-off a lot more power so it’s not as efficient. I’m talking about a linear regulator here, obviously.

[Anonymous No.2776693]

someone recommend me panel-mount waterproof connectors for at least three amps, with 2, 3, 5 pin counts available
size is somewhat a constraint

[Anonymous No.2776695]

>>2776693
What voltage/current rating?

[Anonymous No.2776704]

>>2776695
twelve volts, three amps

[Anonymous No.2776773]

>>2776686
> any halfway modern op-amp circuit running on that
What do you mean, an error amplifier or what kind of a building block? I know there's some additional circuitry to improve LM317 performance, something like "capacitance multipliers" and such.

>The ripple rejection through the pass transistors is going to be far worse than that
Just to clarify, I am talking about high current linear regulators, say 20A 500W. Pass transistors are a must.

> If the ripple on your filter caps is 2V, then the output ripple from that op-amp is like 2µV.
Yes, 120db will do that and that would be ideal.

>going for op-amps with a high slew-rate and high gain,
I simulated a few OAs and some oscillate like crazy on high ripple even in a simulation. Even a typical RC network in the feedback loop doesn't help much. I imagine a real circuit would perform even worse since simulations tend to give more optimistic results.

>>2776692
True but my point is the cost could end up being too high if the required output ripple has to be low. Imagine you need <100uV at a modest 5A and your RR is only 60dB. That would mean your input ripple should be <100mV. At 5A, that needs 200,000uF. Can you imagine how expensive that is? And what if you need 10A? That's just silly. You need >100dB for that. Sure less capacitance means more headroom needed and more ripple voltage to burn at high currents, but adding an extra pass transistor is cheap.

[Anonymous No.2776794]

Hello /ohm/. About to build a power amp with baxandall preamp and found this circuit in the internets.
>could I just replace those pots with 10k ones?

[Anonymous No.2776806]

>>2775738
If freecad devs lived command line you'd be able to "freecad file.fcstd" instead of "freecad $PWD/file.fcstd".

[Anonymous No.2776817]

>>2776794
>replace those pots with 10k ones?

you can change volume and balance
as for bass and treble, you'll change the frequency response, so they'll work in weirding ways

[Anonymous No.2776820]

>>2776817

actually, you cant change ANY pots without affecting frequency response
unless you also change the caps accordingly

[Anonymous No.2776824]

>>2776820
Hmm. Bummer.
The problem is that the amp has an internal 20k GND resistor and as far as I'm concerned would require 1/2 or less that resistance for the volume pot - so 10k I guess.
I just dont want to order a multitude of different pots.

[Anonymous No.2776841]

>>2776626
I'm back with a multimeter, check my replay in my main thread.
>>/g/99652821#p99661157

[Anonymous No.2776844]

>>2776626
Sorry I meant >>>/g/99652821

[Anonymous No.2776847]

>>2776617
But it appears that the port connecting to the battery itself has continuity between many parts of the motherboard, I guess it is supposed to be that way?

[Anonymous No.2776852]

>>2776617
Uhmmm I checked again and it appears there's continuity between the top cover of the new connector and everywhere else on the motherboard

[Anonymous No.2776892]

>>2776693
>>2776704
12 V 3 A
Maybe those connectors for truck trailer hitches.
I was using these plugs I got from some 12 V outdoor lights (I salvaged them) Picrel is very common here for that.

[Anonymous No.2776893]

I've managed with a cheap no name soldering iron thus far. No temperature setting or anything special, now the tip needs to be replaced but I don't know what size/format to buy.

Do I bother looking it up, or do I just buy a complete soldering station (or whatever's it's called in English)? I've only soldered through hole components so far, but I want to get into SMDs eventually. What's a good brand known for their quality and accessories?

[Anonymous No.2776897]

>>2776773
> less than 100uV at 10 A.
Like, isn’t your biggest problem here going to be the massive wires you’ll need?

If you look at a GPU card, it’s basically just a big GPU chip surrounded closely by a dozen or so 12V1.1V DC/DC switching voltage regulators and a “smooth” brand controller chip that you can never find the exact datasheet for.

> pass transistor
Like picrel?

[Anonymous No.2776900]

Are these long hinge microswitches for coin acceptors meant to be bent into the right position for the mechanism you're working with? I'm trying to build my own arcade coin box.

[Anonymous No.2776901]

I plan to mount one of these switches in a coffee can and connect it to a USB encoder to make the coin slot work. I don't need any actual coin mechanism to differentiate coins, just a switch. So I plan on bending the hinge into a rectangle so the coin can reliable trigger it as it falls down.

[Anonymous No.2776904]

>>2776773
>some additional circuitry to improve LM317 performance
Look, if you're making any kind of linear power supply capable of more than 1A, you're best off ditching the shitty LM317 and making your own control loop of an op-amp controlling pass transistors. An LM317 is just a shitty op-amp, shitty voltage reference, and shitty pass transistor, combined with a nice to have thermal overload prevention circuit.
>what kind of a building block
The op-amp topology would probably be a differential amplifier (op-amp plus 4 resistors) that compares your reference to the output voltage, after some division to scale things down. It's best to put the potentiometer on the reference, since that allows you to tune the output voltage all the way to 0. The open loop gain is usually best being around 100 or so, but I'd want to make it higher and just curb off the gain at high frequencies to keep it stable. The control theory chads here would be busting out the equivalent capacitance models for BJTs to ensure their system is as stable as reasonably possible, but I'd just slap capacitance on the loop until it stops outputting an AM radio signal.

So the rough topology would be as follows:
>multi winding transformer with power and analogue windings
>analogue winding goes to rectifier, moderate capacitors, and 7800/7900 regulators that powers the op-amp
>power winding goes to rectifier, bit capacitors, and your pass transistors
>op-amp controls the pass transistors
So PSRR in the op-amp isn't going to matter really, much more of the ripple will be getting in through the transistor collector than the transistor base. As for minimising this, a circuit like a capacitance multiplier is a good start. And the higher gain your op-amp has at 120Hz and its harmonics, the better it will be able to counteract the ripple.
Adding thermal protection to the pass transistors is a good idea too.

Also if I were going to make a 5A+ linear power supply, I'd use a multitap transformer with relays.

[Anonymous No.2776915]

>>2776576
There are high power RGB LEDs. You could just control them using any MCU (controlling a tranny ofc)

[Anonymous No.2776923]

>>2776897
>Like, isn’t your biggest problem here going to be the massive wires you’ll need?
Good point, that 100uV is a random value I could think of just to illustrate the point. But it only makes sense for low currents. I agree the the total loss in the internal / external wiring / connectors could easily be 100-1000 times higher.

[Anonymous No.2776933]

>>2776923
Some high current power supplies use something called remote sensing: they probe the voltage at the load and correct for the voltage drop. That way they correct for both internal and external losses. So there is no reason you can't make a high current power supply super accurate down to microvolts if you calibrate it and compensate for the wire losses. The question is, do you actually need that much accuracy at such high currents?

[Anonymous No.2776940]

>>2776904
> An LM317 is just a shitty op-amp, shitty voltage…
Which reminds me, the adjustable voltage regulators have lower noise than the fixed 78xx versions. So they’re slightly less shitty in that regard.

And when you’re talking about 100db of RR then you might as well start crawling into the whole noise floor hole.

[Anonymous No.2776947]

>>2776852
where do you mean by everywhere else on the board? it could just be grounded

[Anonymous No.2776953]

Is this a waste of money? I see other bench power supplies are 2x as much but look almost identical.
I know i can just make one with an old PSU off a computer and a buck converter, but it'll look ugly as fuck and prob be a risk of shorting out.

>https://www.amazon.ca/Variable-Adjustable-Switching-Regulated-Adjustments/dp/B09DKFFBG5

[Anonymous No.2776954]

>>2776901
Sounds good.
> can’t differentiate coins
Let us know when we can come over and try it. I’ve been an electrician for 35 years and I saved every punch-out slug from every job hoping they’d come in handy one day. Today is that day. Here’s ones I had in my pocket from 20 minutes ago.

[Anonymous No.2776967]

>>2776953
> waste of money?
Not necessarily. I know it looks like it’s going to be the size of your gaming PC tower running 2 4090s, but believe me it’s gonna be a lot smaller in person.

It’s probably pretty noisy, probably an adjustable buck converter most likely.

I usually like (and use) linear fixed output supplies, but the variable ones are fine for things like manual battery recovery, electroplating, figuring out the minimum current /voltage rating of unknown LEDs and unlabelled components.

The ideal set-up (in my opinion) is an isolation transformer, a variac, a beefy raw AC to DC unit, (possibly with active rectifiers nowadays) and smoothing caps, then some linear regulators or buck/boost modules on the ends depending on what I need.
Remember modular Hi-fi systems? Like that.
Then when any one component blows up, you just have to fix that one unit, but it’s rare because they’re usually higher quality individually.

I remember my friend thought he got this great deal on a “compnent” stereo system, and we came over to check it out, it was fake! all attached but the front was molded to look like separate components. Even had fake vibration dampeners like you find on the phono, but it didn’t even have the record player on top, so that was seriously an embarrassment.

[Anonymous No.2776982]

>>2776893
I use a cheapo no name brand with no temperature control but I only do THT. You buy tips by their power rating (the higher the power the thicker the tip), at least in here.
I don't really see the need for me to get a soldering station, it would be nice to have but I can find better things to use my money on. SMD is a bit trickier and better temperature control is very helpful for those, specially so you don't fuck up sensitive components or the traces and the other components around so it might be worth it for you.
There's also different shapes of tips, including ones that are very pointy and thin and might help with SMD stuff. I can't find those special tips readily available on stores here, but I can always order them online. Might be useful for you as well.

[Anonymous No.2776993]

>>2776893
Advice i see everywhere is unless you already have a T12 iron(or T85/T80/T100... the smaller kind with the self contained ceramic tips) or better, you should just put your cash into upgrading to that system over the traditional stuff. If you buy a direct from china one from aliexpress, you can probably get a station with a few tips for under $40usd. Might not include the PSU though

[Anonymous No.2776999]

>>2776953
>other bench power supplies are 2x as much but look almost identical.

40 canadian pesos is way too cheap for 30V 5A supply
disappointment is guaranteed
son

[Anonymous No.2777011]

>>2776953
What do you want to use it for? For powering normal circuits, fixed voltages are usually fine. For load testing, a cheap CC/CV buck converter on a power brick is fine. For audio, get a transformer and make a split-rail supply.

If that one you linked to actually has thermal protections and other desirable features it might be worth going for, but it isn’t going to be a good low-noise supply for audio or instrumentation.

>>2776999
It’s a switcher, not a linear supply with relays.

[Anonymous No.2777035]

>>2776999
Justin Castreau switched to pesos to honor his dad? Based.

[Anonymous No.2777043]

>>2776904
Just use a μA723 like everyone else that needs a precision voltage regulator. It has better ripple rejection among other things. Why try and re-invent the wheel???

[Anonymous No.2777149]

>>2776794
How do I determine the correct volume resistance?
I heard that the volume pot should have 1/2 the input ground resistor, so 10k in this case. Does this make sense for this preamp?
I'd just prefer to stick to 100k for all the pots on the board.

[Anonymous No.2777176]

>>2777149
I think they want you to feed it with a TDA1524 volume control. It has a fixed gain of like 30 similar to the LM386 which is the most popular amp in the world by far and has a gain of 20.
100K or 1M pots are good to have, because you can convert them to any value you want.

[Anonymous No.2777182]

>>2777149
If the load is less than 10x the volume pot, the resulting voltage divider (look this up) will never drop below 9%.
If you set it at 1/2 the ground resistor, you'll never get less than 33% volume. Your amp will always be on.

[Anonymous No.2777198]

>>2776982
>>2776993
Appreciate your input!

[Anonymous No.2777232]

>>2777149
>How do I determine

any reasonable value pot will work, from 1K - 500K, with logarithmic taper
because of the 20K input resistors, higher value pots will deviate the log curve somewhat
but who is even gonna notice?
no one unless they have two sets of ears.

[Anonymous No.2777292]

>>2776947
Uhmmm nevermind, it seems like two leads of the new connector are supposed to touch the ground, the other two have continuity with the far away battery connector, there is no continuity with the ground.

But I am worried about short circuit near the CPU.
Pic related I checked (what seems like inductors?) R22, and R47 and for some reason the multimeter shows continuity as well as near 0 ohms with the ground, is this a short circuit?

[Anonymous No.2777294]

>>2777292
*A potential short circuit, because in comparison this other part of the motherboard (pic related) has no continuity with the ground, shows about 50 ohms.

[Anonymous No.2777296]

>>2777292
>>2777294
Throw it in the garbage and forget it, or use the board for practice. It's toast. That thermal paste is conductive and it's all over the place because you used way too much on the CPU, and failed to clean it before powering the board. The yellow shit on the solder joints is roasted flux. It's over, Johnny.

[Anonymous No.2777297]

>>2777296
The thermal paste was all over the place even before the laptop stopped working, you are a retarded mentally ill tranny

[Anonymous No.2777299]

>>2777297
You tranny'd your board and I'm the retard?
Good luck, kiddo. Maybe /g/ could help. lmao

[Anonymous No.2777311]

>>2777299
But it still works lol, you are a mentally ill schizophrenic tranny on some Mongolian forsaken image board

[Anonymous No.2777314]

>>2777296
I literally didn't use any thermal paste, you are a mentally ill schizophrenic 4chan tranny that makes stuff up about what I do or don't do

[Anonymous No.2777315]

>>2777311
>>2777314
Okay. Keep soldering. It should work eventually.
AAAAAHAHAHA

[Anonymous No.2777324]

>>2777315
But I already put the connector in it's place, why are you saying it's wrong when it's working?
You are a mentally ill schizophrenic incel on 4chan with no degree and skills and no education while I literally took some classes and know more than you in electronics?
Lol who are you know your place schizo

[Anonymous No.2777326]

>>2777324
>It's working
Why are you here again? High self-confidence?
lol

[Anonymous No.2777329]

>>2777324
>AAAAAAA help me!
>it's working
>but I need help
>I know more than you
You seem confused. Strange cope. Probably a tranny yourself.

[Anonymous No.2777330]

>>2777329
and underage at that

[Anonymous No.2777334]

>>2777329
I didn't ask for your help, you are some random mentally ill schizophrenic tranny that assume I put thermal paste somewhere on the board.
Are you underage?

[Anonymous No.2777335]

>>2777326
I asked different questions, the charging might work but I still ask about continuity somewhere else on the board duhh.
And again I haven't asked you any questions you delusional subhuman shitskins.
You are not competent enough, you are not the other anon I asked questions.

[Anonymous No.2777337]

In particular I asked this anon
>>2776947 a question about continuity since he's clearly not an underage college student or some subhuman shitskin or tranny like
>>2776947

[Anonymous No.2777338]

>>2777324
Take some soldering classes and thermal paste classes while you're at it.

[Anonymous No.2777339]

>>2777338
I haven't applied any thermal paste, it was literally like that when I opened it.

You are a mentally ill subhuman schizophrenic tranny, shitskin or an underage college student that makes wrong assumptions about things.

I haven't asked you any questions because you are an incompetent schizo incel on 4chan who has no university educatuon

[Anonymous No.2777341]

>asks for help
>calls everyone mentally ill tranny and whatever
>is a complete retard
Get back to /pol/ and complain about your fried consumer trash there.

[Anonymous No.2777342]

>>2777339
Why are you here asking questions? Is it because you know everything, or is it because you're an absolute retard who fucked up your computer like a toddler?

[Anonymous No.2777353]

>>2777341
>>2777342
But I repaired it, you are a mentally ill subhuman tranny, a shitskin who thought he was smarter than a white men.
Or you are an underage retarded college student who always makes assumptions.
Which one are you anon?

[Anonymous No.2777358]

>>2777353
Underage college student? You -are- retarded. Take some more classes, retard.

[Anonymous No.2777375]

>>2777353
>But I repaired it
Post a pic of it working then.

[Anonymous No.2777377]

>>2777375
with a timestamp

[Anonymous No.2777409]

wow board tourists are fucking cancer

[Anonymous No.2777488]

what the holes should look like.

[Anonymous No.2777489]

>>2777488
his desolder job.

[Anonymous No.2777492]

what the connector in holes should look like.

[Anonymous No.2777493]

>>2777492
his board with connector in place.

[Anonymous No.2777495]

what connector pads should look like on other side.

[Anonymous No.2777499]

>>2777489
>>2777493
I'd hire him just to fire him. Imagine his 12-year-old ego getting crushed. Smol pp energy.

[Anonymous No.2777512]

Would any anons mind helping me with comprehending an unusual soldering problem? I'm experiencing a very strange phenomenon when attempting to solder SK6812 LEDs onto a PCB, and I cannot for the life of me understand why it's happening. For some reason, although I can apply solder to the pads on my PCB, I cannot at all connect to the metal pads on my LEDs. For reference, I've gone from 220C up to 300C (in case temps were a problem), but solder doesn't connect to the LED pads at all, and thus not bridge to the PCB pads either.
>Pic related, weird shit.
Even in isolation with some sacrificial LEDs, it just won't stick. This is quality Kester solder wire I've used before, as well as on this exact same board when attaching diodes using the same iron, and no such issues occurred. It absolutely baffles me; even in isolation, even if I just totally goop a ball of solder on my iron, I can't get solder to stick on the pads. It's like there's an invisible barrier that no amount of flux can fix. These are LEDs from a roll I've used in the past, and they flowed fine in a previous project. Very confused; I'd appreciate any ideas as to why this is happening.

[Anonymous No.2777515]

>>2777512
Try cleaning the pads with a pencil eraser or fiberglass pen first.

[Anonymous No.2777519]

>>2777512
more flux
>through-mounting SMD LEDs
freaky

[Anonymous No.2777531]

>>2777515
>Try cleaning the pads with a pencil eraser or fiberglass pen first.
I gave it a shot with your advice just earlier with multiple LEDs, but nothing changed. I swear, it's like there's some magic bullshit barrier preventing the pads from being adhered to. It isn't like these were dirty loose ones either; they came in a roll with plastic sheet lining covering them from debris and damage. Whether trying to bridge or do them in isolation, nothing changes. I have no idea how I made it work in the past, but I never had any issues with these like this before. The absolute best I could do before is pic related with a sacrificial LED for testing (like rolling over with a blob of solder).
>>2777519
>more flux
Already tried it. The problem is it won't adhere to the metal AT ALL, or if I'm lucky, juuuust barely. It's so inconsistent and impenetrable, it's baffling.

[Anonymous No.2777533]

>>2777531
Do you have a hot air station? If not you could put a thin layer of sand in a skillet and put your board on top of the sand, but you should use medium temp solder paste for that. The sand buffers the heat so it's even across the board. Those LEDs are pretty sensitive to heat, so keep the dwell time short. BTW, you aren't the only one having a hard time with those.

[Anonymous No.2777538]

>>2777533
I don't have any of those on hand, but I sincerely don't think the issue is evenness of temps. I've done this same board design before, using LEDs from the same (clean) roll, the same solder wire, the same iron, everything the same, but I can't get this shit to flow this time, no matter what temp or flux I use. However, it's been a few years. All I can imagine is that something happened to the LEDs in storage, like getting some sort of translucent film, but I don't get how, nor see anything like that, so all I can gather is this extremely strange behavior. At the very least, I'd think that hot solder would bind to bare metal like all the other joints I already made with the diodes.

[Anonymous No.2777543]

>>2777538
I'm just as baffled as you are, Anon. Maybe some fine-grit sandpaper or a magic eraser will expose a wettable surface. They could be defective from the manufacturer too.

[Anonymous No.2777546]

>>2777543
>Maybe some fine-grit sandpaper or a magic eraser will expose a wettable surface.
I'll have to try that later. I tried scraping at the pads earlier with some tweezers, but I think they didn't seem to make a dent since they're too smooth.
>They could be defective from the manufacturer too.
See, I'd totally agree with you that this could be the case...if it was only a few, and wasn't from a roll I'd used (quite successfully) before. Oh well...I'll keep trying. I appreciate you spending your time to help me diagnose this. I'll get at it again with your idea either tonight or tomorrow and reply to you with my results if you're interested.

[Anonymous No.2777548]

>>2777546
>I appreciate you spending your time to help me diagnose this
No problem at all.
>reply to you with my results if you're interested
I am interested because this is going to happen to others and it would be good to know one way or the other. Good luck, Anon.

[Anonymous No.2777555]

>>2777493
Did you ban evade?

[Anonymous No.2777557]

>>2777555
checkin those digits

[Anonymous No.2777601]

>>2777077
NEW
>>2777077
THREAD
>>2777077
PHONEPOSTER
>>2777077
RETARD
>>2777077
GET
>>2777077

please no trannyposting this time ok

[Anonymous No.2777877]

>>2777601
You made the thread too early again you stupid fucking nigger tourist fag.

[Anonymous No.2777953]

>>2777877
I'm not the new thread OP, I'm this thread's OP. The new thread's OP is a fag who didn't even post a link in the old thread, didn't put a "previous thread" link, deleted the "bake at page 8-10, post in old thread" line, and posted a sideways image.

[Anonymous No.2777962]

>>2777953
My apologies. The next time this happens the person responsible needs to be banned for at least 30 days. It's a fair punishment for fucking up OUR /ohm/. This thread could stay up for another 5 days.

[Anonymous No.2778220]

i need to do some SMD soldering and only have a simple iron, and know from past excursions that trying to desolder SMD with a single iron will just lead to ripped pads and many tears shed.
i was looking into soldering tweezers but ran into two problems.
first, i have a weller WS 81 station and the compatible tweezers cost about as much as another one of those stations costs new. some $250 to $350 depending on where you look, which is a little too much for me.
second problem is that i read that soldering tweezers arent even a good tool for SMD and i should just get a hot air station.
what should i do? some cheap tweezers or a hot air station?

[Anonymous No.2778355]

>>2778220
Nobody in the industry (well, almost nobody, gullible idiots are everywhere) uses tweezers.
Get a cheap mini hot air gun from zhang. They’re all basically the same, and realistic temperature controlled soldering irons, while a hilarious concept, is absolutely absurd on a hot air gun.