/ohm/ - Electronics General: Hermetically Sealed Silicon Edition

[Anonymous No.2457597]

NOT(Thread):>>2448754

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

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

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

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

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

>Books:
https://libgen.rs/

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

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

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

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

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

[Anonymous No.2457635]

first

[Anonymous No.2457638]

>>2457635
zeroeth

[Anonymous No.2457691]

Anons could this work?

>pic related

I have a idea for a project that default state can be used to power devices through banana plug jacks, like a benchtop power supply.

I also wanted to integrate a simple multimeter, and thought of using a DPDT switch to go from output 5-12v whatever is needed, and then flip of a switch my banana plug jacks can be used with probes to get a reading, on a device or whatever.

This is just basic concept, I plan to inline a fuse, and a diode for reverse polarity protection.

[Anonymous No.2457693]

>>2457691
what the fuck are you talking about
what does the potentiometer do
and you want a panel meter, not a multimeter

research simple benchtop power supply design first

[Anonymous No.2457695]

>>2457691
Do you want something like this?
https://hackaday.com/2013/07/04/dip-switch-adjusted-voltage-regulator/

[Anonymous No.2457716]

>>2457693
"brightness" is in text above the potentiometer, those things are bright and want to adjust brightness on them.

[Anonymous No.2457724]

>>2457693
How hard is it to understand the concept of having banana plug jack type being able to output voltage to power a device, and flip of a switch I then can measure voltage on a powered device have the voltage display on voltage meter like pic related.

You're annoying as fuck to assume that you never consider yourself to be a problem, its always someone else, I bet you think highly of yourself while you look for the worse in other people. wew lad.

>>2457695
>https://hackaday.com/2013/07/04/dip-switch-adjusted-voltage-regulator/

that could work, I have 18650 I salvaged form a laptop, and have some BMS and was planning to make battery pack or a few from what I salvaged.

>pic related

is something I already had in mind to display voltage, just something simply nothing fancy.

[Anonymous No.2457749]

>>2457716
Yeah but if the panel meter just has Vcc, GND, and V_sense inputs, trying to adjust the LCD brightness by putting a series resistance into the panel meter wouldn't be very advisable at all. Even if it just uses a resistor to power an LED backlight, I wouldn't trust that circuit to get much change in brightness before the voltage dips enough to influence the ADC's measurements.

>>2457724
So you just want it to be fixed voltage? Well why not have two banana outputs for 5V and 12V, instead of a switch between the two? You'll also probably want current measurement, so buy a panel meter with both voltage and current, there may even be ones with a trimpot for backlight brightness, or ones that are easy enough to interdict the current going to the backlight. Voltage won't be too handy for a fixed output supply anyhow, so maybe just current is all you want, but if it's coming from a battery I can see why you'd want it. Though I can't really see why you'd want to run a benchtop power supply off a battery in the first place.
Maybe use a latching comparator and MOSFET for overcurrent protection, instead of a fuse.

Next time explain yourself properly instead of using word salad like "a project that default state can be used". The switch in the pic covered in "load" and "bat" labels is throwing me off, same for the arbitrary use of colours and the fact there's only one voltage output shown in the diagram. To say nothing of the fact your switch does literally nothing in the way you've drawn it. Regardless of what position it's in, the battery goes out to the bananas and the panel meter. Please draw an actual circuit diagram next time (be it gimp, falstad, ltspice, or kicad), instead of using google images.

[Anonymous No.2457790]

>>2457749
Ah fuck it all, ill go do what I want, it was a mistake to ask in this thread anyways.

[Anonymous No.2457888]

>>2457749
wait no it's the output and voltmeter pins that are always connected together through switch contacts (why not just wire them together) while the battery can be disconnected from them
if the aim is to connect a 5V supply to the other side, why not just use an SPDT?

[Anonymous No.2457961]

>>2457888


okay so ill see if I can explain what I want.

when the switch is flipped in one direction the battery will be outputting x volts to banana female jack, and to the yellow wire, this is so I know how many volts the battery is outputting when I flip the switch the other direction, it cuts power to the jack, and the power source for the yellow wire will be whatever im measuring with probes connected to the banana jack connector.

This was an idea I had, and honestly im just going to try it out regardless whats been advised to me anyways.

[Anonymous No.2457963]

>>2457961
>im just going to try it out
please post result

[Anonymous No.2458095]

What's up h/ohm/ies. For a project I'm doing I need to create a fucking strong electric field between 2 capacitor plates, like 1000's of Volts type of strong. I heard there are ways to do this with dead batteries or something? Anyone can tell me what I need to look into to do this or other easy ways to do this? Thanks guys

[Anonymous No.2458099]

What do you think of my portable Class B amplifier?

[Anonymous No.2458107]

>>2458095
whatcha makin?

[Anonymous No.2458130]

>>2458099
Guyse?

[Anonymous No.2458136]

>>2458099
The buttom switch is for choosing between line in and bluetooth, the top one for battery powered and DC plug operation. Works with any regular ~19v power brick.

[Anonymous No.2458172]

>>2458099
nice box

[Anonymous No.2458177]

>>2458099
My favorite part is the green and yellow tape. The hot glue accents are a nice touch. You should put a coffee warmer on the top of the case, coupled to the heatsink.

[Anonymous No.2458183]

>>2458107

[Anonymous No.2458199]

>>2458172
Thx. :^)
>>2458177
Is that supposed to be a compliment?
There's enough space between so the battery won't heat up too much.
I didn't have proper heat shrink for the batteries. The tape works fine.
When you're working with such limited space there's not much room for aesthetics, but I'm open for criticism to male improvements to the design.

[Anonymous No.2458205]

>>2458099
How much power? Output impedance? Is it short circuit protected or safe? True class B or is it AB?
Why class B and not AB or D?

[Anonymous No.2458209]

>>2458095
Electric fly swatter. Replace internal rectifier with walton-cockcroft multiplier which then feeds a Marx-Generator

[Anonymous No.2458212]

>>2458199
>but I'm open for criticism
I was just busting your chops, Anon. If it does what you intended, then it doesn't need any improvements. Fuck it. Everything is gonna turn to dust anyway. lmao

[Anonymous No.2458216]

what is the most accurate circuit simulator right now?

[Anonymous No.2458235]

>>2458216
I'd say LTspice, but spice sims are only as accurate as your models. Try Multisim or Microcap or whatever also.

[Anonymous No.2458246]

This is my box
There are many like it but this one is mine
thank goodness Leonardo Arduino can be read as a joystick because I have no idea how to work with chips

[Anonymous No.2458247]

>>2458205
>power, impedance
it *can* deliver 10w at 2 Ohm. Though I am using an 8 Ohm speaker, so I suppose it's something around 6W.
>short ciercuit protected
You mean the speaker output? I suppose the chip has some kind of protection. It also says in the datasheet that it will shut off when overheating.
>true class B or AB
I don't really know all tha much about those differences and amplifier designs in general, but I think it is class B. The IC is a TDA 2002.
>Why class B
I had this linear amp chip at hand and wanted to build something simple on breadboard by myself. Class B also is supposed to have a particularly accurate sound reproduction and indeed I think it sounds better than that cheap class D amp board I had connected to the same speaker in the past.
I can live with the lower efficiency.
Class D is just kinda boring I think, even if it sounded as good as B or AB ir whatever.

[Anonymous No.2458250]

>>2458099
Had it playing music the whole evening at max.
The only issue I had is the voltage regulator used for the bluetooth mudule.
20 to 5 is just too inefficient. The thing just makes for an unstable voltage supply when it gets hot and thus resets the bt module it alsp draws more power than nessecary.
A buck converter would be best but I don't know of any in that small size.

[Anonymous No.2458252]

>>2458212
>everything is gonna turn to dust anyway
right. There's no point in torturing oneself with overt perfectionism

[Anonymous No.2458257]

>>2458235
im looking for one that can emulate real world op amp accurately(offset voltage and all that error stuff)

[Anonymous No.2458261]

>>2457597
Wtf is that thing?

[Anonymous No.2458264]

>>2458261
exactly one pringle

[Anonymous No.2458266]

>>2458261
That was Rasputin's calculator.

[Anonymous No.2458270]

>>2457963
Ah yes, you hope I fail, maybe I will, this is concept ideas anyways.

[Anonymous No.2458277]

>>2458270
I don't hope you fail. I hope you learn.
I'm a nice guy- really...
unless you cross certain lines, then I'll hang you with your own intestines.

[Anonymous No.2458347]

>>2458277
>unless you cross certain lines, then I'll hang you with your own intestines.

Oh, im cringing, thats too try hard, and middle school vibes.

[Anonymous No.2458403]

>>2458347
We’re all chuunis here, anon-niga!

[Anonymous No.2458416]

>>2458247
AB is more linear than B. Much less zero-crossing distortion, which is the main issue with Bs.

Making a self-oscillating class-D without a dedicated IC makes for a fun project, and should have a pretty linear output compared to class-Ds with dedicated triangle oscillators and no feedback. Pic related is a simplified circuit.

>>2458257
Pretty sure you can download spice models for particular operational amplifiers from the manufacturer's website that include all that stuff.

[Anonymous No.2458429]

How can I make voltage regulators more efficient?

[Anonymous No.2458433]

>>2458347
Oh fuck me that was some top level cringe, thanks for pointing it out. That really put me in a sour mood.

[Anonymous No.2458436]

>>2458429
What's the problem in your case? Too much amps being pulled or too much input voltage?

[Anonymous No.2458445]

>>2458429
Replace it with a switching regulator.
If you're amplifier-anon, there are ways of connecting BJTs to a linear regulator in order to spread the heat-loss out, but you're probably better off using a switching reg.

[Anonymous No.2458446]

>>2458107
Just what I was describing. I read some research online where it was claimed that if you have seeds and put it in a strong electric field for a few days and then plant it it will actually be a prehistoric version of the plant. I want to try this out myself.

>>2458209
Thanks alot for this. I'm gonna see if I can build this

[Anonymous No.2458463]

>>2458436
Too much input voltage - 20 to 5V
>>2458445
Yes, I'm that guy.
I don't have such a small switching regulator at hand. Could you pls post a diagram for that bjt method?

Also the lm25965 CC module is only supplying 500mA at 19v. It's lso getting very hot. Is there a better method to charge li-ion batteries?
I don't thi k it will survive for very long like that.

[Anonymous No.2458465]

>>2458433
>crige
go back to instagramm

[Anonymous No.2458477]

can you do TIG without the gas?

[Anonymous No.2458479]

>>2458463
>I don't have such a small switching regulator at hand
You can buy them, but I can understand if you want to just use parts on-hand. Pic related is an LM317 + three BJTs for current sharing. They're a bit shitty, not very good for widely varying current ranges. Doing a similar thing with an op-amp may be advisable instead, you lose out on the temperature regulation of the LM317 but you gain the ability to have better feedback loops. You could maybe even use LM317s or other regulators as pass transistors, considering your high dropout, and use their thermal regulation to your advantage.

Also that looks like a handy inductance multimeter.

>>2458477
Are you welding in space? Or an area that happens to be drenched in noble gas? If not, no. If you want to ditch the gas, go flux stick welding or whatever.

[Anonymous No.2458480]

>>2458479
ill stick to regular flux things thanks

[Anonymous No.2458482]

>>2458463
https://sound-au.com/project102.htm
always keep in mind, everyone on here is stupid

[Anonymous No.2458506]

>>2458347
>try hard
It's easier than you think.
>>2458433
>sour mood
That's all it took? lmao

[Anonymous No.2458507]

>>2458433
Go back to /adv/

[Anonymous No.2458509]

Sad to think that boards like this are the only thing left from the old internet

[Anonymous No.2458511]

>>2458509
>boards like this
das rayciss

[Anonymous No.2458521]

>>2458509
What, back when only uber-autists were dedicated enough to actually use the internet? Anywhere with enough autism should feel pretty similar. Places like r9k are autistic but also just kinda nihilistic incels, so you need to find someplace with enough of a refined, positive vibe.

That’s what I think anyhow.

[Anonymous No.2458536]

>>2458247
It the amp can put out 10w at 2 ohm, its probably 2.5 watt at 8 ohm.

I like class D projects myself. Currently trying to get a class D VCA to have some analog (class A) sounding mojo going on. What i have on my breadboard sounds like an old radio. (Which is kind of what im after.

[Anonymous No.2458548]


DP to DVI converter
/
DP Splitter - DP to DVI converter
/
DP input - DP Splitter
\
DP Splitter - DP to VGA converter

Generally speaking could this work as long as i keep every screen to a 2k level, or will DP protocol slowly die at the second split?

[Anonymous No.2458551]

Generally speaking could pic related work as long as i keep every screen to a 2k level, or will the DP protocol slowly die at the second split?

[Anonymous No.2458677]

What are some ways to make a living working on electronics without having an engineering degree?

[Anonymous No.2458721]

>>2458677
I'm a cook who likes to burn myself with a soldering iron on the weekends. I earn money by selling colored fentanyl tablets to schoolchildren.

[Anonymous No.2458770]

>>2458677
Either repair people’s hifi setups or radios, or make something you can’t buy from AliExpress that’s in-demand by guitar effects guys, by arduino guys, or by arts majors. Or selling your ability to create prototypes on fiverr. Having a 3D printer helps in any case, be it for replacing stripped gears or making housings for new projects.

[Anonymous No.2458780]

I have some dc motors that can run anywhere from 5-12v (increasing voltage increases rpms).
Should I control the rpm's via voltage or PWM?

[Anonymous No.2458793]

it works, all that is left is putting it in a box

[Anonymous No.2458828]

>>2458793
dont lie, anon, you will keep it on the breadboard, and let it collect dust.

[Anonymous No.2458834]

>>2458793
First time Viagra user?

[Anonymous No.2458835]

>>2458780
PWM is just easier. Don't need all the inductors and feedback from a variable voltage converter, can just run it directly off a 12V source with a single FET and diode.

[Anonymous No.2458838]

>>2458835
gotcha, I figured that was the route, thank you for the insight

[Anonymous No.2458853]

>>2458770
>>2458677
>make something you can’t buy from AliExpress that’s in-demand by guitar effects guys
don't make yet another tube screamer clone unless you can chuck it out faster and cheaper than the chinks (protip: you can't; they're wizards and you can't beat them). Unless you can market it well, eg. "ultra hi-fi, low noise, noiseless, transparent, colorful, smooth, creamy, has character and bite, includes magical, rare component that makes tone go to 11, gold plated contacts, fairy dust-infused pcb, has real tubes in it, ideal for literally every style of music on earth, [famous guitarist] loves it and uses it always all the time even when it's off, true bypass."

[Anonymous No.2458855]

>>2458770
>>2458853
I don't want to make things for mentally ill boomer audiophiles. Isn't there something else you can do with electronics that's profitable?

[Anonymous No.2458860]

>>2458855
Make a ewetoob channel and review a bunch of trash for the consoomer market like all the other fags.
Or be cool like Ben Hanks and make vidya game controllers for people with disabilities.

[Anonymous No.2458878]

>>2458834
Lmao got em

[Anonymous No.2458882]

>>2458853
So long as you get enough publicity before the chinks copy you, you should be able to retain a decent revenue stream just from your reputation alone. Tube screamers are too saturated though, you should aim for something more novel. Still haven't seen a DSP pedal for less than like $40, as you could definitely do with an ESP32.

>>2458855
Hifi systems aren't just for boomer audiphiles, anyone with a stereo they've hooked up to do sound for their television is a potential customer, especially for older stereos that might be starting to show their age. Fixing up broken audio equipment from thrift stores to sell may work too. Audio and radio stuff from the 80s and 90s is going to be the easiest to repair while still being reasonably relevant today, most modern electronics aren't going to be easy to repair at all, unless you're good at microsoldering.

Pretty sure youtub guy Great Scott designed the HackRF One to sell on his online store, maybe you can do the same if you get enough of a reputation doing something.

[Anonymous No.2458902]

>>2458780
>the rpm's via voltage or PWM?

PWM, but not because it's easier, but because it lets you maintain high torque at any speed.
if you just drop operating voltage, it becomes pathetically weak at low speeds.

[Anonymous No.2458905]

>>2458902
So the torque/speed curve for varying duty-cycles does not match the torque/speed curve for varying voltages? Interesting, never knew that. I guess the difference in instantaneous current makes a difference.

[Anonymous No.2458911]

>>2458905
>difference in instantaneous current makes a difference.

sure, it's the difference between being hit occasionally with a hammer, rather than hit constantly with a chopstick.

[Anonymous No.2458933]

>>2458902
thank you so much for this answer

[Anonymous No.2458973]

>>2458933
First and foremost, I'd like to thank my Lord and Savior Jesus Christ. Second, I'd like to thank the academy and Harvey Weinstein.

[Anonymous No.2459080]

>>2457597
any of you guys think I could wire up this old swede military surplus lamp for 120v US outlet? I couldn't find much online about it besides a few old expired ads. It has a diagram on the lid of it running off of batteries in series(?), and the cord ends in a single ended plug. Also, the lamp has "NIFE" on the back. I couldn't find anything on that for a brand so maybe it's Nickel Iron?
What do you guys think? I figured I would ask here before cutting anything.

[Anonymous No.2459084]

>>2459080

[Anonymous No.2459124]

>>2459080
NIFE is probably an old defunct manufacturer. As for powering it, does it have a socket of some sort for a bulb? If so, and if it's a socket you can actually get bulbs from, then yeah go ahead and wire it up for a 120V outlet. I think it could look cooler if you kept the original plug on it, but considering it doesn't look safe for mains you'd better ditch it. Also check with a beeper to be sure if the outer metal of the lamp is connected to the negative/0V/shield of the connector, if so you'll have to disconnect that so it isn't at neutral, or optimally, connect it to GND with a three-conductor cable.

[Anonymous No.2459132]

>>2459124
>does it have a socket of some sort for a bulb?
Yeah, theyre these little phillips brand guys 6v 35w.
in pic related, the socket on the right is built into the box it came in. Maybe I could wire a 3 conductor cable and switch to that to retain the original male plug?

>Also check with a beeper to be sure if the outer metal of the lamp is connected
Electrical isn't my specialty, and this has me a bit confused. What exactly does that mean? I have wired up a socket before, but nothing this old.
I have a multimeter

[Anonymous No.2459134]

>>2459132
Oh, I might not have been clear. The socket on the right is not for the bulb. It's for the plug in the other pics

[Anonymous No.2459141]

>>2459134
>phillips brand guys 6v 35w.

that bulb base is type BA20s; replacements are available in 12V and 24V for trucks and such.
so if you wanna run it at 120V, you'll need a 50W transformer.
about 20 years ago, they made a lot of 12V 50W halogen table lamps with built-in transformer, so maybe look for a used one.

[Anonymous No.2459144]

>>2458853
I don't think that aiming to make money off of effects pedals is realsitic to begin with.

I mainly build weird circuits for myself and people i know. I personally care more about circuit design than i car about the final product. I recently came out with the Xenos Overdrive, which abuses a CMOS output opamp and has the output run open loop while the input stage had NFB to control the gain.

[Anonymous No.2459175]

>>2459132
If you can’t get 120V bulbs for the same socket, you’ll need to remove the socket and replace it with something like an E27. That or run it off a power supply.

>What exactly does that mean
There’s metal that a human can touch on the outside of the lamp housing. Use the continuity or resistance measurement mode to check if there’s a connection between that touchable metal and the terminals of the power connector. Because it’s not considered safe to have something that a human can touch connected to mains neutral, and definitely not live, though ground is meant to be safe to touch. If negative or positive on the connector are connected internally to the metal case, then just changing the bulb and type of connector wouldn’t result in a safe appliance.

>>2459144
Ever designed an effect with a phase-locked-loop in it? What about anything with switched capacitors?

[Anonymous No.2459187]

>>2459141
Thanks, anon. great info.
after looking at these transformers, this is starting to make a lot of sense. So, any low voltage light 50w transformer should work with this? Some specify LED or halogen

Just to clarify. These are the kinds that would probably have built in drivers right? I remember that first one being all over the place in the early 2000s, and the second one is fused.
ebay.com/itm/125262235979
ebay.com/itm/115475943313

>>2459175
>Use the continuity or resistance measurement mode to check if there’s a connection between that touchable metal and the terminals
Okay, I gotcha. Thanks for explaining that.

[Anonymous No.2459197]

>>2458855
sell things people want to buy

[Anonymous No.2459215]

>>2459197
Force Factor? Relaxium Sleep? Balance of Nature?
fentanyl? MyPillows? Covid test kits?
Child sex slaves? Social security numbers? Guns without serial numbers? hookers? cocaine?

[Anonymous No.2459232]

>>2459197
politicians?

[Anonymous No.2459286]

How's a pack of 300 chink ceramic caps for $8.85 sound?

[Anonymous No.2459342]

>>2459286
You should record them with your phone so we can hear them.

[Anonymous No.2459423]

Hi Guys,

So I have pic related lithium polymer pouch cell rated 3.7v 2000mAh. It has a battery protection circuit (PCM) built into it for over charge/discharge. I'm also wanting to connect it to a TP5046 charging port. The TP4056 also has a built in DW01 and dual mosfet to protect, again against overcharge, discharge, short circuit. I'm wondering will the two different protection things work together? like if one fails to work the other will save it? or is it dangerous to work like this with these two different protections in place. Also does anyone know with the TP4056, when I'm charging I'm aware I need to be somewhat present in case of any problems with the battery, however once its fully charged will it simply stop charging itself? like if I accidently leave it in a bit longer after its fully charged it wont have problems right? sorry for questions I am noob

[Anonymous No.2459430]

>>2459286
expensive

[Anonymous No.2459441]

>>2459423
From what I remember, the IC on the charging circuit is also a DW01. There won’t be any major issues with running both, though you could make things more compact by ordering the TP4056 modules without the extra shit on them. Or do what I do and just order TP4056 ICs and USB sockets, so you can make something for any form-factor.

[Anonymous No.2459443]

>>2459441
hmm compactness is not really a problem, I'm just wondering will they work in unison i.e double protection. If there is no issues running both that's good. thank you

[Anonymous No.2459487]

>>2459443
If you actually look at the circuit (picrel) All you've really done is added two more FETs in series with the path of current. The overdischarge protection for the DW01 nearest to the lipo will always kick in normally since it's got direct measurement of the battery voltage, while whether the second DW01 cuts in or not doesn't matter. As for overcurrent protection, the datasheet says:
>The overcurrent condition returns to the normal mode when the load is released or the impedance between BATT+ and BATT- is larger than 500kΩ
So regardless of which DW01's OC protection kicks in, it will properly latch. Assuming the impedance of the second IC isn't less thatn 500k that is.

If it does prove a problem (e.g. shutting off without turning on again), you can bypass the IC on the charger module by connecting the battery to the load terminals instead of the battery terminals.

[Anonymous No.2459531]

>>2459487
Ok thank you for your help

[Anonymous No.2459551]

>testing diy step-up flyback converter
>got feedback networks working properly
>overvoltage and gate drive issues fixed
>still basically no output
>wondering if transformer is saturated
>decide to check that the transformer pinout is correct
>beep out my primary, 3.5Ω
>beep out my secondary, 0.8Ω
holy shit it was backwards this entire time
and the pinout is diagonal so I can't just flip it

any hints on how to amend this? without rewinding. as much as i hate the fragility of mounting tht parts on the front of a single-sided board, all i can think of is putting it in backwards, which would be a pain to solder but is doable. the main issue with that is the form-factor of the board ending up too thick due to caps on the other side, but i guess i can resolder those on the top too. drilling new holes won't quite work, and the pins can't really be bent into shape, at least not after i've already trimmed them to length. i'll wire it up manually just for testing, trying not to delaminate any more traces than i already have.

[Anonymous No.2459558]

>>2459551
>any hints on how to amend this?

you must be new.
just glue it upside down and run wires.
it's called ''dead-bug wiring''
bing it for more pics.

[Anonymous No.2459568]

>>2459558
I don't want to run wires as that increases the high-current loop area. I'm already short a snubber or two and I don't want to tempt fate any more than I have by running a FET with half its elements probably fried.

[Anonymous No.2459573]

jesus christ
rewired the cap backwards hahaha

[Anonymous No.2459575]

>>2459573
>just_fuck_my_shit_up.png

[Anonymous No.2459586]

>>2459551
ok I swapped the windings, it's running stone-cold, outputting 300VDC, and the UV LEDs are lighting faintly. Which is a good sign. It's still voltage limiting from the primary, which is getting up to 60V, the maximum rating of the FET, so I'll need to desolder the FET again, inevitably delaminate more of its pads, and remove some turns from its primary. At the moment I've got 35 turns or so, I'll remove 8-10 and it should be good.

FYI, the current primary is what I hand-wound atop an existing CFL bulb inductor, the original winding of which is the current secondary. I'm not rewinding 225 turns of it.

[Anonymous No.2459587]

>>2459586
>desolder the FET again
I mean desolder the transformer again.

[Anonymous No.2459590]

>>2459586
>try to desolder while it's plugged in
>short 300V cap across finger
this is just not my day

[Anonymous No.2459593]

>>2459590
You're gonna turn into a spiderman now.

[Anonymous No.2459604]

>>2459586
ok now I removed 9 windings, resoldered it on there, and found the primary side voltage was basically identical with the secondary voltage only moving up by 3V or so.
maybe this is saturation, maybe it could be responsible for this anomalous behaviour.

tried a 100Ω/100nF snubber to get rid of ringing, it worked well but for the ringing but didn't touch that over 60V spike that's probably using my fet like a zener

i'm pretty ready to give up on this and just etch another mains linear regulator board, or maybe something with an array of 12V current sinks.

[Anonymous No.2459659]

>>2459573
Probably fine

[Anonymous No.2459709]

>>2457597
I have this tda 2002 amp chip that I want to power using a 5s battery pack.
https://datasheetspdf.com/pdf-file/1156408/SGS/TDA2002/1
Will it damage the chip in the long run if I charge the battery to 20.4v?
According to the discharge curve it shouldn't stay at that voltage for too long, but will still exceed the maximum ratings for some time.
>tl;dr is it really bad to exceed maximum ratings?

[Anonymous No.2459710]

>>2459709
*The chip has a max rating of 18v and an absolute max of 40v for 50ms kek

[Anonymous No.2459712]

>>2459710
>>2459709
I was just wondering what the difference is between 'dc supply voltage' and 'operating supply votage'.

[Anonymous No.2459714]

>>2459575
lel

[Anonymous No.2459727]

>>2459709
>>2459712
That is strange, but I think I would want to stick below 18V. It’s probably fine, but there’s no way to be sure. You can power it through a couple of diodes or an LDO, incorporate a buck regulator, if not ditch the ancient chip for something more modern with better PSRR. Maybe build a discrete class-AB for your 5S setup and use the TDA2002 for a lower voltage project.

[Anonymous No.2459759]

>>2459727
Well, I have already built my amp around that IC.
see >>2458099
I mean if it somehow burns out I still have a box full of those chips and I can replace them easily.

[Anonymous No.2459775]

/ohm/'s official drink

[Anonymous No.2459783]

>>2459775
I like coffee and turtles.

[Anonymous No.2459821]

analog panel meters are cute
digital displays lack soul

[Anonymous No.2459824]

>>2459821
analog = ballpark
digital = precision
If you aren't looking at an analog meter head-on, your readings will be inaccurate. They are cool though and still have their place.

[Anonymous No.2459830]

>>2459821
>>2459824
why not have both? a dual analog-digital meter. maybe i want accuracy on mondays and cool looks on the weekends. the perfect combo of coolness and accuracy.

[Anonymous No.2459832]

Just picked up a Sonos 5 speaker and I'm tremendously disappointed the speakers 3'5mm jack is a converted to digital, processed by the speaker and streamed back to the Linux server the speaker itself runs, so any audio has latency.

Would it be wise to just go ahead and solder an input jack to the TPA3118 audio amplifiers and call it a day?
I guess the speaker does some pre frequency separation as to not send mids and lows to the tweeter amps (the two on the left) can I do that easily?

[Anonymous No.2459836]

>>2459830
>why not have both?
that would be even less soulful

[Anonymous No.2459837]

>>2459830
Your meter, your choice? lmao

[Anonymous No.2459841]

>>2459836
so i have a better idea: two separate meters! a workhorse to get you from A to B and a weekend meter!

[Anonymous No.2459844]

>>2459841
now you're getting it

[Anonymous No.2459859]

>>2459844
nice. did you use generic project boxes? how did you make the labels?
did you put the binding posts on the display frame on purpose or you ran out of vertical room?

[Anonymous No.2459862]

>>2459859
1) both are the cheapest project box i could find on aliexpress that roughly fit my transformer and heat sinking needs. 2) some label printer i stole from work 3) ran out of room. the display frame is a printed carriage that has slots for the display and the control/adc board; it ended up needing to be a bit larger than i'd originally planned.

[Anonymous No.2459863]

Almost done, this is the back cover of the picture frame. I need some screws for attaching batteries and attaching this back cover to the frame and it will be done. I will make shorter cables as well

Arduino reads images from an sd card and draws them on e-ink screen. There is also a PIC there that turns on/off arduino via latch. Hopefully the batteries will be lasting long

It will update once a day or manually using three buttons (next/prev image and clear clear) behind

[Anonymous No.2459883]

>>2457597
Well shit balls. I got a 48v dc power supply and hooked it up to my led bar, and it just blinking. Slower than it was with the original busted power supply. I checked and it is supplying 48v. When it was connected to light bar, the voltage would drop until it blinked, then shoot back up, and drop again.
Any ideas what's wrong?

[Anonymous No.2459902]

>>2459883
>the voltage would drop until it blinked, then shoot back up, and drop again.

kinda obvious: over-current protection is shutting it down.
in my experience, the current rating on chink power supplies is very often exaggerated.

[Anonymous No.2459906]

>>2459883
Are both sides blinking? Disconnect one set of LEDs and test. If it works normally, disconnect those LEDS, connect the other set of LEDs and test again.

[Anonymous No.2459910]

>>2459902
Ffs I only needed like .7A, and it's rated for 3.

[Anonymous No.2459912]

>>2459906
>Are both sides blinking?
With the original power supplies, no. But yes the new one made both sides blink

[Anonymous No.2459926]

>>2459862
great job anon. i want to build something like this. maybe not as nice but at least something that would work.

[Anonymous No.2459943]

>>2459832
Should be possible, but you’d need to ensure that you prevent the original signal from fighting it, and that it’s at the right line level. Also avoid ground loops I guess.

>>2459844
Are those autoranging multitap with relays? Or just a single fixed transformer output with all the slack taken up by the pass transistor?

Also I’m gonna use a VFD for my adjustable PSU, LCDs aren’t edgy enough.

[Anonymous No.2459962]

>>2459912
The only thing I can think of is putting a constant-current buck converter on the output of the brick. XL4015 is adjustable and cheap. The 48V is gonna have to drop to around 46V to keep the buck converter happy.

https://www.amazon.com/Lithium-Battery-Charger-Charging-Converter/dp/B089NG9TFW

https://www.ebay.com/itm/5pcs-XL4005-5A-DC-Buck-Step-Down-Voltage-Converter-Constant-Current-Power-Module-/142632518904

[Anonymous No.2459965]

>>2459962
Nevermind, 48V is too high for the buck input.

[Anonymous No.2459971]

>>2459965
>Nevermind, 48V is too high for the buck input.
Well fug. Constant current "led drivers" are way more than what would make this repair economical, from what I googled at least.

[Anonymous No.2459973]

>>2459971
I'm still looking for cheaper solutions.
https://www.amazon.com/LPF-60-48-Mean-Well-LPF-Driver/dp/B00DEE1TIG

[Anonymous No.2459982]

>>2459973
Bless you sir. That looks fairly promising

[Anonymous No.2459993]

>>2459982

the major diff between a constant voltage and constant current source is the CV one expects to see some resistance in the path.
so, before buying anything else, i'd try to add a resistance in series to limit current to an acceptable level.
like what? a 120Vac drill, or a 12Vdc drill, a toaster, several 120V incandescent lamps in parallel, one or more 12V halogens in parallel, coffee maker, toaster oven, microwave transformer, etc.
the idea being to find what resistance is needed to set the appropriate (measured) current level, then finding a load that's small enough to fit inside.

[Anonymous No.2459995]

>>2459982
The only thing cheaper that I found was from Chiner and it takes months to get the stuff.
>>2459993
Thank you for the helpful advice, Anon.

[Anonymous No.2459998]

>>2459783
I like coffee but hate turtles. I still drink soda with the out-lawed PLASTIC straws.

> Fuck them shielded bois rite up they nose

[Anonymous No.2460000]

>>2459824
Experienced engineers don't need precision.

[Anonymous No.2460003]

>>2459998
But turtles are slow and steady. They win the race (except turtleboy McConnell). Snapping turtles are little fuckheads though. They should be put in labor camps.
>>2460000
Where we're going we don't need roads.

[Anonymous No.2460007]

>>2459844
> FOLD
what does that do?

>>2459859
Friends been using VFD's in a DIY guitar amp. Its a simple 7 segment and it shows the numbers 1 through 5 for all 5 of the channels he built in it.
Way to much effort for a 8watt amp.
> In comparison, i built a 17watt amp with only 5 trannies, only 1 channel with 2 controls, volume and midrange.

[Anonymous No.2460012]

>>2459993
>? a 120Vac drill, or a 12Vdc drill, a toaster, several 120V incandescent lamps in parallel, one or more 12V halogens in parallel, coffee maker, toaster oven, microwave transformer, etc.
How would I add that stuff in series with the LED?

[Anonymous No.2460015]

>>2459993
>d try to add a resistance in series to limit current to an acceptable level.
>like what? a
Oh wait, when I measured the current of the working side of the led bar, I was getting 600-700 mA, that what you mean?

[Anonymous No.2460017]

>>2459993
Thing is, if the blinking is coming from the LEDs, not the PSU (e.g. the voltage doesn't stray from 48V) that means there's some smart circuitry inside the LEDs themselves. But it's such a strange response that I couldn't imagine what it's trying to do, unless it's one of those strange fluourescent tube replacements and the blinking is a ballast or whatever. Anon needs to take apart the LED side to see what's actually going on and what kind of circuitry is trying to drive the LEDs, if it was a dumb LED strip with no current limiting, he'd have already killed it with the 48VDC PSU, not got it to blink.

>>2460007
>what does that do?
Closes the timelike loop to send voltage back in time.

>>2460012
Alligator clips on the prongs, for a toaster or drill you may need to hold them on.

>>2460015
Yeah probably.
Try measuring the current with the new PSU, see if it's higher or lower than that, but it may be difficult to do with the blinking if you don't have a max feature on your DMM. Also measure the min and max voltage across the LEDs while it's blinking.

[Anonymous No.2460021]

>>2460017
>Anon needs to take apart the LED side to see what's actually going on and what kind of circuitry is trying to drive the LEDs,
That was step 1. Consensus seemed to be that this thing was fucked. AFAIK this is the only circuity in there, which I've now removed, just the dumb LEDs left.

I'll go measure the max current and voltage while blinking

[Anonymous No.2460022]

>>2460021
Front side

[Anonymous No.2460023]

>>2460017
>Closes the timelike loop to send voltage back in time.
does that affect signal's causality?

[Anonymous No.2460024]

>>2460022
Wondering whether U2 is a boost converter IC.

[Anonymous No.2460026]

>>2460024
Wut

[Anonymous No.2460029]

>>2460026
???

[Anonymous No.2460033]

>>2460029
Kek yea I'm an electrotard.

[Anonymous No.2460044]

>>2460017
>Try measuring the current with the new PSU, see if it's higher or lower than that, but it may be difficult to do with the blinking if you don't have a max feature on your DMM. Also measure the min and max voltage across the LEDs while it's blinking.
Ok so max current was 615 mA, min just showed 0.
Both max and min on voltage read 0, not sure why, but just looking at the reading normally, it fluctuated between 27 and 19V

[Anonymous No.2460047]

>>2460033
So am I, Anon. So am I.

[Anonymous No.2460049]

>>2460021
Well the thing is still blinking, so there’s got to be some circuitry alongside those “dumb LEDs”. Can you take those apart to see if there are any distributed resistors or current regulators at least?

Reverse engineering the old PSU (all of it, not just that end of it) might be revealing also.
Maybe high capacitance could cause the new PSU to blink from inrush current.

>>2460044
Ok so there’s no high peak current spikes to be wary of, it’s drawing a normal amount of current when on. So chances are it’s intended to run off a constant voltage. But the fact you can’t measure the voltage peak in max or min mode is somewhat worrying. That 19-27V will just be a bumpy average (or rms), which doesn’t help much.

If it gets up to the full 48V but no higher when it blinks on (the 600-700mA output) I imagine the PSU is good.

[Anonymous No.2460052]

seems like all voltage regulator chips have a voltage reference so they cannot regulate below that level. how do bench power supplies produce low voltages down to 0v?

[Anonymous No.2460055]

>>2460052
Some switching regulators take an external reference instead of an internal one, which can be adjusted with a pot or DAC down to 0V. The alternative is introducing your own op-amp differential amplifier to feed the FB pin with the difference between your own variable reference and the output. Add some integral gain to the differential amplifier to compensate for its low small-signal gain, maybe some differential too if it’s prone to oscillation from too high a gain.

[Anonymous No.2460062]

>>2459943
just a fixed transformer and a fat pass tranny. the blue supply uses a stock intel cooler with a fan. the black one is only dissipating ~25W so the tranny is bolted to the case; i haven't finished putting it together but i ran some thermal sims so hopefully it's fine.

i'd like to experiment with adding taps to transformers but taking them apart is such a huge pain in the ass.

>>2460007
>FOLD
foldback limit. i haven't taken measurements but, according to my math, at high line + 0V/3A output the transistor junction temp gets close to the datasheet limit. this is bad for transistor life so i added a foldback circuit to avoid that operating condition unless i specifically need it. there's a backup temperature cutoff too but it just shuts the supply off.

[Anonymous No.2460064]

>>2460055
Is it possible/feasible to roll your own voltage regulator from discrete components that would work as well as, say, LM317, minus the thermal shutdown or any other nifty features?

[Anonymous No.2460069]

>>2460064
yeah it'll just be shit unless you add a lot of compensating/buffering circuitry and hand-match transistors for your current mirrors and differential amps.

[Anonymous No.2460073]

>>2460062
You can buy multitap transformers relatively easily, though you don’t get that many taps in them. One with two seperate secondary windings, each with a centre-tap is reasonably common though, so you can stack those for 4 different ranges.
Also the transformers in tube-testers have a bunch of taps in them that look handy.

As for diy transformers, if I could get a transformer core that’s circular and could easily be mounted on a slow-moving lathe or other rotating tool it would be pretty easy. Maybe just winding around a PVC pipe then slipping the pipe and windings onto an EI core would work well enough, but it wouldn’t exactly be the best for minimising size. A closed-core transformer where the middle is a removable cylinder would be ideal, not sure if anyone makes those but you could probably machine down an existing one easily enough. If you had a lathe, that is.

>>2460064
Absolutely. In fact, it’s preferred when it comes to eliminating high-frequency noise, since common monolithic linear regulators are only really designed to block rectified mains frequency ripple.
Just an op-amp and a sziklai pair and passives is usually sufficient, though if you want to get fancy like the audio guys do and bootstrap the op-amp’s and/or reference’s own power rail for optimal ripple rejection you can get even better results.
It’s also probably possible to use an adjustable linear regulator like the LM317 as a pass transistor in an op-amp circuit to get the good ripple rejection at a wide frequency range, and the thermal cutoff, but considering the kind of funky feedback loop you’d need to get a good frequency response while retaining stability you may be better off just adding a thermistor.

>>2460069
Oh yeah it will be pretty garbage if you’re using discrete transistors. Unless you’re one of those analogue wizards.

[Anonymous No.2460101]

>>2459551
>>2459604
Ok I can get it to be in current limiting mode with a current sense resistor set to 1mA, and with a quarter of the LEDs. In that state I get 27VDC spiking in the primary and ~80V out of the secondary, at the planned 1mA. When I change the current sense resistor to be set for 20mA, the output voltage stays roughly constant, while the primary peaks shoot up to 60V again, where the negative feedback kicks in to prevent it cooking the transistor again. It's like it's current-limiting itself by changing the voltage ratio of the transformer, so I'm guessing I am actually running into saturation. Even though this core has an air-gap. I may look for a replacement transformer.

More likely I'll just design a different board that uses linear regulation on low voltage DC strings directly from an existing PSU. I do have a 48VDC DIN rail PSU lying about, which I guess I could use for this, though it's not exactly plug-and-play safe, and I wanted to use it for my variable benchtop PSU. I'll see what else I can get at work.

[Anonymous No.2460137]

>>2460101
>can't even power 1 LED at 20mA
Jesus this is awful, think I'll replace the transformer with an inductor and see what that does.

[Anonymous No.2460180]

I kinda want to make something like a Cliff Quicktest. Though I'm not sure about the arc-resilience of having knife-edge contacts for turning off loads, so I think I'll go for a design that incorporates a mains-rated momentary switch.

[Anonymous No.2460201]

>>2460101
>>2460137
Replacing the transformer with a 200µH inductor and it's fucking easy to get 20mA into the LEDs. Well, 12 of them anyhow, before the voltage limit kicks in. And I've got rows of 13, so I'll likely redesign an array with rows of 12 of them, plus the resistors and red guide LEDs on it also. Turns out the requirement that you drop 5V across your sense resistor makes a significant impact to the number of LEDs you can handle, compared to more modern lower reference voltages like 1.25V.

[Anonymous No.2460220]

every time i see a video of someone using heatshrink they always use a hot air gun..
the fuck?
i always just use a lighter

[Anonymous No.2460228]

>>2460220
gotta piss with the cock you got

[Anonymous No.2460229]

>>2460220
In theory, slower, more even heating gives you maximum shrinkage. I dunno if that's bullshit or not but that's what I always heard. It does usually look a little nicer. you can also do 5 or 6 at a time if you have the slot nozzle, I've done that for marking shrink on the back of ferrules.

[Anonymous No.2460233]

>>2460229
>ferrules
wait is the plastic at the back of shoe lace ferrules shrink wrap?

[Anonymous No.2460240]

>>2460180
I kinda also want to add a variable current limit too. Fuses work, but a potentiometer that controls a crowbar that trips a breaker would be even better. And a voltage/current panel meter while I'm at it I guess.

>>2460233
aren't those called aglets?

[Anonymous No.2460258]

>>2460180
>>2460240
Enter: the Mains Overcurrent Latch, circuit prototype the 1st

Decided against using a breaker, the only small mains-rated ones I can find locally are thermal so take a while to trip, probably trip the building breaker before that.
Think I'll use a mains-coil relay and a momentary pushbutton instead of the knife switch, that way I can get it rated up to 15A pretty easily. But then to get it to turn off even while the button is being pressed, I think I need to use this TRIAC + optoTRIAC circuit. Not a big deal, assuming the turn-on current of them is compatible with my comparators' shitty zener supply. DC coil relays won't really work due to their high current requirement, but SSRs may be the more elegant solution if they don't need much current to switch. Other solutions like SCRs or FETs inside rectifiers could work, but feel messier and still need some sort of opto.

Hope 0-150mV is linear enough within the range of these comparators, because I'm not going much higher with these tiny current sense resistors. Also need to make them latch, think that's possible with positive feedback, I'll have to mess about in spice. I have two comparators to make it work with, so it shouldn't be too hard.

Chances are I've forgotten something obvious (aside from not finishing the schematic) so feel free to suggest better ways to do what I'm doing.
>inb4 don't put negative voltages into the comparator
I'm sure it's fine, it's going into open-emitter inputs and won't pull them down by anything close to a Vbe drop. Can hardly put a diode there anyhow with an amplitude of 150mV maximum.

[Anonymous No.2460261]

>>2459832
>>2459943
>>2459832
>>2459943
Are pins 3 and 4 the right pins? I'm not used to SMD complex chips.

TPA3118 has internally biased differential inputs - meaning, if you connect any of the inputs straight to ground the TPA3118 will NOT like it (it will drive the FAULTz pin low).

So I guess I can't just straight solder my inputs to the amplifier? I'm going to try to piggyback off the caps on this board and see where that takes me...

[Anonymous No.2460274]

>>2460261
Ok I successfully piggybacked off the caps, only problem now is the speaker only keeps the amp powered and working when it knows there music being played, so I can only hear my analog input when I play content on adjust the volume on it. Any idea how to keep the amp always on?

[Anonymous No.2460285]

>>2460261
You may need a differential output amplifier circuit if you need to feed signals into that setup. A simple inverting amplifier with a gain of 1 may be sufficient.

>>2460274
With any luck that’s just an enable pin being toggled that you can forcefully pull up or down to keep it on. If it’s digital all the way, you may need to feed your main signal into the system some other way, feed a dummy signal into the other end of the digital pipeline, or write a little injector microcontroller to sit on an I2C bus and change one command to another, like what I assume modchips for consoles do.

[Anonymous No.2460292]

I want to use this LED connection to turn on another LED.
Anything I have to look out for when using a transistor as switch? Does the it also need a resistor?

[Anonymous No.2460354]

>>2460220
lighters are unsafe indoors. can cause fire

[Anonymous No.2460370]

Can anyone explain why my tractor battery keeps dying? It's new but it was made 2 years ago, 6 volt tractor battery. The system is 6V positive ground, I redid the wiring so all the wires are new as well as the voltage regulator. I'll charge it up reading 6+ volts and when I crank it the voltage immediately drops to around 5.6-5.8 or lower. Is that a sign of a bad battery? Or is it possible I fucked something up? Can provide pics.

[Anonymous No.2460385]

>>2460370
Needs a desulphinate cycle on a good charger and verify you have water in it. If you overdischarged it you could have permanently damaged it.

[Anonymous No.2460388]

>>2460385
I filled it with distilled water. I'll look up what a desulphinate cycle is. How would I know if it's overdischarged?

[Anonymous No.2460389]

>>2460052
>how do bench power supplies produce low voltages down to 0v?

one way is to send the reference voltage not to ground, but to a negative voltage equal to the ref voltage.
this lets you go to zero, where everyone is happy and gay.

[Anonymous No.2460465]

>>2460292
Apparently the tranny is drawing too much current, so the power LED doesn't turn on anymore.
I already put a 390 Ohm resistor before the base emitter junction, but still.
Am I doing something wrong?

[Anonymous No.2460466]

>>2460465
It's a BC547 if that's important.

[Anonymous No.2460530]

>>2460465
I’d use 10k on the base, and 1k on the collector in series with the new LED. 390 is pretty tiny for a BJT’s base. You could also use a logic level MOSFET and not have to worry about that.

[Anonymous No.2460547]

>>2460465
>Am I doing something wrong?

probably everything.
- do a drawing showing the circuit you came up with. make sure to show transistor part # so we can verify pinout.
(shoulda done this in your very first post.)
- disconnect the wires going to the transistor/resistor and test if the LED on PCB has survived your tortures so far.

[Anonymous No.2460553]

>>2460465
>>2460547

also measure voltage (if you can) on both sides of the PCB LED with respect to battery negative.
this is quite important coz it might be connected in a funky way.
we also need to know battery voltage so we can math correctly.
yeah, it only seems easy to the un-informed.

[Anonymous No.2460696]

>>2460220
>i always just use a lighter

Same anon
>pic rel

[Anonymous No.2460698]

So I have a raspberry pi inside of a 3d printer that I can't get access to so easily, I want to attach webcam to my raspberry pi inside, and provide external power for the webcam,

Would pic related work? the power source I would use is coming from the same power supply the raspberry pi is on.

Also in the schematic should I also wire a ground to the raspberry pi port as well or no?

[Anonymous No.2460704]

>>2460698
Yes, make sure to ground the shield of the USB socket to the Pi.

[Anonymous No.2460707]

>>2460698
>wire a ground to the raspberry pi port as well or no?
The shielding in the USB cable connects the grounds between the webcam and socket.

[Anonymous No.2460710]

What is the point of attempting to add a bias through a capacitor?

[Anonymous No.2460713]

>>2460710
you mean the 47nF in the top left? might be for startup?

[Anonymous No.2460721]

>>2460710
virtual gnd, it's sending toan to ground

[Anonymous No.2460724]

>>2460704
>>2460707

Thanks the usb sockets I have just have 4 pins, so I guess I don't need to worry.

I probably clarify I am basically making a usb_A male - female extension that I want to have accessible from the outside so I can plug/unplug my webcam when needed. I have a powered hub but I don't want to fuck with that and I only use one port

[Anonymous No.2460739]

>>2460710
Lmao anon dont try to understand music people schematics theynare the most fucked up brained people to ever exist. Will talk about how different capacitor manufacturers sound different using terms like warm and open. Completely meaningless and subjective. Do they sound different? Maybe, depends. Does it matter? No.
Anyway the point is if you want to learn, build 8t and try it out. Investigate with a scope see for yourself.

[Anonymous No.2460743]

>>2460724
Nope. USB has 5 connections, 2 signal, 2 power and one shield. If you want to make an extension you must connect the first 4, both signals and both power. The shield should be connectedntoo but it might work without it.

[Anonymous No.2460747]

>>2460743
Anon is powering the device externally so the +5 isn't connected to the USB host.

[Anonymous No.2460764]

repost from >>2460761 ▶


im going to a scrap yard and would like to buy some electronics (computers, HDDs, TVs, printers)
what is a good price to offer or want to settle on? any experiences?
ill be spending mostly the money i get for scrapping my metal

[Anonymous No.2460769]

>>2460764
How much shit do you need? What are you going to do with it? If you're not buying scrap by the ton you aren't getting a good deal. Better to dumpster dive because retards just throw stuff away when it can easily be fixed for $0 and little effort, or the store policy is to write it off and dump it.

[Anonymous No.2460782]

>>2460769
i only have a total of about £50 or so in metals, and there are two or three different types
i can throw in some money too but really i would like to look around the yard
and get as much as he would give me (in the electronics i mentioned) for £50 or so
and id be doing it to salvage parts and also, have a look around

[Anonymous No.2460793]

>>2460782
Sorry. I misunderstood your first post. If the scrapyard is willing to trade then it wouldn't hurt to go have a look around. Just keep in mind that it's all basically junk with a minimum value, even if you have an intended use for it. I wouldn't bother with printers or HDDs unless you want to salvage them for specific things (magnets, HV laser/LED, motors, etc). Depends on what's there, what's useful to you, and whether they're willing to let stuff go for nothing.

[Anonymous No.2460796]

>>2460793
the thing id want to get from HDDs is magnets yeah. theyre very powerful magnets and would be useful for me right now
if i get about £20 for my metals and i have that cash, then i go and pick up a TV or 7 or 8 HDDs (they would probably come in so many from one machine i THINK) and i offer him £5 for a TV or 7-8 HDDs, would that be a good price to offer?
or is it too much? too little? £5 for a probably broken TV etc.

[Anonymous No.2460799]

>>2460796
Make offers on shit that you don't care about first so you can feel them out. When you get to things you want, act like you aren't really interested and make a low offer. You might get lucky, or they might charge you the going rate for the materials. Either way, it's worth it because it doesn't have to be shipped.

[Anonymous No.2460802]

>>2460799
ill keep that in mind
i do get attached to things easily though, haha. but ill try and feel them out and go from there

thanks for your help, i understand a lot more now
but if you have any sort of ballpark figure for a TV or computer, HDD, then that as a guideline would help

[Anonymous No.2460804]

>>2460796
>£5 for a probably broken TV etc.
TVs aren't really worth much unless the screen is intact, or you can test the boards and verify their proper function. Even then, there has to be a market for the parts or they'll just take up space in storage.
>>2460802
HDDs will be weighed because it's basically a block of aluminum so there's a base price. I don't think they give a shit about neodymium magnets unless they're big bois, or you're buying them in massive quantities.

[Anonymous No.2460809]

>>2460804
im very happy with that. they arent too heavy and alum is cheap
maybe it could end up in profits for me after all

i doubt ill come back and show my finds, but i have some of the electronics from TVs i might keep. ill see you around if i ever take up a project with one of them!

[Anonymous No.2460810]

>>2460802
Wish I could give you a ballpark, but it's been 20+ years since I've been to a scrapyard, and I'm not current on metal prices. Have you looked for really old HDDs in local computer shops? Maybe you could get a deal on low capacity shite from the 90s-00s.

[Anonymous No.2460811]

>>2460810
the old HDDs that are like 20GB, 50GB? theyre absolutely worthless now
something funny that happened to me lately, i bought one for £0.17 from CEX, and got it, but they had already stripped the gold from the HDD. lmao. ...

[Anonymous No.2460812]

>>2460809
>ill see you around if i ever take up a project with one of them!
I want to see some kind of dangerous weapon made from junkyard parts like some futuristic dystopian robot nightmare when you come back, Anon.

[Anonymous No.2460814]

>>2460811
>but they had already stripped the gold from the HDD
What a waste of time. Holy fuck.

[Anonymous No.2460817]

>>2460814
i think that something extremely suspicious is going on with CEX - and theyll do things like that, while holding old electronics
ill leave you to think about it and wont spam. but those magnets are expensive - and they might be buying/selling things from themselves, then they can probably tear down and rebuild any hard drive they want.

i dont know, its just hilarious to me that they did too
still
>>2460812
i would love to put an actual OS on the TV boards
a crazy scrap weapon could just be less electronics i think lol

[Anonymous No.2460823]

>>2460817
Hmm. If I own a scrapyard or warehouse (or several warehouses), and thousands of parts that contain precious metals, am I technically a bank vault who can issue loans?
>>2460817
>OS on the TV boards
That would be nice. Probably need to do a side-channel attack to bypass bootloader protections so you can load unsigned code. I hope you actually do it, Anon. Git gud with Ghidra, binwalk, etc. if you aren't already.

[Anonymous No.2460839]

>>2460796
I would take the motors if they're removable, you may end up wanting some and wish you had the ones from HDD

I was ignorant and disposable 10 laptop batteries that all had 18650 in them, and I still kick myself for it, but I salvages all the stepper motors from floppy/dvd drives I had.

[Anonymous No.2460912]

i am making this thing from a tutorial i found online for a diy microphone
but i would also like to add a low and high pass filters to filter out noise since human voice has a limite frequency range, so no need to the mic to pick up frequencies outside of it
can anyone show me where in the circuit can i add those?

[Anonymous No.2460937]

>>2460912
The 22μF+3.9kΩ network already makes a high-pass filter, which you could tune the frequency response of with ease. As for the low-pass, I think I’d probably add it after the preamp IC, since series impedances may mess with the JFET stage. A single series resistor and a cap each to V+ and V- would by my attempt, to try and minimise turn-on-pop.

Also consider capacitance multipliers instead of the PSU RC filters, reduces the 2200μF caps down to be less bulky.

Also are the THAT1512s in stock? And has he not cruelly updated the schematic from the original video?

[Anonymous No.2460939]

>>2460937
>THAT1512s
they are no, so i'm using another ic, 2n4416 is also deprecated but luckily chinks still sell it, as well as that dc-dc power supply so the only part different from the original schematic will be the opamp, this is my current circuit design, i'll just slap the high pass filter on there and it should be done

>Also consider capacitance multipliers instead of the PSU RC filters,
i'm not experienced enough with sound processing to make any improvements in the circuit on my own, since any changes i make could likely make the sound quality worse

[Anonymous No.2460942]

>>2460939
one thing i'm not sure about the original diagram however is,
is the microphone casing (the metal mesh)
audio ground
5v power ground
and the ground for rest of the components all the same connected ground i should connect together?

[Anonymous No.2460947]

>>2460710
reduce startup pop maybe? idklol

[Anonymous No.2460956]

>>2460939
Firstly, you’ve biased your JFET to GND on both sides, instead of tying it to +/-15V. The lack of grounding resistors on the inputs of the instrumentation amplifier IC has me somewhat worried. Could get a wonky DC output potentially clipping your wave.
I also wonder about the JFET biasing of the original circuit, strangely symmetrical for a JFET that usually wants its gate voltage sitting kinda negative with regards to the drain-source channel.

>>2460942
Yeah they’re all commoned.

[Anonymous No.2460970]

>>2460956
thanks, yeah whoops, i didn't notice i fucked up copying the resistors
can you draw into the schematic the changes you would make personally? if you have a 4chanx you can just edit the image in the browser easily, or even just ctrl+v it from mspaint to the form here directly
the original schematic has grounding reistors tho, 3.9k on each leg

the original circuit was designed by a guy who isn't an electronics engineer so it's definitely not perfect, but the audio from it still sounds great so it seems good enough for sure altho i'm not opposed to improvements

[Anonymous No.2460971]

>>2460547
It's a PSP mod. I want the shoulder buttons to light up when the system is turned on.

[Anonymous No.2460972]

>>2460530
I can't test it now, but I'll be using the 10k resistor for the base then. I don't have such small MOSFETs.
>>2460547
It's a PSP Mod. I want the trigger buttons to light up when the system is turned on, so the ground between the power switch PCB and the battery I want to connect the trigger LEDs to is common.
>>2460553
The voltage on the LED I want to turn the turn the transistor on with is around 2v.

[Anonymous No.2461060]

>>2460972

if the PCB LED has a high-side resistor, then your drawing is fine, coz the emitter will connect to ground.
but if it has a low-side resistor, then it may or may not work.
to fix, find a ground (i.e. battery negative) and connect the emitter to it, instead of cathode of LED.

[Anonymous No.2461067]

>>2460769
The recycling center I got electronics for free in the past has completely sealed up .
Where do I find dumpsters? In germany it seems like most people drive to "recycling centers" to dispose of their "broken" electronics. Corrwct me if I'm wrong.

[Anonymous No.2461077]

>>2461067
Germany has crazy laws about dumpster diving. It's illegal in many places, but people don't seem to give a shit about retail stores or offices unless you're dumping your own trash in their dumpster (illegal everywhere, dumpsters are leased). Totally unfamiliar with seasonal shopping habits in Europe, but in the US it was Black Friday and the day after Christmas when the good stuff was plentiful. Early bird gets the worm.

[Anonymous No.2461109]

>>2461077
So I should look out for office dumpsters?
Aren't they locked most of the time?

[Anonymous No.2461117]

>>2461109
Depends where it is, I suppose. There are innumerable strip malls in the US with a lot of offices, and they usually have the dumspters outside in an alleyway. If it's a building full of offices, then they usually have everything locked up so you need a keycard to even get in. Sometimes you can get an entire office full of computers/monitors when they upgrade if you offer to destroy their old drives and haul everything away. Small businesses are your best bet. Corporations contract it out to their butt buddies.

[Anonymous No.2461140]

>>2460721
oh yeah and that's also part of the feedback loop, minimum gain is about 12, max is about 118
What part about it is hard to understand?

[Anonymous No.2461199]

[Anonymous No.2461200]

>>2460970
Pic related. The two grey circles at the front can have their RC constant adjusted for a 20Hz cutoff (it's 1.8Hz at the moment, so reduce the capacitance to 2.2µF for 18Hz). Then, since C9 has a maximum impedance of 360Ω, we should aim for the low-pass filter to have a series resistance of somewhat less than that, so 100Ω. To get a 20kHz cutoff with 100Ω, we need something around 80nF, 82nF is the closest normal value.

I'd also experiment with tying the condenser's GND to -15V (with another RC filter), to see if it gives better sound.

[Anonymous No.2461248]

>>2461200
thanks, i added the filter to my circuit
can't wait to test it out once rest of the parts arrive

[Anonymous No.2461263]

>>2460912
Oh it’s an electret? That explains the strange biasing. Disregard my statement about tying it to -15V.

[Anonymous No.2461334]

Speaking of the voltage reference in linear voltage regulators. It occurred to me that it doesn't have to be super precise as long as it is stable enough. Let's say even if it is off by a whopping 5%, it shouldn't really matter as long as the power supply displays the actual measured output voltage not the set voltage, right?

[Anonymous No.2461367]

>>2461334
the Vref of lm317 adjustable regulators varies by more than 5%
I usually have to lower the positive rail to match the negative one because the negative 337 regulators vary even more

[Anonymous No.2461387]

>>2461334
For variable power supplies, if you’re using a DAC for reference setting you can just use an ADC to correct for drift and offset. Not like you still don’t want to do the measurement at the output though. If you’re using a hand-adjusted potentiometers then there’s no precision other than the output anyhow, aside from putting numbers on the dial.

[Anonymous No.2461394]

i want to build a discrete under voltage lockout circuit with hysteresis. i'm powering an LM339 with 12V zener, which i'm dividing down to 4V for my reference. i'm comparing my DC bus, which varies between 0V and 30V maximum, to this 4V reference. i want under voltage lockout set at 14V.

i know the LM339 requires ~2 volts of headroom and i want to divide my 30V DC bus maximum down to fall under 12V, because i've read conflicting things about reliability when input voltage is greater than IC supply voltage.

i started without hysteresis and calculated out some resistors values. pic related seems to work reasonably fine (at least in simulation) despite me being shit at electronics.

i then tried adding 1V hysteresis to the 14V lockout. i followed https://fscdn.rohm.com/en/products/databook/applinote/ic/amp_linear/comparator/gpl_cmp_hysteresis-e.pdf "Calculation of threshold voltage (resistance division type)" but the numbers aren't making sense. i calculated that a feedback network with a 3M and 17K resistor should get me ~0.33V divided feedback or ~1V undivided feedback. instead i get some crazy ass oscillations when i run the simulation. what am i messing up?

[Anonymous No.2461395]

>>2461394

[Anonymous No.2461397]

>>2461394
>>2461395
just realized i left the ground to the mosfet disconnected while dicking around and taking these screenshots, but the impact to the functionality of the circuit is negligible. oscillations still persist.

[Anonymous No.2461403]

wtf did i do, bros? what else do you call them? is ali going to email me 18+ offers?

[Anonymous No.2461410]

>>2461394
Oscillation comes from negative feedback with a delay.
Try a small capacitor (100p-1n) from the UVLO_IN node to ground. If that doesn’t work, put the cap from UVLO_IN to the comparator output. This should snub any negative feedback, maybe.

Also check Vcc to see how much it varies, the path from Vcc to the inverting input may possibly be causing the oscillation. Try using seperate series diodes for the voltage sense divider and the load. Increasing hysteresis may also dampen oscillation.

If neither of these help, I suspect your comparator model is shitty, and that you try a different one.

[Anonymous No.2461449]

suggestions on not shit FOSS, PCB design software? planning on building out 4x of picrel and dont wish to dick arounds with point to point anymore

[Anonymous No.2461456]

>>2461449
Altium Designer

[Anonymous No.2461462]

>>2461456
isnt that the one you need to create an account to even use? i want something that just frekin works like paint or logisim

[Anonymous No.2461463]

>>2461449
KiCAD is brilliant at everything that isn’t autorouting. If you want that try Circuitmaker, it’s free but not open-source. Same for the demo of Eagle, it’s just size limited.

Also add some fucking pull-down resistors. Bypass caps too.

[Anonymous No.2461467]

>>2461449
just pirate altium, weirdo

[Anonymous No.2461468]

>>2461467
>just pirate a niche software that litrally who cares about bro

[Anonymous No.2461473]

>>2461468
The professionals use MS paint for all of their designs. Are you professional enough to save the president?

[Anonymous No.2461533]

>>2461449
I used eagle before but am willing to try kicad in the future. Making the footprints for stuff was the hardest part.

[Anonymous No.2461536]

>>2461533
You can import footprints and symbols en-mass from digi-key:
https://www.digikey.co.nz/en/resources/design-tools/kicad

[Anonymous No.2461578]

>>2461449
the cern particle collider was designed in kicad
only braindead tars would use anything else

[Anonymous No.2461582]

>>2461578
unless you want autorouting

[Anonymous No.2461697]

>>2461578
Does the CERN poster over your bed have jizz stains on it?

[Anonymous No.2461726]

>>2461410
isn't the whole point of hysteresis to avoid capacitors?

[Anonymous No.2461735]

>>2461117
So do I just ask the offices or what?
I used to work for one of those computer refurbishing company that have contracts to majpr companies where they apparently get the stuff for almost free and it was really disturbing to see all those LCDs computer monitors with bad caps going straight to the recycling. It would have been a matter of 30min max to repair them.
I mean one doesn't even have to have any understanding of electronics. In 90% of cases it's just bad caps and a retard could replace them.

[Anonymous No.2461822]

>>2461726
Not really. If you want to avoid oscillation with a high-gain amplifier, you need to ensure there’s no delayed negative feedback loop. The only loop I can see would be through the output of the amplifier, through the FET pulling down the supply rail, through the supply rail and into the voltage divider to the inverting input. A capacitor from this voltage divider to ground should serve to dampen any signal making it through that loop, while a capacitor from the voltage divider to the output should decrease the effective high-frequency gain.

Either capacitance can increase the delay and thus make oscillation more likely, so it would be better to eliminate the negative feedback loop some other way, like preventing a signal from going from the output to the power rail. But I’m not sure how that signal would even exist at 1V hysteresis, so maybe the issue lies elsewhere.

[Anonymous No.2461841]

>>2461735
>So do I just ask
Yes, but it'd be easier to leave flyers or business cards that advertise your "services" that consist of on-site data destruction and taking their bulk waste away "free of charge" lmao

[Anonymous No.2461872]

>>2458099
>>2458247
I've been noticing those ~4w I'm getting with the 8ohm speaker sound very full and it's more than enough to sonicate an entire room, almost too loud.
Comparing this to a cheap class D module with the same power, the D amp somehow sounds much thinner and not as full.
I wonder if the amplifier power ratings really have that much significance.

[Anonymous No.2461889]

>>2461872
>power ratings really have that much significance.

of course.
10W will vibrate your floor
100W will vibrate your house
1000W will bring down the twin towers

[Anonymous No.2461893]

>>2461872
Sounds like it’s more harmonic content than amplitude, if not you can compare RMS signal output to the speaker in both cases. It could just be that the gain of the class-D isn’t as high, or maybe the output can’t be as high if the voltage rails are smaller for the same speaker. Class-Ds can usually go closer to the rails for the same voltage, and are more likely to be differential output amps than Class-ABs so would likely be able to output twice the voltage swing, but if it’s just running in 5V compared to 12V or higher for the AB that’s going to make a difference.

[Anonymous No.2461895]

Does anyone know what happned to my nigga? Has the coof got him?

[Anonymous No.2461908]

>>2461895
He was recruited by the Taliban to teach Afghan girls how to build z80 based breadboard computers.

[Anonymous No.2462010]

>>2461895
i think he has just run out of contect. his videos are so invested with time.. that and he prolly got a new job is focusing on work

[Anonymous No.2462018]

I need to power UV LED strips in my nursery. I do not want to get an additional ballast to power them specifically (they need 24v@ 1A max).
Is there a way I can feed them, adjust-ably, off the ballast that powers my lights (it outputs 24volts), or should I look into getting multiple ballasts to power them specifically?

[Anonymous No.2462037]

>>2462018
Please do not expose your infant to UV rays

[Anonymous No.2462079]

>>2462018
Are you referring to those LED replacements for tube fluorescent lights? Those do need to run of ballasts, but standard LEDs will just run off constant-current power supplies or “LED drivers”. You can definitely get variable constant current “LED drivers”, be it through PWM or varying current.

Post link to LED light

[Anonymous No.2462109]

>>2462037
im retarded i meant infrared..
>>2462079
strip is picrel. power requiremetns: 24v @ 1A max. you mention PWM, if that is the case that would be awesome. What about something like this: https://www.aliexpress.com/item/2251832826410547.html?

[Anonymous No.2462133]

>>2462109
Are there any resistors or other current regulating parts on that kind of lamp? If not and there’s just LEDs, you’d need to power it with a constant-current power supply. From what I understand, it’s probably not the best to put a standalone PWM circuit like your Ali link after such an LED driving PSU, but I’d do some research into that, both generally and in regards to your chosen fixed current LED driver.
A better option is to buy a constant-current LED driver with a PWM or current varying input or potentiometer on it.

[Anonymous No.2462167]

>>2462133
>Are there any resistors or other current regulating parts on that kind of lamp?
no just the diodes on a pcb
>buy a constant-current LED driver with a PWM
my driver is just that (I have the HLG-240H-24A - 'A' means 'adjustable').. I just wanted a way to dim the input to 1 specific LED strip with out dimming all the rest connected to the driver.. I was hoping I could do that with the PWM connected to that 1 strip that needed additional dimming separate from the rest instead of buying more drivers, running more wire, cluttering etc

[Anonymous No.2462174]

>>2462167
Ah, but how are they connected together? Series or parallel? Are there individual outputs for each LED strip? If so you may need to take it apart to show how they’re internally connected.

[Anonymous No.2462207]

connected in parallel
there is nothing to take apart, the LEDs are SMD and are soldered onto a thin PCB (aprox 1mm thickness,white solder mask), which then adhered to a heatsink.
>internally connected
like how they are routed on the PCB, or how thick are the traces? otherwise they are all connected to Vcc the other GND

[Anonymous No.2462215]

the strips are connected in parallel to the driver; the 96 LED's on the strips are connected in series
there is nothing to take apart, the LEDs are SMD, soldered onto a thin PCB (aprox 1mm thickness,white solder mask), which then adhered to a heatsink.

[Anonymous No.2462230]

>>2462215
They’re directly connected in parallel? Then when you turn off one strip momentarily with a PWM circuit you’d just end up sending that current to a different LED string. Technically you could arrange a circuit where you PWM them all synchronously at the same time as adjusting the constant-current driver, but that would be quite the pain if it would even work. More likely to work would mean using some current sinks instead of PWM circuits, but that would be even more of a pain to get set up.

There’s basically no easy way of splitting up the current going from one constant-current driver into multiple LEDs, so I’d recommend just buying seperate adjustable LED drivers for the strips you do want to adjust. Reduces the chance of thermal runaway between strips too.

Also by “take it apart” I meant the driver, not the LED strips.

[Anonymous No.2462238]

I’m trying to fix a neon/light box sign for the motel I manage. Everything works, but the previous manager had the switch removed that controlled the power, as well as switching the “no” in “no vacancy” on and off.

There’s a pool box under the sign with access to the wiring. It looks like there’s some wires that were capped. What would be the best way to go about testing these wires to see if they’d be what I’m looking for?

[Anonymous No.2462246]

>>2462215
>>2462230
Thinking more, it would be reasonably easy to have a variable current sink on each LED string (probably doesn’t have to be HV rated), a zener or TL431 across the whole network to prevent the driver outputting too much voltage, or better still an op-amp circuit feeding into the adjustment pins on the driver (or bodging into the potentiometer’s pins) to turn the output current down when the voltage drop across regulators gets too high. You could get the minimum voltage drop across the regulators with diodes and use that for feedback quite easily. If the output voltage can’t get more than 5-15V above the standard LED operating voltage then this won’t be required, but I’d still use such a circuit to make it run cooler.

Pic related. Use the more complicated adjustable current sink if you're at risk of frying the potentiometer over an LDO's reference voltage, or want to reduce the minimum dropout.

[Anonymous No.2462264]

>>2462238
>What would be the best way

1) find the breaker/fuse that powers the sign. if it's not labeled, then you have to switch them off/on one by one.
2) uncap the wires and measure voltage while someone else flips the breaker on/off

if you have no voltmeter, dont rely on those silly beeping voltage testers. wire 2 incandescent lamps in series, connect one end to ground and the other to the wire under test. lamps should appear half-bright. if they're fully lit it means you have 220V.

[Anonymous No.2462279]

can I put several of those piezos in parallel?

[Anonymous No.2462286]

>>2462279

takes less time to actually try it and see than to solve one recaptcha.

[Anonymous No.2462294]

I asked last thread about a GameCube controller
I got it fixed, couldn't find any damage to the trace but I found a guys website that documented the pin out of the chip and traced all the traces so I ran a bodge wire.
Works great

[Anonymous No.2462303]

>>2461449
>waaah it's not pretty, mommy i don't like it! give me something with animations and flashy colors!

[Anonymous No.2462309]

>>2462279
Yeah, but if you’re already pushing marginally high peak-currents you may want to reconsider, or else put a choke in series with the lot.

>>2462294
Nice work!

>>2462303
Are you implying that the software in his pic is what he’s complaining about? Because that’s just LTspice, it doesn’t do PCB design at all.

[Anonymous No.2462339]

>>2462286
Only have one piezo.
>>2462309
It's only drawing around 100mA. The MOSFET is barely getting noticably warm.

[Anonymous No.2462521]

>>2462339
what's the voltage ripple on your power rail though? may well need a lot of capacitance on it

[Anonymous No.2462549]

If you leave the cover open and apply formic adhesive then you can etch smaller circuits than a breadboard.

[Anonymous No.2462550]

>>2459863
>>2462549

[Anonymous No.2462644]

If I'm winding a motor does it make any difference to have all the individual coils in a phase in parallel rather than series?
I'm having trouble with too high current draws

[Anonymous No.2462649]

>>2462644
>too high current draws

easy to fix.
decrease the voltage or increase the number of turns.
wiring coils in series will do that, as will mixing series and parallel.

[Anonymous No.2462669]

What the fuck am I doing wrong with this thing. I can't seem to get it to properly route wires and it seems like rhe fill tool is what I need for the larger power trace but it's just overlapping with the ground. It took an hour just to get the fucking sata differential pairs to work with a center ground and not look too much like shit.

[Anonymous No.2462713]

>tfw you find out that there's a smaller version of a connector because you ordered it by mistake

[Anonymous No.2462755]

>>2462669
In general, don’t try to use a fill for something that isn’t a fill. Use net-based trace sizes, or else use copper polygons that you set the net of.

>SATA differential pairs
>not length matched
ishygddt
also sort out your pin grid alignments, looks like either you’ve placed a component such that its pins aren’t aligned, or you picked the wrong grid.

>>2462713
You got yourself a banana litter. Gonna give some away?

[Anonymous No.2462779]

>>2462755
Grid is retarded, the micro sata connector is centered to the board perfectly because it is necessary for it to be so it slots into the drive bay.
I found it incredibly annoying that nothing snapped to out of grid components but managed anyway
Length match may not be out of tolerance, it is sata 2. I'll check it though

I'll try polygons and see if I can get that to look good. This is my first time ever using a pcb cad so

[Anonymous No.2462820]

>>2462779
You can create custom grids, and by temporarily placing the centre of the coordinates on one of the connector pins you should get good alignment.
Try Mitxela’s PCB melting plug-in while you’re at it, though I’m not sure if it works with length-matched traces, which kicad can do itself IIRC.

[Anonymous No.2462853]

>>2462820
Which footprint looks lit of alignment to you though? The micor sata I made myself using the spec sheet of the connector and a significant modification to the Normal sata footprint
The m.2 was pulled from someone on github and I just deleted some pads to get the traces through, it is also to spec. The top row of pins are offset from the bottom if that is what looks wrong

[Anonymous No.2462857]

how the fuck is the new iphone doing the thing where you can point it at a satellite and send sms through it with no big antena ?
thats cool as fuck, it gives you coverage anywhere on the planet

[Anonymous No.2462868]

>>2462853
The side-to-side wiggling of traces going to the bottom footprint look funky.
Is it a single-sided design? Not sure how edge connectors work in that situation, but I suspect it’s fine.

>>2462857
I suspect there’s a small phased array grid of antennae behind the back cover in order to make it accurate enough, but even a small one would take up a fair bit of space.
Also it’s a subscription thing, and there’s still no headphone jack or SD card slot, so that’s a no from me. I’d rather bounce CW off the ionosphere.

[Anonymous No.2462872]

>>2462521
I can't tet that unfortunately. So I should just put a big cap in series with the output?

[Anonymous No.2462877]

>>2462872
You could try pulsing it at 50Hz and measuring the AC voltage across the power rails with a DMM. Depending on how it's measured, it may give you some idea of the ripple.

Big cap on the power rail is a good idea, same for putting a choke or small resistor in series with the piezo. If you use a 1Ω resistor, and it's a 12V power rail, then you know the peak current isn't going to be more than 12A, and you can calculate the RC constant which tells you how long the pulse will go for, from which you can figure out how much ripple voltage there will be.

[Anonymous No.2462878]

>>2462872
>>2462877
Wait disregard that and post a circuit diagram, that pic doesn't look like a simple push-pull driver but rather something single-ended with an inductor thrown in for voltage boosting or even resonant driving.

[Anonymous No.2462920]

Think the pcb place would ban me for ordering this? It's my art.

[Anonymous No.2462922]

okay, big brain time
this is current sampled over 2ms
the current is 5V between 10 and 20ma, (if we don't include the big spikes),
i need to measure that current but NOT any of the spikes
how can i do that?
Basically it doesn't matter at which point in the graph i get the reading from as long as it's not the spike.
do you think a simple braindead circuit with one op amp like lm324 and esp32 ADC pin will do the trick?

[Anonymous No.2462924]

>>2462922
man plan so far to eiminate spike readings is to take say 5 samples and then take median from them

[Anonymous No.2462927]

>>2462922
oh and by opamp and mean taking the rading via 0.1ohm or 1ohm shunt

[Anonymous No.2462941]

>>2462920
>Think the pcb place would ban me for ordering this? It's my art.

Why would they? Most of the people in the "pcb place" wouldn't even be able to read it.

[Anonymous No.2462943]

>>2462941
one of the few things china has done right. that robot is fucking cute, way better than your average restaurant worker

[Anonymous No.2462970]

>want to convert lathe to CNC, long term project
>uses a 3ph motor, VFD, and a contactor
>dont understand how it all works, but slowly buying the parts

I bought the Siemens contactor on Ebay with a low bid.
They screwed up and sent me a "Time Relay" and two "Monitoring Relays"
I told them, they ended up sending the contactor and told me to keep the other shit.

So now I have
https://mall.industry.siemens.com/mall/en/WW/Catalog/Product/3UG4512-1BR20

https://mall.industry.siemens.com/mall/en/ww/Catalog/Product/3UG4633-1AL30

https://mall.industry.siemens.com/mall/en/WW/Catalog/Product/7PV1540-1AW30

What exactly is this stuff used for, and would it be useful in any way to shove into a lathe control box.?

[Anonymous No.2462973]

>>2462970
Do you have 3-phase AC service installed in your shop?

[Anonymous No.2462981]

>>2462973
No this is a garage on a small lathe
The VFD is going to be a single phase 110v which is used to control the 1hp 3phase motors speed and direction through pwm

It’s only using 3 phase for control, not for large power

[Anonymous No.2462986]

>>2462981
This is the VFD that it calls for, so there will be only 3phase between that and the motor

https://www.wolfautomation.com/vfd-1hp-115v-single-phase-ip20-med-duty/

[Anonymous No.2463022]

>>2462981
>It’s only using 3 phase for control, not for large power

that's a weird statement.
the only high power thing in a lathe is the motor, and it's 3-phase.

anywayzz, what you normally use the contactor for is safety, since you already have manual on/off control on the VFD.
example: you pass the 3-phase power from the VFD to the motor thru the contactor.
and the contactor is activated using an emergency shut-off push-switch.
when the lathe grabs your sweater and starts killing you, you use your free arm to hit the switch, and you end up disabled instead of dead.
which makes your wife even angrier coz she woulda preferred the life insurance money.

[Anonymous No.2463051]

Ok I cleaned up the traces and also populated the rest of the components. Now I have to fit a usb 2.0 hub in here, a Usb3 hub, type C controller/PD whatever, a connector for Usb3 input, and usb2 input, and find out if the 5v rail provides power at the microsata connector.

Yes the type c is nearly rubbing on the SD card slot, no its not my problem

[Anonymous No.2463054]

>>2457724
>You're annoying as fuck to assume that you never consider yourself to be a problem, its always someone else, I bet you think highly of yourself while you look for the worse in other people
Welcome to /diy/.

[Anonymous No.2463057]

>>2462941
OSH park is 100% made in USA. Pretty sure they have to separate them and mail them by hand, too.

[Anonymous No.2463163]

>>2462878
Sounds fairly complicated. Well thanks for the effort anyways. I'll just try it when I get more piezos.
https://www.instructables.com/Make-Your-Own-Super-Simple-Ultrasonic-Mist-Maker/

[Anonymous No.2463165]

>>2462922
Diode clip the spikes down before the ADC and a second amplifier so you can get a higher SNR.

>>2463051
What kind of board is it? I could maybe understand the card reader and M.2, but the USBs as well? Also don’t forget the fastening screw hole(s) for the M.2 part.
>through-hole part
So is it a 2-sided board after all? Gonna want to add a rear ground fill and some stitching vias. And bypass caps, but I guess you’ll get to that.

Any references for custom SATA boards? Ever looked at custom PCI/E boards?

[Anonymous No.2463209]

>>2463163
Oh I get it, it’s a class-C “amplifier” that relies on the Q factor of the piezo to stop the IRLZ44N from overvolting. At 50% duty-cycle. I’m not too sure how adding series piezo will impact things, but I think it will just make the whole thing quieter. Parallel piezos will make it louder, but you’ll likely get mismatch between the resonant frequency of the LC circuit and the mechanical resonance frequency of the piezo, assuming you’re not getting that already.

I can suggest a better driving circuit if you’re interested.

[Anonymous No.2463255]

Anyone know how I could tell how much current this thing would draw? I want to power it with maybe 1000mah of battery maybe, would it only last a couple hours or?

[Anonymous No.2463258]

>>2463255
Soz pic

[Anonymous No.2463264]

>>2463165
>What kind of board
Hard drive adapter for a thinkpad x301/x300 and in theory a t400s
I will print a new drive bay cover and it will add USB and SD card to the laptop.
It was going to be a 1 sided board if possible but it is no longer worth it
Still haven't measured for a screw clearance yet but that is planned yea
What's a bypass cap do?

[Anonymous No.2463265]

>>2463258
>not providing link an only having a picture

this is equivalent to describing your problem and not having your laptop with you and expecting us to provide help.

[Anonymous No.2463268]

>>2463265
Here soz
https://m.aliexpress.com/item/4001032411407.html?aff_fcid=9c0452064df644a19dcbf039add780d7-1662683278934-00425-_A4gUEH&tt=CPS_NORMAL&aff_fsk=_A4gUEH&aff_platform=portals-tool&sk=_A4gUEH&aff_trace_key=9c0452064df644a19dcbf039add780d7-1662683278934-00425-_A4gUEH&terminal_id=e5a2f9257eeb4fd0ae2bc07c251b10a4&afSmartRedirect=y

[Anonymous No.2463270]

>>2463268
>https://m.aliexpress.com/item/4001032411407.html

just for future reference, anything past "?" in the link is not needed.

I managed to find same one on amazon and question was answered by same seller name on ali and shows that 2A is the max

[Anonymous No.2463277]

my google-fu is failing me
I'm trying to add a custom symbol and footprint to KiCad. Do I really need to manually measure and place the pads? Is there really not just some way to plop down some standard boilerplate TS8 SMD footprint and just label the pins?

[Anonymous No.2463283]

>>2463277
let me rephrase.
Is there really no way to select boilerplate component templates in the footprint editor? I really need to re-create the entire thing from scratch, even if it's a standard component footprint like TS8/TSOT-23-8?

[Anonymous No.2463285]

>>2463283
>>2463277
nevermind I'm fucking braindead and illiterate evidently

[Anonymous No.2463306]

>>2458250
thats cool ive always wanted to build a bluetooth speaker. did you use your own random parts or was it a kit?

[Anonymous No.2463315]

How hard is it to solder to these button cells?

[Anonymous No.2463317]

>>2463315
no

[Anonymous No.2463319]

>>2463315
I'm trying to replace the battery in these Bose earphones, you can get the stock batteries from aliexpress for a reasonable price, or you can pay out the ass for these ones with tabs welded onto them for soldering. It's viable to replace the battery only for the cheap non-welded version in terms of cost.

Is it even possible to solder those batteries directly? Any way around this?

[Anonymous No.2463322]

>>2463264
Sounds neat. A bypass cap is just to get rid of ripple on your power rail. I recommend putting one at the VCC going into the M.2 and the VCC going into the SD card. Use a ceramic of at least 100nF, maybe up past 10uF, but read the spec sheet for SATA before committing to large caps. A choke may be a good idea.

>>2463319
You can solder onto them, the amount of damage you do to them will depend on the chemistry as well as your technique. If you do solder them, cover the surface in rosin flux paste (or even solder paste) before tinning, and reflux before joining the wire. Go in at a high temperature so you can get out fast.

A much better idea is to buy or make a little spot-welder. $30 on alibay last I remember.

[Anonymous No.2463325]

>>2463319
https://www.amazon.com/uxcell-LIR2032-Button-Battery-Holder/dp/B00QC2XJ9O
Solder the battery leads to that and put a LIR2032 in there.
>>2463322
>spot weld a coin cell
Stop smoking crack, Hunter.

[Anonymous No.2463331]

>>2463325
>LIR2032-Button-Battery-Holder/dp/B00QC2XJ9O

that's never a viable idea coz of space constraints.
unless you have the battery hanging out by its wires, or glued to the outside, in Mad Max style.

[Anonymous No.2463332]

>>2463325
Pretty sure there is nowhere near enough space for that inside the little earphones.

>>2463322
tfw I have full size welders but not a spot welder. I could maybe ask someone at a nearby vocational school to use theirs. What profession normally uses those?

[Anonymous No.2463334]

>>2463331
>>2463332
>it won't fit
You'll never be a good rapist. I mean hacker. HACKER.

[Anonymous No.2463335]

>>2463334
brb changing classified ad to "the electronics rapist"

[Anonymous No.2463343]

>>2463335
That'll just scare the victims away. Call yourself "the electronics therapist" because rapist is still in the name and they can't say you didn't warn them.

[Anonymous No.2463345]

>bucketful of free 24V 2.5A power bricks at work
Cool, I’m grabbing 4, will grab more if I can think of enough projects that would use them. 24V is good since it’s nice and high for analogue circuits (with enough headroom for a pass transistor to kill the ripple) while being not too high for JFETs and op-amps. And they’re isolated so I can chain two in series.

>>2463325
How do you think they get the tabs on there in the first place? It’s not much smaller cross-section wise than spot-welding an 18650, which is what those eBay spot-welders are designed for, shouldn’t be too tough, you’d maybe just need to bend the copper electrodes a little closer together.

[Anonymous No.2463347]

>>2463345
>How do you think they get the tabs on there in the first place?
They have little elves glue them on with red blood cells and leaded mucous.

[Anonymous No.2463354]

>>2463209
Well, I can use the trimmer to accurately adjust the frequency of the piezo.
I don't know if I can implement it.
As for your auggestion, if it's something time intensive you have to draw yourself, you're propably wasting your time on me.
I would try looking into just adding more piezos first and if that ends up producing more fog, it'll be good enough for me.
If it doesn't work I would try implementing modifications to the circuit.

[Anonymous No.2463357]

>>2463306
Random parts I had lying around

[Anonymous No.2463368]

>>2463354
>if it's something time intensive you have to draw yourself
Pic related, a resonant half-bridge driver, using an inverting comparator and an oversized (but non-polar) capacitor. I've built a couple myself. Can probably replace the gate driver and FETs with something like an L298N or equivalent. I suspect it will resonate at the mechanical frequency of the piezo.

[Anonymous No.2463376]

>>2463368

i dont think that'll work.
to get the piezo to do significant work you gotta hit with 50-200V.
that's what the coil in the original circuit did.

[Anonymous No.2463377]

>>2463315
it is technically possible, I soldered to a 1616 once upon a time by scratching it first
don't do it it's bad, you're supposed to spot weld them

[Anonymous No.2463382]

>>2463376
The thought is that the mechanical resonator itself acts as an inductor to some extent, but maybe that's a false suggestion. You can just put an inductor in series with the piezo, but that again has the issue where it makes an electrical resonance frequency that might be different from the natural mechanical resonance frequency.

[Anonymous No.2463385]

>>2463383
>>2463383
>>2463383
belated /mcg/ microcontroller general refresh

[Anonymous No.2463393]

>>2463315
don't use solder or a temp above 360C, these cells tend to explode if over heated

[Anonymous No.2463397]

>>2463385
Fuck digital. /ohm/ is more helpful and more repairable than your gay threads.

[Anonymous No.2463402]

>>2463397
I OP for both generals, you can see it as a containment thread for tarduinos, or as a place for dumping code that would clog up the busier electronics general. It serves its purpose, and there's some interesting conversations there on occasion.

Just posting the refresh link here because /mcg/ died before I got home and I imagine a lot of the people there are on here also.

[Anonymous No.2463478]

>>2463382
How big do you suggest the inductor should be?

[Anonymous No.2463480]

>>2463397
I lurk over there, and all arduinos serve is making it easier to access the pins (all or most) on the chip and ELI5 level of understanding of the pins. and adding some quality of life as well.

On the other hand, whenever my google searches end up being a thread on arduino forum, the people replying to those threads act like they are practicing the purest form of electronic creation, and have the biggest inflated egos.


>>2463402
I do both if space is not a issue and its not a dip type I'll most likely use a tarduino, I do however wanna try my hand at pcb design and see what it takes to make a pick n place machine, and just build a smallish one.

[Anonymous No.2463557]

I'm building something. Please clap.
https://www.youtube.com/watch?v=-Wp95itbiAA

[Anonymous No.2463570]

why is type c so fucking hard to implement?
Why doesn't anyone just make a fucking ic you drop in, connect up all 18 pins to, connect your system io and it does everything
I'm too dumb to figure out all these separate components wtf

[Anonymous No.2463595]

>>2463570
aaaaaaaaaaaaassaa
>find cypress CYD4155
>does almost perfectly what I want
>out of production out if stock

[Anonymous No.2463596]

>>2463570
Buy some cheap piece o' shit with the stuff you need on it and patch it in until you can get the right parts.

[Anonymous No.2463597]

>>2463596
that's hard when I want to produce and sell them in low volume

[Anonymous No.2463599]

>>2463597
You're going to sell zero as it is.

[Anonymous No.2463600]

>>2463599
bet

[Anonymous No.2463606]

>>2463600
Okay. Zero dollars says you sell zero devices unless you find a way to plug them in and use them.

[Anonymous No.2463626]

>>2463606
The type c isn't all of it. I could just axe it and use normal usb 3. PD is nearly pointless and why tier, also has better solutions already, but I want it anyway fuck

[Anonymous No.2463635]

>>2463626
What are you building? Can you get away with using wireless for comms and a big honkin' DC barrel jack for power?

[Anonymous No.2463652]

>>2463478
I’ve no idea. A simple guess would be making a resonant oscillator with the capacitance of the piezo at the desired frequency, but if you need to match the two resonances the only option would be sifting through the differential equations.

Also there’s a resistor and diode clamps before the comparator input, to not overvolt it.

There are resonant driver circuits out there, like this one for an ultrasonic cleaner, looks like they use transformers.
You might be able to use my circuit without an inductor but with a transformer, but it’s kinda sketchy in my experience.

>>2463626
If you’re Lenovo card anon, wouldn’t the voltage boosting circuitry required by USB PD not be the kind of thing you want on an SSD board? If you’re not doing voltage boosting, then I’d just use USB As. Can’t see how you’d wire them up either way though, unless they’re going to a connector other than the SATA.

[Anonymous No.2463670]

>>2458677
Field service technicians or engineers
I'm an fse with no degree though I had a background in electronics from my time in the Navy as an ET

[Anonymous No.2463673]

I have a very "hot" transformer that trips the circuit breaker and another one that "only" makes my monitor flicker when I turn it on and off. I didn't have much luck snubbing the inrush current with NTC thermistors I have on hand. I searched through a catalog, and it doesn't seem practical since the cold resistance needs to be quite high, about 100 ohm and the steady state current is about 5A and this combination is hard to find. So I tried a regular resistor to confirm that 100 ohm is enough. I then decided to make a relay delay circuit and simulated several different ones. I first tried with just one NPN transistor and RC, and I didn't like the idea since I cannot make R very high or the base current won't be enough for bjt to saturate. I then tried with a pair of PNP and NPN transistors and that worked well. But I like the comparator approach better. It is simpler and C can be much lower. the 3rd one with a PNP transistor seems a bit more preferable since the relay is referenced to ground and sees (almost) full 5v. What do you think? Also, is relay+resistor a sound approach? The resistor would have to drop a few volts for a brief second and it can get hot but once relay turns on, it is completely bypassed. But then what if the relay fails? I would probably need to make sure the resistor is large enough to remain cold. And I cannot increase its resistance too much, or it will drop most of the voltage and this whole approach won't work.

[Anonymous No.2463737]

>>2463652
>not doing voltage boosting,
Correct
USB pd is just the controller asks the power brick nicely to output 20v at 3-5A, and then it isolates that to the 20v rail.
Yes its for the card. I don't think it's going to cause any interference.

The 20v will be it's own connector which routes via a cable to the laptop power connector.
The USB 3 and USB 2 source are actually also coming on their own cables. Rather conveniently there is a mpcie slot about 8cm cable length away and an existing pass through in the chassis.
Not only am I making this card but also making a mpcie card with a Usb3 controller and hopefully tapping usb2 from the socket.
The microsata is just for the storage drive and a very convenient power source for 3.3 and 5v at quite a lot of fucking amps.

After tons of research and pain, I stumbled across a project made by popcorncomputer. Their project itself is very interesting, but I am infinitely thankful that they open sourced their schematics and are using the same PD controller I was going to use.
I'm going to basically clone their entire implementation of USB c. Between that project and a dead thinkpad t495 board I have that also has the same ic, I should be able to get this working.

I haven't made much progress today as it was all research, I did finish up the SD card controller. I'm currently planning on using a MA8121N because of shear laziness. It is a entirely integrated solution that I basically whack on and run traces. It's out of print but still greatly available. My ender 3 pro came with a USB to microsd which uses this chip, which is how I found out about it.

[Anonymous No.2463762]

>>2463673
>But then what if the relay fails?

- buy model that never fails. just add "NF" to end of part number when you order it.
- tie wrap a thermal fuse to the resistor, and wire it in series.

[Anonymous No.2463891]

tht-mounted spade connectors are pretty neat actually

[Anonymous No.2463933]

pre-testing schematic for >>2460180 with adjustable current limit latch-off. the reed switches and the actual coil to the relay will be connected as some sort of wiring loom. neutral is my 0v rail for logic for current sensing purposes, but i've got protective earth wired to the chassis of my switch and potentiometer. which should be reasonably safe, maybe. i'll see if i can bust out the insulation tester to see for sure.
fully tht single-sided board

breadboard later, sleep now

[Anonymous No.2463991]

>>2460180
>Cliff Quicktest
What's the point of it? Is it just a glorified extension cord for bare wires with a fancy knife switch instead of a regular switch? Seems quite expensive for what it does.

[Anonymous No.2463992]

>>2463991
The switch affects the color temperature of the Dubai LEDs. Every lumenophile knows that.

[Anonymous No.2464036]

>>2457597
Is it normal to have 120v ac across two heatsinks?
I didn't think having mains voltage across massive side by side heatsinks was kosher
Pic semi related, looks like it might actually be the board, the two heatsinks on the right are full mains

[Anonymous No.2464038]

>>2464036
Sometimes the tab is not isolated in which case it's normal. Did you or anyone else do any work on the board?

[Anonymous No.2464039]

>>2464038
>isolated
insulated (from the heatsink)

[Anonymous No.2464049]

>>2464038
Pulled from the trash, it's a peltier fridge, appears to be a 12v power supply but the whole unit is a mess.
This power boards hit and miss, sometimes it won't turn on and only outpost like 1v and makes a hissing whine.

[Anonymous No.2464051]

>>2464038
When I say the heatsinks are 120v I don't mean potential to ground, I mean if you touch both of them at the same time it shocks you

[Anonymous No.2464060]

>>2464049
>Pulled from the trash
Return it to the trash.
>>2464051
>if you touch both of them at the same time it shocks you
Stop touching the heatsinks, Anon.

[Anonymous No.2464062]

>>2464060
>Return it to the trash.
but salvage the copper and aluminum before you do.

[Anonymous No.2464068]

>>2464060
>Return it to the trash.
No it's actually kinda nice and it's easily fixable. I'm probably going to put stronger piezo coolers in it. Things listed for like $700 new

>Stop touching the heatsinks, Anon.
I wanted to know if the chip was working or scolding but checking the temp,
My question is should heatsinks be at 120v right next to each other or is the chip screwed to it failed or is something else fucked up

>but salvage the copper and aluminum
It's thermoelectric type of fridge, no copper worth a crap but pretty heafty heatsinks
It's supposed to be a 60 watts, I'm going to open up the heatsinks and swap the peltier thing out with one that's two or three times more powerful
But it started working yesterday but still says error so I might just use it as is if I can figure out why it intermittently doesn't work

[Anonymous No.2464081]

>>2464068
>$700 new
The whole fridge or just the PSU?
>should heatsinks be at 120v right next to each other or is the chip screwed to it failed or is something else fucked up
Need schematics- or you could reverse engineer the board if you're so inclined. Using an o-scope will cut down troubleshooting time.
>But it started working yesterday but still says error
Maybe one or more of the Peltier modules is failing or dead?

[Anonymous No.2464094]

>>2464081
The whole fridge
It's a wine cooler, probably holds a couple dozen bottles
It's over priced for what it's made from but it's eye candy for upper middle class people

The internals are a joke, two mid range peltier coolers, two power supplies, and a ln mcu on it's own board to turn the two coolers on and off for the desired temp, two thermometers, assumedly, and the MCU board has two 2x8 segment style blue displays.

The pelteit can be upgraded to one that's 3x more powerful l, the MCU board was made by matures and can be replaced by a $2 chink Arduino, at this point I don't care about the thermometers or the power supplies.
The display is throwing "er" for one zone.
The whole fridge works sometimes and other times neither zones will turn on.

Im planning to replace the power supply, upgrade the peltier, and set it to always be on medium low. Screw everything else.
But it turned on yesterday and kept two liters of water kinda cool while sitting in the sun all day so I might just leave it as is if I can make it stay working

I just don't think mains on the power supply heatsinks is kosher and neither is that hissing noise one makes

[Anonymous No.2464099]

>>2463991
it's so you can pretend to be a big gay youtuber

[Anonymous No.2464118]

>>2464094
You could run everything from an ATX power supply pretty easily, and you'll have +5V standby voltage for your tarduino to switch the peltiers on and off.

[Anonymous No.2464120]

>>2464118
>for your tarduino to switch the peltiers
to switch the ATX +12V. You know what I meant.

[Anonymous No.2464122]

>>2464118
I thought about that but I scored a few of these things in the trash too and I'm short on atx supplies atm
>Shutting the peltiers off
I'm not sure I'm even going to bother, I might same day but I just want cold drinks while I'm processing the curb find hoard.
Probably just undervolt them and let inefficiencies in the fridges design even the temp out for me

[Anonymous No.2464123]

>>2464122
>>2464118
Forgot pic, got a few of these things

[Anonymous No.2464131]

>>2464123
That looks like it'd work fine. Maybe plug it into a timer for safety.

[Anonymous No.2464139]

>>2464131
All I'm hearing is wire it directly into the mains and bypass the circuit breaker

[Anonymous No.2464156]

Looking for a small switch/button than can be activated by the touch of a pen. Not sure what the best way to accomplish this would be. I can't find a button small enough so I'm considering just running wires through a 3d printed board that will keep them just close enough together that sticking the tip of my pen between them will make the connection. Does anyone have a better idea? I'm just wanting to make a very compact button panel

[Anonymous No.2464162]

Does decoupling capacitors actually matter or is it a big capacitor corp lie

Seriously though, when/why should i use them?

[Anonymous No.2464164]

>>2464156
Halleffect sensor connected to a transistor with a magnet in the pen
Or a magnetic door security switch thing that has magnet on the door

What are you doing

[Anonymous No.2464166]

>>2464156
micro tact switches

[Anonymous No.2464169]

>>2464166
He wants Le magic pen though

[Anonymous No.2464170]

>>2464162
It's all about keeping the niggers out of the woodpile. You'' understand when you're ready.

[Anonymous No.2464173]

>>2464169
Like Bill Cosby's doodly-doo pen?

[Anonymous No.2464174]

>>2464173
Mortimer Ichabod Marker
>squeaky-squeaky-beep-boop-squee

[Anonymous No.2464177]

>>2464173
>Bill Cosby's doodly-doo pen
I have no clue what you talking about but I'm assuming he wanted to larp Harry Potter or doctor who

[Anonymous No.2464184]

>>2464177
https://www.youtube.com/watch?v=7eIq8R3Pjeg&t=233s
It's like Harry Potter except black and rapey.

[Anonymous No.2464194]

>>2464184
Idk what it is
It being from a different era
Me being racist
Knowing you shouldn't let niggers around your kid
Knowing bill Cosby is a rapist
Or if it just looks like they tried to polish a thug
But this video is really creeping me out

[Anonymous No.2464197]

>>2464164
I was looking at those bot it seems like it would be too difficult to activate just one of them. I'm going to cut off the bottom of a scientific calculator and make a new, compact keypad for it. I just want a smaller calculator instead of something the size of my phone.
>>2464166
I can't find any that seem like they would be the size I want. The smallest I found would be 6x6mm which comes out to a good size (picrel) but it would be nice if I could come up with something more compact
>>2464169
Not necessarily but I want to be able to go from writing to using a calc without putting my pen down. It seems like it would be more efficient
This is my first electronic project though so it's really just for fun and to get into making something. I'm gonna 3d print a new case

[Anonymous No.2464201]

>>2464194
Probably all of the above. Kids used to carry severed rabbit's feet (dyed fun colors!) in their pocket for good luck back then. Star Wars, pet rocks, disco. It was a weird time overshadowed by the excesses of the 80s.

[Anonymous No.2464211]

>>2464156
Capacitive touch pads? Well at that point it’s basically a capacitive touch matrix like what you’d find on a screen, but yeah with common PCB fabrication techniques it would be pretty easy to get 2mm x 2mm “buttons”. The harder part would be getting the stylus to trigger them, I think a small conductive rubber dome would do the trick.
Even better would be a custom membrane keypad, which may be feasible through some online prototyping services.

>>2464197
>assio
nice

[Anonymous No.2464229]

>>2457597
>>2457724
>>2458416
>>2458463
is it a bad idea to use an mdf box as a case for a power supply?

[Anonymous No.2464242]

>>2464201
As a 90s kid, I found a lucky rabbits foot at school ones and was freaked out when I found claws on it.
I thought it was weird they'd go into so much detail on a reproduction
It is a reproduction right
Right?!?!
Surely a poor animals foot wasn't used to make a fucking keychain accessory
Mfw I found out nobody's making fake rabbits feet keychains
Appalled.jpg

[Anonymous No.2464245]

>>2464229
Should be ok, but I’d recommend sealing it with PVA or paint or whatever to ensure it can’t soak up humidity.

[Anonymous No.2464268]

>>2463991
It's just a quick and safe way of connecting mains appliances that don't come with plugs. I've got a cat that likes to get her nose into whatever I'm doing so it's good for avoiding that. Might be useful at work also. Beats soldering an IEC socket onto everything.

[Anonymous No.2464288]

>>2464242
>Surely a poor animals foot wasn't used to make a fucking keychain accessory
Just wait until you find out how they make Rubik's Cube. Those poor babies.

[Anonymous No.2464327]

Is there any way to get a video feed from these detached phone cameras?
I'd like to use them for home security.

I'm sure there are many more viable ways of doing this, but not only do I already have 5 of these, but I told a friend I'd try to make a security system with them, so I'd like to do it and show him that his old phone cameras have been put to use.
It also seems like a nice project for a beginner.

This is my first time in this general, any help is appreciated.
I am yet to buy a soldering iron.

[Anonymous No.2464332]

>>2464327
Arent they behind 7 DRMs

[Anonymous No.2464334]

>>2464332
No idea, altough there might be a high chance.

[Anonymous No.2464337]

>>2464327
You need a host computer and hardware drivers for the host OS.

[Anonymous No.2464339]

>>2464337
That I can manage. I have a few laptops and I could try some generic Linux drivers.

[Anonymous No.2464340]

>>2464339
Try to find pinouts or datasheets for your cameras. If you see D+ and D- then you can just solder a USB cable to those traces/pins.

[Anonymous No.2464344]

>>2464340
>Try to find pinouts or datasheets for your cameras.
Do the manufacturers even release them? I've no idea what to search for to find such a thing for these cameras.
>If you see D+ and D- then you can just solder a USB cable to those traces/pins.
Got it.

[Anonymous No.2464349]

>>2464344
>Do the manufacturers even release them?
Usually not, but people sometimes figure it out and post it on the internets. Elektrotanya has a lot of documentation for various stuff. There are Russian sites with a lot of technical info (use Yandex & translate).

[Anonymous No.2464351]

>>2464349
I see. Thank you for the help.
Seeing as I have quite some cameras I'm hopeful I'll find something.

[Anonymous No.2464357]

>>2464351
You're welcome, Anon. I'm sure you'll get at least some of them working as long as you don't vaporize them with your iron.

[Anonymous No.2464392]

>>2464327
>a nice project for a beginner
Honestly I'd say it's more of an intermediate project, possibly advanced if there are any big hurdles I haven't thought about. Something I personally wouldn't undertake unless I had a pretty good reason. Finding the ID of that tiny SMD connector, then having to solder it on a self-designed PCB, figuring out and converting the protocol, getting the video data to travel a long distance without altering signal integrity, and probably not having a flood of existing well-documented similar projects to go off of all make it not terribly beginner friendly. Well it would be easier if there happens to be an existing protocol-transforming board on ebay or whatever.

Got a part number for the camera?

>>2464340
Laptop webcams are often USB because it's easy and carries well across the 10-30cm from the webcam to the mobo, but phone cameras? Especially the kind of high-definition cameras that likely output more than 480Mb/s? Just 1080p at 30fps with 24-bit colour is 1.4Gb/s, so the forward-facing main cameras of modern phones would be well into the domain of USB 3. Considering the short distance to the board, I'd bet they'd not be using USB at all but some sort of more specialised camera protocol, perhaps like what those raspberry pi cameras use.

[Anonymous No.2464401]

>>2464392
So easy an Indian caveman can do it. Granted, this won't work with proprietary appleshit 64K camera arrays that upload to the clood through Hillary's closet email server relayed to a chink spy satellite. Starting at $5999.99 and needs a proprietary cable ($499.95, sold separately).

[Anonymous No.2464404]

>>2464357
>I'm sure you'll get at least some of them working as long as you don't vaporize them with your iron.
Let's hope!

>>2464392
Well, let's take it as a possibility to learn about new things in a new field, even if it takes time.
If I do get far enough with this, I'll be sure to document whatever I find.
>Got a part number for the camera?
Here're the part numbers from 2 cameras, back and frontal, taken from a Xiaomi phone, each with 2 serials (I imagine these aren't exclusive to Xiaomi and might be found on many other phones of other brands):

Cam 1 front: L02AF10
Cam 1 back: 1YJ341

Cam 2 front: L4H7FB0
Cam 2 back: 1HJ381

>>2464401
I've watched an Indian guy on youtube do something similar across many videos. Even without knowing how to solder I can tell they're fake. It's quite funny watching him trying to mimick a pre-recorded video as it falls out of sync with his movements, lol.

[Anonymous No.2464428]

>>2464392
>perhaps like what those raspberry pi cameras use
You're right. I wrongly assumed the camera would fallback to a lower resolution and framerate due to lower bandwidth of USB 2. Pi or similar SBC with MIPI CSI interface would probably be the least work.
>>2464404
I was wrong. Forget everything I said unless your cameras are 20 years old.

[Anonymous No.2464434]

>>2464428
May still be the case for the shitty selfie camera, though it only makes sense from a design perspective if they've already got a USB hub IC. I think my laptop uses such a hub IC for the webcam and for the bluetooth controller, but I'm unsure if mobile phones have such an IC.

[Anonymous No.2464455]

>>2464401
Can you fucking do that?
I though you need to do mipi to USB or something

[Anonymous No.2464466]

>make simple comparator circuit
>it oscillates
it's all so tiresome

[Anonymous No.2464473]

Is it easier/cheaper to send off two separate cad files for two separate pcbs to be made or join them with a simple break off point and send one file to be made?

[Anonymous No.2464478]

>>2464229
>bad idea to use an mdf box as a case

yes, very bad.
which is why nobody uses 'em.
- ugly
- joinery needed or else gets even uglier with screws and butt joints
- heavy
- ventilation needs lots of big ugly holes
- thick -- cant mount switches, pots, banana jacks, meters, or strain relief for power cord

if you have an old/dead PC power supply, that's makes a cheap and easy-to-make case.

[Anonymous No.2464499]

How do I make these soil meters waterproof ? I need them to stay in my garden under all weather conditions.
>https://a.aliexpress.com/_m0kIv6G

[Anonymous No.2464503]

>>2464499
this is easy. umbrella or condom.

[Anonymous No.2464504]

>>2464503
I'll use electrical tape, put a condom over it and then add a cute little umbrella :)

[Anonymous No.2464505]

>>2464499
They're not good at all. The plating will decay rather quickly from the current flowing between the electrodes if you run it constantly. Even if you turn it off most of the time, you'll likely not find it last more than a year or two. The capacitive ones are also bad since the solder mask tends to flake off after a while of being surrounded in soil. Google things like "soil moisture module corrosion" to see what comes up.

[Anonymous No.2464511]

>>2464505
Thanks a lot for the tip ! I found these bad boys, they're waterproof too.

>https://a.aliexpress.com/_mL6IsvI

[Anonymous No.2464522]

>>2463933
Wait a minute, my current plan is to detect overcurrent from the peak of the waveform, but of course current isn't necessarily sinusoidal. Does anyone know if appliances (especially ones with a rectifier) tend to just draw short sharp spikes that could peak more than 21A?
The alternative circuit is something with an operational amplifier to add gain (and an offset since the comparator doesn't operate well below 1V or so), then do an ideal op-amp "rectification" such that the output can push and pull current, then feed that into a ~10Hz low-pass filter so I can trigger the comparator off the average current instead of the peak. It's not RMS but who cares it's just for a potentiometer to trigger off. Also allows me to set it to 10Arms instead of 15Arms since the lpf will also take care of turn-on spikes. But the delay may mean it's not enough to prevent from tripping a breaker.

Pic related is what I think will work as a solution.

[Anonymous No.2464535]

>>2464511
>they're waterproof too
really doubt it
Those work by running a current directly through the soil. metal + current + electrolytes from soil = recipe for corrosion. Even stainless steel will probably rust in those circumstances. And of course once your probe is coated in corrosion it's not going to be accurate.
Other anon says the capacitive meters fall apart, but with those at least there is no metal-soil contact. So your best bet is probably to try making those last longer by coating them in polyurethane or some other material.

[Anonymous No.2464539]

>>2464535
If they run a really low AC current, or otherwise just turn it on for a really short amount of time, then the galvanic corrosion will be negligible. Though I bet that little PCB with the trimpot shown gives them the full voltage all the time, so you'd be advised to add a transistor that can cut power to the module itself, or just power it through a GPIO, which might be fine if the current is less than 10-50mA or whatever.

[Anonymous No.2464540]

>>2464535
>coating them in polyurethane

they work by measuring conductivity of soil, so insulating the probes is not recommended.
but if you make the probes easy to replace (you can get a spool of zinc-coated wire at the dollar store) then oxidation becomes a non-issue.

[Anonymous No.2464541]

>>2464540
No he's talking about the capacitive ones. So long as the layer is thin enough, the capacitive coupling remains. By sending a high-frequency AC signal across the two probes, you can measure the current, which will be a function of capacitance and ESR. That ESR will largely be made of the soil, so if you set the frequency such that the capacitive reactance is significantly lower than the ESR will be, it's pretty easy to measure with an envelope follower and an ADC.

[Anonymous No.2464554]

>>2464553
>>2464553
>>2464553
THREAD NOUVEAU

[Anonymous No.2464578]

>>2464522
>Does anyone know if appliances (especially ones with a rectifier) tend to just draw short sharp spikes that could peak more than 21A?
Is the load inductive? Is the circuit mitigating inrush current with thermistors or capacitance? Does it start normally and spike during operation, or does it start up all goofy?

[Anonymous No.2465049]

>>2464478
I dont have shit at hand except wood, and aluminum has been really expensive around here since the incident
what else could I use?

[Anonymous No.2465057]

>>2465049
plastic, either 3d printed or sheet acrylic, maybe even recyclable plastic food containers made from hdpe or whatever
metal tins too

[Anonymous No.2465630]

found these ebay and want to try something with all 3 together as 1 big mat. Should I wire them in parallel or series?

[Anonymous No.2465639]

>>2465630
response in new thread:
>>2465638