[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/diy/ - Do It Yourself

[Advertise on 4chan]


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


File: 555 stack.jpg (868 KB, 4032x3024)
868 KB
868 KB JPG
Thread detonated:>>2128102

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

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

>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
>Some guy’s list of electronics resources:
https://github.com/kitspace/awesome-electronics
>Microchip Tips and Tricks PDF:
https://ww1.microchip.com/downloads/en/devicedoc/01146b.pdf
>Li+/LiPo batteries required reading:
http://www.elteconline.com/download/pdf/SAFT-RIC-LI-ION-Safety-Recommendations.pdf

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

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

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

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

>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
>>
File: peaky.png (188 KB, 1688x738)
188 KB
188 KB PNG
Thread brought to you by the M06782 1.583GHz SAW filter
>>
>>2136785
wait oops it's the MP06782
>>
File: 1578012229509.jpg (92 KB, 1198x369)
92 KB
92 KB JPG
Reminder to serve me breakfast.
>>
>>2136757
yeah I put that resistor there mostly to protect the green LED that activates with the button but it will be removed once I attach a real frequency to the chip.
Thanks again!
>>
>>2136782
Ive finished reading electronics for dummies v3.

Most of the parts for my projects are in the mail. What should read/do to further my knowledge?
>>
>>2136861
Datasheets and example circuits.
>>
>>2136861
Learn to code (C and assembly).
>>
>>2136782
>what
OK, I'm a retard. What on earth would you do with a stack of 555 timers like that?
>>
File: 1521863139109.jpg (37 KB, 800x734)
37 KB
37 KB JPG
>>2136883
Take a picture of it and show it to people on 4chins.
>>
>>2136883
The only rational reason I can come up with is to drive something that requires more amps than a single 555 can provide, in a world where all transistor disappeared and only 555 timers are left.
>>
>>2136883
to make a 2.14k resistor
>>
>>2136923
>in a world where all transistor disappeared and only 555 timers are left.
I want to live in that world
>>
reminder to serve the ohmnissiah
>>
>>2136923
Ask /g/, I think someone did it for more current.
>>
File: 1614518315892.jpg (110 KB, 594x1024)
110 KB
110 KB JPG
>>2137050
I want a 555-based mobile phone.
>>
Wait I thought Photoninduction electrocuted himself to death?
https://youtu.be/0mGhhdPgXG8
>>
>>2137254
I thought he was fighting with customs about his mail-order indian bride
>>
>>2137287
It's over now, pajeet has been imported successfully. I think she's only legal for 2 years then she has to reapply again which means photon probably stops making videos again in 2023. I liked the old misses though she did look like a bit of a gold digger.
>>
>>2137075
>reminder to serve the ohmnissiah

the ohmnissiah is a fraud.
to ensure your dick-tip is brown for eternity, you must serve ONLY the sodOHMite god.
and sing his praises far and wide.
long live the god of sodOHMy.
>>
>>2137050
but there are transistors in the 555
>>
>>2137329

substantially every semi has transistors.
so they're easy to find, hard to extract with pliers.
>>
>>2137329
>he doesn't use vacuum tube 555's
>>
File: creb.jpg (1.78 MB, 4032x3024)
1.78 MB
1.78 MB JPG
Now I have to do it all again
this time with soldermask, because that etch resist ink is horseshit
>>
>>2137350
goddammit why does everyone but me have one of those dumb crabs
>>
File: file.png (537 KB, 449x401)
537 KB
537 KB PNG
>>2137359
>he STILL doesn't have his crab
>>
>>2137369
I-is that like a rubber duck for Electronic engineers?
I'd rather have a "snippy snap :DDDDD" spurdo one to be honest.
>>
>>2137371
It's the yee claw meme crab. It's not even a EE meme. The real ones you have to get from Japan or some shit.
>>
>>2137372
>japan
lmao got mine off aliexpress from the recommendation of >>>/g/csg

alongside my programming socks
>>
>>2137350
the presensitized boards are pretty cheap and work well btw
but you're probably just doing it for the experience so good luck with it

I've had best luck with presensitized boards and transparency, toner transfer is not good
I have an old 3D printer that I want to modify to mount a scribe to, then I can put a light coat of spray paint on the board and have it scratch out lines where it will be etched
>>
>>2137438
I mean, I had better result on boards where I drew traces, pads etc with a sharpie and fucking hydrogen chloride etched it (no peroxide added since thin copper)
>>
>>2137438
>presensitized boards
That would require some way of making a photomask though. Ideally, I'd use laser jet transparencies, but I have no printer. When my laser gets past customs I'll bolt it to my 3D printer and do the laser ablation method. Until then, it's paintbrush time.
>just doing it for the experience
If I weren't I'd probably just use JLC or whoever
>scribe
You mean those springloaded bits? They reasonably good, but I think a laser would be better.

>>2137441
>sharpie
Not fine enough, and it seems to not be resilient enough anyhow. At least not in heated ammonium persulfate etchant. Once I've got a consistent laser ablation method sorted I'll branch out into different etchants.
>>
>>2137446
I meant a permanent marker, a lot more finer than what you got on that board.
>>
>>2137448
Yeah but 80% of the permanent inks don't hold up, it's hard to be picky about both size and chemical quality at the same time. Maybe some sort of ink pen would work though.
>>
>>2137450
How odd, hydrogen chloride is polar, so we can only blame additives inside the both products.
The store bought stuff might have anti-corrosion agents since it's used in cleaning and it drains from metal pipes, which could easily erase cheap permanent maker.
>>
File: CrATtiny.jpg (1.61 MB, 4032x3024)
1.61 MB
1.61 MB JPG
And redone again. What a waste of a day.

>>2137452
I don't use HCl. I suspect the ammonium persulfate I use is worse than more standard etchants like ferric chloride, though I'm not too sure where on the polarity scale a weakly soluble ionic compound sits, or a copper-ammonia complex for that matter. I've been doing heated etchings for faster speed, so that may be part of the reason, but I know Clive and others heat up their etchants.

I should hit up Bunnings for some "pool chemicals" either way, using vinegar to lower pH is a bit shit.
>>
File: whomp.jpg (13 KB, 228x162)
13 KB
13 KB JPG
>>2137465
>>
>>2137350
looks like SHIT
>>
>>2137350
>>2137465
why dont you just get a pcb made by the chinks for 5 bucks. i never understand why everyone wastes so much time, effort, and money to do shit like this...plus it also looks like utter garbage
>>
File: eat pant niggy.jpg (1.5 MB, 4032x3024)
1.5 MB
1.5 MB JPG
>>2137519
>time, effort
See: >>2137446
>>just doing it for the experience
>If I weren't I'd probably just use JLC or whoever
In the same way plenty of people here enjoy the act of designing their own amplifier circuits and such instead of buying them, I enjoy the act of designing and etching my own PCBs instead of buying them.
And once I've got my CNC laser setup going, it will be far faster than waiting for a board from offshore, and will facilitate very rapid prototyping. This hand-painting is just practice and testing for that.
>money
It's cheaper to do it this way, I can guarantee that. If I got them from the chinks it would be like $10 minimum including shipping. Plus there's a month delay for that shipping. With my own chemicals it's probably less than a few dollars per board, and most of that is in the copperclad.
>looks like utter garbage
That's why I use soldermask. What I posted is pre-etch, for your information. After etch it will look much more uniform, similar to pic related.

Also why do you nerds come out of the woodwork every single time I post PCB progress here? I do it like once or twice a month, you'd think someone would have twigged that trying to convince my mind for the 5th time on an etruscan rodent skinning forum is an endeavour of futility.
>>
File: 20210611_154000.png (125 KB, 1635x991)
125 KB
125 KB PNG
Hated working on this PCB but really like the way it's turning out. Thanks anon that helped with the gold contacts.
>>
File: markup.png (142 KB, 1553x548)
142 KB
142 KB PNG
>>2137620
/ohm/ will approve your board once you fix these items
>>
>>2137629
Thanks for the pointers, I'm pretty new.
A few I had already corrected and the rest I'm waiting to see how the copper pour irons them out.
The original board is full of 90° bends and weird layout decisions so I'm not very concerned with optimizing traces.
>>
>>2137639
not a single one of those items actually matters to be clear, it's just aesthetics
>>
File: 20210611_161124.jpg (856 KB, 2560x1280)
856 KB
856 KB JPG
I'm making an instrument (an inductor checker) that can be powered with a voltage that can go from 6V to 16V, with the most convenient voltage being 10V. I want to add three led that show when the device is on (yellow), when the voltage is too high (red), and when the voltage is optimal (green). I concocted this small circuit to turn on the yellow led at about the right voltage and keep it from become blinding as the voltage increases. For the red led it's just a matter of a single zener and a resistor, but for the green led I'd like for it to be on only exactly at the 10V mark, with say a 1% leeway. I found a few circuits that can turn it on at the precise voltage with a couple of bjt, but nothing that will turn it off if the voltage is exceeded. Any suggestion?
>>
I'm looking for an IC for my waveform generator project. My primary goal is testing audio equipment but also general use would be good.

I am considering AD9833 but not sure
>>
>>2137620
pull back traces from the edges by at least 1-2mm
add ground planes to both sides and stitch it with vias
remove <90 degree enclosed angles from traces
use one trace width unless you need higher power traces
>>
>>2137528
Looks janky as fuck, love it.

>>2137654
Rule of thumb it not for every PCB, you gotta remember that.
>>
>>2137431
you have to go back
>>
>>2137465
>I should hit up Bunnings for some "pool chemicals"
Good luck. My local pool store has literally nothing except liquid bleach, and they're limiting purchases of that. I don't understand why there are still supply problems on this shit.
>>
>>2137519
>i never understand why everyone wastes so much time, effort, and money to do shit like this
shit why even buy a board? just buy a finished product that will probably work fine.
>>
>>2137667
I'm the HCl anon, different countries have different rules on these so I'm gonna assume that the pool chemicals store in down under should be terrible.
But if your country is a little more "dodgy" let's say, you can easily get 10 liters of 30 percent hyrdogen peroxide (medical grade) online, HCl is common place but you gotta read the label for purity (they're for general cleaning not pool related), usually they don't even bother selling small bottles online since it's a liquid. The local pharmacy clerk (he's got a medical degree of course, otherwise can't sell) thought I was a med student and insisted on 3%, had to say "it's for oxidising copper"

>>2137668
It's fun to design your own board, how you produce that is up to you.
>>
>>2137652
try a window comparator
>>
>>2136782
Midwit here. I'm trying to wire an inductive probe to my SKR Mini E3 V2.0 but it's not working. The probe is a DC 5V NPN NO (or maybe NC) LJ8A3-2-Z/Ax-5V. It seems like this probe it needs an external pullup resistor but I'm not sure what that exactly means. It's either that or it needs some type of diode wired to it to prevent the 5V signal from going to the signal pin.
I've tested the probe with a multimeter testing the resistance and it does indeed do something when I put the probe near metal. What exactly, I'm not sure...
Anyways, one of the solutions is to install a certain diode between the 5V and the Signal wires. The only diodes that I have are LEDs. Can I use one of these do you think?
>>
>>2137792
Too much voltage drop across them, also forget about leakage current since it might blow on you.
>>
File: wall warts.jpg (187 KB, 1200x600)
187 KB
187 KB JPG
>>2137792

you can pull diodes out of old wallwarts.
older and heavier ones will usu have four.
the problem is you have no idea what's going on with your probe.
you need to read more and get clearer instructions, before you fuck it all up.
>>
do you have a recommendation for learning electronics from scratch? i'd like to read "the art of electronics" but at a glance i'm worried it won't be absolute beginner friendly (or is it?). i don't even know what most of the electronics terms/acronyms i come across mean.
>>
>>2137919
the book
>Getting Started in Electronics
in the OP appears a good start, didn't catch these my first read, sorry. still interested in any video/series recommendations if anyone has any.
>>
>>2137919
I was in almost your exact situation a year or so ago. I started with Getting Started in Electronics, which is quick and fun, and it's perfect if you have literally no idea what most of the stuff means. I then followed reading Practical Electronics for Inventors and even bought it, as it's relatively cheap and well written. In the meanwhile I started getting the tools for some real life experience: a cheap DMM (the Aneng AN8009 is great value), and an amazon electronics starter kit (basically a breadboard and a bunch of components); after playing a bit with the most basic circuits (like a battery and a resistor) I got a DSO and a XR2206 signal generator and started upping the complexity of the circuits. Then I decided to make a guitar pedal, a Big Muff as it has a really simple circuit, got the parts (always buy more, so you have some left for when you fuck up and for future projects), got a decent soldering station (chink T12 soldering stations are great value) and a hot air one of the same brand (ksger) too since I was at it, soldered everything up, tested, realized I fucked up, fixed, tested again, gifted it to a very happy friend as I don't play guitar. Now I'm pursuing another project, which is more complicated and I'm down a very deep rabbit hole: every time I realize I need to know more I do more reasearch and learn new things so I can make progress. It may not be a optimal university-tier learning experience, but it works for me. I decided to skip The Art of Electronics, as it's a quite difficult and boring read, also the author is a literal jew who split half the explainations on another book for some fucking reason, and uses so many footnotes (half of which to publicize the other book) that the book becomes even harder to read and follow.

tl;dr Getting Started, then Practical Electronics, then pick a project buy tools and components, and go from there.
>>
>>2137939
And he mustn't use the art of electronics as a starting point, it's after he read some amount of books and that online book series.
>>
>>2137640
I don't know how much it affects outside of theory land but I try to keep traces as straight as possible, avoid 90° bends, and avoid forks
>>2137654
Only ground traces are 2mm close to the edge the board is huge so dimensions look weird what's a good trace width calculator?
>>
>>2137865
I'm using it for a filament sensor, if I can't figure it out, I'll probably just print something that uses something simpler like an endstop switch. You are right though, I haven't done enough research on the matter yet, but I'm pretty sure I can make it work.
>>
>>2137792
Try a pull-up resistor first, never heard of using a diode for this before. The resistor (1-100k) should go between the signal output and the 5V rail. Read the datasheet for exact info.
>>
Why the hell are the majority of comparators open-collector instead of push-pull? If I want to use a comparator output to source and sink current into an analog circuit, or to turn a MOSFET on and off, having a damn resistor there just serves to make the whole circuit less ideal. Not to mention the quiescent current. I'm this damn close to somehow using 555 timers as comparators. Though they have poor output voltage range and also have poor quiescent current, among other issues.
Only if I want to AND gate the comparators, or switch a PNP transistor, would I actually find the open-collector output useful. I just want a nice CMOS output.
>>
>>2138178
>Only if I want to AND gate the comparators,

it's actually OR, not AND.
and everyone ORs their comparators except, apparently, degenerate weirdos like you.
>>
>>2138186
>it's actually OR, not AND
Oh yeah it is OR.
>everyone ORs their comparators
Fuck if you want to OR your comparators just chuck a diode after it so it can't sink current. Lets you AND them too just by flipping the diode.
>>
File: 1613070765787.webm (1.02 MB, 406x720)
1.02 MB
1.02 MB WEBM
Hey I'm kind of new to this and am looking to use an arduino for a minor automation project to turn a heat lamp on when a thermistor reads below a certain temperature and turn the heat lamp off when it reads above a certain temperature. The only part I'm having trouble finding (because I don't really know the name of what I'm looking for) is an output that can act as a switch that the arduino can turn on and off. I don't even know if that is the best way to accomplish what I'm thinking or if people generally do something else in this situation.
>>
>>2138208
>is an output that can act as a switch that the arduino can turn on and off.
relay or solid state relay.
If using a plain old relay you need a transistor to drive the coil.
Though there are ready-made solutions to do what you are doing: https://www.amazon.com/Inkbird-All-Purpose-Temperature-Controller-ITC-1000/dp/B00OXPE8U6
>>
>>2138218
> there are ready-made solutions to do what you are doing
This, also consider a PID controller.
>>
File: mechanical thermostat.jpg (120 KB, 800x842)
120 KB
120 KB JPG
>>2138208

rampant 'duino insanity.
what you're looking for is one of these -- for a couple of bucks at the thrift store.
advantages: doesnt need power, doesnt crash coz has no software issues, will work for the next 50 years coz it's so simple, intuitive UI can be used by your granny when you go on vacation.
>>
>>2137446
>laser ablation
that sounds very interesting
laser engravers also have reasonable resolution so maybe I'll keep an eye out for a bargain second hand unit
I think the process would go as
1. lightly spray paint copper clad board with black paint
2. ablate paint with laser
3. etch and then use acetone to remove paint

should be much more independent than sourcing presensitized boards. I have also thought about making a spin coater to apply my own photoresist, as it is the application method with best results
for transparencies I had the local copy shop use their expensive printers to first print a test page with a grid on it (because a 1200 dpi printer may be +-2% in either dimension). then I scale my PCB image on my laptop while I'm at the shop to correct. it is important to mirror the image so that the toner will be on the PCB side of the transparency, for maximum sharpness. the difficulty of photoresist method is calibrating the process for your bulb to get the right exposure. most important is to make it repeatable so that you can get it right once and repeat it
>>
>>2138293
>'duino insanity
I call them "blinkies".
As for the thermostat, they might have too much hysteresis for some purposes, which is why I prefer things like >>2138218, so long as they are mains powered. Having the setting dial apart from the thermal sensor makes them much more versatile too, can use them to make a sous vide if you want. Though a PID might be a requirement for optimal performance.

>>2138296
>I think the process would go as
>1. lightly spray paint copper clad board with black paint
>2. ablate paint with laser
>3. etch and then use acetone to remove paint
Yep that's exactly how I've heard it described. I've seen another anon do it here 6-12 months ago, and I've also seen hackaday articles and even an article in Jaycar Electronics' jew magazine. I didn't buy the magazine, they gave it to me because it was old stock they wanted to be rid of. I think Marco Reps even did it.
Black spray paint might not be ideal though, you may be able to get a more even surface finish with some sort of thermoset squeezed against it with a piece of glass to cure. Not that I've heard any issues with black spray paint.

I'm also planning to bolt a drill motor+chuck to the 3D printer too (it already arrived and is pretty useful) for drilling the THTs and vias, assuming that's easy enough to get working. I don't think the mechanical loads will be significant enough to wear the printer, I'd be more worried about it putting too much weight on it, so I'll try to make the mounts repeatable enough that I can just take the laser and drill off when I'm not using them.

Here's hoping the ~1W violet laser doesn't destroy my bedroom when it hits polished copperclad.
>>
Does anyone have a guide on removing the Roku smart components from a Roku TV? I know you can disable the Roku features, but I want to try removing all the actual hardware that does the Roku stuff.
>>
File: IMG_20210612_101712.jpg (289 KB, 1500x1125)
289 KB
289 KB JPG
I made a sweet probing station for my pcbs to easily attach probes, program chips and shit like that to any part of the pcb.
No more soldering tiny wires, or making single use breakout boards, or deadbugging like a fucking medieval peasant.
We cyberpunk nao
>>
>>2138322
I fucking love it.
Got an STL?
>>
>>2138322
it's somewhere from thingyverse should be easy to find
>>
File: matrix.jpg (76 KB, 480x202)
76 KB
76 KB JPG
>>2138322
He's...the one!
>>
>>2138348
was for >>2138343
>>
>>2138322
this is great, you should order some thumbscrews off ali though. they're way less annoying than wingnuts.
>>
File: alijunk.jpg (189 KB, 800x800)
189 KB
189 KB JPG
>>2138322
so basically one of these?
>>
>>2138322
What kind of probes are they? Really long pogo-pins? They look to thin and flexy to make reliable contact without solder.

>>2138370
That looks neat too.
>>
File: hoverboard motor.png (136 KB, 409x429)
136 KB
136 KB PNG
Is there a good guide on designing mosfet half-bridges? I want to make my own driver for one of those Hoverboard motors. They are 36V, the mosfets on the original board are rated 80V 100A.
I have previously done motor control on smaler Hall comutated BLDCs but for this one I don't think an integrated 3x halfbridge will do.
>>
File: Capture.png (33 KB, 262x156)
33 KB
33 KB PNG
Is there a standard way to carry output from a power supply, or can I just slap 30V between two header pins and call it a day? Seems a little sketchy.
>>
>>2138458
Check the voltage rating of connectors, also their current ratings, and their contact resistance.
You can surely do that, assuming it's like never about 30V 1A.
>>
>>2138396
The one on the picture requires a three phase BLDC driver and the feedback circuit (small square connector)
>>
>>2138478
Yes, I am aware of that.
>>
>>2138458
Those headers should do 30V no problem. If you need to draw a lot of current use multiple pins for your power and ground.
>>
>>2138380
>What kind of probes are they? Really long pogo-pins?
nope fren
they are acupuncture needles
>>
>>2138370
yep looks similar in principal, but mine cost about 10x less money to make and DIYing it myself felt good
>>
>>2138369
yeah i will probably buy some in the future, i bought parts in the local store and they don't sell no fancy thumbscrews here
>>
How fast is thermal camera technology advancing? I would like to get one of those that plugs into a smartphone, but will I be able to get a much better or cheaper one three years from now?
>>
>>2138178
so you can compare output to any logic level without a buffer or level shifter.
>>
>>2138322
nice one, how does it work without being flexible? doesnt it detach the moment you tight the nuts?
>>
>>2138546
the needles are super flexible so i simply tighten them while they are slightly bent and then use tweezers to place them on the contact i want and they are sharp so they just stay there
>>
another fucking project destroyed because i forgot to switch the fucking multimeter probe cable from the current slot to voltage slot, this is so fucking gay i can't believe nobody managed to solve this shit yet, it happens literally all the time
>>
>>2138576
>this is so fucking gay i can't believe nobody managed to solve this shit yet
use two different meters?
>>
>>2138576
>i can't believe nobody managed to solve this shit yet
There are multimeters that mechanicly shut the current sockets if set to voltage mode and you cant select voltage mode while there are probe cables in there.
>>
>>2138576
>another fucking project destroyed
quit designing your projects with power supplies that can't survive a simple short.
>>
>>2138576
How many times have you made that mistake?
>>
File: Diode-Open-768x269.png (39 KB, 768x269)
39 KB
39 KB PNG
Can I use a 1N4007 diode in place of the 1N4148 diode in this circuit? The only difference between this circuit and the one I am working with is the voltage, mine would be 24V but I do have a buck converter that I could use to bring the voltage down to 12V. Also, my sensor is NC but I'm not sure that matters to the diode.
>>
>>2138693
Main difference is that 1n4148 is faster, to put it simply, and withstands lower current than the 1n4007.
It'll probably work.
>>
>>2138302
>so I'll try to make the mounts repeatable enough that I can just take the laser and drill off when I'm not using them.
They have mounts like you are thinking. X-Change is a commercial product or you can print something likepic related
https://www.thingiverse.com/thing:3369444
>>
File: 1623525473269.png (29 KB, 768x269)
29 KB
29 KB PNG
>>2138693

what is that box that is labelled "black"
>>
>>2138706
>REDACTED
It's classified.
>>
>>2138700
I also have a SB340 or 1N4001 if either of those would be better.

>>2138706
Thats just the black wire, I'm not sure why the author did it like that, so I guess you could call it a label.
>>
File: harbor freight DMM.jpg (122 KB, 1402x623)
122 KB
122 KB JPG
>>2138603
>use two different meters?

you must be confusing me with one of your millionaire friends.
>>
File: isabela.iweb.co.uk.jpg (17 KB, 600x405)
17 KB
17 KB JPG
I feel incredibly stupid. I have two 12v 2.8Ah batteries (np2.8-12) I picked up from my brother, but I have no idea if they work, they're completely uncharged. What kind of charger can I use with these? Can I DIY it? Is there any way of testing if they're still good?
>>
>>2138750

read the Ah rating.
divide by 10
run that much current thru it for 10 hours
(change the 10's above to 5 if you're in a hurry)
>>
>>2138744
>>2138700
Actually nevermind, there is an electronics parts store nearby and they have the diodes in stock there. Which is great, only $0.25 each. What a steal!
>>
Can I put two PTC in series to double the voltage rating?
>>
>>2138765
probably not. the voltage rating is really a transient power rating as the thing enters cutoff, sort of like switching loss in a FET. PTCs have huge tolerances so it's likely that just one of your series PTCs would absorb all of the cutoff transient and would fry.
>>
>>2138750
>rg
For regulating voltage and current as >>2138754 says, I tend to use a cheap chink XL4015 module.
>>
>no name inductor
>tinned copper legs
>high temp lacquer
>cheap
>mfw it's from 20 years ago
I need to take extra care with (new) cheap inductors because of their steel fucking leg. Just making sure the copper wire wound around the steel leg and tinning it is enough to bypass them.
>>
>>2138576
>imagine ever using the current measuring mode of your multimeter
just put a resistor in series with your power supply and measure the voltage across it instead

>>2138704
Looks perfect anon
>>
>>2138792
>just put a resistor in series
Ah yes, you want me to chance the tuned copper-alloy low ppm stable current shunt of my multimeter with a fucking low ohm resistor.
Or the very expensive vishay current shunt that the agilent benchtop multimeter has.
Genius.
>>
File: image.jpg (1.47 MB, 4032x3024)
1.47 MB
1.47 MB JPG
Hey nerds, I got a question…

Fucking chink jumper brick, I have a feeling they did a shit job designing the charging board for this thing, so I have an S3 board on the way tomorrow.

So it already has the wires in there, if I wanted to charge it with a power supply and send 12.6V/1A across the #1 cell - and #3 cell +, it will end up unbalanced? Or if the cells are in good condition, it should end up close?

And the other question, so like theoretically if black to yellow is 3.70V for cell #1, yellow to white is 3.70V for #2, and white to red is 3.70V for #3, can I hook the power supply up to black and yellow (#1) and give it 4.20V while the other two are at 3.70, and then go on to #2 and then #3? Or is that going to cause problems charging one at a time while they’re in series?
>>
>>2138804
Sure it won't be accurate, but 99% of the time you just need ±5% or worse. I'd suggest to use a DC clamp meter instead, but those are probably no better than ±5%. Either way it's preferable to nuking your ICs.
>low ohm
I've used 1kΩs before, when current is low.
>>
>>2138805
You need to charge them all together which calls for a balanced charger, sadly. Also the way to charge lipo/ion is constant current.

>>2138812
Voltage drop and the literal resistance itself can cause the more smarter ICs to shutdown and reboot themselves over and over. Just take more vitamin B12 to not forget. (or use two multimeters)
>>
>>2138874
>Voltage drop
That's why you select the resistance so the drop is no more than 10mV or so. 100mV might be ok for a dumb load. Don't know of any IC (other than something like a BMS) that can tell 10mV difference and decide to shut down.
>>
I needed a transistor for replacement and got a box with various transistors off amazon, as it was cheaper than the dedicated e-shops. Is there a resource to easily and quickly compare them with just a name search so I know what's best to use on my next projects?
>>
File: PSX_20210612_143535.jpg (57 KB, 1161x837)
57 KB
57 KB JPG
doing a usb-c conversion on a somewhat cheap motorcycle headset. i'm about finished but i need a suggestion. would you more experienced folks rather use 24awg solid wire or stranded wire for the breakout board solder pads?
>>
>>2138874
I know about the constant current. I guess tomorrow I mess with the BMS when they show up.
>>
>>2138893
I personally never use stranded wire for anything.
>>
>>2138893
Are you running the usb on 2.0 speeds?
Anyhow, I would use stranded wire and you might wanna put some woven sleeve around it in case something cuts the conductors and kills your headset. (Not a problem you really encounter on home usage)
Also what's the current draw of the device? Is it PD supported?

>>2138894
So they're arriving tomorrow? Really not worth the risk of damage/explosion.
>>
>>2138882
bustedgear.com/faq_transistor_substitution.html
alltransistors.com
>>
>>2138898
there won't be any data lines on this. i mean, technically there are on the original connector which i guess they used to flash the device, but the usb will be just for charging. I have no way of measuring the draw but it has a 3.7v 800mAh battery in it and takes maybe 1 hour to charge i guess. sorry i'm such a noob but i hate micro usb so this seemed like a cool project. DIY Perks on youtube gave me the idea
>>
>>2138903
Apply all the sugru.
Anyways, I'm gonna assume 500ma, in that case 24 awg is fine.
>>
>>2138893
>>2138897
Solid wire might work harden and become brittle faster near the connections if there's enough vibration. Nothing else really comes to mind.
>>
>>2138910
Conductor skin effect for AC and even more so for high frequencies.
Also it's more flexible and you can flatten a stranded wire for a slimmer wire, for solid core 4mm2 means 4mm2 forever.
Coax cables, if not meant to handled (like a probe) are also solid but that's just black magic, I don't know.
>>
thanks for the help guys. hopefully i can get it soldered up. i did a socketed EEPROM ecu conversion before and also 0603 LED xbox 360 ROL color swap. hopefully this goes that well.
>>
>>2138922
>0603 package
Always a pain to hand solder, I use ceramic tweezers for small stuff.
>>
>>2138898
I mean I’m watching the voltage. The 3 cells are all pretty close. I got 3 of these jumper packs because they had like 5,000 reviews and were pretty good, but I guess that was before people owned them for a year or two. There is something fucked up with the charging setup, this chip just burnt up by the 15V/1A input, and I used them maybe 2-3 times each to jump cars plus once every few months plugged them in to charge. Upside is the actual power doesn’t go through that shitty board, and the power switch on the board doesn’t really do anything besides light up the battery indicator LEDs.

Second jumper pack doesn’t want to charge above 4 bars now, just over 12V. I think it’s the same damn issue, it’s getting warm in the exact same spot.
>>
>>2138927
Do tell me that the jumper cables exists, and they connect directly to the battery and the cable has a chunky array of died protecting it from reverse charge.
>>
>>2138926
tin one pad. apply flux to tinned pad. with good tweezers, hold 0603 to pad and reflow with iron. solder second pad as usual. you can go down to 0201 like this.
>>
>>2138931
Yeah that’s why it should still work. The board that burnt up seems to be nothing more than BMS, plus that stupid flashlight, LEDs for charge level, and the USB outputs. If I ignore the USB outputs, I only need the new BMS to charge the cells and figure out a good way to feed it like 13V.

There’s 2 beefy cables to the 12V output, and then the clamps have some circuitry in there, the reverse polarity light plus a little “boost” button, so the fried board inside the pack doesn’t really have anything to do with the jump starting.

Also I’m curious about the voltage at the clamps. If I plug in the clamps, there’s 0.0V on them until they’re hooked up to the vehicle and it sees the correct polarity. I’m wondering if it’s only sending out ~11V or like 12.5V max, or maybe that boost button steps it up? Can’t be too many electronics inside those clamps so it probably is just low 12’s making up for the big voltage drop of a dying car battery.
>>
>>2138944
>>2138931
Haven’t popped this open yet, all I know is it won’t let any voltage into the clamps until they’re on a battery. I stuck em on a super dead little lead acid that was only reading ~5V and it was enough to get the green light on.
>>
File: gon look neat.jpg (525 KB, 1874x884)
525 KB
525 KB JPG
final solder mask being cured
>>
File: look neat.jpg (2.82 MB, 3264x2448)
2.82 MB
2.82 MB JPG
>>2139074
And the vias are stitched now. Not looking too bad, though I'm unsure if I should add soldermask over the vias. Maybe I should just buy some of those shitty rivets, though they all seem to be rather wide (not for 0.8mm holes) and designed for making plated through-holes. Also they're difficult to rivet flawlessly without a proper riveting tool, which is rare and costs like $20. I'll try electroless silver plating once I've got my laser setup.

I'll check the board's net continuity tonight, and solder the parts after work tomorrow.
>>
shieeeeeet turns out when i try to measure current with a multimeter it's harder than one would assume.
the micro uses ma when it boots and then ua when i goes to sleep, the issue is that to get precise ua readings i have to put the multimeter into ua mode, but the multimeter has to use a fat shunt in order to be able to read such small voltage drops and because the shunt is such high value the micro crashes before it gets to deep sleep because the multimeter shunt is dropping to much voltage
all this shit you learn in school
>oh meters have infinite impedance in voltage mode and zero in current mode and so on
IT'S A FUCKING USELESS unless all your projects never leave the paper drawing. It will never work in the real world because of all the gay shit like emi, voltage drop, impedance,
i am so fucking tired of debugging trash like
>oh your shit isn't working properly because there is a motor running in the next room
>>
>>2139192
large dynamic ranges for power consumption are measured with an integrator
>>
>>2139200
if it was this easy they wouldn't sell $700 devices for this purpose
>>
>>2139209
everything that costs $700 or $70000 is because of NRE and characterization. you can nigger rig a circuit to get close enough for your needs.
>>
I want to power an led circuit off of a 120V AC outlet from my house. I would need 6000 ohms of series resistors but I'm concerned about the power going through them. Would they still blow up?
>>
>>2139229
Well that 6kOhm resistor is gonna dissipate about 2.5W so take one that can handle that.
>>
>>2139241
Would putting multiple resistors with lower rating be able to handle it? I don't have a 6kOhm resistor rated for 2.5W
>>
File: retard resistor.png (5 KB, 360x328)
5 KB
5 KB PNG
>>2139243
Assuming you only have 1/4W 6kOhm resistors you could do pic related. This way each resisor will dissipate .15W.
>>
>>2139192
I hate non ideal shit too. Everything is thought and taught like everything's ideal, and the implementation is always a good enough deal, but the moment you need to make a single step towards accuracy everything falls apart.
>a resistor followed by a zener should clamp the voltage steady
>it literally doesn't
I spent a day measuring and researching. In the end after everything failed I flipped through Practical Electronics and found pic related, which works a treat and I can now finally have a decent reference voltage, independent of line voltage, that I can hook up to a comparator.
>>
>>2139229
Post circuit.
Use a capacitive dropper instead.

>>2139393
>it literally doesn't
What was the issue? Did it have a curve that wasn't as steep as you'd want? A TL431 based solution is what I'd use either way.
>>
>>2139393
>. Everything is thought and taught like everything's ideal

And that is made perfectly clear, and you should understand why that's how material is taught. And when you are in the lab, or tinkering at home, or on the job, you show your brilliance by accommodating reality which is pretty much never ideal.

Sorta related, when I was in uni I would write down all the digits my calculator displayed, and never really understood why the instructor bitched about it. And now I design actual electromagnetic devices and realize that you are lucky if three digits are actually significant for just about every parameter other than the mechanical ones.
>>
File: 1623105663018.gif (1.12 MB, 260x454)
1.12 MB
1.12 MB GIF
>>2139508
I learned the scientific number notation or some bullshit early just to troll people.
>this shunt resistor here is 1.0E-10 giga ohms
>gif related
>>
>>2139534
mixing SI prefixes and scientific notation is hell
>>
File: whatsapp_car.jpg (48 KB, 1080x991)
48 KB
48 KB JPG
>>2136782
Rolling for challenge from https://github.com/Rocheez/4chan-electronics-challenges/blob/master/list-of-challenges.png
>>
>>2139574
Rolling again for another project. Parabolic mic annnnd....
>>
>>2139574
Oy, lemme roll as well.
Since the list has no devil-ish projects for trips or quads, I must add them.
333 Class D audio amp.
666 Class B+A Audio amp.
777 Make a pwm brightness and I2C color temp floodlight with no aluminium substrate (whatever power leds you have, dude)
6969 Portable headphone amp, must be hifi, analog or not
888 8 layer board that combines everything sean hodhings (forgot the name, volumetric display guy) did.
999 Re-design ottercast amp for car use
111 12v to USB PD, 100w+, no idle current limit
222 DANTE VoEthernet bootleg card.
444 10W continuous power laser driver, must be heatsinked and open source.
555, 556 powered double PWM half-bridge (I designed one way back with a single 555 of course, 10A cont.)
>>
>>2139584
>fm bugger circut
Fuck off, already did a very illegal GPS jammer.
Rollen'
>>
>>2139590
I've done programming shit but I'm NEW to electronics. Besides an LED circuit and taser, haven't done shit. Parabolic mic and rain alarm should be a good start, not too complex.
>>
>>2139591
>led circuit
Exactly what? I had fun with a multi vibrator circuit, it taches you to pick the correct value for different single led colours. Normally you would slap 470ohms for the leds and 10K to charge the caps, I fiddled with changing cap values, different voltage caps, had no variable cap but later I learned that I could salvage one from radios, and no pot.
All of this was done on a copper board that I etched with HCl and permanent marker as etch resist, then I trilled every hole with like 1mm or so drill bit on a miniature drill press meant just for pcbs.

Make sure to tune the rain alarm for your local rainwater contents, rich in sulphur, rich in minerals, natural etc.
>>
>>2139468
>What was the issue?
>Did it have a curve that wasn't as steep as you'd want?
Basically, yes. I spent the day backtracking the informations I actually needed; I assume that when they talk about line regulation the assumption is that you want to smooth out ripple for a load, and not step down voltage from a variable source with basically no load.
>A TL431 based solution is what I'd use either way.
Yeah, well, at first I didn't want to buy yet another box of components, but then I got fixated in using what I had.
>>2139508
I'm an engineer branching out towards electronics because it's cool and I can thinker with it at my desk: I know. My issue is that the moment you want to step out of the ideal world you're pretty much on your own. The knowledge required is often at a master degree level, which creates accessibility issues: not that the informations are not publicly available, but they're hard to find if you don't know where to look in the first place. If you try to look for something all search engines will shove you basic stuff in your face, while looking on Google Scholar will give you state of the art research (and trash papers from the usual Pajeet/Chang), no in between. Also, all the basic stuff give all working hypotheses for granted, and that's a VERY bad habit.
>in uni I would write down all the digits my calculator displayed, and never really understood why the instructor bitched about it
Usually using all the digits in the middle of all calculations is fine, but in my uni experience writing more digits than made physical sense was counted as a grade-lowering error.
>>2139554
Mixing measurement system is hell's hell. I never really cared for the whole American system as it's stupid, but mostly harmless, until I started studying aircraft performance: the industry uses the American system, and there I saw the most insane shit ever devised. I swear every time I see anything with slug in it I get my PTSD triggered.
>>
>>2138322
Lmao. Why not just use a breakout board? They sell them on Amazon you just need to solder the chip and some pins to the board. Should be really easy to do with that package.
>>
I’m lost

I’m 20 and currently in community college. I’m attending the college through a promise grant so I thought I might as well get some free credits. Because of Covid and the state I live in, the pandemic has shut down our community colleges and forced us to go online.

A lot of the engineering classes that the college originally held were either removed or drastically downgraded to fit within an online course. My next semester is the first semester to include in-person classes, but most of them are classes where they meet 3-5 times during the semester

How do I, a 20 year old anon, make progress in becoming an engineer? I don’t know how to code and I live on a military base, so there’s no way for me to have a workshop to build big things. I bought an Arduino kit and so far, I’ve learned how to change RGB lights, read potentiometers and joystick, change a 7segment display, and turn on a buzzer. I also know how to use Autocad and Fusion 360. Is this a good start? Should I be doing something else with my time?
>>
>>2139612
Pick up "Practical Electronics for Inventors". You don't need to build big things. With a bread board, the book, and that Arduino kit (assuming it came with jumpers, resistors, capacitors, and other common components), you can learn a lot and play around with DC circuits. You really only need the Arduino for its 5V or 3.3V supply, and maybe to generate some PWM. Unless you're going to become an EE with specialization in Power, you don't need much and can create all sorts of things with limited space and budget.
>>
>>2139612
Anon, that's a good start, BUT
They teach me autocad and electronics when I was in high school, either the education system which you're under or the schools you have been so far wasn't really equipped to spit out engineers. To become and electronics engineer you also need to pick a sub-category to electronics, they might be called sub, but they're fully actualised and proper fields.
Not everyone needs to be an engineer, and playing with Arduinos should be cut short.
>>
>>2139619
>>2139620
Thanks. I am currently undecided as to what I want to do. I like to 3D model, both on CAD and Blender, but polygon based modeling is highly competitive and I doubt I have what it takes to make it far

I’ve always loved math and science as a whole, and I love learning about how things work and operate, as well as the physics behind it. I just hope I can find my path soon
>>
>>2139626
>I just hope I can find my path soon
Get a machete and make your own path, nigger.
>>
>>2139602
>you're pretty much on your own
Just read the datasheets, and any relevant appnotes you can find. They should cover all the nuances, and you can google any term you don’t understand. For example, the I/V curve should be on any decent zener diode datasheet. For low voltages zeners tend to be kinda shit, only at higher voltages does that curve get steep.
>accessibility issues
The only accessibility issue is that test equipment is expensive. If your circuit isn’t working the way it should, wire up your scope to it and see why. Sometimes just a DMM is enough.
>no in between
The most basic page for the correct search term is usually sufficient for anything a hobbyist needs. Appnotes specific to the topic are the gold standard, and should go sufficiently in-depth for proper engineering work. Knowing that correct search term is the hard part, but again the datasheet and any referenced appnotes should have what you want.
>>
Kinda retarded question i needed to extend my usb cable for my raspberry pi cam i used speaker cables. Now the voltage jumps up and down continuously but if i measure below the speaker cable its a solid 5v what gives?
>>
>>2139658
>below the cable
What you mean actually? The starting point of the cable where the power supply resides? If so then you have voltage drop on the cables.
How thick are the conductor inside cables?
>>
I have humming/ground problem on my left Turntable that I do not have on my right. They are Technics 1200s MK2 w/ Pioneer DJM mixer. I am unsure next step. I have done the following without improvement:

-swapped cartridge/headshell
-swapped pre-amp inputs
-swapped AC outlets

So I'm fairly confident the issue is specific to the TT. Symptoms:
-low level hum
-touching TT causes pop and change of hum
-touching mixer / Right TT affects hum
-Hum appears on right speaker only
-Removing headshell creates hellish noisescape if I have volume turned up to better hear hum.

Right TT does not have hum.

Does this suggest anything specific?

Because I have trouble with the basic concepts it is hard for me to problem solve electrical issues. My next step would be to disassemble the TT and check the ground connections internally. I'd like to get opinions before I start performing exploratory surgery :)

I struggle with electrical concepts, such as a shared ground ( I don't know why, but please keep that in mind ). Like this makes my head hurt: https://electronics.stackexchange.com/questions/253546/how-does-stereo-audio-signal-works-with-shared-ground-e-g-headphone-cables but I guess that should be a separate question.

I really can't visualize why me touching the turntable causes a change in the energy field. I mean is it stealing my electrons? Am I suddenly sharing elections back and forth with the turntable? Why does this black magic of electricity not make sense to me. Yes, I get that I=V/R but the way it all comes together just will not click. I have a DMM and oscope if there is something I can measure that will provide insight.

Any practical or theoretical information appreciated.
>>
>>2139658
>voltage jumps up and down continuously

you need fatter cables, and stand-offs so they dont touch the ground.
>>
>>2139692
>Wood
Do audiotards know that grounds?
>>
>>2139687
>Does this suggest anything specific?
Change the power supply/adapter.
Check the grounding of your home. Make sure your home is not "zeroed" meaning some retard connected floating ground to the neutral.

Try to take it apart and directly power it with batteries. Try to slap some DC blocking caps.

>>2139692
AC signals and AC power lines should be well above ground, and that means suspending them with ceramic or PTFE, not on the ground with fucking grounded wood.

>>2139695
An audiotard knows nothing. There's zero information in their heads.
>>
>>2139698
>Check the grounding of your home.
Wouldn't this issue impact the right turntable?

>Change the power supply/adapter.
You mean the internal PSU? I'd like to isolate the problem more before I take that step

>slap some DC blocking caps
Where would these go? what values?

>>directly power it with batteries
Do you mean running a separate 120V using batteries, or using 24V batteries and bypassing the PSU?
>>
>>2139687
Try different RCA cables and check for cold solder joints on the phono jacks.
>>
>>2139687
>>2139712
Also, swap the left turntable and right turntable inputs on the mixer to narrow it down.
>>
File: image.jpg (2.28 MB, 4032x3024)
2.28 MB
2.28 MB JPG
>>2138944
Oi I need more practice soldering. Trying to think about how to leave this and recharge it off ~13V running car power. Let’s see if this test run starts it on fire!
>>
File: louis.png (519 KB, 716x672)
519 KB
519 KB PNG
soft question: does /ohm/ have any strong opinions on this lad?

(if this is irrelevant just tell me and i'll delete the post, apologies in advance if i'm shitting up the thread)
>>
File: image.jpg (1.38 MB, 4032x3024)
1.38 MB
1.38 MB JPG
>>2139717
Sorry bad pic
>>
>get transmissometer
>bust balls installing it
>pour concrete pads
>mount and align it
>run cables underground for AC power
>get it all up and running and systems check out in the maintenance line
>has tx/rx hookups in the terminal box at the bottom
>make datalogger for it with shitty little waterproof box and a rpi
>hook up tx/rx
>only garbage coming in, but it's consistent garbage
>rpi suddenly dies and SD card corrupts

I think that I messed up /ohm/bros, it might actually be RS232 coming out of the terminal box. I guess it never clicked in my head that the voltage swings were different between serial formats.
>>
>>2139723
Wait why did I post this in /ohm/ I shoudla posted it in /mcg/, sorry brain is totally fried
>>
>>2139719
I like Louis. He knows his shit.
>>
>>2139723
Anon painfully learns that data level translation is a thing and isolation is good.
>>
>>2139723
>transmissometer
wtf are you doing with that thing?
>>
>>2139746
measuring transmissos
>>
>>2139746
/diy/ airport
>>
File: 1617879714692.jpg (30 KB, 472x461)
30 KB
30 KB JPG
>>2139754
>>
>>2139712
>Try different RCA cables and check for cold solder joints on the phono jacks.

The RCA cables are soldered to the TT. How can I check this using a DMM or scope? I assume I'll need to open up the TT.

>>2139715
>Also, swap the left turntable and right turntable inputs on the mixer to narrow it down.
Yes, did this

Considering I have access to the cartridge and 4 pins there, are there any diagnostics I can do by checking resistance or waveform between the wire at the TT cartridge and the RCA out cables?

I'm assuming I would see the hum wave if I looked on the right ground or right tip?

does that tell me anything, or am I still left hunting inside for something else?
>>
>>2139687
Take apart, check ground connections, check the big capacitors. Scope the noise too.

>>2139719
Most of what he does is really specific to Apple repair so doesn’t apply that well to this thread, but he’s definitely a good solderer and his right-to-repair videos are kinda relevant (and hella based).

>>2139723
Hahaha
Dedicated RS232 level shifting isn’t completely necessary, all you need is a resistor and a couple of diodes to clamp it to your logic rails. One diode and one transistor with the resistor is a bit safer, and two resistors one optocoupler is even better. That’s for logging at least, if you need to spit something onto the lines you’ll need something a bit more sophisticated.
>>
>>2139771
>check ground connections, check the big capacitors. Scope the noise too.
Thanks I appreciate these next steps, Can you be a bit more specific about "scope the noise"?
>>
>>2139771
>sophisticated
Like a bidirectional logic level translator, automotive grade or rad-hardened for some cool points (just a joke, about the diy airport part)
>>
Is there a way I can make it so that the output only receives current from the capacitor and not from the input going into the capacitor?
>>
File: 20210613_205906.jpg (304 KB, 931x698)
304 KB
304 KB JPG
>>2139796
Fuck, forgot my shitty simple little schematic in case you couldn't understand my question
>>
>>2139799
Put a switch, relay, solid state relay, or a mosfet before the capacitor and disconnect it.
What's that tho? A supercap?
>>
>>2139796
are you trying to draw peak currents from the capacitor in excess of the upstream input's current rating? if so then add a current limit or just a resistor to the input.
>>
>>2139785
>Can you be a bit more specific about "scope the noise"?
First do a continuity check between the ground lug of the power cable and the 0V rail of the power supply. Might be split-rail, might be single-rail, but either way the 0V should be easy to find at one of the RCA outputs. If there is continuity, it will determine that it’s safe to clip a scope up to something in the circuit. It should always be safe, but there are some fringe cases where your ground clip will short something. Clip the scope ground to the outer part of the RCA connector (probably the side within the case) and the scope probe itself to the centre part of the RCA connector. Ensure the volts/div are at a sensible level for the signal strength you’re measuring (idk 1V/div should be fine).

If there is no ground continuity, then you have a floating ground, which is probably not intentional.

Whenever you’re scoping inside an appliance, you’ve got to be careful about mains electricity frying you or your equipment. Usually the power supply of the board is on a seperate board or even in its own safe enclosure, either way ensure you know where it is and which wires could be high voltage. Power switch included, gave my finger a bit of a belt when messing about inside test equipment once when my finger brushed the switch contacts and my finger shorted the open switch.

>>2139799
Use a low-pass LC filter. RC might be fine if you’re not wasting a bunch of watts on it.
>>
>>2139799
Yes, just cut the wire going from in to the capacitor. Then current will not flow from in to out.
>>
>>2139611
probing stations like that one are intended for things like debug or testing with finished circuits with chips already soldered onto them and to easily rapid dev with chips where they only need to be used once so making a breakout board would be a waste of time
once you advance your ee skillset beyond the basics you will understand why tools like this are extremely useful
>>
File: Capture.png (200 KB, 1457x495)
200 KB
200 KB PNG
I'm always paralyzed right before submitting a PCB order I spend days and days postponing thinking about any mistakes I may have made or any improvements I can do how do I just say "fuck it" and pull the trigger?
>>
>>2140103
what the fuck am i looking at? load this into oshpark, it has a better preview
>>
>>2140103
You can always botch some wires/cut some traces if you fucked up.
It's an iterative process, there will always be fixes and improvement you can make.
>>
File: Capture.png (282 KB, 969x739)
282 KB
282 KB PNG
>>2140105
Holy shit, the difrerence is day and night, I still got to fix that SMD cap silkscreen on the top layer

>>2140109
I mean, yeah. You're right, worst case scenario I have to reorder if I want it neat. But hell, I'm still hovering over the order button with hesitation.
>>
>>2140111
Gotta think practical.
What wastes your time more? Fretting for days or fixing a mistake and ordering a new one.
>>
>>2140111
they won't be able to cut some of those nicks in the board. they use a round end-mill, probably 1/16th (1.5mm) or larger. you'd have to cut them yourself with a file.
>>
My JLCPCB order came with a promotional jigsaw wtf.
>>
>>2140120
I'm not expecting 100% results with the corners but thanks for the heads up, I'll have a set of hand files ready for when the PCB arrives.
>>
>>2140116
Yeap, time is money and I'm sure waisiting it like I got time to spare
>>
>>2140134
Tell them you'll give them an extra dollar to make the child slaves file it to spec for you.
>>
>>2136782
I read somewhere that amplifiers have a really big capacitor for when they need to draw power for bass. But I'm a noob and idk where that capacitor is supposed to go for my diy amp. To ground? Parallel with the power lines? Pls help.
>>
>>2140141
connect the negative cap terminal to the negative input and the positive terminal to the positive input.
>>
>>2140141
>I read somewhere that amplifiers have a really big capacitor for when they need to draw power for bass.
I've been wondering about this.
I get the sound perception thing but I also heard people claim that for lower frequencies the speaker has to push more air, requiring more energy. But that seems wrong to me.
Wouldn't how much air it pushes be the amplitude, not the frequency?
>>
File: amp.png (7 KB, 886x468)
7 KB
7 KB PNG
>>2140145
Like pic related?
>>2140153
Afaik speakers are less efficient at lower frequencies combined with the fact that our ears are less sensitive to low frequency which requires lower frequency sound to be louder, requiring even more power.
>>
>>2140237
correct, and make the wires from the cap to the physical input of the amp as short as practical.
>>
>>2140153
Could be that an amp that responds well to lower frequencies is more suceptable to power line hum
>>
>>2140111
Did you design that board yourself? What software? Looks nice.
>>
>>2140324
Thanks man, the software was the absolutely shit EasyEDA because I want JLCPCB to do my SMD work.
I don't know if I can use KiCAD or Altium to get SMD with JLCPCB.
>>
>>2140111
do you do con badges or something? what are you working on?
>>
>>2140411
I’m pretty sure the file formats are highly standardised, unless JLC are deliberately making things proprietary, which I doubt. EasyEDA is a tool for them to get more noobs making PCBs through them, it would be a bad business move to try to force people into using it. They’d lose the more professional customer-base.

Looks like a pretty complicated board layout with those decals and wonky outline. Though now that I think of it, I wouldn’t be too sure how to have that board outline in KiCAD either. I guess I’d make the outline in some vector program (Q-CAD) and export it into the right layer?
>>
File: file.jpg (3.5 MB, 3471x2825)
3.5 MB
3.5 MB JPG
there are no commercial battery clips/contacts suitable for very high currents (eg 16 amps from an 18650) so a /diy/ contact is mandatory. nickel-nickel contacts under high pressures have excellent contact resistance; neodymium discs are nickel plated, dirt cheap, and in the perfect form factor.

i used a shitty test setup and got a 0.035 ohm total resistance, so ~3 mohms per contact which is 50x better than the leaf and coil springs on digikey. it didn't seem to care about the clamping force beyond a nominal amount. appears to be a good solution for anybody needing low ESRs.
>>
>>2140533
Usually you would spot weld them together and the last two strips would be soldered to cables and then get spot-welded again.
Having solid strips is recommended of course.
Are those pure nickel or mostly iron ones?
>>
>>2140535
china says they're "pure nickel" but stainless core strips are alright for my application. i don't want to spot weld them because i'm not going to integrate a charger into my project. the plan is to print a removable housing for the cells and use a wave spring to provide a consistent clamping force.
>>
File: ground.png (45 KB, 286x558)
45 KB
45 KB PNG
Attaching something to my car and it has this kind of end. Is this for screw terminals. Im probably going to cut it off and put a quick disconnect on it but I wanted to know if there was a way I could use this with some sort of tap.
>>
File: Capture.png (94 KB, 331x281)
94 KB
94 KB PNG
>>2136782
is there a name for this style of spacing for surface mount board to cable connections? it's 0.1" / 2.54 mm pitch but it's offset in this weird fucking way. the datasheet gives wurth 690367280876 or te connectivity 7-2178711-8 as compatible parts but i don't feel like getting jewed with proprietary ribbon cables.
>>
>>2140536
>stainless core strips
Either terrible stainless or mostly iron. Yes, good old Fe, when carbon is too expensive.
Why? Magnetic.

>>2140538
That's for a camera system right? If not then they're common for vehicle cameras, perhaps you could find a better one and yea you can put whatever connector you want there, make sure it's polarised.
>>
>>2140541
It's staggered, those pads aren't meant to be high current anyways or you can definitely make the outer ones like that.
Just replace them if you don't wanna buy from TE or Würth, best ones in the market.
Würth for connectors tho... unnecessary.
>>
>>2140543
Yeah its for a vehicle camera. But theres nothing that this connects to and it doesnt come with any attachment for it. What do you mean polarized?
>>
>>2140536
Decent idea, but from my experience if you want high current you use a Lipo instead. That way it’s just a matter of unplugging the XT60 and taking the pack out of an elastic or velcro strap.

>>2140541
Post what the connectors themselves look like. All I can imagine is that kind of footprint from a single-row pin header with the pins bent alternatingly. Also just design your own PCB for any connector you want.

>>2140556
Polarised means it can’t be plugged in backwards. A figure-8 IEC jack isn’t polarised, while an XT60 is. An XT30 or mini-T might be a decent choice if it’s just a power connector, otherwise there are probably more standard automotive power connectors you could use, assuming you don’t have to pay out your ass to get the right crimping tool.
>>
>email post service after 2 weeks of no laser
>no reply
>tracking immediately updates
oi
>>
Alright, I'm a newfag. Rolling for the challenge.
>>
>>2140574
Unlucky
>>
>>2140570
There are industrial, be it aerospace, defence, automotive connectors that either come with a solder option or are designed to be soldered.
But of course, crimping is the way to go.
>>
>>2140570
>use a Lipo instead
why did i not know lipos had such low ESRs? still i think i'll stick with the 18650s because they pack in a cylinder so well.
>>
>>2140597
>why did i not know lipos had such low ESRs?
I didn’t know either until I ventured on a rather long project (still not done because IRL is an ass and so am I) that needed high battery currents. I asked here, and was directed to the remote control general, where they fly FPV drones. Turns out, 100C lipos are commonplace. That means you can drain the battery completely in 36 seconds. Picked up a pair of 6S 1.3Ah CNHL lipos from AliExpress, which can sustain 130A of current, more than I’ll ever use. More than twice what the XT60 that comes with them can handle.

If you’re interested in some cutting-edge shit, fpv drones have a fair bit.
>inb4 lmao analog video
>>
>>2138693
>>2138762
Not that anyone gives a shit, but all I had to do was add the diode and the output voltage was within range for logic high (or maybe low). Either way, its triggering and registering now so I'm happy.
>>
>>2140126
i keep getting puzzles of their factory
>>
File: halfdone.jpg (1.2 MB, 3538x1415)
1.2 MB
1.2 MB JPG
The PCB works!
The code doesn't. Also the cantilever of the OLED is pretty precarious, once it's debugged I might desolder the socket it's plugged into and just hard solder it with a bit of gap to the VOC module.
>>
>>2140499
Thing is with easyEDA you can pick SMD components from their online catalog and import them directly into the CAD and when you export the Gerber it gives you an additional pick and place file.
What workaround can I use on other software.

For the board shape I drew it on Photoshop and imported it, then I zoom in like a MF and drew the board outline following the imported image and deleted the image.
>>
>>2140734
>pick and place
Oh are you doing their assembly thing too? My suggestion would be to see if you can import KiCAD's gerber into EasyEDA. Maybe there's a plugin or something for KiCAD that generates the pick and place files also.

>zoom in
>drew
that sounds very tedious, my heart goes out to you
>>
>>2140734
I have some ideas for PCBs like this, any recommendations for where to begin learning how to design them?
>>
>>2140755
Just trow together simple projects till you get the hang of it
>>
File: ups.png (5 KB, 400x400)
5 KB
5 KB PNG
so basically if a diode has say 0.7V voltage drop that means that if the voltage in front of the diode is not 0.7V lower than the voltage behind the diode it won't conduct anything at all right? kind of like zener

why am asking is, i have usb wall wart at 5V and i want to add a battery to it to serve as backup power in case power goes out like in the picture, will that work?
>>
>>2140760
The powered device will be just a small chip like arduino uno for example
>>
>>2140760
>>2140761
A diode works like a check valve, and the diode drop indicates the voltage needed for current to flow in the allowed direction. In other words if voltage difference from anode to cathode is lower than 0.7 V no current will flow; conversely if it is higher than 0.7 V current will flow, but the voltage drop will remain. In practice your drawing will work, but if the battery takes over the device will have only 2.3 V available (3-0.7). You could use a Schottky diode as it has a lower forward drop, but there are better solutions like using a P-channel MOSFET to switch the battery off the circuit when not needed.
>>
Is this a good 12V power supply?
>>
>>2140796
it'll do. for lab work you want variable voltage and current.
>>
>>2140788
>but if the battery takes over the device will have only 2.3 V available (3-0.7).
thats fine the chip can operate between 5v and 2V no problem i suppose there will be a bit of energy wasted ude to the diode drop but this is just microamps so it's shitall, mosfet would probably be better but that is adding a bunch of parts while with diode it's literally just one single little part
>>
>>2140570
>Also just design your own PCB for any connector you want.
thought about it but the ic is $7 plus whatever the cost of making the pcb plus assembly costs (i don't have the skills/equipment to solder qfn-16 by hand.) or i can buy the development board for $10. no idea why they chose to use such a retarded pad layout. https://www.monolithicpower.com/en/ma330.html
>>
>>2140803
What's the ideia behind the values of the capacitors? The datasheet of the 7812 IC says 0.33uF and 0.1uF
>>
File: 1597388487103.png (114 KB, 635x521)
114 KB
114 KB PNG
Building out a home weather station to setup next to a small solar array, and trying to figure out what sensors I am best with using for a few different measurements, curious if I am going in the right direction for an initial build if anybody has experience with these.

> Temperature
RTD Temperature Sensor w/ Probe

> Wind Speed
Was going to 3D print a cup-anemometer, and measure rotations with a hall sensor and magnet planted in the anemometer body.

> Wind Direction
Similar 3D print of a wind-vane, on-top of a rotary encoder, w/ necessary bearings, etc.

> UV Index / Intensity
Adafruit UV Index Sensor or something similar.

> Humidity
Standalone sensor, maybe an SHT4x/3x or so.

> Remote Output
Bluetooth module to export data.
>>
>>2140823
the 1000uF cap stores energy between mains cycles; it "smooths" the input voltage to the 7812. the output capacitor is largely arbitrary. i'd honestly prefer it to be smaller but it doesn't matter for your application.
>>
>>2140817
>mosfet would probably be better but that is adding a bunch of parts while with diode it's literally just one single little part
It would be a single part with the mosfet too, but on a second thought I agree, it would be overkill.
>>
>https://www.thonk.co.uk/shop/plinky/
Anyone know what this is using for touch?
>>
>>2140876
BME280 can do pressure, humidity, and temperature all in one. accuracy might not be as good as a standalone unit but for home use it's probably good enough?
>>
>>2140901
>It would be a single part with the mosfet too,
can you show an example schematic?
>>
File: wow.png (1.45 MB, 1322x780)
1.45 MB
1.45 MB PNG
>>2140908
>>
>>2140949
I can't read very well, but what is it? Looking for that text specifically just brings me back to the store page.

I can't find an explanation of what it actually is anywhere.
>>
>>2140985
Open-air capacitors that change value when you touch them.
>>
>>2140985
>>2140990
This. You make an oscillator with the touch pad as the capacitor, and as a finger is pushed harder or softer into it the frequency will change. There might be dedicated analog capacitive touch ICs to convert this frequency into an analog value or send it to an MCU via I2C, or maybe it’s using some sort of lightweight frequency counting IC, or just sending the modulating inputs directly into the MCU for counting. In the latter two cases each button still needs an oscillator circuit of some sort.

Now I really want to make a MIDI PCB keyboard with proper key velocity. I’ve gotta look into analog capacitive touch myself.
>>
File: d707.png (251 KB, 626x621)
251 KB
251 KB PNG
Need to replace diode in an old CRT monitor (no continuity either direction).
The diode itself has no markings whatsoever. The service manual has 3 different part numbers in the schematic and 2 different ones on other pages (pic related). I could find ratings for the first two of the three diodes mentioned in the schematic (the other two seem like misspellings of the first one. the third in the schematic seems to be an incomplete name. it should be ESJAxx-02), but they all differ by huge amounts:
VRRM | IF | VF | IFSM
ERB26-20 | 2000 | 200 | 3 | 10
ERB06-15 | 1500 | 1000| 1.5 | 50
The list goes on. How should I go about choosing a replacement? I'm worried because it has a safety symbol ext to it which means "REPLACE COMPONENT WITH IDENTICAL TYPE"
>>
File: file.png (146 KB, 943x599)
146 KB
146 KB PNG
>HF solar panel for cars
> youtube video says it can output ~18 volts
Could I hook this up to a cigarette lighter phone adapter, or would that overload the adapter?
>>
>>2141073
It won't be 18V under load, and the car battery itself (if connected) will clamp it down.
>>
>>2140924
Nevermind, I had a brain fart.
>>
I want a northridgefix / rossman style inspection camera on a stand or swingarm. I've trashpicked me a nice working 4K smart TV for a monitor. Need to get an HDMI camera w/ macro lens / ring light / stand etc. on the cheap.... suggestions? anyone else done this without paying $500-$1000? the cheapest ones I've seen ($130-ish) seem to come with smallish looking stands and fixed distance mounts, but I might be working with boards as big as xbox boards. I don;t mind fabri-cobbling some parts of this setup as long as it provides a stable image.
>>
>>2141061
>identical type
You just need to match the spec.
I also a few of those old, maybe high voltage diodes and they drop about 0.72-0.68.
>>
>>2141206
See what ubiquitous 4K camera modules there are from smartphones and such. The ones common enough to be put on a pcb for blinkies to plug them into their raspis or drones or whatever are going to be the cheap ones. If you have access to 3D printing then fabricating a solid mount for a macro lens shouldn’t be an issue. Ring light even less so. The only problem remaining would be getting its data converted to HDMI format, for which there should be some combination of available adapters that will suit your needs.
>>
File: Inductance Checker.png (93 KB, 2301x566)
93 KB
93 KB PNG
I should really stop trying to make PCB as small and compact as possible. Spent the whole day playing tetris, damn autism. On the bright side, design phase is done, tomorrow is solder time.
>>
>>2141240
Fucking autist, at least check the actual size and shape of the components you gonna use or switch to SMD.
>>
>>2141240
I just realised after posting, who the fuck teached you layout? Ants?
>>
>>2141276
>at least check the actual size and shape of the components you gonna use or switch to SMD
I did. No SMD for now, as all my components stockpile is TTH and perfboards.

>>2141278
>who the fuck teached you layout?
Literally nobody. Gimme a break, it's only the second layout I make.
>>
File: 1579612956651.jpg (791 B, 20x20)
791 B
791 B JPG
>>2141286
https://www.youtube.com/watch?v=GgmaFPR17qY
>>
File: file.jpg (3.64 MB, 3036x2747)
3.64 MB
3.64 MB JPG
behold, a capacitance decade box.
>>
>>2141061
>The list goes on.

a component with various part numbers usu means ''we used whichever of these we had in stock that day''
>>
>>2141240
DRC: warning, courtyard overlap!
>>
Was trying to think of the most simple method to have a bunch of variable capacitive buttons, for making a midi PCB keyboard with key velocity. I’ve gotten to using something like a 74HC14 hex Schmitt inverter with a 100k-1M resistance across it as an oscillator. Assuming the differences in finger force will lead to sufficient frequency variation via the flattening of the finger surface. But I’m not sure the best way to have an MCU receive a couple of octaves of data. Would it be better to convert that frequency into a voltage with a filter and detector or a phase frequency detector? Or would I be better off leaving the MCU to just count the frequency of a bunch of IOs?

I think it might be nice and elegant to have a single MCU for each octave board, and be able to plug multiple octave boards together. With a sufficiently rigid connector, that is.
>>
>laser arrived
AW YEAH

Hope 0.5W is enough, should be. Shame it won’t cut through polycarbonate since it’s not IR.
>>
>>2141240
Thicken up your tracks
>>
>>2141388
I'm thinking that frequency counting is going to be better, plus it doesn't require extra components or as many analog inputs as keys. But you'd need to know how to program the MCU such that it can count multiple keys at once. That way you can play chords.

Be nice to have some sort of touch-based sliders too, for pitch-bending and the like. I'm thinking long triangular touch pads, where sliding the finger up and down naturally changes the capacitance. But there's no repeatability or physical memory, so you might want to put LEDs behind the pad or something like that.

Some sort of capacitive touch rotary encoder would be great too, where you'd slide your finger around it like an old ipod. Just having 3 or more pads in a circle would work, but if the SNR is good enough it might be worth making tapered overlapping touch pads to get full analog rotation.
>>
File: 1622198871023.jpg (113 KB, 474x879)
113 KB
113 KB JPG
Ayo. I'm the anon from a few threads back asking about oscilloscopes and mentioned my mate has some in a throw out room in a hospital. He sent me some images and they have two Tektronix 475A, but I'm not sure what these other 2 are?
1/2.
>>
File: 1616716446292.jpg (110 KB, 918x527)
110 KB
110 KB JPG
>>2141592
2/2. Orange one to left is an old ECG machine. Just thought it'd be neat to include. Right one may also be medical for all I know though.
>>
>>2141593
The right one is obviously the machine that goes *ping*.
>>
>>2141595
I couldn't quite tell if the orange ports were ports or nobs. But yeah an ecg is what goes "ping" with your heartrate.
>>
>>2141592
This one says "storage oscilloscope" on it, meaning it may have a fair bit more utility than a normal analog scope. Like my Hitachi scope, which has the memory to store a single screen's two traces and transmit it to a plotter. Very handy for analysing single-shot events, though the lack of a longer capture that I can zoom into means I still rely on my logic analyser a lot.
There's a small possibility it just has a long-lasting phosphor screen that "stores" traces, but it doesn't look old enough for that.

Looks to have quite a lot of time/div numerations, meaning it should be capable of reasonably high frequencies, but you'd be better off with a high quality image of the front panel to examine any specs properly. There's almost certainly a brand name and model number on it somewhere, if not the front then the back. A scope of that vintage should have a service manual online somewhere, my Hitachi VC6545 did. If the scope has been sitting around for years without being turned on, you may need to give it a full calibration, in which case that service manual will be very necessary.

Other features you'll want will be external triggering (I assume that's what the right-side BNC is), horizontal zoom modes, XY, and cursors can be very handy too.

The 475 is a 250MHz (fast!) scope, and is designed to be rather portable. Looks like a solid scope.

>>2141593
This scope looks considerably older just from the form-factor and darker phosphor, chances are it's slower too. But it probably has enough features anyhow.

To be honest I'd get more than one, room permitting, having two scopes isn't a terrible idea.
>>
>>2141593
Looks like Tektronics 5000 series storage scope. The have a low 2MHz bandwith, so limited usability. If you were to sell it you would probably get 100-200 bucks.
>>
File: seerkeet.png (5 KB, 389x203)
5 KB
5 KB PNG
Sup, /ohm/. I'm about to design my first (-ish) piece of circuitry for my labwork at uni. I'm looking to make an amplifier, that cuts out 50 Hz, 100 Hz and anything above 1 kHz.

I'm probably going to use something like pic related, the first capacitor to filter out anything above 1 kHz, after that a resonant filter to filter out 50 Hz, then the op amp to amplify. Is this viable?
>>
File: interpolated touch wheel.png (240 KB, 1324x1046)
240 KB
240 KB PNG
>>2141490
>it might be worth making tapered overlapping touch pads to get full analog rotation
Looks like Microchip beat me to the punch. It also has a bunch of info about ground planes and shielding and such. Good appnote.

Also single finger presses are on the order of 1pF, so this might not be ideal with a 74HC14 out of the parts bin, but I'll give it a shot all the same.

>>2141620
Pretty sure your circuit is closer to a band-pass filter than a band-stop filter. And the op-amp being completely open-loop is nonsense. Inductors have unideal ESR that limits their Q factor, so I'd advise not using them unless you have a power application (speaker crossover, mains line filter, class D amplifier output, etc.)
>cuts out 50 Hz, 100 Hz
Consider a twin-T notch filter with feedback, they've got really high Q. Two would probably be necessary to get both notches cut out. I often see them being used to cut out mains hum. Then for the low-pass stage I'd probably go for a Sallen-Key or two, with whatever kind of roll-off profile you need. Your filter design depends on the distance between the passband and the stopband, the attenuation required at the stopband, and the linearity required at the passband. If you can get numbers to those three, then it would be rather straightforward to select an appropriate filter topology.

Whatever you do, this is a perfect situation perfect for using a spice's AC simulation. AC simulation with LTSpice is free and should only take all of 15 minutes learning how to do.

Good luck!
>>
>>2141622
Grand advice! I'm googling these filters right now and diving into LTSpice when I'm back from work, the interface looks as old as dinosaurs, but I'll figure it out.

Thanks a lot!
>>
I need to learn Pspice/LTSpice, is there any guides/resources that my ohmbros would reccomend?
>>
>>2141729
lrn2 use F2-F9 keys, fft analysis, and "ac 1" frequency sweeps
also transformers using K1 L1 L2 1 statements, oscillators using ".ic V(n001) = 4", etc.
>>
>>2141729
FesZ Electronics on youtube has an amazing series of tutorial videos on ltspice, they helped me a great deal. Other than that find or make a cheat sheet for all keybindings and keep it at hand, as controls are old school. For instance if you want to move something you have to press F7, which activates the movement cursor, and then you can pick and move; if you want to undo you have to press F9, and not Ctrl-Z which activates the zoom cursor. It's very unintuitive at first, as today's controls are very different and quite consistent across programs, but once you get the hang of it ltspice is quite an easy and robust program to use.

P.S. The help page (F1) has all the documentation you need.
>>
File: s-l500.jpg (20 KB, 500x500)
20 KB
20 KB JPG
What is the best solution for troubleshooting fused circuits that are blowing fuses? I seem to remember plans in an electronic mag years ago for an adjustable-current electronic "fuse" for troubleshooting. Can't find anything quite like this except chinese adjustable current limit breakers for AC line (picrel). Should I just get a bunch of breakers mount them in a project box and have banana jacks to each for connecting test leads? Is there some commercial solution that works for AC and DC circuits?
>>
>>2141846
>banana leads
You gonna shock yourself to death. Get a clamp meter, stuff like kaitweets will record the peak current, peak watts etc.
Also how you exactly plan to troubleshoot them?
>>
>>2139719
he seems like a nice dude, i actually used his videos to learn how to solder smd microchips
although what he does isn't really EE, his job description is solderer, he removes one chip and replaces it with another identical chip he took from a donor board
i could actually never do work like that because i find it incredibly tedious, it's like fixing dents on an existing car car VS actually designing and then building your own car from scratch
>>
>>2139723
just use bidirectional level shifter they cost few cents
>>
File: gl240-data-logger.jpg (99 KB, 1200x758)
99 KB
99 KB JPG
>>2141846
For AC get a clamp current transformer and attach it to some kind of datalogger
>>
Why is a diode required between a (switching) power supply output and a battery. Many say: so the battery power cannot flow back into the psu when powered off. Me: but the psu does have a diode bridge that will prevent current going back into the secondary coil. Yes it will charge the output caps, so what. What's the problem?
>>
File: 1510509958631.png (205 KB, 848x579)
205 KB
205 KB PNG
I'm attaching a heatsink to a pcb for the first time. I'm going to screw it on to a vertically oriented (that is, straight up-and-down from the board) voltage regulator, such that it's bolted to the part, but not attached to the board. Is there any reason I can't just hot-glue the heatsink to the board after I screw it in? Also, how important is thermal grease?
>>
>>2142061
It's considered good practice to mount the heat-sink solidly to the PCB, or to something that the PCB is mounted to. Otherwise the heatsink could vibrate around and put a lot of stress on that voltage regulator's leads and solder joints. Usually, a dedicated TO-220 heat-sink has holes on the bottom to bolt the PCB to it, or solder lugs to solder it to the PCB. I'd advise getting one of these if you're not stuck with your current one. Hot glue might be ok, but remember that hot glue softens and melts under heat, and you're putting it on a heat-sink, so I'd avoid it if the heat sink will be above 50C or so.

Some thermal interface material is necessary (unless you lap your surfaces) but it doesn't have to be thermal grease. A thermal rubber pad can be ok if you're not pressed for power, and indium pads are great if you don't need to worry about electrical contact. In your case you've got a bolt to clamp the interface, but in some instances you need to use some sort of adhesive thermal interface material. Like an adhesive thermal pad or thermal epoxy.
>>
>>2142077
Thanks for the reply. I'm actually going to cast a custom heatsink out of aluminum, for fun. I'll play it safe and bolt it to the board. The hot glue idea was admittedly a little retarded.
>>
>>2142087
>cast a custom heatsink out of aluminum
Ooh, complex stuff. What are you going to cast it into?
>>
>>2142098
frozen butter
>>
File: 20210616_180525.jpg (272 KB, 690x504)
272 KB
272 KB JPG
>>2142098
Going to make the model out of balsa wood and do a sand mold. I've done some pretty neat stuff before, picrel was a sand mold cast of a glass ashtray I found at a thrift store. Going to try casting some brass stuff after the heatsinks
>>
>>2142112
I'm not sure if it will be possible to make thin enough sand structures for forming the fins that can withstand the casting process. I'd recommend reinforcing the sand with wire mesh. What sort of sand are you using? Sodium silicate? Oil sand?
Mixing a bit of copper into the aluminium might make it have higher thermal conductivity, and maybe have a lower melting temperature also. Probably still won't be solderable though.
>>
File: 20210616_190156.jpg (3.72 MB, 3072x2546)
3.72 MB
3.72 MB JPG
Troubleshooting this Toshiba RT-200S Boombox

Wondering if anyone can help me trace this wire to its origin? The black wire I'm holding is the same one that goes to the 4 female connector. The white, red, and orange from the pinout goes to the cassette portion to the left, so I am assuming that's where the black wire goes as well?

I can't spot where on the cassette portion that the wire broke off though. I'm guessing it's on one of the pinouts where I circled but I have no idea and cant find a schematic.
>>
any recommendation for a fab and assembly company. i am shit at soldering and want to work with some surface mount only parts for a little project I am cooking up. usually i just choose through hole parts but my selection of parts is always trash if i do so.
>>
Basic question.

How does a DMM measure current again?
It has like a super small resistor inside right?
>>
>>2142201
it has a resistor inside but it's not *that* small and can sometimes cause problems with measurement. google "burden voltage". my good meter's 10A range uses a .04 ohm resistor.
>>
>>2142164
Is there a service manual available? If not, still try to trace parts of the circuit. As a black wire, there's a reasonable chance that it's 0V or otherwise a negative rail of some sort. So if you trace it to a ground plane, then all you need to do is trace the wires from that point-to-point mount and see which one of those also goes to ground.

>>2142174
>assembly
Learn to hand-solder SMT, it will broaden your horizons. 0.05" pitch components like SOIC are pretty easy, in fact I'd say they're easier than DIP in some ways. So long as you have a non shit tip geometry, I'd recommend a 1mm-1.5mm screwdriver tip. Then passives are a bit more of a pain, but still no big deal. 0805 should be solderable by basically anyone with good enough tweezers, 0603 needs a bit more finesse and decent eyesight, while 0402 probably needs a magnifier and ceramic tweezers.
t. does analog circuits and can find 99% of my needed ICs as THTs
>>
>>2142061
>hot-glue
>heatsink
am I the only one that actually feels sad reading things like this?
>>
>code doesn't work again
>reupload it into microcontroller
>still not working
>reupload it like 5 more times
>eventually it works
>unplug it from programmer
>plug it in via its usb power socket
>stops working
>plug it back into programmer
>still not working
>spam reset like 50 times
>eventually works
>further resets and it remains working
this project has brought about feelings of self loathing I never knew were possible
maybe I can just gonna make a loop to test if its reset properly, and to continue resetting itself until its reset properly
>>
>>2142297
did you by chance buy the micro from aliexpress or eBay?
>>
i wanna put a camera flash inside a watch
>>
>>2142323
Yes, but it happened with two micros from different sellers in different packages. So i don't think it's a hardware problem. Obviously something wonky is happening in the startup code, or with the memory of the module I'm getting the data from. When I inspect the traces with both my oscilloscope and logic analyser I was seeing missing clock pulses on occasion, but haven't yet figured out when and why that was happening. Missing clock pulses on occasion is something that shouldn't at all be happening with my code.
>>
>>2137653
Me again. I'm back with some more descriptive requirements :
Amplitude (50 ohm): 200mV - 5V
Waveforms: Sine, triangle, square 100uHz - 20 Mhz (possible from Atmega328P?)
Offset: +/ 5Vpk DC
waveform + DDS IC: AD9102
I'm trying to make it as good as possible for hi-fi audio equipment testing... I am restricted to using Atmega328P but can add additional hardware.
>>
do reed switches need debouncing?
>>
>>2142465

yes.
the ones that are mercury wetted need less, tho.
coz lubrication is life.
>>
>>2137653
>>2142420
Huh, didn't know there were readily available sinusoidal output ICs. The MAX083 is another one, though I wish it had a better phase detector.

>100µHz
for what purpose
>>
>>2142485
oops I mean meant MAX038
>>
>>2142467
>yes
it's all so tiresome
>>
>>2142514
If you use an SPDT reed switch (assuming they exist) you'll be able to put a small cap (100pF maybe) from the middle pin to GND, tie NO and NC to VCC and GND respectively, and just use that. The cap will hold the charge, and regardless of what bouncing happens it won't move because it requires hard pulling up and down by both sides, without resistors to pull it when floating.
>>
>>2142526
That would be great if it works, because otherwise i would have to do software debouncing which normally is very easy but in my case very annoying and hard because my project is battery powered which means i deep sleep between interrupts, which means no timers are running and i have no way of measuring how far apart the reed switch pulses are and thus are unable to tell if they are a legit triggering or just a contact bounce. The only software solution i came up with is convoluted and i hate it. If a small cap solves the problem that would be great
>>
>>2142529
although the probes i use have spdt reeds in them so i guess i can't use it in such a simple way
>>
>>2142533
>spdt
*spst
>>
>>2140484
It's going to be an overdrive guitar pedal but I'm stupid and I want people to be impressed when they change the battery.
I started this project 2 years ago and was displeased with how unprofessional veroboard looks so I took it from there and now I'm here.
>>
I'm trying to build a square wave inverter using a transformer I found inside a broken computer UPS. It can switch 12VAC to 240VAC. On ripping the protective tape on the transformer, I was able to find and wire up a center tapping on the low voltage side. This means that I could probably use a pair of FETs to alternatively switch the transformer to make 240VAC.
I am however curious, how do I avoid fucking up my FETs in case I go full retard and switch on the switcher on no-load on the secondary? That would be pretty bad right? If I'm going to measure the load on the secondary anyway, I might as well build something to regulate the switcher duty cycle according to the load
Also I'm also curious, do I really need one of those gate driver ICs in case I want a full H bridge with just N-FETs? Do I really need a few volts above Vcc to switch the high side FETs? Couldn't I use a voltage doubler for that?
>>
File: file.png (120 KB, 1500x771)
120 KB
120 KB PNG
so i decided to look at the reed switch through my oscope to see if there was any contact bounce going on when it triggers and... the fuck?
I am feeding 3V into the reed switch with current limit set to 10m, from me lab bench power supply.
One vertical column in the graph is 500us, with single trigger being set to capture the moment the reed triggers.
I took a small magnet a put it on the reed switch and how the fuck is this even possible?
The top pink line is 3V the bottom pink line is 0V
How in the name of fuck are there spikes above 3V? the top spikes go off screen at like 9V where the fuck is that voltage coming from?
Also the pattern looks a bit like a square wave, is that the contacts bouncing? Which would mean they bounced like 80 fucking times.
What in the name of fucking fuck am i seeing here?
>>
Ohmbros, I need your help.
It's a tangentially related question.
I'll be working with electronics in my new workplace and they need:
ASTM-F2413-11 shoes (OSHA-standard).
Any recommendations? Got a massive company budget of close to $200
>>
>>2142719
You need steel toe footwear? Just find your local Red Wing shoe store and try on their safety toe shit, see how you like it. Make sure whatever you get is made in the USA.
>>
>>2142742
Yeah, I'm going to be working in the field to check electronics in people's homes so we need to value safety above all else. I'll be learning a lot both from my bros here and the job, hopefully I'll be able to contribute more into the future.
Kinda excited desu. Will be using everything from PSpice to Autocad and etc.
>Make sure whatever you get is made in the USA.
Oh good point, I completely forgot about that. I can't risk my foot with stuff made outside of US standards.
>>
>>2142757
Awesome dude. All I can say is my grandpa swore by Red Wings (he was an ironworker for ~40 years). Just make sure whatever you get is comfy enough to wear all day. Steel toes are not particularly comfy.
>>
>>2142764
Neaaaat, I just called and they have a location nearby. Thank you again anon!
>>
I have an 1A analog ammeter. I want to extend its range to 1.5A. I figured I have to shunt it with a resistor having 2x the resistance of the ammeter. I don't know the resistance of the ammeter, my tools aren't enough to measure it, I have only a cheap multimeter, but it's small, bellow 1 ohm.
My question is, how do I go about picking a shunt for my ammeter?
>>
>>2142996
run current through it. measure the current with another ammeter, and measure the voltage across the shunt.
>>
>>2142996
Measure the voltage-drop of the ampmeter. Then you know the resistance.
>>
>>2142600
Look at the datasheet of something like an IR2183/2184, you'll see how it bootstraps the gate properly. That kind of bootstrapping circuit is reasonably trivial to do yourself. More difficult however is the gate drive current. For minimising switching losses, the current into the gate can be up to 2A or so with that particular half-bridge IC, which won't be that easy to pull off with a discrete circuit. Consider doing the calculations to see how much switching loss you can handle, but I'd still recommend using a dedicated IC since it's simpler anyhow.
Also make sure to simulate the driving of the transformer. I ran into issues attempting to drive a square wave into a transformer before.

I'm unsure if a centre-tap is even necessary, it's not like your transformer isn't isolated anyhow.

>>2142652
Current limiting power supplies may take a certain amount of time to respond to sudden transients, and their output capacitance will be after their current limiting circuit. Use no current limit and a pullup resistor instead.

>>2142996
>I want to extend its range
Doesn't that require printing out a new label for the display? Plus you'll need to get a specifically low tolerance current sense resistor. And since the coil is a reactive load, the added resistor may interfere with the measurement of AC currents.
>>
>>2142384
when you develop on a new platform spend the extra money on the manufacturer's devboard or a devboard from a trusted second party (adafruit, sparkfun, arduino, etc).
you need a known baseline so when your code doesn't work you know it's not hardware.
saying something like "i bought from two different chink sellers so I couldn't have gotten scammed twice" isn't true considering china is the fake capital of the world.
once you know your code works then you can pinch pennies in production.
>>
>>2143073
this isn't necessary. just assemble your project and then put it away forever when your code doesn't work on the first try.
>>
>>2143073
Good thinking I guess. But instead of getting a dev-board (since it doesn't exist) I'll just buy one of the microcontrollers from the brand themselves, and socket or solder it to a diy dev-board. Are Atmel parts on LCSC legit, or should I get them from Arrow instead?
>>
>>2143209
Allegedly, LCSC buys their stuff from authorized sellers. Since everything is made in china or in taiwan (belongs to china) then yea, they can get the real stuff if they want.
>>
>>2143209
LCSC from what i've bought is legit though the part's themselves usually contain cn partcodes. not sure what the difference is but never had an issue.
>>
File: amp_filt.png (51 KB, 1274x1230)
51 KB
51 KB PNG
Sup, /ohm/. I'm designing a circuit to filter out 50Hz, 100Hz and >1kHz noise and amplify the output, all in order to measure the I-V curve of a device. I've followed the advice of a kind anon and used two twin-T and two Sallen-Key filters, followed by an op amp, pic related.

After simulating it, it seems that the phase of a ~10Hz signal is shifted by ~50 degrees, does that mean I should likewise shift my V(t) measurements by a corresponding ~0.166 of a second, if I use a 1Hz saw signal as my voltage? Also, how bad is it that I attenuate the semi-high frequencies of 10-40Hz a bit? Should I compensate for that?
>>
>>2143336
Also, now that I'm checking the output, it doesn't seem to amplify it - 50mV in at 1Hz, 50mV out
>>
>>2143336
Original suggester here. I'd definitely consider putting an op-amp buffer between the twin-Ts. Or using active twin-Ts for higher Q, should make your response more linear. Seems you've gotten the hang of LTspice in no time.
Also I'm not too familiar with the internal model of the op-amp, but it may help if you have two 6V supplies to it, and add a ground reference between them. Just to ensure that your signal is sensibly referenced in the middle of the power rails.
>all in order to measure the I-V curve of a device
I suspect you're going about this the wrong way. With sufficient grounding, electric shielding, and magnetic shielding, this kind of filter shouldn't be necessary. Unless your device is too large to be shielded, or already referenced suboptimally. Here I was assuming you were making a radio to listen to Schumann resonance. Plenty of curve tracers I've seen just use a normal mains transformer as an isolated signal source, with a resistor in series for current sensing. Just put the ground of the oscilloscope between the resistor and the component being inspected, clip the probes themselves to the components on either side, invert one channel, and flip your scope to XY mode.
>>
>>2143336
>>2143368
Oh and let me clarify. A filter like that will distort your output data somewhat. If a harmonic of your driving frequency lies somewhere that's below 0dB, any nonlinear component (like a diode) will produce some of that harmonic, and so that harmonic will be attenuated to some degree. To avoid these kind of artefacts, I'd instead try to use a digital filter. Since you're talking about shifting your measurements' phase, I assume you're processing the data in a microcontroller or computer after taking them. In this instance, keep a sufficiently aggressive low-pass filter to get your full bit depth beneath your Nyquist limit, and ditch the twin-T filters. For those, you can just fast Fourier transform your data, delete any big spike near 50Hz (and precise harmonics of that spike), and inverse fourier transform it. What comes out will have no phase shift to it and very little distortion. Note that this requires that the 50Hz/100Hz noise is within your bit-depth in the first place, while keeping sufficient resolution across the range of the curve tracer's amplitude. This should be the case, but if not, you'll need to either increase the resolution of your ADC or add some sort of analog notch filter. The phase shift and other artefacts from that analog filter can be taken out with some somewhat simple Z-transform-based digital filtration.

Z transforms are kinda fun. Much luck!
>>
File: I-V sample.jpg (52 KB, 454x749)
52 KB
52 KB JPG
>>2143374
I'm working on plasma diagnostics, using a retarding potential energy analyzer - it has a grid, the potential on which prevents a portion of charged particles in the plasma from reaching the collector. So the current is determined by how many particles reach the collector at a given voltage on the grid, that's the I-V curve I'm taking. There most assuredly is a plethora of problems with shielding - there's a 150W antenna operating in the same room, the building's ground is somewhat sloppy and there is active wiring held together by tape all over the place.

Pic related is a sample of the curves - the top one's the current, the bottom one is the voltage, we used a 50Hz sine back then, the filter-amplifier circuit is for the new setup with a 1Hz saw signal.

I've been advised by my predecessor in the lab to use an analogue filter with further digital filtering afterwards, I'm not sure why though, so I'll ask about his reasoning.

In the mean time, I'll try to design a simpler amplifier without the filtering then, maybe it really should all be done digitally. Thank you for the stellar advice!
>>
>>2143391
>there's a 150W antenna operating in the same room
shit yeah that would do it

>to use an analogue filter with further digital filtering afterwards
>I'm not sure why though
Likely for the reasons I stated, that digital filters can have arbitrarily steep cutoffs and minimal distortion of both phase and amplitude in the pass-bands. Whether or not you'll need analogue notch filters will depend on your signal to noise ratio, seeing a sample of a signal without filtration would be useful for this purpose.

>cyrillic writing
Are you a slavanon? Having a massive antenna and plasma physics in a poorly grounded room sounds sovietcore as fuck. I'd like to be doing electronics in a physics lab like that, but my physics grades are too shit to get hired as a lab boy and nobody in academia looks at my actual experience.
>>
>>2143401
Yeah, I'm one of the slave boys germinating in Putin's plasma labs. If labwork in your country is anything like what it is here, they are retards for not taking you in - half the problems I encountered while doing research for my thesis was caused by either a shitty chinese oscilloscope or diy diagnostic instruments inducing epilepsy in the ADC. That's actually why we use cheap chinese scopes instead of quality ADCs - they're cheaper to replace and handle RF noise better.
>>
will there be any problems if i connect i2c devices with 2 meter cables? i will use 10k pull ups, or even 4.7k if needed
>>
>>2143569
for the cables themselves i will use wires from ethernet cable, i was thinking two twisted pairs, each pair will have SCL and GND; SDA and GND
>>
File: bulged_capacitor.jpg (64 KB, 600x600)
64 KB
64 KB JPG
I'm wondering if there's a way to tell if electrical components are "starting to fail" but haven't failed yet and if it's better to check for that than wait until it fails. Or does it not matter with electronics?

I'm asking because I have an old LCD monitor that works well but it's probably already 15 years old but I'd like to use it in the future too. Wondering if I should open it and perhaps look to see if I can find signs of components "starting to fail". Bulged caps are probably the easiest to identify and encounter, but what about other components?
>>
>>2143583
everything non mechanic except electrolytic caps will probably last longer than you will
i suppose the display itself will wear over time but replacing that has no point
>>
>>2143569
you will need lower pull ups to drive the cable high faster, but should be fine otherwise. start with 4.7k.
>>
>>2143588
>everything non mechanic except electrolytic caps will probably last longer than you will
Drive transistors that get hot are also likely suspects for premature failure.
>>
>>2143752
>>2143752
>>2143752
NEW THREAD



Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.