/diy/ - Do It Yourself

>>2630793
>Since neither are on, the LED gets the full current from 5V via the 2k resistor.
I understand this, but I don't understand why the LED is getting any curernt, since the current that controls Q3 gets sucked away. What makes the current going to diode so special that it doesn't go the route of the current that nromally opens Q3?

>What are you trying to do, make a minecart station with two pressure plates?
I'm just learning ttl

how difficult is it to reflow a usb C connector?

>>2631096
It's easy with the right tools. Use copious amounts of flux and a horseshoe tip for your iron. Keep some desoldering wick handy just in case, and don't overheat the pads. Try to keep flux out of the connector and clean everything with ipa when you're finished.

Watch videos about drag soldering.

>>2631184
>horseshoe tip
bevel tip

>>2631184
>drag soldering
but what are we in the shadow of Gods?

>>2631198

>>2631036
I have a horde of CRTs I've been slowly going through and cleaning out. Stumbled upon a box of tubes and this tester in a storage closet at works. Surprisingly it works and some of the tubes are good.

>>2631283
Heres most of the tubes I tested.

>>2631285
I'm not sure the "science" of it, but a lot of times tvs seems to run better after being opened up, grounded, and blown out with some compressed air. Only a couple times have I resorted to changing a tube or reflowing points on the board. Fuck combo units. The older ones with vcrs arent so bad, but the dvd player ones seems to be super finicky. I've experimented a lot with rehousing tvs, and so far I've fried every dvd combo tv I've tried to disassemble.

Pic related is a tv I cleaned out for an installation at work. Main thing is that these tvs will run 8+ hours a day 7 days a week, so built up dust and hair causes a very real fire hazard if not taken care of.

>>2631293

>>2631295
The simplest form of analog glitch hardware you can build is a 1k pot with left and right leg wired to video input and center leg wired to video output. Called a Dirty Mixer. Made popular by Karl Klomp. I work in an enviorment with a lot of children and they really like to twist the knobs on the tvs and fiddle with all the buttons. Decided to give them something that will hopefully draw their attetion so I stop getting calls to come turn a knob to change the channel because zoomers cant figure out analog tech. It bounces between two 12v cctv cameras. 12v transformer is attatched to the back. Added two 12v led bolt lights to the bottom for the hell of it since i had extra room on the transformer

>>2631301
You work in some sort of tech museum? Sounds like "look mum no computer" kinda stuff, real neat.

>>2631301
You mean two different video inputs or the same one in parallel?
Got any PVMs?

>be me
>know nothing about electronics
>computer mouse died, the button started to misclick
>rather than completely trash a whole mouse and outright buy a new one, decide to replace the broken switch
>how_hard_can_it_be.jpg
>get a "My First Soldering Kit" and a bag of chinkshit switches off ebay
>desolder the broken switch off, have to use pliers and just rip it off the board because I couldn't get it out for some reason.
>pop in the new switch, solder it in and button it up
>mfw it works
For every click I make, the electrical signals pass through a solder joint I made with my own hands feels good man.

>>2631367
great work not destroying the board when you ripped it off
next time try to cut away the plastic connecting each pin so you can desolder each bit separately.

also your solder joint better be somewhat shiny looking.

Is it true if you take apart the CRT of an old TV it can discharge enough current to kill you? Why doesn't the energy discharge over time? Can capacitors store charge forever? Thanks.

>>2631373
They can hold charge for a while, I don't know how long. It's a real danger. But the main danger with CRTs is that the tubes can implode and then explode. Even if you don't want to open the tube (I did to get the electron gun) be aware that the glass parts around there are relatively fragile.

I'm replacing the screen on my OnePlus 6t and the old one will not come off easily. I watched a lot of YouTube videos and read on forums but they all say just about the same thing. I used a heat gun at 90°c for 5 minutes all around the edges twice and it still won't budge. I'm using 99% alcohol on it now, should I use the heat gun again?

>>2631367
fixing your own shit is the most satisfying
great work anon

>>2631376
Is the phone actually getting hot enough? If not, try the microwaved damp towel method instead, the heat conductivity from a heat gun at low temperature isn't that great, and it's not like you should be turning it up much higher for fear of damaging the phone. If it is getting hot enough, then maybe you're not applying force in the right way or in the right area.

>>2631301

geez louise!
driving screws into a classic tv case can get you exfoliated by the purists.
to avoid that horrendous fate, try tie-wrap mounts next time.
they're very strong but off clean if you're careful.

>>2631301

geez louise!
driving screws into a classic tv case can get you exfoliated by the purists.
to avoid that horrendous fate, try tie-wrap mounts next time.
they're very strong but come off clean if you're careful.

What the fuck am I doing wrong?
I have three (3) inrush resistors for capacitive dropper now in series and shit still arcs over, if there is any sparking when you connect shit to mains.
Like each 1/4W resistor is rated 250V, and 750V is much bigger than 2x310V which you'd have in worst case scenario. Did I get chinked and resistors are not rated 250V? Or is it because Im dumb and I didnt account for tolerance (which should be 1%)?
What should I do? Bodge 4 resistors, preferably bigger ones, like 1W ones?

Given that IR cameras are really fucking stinky stupid expensive, how do I track living targets?

>>2631390
>build your own
if that’s too expensive you might consider just using your eyes

>>2631390
Do you mean thermal cameras are expensive? IR cameras are dirt cheap. You can build a high-power IR illuminator to spotlight a wide area, but it will be obvious to anyone with IR vision.

>>2631308
Kind of. We are a very unconventional place. Look up "City Museum" and there you go.

>>2631309
Here is the OG diagram. I decided not to use switches on this one, but same idea. >>2631367

>>2631373
If you are really concerned, get some rubber electrical hazard rated gloves and a good insulated skrewdriver. I am the other TV anon in the thread. Ive discharged maybe 15 tvs at this point and never had an issue. The big thing is to not just discharge it once and call it good. I leave a tv unplugged for at least a week, discharge it, disconnect the plug from the back of the tv, then wait a day or so and discharge again.

>>2631309
Forgot to answer. Yes I have a few PVMs. Mostly black and white but I have a JVC pvm i use for personal use at my home for editing and what not. Otherwise most of my tvs are consumer grade stuff i get for cheap in the local area.

>>2631390
If detecting presence is enough then PIR motion senor or mmWave human presence module. If you need tracking then scavenged webcam + any computer that can do enough tensorflow for facial detection.

>>2631386
I pulled this tv out of the trash when one of the residents of the bulding was throwing it away. Had to do some creative wood work on the inside to keep it standing straight. Its sitting out on the floor where at least 1k guests a day will see it and mess with it, so its gonna slowly be ripped apart. Had to do what will be the strongest. I do have several table top console tvs that I have repaired and kept clean and og

>>2631411
>so its gonna slowly be ripped apart
Take inspiration from arcade cabinets to mitigate the damage.

>>2631415
>>2631415
Too much time involvement. If I had unlimited time and budget, sure, but this is all stuff I work on after I finish my normal maintenance and construction tasks at the museum. About 6 hours or less a week is what I spend on TV projects so I gotta make the best of what I have. Trust me, if it was worth the care I would take care of it. Its just a generic colortrak made in the late 90s with some questionable parts of the board where water damage hit it. I'm just happy she works after litterally being dumped in a dumpster and the frame no longer being rectangular when I pulled it out.

That being said, we do have an arcade and pinball hall. Would be cool to go talk to my coworker who takes care of it and see if he has some spare parts laying around for joysticks or buttons. I've been hunting for a gutted cab for a while now to build my own glitch photobooth

>>2631419
Seems like a cool job. Here's a thing from long ago that may interest you:
http://www.mediumrecords.com/vs0/vs001/vs001.html

and

https://www.lofifuture.com/gbs8100-diy-video-synth-project

I hope battery power is a part of Electronics General thread.

I'm planning to build a homemade battery stack with solar array for my house. And since I'm poor it's going to be fully diy thing with parts from aliexpress. There are some questions that I need to figure out beforehand:
1. What DC voltage should I aim for?
2. Since the voltage is 'constant' that means max. current in the system will rise with bigger capacity?
3. How to choose a BMS and invertor knowing the voltage and max. current?
4. If my house burns because of my contraption, is it an insurance case lol?

I love videos so happily take youtube channels that answer those questions. Thanks anon.

>>2631465
LiFePO4 are pretty good at not bursting into flames

>>2631465
> 1. voltage
probably 48 volts DC.
> 2. current
For something like a hair dryer or a stove, yes you’re going to need wire twice as thick. Or you could use an inverter and up the voltage and use standard gauge wire.
> 3. BMS/inverter
Are you going to buy it or build it? I’d suggest you start small and add more BMS/inverter modules as you need capacity.
> 4. insurance
Too many variables to know. In general, modern insurance companies already expend almost all of their time and effort looking for ways to not pay out rather than actually insuring people.
>>2631575
> LiFePO4
Yep. Even very popular even with /rcg/ for many reasons.

Is there such a thing as a multimeter or continuity tester that I can plug headphones into?
I do a lot of reverse engineering PCBs and spend a lot of time checking what points are connected together and I'm pretty sure it's annoying as fuck for other people working around me to hear the constant pinging of the multimeter.

>>2631671
>continuity tester

build your own.
for headphones you dont need the 2 transistors.
just wire headphones to pin 6 in series with 100uF cap.
for stereo headphones, put speakers in series, not parallel.
other advantage is it only puts 40mV across probes against maybe 2V from DMMs.

>>2631428
i have all the parts for these projects sitting in a drawer at the shop. I really need to set down and put it together.

Here is something you might also be interested in.
https://www.etsy.com/listing/700238686/diy-analog-video-glitch-fx-mutant

>>2631733
I love this shit. I wonder what kind of effects you could pull off with a 3D display and active glasses. Hell, even the old cellophane red & blue glasses.

Tips for picking up a scope? Mostly focused in the realm of of MCU/MPU and supporting components. Thinking of grabbing used, but don't know what to look for and prices look to swing wildly on eBay from under $200 to over $10,000

Or hell do I even need a scope at all or should I just upgrade my cheapass USB logic analyzer to something not sparkfun

Planning out a little getting-my-feet-wet project. Haven't done any embedded since that one college class from which I still have an ancient PIC something-or-other lying around.
I was eyeing some fruit pi arm boards or some sifive risc-v's to hopefully make it easier to get compatible peripherals without having to think too much on this one, but then I've been wondering, do I actually have to contend with proprietary tools to program the chip or can I just slap a raw binary at address 0 using whatever magic sequence to turn on the programmer(? or can I use any serial-to-usb)?

>>2631793
When I bought one I bought the cheapest 100 mhz two channel Tektronix I could find that showed the self-test calibration signal in the photos. Came out to $150 shipped but that was years ago. It works great.

im trying to make a calculator that i can program,connect to my phone and use as macro pad with a Esp32 and an old calculator.
can i use a multiplex for the keyboard? i have some shift register sn74hc595n and multiplex from stuff that a friend gave me the multiplex seems simpler to use i dont know how the other one works is it better?
i checked with a led to see what parts are connected and i think it is divided in 14 x 4 lines, can i put the 14 as inputs and and the 4 as outputs and check every x time the line 1x1 then 2x1 .... then 1x2
to know what key was pressed?
suggestions?

>>2631858
> make a calculator that i can program,connect to my phone and use as macro pad with a Esp32 and an old calculator
so, you want to take a calculator and turn it into a calculator. got it.
hey, I got another idea. make something useful.

>>2631793
Unless you’re doing a lot of high frequency analog/peripheral stuff, yeah a logic analyzer.
You need at 6-digit $ scope to look at USB 3.x.
I usually just use another faster MCU instead of an actual logic analyzer though.

>>2631925
yeah, i know how it sounds....
well i want to be able to program it, so it can show me all the steps for lets say matrix multiplication.
and if i can connect it to my phone, it should be possible to search things in it and there is many other things, like using it as a macro pad.

>>2631687
Interesting that they used a transistor speaker output topology where you generally don’t need the cap, but then they put one in. I’m guessing that’s where R9 and R10 went.

>>2631927
A good use for those calculator keyboards is to make what we used to call µC “trainers”

If I were you, I’d make an external interface for it so you could plug it in anywhere.

And yes, the 4 lines could be used as outputs to “scan” for a keypress on the other lines to do a simple multiplex. This is where “scan codes” come from on a PC keyboard.

If you want to use less lines, you can do tricks with diodes and resistors.

Instead of 14x4 I’d just make it 8x8 as the codes are going to be somewhat arbirary. PCs had a keyboard rom to lookup what was what depending on the style/language of the keyboard.

>>2631390
Get a cheap PIR sensor (e.g. a thermal motion sensor), replace the diffractive optics with a narrow aperture, and put a galvo in front of it. Scan it across the desired area, sampling at each direction. Not sure how long each sample will take to be stable though. The other option is making an array of static PIR sensors, but that's probably going over $100 to approach a decent resolution. Each sensor has 3 pixels in it, but I don't think you can sample them individually.

>>2631465
>voltage
Higher means thinner wire and less complicated electronics for stepping up, but more complicated electronics for stepping down. Higher voltages generally have less available equipment for dealing with it, and for this reason the equipment may be more expensive.
>current in the system will rise with bigger capacity
Maximum allowable current will, yes. But chances are you'll be limited by energy capacity, not current capability. Current draw is decided by the load.
>bms and inverter
Take a look at what Daly have on their store page / website. They're a chinese brand but with a pretty wide range of stuff. Never bought from them, but a client decided to go for them to cheap out (was making his own pack from a bunch of salvaged 18650s). They might do inverters, I'm not sure. Generally the inverter will be rated at some voltage input, while the BMS and charger(s)/charge controller will be rated as so many cells in series (e.g. 12S, 16S) and for your chemistry of choice. As the other anon says, go for LiFePO4.
>insurance
Get this setup through your insurance company. Might be a pain, but it's worth it in case anything goes wrong. Even if it's not your battery shit that goes up, if a fire happens for some other reason it will complicate shit. The only risks if using LiFePO4 would be electronics and contacts getting too hot. So ensure your wire terminals are properly crimped, and that your inverter and such are in a reasonably fire resistant area.

>>2631935
ohh thanks.
scan codes, i will look at it.

>>2631842
There's a variety of open source tools available for working with different microcontroller families. Like AVRgcc and AVRdude. Ultimately it's quite doable to write your own programming hardware and software, but for that reason people have already done so. Things like the USBasp and SerialUPDI. I think there are similar things for PICs, while for STM32s I think the STlinkV2 is a widely cloned design anyhow. PICkit clones exist too. Not too familiar with Risc-V, IIRC some manufacturers were making their own programmers but I'm not sure how compatible they are.
For more complicated debugging tasks, I think you'll find that diy hardware doesn't quite cut it, people still buy the ATMEL ICE and such for that reason. Which for some reason don't exist as clones or diy projects, chances are through locked down firmware. But there is the capability for free as in freedom debugging tools via GDB, with some level of support for a variety of microcontroller families, even if you often have to use proprietary hardware.

My only advice is to not skimp on hardware design to "figure it out in firmware", that just leads to more pain further out. Read your chip's pin multiplexing capabilities if nothing else.

>compatible peripherals
Anything that uses SPI or I2C or UART will be compatible with 99% of non-shit microcontrollers made in the last 30 years. Clock speed is the only thing that might matter, for things like displays. I'd recommend the Pi Pico for messing about since it's pretty good and pretty cheap, but if you do design your own boards for it you'll need all the supporting componentry.

>fruit pi arm boards
Does your project in mind require a full computer?

Also see >>>/diy/mcg

>>2631928
>you generally don’t need the cap
Because the output of the op-amp would default to 0V anyhow? Maybe the asymmetric resultant waveform didn't sound right, or quiescent leakage current seeped through anyway.

The crystal added to ground as a modification is interesting though.

>>2631951
> figure it out in firmware
“firmware” was invented by the hardware guys so the buck could be passed to it.
Luckily, the firmware guys quickly invented “software” so they, themselves could pass that buck…and the modern, buggy-as-hell world we’re living in was created.

the duality of man

>>2631951
>Maybe the asymmetric resultant waveform didn't sound right

right, a cap (of the right size) can turn a square wave into a triangle which is less grating on the ears.

>>2631960
>“firmware” was invented by the hardware guys so the buck could be passed to it.
>Luckily, the firmware guys quickly invented “software” so they, themselves could pass that buck…and the modern, buggy-as-hell world we’re living in was created.

How do you make it through the day without blowing your brains out, anon?

>>2631960
>and the modern, buggy-as-hell world we’re living in was created.
Deliver bug-free hardware to us before you say that, dumb EE.

>>2631951
>Ultimately it's quite doable to write your own programming hardware and software,
I want to do it for my own edification. If there are open-source tools then I can gaze into them and figure out how they trigger payload delivery, which is really all I need at this stage (I wrote my own little programming language for fun that can do either raw binary or ELF, easy enough to modify for other targets than x86, too).
For debugging, I'll have to experience it first to see how I can reproduce something useful. Maybe go with a simulator for whatever board I end up with might be a fun project.
Thanks for the software + hardware references, that's just what I was looking for!

>>2631951
>Does your project in mind require a full computer?
Basically no, and especially, it shouldn't require one on-chip. Basically sensors would report to a connected phone or computer which would perform some pretty extensive processing. The control board needs to operate an automated feedback-based system at very high speeds, and to operate in a few different modes (which may be changed or updated later).

>Also see >>>/diy/mcg
I tried asking a few questions but got no response at all. I'll give it another try when I get into it more.

i lost my last 1nf 0603 capacitor that i needed to finish a circuit, i have nothing else to buy and didn't plan to do more projects until September
(¬-¬)

I want to power my old Galaxy S3 phone with alkaline AAs because I'm tired of buying lithiums that puff up like balloons after 6 months. I rarely take my phone off the charging cord so rechargability/capacity isn't really a concern. Been looking at this schematic for a plug-in replacement (https://www.electroschematics.com/mobile-phone-battery-emulator/)) and thinking about adapting it. Has anyone else done something like this? Any thoughts or suggestions?

>>2632306
>tired of buying lithiums that puff up like balloons after 6 months.
>rarely take my phone off the charging cord
You should unplug your phone and let the battery discharge normally. Turn it off when you plug it in to charge so it doesn't take as long.

>that chinesium pipe USB microscope finally stopped working
>because the aluminum wire cable literally dissolved
>try to disassemble it
>now it's destroyed
>that garbage is twice as expensive now
Did anyone DIY a microscope? The problem is you need optics too.

>>2632306
The “great scott!” Channel did something like this not too long ago

>>2632317
You can do an optical microscope for $1, see the foldscope.
You can do a nanometer-resolution microscope using either STM or AFM modes with just $50 of material (probably less). It actually requires only simple control electronics, too. See
http://web.archive.org/web/20090304013944/http://www.bsc.ustc.edu.cn/~jlyang/research/STMWebPage.html
https://www.angelfire.com/electronic2/spm/index.html
http://www.e-basteln.de/other/stm/mechanics/
https://dberard.com/2015/01/12/a-home-built-scanning-tunneling-microscope-with-atomic-resolution/
Modern microscopes aren't even optical anymore because of the physical limit from photon optics. Tricks like fluorescence-based hacks or 4pi don't go that far, and interferometric microscopy or raman microscopy is also too limited even though it's better. The future will eventually be atom optics (using helium exploiting wave-matter duality), but for now we have scanning probe microscopy (and fuck SEM/TEM).

>>2631980
Better yet, use a piezo electric transducer… which *is* a capacitor (although that’s not especially relevant)

>>2632306
>Galaxy S3 phone with alkaline AAs

AA's wont fit inside the case, but i guess that's not a big deal.
i got a 2 dozen new lipos for tablets, super cheap, so i use 'em for everything.
phones, flashlights, cameras, scales, radios, thermometers, etc.
unfortunately they cant put out a lot of amps or i'd use em for tools as well.

replacing a lipo by another lipo requires no changes to the charging circuit.

>>2632323
I want a magnifying glass, but one which shows its output on a screen. These "USB microscopes" are pretty shitty for their "intended" use as microscopes, but they do well as glorified magnifying glasses, like for looking at PCBs. It needs to be optical and output digital data.

>>2632335
Then you can just slap a camera over a sapphire bead depending how much you care about artifacts (if you do, I actually recommend doing software correction, but otherwise you have to do the lens-tube length-convex-concave gymnastic). It's mostly meant for biology but here are example images you get with that kind of cheap setup.

>>2632273
bodge a tht ceramic there instead

>>2632306
>not modifying it for replaceable LiFePO4 cells
ngmi
Also never charge a lithium ion cell over 4V if you want it to last long, I think droid phones have the ability to set the charge cutoff voltage in software. Leaving them plugged in for ages without changing that setting is just asking for them to die.

>>2632306
alkalines leak.
get LiFePO4 or NiMH

>>2632313
That's impractical
>>2632318
Thanks, I'll check it out
>>2632331
Yeah that might be the smarter route. I'll think it over
>>2632350
I didn't know this. Thank you for the heads up. I wonder if that's why I've been having so much trouble. Maybe I'll order one last replacement battery and try this before I break out the soldering iron and make things complicated.

kicad 7's knockout text is really useful, same for different ground fills not intersecting each other
but the small hitboxes for selecting symbols is kinda strange, i preferred the kicad 6 behaviour for that. when you're really zoomed out it preferentially selects wires instead of symbols.

>>2632313
>You should unplug your phone and let the battery discharge normally.
no you shouldnt, idiot

>>2632392
Yes you should, retard.

>>2631389
used 4x1W carbon resistors, should work now

>>2632323
I've been meaning to ask, wtf is the point of the tuning fork assembly? Is is just a matter of gain? I know some piezos have a natural resonance of like 2.5kHz

>>2632411
It's to get stable sampling rates at high speeds across the sample isn't it? The frequency is a lot more stable and it can be used for longer due to better temps compared to a silicon cantilever for example. It's also easier to drive. They're also very cheap. What else would you use? It's pretty much either tube, fork or buzzer.

>>2632397
Lithium batteries are not NiCd batteries. Recharge cycles destroy it. Yes, keeping it at 100% would also destroy it.

Ideally the battery would be kept at 50%.

>>2631198
What kind of flux, or applicator, is that?

>Check out an instrument store for some parts
>They sell pots
>$10 each
>Lots of reviews
hehh

>>2632530
>$10 each

every crumb sees high prices as an annoyance, when it's actually an opportunity.
buy 10-cent chinese pots, sell 'em for a dollar + shipping.
1000% markup means you'll be a millionaire in 6 months.

>>2632471
>What kind of flux, or applicator, is that?

just regular old flux and regular old squeeze bottle you find in every regular old home.

that wasn't worth a repost

>>2632468
>Lithium batteries are not NiCd batteries
Thatnks for pointing that out, Anon. I never could've figured it out. I'm gonna write a book now.
>Recharge cycles destroy it.
Well fuck my ass. I had no idea that USING THE BATTERY DESTROYS THE BATTERY. lmao go back to /g/ or plebbit or hackadong, faggot

I want to replace these failing membrane buttons for their mechanical equivalent. Any good recommendations for durable mechanical buttons with transparency? The panel is under-lit. I'm not paying 500$ for a board.

>>2631036
are there any cheap "surplus" puck SCRs that have data sheets? i want to buy one for dicking around with autistic pulsed power experiments but i don't want to pay $500+ for something new/modern. i did a quick ebay search and i can't even find a data sheet for most of the ones that are showing up.

>>2632586
try AOL keyword "SCR TO-66"

>>2632563
I think I will just get Cherry MX keys and solder them onto a prototyping board, like this

>>2632563
One can literally take an off-the-shelf prototyping board and make better buttons than this shit. But I'm going to go to town and use actual computer keyboard buttons that fit into a board as through-hole components. any comments welcome.

i have a shitty evaporative cooler which has a single phase AC capacitor run motor in it for the fan. the current problem is that the motor heats up a large amount after running for a while and also slows down a little (that's just the symptom though the main issue is the overheating) i can actually start to smell some stuff after about 5-10 minutes of operation, feels like idk, 55-65 deg C. what might cause this? the bearings are ok.
history
>stops working
>call motor guy
>changes the cap (4 uF)
>runs
>breaks down again the next day
>remove motor, somehow works on the table, still a little too slow
>add another capacitor in parallel with the existing one (2.5 uF)
>works for a few weeks
>stops working
>add capacitor (2.5 uF)
>stops working
>add 2 capacitors (picrel only shows 4 it was a wip) (2.5, 2.5 uF)

>just change the motor bro
no i am not spending that much money to keep this rusted out cooler running.

>>2632615
Landline telephone pads are pretty robust.
>>2632619
Use the metal can caps.

>>2632622
>Use the metal can caps.
what would that do?
Do you think the issue is with the caps?

>>2632623
They're more resilient than the caps in the plastic case. Have you tested any of the old caps? Does the fan spin freely when you push it with your hand?

>>2632626
>>2632626
>Does the fan spin freely when you push it with your hand?
Yes, and when it fails i.e. just vibrates with no motion, it doesn't. What does this signify, in general?
>>2632626
>Have you tested any of the old caps?
I did just yesterday, two of the 2.5 uF caps were reading 3.7 uF. But the fan spins so i guess the current capacitance is ok.

>>2632627
>vibrates with no motion
The motor could be in the beginning stages of failure, but it's more likely crappy caps. If the fan doesn't spin easily by hand (when off) then it needs replaced.

>>2632632
>If the fan doesn't spin easily by hand (when off) then it needs replaced.
it does spin easily when off
>>2632632
>The motor could be in the beginning stages of failure
but that's also what happens when i leave the cap wires open i.e. no capacitance and an open loop for the auxilary windings, so i have always thought it to be some bad capacitors. Anyway what might be the reason for the motor refusing to spin by hand freely? is it jsut the bearings? i thought it was something like an electromagnetic brake but then it dawned on me that since there are no permanent magnets in the motor and the current is off there is no magnetism.

>>2632635
If the fan spins freely when cold but doesn't when hot, the bearings/bushings are probably overheating and expanding.

>>2631976
There's an Indian engineer in the electronics manufacturer I work for and his name is Deepak.
I always smirk when I hear his name.
It was his destiny to design PCBs

>>2632543
It's clearly gel flux from a syringe

>>2632671
>his name is Deepak.

I always liked Jugdish in the loser's group in Animal House. No way could you make that movie these days.

>>2632563
>>2632609
>>2632615
Get on my level

>>2632635
>what might be the reason for the motor refusing to spin by hand freely?
Rust and loss of lubricant, which leads to deformation of the ball bearings and increases friction.

>>2632680
Nice, that's a Commodore 64 keyboard from the 80s. I'm going to do something like this but using off-the-shelf keys that fit into a board. Not the cheapest thing possible but also not a 500$ module.

>>2632680
The panel is under-lit with LEDs, say if you enter a code wrong it light up purple. I will just make a hole where every RGB LED is.
It amazes me how a projector manufacturer makes such a cheapo membrane button on a cinema projector, but I guess these machines are now obsolete.

>>2632688
You could do leaf switches and spring-load the key caps.

>>2632694
Yes, I'm thinking of overlaying the green board with a piece of plastic that just has holes for standard keyboard switches that are transparent and ditching the membrane panel.

>>2632696
That'll work fine. Do you need to waterproof the panel?

Hey, I need to trigger a 5v circuit using a 3v supply, is there a very small kind of relay I can use for this?

>>2632711
You can get level shifters or just you know use a transistor

>>2632698
I don't really need to, I think recessing the switches in the board will be waterproof enough in case someone splashes antifreeze around. The bigger problem is with corrosive humidity. After the soldering is done, I think I will use nail polish on the solder points. The designer of the projector thought that people won't really use the buttons because they have software to control the machines, but it turns out that they actually do, to a point the membrane buttons wear out.

The shift, focus and zoom buttons are not under-lit therefore I have free reign as to how I lay them out. I'm open to brutal criticism and suggestions. The most used button is power, sometimes they press 'dowser' which is a flap that blocks the light from coming out. I would move away the zoom UP button from the DOWSER button because they might press it by accident.

This is what I think I will do. The buttons will be grouped together based on their function: basic controls, image adjustment, pre-sets 1 to 6. I'm not sure about the image adjustment section, maybe the buttons could be laid out better so it doesn't look ghetto. The basic idea is to replace membrane buttons with real ones but without making it overly expensive or look like trash.

Is an unvarnished MDF surface an ESD safe work area? I heard cardboard is basically ESD safe, suppose we could extend that to bare wood?

>>2632558
well now you know, retard. Look up recharge cycle ratings for Li-Ion and you'll know how many recharges it takes before the battery is dead. and of course, halfway to that number you'd lose half the capacity to when its considered dead.

quickest way to kill a lithium ion battery is probably to fully charge it and fully discharge it constantly. Even faster than keeping it at full charge constantly.

>>2632846
>quickest way to kill a lithium ion battery is probably to fully charge it and fully discharge it constantly.
We're talking about a phone, not a quad or vape pen. Phones have integrated charging and battery protection circuitry which prevents full discharge (among other things) even if it says 1% on the display, and the drain is mild compared to vapes and drones.

Dumbass.

>>2632586
Maybe look on LCSC? Google translate may be a requirement for those datasheets.

>>2632750
>I think I will use nail polish on the solder points.
You could also use conformal coating.

>>2632857
>which prevents full discharge
completely irrelevant. a "full discharge" implies about 3.2V or so, not 0V. You have to be pretty retarded to think I meant 0V.

You still dont seem to know what a charge cycle is or acknowledge that lithium cells are rated for a limited number of them. a "full charge and discharge" is 4.2V down to 3.2V. This is a full charge cycle.

if you want to kill a lithium battery quickly, then you will constantly charge and discharge it.

>>2632934
Your dumb pedantic ass still doesn't realize that batteries are consumables and die no matter how lightly you suck their dick. Your shit also runs hotter when you're charging and using the device at the same time. Go cut your dick off for science, faggot. You smell like a zoomer tranny.

>>2632543
I was the guy asking. Thanks the info anon.

>>2632954
>things die eventually so you may as well just abuse them so they die faster for no reason
no thanks, ill use my batteries and everything else in ways that do not accelerate their destruction, unless I have some compelling reason to

Thoughts of my 2S BMS circuit?

>>2632996
>unless I have some compelling reason to

one compelling reason is to not appear to society to be an obsessed OCD autist.
to the bitches, that's even worse than being short.
another compelling reason: you wont survive the zombie apocalypse on a 50% charge.
unless you carry one of these all the time.
like i do, as a fellow autist.

What are some good electronics related home businesses? Anon in the cell phone repairing thread says it's not worthwhile anymore.

>>2633079
>home businesses?

one company i worked at hired 1-person outfits to do PCB design, and to develop embedded processor designs that the resident engineers couldnt handle.

>>2630912
>monitor just died again
>take it apart
>the new capacitor is bulging and needs to be replaced
What could be causing this? I used a 470µF 16V LESR cap. The maximum temperature is the same as the old one 105°C, but the old one lasted for years and this only lasted for a few days.

>>2633113
Maybe the ESR isn't low enough and it's heating too much? I'd up the voltage rating too, just because a wider ratio between the exposed voltage and rated voltage usually means a longer lifetime.

>>2633084
I'm afraid I'd have to register as self employed to do that and I'm not going to bc of my country's politics. I want something to offer to the final customer so I can have an illegal workshop. That's why I was thinking about phone repairing but it seems it's a bad idea now. Maybe small appliances or something?

>>2633127
>Maybe small appliances or something?

computers would be your best bet.
substantially every owner is dissatisfied with some aspect of their system.
mostly it's software tweaks, so no need for much inventory, except knowledge.
another huge opportunity is when windows slows to a crawl.
a fresh windows install on a new solid-state drive will make owners giddy.

>widget doesn't work
>tfw have to painstakingly check every node now
fuck me i'm not autistic enough for this shit

i have solar panel and i got a car battery and pwm power controllers (since mppt is too $$) so basically what sort of power is coming out of that thing? stable 12V? since the car battery is 13V tops? can i just hook a buck to the output of the pwm controller to get nice clean 5V to charge my electronics?

Guys, i have two wall mounted speakers with a 12V supply in hand, is there a board i can get with bluetooth to make these speakers smart?
Would any of the following be any good?

https://aliexpress.com/item/1005002253796079.html?algo_exp_id=476ef969-423d-484e-a166-34220f748c73-47&gatewayAdapt=glo2bra

https://aliexpress.com/item/1005004830515523.html?gps-id=pcDetailBottomMoreThisSeller&_t=gps-id%3ApcDetailBottomMoreThisSeller%2Cscm-url%3A1007.13339.291025.0%2Cpvid%3Abb42ce7f-22a1-49aa-a921-84b273baae5a%2Ctpp_buckets%3A668%232846%238116%232002&gatewayAdapt=glo2bra

>>2633181
>Would any of the following be any good?

i wouldn't chance it.
- you dont know if bluetooth is well implemented, so you actually get a stable long term connection
- you may send your money and never receive anything

what i'd do instead is get a thrift store unit with a dying battery, and pull the electronics.
and i'd power with a 5V 3amp regulator in series with a diode to get 4.3volts.
of course you want the stereo units, not the more common mono ones.
the one on RHS of pic is 20W stereo and cost $6.95

>>2633199
>get a thrift store unit with a dying battery, and pull the electronics.
I do this with old BT earbuds, and power them with a single 18650.
>5V 3amp regulator in series with a diode to get 4.3volts
Some BT boards have a regulator on the PCB so they'll be ok with 5V.

>>2633061
Learn to schematic

>>2633159
No, it will the spikes generated from the pwm will likely get through the buck and destroy the electronics it’s plugged into as a bonus.

>>2633115
> lower esr and higher voltage
This ^^^ get the highest voltage you can stuff in there.
If there area is too hot, run some wire to your bigger cap you’ve ziptied to something in a remote, cool location.

I have an electronic instrument that blew its fuse. According to the service manual it needs "Type GMC, 5mm x 20mm, Time Delay, 250 Volt, 1.25 Amp fuse".
I have all these fuses in possession but I don't think they say their specifications like this. They look to be the same size I want, some of a wire connection inside, some have a long steel tab. The fuse I blew is a wire. Are these safe to use?

>>2633273
the output cap on the buck should catch these spikes

>>2633280
https://www.amazon.com/5Qty-Littelfuse-T1-25AL250VP-1-25A-Slow-Blow/dp/B07L5FT1RF

Look at the metal end caps with a magnifying glass.

>>2633285
All the wire types I have seem to be 250v, and on one end of the caps say "buss agc" followed by a number. I guess I should find one that would say 1.25 or 1 1/4?
>https://www.amazon.com/5Qty-Littelfuse-T1-25AL250VP-1-25A-Slow-Blow/dp/B07L5FT1RF
I was checking out amazon but couldn't find "type gmc" in any listing. Does that matter?

>>2633294\
>gmc
>Does that matter?

no.
seems to be a Bussmann-specific thing, where M is for medium-speed slow-blow.
as opposed to GDC which is ''normal'' speed slow-blow.
must be metric. or communism.

do small 3 phase AC motors exist? i want to play around with making my own variable frequency drive (running it off a 30V DC power supply or something) but i don't want to drop $500 on a 40 pound induction motor that's going to bend my shitty plastic desk.

>>2633324
Any RC plane or quadcopter motor would work. Or you could probably even salvage one from old hard drive.

>>2631036
I have a bunch of MLCC but no electrolytics. Can I solder them in parallel and use them in place of a larger electrolytic capacitor?

>>2631390
The USA makes the production of proper bolometers illegal. But the Chinese do it anyways.

>>2633334
sorry, i meant 3 phase AC inductor motors (asynchronous) as opposed to 3 phase BLDC motors (synchronous.)

>>2633337
depends on the application. sometimes the much lower ESR of MLCC can cause issues that otherwise wouldn't be a problem.

https://www.youtube.com/watch?v=2287hcWFqe8

>>2633342
I need to use them as storage caps for the power rails of a TV PSU

>>2633268
eat my ass

>>2633357
that seems like it would be a shit ton of MLCCs?

>>2633324

They're kind of uncommon, at least in terms of cheap, plug-and-play motors. You can certainly find them if you're willing to shell out a bunch of cash, but that's waste when you're just experimenting.

Try looking for used AC servos. Small ones can be had for fairly cheap. Short of that, they can be sometimes found in industrial pumps and fans.

>these extremely thin ribbon cables that are more like flexible PCBs, often used to connect LCDs
I have 2 pin one, which is a more simple case. It's cut in two, any ways to repair it? The mean part is that it's basically a part of an actual PCB, really hard to access physically.

TL;DR how can I lower the output leakage of an analog switch
I'm building a multiplexer for an LED spectrograph (I posted about it last month if anyone remembers somehow)
I'm starting to breadboard the circuit. Just for breadboarding, I'm using a CD4017 to drive the different switch control inputs of the analog switch, a DG201. (I'll be using a microcontroller and better digital switch for the actual build)

But, there's a pretty large amount of output leakage. For a 5Vpp sine wave, there's 1.24 Vpp at the output when the switch is "off". I'm 90% sure I had everything wired up right, so what can I do to lower the leakage?

pic is of a scope reading from the breadboard with the aforementioned 5Vpp sine wave (in green), and another sine wave whose amplitude I forget. You can see both have significant leakage during their "off" phases

>>2633341
Boil a BLDC past its curie point, then it will no longer be synchronous. Also probably wouldn’t work well at all. Trying to make your own squirrel-cage rotor from some silicon steel laminations could be fun.

Rewinding a shaded-pole or two-phase motor might be doable.

>>2633392
Could try wire glue if it’s not high-power, but I’d want to solder using a bodge wire to bridge the gap. Either way I’d use epoxy to provide strain relief to the whole joint. Or just run a separate pair of wires.

>>2633426
Do you have load resistors on the outputs? Something like 1kΩ.
You can also use (J)FETs to ground the outputs synchronously with your switching.

>>2633462
The outputs of the switch were completely unloaded save for the scope probe. Would that have caused the problem? I would have expected the opposite to happen.

Does this mean I shouldn't buffer the outputs with an op-amp?

Has any of you built a DIY Segger jlink debugger? There are schematics online, I built one, and it is recognized as a Jlink device and it prints some diagnostics to screen but it can't connect to the target attached to it.

>>2633426
I don’t know why people use analog switches when you can just put a couple of mosfets back-to-back and achieve the same thing with 1000 times less “on” resistance.

>>2633496
because I'm dumb and gullible and don't trust my lackluster skill in using the mosfets as bidirectional skills to the best of their ability and would rather defer to a pre-packaged solution that more or less guarantees correct functionality and operation so long as I follow the datasheet

>>2633497
>bidirectional skills
switches

>>2633482
>Would that have caused the problem
Very likely.
>Does this mean I shouldn't buffer the outputs with an op-amp?
Buffering is a good idea after any analogue switch, but I'd still add a load resistance. I'd combine the two and use an inverting amplifier, since inverting amplifiers are better at avoiding nonlinearity of their inputs.

>>2633485
Sounds like a really useful project, I'll have to look into that soon. No clue about your problem, aside from simple shit that isn't in your firmware, like output pins getting swapped, computer software settings, or dip-switches being set wrong.

>>2633496
DG series analogue switches basically are that. But they use 4-terminal MOSFETs, which are better for some reasons. Lower diode capacitance if nothing else. Go for a DG4 or DG6 and you'll get some damn good specs anyhow.

>>2632404
>>2631389
Anyway, somebody asked a spectrum and CRI or chink LEDs? I've finally found a CD to see what is actually going on.
I don't really have the spectrum, but it correlates to whatever seller told me and to what I see really well.
Maybe you can extract spectrum from it somehow, but I can't be bothered.

r8 my solder job

Also how the FUCK are you supposed to splice tiny gauge wire, like for USB or headphones? I can't get the wires together without breaking them. And then how are you supposed to keep them from shorting when you're done? Heat shrink doesn't come in tiny sizes.

>>2633617
>rate
fucking terrible
>how to
like every other multicore wire, or put more simply, just stop being a tard
>no small heat shrink
yes it does. go buy some for the appropriate gauge of wire

>>2633620
I will have you know it works perfectly well.

>>2633617
>tiny gauge wire
>like 24awg
>tiny

>>2633617
its fucking ass, git gud

>>2633617
fold electric tape around one wire, wrap the excess "flag" of tape around the pair, then heat shrink the whole bundle to keep it in place

i hear these reverb ics are tough to come by, but can't you just use a pair of record/playback ics like the isd1820?

>>2633589
cool stuff anon

>>2633617
it is bad
use more flux and more solder

>>2633617
>remove isolation
>drill wires together
>put solder on your iron, immediately wet the wires with it
>drop hotglue on it
>practice and repeat until it doesn't look like gore

>>2633617
Replace the entire cable. Don't be cheap AND stupid.

>>2631036
How do we make a gun? None of that ak or ar glownigger shit.

>>2633657
Roll up a National Geographic magazine into a cone shape and wrap tape around it. Stuff a small piece of tissue in the small end with a wick, and pack some black powder into the open end of the cone. Add shrapnel, point cone and light fuse.

Okay, it's more of a spicy nerf gun, but you're in the wrong thread and on the wrong board, so go fuck yourself.

>>2633657
>gun
That's a vacuum tube display, like an old TV.

when i try to charge my phone with lab power supply and set it to 5V the phone is only drawing 800ma even tho the power supply can do up to 6A, how come?

>>2631036
>image of an ancient scope
>even transistors have a socket
Did they do this for easier repair and maintenance?

>>2633617
This HAS to be bait. Right?

>>2633694
>how come?

coz 6A would make the battery sizzle and smoke.
the phone only takes what it needs, and gives according to its ability.
coz Karl Marx said so.

>>2633718
thats not true, phones normally draw way more than 800ma to charge otherwise it would take all day

>>2633697
>socketed transistors
>Did they do this for easier repair

it's coz, in the tube days, transistors were so rare and expensive that people would pull them from one device to use on another.

>>2633721
>otherwise it would take all day

say the battery is rated 1000mAh
that's 1.25 hours at 800mA for a full charge.
that's only 1/8 of a day on Jupiter.

>>2633694
5V @ 900mA meets USB 3.0 / 3.1 standard. Are you charging from a USB port on the power supply or the terminal outputs?

>>2633733
i just cut open usb cable and connecrted + and - to 5 volts
i was hoping i could get fast charging or such since the power source can easily handle it

>>2633735
Put a 150 Ohm resistor across D+ and D- at the cut end of the cable.

>>2633694
Is it USB-C?

>>2633740
yeah

>>2633736
i already tried that and had no effect

>>2633764
do you have an iphone

>>2633768
i'm not a fag

>>2633770
just checking

>>2633764
I don't remember the details how PD handles current, but you might need to talk to the PSU with the PD protocol.

>>2633617
Solder shrinks, my dude. It's like a regular heat shrink but has a ring of low temp solder inside that melts when you shrink them.
Here's an example, you can find such kits just about anywhere.
https://www.aliexpress.com/item/1005005191631448.html
If you can't solder wires and you don't have a bunch of these boyos around then you can go claim your retard certificate from /b/.

>>2633617
probably the worst solder job ive ever seen

>>2631036
Hello /ohm/ I'm a noob. Plz recommend a soldering iron.

>>2633891
> slobering iron
Just get a standard 27 watt cheap-ass soldering iron for $10 or $15 like everybody here had when they started. I still have my ungar for christsakes.

Started work on a behemoth today. I'm implementing the computer from NAND game electromagnetically with handmade solenoid pistons to act as the relays for the NAND gates. Ordered a couple thousand ferrite beads to construct core memory with, figured I'd have fun with embroidery while I'm doing it.
I'm building it into half of my bedroom and powering it with a water wheel in the creek next to my house. Gonna convert electric typewriters into terminals around the property and build a "screen" where it's housed out of wood pixels.
Got a lot of gold leaf, so I think I'm gonna use that on wood for my circuit board.
So far the only thing I've had to spend money on has been the ferrite beads. I guess I could also just do the memory with more solenoid relays, but I like how core memory looks.
I figure I'll use it to automate a few simple tasks around my farm, as well as have a fun way to communicate between the different parts of the property. I've got a ton of radio equipment to cannibalize, too, so I eventually want to get radio communication going on between the mainframe and the peripheral resources. It's gonna be big and noisy - it eventually might want its own shack to live in - but I was also thinking of adding different sized bells to the solenoids so that it generates music while in operation - you'd be able to not only watch the logic gates firing, but also hear them.
What if Ted Kaczynski build a computer instead of bombs? How low-tech can a computer be and still be useful? I haven't gotten to the point of doing chip fabrication yet - although I do have a ton of quartz gravel and could pretty easily build a submerged arc furnace, producing silicon wafers is a project for another day.
Fuck Intel, Fuck AMD, fuck globohomo society, fuck the elite that control the automata, and fuck the government.

>>2633973
>chip fab
I mean semiconductor fabrication, clarification. I do etch but as I said I'm using gold leaf on stained oak for the board

>>2633973
Not sure if serious, but this sounds very interesting. Do you maintain a blog somewhere? Do you keep track of what you're doing at least? So you could share the story later if you have no blog.

>>2633589
Yeah. LED, you will never be a real lightbulb.
I hope now you know which LED to use for workbench. An incandescent one.

>>2633988
I'll always be surprised when that's news to somebody. Propaganda's a hell of a drug I guess.

>>2633996
You know, after coronavirus I don't consider general public to be human anyway. Just sucks to live with them in one place.

>>2633988
>REALLY hot, breaks all the time, sometimes explodes
I don't miss them.

>>2634002
What I don't really miss are CFLs.
As for halogen/incandescent - their spectrum is unmatched by anything.
Sure, for general lighting it doesn't make sense to use an incandescent, but for a desklamp it makes a perfect sense.

>>2633997
this but for the right reasons

what the hell is "impedance trace to copper" on jlcpcb? i'm trying to calculate trace width/spacing for USB 2.0 high speed requirement of 90 ohms differential impedance. is "impedance trace to copper" even relevant here?

>>2634008
> general lighting
For many decades most lighting was done with hid lights which are very efficient. Leds are only catching up now. I’m talking about schools, warehouses, factories, big box stores, buildings, etc, and street lighting was sodium, even more efficient.

Most people look at the efficiency chart of LEDs and don’t realize that the most efficient LEDs are things you can’t get ahold of… one offs made by research facilities or NASA.

>>2634021
>sodium lamps
Now that is some horrible bullshit I won't miss.

>>2634015
also why does kicad's calculator tell me that would actually be 120 ohms differential impedance?

>>2634021
>For many decades most lighting was done with hid lights which are very efficient.
Sure, but most discharge lights are awful. Mercury vapor lamps had shitty spectrum. Sodium ones have awful spectrum.
Metal-halide ones were also quite awful.
Best one were probably xenon ones, but I think they were mainly used for filming and photography. Oh, and car headlights.
>Leds are only catching up now.
Well, they caught up in 2013 probably, because most discharge lights are no more than 120 lm/w.
> I’m talking about schools, warehouses, factories, big box stores, buildings, etc
I was talking home lighting mainly.
>and street lighting was sodium, even more efficient.
Well, what can sodium light do? 100-200 lm/w. But at what cost? Spectrum of sodium light is useless, as it is nearly monochromatic. I liked mercury vapor lights more.
So yeah, I won't miss those piss yellow lights. LED streetlights are illuminating much better, and they have much better light beam too.
>Most people look at the efficiency chart of LEDs and don’t realize that the most efficient LEDs are things you can’t get ahold of… one offs made by research facilities or NASA.
Commercial LEDs are getting there. I think atm the most efficient LED you can get is Samsung LM281B+ Pro... Which is 213 lm/w @ 65 mA I believe? Which would mean 65% efficiency.

>>2631036
I've got some old graphing calculators I'm trying to fix. The display cables are so old, they basically just flaked off, and there's still bits stuck to the PCB. What's the best way to break down the adhesive and scrape them off?

>>2634053
Gentle heat, isopropyl alcohol, and a thin, soft piece of plastic to use as a scraper.

>>2634054
Thanks m8

I want to make a switch for various headphones to various devices. It will have several inputs and several outputs and 2 switches, one to select input source and one to select output destination, I don't care if the switches are rotary or slide type, (although rotary would probably be cooler) I am just not sure what type of switches they would be called. Can anyone suggest switches, or what to search for nomenclature-wise? Also what type of wire should I use and can my choice of wire degrade sound quality?

>>2634073

Rotary Switch 2-Pole 6 Positions
you need 1 pole for mono, 2 for stereo, 4 for quadraphonic headphones (2 speakers on the side, one each in front and back)

>>2634065
No problem, Anon.

>>2634081
so if I get (2x) 2 pole 8 positions I will be able to do 4x stereo in to 4x stereo out in any combination?

>>2634088
oh wait i confused poles with positions

>>2634088
I know this is diy but mixers already exist.

>>2634088
>do 4x stereo in to 4x stereo out in any combination?

you'd need 2 switches of 2 poles, 4 positions.

>>2634090
But a mixer will only have one or two outputs right? I have a yamaha mixer and it has one output. I want to keep 4 different headphones plugged in and switch between 3 or 4 different devices

>>2631036
How do I precalculate diodes into my circuit? Their resistance is not constant, I don't know what the voltage drop will be? Do I just refer to forward voltage on the datasheet and assume that much voltage is gone after the diode?

Btw how do I correctly find the forward voltage without the datasheet? I bought a kit of random parts and there was no reference to the diodes model. It's variety of colors in 3mm package

>>2634120
You don't make your circuit rely on diode voltage drop. With zener diode being an only exception.
When necessary, consider range provided by datasheet by looking at volt-amp curve or in case of LED it may be listed as Vf min max.
>Btw how do I correctly find the forward voltage without the datasheet?
Variable load (e.g. potenciometer), voltage source, voltmeter, ampmeter.
Or just assume ballpart values.
For ex. silicone diode drops 0.7V, germanium diode drops 0.3V, red LED drops 2V, and blue LED drops 3.2V. Schottky diode drops 0.1-0.5V.


Here is how you design a simple capacitive dropper LED driver.
In case you dont understand anything.... No problem, start with simple circuit consisting of LED, resistor and 9V battery. Swap LED colors and note the current differences, and voltage dropped on LED.

>>2634139
Oh, other thing... Did you notice that I didnt consider the full bridge rectifier voltage drop? Because 0.7V is <<<< than whatever LED voltage.

>>2634120
diode equation

>>2634101
A theoretical mixer would only have one output.
In practice, they have multiple outputs such as “monitor” which is a headphone jack in addition to the “line out” which goes to your multi-track or whatever.
Obviously you could split it with a high-impedance headphone amp or something. These things are modular that way.

I have a single solar cell that I am looking the datasheet of. It's an ANYSOLAR sm141k05l. Picrel is the I-V curve of the cell.
My question is why do the chart's only go to 0.7V instead of the Voc of the panels?
I want to connect it's input to an SPV1040 MPPT controller and I want to see the voltage and current capabilities of the solar cell at 2 volts because that's when MPTT turns on. Can I somehow do the calculations?

>>2634254
Oh wait I just though. Is this the characteristics of a single cell? If the i just multiply by the cells that exist in the package

There was a thunderstorm and a loud lightning bolt and then a very loud pop. Now all the tvs won’t turn on and the modem/router won’t turn on so I have no internet. Everything else works though. What happen and What do I do

>>2631036
NOT gate. Why 2 works but 1 doesn't?

>>2634120
There are rules of thumb. Mainly these are:
>silicon diode - 0.7V drop
>schottky diode - 0.2V drop but higher at high current
>germanium diode - 0.3V drop
>IR LED - 1V drop
>red LED - 2V drop
>orange or yellow LED - 2.2V drop
>green LED - 2.5 drop
>blue LED - 3V drop
>UV LED - 3.5V drop
These are all approximate, since they're a function of manufacturing method, temperature, and current flow. But they don't change THAT much. If it's critical (e.g. driving a bunch of series LEDs) assume +20% to the voltage when calculating how much voltage you need to turn X LEDs on (or how many LEDs you can run on Y voltage), and assume -20% when calculating how highly driven they'll be. Chuck them in series with a resistor to measure them at the kind of current they'll be in.

>>2634254
>>2634256
Yes it's just a single cell.

>>2634264
Something tripped, be it a breaker or RCD (or fuse) in your breaker panel, or a breaker or other safety device somewhere else, like in a multi-board. That's what the little black button on the side of some of those multi-boards is.

>>2634265
Because 1 is trying to shunt the power rail to ground, but only has 5-30mA of drive to do so, and your 5V rail can source a lot more than that. So the right-most note remains at 5V, because that's what the power supply is set to.

>>2634270
>Because 1 is trying to shunt the power rail to ground, but only has 5-30mA of drive to do so
I'm not sure I understand. Shouldn't even less than 5mA be enough to drive 50x times more via Collector to the Emitter? I tried changing the resistor on the Base down to 100 ohms and Circuit 1 still wouldn't work as a NOT gate. It only started to once I removed resistor on the base, but that's obviously a crazy thing to do.

>>2634284
Okay, my bad, I just had a switch malfunction. Lower resistor actually fixes the problem

Discharging

I've got a TV/VCR combo where I need to repair the VCR portion. Looks to be made in 1992~ish.
Theres a repair video on youtube - done by "12voltvids" - but he doesnt seem to discharge any of the TVs he works on. BUT I know working on the old 9inch macs you ALWAYS had to discharge them.

>>2631407
Mentions to always discharge. Do these CRTs discharge after being off for X days, or do I need to do the flathead + lead discharge

>>2634482
>he doesnt seem to discharge any of the TVs he works on

that's coz he's not a moron who falls for stupid memes, like the ever-popular discharge meme.
one stupendously stupid variation of this meme is people who suggest you short the plug on your computer before opening it.

>>2634284
>Shouldn't even less than 5mA be enough to drive 50x times more via Collector to the Emitter?
No, 5mA IS the collector current. The base current is (5V - 0.7V) / 47000Ω ≈ 100µA.

>>2632842
>esd protection
pussy
if you're that worried about it put an esd mat on it.

Could anyone draw this out for me? I don't get why v1 is positive

>>2634617
Show your incorrect reasoning so you can be taught to fish, at least.

Holy fuck, why are analog joysticks so expensive? Everywhere I look a standard 5 or 10k, 2 axis analog joystick is like $150. I used to be able to pick up CH joysticks for like $30 a pop. Seems they only make computer flight sticks and such now. So, I either buy used or get some Chinese shit that will wear out in a few thousand cycles. What happened?

>>2631301
>Dat photo
Back in my day we could see the pixels from where we sat on the couch across the room.

>>2634619
I figure 1V source is too weak against 2V, so 2V does a ccw loop from + to - along the Amp direction. But considering I have to measure it in opposite direction, wouldn't it be voltage rise? Or it's considered a drop, because it's traveling towards the -?

And if that's the case, why would 1V source going agasint v4 measurement be negative? Does that mean 2V splits also to the left and blocks 1V at the mere terminal? Why would it split to 1 and 2 volts and how does that bottom amp direction sign go along with all of that?

My reasoning: I have no fucking reasoning, I don't understand what's happening in the picture and have more questions than assumptions. I thought someone could just spend a minute on drawing the flow and simple calculations and post it so I could understand the logic behind it, instead of asking me to explain myself and then deflect whatever I write with a short dubious mental-shortcut tier answer I won't understand

>>2634626
>What happened?
Retro gamers. Your best bet is to get the chinesium ones and replace the shitty pots. Are you trying to build an arcade cabinet, or a 70s/80s hobbycomputer game?

>>2634617
Looks like the notation is "final subtract initial" with respect to the arrow, or "head subtract tail". That explains both V1 and V4, since V4's tail is 1V greater than V4's head, and V1's tail is 2V lesser than V1's head. If you have a hard time getting that, then ignore the currents and just look at voltages at each node. Don't have to think about directions at all that way, just set one node as 0V (the top one is best in this case) and figure out the relative voltages as a result of the voltage sources.
Personally I'm not too fond of the arrow notation used here, they'd be better off drawing voltmeters with explicitly positive and negative terminals.

Never seen a battery drawn with a line between its cells before, kinda interesting. Odd that they used current sources instead of resistors though, the current sources might as well be open-circuit for all the defining of circuit parameters they do.

Trying to understand this logic board to a jump box disconnected everything without taking a photo like a scrub need help

front of logic board

>>2634677
A few years back I made a custom Vectrex controller using a JH-D202X-R4 joystick. Picture related. It is a smaller ~4 inch joystick. They can be had for anywhere between $5 and $35, depending on how much of a sucker you are, but they are all the same. Sometimes they change the name of them, like Adafruit did with their "Mini Analog Joystick - 10K Potentiometers". They are junk. Every few months, I have to open the case up and adjust the pots on the stick. Also don't buy from Adafruit, they mark up their shit like crazy.

I've gotten back into Apple II games and am trying to put together a joystick. Apple II sticks are fucking expensive as hell even used. You can get adaptors for IBM sticks but even those cost $40 unless you get some Chinese trash. I've found a few sites that used to sell DIY kits for adaptors but they are all discontinued or have been sold out forever. All told, to get a non-flightstick IBM joystick plus an adaptor I am looking at spending around $100. Fuck son, I can put everything but the joystick together for less than $30 and that is using Sanwa buttons!

When I was a kid the hot stick was the CH Products MACH series and they used a standard analog stick right out of their industrial products catalog. I'd used them in other projects back in the day and they used to be reasonably priced. Not anymore. Jesus.

>>2634692
>6 identical 2-pin jst connectors
is that the main problem? what are on the other end of the connectors?

I'm only interested in the 2 pin jst's, Ive got a regular switch, a solenoid , two probes to the connections the solenoid makes, two probes to the negative and positive ends of the internet battery( I believe it uses this to power the board power, power from an ac to dc box ( i believe it uses this to charge the internal battery's) j3 or the top right is a known connection

i want to change all the switches on my "Logitech G203 Lightsync 2nd Gen", but the middle mouse button is different from the rest. is there any site where i can get a parts list, so i know what to look for?

also where to buy them from europe, the shipping from us/china is like 10 times the cost of the switches.

also mice recommendations, if you know a mouse that isnt total garbage for less than 50 euros.

>>2634718
>regular switch
Regular switch would just have a resistor tied to Vcc or 0V on the one pin connected to the MCU, with the other pin tied to 0V or Vcc respectively. Where Vcc is the voltage rail used directly by the MCU. If the other pin is tied to 0V, they may be using an internal pull-up resistor on the MCU, and so no external resistor would be present.
>solenoid
Solenoid should have a diode in parallel with the pins, likely with one pin tied to some sort of positive voltage rail via a transistor. It may also have thicker traces than the switch connectors.
>two probes to the connections the solenoid makes
Do you mean like limit switches, or a relay? Assuming it's a relay, I guess it depends entirely on what the relay is switching. It will likely have thicker traces, and I wouldn't be surprised if one of the leads of each connector were connected to 0V or the positive power rail. J4 looks somewhat promising.
>two probes to the negative and positive ends of the internet battery
As in, two wires going to each one? If they're where the power for the circuit comes from then it should be simple to follow the power wires, maybe via a linear regulator or BMS IC.

Also it looks like at least three of those JSTs are labelled:
>power
>output input - J2
>sp

>>2634722
It should say on the side of the switch what its brand and part number are.

>a mouse that isnt total garbage for less than 50 euros
The Logitech Pro Gaming Mouse, I think I bought it for below 70 euros when on special but I'm not too good at currency conversion. That was like 4 years ago though, inflation's a bitch. Got a good long lifespan on it switches, apparently. But the investment of a soldering iron and a handful of good Omron switches will pay itself back way better than getting another mouse. Maybe a vacuum desoldering iron (or desoldering needles?) if you don't want to wreck the board.

>>2634738
>the investment of a soldering iron and a handful of good Omron switches will pay itself back way better than getting another mouse.
yeah, if i buy another mouse chances are ill get another lemon and its back to square one.
its grim that to get a functional mouse you have to diy.
another sign of late stage capitalism i guess.

>>2634741
>another sign of late stage capitalism i guess.
More like communism, Anon. Nobody has been operating as a capitalist economy since the 50s. The few at the top reap the harvest while the masses eat shit. That's exactly what communism is in practice.

>>2634745
>The few at the top reap the harvest while the masses eat shit. That's exactly what communism is in practice.
i hope you are just trolling bud XD

>>2634749
I hope you're just trolling, bud. Take a look around. Your great-grandpa was able to comfortably retire from a manual labor job not too long ago because he was a shareholder in the company. Now the major shareholders are Blackrock and Vanguard who extract the wealth and use it against regular people.

>>2634751
>Blackrock and Vanguard
who wouldn't be able to operate without commie policies being forced onto an unwilling population.

>>2634749
When corporations use the power of government to cement their monopoly positions, is it really capitalism? Things like patents and copyright alone cause corporate fraud to go unnoticed because they spin it as if it's there to help you. It's not. Patents on cancer medicine aren't going to help your dying loved ones when they can't afford a dose. And all those record label lawyers are using parts of chord progressions to tie up other labels in court are using your tax money.

>>2634745
In the end it's all just tribalism. If you have connections and belong to the right group your *ism good other *ism bad. Me beat other *ists with a club. Ooga booga.

>>2634767
>If you have connections and belong to the right group
Cool it with the antisemitism, Anon.

>>2634745
The idea of things like communism or capitalism is the economic model, not the "practice" (in "practice", capitalism IS about the rich getting richer at the expense of the poors by definition, too, but you wouldn't say the "definition" of capitalism is that, so why do you have a seizure when it comes to communism?). In addition, it's always funny to me that you amerilards make this dichotomy as if there exists no other system in the world.
Globalized production is 0% communistic and 100% capitalistic. It is a pure cost cutting method, which is the truest sense of capitalist endeavor (as opposed to, for example, producing fewer but better items, which is mercantilistic instead, or producing exactly as much as required by the ingroup, which is what communism aims to do).

>>2634753
>when a capitalistic entity pays another capitalistic entities to provide a service, is it capitalism?
Yes, dumbfuck, yes it is.

>>2634777
Capitalism is an economic model that allows for peasants to advance to middle or upper-class status, while taking into account the negative (predatory) aspects of human nature.

Communism is a power grab that impoverishes everyone while promising the world. We've seen this play out many times, but this time it'll be different. Right?

lmao

>>2634778
>theft, bribery, and extortion is capitalism

>>2634741
>its grim that to get a functional mouse you have to diy.

or just visit a thrift store.
here are 5 of the 20 or so i bought there.
two still have price stickers, $1.50 and $2.50.

>>2634783
>I don't like the merchant I'm dealing with so it's theft!
The absolute state.
And yes, theft, bribery and extortion is quite literally the very backbone of crapitalism. After all, it is literally defined as capital = power.

>>2634784
He said "functional", illiterate faggot.

>>2634782
>Capitalism is an economic model that allows for peasants to advance to middle or upper-class status
3/8 b8
>while taking into account the negative (predatory) aspects of human nature.
0/8 b8

>>2634785
>>2634788
>a true believer
I will never stop lmao at "people" like you.

>>2634789
>the only thing that exists beside crapitalism is communism
>anyone who's not braindead must therefore be a commie
The absolute state of crapitalist amerilards

>>2634786
>He said "functional"

never got a single dead or defective mouse.

>>2634790
>national socialism
Can't do that in a multicultural pot of turd-worlders.
>muh Norway
Yeah, the happiest people on Earth. heh

>>2634795
>communism is the only thing that exists beside crapitalism
I lol every time.

>>2634796
You can't even read. lmao

How much time would it take for a professional to handplace 0603 0402 components and couple of MCUs on a 2x2" board with soldering mask on it? Obviously your estimation.

>>2634804
Depends on your setup. Hot air? Infrared? Solder paste and stencil, or iron and roll of solder?

Best case - 5-10 minutes if your workbench is set up and ready to go beforehand.

>>2634804
>How much time

we had a guy at work who did this all day long using a soldering iron.
took about 3 mins per 80-pin chip, and 20 secs per SMD component.

is there a way to implement this on both upper and lower FETs in a half bridge without using an isolated DC/DC converter? i'm using a bootstrapped high side MOSFET driver (IR2184.)

https://sci-hub.se/10.1109/EPEPEMC.2010.5606838

>>2631036
Good evening /ohm/ i got this old crt (a philips 21pv275) that i'd like to mod with stereo audio.
At the moment there is only one speaker but there is an unmounted connector on the pcb for second one and also on the chassis there is screwholes for an additional speaker.
However i dont know much about audio or electronics in general so im not sure if it is as simple as soldering a second pair of speakers to that connector or if something else would have to be done. If somebody here could point me in the right direction for attempting something like this that would be appreciated.
See pic related for schematics, the purple traces/speaker is the electrical connection for the unmounted connector

>>2634901
IC151 (LA4224) says it's a monaural amplifier, and you can see both speakers are wired to the same input on pin #1. if you're trying to get true stereo audio (left and right channels rather than two speakers outputting monaural) then you would need a in-depth rework of the internal circuitry, considering that amplifier is fed from a monaural source (pin #50 on the large IC at the bottom center.)

>>2634918
I see, that's what i feared. I was able to find a schematic for the big ic (Mitsubishi M61029) or at least something very similar (M61203) and it seems like there is only a single audio output pin so i guess true stereo audio is out of the question here. Oh well.

>>2633694
Generally speaking, phones won't draw the full charging current they're capable of using unless the power source signals (either passively, with resistors across certain pins, or actively, with a protocol like USB Power Delivery) that it can safely provide the higher current.

This page lists all the common ways of doing such signaling:
https://lygte-info.dk/info/USBinfo%20UK.html

>>2633735
>>2633764
For USB Type-C, with the cut cable that you described, I believe you need 10kΩ resistors from the CC wires to VBUS (+5V), to tell your phone it can draw 3A. This is in place of the 56kΩ resistor in the attached picture, which is what should be used in a Type-A to Type-C cable that can be plugged into arbitrary ports that might not support 3A.

>>2634782
>falling for poorly-made fake screenshots
ngmi

You can see the real article this was made from here:
https://nypost.com/2019/02/27/man-delivering-food-allegedly-dips-testicles-in-salsa-over-low-tip/

Is there a simple, fast and easy way to find compatible resistor values if you know the ratio of say R1/R2?

>>2635002
>fake screenshots
oh no
not fake screenshots
noooooooooooooo
AAAAAAAAAAAAAAA

>>2635005
in the context of resistor dividers? i use this. https://www.ti.com/download/kbase/volt/volt_div3.htm

>>2635014
Never post fake screenshots, Anons. It violates the communisms lmao

>>2635015
that's exactly what I was looking for, thanks!

>>2635014
>>2635018
Dial 8, don't die l8.

>>2635034

>>2634809
>3mins for 80pin chip
yikes my brother, hope you fired that useless cunt

I know some basic stuff about electronics from working on cars, but I want to learn how to build a modular synth, keyboard, and sequencer from scratch.
What book do I need to read, and are there any sites that have free schematics for various synth modules that I can try making?

>>2631285
6x4 is a good high-ish voltage rectifier. 6AQ5 is a good high gain amplifier tube.

>>2635151
Thank you for the info. I have been too lazy/busy with other projects to look into most of the tubes in that picture.

>>2631036
All found materials we had around the storage floors

Fellas what am I getting wrong here? Is the generator fucked or I'm missing something?

>>2635173
okay nevermind. I added Req to R1 instead of R1+R2. I sat down on this for an hour and noticed it just now

In VFD systems when they show power coming in on the network bridge and a motor being driven on the machine bridge side.
If the VFD is being used to connect a generator to the power grid, does the generator count as being on the machine bridge or the network bridge?

>>2635213
>If the VFD is being used to connect a generator to the power grid
does it list this as a feature in the user manually? generally they can't do that. they don't generate sinusoidal output. to anything that isn't a motor (which naturally filters the waveform) it will look like absolute shit.

>>2635162
Sick.

>>2635239
That model is used in many Siemens wind turbines but the data sheet only shows example circuits where they're used to drive motors.

>>2635289
my guess would be the wind turbine output would go into the left side. it looks like network bridge is an active rectifier. i doubt they would bother to make the machine bridge (six transistor bridge output) active rectifier capable as well.

>>2635239
> naturally filters
I don’t know about that, my vfd’s make the motors emit this truly maddening high-pitched whine.

>>2633980
Very serious. I've been keeping noted all via typewriter, but I do have server space via sdf to do a blog/phlog (good idea, it's my day off).
Been wrapping solenoids all day, decided to do the chimes as a separate layer that can be cranked into and out of place to the solenoid layer. The modularity of this project has been making prototyping very easy, since each NAND gate is essentially just two solenoids on wood with conductive paint.
Gonna start on some core memory later. I'd like to convert my extra piano to a player using core memory instead of paper rolls.

>>2635534
>I do have server space via sdf to do a blog/phlog (good idea, it's my day off).
Yes please. That sounds fucking awesome (and awesomely autistic, which is always a plus).

>>2635162
>>2635271
I finished this much simpler piece to replace another light. Used scrap MX wire that was waaaay too brittle and thick for 1/2 iron pipe. I tried to be careful but I guess I nicked the wires while pullin them through because this thing just lit my ass up.

>>2633140
>Computer

I'm not a fucking nerd faggot

>>2635556
>I'm not a fucking nerd faggot

then you wont make any money.
i can walk half a mile down the street and find 3 computer sale/repair shops.
i wouldnt even know off-hand where to find one single electronics repair shop.
first rule of economics.: if there's no supply, it means there's no demand.
you might try blacksmithing instead, it may have better prospects.

>>2635576
>if there's no supply, it means there's no demand.
That's the opposite of any principle of economics. What bizarro world are you even from?

>>2633973
Have you considered using a flip disc array for display output? Seems like the best way to have an electromechanical display that matches the theme of your build.

>>2635577
>>if there's no supply, it means there's no demand.
>That's the opposite of any principle of economics

blockbuster says you're wrong.

>>2635607
b8/8 or clinically retarded?

>>2635607
Act now. Supplies are limited.

>>2635607
i dreamt i was in a blockbuster again. they had reopened some locations temporarily for some promotion

>>2635614

when my local store closed, they threw out 1000's of tapes in the dumpster.
like a dummy i filled up a backpack full.
never watched a single one, but kept some of the packages with cool artwork.
still have 'em.

>>2635607
You were implying that a low supply causes a low demand with your poor wording. The other anon responded by stating that your causality was backwards. Be clearer next time.

>>2635631
I'm the other anon and I still don't understand what the fuck he's trying to say.

>>2635632
Midwit description of “there is some sort of positive correlation between supply and demand”, to which “the opposite principle” implies its negative instead. How to make an argument out of thin air.

>>2635639
If that's what he was going for, then he's even more retarded than I gave him credit for.

What is this?

>>2635666
Spare solder. Break in case of emergency.

>>2635639
>How to make an argument out of thin air.

geez, such pedantry.
if lack of demand leads to lack of supply, then logically a lack of supply might easily INDICATE a lack of demand.
among other things.
which is a valid point in its original context.
if you dont see a lot of X for sale, when it could easily be made available, it's probably because nobody wants X.
i.e: there is no supply coz there's no demand for it.

>>2635666
It's a fuse.

>>2635676
Take your meds, schizo

>>2635676
>it's probably because nobody wants X.
No, because supply shocks exist. There was a real demand for STM32s 2 years ago even though there was fuck-all supply. And there’s a non-zero demand for 4-terminal MOSFETs but the fabs just round it to zero. There’s demand for MacBook charging chips but the exclusivity agreements prevent Intersil from selling them to anyone other than Apple. There’s demand for the internal parts of spring-loaded measuring tapes so wood-turners can make custom tape measures, but they’re prevented by the shadowy cabal that is “big foot”. There’s demand for free energy and honest politicians but those don’t even fucking exist. If this were consumerist /g/ then I could accept your generalisation, but this is /diy/, where we’re constantly making shit because it isn’t made, and even selling it to fill an ignored niche. The exception is more important than the rule.

>>2635676
what the fuck are you talking about, how did you graduate high school?

Is there a brand name for this 14-pin ribbon cable connector? how do you find it in KiCad? I know that you use the pin pitch to characterize these but each vendor probably makes a slightly different one.

>>2635773
looks like a variant of jst-gh 1.25mm or 1.27mm. molex brand as a few of these too. as for kicad, it doesnt have many specialized connectors in its library. same goes for other pcb editors. find the part, make the footprint yourself

holy shit these fucking connector wizards man

also they're probably on grabcad or something. octopart usually has links to files.

Is it possible to have an electric motor with lots of torque but slow RPM (no gearbox obviously)? I've heard about FOC but don't understand it fully, and I've seen stepper motors, but not sure if they have proper slow speed torque.

>>2635791
If your number of pole pairs and slot pairs have a really high lowest common multiple, you naturally end up with a low speed, I think. Go through a full electrical cycle on paper and see what falls out, but when I tried doing that to one of my BLDCs I found its actual rotation frequency to be even lower than my calculation. There's probably documentation out there on the relationship between slot/pole numbers and speed or Kv.
An extreme form of this can be seen in a stepper motor, though it's 2-phase not 3-phase so the execution is slightly different. Stepper motors themselves are an option, and their benefit is they generally don't need feedback, but their power density and specific power fall behind BLDCs.

>>2635793
> but their power density and specific power fall behind BLDCs.
NTA, but do steppers have better torque density and specific torque, compared to BLDCs?

>>2635791
>but not sure if they have proper slow speed torque.
AFAIK they only have low speed and holding torque, since it drops off at higher RPMs. That might not be the case with different methods of driving them, like FOC.
>>2635791
>I've heard about FOC but don't understand it fully
You use a sensor to measure the angle of the rotor, then you supply the ideal amount of current to get the movement you want. Minimal losses, great amount of control and accuracy, smoother motion and torque than normal stepper drivers. It's basically a servo motor.
Since it sounds like you want an FOC stepper, I'd recommend looking into either the Arduino Simple FOC project (https://docs.simplefoc.com) or the TMC4671 chip.

By the way, why do you not want to use a gearbox? Would probably be simpler to implement.

>>2635778
I think I will use an adapter board instead, these ribbon cables look flimsy.

I'm interested in building a small 100-200Wh power bank using lifepo4 batteries. It would only be used for a couple of USB-A and one or two 60W USB-C PD ports.
Could I use one or two of the smaller lifepo4 prismatic cells? I've seen some pretty compact 3.2V 105Ah cells. I'm interested in lifepo4 cells specifically because they have a huge lifespan and can be charged/discharged very quickly and generally seem safer.
I've never really soldered anything before so I'm a bit wary of all this and how to source BMS and whatnot, I know chink-ebay has an abundance of boards but wouldn't know how to weed out the rubbish/dangerous ones.

>>2635791
what are your torque and RPM requirements?

>>2633988
imagine being this much of a picky bitch about lighting

>>2635890
You know, when resistors turn from
>orange, red, violet, black, blue
to
>red? red? black? black? black?
this is really bad. And this is all LED lights so far. Except these.

>>2633988
>>2635900
What is the "correct" spectrum? Even sun light changes its spectrum during the day due to the atmosphere.

>>2635814
85 Nm from 0 to 100 RPM

>>2635799
Thank you. If FOC is so good why nobody uses it except camera gimbal motors?
> By the way, why do you not want to use a gearbox? Would probably be simpler to implement.
Weight and size constraints, but honestly it's just what I want. It's all hobby and diy anyways.

>>2635793
I assume steppers won't work nicely as a main motor for a vehicle?

>>2635914
Sun would be the correct spectrum, but black body radiator spectrum is a good approximation for it.
Which means it should be as continuous as possible in visible range without random dips.
Halogen/Incandescent bulb fit perfectly there, but they are increasingly harder to come by, and at some point you won't be able to order one even from the aliexpress. Not to mention that I don't want to pay for IR that I don't need.
Other thing is that CRI is kinda flawed, as it takes in the account only 15 colors.
TLCI is better as it uses 24 colors, and TM30-15 is even better... But nobody seems to use it. Which is the problem, because even CRI 95 conventional LED doesn't produce quite enough of violet light and produces too much of blue, which more than definitely fuck up a bit with how you actually see things. Not as much as in CRI 80 bulb, but still I'm pretty sure that you can see it if you look closely.

Most interesting part is that I didn't notice that much difference in red colors (they just got more saturated), but I've noticed the difference in blue/violet colors. I have an african grey parrot, and he has blue-ish violet "black eye" around his eyes, and with normal LED they just disappear. With incandescent, sun light or chink full spectrum LEDs it is there. Also ends of his grey feathers are a little bit violet in the sunlight, same with incandescent and full spectrum LED.
So dunno. Fuck eco greenfag faggots. If they didnt ban incandescent, I wouldn't be dicking around with special snowflake LEDs.
>>2635931
>Thank you. If FOC is so good why nobody uses it except camera gimbal motors?
Ninebot e-scooters use FOC for PMSM motors. Very common on e-bikes, e-scooters, etc.. Not so common on mains motors or steppers for some reason.
>>2635791
BLDC or PMSM with low KV. Maybe steppers.
>>2635931
>85 Nm from 0 to 100 RPM
That is too much. You're better off with gearbox of sort.

>>2635939
>That is too much. You're better off with gearbox of sort.
>>2635931
Maybe e-bike motors would work. 1 kW maybe. But yeah, you want sensored motor, because BEMF detection sucks at really low RPMs, which is a problem. And then you can use simple BLDC commutation controller (torque ripple and noise, but is cheaper) and it would go to low RPMs, or FOC (less torque ripple, less noise, but controller might get costly and tuning motor for FOC controller might be challenging).

>>2635939
>>2635957
85 Nm is what I heard Bosch uses in their electric bike drives, that's why I asked. I wanted to do something like that, a middle drive inside a bicycle frame, but obviously diy way and without gearbox, like a single gear bike, maybe with belt instead of chain.

So if I buy or design FOC board based on SimpleFOC and stick it to PMSM or BLDC motor I should be fine?

>>2635969
>85 Nm is what I heard Bosch uses in their electric bike drives,
Aren't they geared? But yeah, 85 Nm 100 RPM seems like a good 1kW or 1.5 kW direct drive hub motor to me.
>So if I buy or design FOC board based on SimpleFOC and stick it to PMSM or BLDC motor I should be fine?
Well. BLDC - you can buy any generic $15 (well, for 1kW probably not $15) motor controller and it would work fine. For FOC, it is hard. Designing FOC is hard. Check uh. https://vesc-project.com/
I think you can buy controllers that are compatible with VESC on alie, but they are like $100 or more dollah.

I suggest you to make the mechanical part first, with generic cheap ass BLDC controller, and only then think if you need FOC.

>>2635791
Also note that torque is proportional to motor radius, so try to go for a pancake-shape rather than a spindle-shape. Maybe a 6215?
Other motor types, be they induction or reluctance or something else, may have better torque or power density than one using permanent magnets, but permanent magnets are likely the optimal method when considering weight. Except for synchronous motors that use a stator and rotor coil, and use a transformer across the rotation gap instead of slip-rings. I'm guessing you don't want slip-rings. Using no permanent magnets means you can reach higher temperatures as there's no curie point to worry about.

If you want a gearbox but sexy, consider a planetary drive. Or even a harmonic drive.

>>2635799
>NTA, but do steppers have better torque density and specific torque, compared to BLDCs?
Not as I understand.
>TMC4671
Holy shit this IC looks great, though I'll have to read deeper to see if it's fast enough for my uses. Their 3-phase gate drivers look decent too, not that I wouldn't couple them to higher current gate drivers like I was already planning on doing to the DRV830x.

>>2635810
Sounds like a neat idea, the portable power centres I see are always either 300+Wh and focus on having an inverter, or are under 100Wh and use lithium ion batteries for maximum compactness. But there's definitely a market for something in between. Using prismatic cells instead of a bunch of 18650s sounds like a good idea to me. Consider how many series cells you want though, since 60W at 3.2V is almost 20A. I'd probably want to go for 3-4S, or even higher, to increase the efficiency of the conversion, but ultimately you're probably restricted by the form-factor of your cells. More than 1S means you need a more complicated BMS, one that can handle balance charging.
I saw some Daly branded chinky battery management stuff that looked decent, well documented too.

>>2635900
>>orange, red, violet, black, blue
>to
>>red? red? black? black? black?
works on my machine. get your eyes checked.

>>2636075
Doesnt work on my eyes and aliexpress resistors of shittiest quality. Sunlight works on my eyes though.

>>2635801
ribbon and fpc are different things. fpc connectors are usually for that flexible pcb shit with gold fingers/conductive strips on one end. and those things are not meant to be stressed and strained. think of them as one time use, plug in, lock, and forget. ribbon cables usually are much stronger and have actual crimped connectors. if you wanna see the flimsiest shit, open up the nintendo switch controller and look at the connector and cable for the joystick

>>2636054
>Not as I understand.
Are there BLDC motors with high torque available at a reasonable price? If not, how difficult would it be to fully DIY a motor like that? How can I maximize torque?
My requirements are: at least 3-5 Nm, at under 100 rpm, the least cogging possible, and I cannot use a gearbox.
I was thinking something along the lines of a 75-100 mm diameter axial flux motor, with 30-50 pole pairs, but I don't know how to decide the number of magnets to use, or anything about the coils I would have to wind. If anyone can direct me to a guide and/or calculator for these things I'd much appreciate it.

>>2636319
>Are there BLDC motors with high torque available at a reasonable price?
Yeah.
>If not, how difficult would it be to fully DIY a motor like that? How can I maximize torque?
DIY a motor is hard, unless you know where to buy iron.
>My requirements are: at least 3-5 Nm, at under 100 rpm, the least cogging possible, and I cannot use a gearbox.
Just buy Bafang or Bosch or any hub motor

Did I get the schematic correctly based on the connections above? This is Jlink Arm debugger probe and Vref is the voltage on the target attached to the probe that Jlink uses as a reference VEE for level shifting and as input to ADC to sense Vtarget. But why is it buffered?? Also why do they use a quad opamp if only two inputs are used?? And what's the point of connecting two buffers in parallel? Is it some kind of a short circuit protection, essentially protecting the attached target if VEE is somehow shorted to ground inside the probe?

>>2636319
>Are there BLDC motors with high torque available at a reasonable price
Yes, I'd recommend hub motors or agricultural drone motors, radial flux outrunner motors with a large as possible diameter. The output torque will basically be proportional to the amp turns you can force through it without overheating the magnets or windings, and the radius, and the magnetic field strength.

>how difficult would it be to fully DIY a motor like that?
A diy motor could just mean finding an existing stator and building a rotor around it. Which is doable, see:
https://www.instructables.com/id/Make-Your-Own-Miniature-Electric-Hub-Motor/
This guy rewinds his stator for a different Kv, which you may want to do if the only motor you can find isn't rated for a convenient voltage, but otherwise you should be able to just select a stator and set your battery voltage accordingly. There's also this guy:
https://youtu.be/EZELVYpDbVk
Who uses a halbach array to increase the effective magnetic field strength, another reason to make a custom rotor.

But if you can't find a stator of the right size, you may have to make it. This has the problem that it needs silicon steel laminations, which are kinda hard to get a hold of. That said, if you can get them laser cut or water-jet cut you'd have a decent chance of making a custom stator. This would give you the freedom to add water cooling channels inside the laminations. You may also be able to make a stator out of cast ferrite epoxy, but I'm unsure if it would be strong enough, you may need to reinforce it with fibres of aramid or glass in key directions.
Magnetics simulation software recommended.

>>2636600
LM324s are rated for continuous output shorting, so I guess it's protection. It definitely needs buffering though, since I imagine the reference voltage comes out of a DAC and can't drive the hundreds of mA that the buffers can. No clue why they used a quad instead of a double though.

>>2636600
>why do they use a quad opamp

everyone i know already owns a dozen LM324s

>what's the point of connecting two buffers in parallel?

doubles current capacity.

>>2636603
>the reference voltage comes out of a DAC
I wasn't clear, it is not really a true voltage reference. It is just VCC of the target MCU connected to the probe so the probe could detect if the target has power and check its voltage and also do level shifting if the target VCC is different than the probe VCC. A lot of times the probe itself can power the target in which case it is essentially reading its own voltage.

>>2636603
>>But if you can't find a stator of the right size, you may have to make it. This has the problem that it needs silicon steel laminations, which are kinda hard to get a hold of. That said, if you can get them laser cut or water-jet cut you'd have a decent chance of making a custom stator. This would give you the freedom to add water cooling channels inside the laminations. You may also be able to make a stator out of cast ferrite epoxy, but I'm unsure if it would be strong enough, you may need to reinforce it with fibres of aramid or glass in key directions.
This is why I mentioned axial flux, from what I've seen I only need a rod of the appropriate material, which I can cut to length and wind the coils around. Correct me if I'm wrong.
>Magnetics simulation software recommended.
Is there a specific one you would recommend? This is what I found with a quick search: https://github.com/NanoComp/meep
>Yes, I'd recommend hub motors or agricultural drone motors
>>2636428
>Just buy Bafang or Bosch or any hub motor
These seem both expensive and geared more towards power rather than torque, so they end up being too heavy for what I need.

Am I the only one who thinks using Wh instead of Ah in description of a battery capacity is marketing bullshit, which aims to obfuscate the real specifications?

what's the cheapest/easiest way to generate 5V output, 0-50 KHz, 0-95% duty cycle PWM with "good" resolution?

i tried dicking around with jellybean op amps in ltspice to do it the traditional way (op amp square wave oscillator --> integrating op amp to yield a triangle wave --> comparator to yield pwm) but e.g. lt1014 (more modern lm358) was too shitty to get anywhere near 50 KHz with clean edges.

i'm kind of a noob here so i'm not sure if it's just a matter of finding faster slew rate? or if there's an easier approach that gets me at least like 12 bits of resolution (or analog) on the full frequency and duty cycle range.

>>2636663
The correct unit to measure battery capacity is the J. Anything else is objectively wrong.

>>2636667
Joule is shit unit as well, because for that you need to know the voltage just like using Wh. While Ah is universal across all baterries and is a perfect measure of capacity.

>>2636672
I refuse to believe you are this retarded.
But in case you are not trolling, no. With Ah you need to know the voltage, because two batteries with the same Ah rating and different voltage can have vastly different capacities:
If you want to power a device that consumes 1A at 5v (5W), a 1Ah 5v battery will last much less than a 1Ah 20v one, assuming you step down the voltage and don't burn the device.
With J and Wh, you get the pure capacity, regardless of voltage: the first battery would be 5Wh, the second one 20Wh. So you immediately know that the second battery would last 4 times as long. Of course, the Wh is a disgusting unit that shouldn't be used, so multiply by 3600 to get J instead. First battery is 18MJ, second one 72MJ.

>>2636683
Correction: 18KJ and 72KJ. Should've picked more realistic numbers.

>>2636667
I prefer 9.48×10^-4 BTU

>>2636684
This is why we need to measure batteries in BTU. No confusion between KJ vs MJ, it would simply be ~18BTU. Brilliant.

>>2636683
Nah, that's bullshit, does not work and gives precisely ZERO useful information especially when voltage changes from 0 to a lot like it always does with DC motors.

>>2636663
Wh is less misleading. Where I work we sell a portable power station that's meant to be a combination of a house/utility battery and an inverter. But even though it charges off 12V and has a 12V output, the mAh rating on the side is actually for its 3.7V battery, so you have to divide by a factor of ~3 to figure out what the comparison to a traditional utility battery.

If you're working with a fixed voltage battery (e.g. 12V automotive), then it doesn't matter which unit you use, both convey equal information. If battery voltage is significantly variable, then you usually have to know it anyhow. But in the event you don't either you're:
>dealing with constant current loads - it's better to use Ah
>dealing with constant power loads - it's better to use Wh
I refuse to believe anyone runs into constant current loads where battery voltage is significantly variable. Either voltage doesn't vary enough such that your load current remains approximately constant even without control circuitry, or you have even rudimentary control circuitry that ensures power gets fed to your load at the desired rate (e.g. automotive LED spotlights with a built-in SMPS).

>>2636716
>when voltage changes from 0 to a lot like it always does with DC motors.
What does this mean? DC motors will decrease the amount of current they pull as a function of voltage, unless your load is a constant torque and it's just the speed that changes throughout battery voltage changes. I can see this being slightly more true to reality when given something like a winch motor, but you're not going to be putting a 12V winch motor on 24V or vice-versa.

Furthermore, Wh can be converted seamlessly to Ah using the battery's nominal capacity, which is only inaccurate in the event that your parasitic series resistance is wasting a significant amount of power. We're EEs here, we try to avoid that unlike the morons using 12awg to power their kilowatt 12V subwoofer amp.

>>2636753
> Wh is less missleading
Yeah, especially in light of, say, some power tool manufacturers who lie about their 18v tools and call them 20v. Like dewalt.

>>2636656
>These seem both expensive and geared more towards power rather than torque, so they end up being too heavy for what I need.
If you're making an e-bike, this is amount of power you need. Less would be useless, I tried.

>>2636886
No, I am trying to build a force feedback controller (probably with OpenFFBoard), where having that much power is not only wasteful, but actually dangerous. Though cost is the biggest (read: only) factor in my case.

I asked this on /sci/ but got no response

An analog integrator is identical to a LPF. A standard 1-pole LPF decreases by 20dB per decade after the cutoff. However, numerical integration only decreases by 10dB per decade (compare sin(x) with sin(10x)).

Why aren't analog integrator circuits designed with, or in some way compensated to, a -10dB/decade cutoff region?

>>2637107
A standard electrical LPF is only -20dB/dec because we define 10V/V as 20dB. It's an RMS/resistive power related thing.

Getting a circuit with a natural 10dB/dec slope is impossible as I understand, so it's approximated with multiple slopes (see pink noise circuits).

>>2637457
>>2637457
>>2637457
NEW THREAD