/ohm/ - Electronics General: IPC Certified Edition

[Anonymous No.2581966]

Thread graduated:>>2575936

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

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

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

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

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

>Books:
https://libgen.rs/

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

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

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

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

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

[Anonymous No.2581976]

> 0 Ohm resistor

[Anonymous No.2582039]

>>2581976
> 0 Ohm resistor
what's that?

[Anonymous No.2582053]

>>2582039
exactly what it sounds like
a very poor current-capacity wire guaranteed to be 0 ohm within some tolerance band.

[Anonymous No.2582055]

>>2582039
undergrad fuse

[Anonymous No.2582058]

>>2582055
free use 0 Ohm slut MELF

[Anonymous No.2582070]

>When you step on a 555 timer
>>>/wsg/5008679

[Anonymous No.2582085]

>>2582039
pick-n-place friendly jumper wire.

[Anonymous No.2582110]

>>2581966
Look at that sexy solder job!

[Anonymous No.2582112]

>>2582110
Fuck you for making me open that image. I don't solder often, but even I consider it gore.

[Anonymous No.2582113]

>>2582110
>>2582112
OP used silver cake frosting in that pic to trick us.

[Anonymous No.2582117]

>>2581966
If it possible to build a CPU?

[Anonymous No.2582118]

I need a relay I can trip with gpio pins that can safely survive 500 watts at 120v. Any suggestions?

[Anonymous No.2582132]

>>2582117
Nice gore. You can easily run a softcore CPU on a FPGA. You can even write a softcore CPU yourself, if you have the HDL skills.

[Anonymous No.2582143]

>>2582118
Pico powering the relay or something else? Might need an intermediate transistor. Otherwise just look at specs of pico and specs of relay until you find what you need

[Anonymous No.2582162]

>>2582117
Yes
https://youtu.be/lNa9bQRPMB8

[Anonymous No.2582166]

>>2582117
People have made discrete CPUs before, assuming a 6502 counts. Make a whole discrete NES.

[Anonymous No.2582172]

How would you mount picrel to the panel?
Solder it to PCB then mount like a standard pot with a nut, or attach the PCB to the panel with screws as well?

[Anonymous No.2582202]

>>2582172
If it doesn’t have an alignment lug then screwing the board on will help keep it the right way up, though it doesn’t matter as much with rotary encoders. If you have a heavy PCB then screwing it on will help with stability, on the other hand if you don’t use the right spacer size the fastening may warp the board and make the traces more likely to break.

[Anonymous No.2582221]

>>2582172
mount the pot to the pabel with its nut and washer. mount pcb somewhere else. solder cables with a header/plug at one end onto the pot. plug cable into socket on pcb

[Anonymous No.2582237]

>>2582172

[Anonymous No.2582244]

what is this green glue that's on all the screws in my 1973 made in japan quadraphonic?
I want to say loctite but loctite doesn't dry outside of threads as far as I know, this stuff was put on the outside to just give keep things together far as I can tell

[Anonymous No.2582248]

>>2581966
WTF IS THIS SHIT

[Anonymous No.2582249]

>>2582244
threebond
they make both anaerobic screw locks and aerobic

I worked for Sony for 4 years and Canon for 3 and both used it for applying externally to the head of the screw. its really great stuff, the way it dries over the surface is incredibly satisfying

[Anonymous No.2582250]

>>2581966 (OP)
WTF IS THIS SHIT

electronics for boring people, i think of so many other better projects that are more fun

[Anonymous No.2582278]

I refuse to pay MOSFET gate driver IC kikes for their shitty overpriced pieces of shit.
Why can't I just use a charge pump voltage doubler IC (for high side NFET) + push-pull made of BJTs
Or should I go even further and wind my own gate driver transformer? (which I don't know how to)

[Anonymous No.2582280]

>>2581966
I used to solder like this a few years ago
I have a very bad shaky hands so I can't do anything precise but I can solder LQFP-48 chips

[Anonymous No.2582283]

>>2582172
>Solder it to PCB

they do that on commercial products coz it reduces labor costs: no hand soldering
but it causes a lot of grief coz, if the pot/switch/connector/whatever gets hit, it transfers force to PCB, which can break traces, or the PCB itself.
using flexible wires limits the damage to the exposed part alone.

[Anonymous No.2582296]

I need to measure the pressure at the bottom of a grad cylinder (yes the laboratory glassware thing) that is increasingly filling with water at a slow and steady rate.

would this suffice?

https://www.digikey.com/en/products/detail/nxp-usa-inc/MPX5010DP/464054

I need it to be as accurate as possible and be able to detect minute changes in pressure.

[Anonymous No.2582314]

I want to start a career designing PCBs.
I repaired or manufactured electronics for 15 years. I recently suffered a double hand injury and it looks like I will not be able to continue using tools at work.
Can i get some recommendations on youtube channels to watch with hours upon hours of PCB design content?
Any other suggestions on how to enter this career field greatly appreciated

[Anonymous No.2582326]

>>2582249
wow thanks!

[Anonymous No.2582332]

>>2582314
Most PCB design jobs seem to require at least 2 year associates degree in EE so acquiring that would be the fastest way to get a job in this field. Look at requirements for job positions and see what sorts of things you need to get before you can start applying for jobs.

[Anonymous No.2582334]

>>2582314
https://www.youtube.com/c/PhilS94/videos

[Anonymous No.2582353]

>>2582249
I think it's three bond 1401b..seems to be pretty hard to find any to buy here

[Anonymous No.2582360]

>>2582353
Sometimes they offer multiple colors for the exact same stuff. We would sometimes get red, blue, and clear, and it was all the same. It might have been 1401, cant remember. it was in the metal squeeze bottles. But both jobs were able to easily order it whenever and have it delivered in the US. I cant remember the sources, but you should be able to get it in the US relatively inexpensively.

Really cant speak highly enough about the stuff. I only havent pursued any for personal use due to laziness/crippling depression, but I do plan to get some.

[Anonymous No.2582383]

>>2582353
Can you find nail polish?

[Anonymous No.2582394]

Does anyone have a good guide on how to diagnose problems with electronics? As in how to diagnose what's causing a problem/short on a PCB? The amp in my car is no longer made (even when it was it would've been $2500) and it can't easily be replaced with an aftermarket amp without replacing the entire sound system. Something is wrong with it, a few months ago it stopped working about 90% of the time, but on cold mornings it will occasionally work as normal. I'm assuming some sort of solder connection went bad from my research, but even researching the issue has been very limited in results since almost every website says to just replace it. There's people that repair this specific amp for a few hundred dollars plus shipping, but if it's just a solder connection or replacing a capacitor I could do that on my own. I just don't know the best way to go about diagnosing it or finding the source of the issue. I'd like to try on my own before sending it off, mainly so I can learn and get better with electronics. I have a soldering iron and a multimeter.
Any advice on where to start with diagnosing the issue? Or any decent guides to share or videos? The few videos I've found when it comes to amp repair are people repairing the amp over the course of an hour but they don't describe what they're doing or what tools they're using or what they're looking for so I'm lost the entire time.

[Anonymous No.2582397]

>>2581966
jesus, how do you fuck up soldering that badly

[Anonymous No.2582399]

>>2582394
Post picture of PCB. Should be able to get a general idea of where the preamp and power amp sections are. Does it work at all, or is it just totally dead when the issue occurs? If it still kind of works, does the audio sound distorted or is there a noticeable difference in audio levels when it's messing up?

If it's totally dead when it isn't working, I'd be inclined to say it's a bad solder joint on the power input somewhere. If it's kind of working, you're gonna need more test equipment like a scope and signal generator, and even then it's kind of a crapshoot.

[Anonymous No.2582400]

>>2582397
You can assume a software developer did it.

[Anonymous No.2582412]

dpak pin spacing is about 1.5mm, which corresponds to clearance for around ~300V per IPC-2221B. how, then, does a 600v rated dpak mosfet work? or to-220s for that matter? i've even seen to-92s used as the power transistor in tiny flyback converters that see 400v+ reflected, and that package has <1mm pin spacing. is IPC-2221B just bullshit like every safety standard? what is the ACTUAL minimum practical spacing for X volts?

[Anonymous No.2582424]

>>2582412
You don't need to care about clearance if failure mode won't be able to kill anyone. Sure transistor could short, but that would only cause fuse to blow. What's important is isolation between primary HV side and secondary or anywhere that could cause harm if breakthrough happened. That's where clearance is important.

[Anonymous No.2582425]

>>2582424
personally i don't mind so much if my garbage circuit kills me, i just want it to be a robust design that will last for 20 years even when it's badly covered in dust so i want to know what i "really" need rather than what some bureaucrat thinks is a big enough number to exempt me from liability.

also on the topic of dust i designed a 20A wireless split-phase switch for a buddy's woodshop a few years ago, i looked at it recently, he'd left the cover off and the pcb was absolutely caked in wood dust but it still worked. i'm glad i never opened the breathing holes on the relays.

[Anonymous No.2582436]

>>2582399
I'll post a picture once I pull it from my car, I've been doing as much research as I can before I start fucking with it so I don't accidentally break anything.
>Does it work at all, or is it just totally dead when the issue occurs? If it still kind of works, does the audio sound distorted or is there a noticeable difference in audio levels when it's messing up?
When the issue occurs its entirely dead, there's no power to the speakers and the car diagnostics shows there's no connection to the amp at all. I've reseated the cables to be sure that wasn't the issue along with checking the relevant fuses in the car. When it does come on every now and then, it works perfect, it sounds great, there's absolutely no issue at all.
>If it's totally dead when it isn't working, I'd be inclined to say it's a bad solder joint on the power input somewhere
Thanks for confirming my suspicion, that's kind of what I've been able to gather through researching it.
>pic related
This is a screenshot from a video of a guy working on the amp in question, I'll link the video too. But it's not great video quality, and the guy doesn't talk about what he's doing, and he doesn't state what the issue is in the first place, it's just silence, so while it gives me a look at the inside of the amp I don't know enough to really glean anything from it.
>https://www.youtube.com/watch?v=BuzPSBdWfxQ

[Anonymous No.2582440]

>>2582332
>>2582334
thank you very much

[Anonymous No.2582446]

>>2582400
I'm from a software engineering background and I am absolutely flourishing in hardware design.
I whip up elaborate spice simulations with almost real world conditions (simulated using many, many spice commands). I use oscilloscopes to record real world load transients and other behaviour of our RF product and then I plug them in right into the simulations to refine them immensely. Our power source is a high impedance lithum cell which I have successfully been able to characterise and model in spice all thanks to my software dev skills.
The knowledge/skills of a software developer can be immensely useful for hardware design. Especially the ability to handle data.

[Anonymous No.2582447]

>>2582383
it's gonna go on a heatsink for output transistors are you certain nail polish is a good idea

[Anonymous No.2582454]

>>2582446
EEs can do all of that, but we don't because we can arrive within 5% of the answer with back of the envelope calculations and hand-waving. Having elaborate precise models when you don't need them is bad engineering.

[Anonymous No.2582456]

>need to drive a 12V MOSFET from a 3.3V microcontroller to control a motor
>recently read about voltage doubler circuit on the internets
>get a few diodes and caps to make one with my microcontroller
>generate 12V and drive the MOSFET
Why am i such a genious

[Anonymous No.2582459]

>>2582283
I think this is a different situation since the USB jack is only attached to the PCB but not to the case and just sticks out through the hole, But If a rotary switch is screwed to the panel, it is not transferring any stress to the PCB.

>>2582221
>mount pcb somewhere else
I guess it is a choice. I see a point doing it this way: >>2582237
The other side of the cable can then be attached to the main board. That way there would be no need to add a separate little PCB just for the encoder. So basically solder wires to the encoder pins as you would to a button or a switch that is attached to the panel.

>>2582202
>you don’t use the right spacer size the fastening may warp the board and make the traces more likely to break.
Good point.

Sounds like an extra PCB just for the encoder could be avoided. However the encoder pins appear really fragile and looks like >>2582237
could break easily if the cables attached to it could move. I think a little tiny PCB should be the best approach. After all it is meant to be mounted to PCB.

[Anonymous No.2582462]

>>2582446
>I have successfully been able to characterise and model in spice all thanks to my software dev skills
What's the connection? You model something in spice by connecting multiple elements together and setting their parameters, such current source or voltage source, parasitics, etc. What does that have to do with software dev skills? Sounds like spice skills and EE knowledge. You know what to expect from a hardware perpsective and you know which buttons to click to simulate that in spice. You dont need to know how to code to do that. I would agree you would need to have good analytic skills that are useful for soft dev. But pure EEs should have those skills as well. There is some overlap in how a CS and an EE brain functions.

[Anonymous No.2582464]

I just wanted to thank anon who is helping here for free with stupid questions. I'm coming from russian-speaking forums and imageboards with radio threads and topics, and all of them are bunch of roten elitist faggots who spit they shit on each hobbyist.

[Anonymous No.2582469]

>>2582464
4chan is a magical place, anon

[Anonymous No.2582470]

>>2582464
We are all retards here, so any stupid questions just blend in naturally.

[1 No.2582475]

>>2582436
>and the guy doesn't talk about what he's doing
Basically he checks the connection from the socket connector to the IC's. He uses a heat gun to melt the solder points.
He opens the coil covers to measure the coils and he checks IC connections.

[Anonymous No.2582476]

>>2582436
It looks like he's troubleshooting a bad power amp section. The long black rectangular section in the middle with the heatsinks and the MOSFETs is what's actually amplifying the audio signal. Since you said it works perfect with it does work, this section is probably fine.

If you pause at 7:51, the solder joints along the bottom of the board where the pins come through from the white connectors, just above what looks like two hex screws, that's the area I would inspect first. It should be fairly obvious if one of the solder joints is bad. Just remove the old solder with some solder wick and flow new solder on any joints that look questionable. While you have the board out, just give it a once-over on all the other solder joints and look for anything obviously damaged. Hopefully none of the pads are lifted because if they are, the repair gets a lot more complicated. Also look for shit stains from components that may have failed or any kind of green corrosion. Corrosion you can usually clean up with some contact cleaner and an old toothbrush.

If you don't find and bad solder joints, I'd start following the traces from whatever pins provide power to the amp and test for continuity in case you blew a surface mount fuse. They're not impossible to replace with just a soldering iron if it's a bigger fuse, but you really need a hot air station if you're going to start replacing surface mount shit. You could always just bridge the fuse with solder, but you run the risk of damaging other components if the fuse blew for a reason and not because it's just old.

[Anonymous No.2582480]

>>2582278
Common ~0.5A half-bridge drivers are pretty cheap. Consider paying the MOSFET gate driver IC chinks instead, something like the EG2131. Add a ZXGD300x totem-pole on the end if you want even more current.
>gate drive transformer
why, are you designing an isolated power supply?

>>2582296
Looks like it would work, though they give it a 5% accuracy rating, which is 0.5kPa = 51mm of water. You can probably do better, maybe with technologies other than a piezo disc. Maybe go for a less sensitive sensor, with a full-scale range closer to 100mm of water, as a 5% rating there would give you only 5mm of variance.

>>2582412
If in doubt, cover it in schmoo. There's differences between clearance and creepage, and requirements vary with dust ingress and humidity. Keep it all out with silastic and you'll basically have a situation where the wires can be as close together as soldermasked traces themselves can be.

[Anonymous No.2582483]

>>2582464
I'm self taught, so if I can do it, anybody can. I mostly work on guitar amplifiers, but started branching out into solid state amps since they're becoming more and more common in the music industry. Learning to work on vintage electronics really helped me learn the theory side of things as most functions are performed by discrete components and it's easier to visualize things.

[Anonymous No.2582485]

>>2582462
>>2582454
Its kinda difficult to explain in words. Its an attitude thing. All my EE coworkers treat CAD and spice simulations as the "big guns" and most believe it to be useless for anything that is not analog. Basically simulations are seen as more "academic" and not something that can help with real world issues. I personally feel the real power of spice simulations is when you can input real world data in transient simulations.

[Anonymous No.2582487]

>>2582459
>switch is screwed to the panel, it is not transferring any stress to the PCB.

not true.
the last case of this not being true was a bench power supply i fixed last month after it fell on the floor.
the front of the case, being plastic, flexed and a binding post cracked the PCB despite being screwed in.

[Anonymous No.2582492]

>>2582456
>Why am i such a genious

all you really needed was a simple NPN and a couple of resistors.
any voltage over 1V would have done the job.

[Anonymous No.2582493]

>>2582480
>Common ~0.5A half-bridge drivers are pretty cheap. Consider paying the MOSFET gate driver IC chinks instead, something like the EG2131. Add a ZXGD300x totem-pole on the end if you want even more current.
You're right the chink gate drivers are quite cheap. Man why are all the western chips so fucking expensive.
>why, are you designing an isolated power supply?
No I just want a circuit board with a generic H bridge driving circuitry. Eventually want to play around with switching a 12V to 120V 10A torroid transformer at mains freq

[Anonymous No.2582496]

>>2582492

naturally, i forgot the pic.

[Anonymous No.2582501]

>>2582485
If you spend a lot of time modelling real circuit simulations can reveal problems that could be difficult to detect otherwise, but this only applies if your models are accurate. For most applications it's more cost and time effective just to build the thing and then tweak it in real life. Where simulations make sense is if you're dealing with expensive things like IC design, where extra time spent on modeling pays off because messing up on that level is just too expensive - that's what spice (Simulation Program with Integrated Circuit Emphasis) was made for.

[Anonymous No.2582523]

>>2582493
>Eventually want to play around with switching a 12V to 120V 10A torroid transformer at mains freq
That can be done all low-side if you have a centre-tap transformer, so you can just use BJT totem poles to get high-current gate driving. Maybe even BJT pairs.

Pic related seems to be the method to get higher current out of that pullup resistor. See more here:
https://www.circuitden.com/blog/11

>>2582496
Ignoring that clusterfuck of an image, what if the anon doesn't have a 12V rail in the first place? Say he wants to switch a power load using a common MOSFET on a 1S lithium ion battery? Getting a power FET that turns on fully at 3.2V is kinda tough.

[Anonymous No.2582525]

>>2582487
I am trying to imagine the mechanics of it.
It seems like if something is rigidly attached to the panel, and that "something" is sitting on a PCB, AND the PCB is not attached to anything, wouldn't that something and the PCB make a single part and would be moving together, not separately with respect to each other?

> after it fell on the floor.
oh. that's not a very good use case. anything could happen depending on the impact.

it would be more interesting to know if there is a long term damage as a result of normal use.

[Anonymous No.2582528]

>>2582525
Sounds like in his case the binding post was soldered to the main circuit board, which was itself connected to the chassis. If you have a daughter-board it isn't a problem. If your encoder leads are as flimsy as mine are, then yeah I'd definitely use the daughter-board, though soldering wires directly and bulking it up with silastic/silicone might be fine.

[Anonymous No.2582550]

any recommendations for a cheap oscilloscope for beginners?

[Anonymous No.2582553]

>>2582496
I am using an N-channel MOSFET connected above the motor connected to 5V

[Anonymous No.2582555]

>>2582550
Anything from aliexpress, like
https://a.aliexpress.com/_mKKMIC2
or
https://a.aliexpress.com/_mt40FHu

But how cheap is cheap for you? Obviously the more expensive the better.

[Anonymous No.2582556]

>>2582528
>the binding post was soldered to the main circuit board, which was itself connected to the chassis.
Yes, makes sense. I wondered about this since you could have a main board mounted elsewhere, typically on the bottom of the project box and then a bunch of cables from it are run to the panel mounted terminals, buttons, LCD and what not. Or you could have everything mounted on the main board and the main board itself is then attached to the panel. I really only tried the former approach. And yes my encoder leads are really flimsy, nothing like typical pot leads that are thick and sturdy.

[Anonymous No.2582557]

>>2582525
>I am trying to imagine the mechanics of it.

no need to strain your imagination muscle.
coz there are pics.
explicit as always.
yellow arrow = break, red arrow = crack.

[Anonymous No.2582560]

>>2582557
Ah OK, so the board itself was also attached to the chassis. That would explain it.

[Anonymous No.2582563]

>>2582523
>Pic related seems to be the method to get higher current out of that pullup resistor. See more here:
I don't think Q1 and D1 are really needed in this circuit

[Anonymous No.2582576]

>>2582550
old analog scopes.
I got a free 5-channel one from a coworker.
I also got the same model used in Strange Days for $50 on eBay

[Anonymous No.2582581]

>>2582576
I was working at a nuclear plant doing board rework and my boss asked me to take an old techtronix analog scope that was sitting around to the dumpster because the power cable was damaged. I took it to the parking lot and put it in my trunk, lol. Cost me $2 to fix it.

[Anonymous No.2582583]

>>2582576
>>2582581
Cool story anon, I just checked our ebay and the old analog 15 MHz devices cost the same as new digital 100 MHz ones.

[Anonymous No.2582588]

I found a picture of it. I think it's a Tektronix 465. Super handy for diagnosing vintage audio circuits, but that's about all I use it for.

[Green Lantern No.2582591]

>>2582172
>How would you mount

/DO IT YOURSELF/

[Anonymous No.2582597]

>>2582583
analog is better than digital.
i see plenty of quality 100mhz analog scopes on ebay for $100.
i dont think a chinesium 100MHz digital scope is going to compare to an old Tektronix

[Anonymous No.2582599]

>>2582583
Tektronix 100MHz analong $70 shipped
https://www.ebay.com/itm/234923020971

HP 100MHz analog $75 shipped
https://www.ebay.com/itm/225423192906

Tektronix 100MHz analog $75 shipped
https://www.ebay.com/itm/185804440538

Agilent 100MHz analog scope $80 shipped
https://www.ebay.com/itm/295429374991
https://www.ebay.com/itm/234828544695

[Anonymous No.2582620]

>>2582469
I cut a boot off my car when I was at a training base in the Navy. MAs were complete dickheads about parking, and they booted my car. So I bought a cheap grinder from Lowe's and cut it off in the middle of the night. Never sweated harder in my life. Literally could have gotten kicked out if I got caught. Good times.

[Anonymous No.2582682]

I want to make an audio interface, USB to balanced jack.
I know I will need DAC to turn PCM signal into analog, but how do I tackle the USB part of the design?
In the past I have only dealt with Arduino and ATmega128 that I only programmed over USB, and then they ran on their own. I'm not sure how to tackle making a USB device and where to even start.

I haven't dealt with electronics in a long while either, what would be more suitable for my project if I want to work out the design at first? KiCAD or Logisim Evolution?

[Anonymous No.2582692]

>>2582563
Look at the sourcing current with and without Q1/D1. For simplicity's sake, we can look at it without the final stage, which just amplifies the current anyhow. Without Q1/D1 you have R1 between the 12V rail and your FET gate, which will limit the current to 12mA. Only 12mA into a FET gate isn't going to turn it on quickly at all, while you get ~500mA of sink current (with a 3.3V input signal). With Q1/D1 you get ~700mA of source current, the sink current is unchanged. Adding Q3/Q4 multiplies the source and sink currents by the hFE of these new transistors.

>>2582682
You use a dedicated IC, like the PCM2900. Otherwise it's an option to pair a microcontroller with both I2S and native USB (e.g. RP2040) with a nice audio-grade I2S DAC (e.g. PCM1780), but it's going to cost twice as much and you'll have to mess about with getting the right clock generated for it. Both the parts I mentioned are single-ended, there may well be nice differential output products made by BB/TI or someone else, though with decent board design it would be pretty easy to have a low noise floor after a pair of differential amplifiers.

Logisim is not software that can do this, you want actual EE CAD software. So KiCAD, Altium, or Eagle. The former is free and open-source and pretty decent, the latter two are closed-source with varying licenses that may or may not be free. Altium in particular has a suite of features for advanced users, but for the common hobbyist I can't imagine you'll use any of them. There's also EasyEDA and Altium Circuitmaker, which are free and closed source. I've used EasyEDA once and it was a shitty browser-based interface. All of these EE CAD programs have a section for designing a circuit diagram, and the ability to import this into a PCB layout. Some also have circuit simulators.

[Anonymous No.2582804]

I read that an electric eel can dispatch voltages of 500V at upto 1A
What would happen if someone was to shot its electric organs? Would it be hurt because of immense current draw or would it limit current to save itself?

[Anonymous No.2582808]

>>2582804
Go fish an electric eel and test it yourself.

[Anonymous No.2582866]

I have a circuit with pico MC on it and normally I want to power it using lion battery with a switch.

However while debugging/reprogramming pico via usb, I want to detach it from the battery and use USB as its power source.

Is it safe if I just disconnect VCC of the lion (so just turn the switch off) but leave GND connected? It would GND of battery and usb would be connected together but since VCC is cut off, it should be fine? Or should I disconnect the battery completely

Also not sure if related, I am not directly connecting the battery, there is a charging circuit that is connected to battery and my pico will be connected to the charging circuits gnd & vcc output (via a switch between VCC)

[Anonymous No.2582876]

I want to learn impedance matching and all that black magic. Of course, not at a greybeard's level but enough to be able to read the datasheet of an RF tranreceiver chip (say CC2500) and be able to design the matching circuit

[Anonymous No.2582900]

What happened to my flux

[Anonymous No.2582905]

>>2582900
It looks crystallized. Either that or your cat was soldering while you were away. Mix it with a little isopropyl alcohol and stir it up.

[Anonymous No.2582906]

anyone knows if they make light switches for us homes as a single pole double throw?

[Anonymous No.2582907]

>>2582866
>Is it safe if I just disconnect VCC of the lion
Yes, it's fine.

[Anonymous No.2582918]

>>2582900
That's a nice flux brûlée. Looks delicious.

[Anonymous No.2582926]

don't die

[Anonymous No.2582927]

>>2582866
rawr >:3

[Anonymous No.2582934]

>>2582906
>light switches for us homes as a single pole double throw?

they call 'em 3-way switches, to let you know they're kinky.

[Anonymous No.2582945]

Dumbass homeowner here. I don’t want to make a new thread. I had a single double receptacle that was a switch on the top and plug on the bottom. I wanted to change it to just a standard receptacle. Top outlet does not work with open neutral. Bottom is “wired correctly”. Pic related is a very rough diagram.

There is a junction that goes to a switched light and a pair of switched lights in my basement as well as this receptacle. From the junction it goes to the fuse box. I have tried a few combinations of wiring with no luck. What do?

[Anonymous No.2582946]

>>2582945
Current wiring.

[Anonymous No.2582947]

>>2582946
I have tried other variations to no success.

[Anonymous No.2582952]

>>2582946
Put one neutral on each terminal.

[Anonymous No.2582965]

>>2581966
rollan for a project

[Anonymous No.2582980]

>>2582952
I did. Soon as I swapped to see I now have an open neutral on the bottom outlet

[Anonymous No.2582981]

>>2582980
Wait. Now it’s hot/grd reversed

[Anonymous No.2582990]

>>2582980
>>2582981

[Anonymous No.2582991]

>>2582990

[Anonymous No.2583001]

>>2582990
>>2582991
I’m sorry I’m just a dumbass. I can’t seem to make any sense of those in relation to my wiring. Getting frustrated at other things wishing my dad was still alive to help… anyway pic related is as far as I can tell. I’ve tried about every combination on that light as I can. Circled was formerly a switch and single outlet. All I did was copy the wiring from prior and make it a dual outlet.

[Anonymous No.2583012]

Why doesn't the led turn off if you hold the button? The capacitor gets charged and goes over the treshold and should reset the latch that wants to stay set due to the trigger being 0 while the button is pressed?

[Anonymous No.2583019]

>>2583012
Having both R and S high is an undefined state and will cause output to lock in a certain state depending on how latch is actually constructed.

[Anonymous No.2583035]

>>2583012

maybe pin 4 is unhappy about being left floating and alone.
tie it to Vcc always.

[Anonymous No.2583043]

>>2583012
If you do want that behaviour, I think you need to use the reset pin for your button instead.

[Anonymous No.2583048]

I want to charge a battery with a dc-to-dc charger.
Should I power my dc-to-dc with a cheap used atx power supply or with a cheap alixpres power suply?

[Anonymous No.2583053]

>>2583048
depends on the battery. particularly, its charge rate

[Anonymous No.2583055]

>>2583053
Why?
Lifepo4 battery.
From what i've read the dc-to-dc converter will always try to draw it's max amp, since the battery will draw as much current as it can, as such as long as the wattage of my power source is greater than the max watt of my dc-to-dc converter (+ ineffiency), it shouldn't matter? Aren't they both regulated power supplies?

[Anonymous No.2583060]

>>2583055
>the dc-to-dc converter will always try to draw it's max amp
Are you trying to charge a LiFePO4 just by connecting a constant-voltage switching regulator to it? Something is gonna overheat, man. Those things aren't a CC/CV switching regulator, they're a CV regulator with a maximum allowable current. If you try to draw more they won't just regulate the voltage down like a good boy, they'll get hot, maybe thermally throttle, maybe not, definitely decrease the lifespan by running hot for so long. Get a proper CC/CV regulator.

[Anonymous No.2583062]

>>2583060
don't worry it's a dc-to-dc charger made for the market who want to charge their batteries from a car or generator. It's compatible with lifepo4

[Anonymous No.2583065]

So this DRV8301 has a built-in charge pump that converts the <60VDC from a battery down to 10V or so for the gate drivers. But I want to drive them with more like 12V, so I have more headroom to add a totem-pole on the output. They also have a "trickle charge pump" to feed the gates and allow it to work with 100% duty-cycle, which half-bridges can't normally do. The buck converter is just intended to provide 3.3V or whatever to your MCU, but I want to use it to provide this 12V. Does anyone know what would happen if I fed 12V directly to the GVDD pin? I don't really understand what a step-down charge pump entails so it's hard to imagine. I assume since in this case the input voltage can vary a lot there will be a feedback circuit in there, so feeding it a fixed DC voltage above its set-point would just disable it?

[Anonymous No.2583074]

Is there no indication on how hot this thing actually gets, or am i just retarded?

[Anonymous No.2583124]

>>2583055
>Why?
because many batteries can be charged at more amps than "a cheap aliexpres power supply" can provide. ATX PSUs are capable of very high amp outputs.

[Anonymous No.2583286]

>>2583074
Well you can use the efficiency/current values or curve to figure out how much power it's wasting at a certain output level. Though that won't tell you the temperature it gets up to.

[Anonymous No.2583367]

Can you just wire correlating pins from ide to pcmcia and get fe a cd drive to run?

[Anonymous No.2583373]

>>2582927
Cute

[Anonymous No.2583375]

>>2581966
>Image from reddit

[Anonymous No.2583376]

>>2582456
Wouldn't the higher stages where the voltage across the cap is higher than the maximum that the MCU can handle hurt the GPIOs of the MCU?

[Anonymous No.2583377]

>>2583367
I'd imagine there's silicon in there doing some smart shit. Read into the standards and see for yourself what it would take to convert data from one to the other.

>>2583375
Yes. Is there a better place to look for retarded electronics practices?

[Anonymous No.2583380]

>>2583377
The other way looks pretty straightforward in some cases

[Anonymous No.2583391]

Hey guys I'm making a programmable load with an atmega8. I intended to use the PWM to switch a 2N7002 MOSFET based boost converter to feed the base of a power NPN transistor with steadily increasing current (higher duty cycle = higher boost voltage = higher current through 100k going into the NPN base) but the datasheet says it can only handle the following
>max base-collector voltage 100V
>max collector-emitter voltage 100V
>max base-emitter voltage 5V
Will my NPN transistor die if i subject the gate to a voltage higher than 5V? Seems retarded desu

[Anonymous No.2583396]

>>2583391
>if i subject the gate to a voltage higher than 5V?

that's not what it says.
the base-emitter connection is a diode, so it's a nominal 0.7V drop.
to get it as high as 5V you gotta drive some very heavy current through it.
as for base voltage, you can go as high as 100V as long as you limit base current to it's max value.

[Anonymous No.2583397]

>>2583396
Thanks anon, yeah that's about what I thought
Its funny how as a hobbyist I've played around with everything digital but barely anything analog

[Anonymous No.2583402]

>>2583286
I already thought of that, but i'm not smart enough to dive into deeper physics. Ordered my parts ahead now and will just test it on a constant load.

[Anonymous No.2583560]

Hey guys where do i get iPhone/Samsung boardviews from? Friends told me he would suck my dick if I find them

[Anonymous No.2583566]

>>2581966
i found this schematic on eevblog. the guy built a boost converter using an UC3843 to drive an LED, and so he wanted current limiting feedback to properly drive the LED.

this is realized by R9 and op amp network that feeds into "VFB" on the UC3843. the voltage of the boost converter output is regulated by feeding into "VFB" as well.

my question is: why is this even necessary? shouldn't R1 and "ISENSE" be sufficient to control all of the current that flows through the LED?

[Anonymous No.2583586]

>>2582117
Sure, if you want to wire up 500 DIP transistors by hand. But maybe you'd want a FPGA or a simple micro controller for your task.

[Anonymous No.2583624]

>>2583566
>shouldn't R1 and "ISENSE" be sufficient
yeah. if you tie VFB to VREF to disable the voltage loop then ISENSE will have a fixed 1V limit that will be related to the led current. you could lower the limit with a resistor on COMP to avoid too much sense resistor dissipation. or just use a switching controller designed for leds. often even those will have direct led current sense for fault detection (short/open led).

[Anonymous No.2583625]

>>2583624
also i'm not sure what this would do to loop stability. i forget if/when boosts need slope compensation, how that's getting injected, and how this would mess with that. you'd probably just have to build it and see if it oscillates.

[Anonymous No.2583628]

>>2583624
>tie VFB to VREF
also i mean tie VFB to ground
i think

[Anonymous No.2583638]

>>2583624
>>2583625
>>2583628
i'd still want voltage and current feedback. so leave VFB tied to the R7/R8 voltage divider (but get rid of that diode since it's no longer necessary, once the op amp network shit is deleted) and ISENSE to R1 shunt resistor.

ignoring the resistor values (can't remember UC3843 ISENSE and VFB voltages off the top of my head), this alone should provide adequate current control to keep the LED safe, right?

i guess i don't understand why the guy who drew this schematic added an R9 shunt resistor and all of the ensuing op amp stuff when, if i'm understanding this correctly, measuring current through R1 shunt resistor is more than adequate.

[Anonymous No.2583651]

>>2583638
>and ISENSE to R1 shunt resistor
ISENSE is already tied to R1. the rc network between R1 and ISENSE is only to filter out switching spikes, it's not really averaging out the Ids ramp or anything

>leave VFB tied to the R7/R8 voltage divider
in the "intended" voltage-limiting operation the current limit is an inner loop with a setpoint determined by the outer voltage loop. if the load is too hungry then there's a 1V zener that stops the outer loop from setting the inner loop setpoint too high. you can't have led voltage control and led current control at the same time, on the contrary, you don't want led voltage control at all except for ovp. the gay opamp scheme converts the outer loop from a *voltage limit* into an *average led current limit* while keeping the inner loop for reasons of stability.

>measuring current through R1 shunt resistor is more than adequate
you are correct (fault detection aside)

[Anonymous No.2583655]

>>2583651
thanks for taking the time to explain.

[Anonymous No.2583661]

>>2582934
They have 3 position (up, down and a middle position). Are there any switches SPDT with only 2 positions?

[Anonymous No.2583671]

>>2583566
If the LED goes open-circuit for any reason, the R8/R7 network will prevent the voltage going so high the converter overvolts itself.

>>2583651
R1 is just the current through the transistor, as it's a current-mode converter. It's correlated with but not directly proportional to the DC current through the LED. The peak transistor current is basically a measure of the energy stored in the inductor each switching cycle, but since it's not a fixed frequency converter, the number of pulses of that set energy per second can vary, and hence actual output current can still vary. An LED is arguably a constant enough load for that not to matter, but as the LED heats up, the voltage across it will decrease, which will result in the converter switching faster(?), which will result in more current going to the LED, which is a positive feedback loop. There might be a reason you've never seen an open-loop current-mode converter before.

Actually maybe limiting the switching current doesn't even influence the output power much at all, a lower switch current means less time to charge the inductor, which means it will pulse more times per second anyhow. Even if it works you're likely switching faster than you'd typically want to, incurring worse switching losses.

[Anonymous No.2583685]

>>2583671
>pulse more times per second anyhow
the frequency is fixed by an RC oscillator, ISENSE only controls the duty

[Anonymous No.2583690]

>>2583661
>They have 3 position (up, down and a middle position). Are there any switches SPDT with only 2 positions?

are you sure of this? The whole point of a 3-way switch is that either one can turn the light on or off at any time, like in a stairwell or a hallway. Your claim defeats that purpose if one switch is left in the middle as yo claim, and furthermore what good is the center position in "your" switch? I think you are wrong.

tldr, 3-way does not refer to the positions of the toggle.

[Anonymous No.2583706]

>>2583685
>the frequency is fixed by an RC oscillator
>on a current-mode converter
huh, never seen that before

[Anonymous No.2583716]

>>2583706
yeah i guess most of them are PFM now aren't they

[Anonymous No.2583721]

>>2583716
Yeah, I've heard PFM is more efficient. Though some converters switch modes depending on output power, like the one on the Pi Pico.

[Anonymous No.2583727]

>>2583690
Sorry, you're right. I looked at SPDT light switches and for some reason all of them showed up as 3 position switches, but there are also SPDT switches with 2 position, it's just they were all the way at the bottom of the list. Thanks.

[Anonymous No.2583736]

Can i use ceramic capacitores in a 7805 montage?

[Anonymous No.2583791]

is there a name for 3>= input xor that doesn't do the gay thing of odd inputs high = out high, but instead only 1 input high = out high

[Anonymous No.2583827]

>>2583791
>only 1 input high = out high
uh
OR

[Anonymous No.2583829]

>>2583827
nevermind in my rush to make fun of you i misunderstood your post. draw a truth table and build from that.

[Anonymous No.2583843]

>>2583829
I know what the truth table is, I'm wondering if there's a component that does it and what it's called.

[Anonymous No.2583867]

>>2583791
Reading this:
>https://www.ti.com/lit/ml/sdyz001a/sdyz001a.pdf
It seems like the "=1" definition of a 2-input XOR applies to you. I'd browse through some common logic chip families to see if any 3+ input XORs do this. If not, no clue.

[Anonymous No.2583897]

I hope I'm posting this in the right thread... if there's any electricians here, I could use some feedback.

So I'm desiging a small hoist system. It's small enough that I can basically have the entire cabinet on a single sheet. However, I realize that my drawing isn't exactly a standard wiring diagram. Rather than the abstract with standard electrical symbols, I've been making a simplified wiring. Besides a couple of power supplies, a breaker, a motor driver, and a PLC, there arent any parts besides the terminal blocks. Should I start over doing a standard electrical wiring diagram? It would be extremely simple since i only have a few hard inputs/outputs.. what would you even call what I have here? A half wiring diagram half schematic? It's not complete, but I'm nearly done. Since it doesn't need 20+ pages like most systems, is this style acceptable in a semi professional setting? Any advice?

[Anonymous No.2583915]

We're trying to find some LED strip lights for our basement, to go around the trim at the ceiling all the way around the room. We'd like them to check the following boxes:
>Single-strand, or can be extended to be such.
>Can cover 50 feet of wall
>Has both RGB and White
>Can interface with smart devices like google home
>Could potentially have enough Lumens to light up the room on their own, but not a necessity.

So far, our searching has come up short with all results checking some of but not all boxes. Any suggestions?

[Anonymous No.2583918]

>looking at connections to UPS AC-to-DC converter
>converter transformer output is tied directly to the battery via a rectifier
>looking at switching controller (UC3843BN)
>there's a BJT base tied to oscillator pin (4)
>collector tied to reference pin (8)
>emitter tied via diode and 2.7k resistor to switching current feedback pin (3)
>additional base-collector resistor
what the fuck is this circuit doing
anyhow the feedback pin is connected to the battery voltage (via some kinda strange transistor circuit I'll have to look at) using a voltage divider, and that divided voltage also goes to the MCU without an RC filter, and since the MCU has no DAC it's safe to say that's just an ADC input. I'll have to see if there's any proper way of the MCU regulating the charging.

That switching controller is a current-mode fixed-frequency jobbie, so maybe they're using that limited switch current to regulate the charge current. Kinda strange if so, but we were just discussing that in the thread.

[Anonymous No.2583919]

>>2583915
Add extra thicker wires behind the strips if you want a single string to cover 50 feet.
>google home
I think you're best off finding an RGBW LED strip controller seperate from your LED strips. That way you have more freedom as to the CRI and length of the strip you're buying.

[Anonymous No.2583923]

>>2583918
is the DC-to-AC circuitry completely separate?

>there's a BJT base tied to oscillator pin

[Anonymous No.2583924]

>>2583919
Interesting. I hadn't considered a separate controller and then getting LED strips to connect to it. We're very new to LEDs and their limitations and aren't sure why it's been so difficult finding what we want as an out-of-the-box solution.

Is there a technical reason as to why getting a single-strand, 50 ft. strip with the color and Lumen options we want seems to be so difficult?

[Anonymous No.2583949]

>>2583923
Ah it's deeper in the datasheet. The title is "slope compensation" so I guess it's for stability at higher duty-cycles, like is a problem with these current-mode converters. Can't find any other connection between the MCU and the feedback loop, but on the other hand the only connection from the feedback loop to the battery voltage is via an op-amp controlled PNP pass regulator. There's a voltage divider from that regulated output to ground, with the middle taken off to the feedback input, just like I'd expected the battery voltage to be measured, but there's no actual signal getting through. I'll try tracing out these MCU pins to see where they go next.

[Anonymous No.2583951]

>>2583924
The problem with high-power long LED strips is you end up with a lot of voltage dropped along their length. Bright at the start and noticeably dimmer at the end. So going for multiple smaller strips, each with power cable going back to some central PSU is what I'd want to do.

[Anonymous No.2584011]

Can I toggle an nMOS off by shorting its gate to ground with a BJT?

[Anonymous No.2584021]

>>2584011
Yes, if by "toggle" you mean turn off for as long as you power the BJT. If there's still something pulling its gate up the BJT will have to pull harder than that, and if it's persistent then the nMOS will turn back on again once you stop feeding current to the BJT base.
Technically speaking, an nMOS will have a gate source threshold voltage between 1V and 6V, and a common BJT should be able to pull the gate below 0.4V. It's also an option to use a logic-level MOSFET (e.g. BSS138) instead of a BJT.

[Anonymous No.2584046]

People etching pcbs at home via acid, how do you get rid of the acid? Guess you can't just pour it in a drain or next to a tree

I am planning let it evaporate and double package the remaining solids and just dump it but it is too cold here for letting it evaporate

[Anonymous No.2584050]

>>2583897
>A half wiring diagram half schematic?

what you have is a wiring diagram.
you'd give that to a wiring person and tell them to wire up the system as shown.
if you gave that to a technician or an engineer, they'd give you a funny look, and walk away shaking their heads.
coz to them, a diagram should show function, not a rat's nest of wires.

[Anonymous No.2584052]

>>2584046
>Guess you can't just pour it in a drain

oh?
then why does everyone who makes PCBs have laundry sinks that look like this or worse?

[Anonymous No.2584056]

>>2584021
Thanks anon.
>if by "toggle" you mean turn off for as long as you power the BJT
That's the plan; The BJT/MOSFET combo is meant to keep a 24V battery pack disconnected while mains power is on so that its buck convertor isn't jostling currents with my router's 19V DC adaptor.

[Anonymous No.2584057]

>>2584046
Neutralize the acid. Assuming you're using some sort of HCl + H2O2 mix, you could throw in baking soda to create salt NaCl + CO2.

[Anonymous No.2584065]

>>2584046
FeCl3 etch produces CuCl2 in an acidic solution. SLOWLY feed it strips of aluminum until the blue color disappears, and a brick red copper powder will precipitate. Neutralize with baking soda, strain out the solids and pour down the drain.

[Anonymous No.2584068]

>>2584050
So if I'm designing a system for a small business that has its own fabricators, would a more technical wiring diagram be necessary? Or perhaps just a ladder diagram type?

[Anonymous No.2584073]

>>2584057
>>2584065
I have sodium persulfate

[Anonymous No.2584079]

>>2582118
Don't do that shit. You'll blow your gpios out. Use a logic buffer of some sort. There are lots of good MOSFETs out there and a wealth of instruction on how to build. Also, don't forget the flyback diode on the relay inputs.

[Anonymous No.2584106]

do you solder pcbs or reflow them?

[Anonymous No.2584107]

>>2584106
for smds i solder and then reflow to make it pretty

[Anonymous No.2584108]

>>2584107
why though? you could just place them on with tweezers and reflow all at once, soldering is just extra work at this point

[Anonymous No.2584109]

>>2584108
i don't use paste because it goes bad

[Anonymous No.2584110]

>>2584068
>would a more technical wiring diagram be necessary?

you essentially have to give them the keys to the city:
schematics, wiring, bill of materials, parts placement, source code (or equivalent)
everything that'll allow any person with knowledge of the art to diagnose/repair/replace the entire system.

[Anonymous No.2584117]

>>2584109
>i don't use paste because it goes bad
that is what Big Paste wants you to believe so you keep buying more.
I am using a big tub of 4 year old solder paste i keep in my fridge and it still works good as new

[Anonymous No.2584118]

>>2584117
i always wondered about that. the fucking jews had me fooled all this time.

[Anonymous No.2584156]

>>2581966
I'm testing some electric field meter
>Trifield TF2
- Not handling the meter it's 16V/m
- Handling the meter in hand, pointing house floor, it raises up to 120V/m.
- Grounding myself, handling the meter, it goes up to 210V/m
Can someone explain me this ?
I got weird crackling sensations under feet and sometimes spasms.
I'm worried by the situation.

[Anonymous No.2584170]

I want to make sure that the input voltage of an analog video signal is always 0.7Vpp, the input can sometimes be 5Vpp or at other the expected 0.7Vpp. How would I go about achieving this?

[Anonymous No.2584173]

>>2584170
>or at other the expected
can you re-word this please

[Anonymous No.2584175]

>>2584170
Sure thing, The input can come from different sources, which use different reference voltages. Some source come with 0.7Vpp and others with 5Vpp. I want to make a small circuit which makes sure that no matter the input (0.7 or 5) the output is going to be 0.7Vpp.

[Anonymous No.2584176]

>>2584175
Look up "clamping diodes"

[Anonymous No.2584184]

I want to learn about protection circuits for industrial aplications like Proximity switches.
Basicly what do I need to run an MCU with simple industrial interfaces like Digital IOs and 0..10V or 4...20mA analog interfaces on PLCs?
(No fieldbusses for now).
Supply voltage will be 24V but I assume I'll need some protection on that aswell.
I'd love to take a peek inside those sensors and see how its done but they usually are potted.
Got anything I could read on that subject?

[Anonymous No.2584185]

>>2584046
I regenerate it by electrolysing it back into ferric chloride, using a platinum anode and a copper cathode. Going slow (e.g. <100mA) as to not oxidise the anode with reactive chlorine species, and to not produce chlorine gas. Then I just have copper powder to dispose of.
>>2584073
>persulfate
That's even easier, there's basically no risk to damage your anode or to produce toxic gases, just set the current such that you don't get any bubbling at the cathode. You should be getting bubbling at the anode as oxygen gas escapes. The result will be a mixture of sulfuric acid and sodium sulfate, which can't be reused as an etchant, but is useful about the shop. If you want to dispose of it just add baking soda until it stops bubbling, or just substitute out the copper with aluminium foil in the first place. I think that works.

>>2584056
Use a nice big cap to smooth out the changeover period.

>>2584170
>>2584175
If you don't care about distortion, use clamping diodes as the other anon said. If you do care, look at Moritz Klein's video on making an audio compressor.

[Anonymous No.2584188]

>>2582118
>gpio pins that can safely survive 500 watts at 120v.
Not going to happen.

[Anonymous No.2584201]

>>2584175
>>2584185
Thanks anons, I'll take a look. Thanks for the pointers

[Anonymous No.2584207]

>>2584188
Presumably he just means able to handle 4.2Arms when closed, and 120Vrms when open. Not dissipating 500W directly. Why would you dissipate power in a relay anyhow?

[Anonymous No.2584211]

>>2584207
Why you try to put 500w through a GPIO pin? His question is poorly worded, indicating he doesn't understand what he's doing and is trying to cargo cult his problem.

[Anonymous No.2584220]

>>2581966
Shit reminds me of when I got handed this in the first month of starting work
>hi anon glad to have you
>yeah you're going to take over this project
>intern was working on it during the summer you can pick up the info and prototype from them to get started
>hey intern i'm anon how's it going i heard you had some stuff for me
>greeted with this shit
it's all so tiresome

[Anonymous No.2584222]

>>2584220
Just start from scratch. That's retarded.

[Anonymous No.2584247]

Brainlet here, I wanted to swap out the power cord (UK) on this grow light for a longer one but inside the power supply they slathered it all in some weird rubbery shit. Am I ok to cut the wires and join them to the new cord with some heatshrink or am I gonna die doing shit like that?

[Anonymous No.2584249]

>>2584220
So this is the power of lockheed-martin

[Anonymous No.2584262]

>>2584249
No that would be Boeing. What you're looking at is early prototype of 737 MAX MCAS.

[Anonymous No.2584265]

is it me or are Weller soldering irons total fucking TRASH nowadays???
>recently buy one from home depot
>stops getting hot within a month of on and off use
like it still lights up and stuff but the heater seems to be built like total fucking garbage... sdo i suck it up and buy a real station with the replicable tips? if so, any recommendations for under $100?

[Anonymous No.2584270]

>>2584265
>buy soldering iron from Weller Changdong Electronic Co. Ltd.
>why is it built like total fucking garbage
https://www.youtube.com/watch?v=OjNpRbNdR7E
Buy european or japanese tools next time.

[Anonymous No.2584274]

>>2584265
Hakko FX888D
>under $100
Stop being poor

[Anonymous No.2584294]

>>2584265
ksger t12 or aixun t3a/b

[Anonymous No.2584312]

>>2584274
>>2584294
is the pinesil any good? i think that also takes t12's too

[Anonymous No.2584315]

>>2584312
it's the pine-sol of peencels.

[Anonymous No.2584348]

>>2584211
He means the relay can handle those, not the GPIO pin, learn reading comprehension.

>>2584220
Was the code at least understandable?

>>2584294
What’s the T3b like?

>>2584312
It’s good if you’re in the USA and can get the cheap shipping. Otherwise the TS101 is a good portable iron.

[Anonymous No.2584363]

>>2584247
yeah, just unplug it first. wrap your connectors in electrical tape for good measure and put some heatshrink around it.

[Anonymous No.2584420]

>>2584363
>wrap your connectors in electrical tape for good measure and put some heatshrink around it
https://www.youtube.com/watch?v=oDGTCULn6P0

[Anonymous No.2584434]

>>2584420
that's about how it goes when I cook. how do you know me so well, anon?

[Anonymous No.2584438]

>>2584434
I listen to you through your sail foam. Actually, Chat-GPT listens and curates a list of the important details.

[Anonymous No.2584487]

I unironically solder like OP's pic how do I not be retarded? Videos and guides make everything seem effortless.

[Anonymous No.2584497]

>>2584487
because it is effortless. any monkey can do it. if you cant, this field/hobby isnt for you

[Anonymous No.2584501]

>>2584487
Gotta get in quick and out quick. Don't stay for so long that you burn away all the flux from the solder wire, or overheat the parts on the board. Ensure you have an appropriate tip geometry for making good thermal contact to both the pin header AND the pad it's being soldered to. If you're using a conical tip, this basically means it has to be on its side. Screwdriver tips are objectively better than conical tips. To ensure speed, you need to have your iron/station set to an appropriate temperature (300-350C), or if it's not temperature controlled, be powerful enough to work on that kind of workpiece (at least 20W, preferably over 30W). Good workholding to ensure you have two hands to use the iron and solder wire is also greatly recommended (shove the arduino in a breadboard and blu-tak that to the bench if you have to), and ensure you can properly see what you're doing with plenty of light. Lastly, your solder alloy may not be making things easier, leaded solder is easier to solder with than common lead-free.

[Anonymous No.2584595]

>>2584487
raise the temperature

[Anonymous No.2584651]

>>2584497
use flux.
make sure the parts you are soldering are hot too.
hold the tip at your solder joint for a long time to make sure everything gets hot

[Anonymous No.2584686]

>>2584312
i don't have any experience with it, but i've heard it's good.
>>2584348
>What’s the T3b like?
i've seen youtube reviews and it looks really nice. i have the ksger t12 and i'm happy with it. if it shits the bed, i think i'm getting an aixun.

[Anonymous No.2584921]

>>2581966
Arduino bros, please help me. I modified a servo motor to rotate continuously as per the instructions of this guide:

https://www.instructables.com/Converting-Micro-Servos-for-Continuous-Rotation/?amp_page=true

It works but spins too fast; about 480 degrees in one second is the least I can make it spin through code. Is there a way to make it to spin slowly?

[Anonymous No.2584984]

>>2584921
Thing is, that's not a servo anymore, it's just a gearmotor. There is no feedback loop. You might be able to just PWM it (analogWrite if you're a blinkie), but I'm not sure if the internal control circuit will take that or not.

[Anonymous No.2584992]

>>2584921
check out that page you linked because it says exactly what to do.

>180 causes your servo to spin clockwise at full speed. 0 causes it to spin counterclockwise at full speed. Values of 45 and 135, for example, have the servo spin slower in their respective directions while 90 theoretically should keep the servo at rest. However, we say “theoretically” because the servo is never really perfectly centered.

>>2584984
>just a gearmotor. There is no feedback loop.
completely wrong. The only difference here is that it never reaches its setpoint. It can continue to spin as quickly as possible but never "satisfy" a move and stop, it doesn't have a stable point there. So indeed the position now maps to speed, because the control algorithm increases the motor's current based on the difference between zero and 'desired' position, and the 'desired' position is user-controlled.

[Anonymous No.2585010]

>>2584984
>>2584992
So I’m a complete amateur but I had a thought: could I solder a resistor to the power wire of the servo? Would half the power make it half as fast? It’s connected to the 5v pin so would a 2.5 kilo ohm resistor halve its speed?

[Anonymous No.2585040]

>>2584992
>the position now maps to speed
Having what's basically a PWM speed controller doesn't mean there's a feedback loop. The potentiometer is no longer connected to the motor; QED it's operating open-loop.

>>2585010
Yes but it's a stupid idea.

[Anonymous No.2585106]

what is the simplest way to detect movement?
i just need some thing for my micro to detect if user is wearing it or if he set it down on the table or something so i can shut it off automatically.
I bought vibration sensors which are basically little spring on a stick and found out they are useless because they only detect shock when you say drop the thing on the ground.
I have some mpu6050 chips but that seems like a large overkill just to detect movement

[Anonymous No.2585109]

>>2585106
You could try using one of those spring sensors as a capacitance sensor instead. Hook it up as the C in an RC relaxation oscillator (e.g. with a 74HC14) and measure the frequency. If the spring gets 10% closer to the wall you should be able to measure that. Otherwise you can buy pure accelerometers (no gyro) that are cheaper than whole MPUs. A single hall sensor would, similar to a gyro sensor, be able to measure changes in angle relative to the earth, which won’t be quite as good as an accelerometer but might be good enough. Some MCUs like the ESP32 have a built-in hall sensor.

[Anonymous No.2585110]

>>2585109
>Some MCUs like the ESP32 have a built-in hall sensor.
Seriously? i have been using ep32 for years and never knew that, gotta check that out
> C in an RC relaxation oscillator that would be getting too complocated, 6050s are dirt cheap and tiny so at that point i would just use that
but i hope i can get the hall sensor thing working

[Anonymous No.2585113]

>>2585110
>6050s are dirt cheap
Are they? They're $16 on digikey, if you're getting them much cheaper I'd suggest they might be fakes. But hey if they work who cares.
>i hope i can get the hall sensor thing working
Don't know how well the hall sensor works through the metal can of a WROOM module. Good luck.

[Anonymous No.2585114]

>>2585113
got a bag of like 10 6050 breakout boards for i think $3 from chinks
and they work fine from when i used them

So i got the hall to work but sadly won't work for my purposes, it can detect magnet fine, but it is not sensitive enough to react to earth magnetic field when i move it around by itself

[Anonymous No.2585115]

>>2585114
>not sensitive enough to react to earth magnetic field when i move it around by itself
Ah, shame. Try using the ESP32's capacitive touch sensing abilities with the spring sensor, that way you won't need the external oscillator.

[Anonymous No.2585169]

>>2584184
Controls Engineer here. I've seen this exact prox sensor used before. We use it with simple CB protection inside the main panel with all our IO.

[Anonymous No.2585181]

>>2585106
large capactive touch pad
flexible resistor in the lining that bends when the user puts it on
sensitive membrane switch
lightly spring loaded contact switch

[Anonymous No.2585296]

where do i learn to solder and what equipment besides iron do i need?
some good lighting and a stand i presume

[Anonymous No.2585317]

>>2585296
Ventilation

[Anonymous No.2585329]

>>2585296
Brass wool for cleaning tip
Holder for brass wool
Flux
Fan for blowing the smoke out of your face
Desoldering braid/wick
Variety of solder iron tips

[Anonymous No.2585334]

>>2585296
Light, good eyes, table that you can burn, or silicone mat.
Solder (use leaded, 60/40 or 63/37, avoid Chinese solder, but if you can't get it, lead-free is usable but for a newbie, it would be harder);
Flux (get couple types. Pine rosin (it is a flux, yes), liquid flux (store-bought or mix rosin with alcohol), gel flux (all of them are OK apparently, but for BGA I'd suggest $30 a syringe one because lol) , plumbing flux (zinc chloride) for those weird metals that don't solder)
Roll of toilet paper to wipe the tip
Alcohol, both to clean the flux, or drink in case of shaky hands
Solder sucker (get good one, like Goot, japanese)
Soldering wick
Tips for your iron
Some soldering kits to get experience
Tweezers

Watch videos about soldering. (search "Pace soldering video")

[Anonymous No.2585344]

>>2585296
>what equipment besides iron do i need?
cheap fan to blow fumes away from face, window for ventilation, empty soda can to discard beads of solder and lead clippings.
flux is for fags, the solder should come with enough

[Anonymous No.2585422]

Holy shit I downloaded whatever was the current version of KiCad recently and it turned out it was v5 and they are on v7 now? Wow. I didn't learn much anyway, should I just uninstall v5 and use v7? Is there any reason to use older releases?

[Anonymous No.2585429]

>>2581966
what are the rules of thumb when using a zener diode in reverse bias, the board I am working on is using a 10v zener diode to drop 35-40 volts down to something more tolerable for a 7805 regulator, the circuit the regulator is powering only draws about .03A but the regulator still gets quite hot. I'd like the change the 10v 1/2 watt zener to a 15v 1 watt. ultimately I am trying to improve reliability of this board by reducing the thermal cycling of the components.

[Anonymous No.2585434]

Why in the DC the second wire is called minus (-) if it's the ground? Confusing as fuck.

[Anonymous No.2585435]

>>2585434
Floating supply I'm guessing? See if it's shorted to earth

[Anonymous No.2585436]

So my boss is kind of pushing me to put together a solar system that I'm completely unqualified to do. I figured consulting 4chan is probably my best bet.

So I've got 2 100-watt 12v solar panels connected with Y-branch connectors. Wires are 12 AWG since the panels together only produce about 17 amps. 20-amp fuse on the positive between the panels and the charge controller.
20 amp charge controller is wired to two 12v 100ah batteries which are in parallel, again with 20 amp fuse and 12AWG wire.
Batteries then wired to 12v-120v inverter, again 20 amp fuse and 12awg wire.

Does any of this set off red flags? I figure popping a 20amp fuse on wire rated for 25amp limits my risk of setting something on fire pretty significantly. Planning on heatshrinking all splices of course. Anything I'm missing?
I'm trying to convince my boss to bring in a licensed sparky but it's like pulling teeth since he's convinced I know what I'm doing.

[Anonymous No.2585445]

>>2585429
The thing with zeners is they don't have internal temperature regulation like a linear regulator does. So I'd want to design it such that the zeners are under less thermal load than the 7805. If the 7805 is rated at 1.5W in its thermal configuration, and the zeners are rated 1W, then I'd want to have the zeners dissipating half the power of the 7805 at most. In your case I'd consider using two 10V zeners in series, spaced sufficiently.

Though honestly if you're dropping 40V at enough current to heat up a 7805 you're probably better off using a buck.

>>2585434
Voltage is relative brother. Sometimes there are bipolar power supplies, sometimes there aren't, but the convention of calling the most negative connection "V-" is pervasive regardless.

>>2585436
Looks good, only how much power is the inverter rated at? It isn't unusual for an inverter to be rated at 10 times the power rating of the solar setup, if just for peak loads like microwaves that only run for a dozen seconds. So having thicker wire and hence a larger fuse coming from the battery to the inverter is common. In your case I'd check what fuses the inverter uses, the shitty inverters I deal with have a bunch of parallel 35A blade fuses inside them, shoved really hard into some PCB-mount spade sockets. Go for something maybe 20% lower than the total current rating of your fuse(s), if there are any. Otherwise just calculate it from the peak power rating of the inverter.

Also use a heavy-duty crimp tool for the thick eye connectors, though if it's heat-resistant insulation you can solder them with a torch.

[Anonymous No.2585449]

>>2585435
> Floating supply I'm guessing? See if it's shorted to earth
>>2585445
> Voltage is relative brother. Sometimes there are bipolar power supplies, sometimes there aren't, but the convention of calling the most negative connection "V-" is pervasive regardless.
Well my logic goes like that.

1. The output voltage is 5V DC. If you measure the circuit it will show you 5V on the oscilloscope, there will be no negative on the display.
2. The output voltage is 230V AC. If you measure the circuit it will show you oscillation between 230V+ and 230V- as a sine, there is negative there.

[Anonymous No.2585458]

>>2585449
> 230V AC. If you measure the circuit it will show you oscillation between 230V+ and 230V- as a sine
false, it will oscillate between +/- 325V

[Anonymous No.2585462]

>>2585449
>The output voltage is 5V DC
You're just measuring +V, not -V. To measure -V you need to measure the more negative of the two outputs of the 5V power supply. But then what do you clip the ground clip of the oscilloscope to? Turns out, oscilloscopes are also relative!

[Anonymous No.2585463]

>>2585445
I have replaced the 7805 with a drop in switch mode type on one of the boards, the voltage at the input is at max allowable for ether the 7805 or the switcher. bodging in two zeners in the place of one will be ugly but not as ugly at the 2 watt 220 ohm resistor I have in there now.

[Anonymous No.2585466]

>>2585462
> You're just measuring +V, not -V. To measure -V you need
But that does not make sence. There's voltage of the circuit, right? There exist no voltage outside of a circuit. And if you measure that voltage it will be 5V flowing from one side to another in a loop. And if you subtract the measured voltage from you output voltage you get 5V - 5V = 0V. So the minus (-) should be technically zero (0) in DC, right?

[Anonymous No.2585470]

Is it KeeCad or KikeAd?

[Anonymous No.2585495]

here is the ugly bodge, I replaced a failed 1N4004 with a BYW56-TR. the extra lead length of the zeners help cooling. when the board is sleeping voltage at the regulator is 18v and when the board is running the voltage is 14v, thermals are much better now.

[Anonymous No.2585501]

>>2585470
Depends on your company.

>>2585495
>1N4004
I don't really understand. You're >>2585429, right? Are we talking about the glass ball case, or the two diodes soldered to each other?

[Anonymous No.2585503]

>>2585470
no

[Anonymous No.2585520]

What does it mean if a lithium ion battery drops in voltage when it's only connected across my voltmeter?
>battery removed from device
>measure ~3.8V initially
>decreases to like 2V over the course of 10 seconds or so
Is it fucked, or can I recharge it and keep using it? Does it matter if I put it in parallel with its healthy buddy for structural reasons?

[Anonymous No.2585524]

>>2585501
yes, I'm the guy asking about the zeners and has the 2 10v zeners soldered together, this board is on my bench because of a failed 1N4004. This is the control board for my gas furnace and it's replacement cost is $480, it developed an intermittent fault related to the 1N4004, but because it was intermittent I couldn't pinpoint the fault, but suspected a power supply issue.

I changed out a few garbage brand filter caps that had sketchy ESR and lower than marked capacitance, the fault still remained. I then suspected the 7805, it gets too hot to touch with the furnace is running, so I replaced it with a switcher, the fault still remained. After months the 1N4004 finely failed as a short. the 10v zener, 1n4004 and the 7805 are getting way to hot in my opinion and I'm pretty sure the 1n4004 was going to some sort of thermal runaway, so I'm doing a little rework to make the board more reliable.

bench testing the board, the parts I was concerned about are now just warm to the touch, at one point I measured the 1n4004 at 130c, this part is idle at 50c and then increases to 130c 50-60 times a day for 11 years, the zener diode and 7805 experienced similar thermal cycling.

[. !qaLOSTEwAs No.2585534]

>>2585470
kikecad, as there is no *good* FOSS eda software, but its been accepted as a "lesser of the two evils" sort of thing

[Anonymous No.2585547]

>>2585520
It's fucked. Even a 1MΩ input impedance on your meter shouldn't be enough to significantly drain a lithium ion cell, rather it's self-discharging instead. Take it out of any parallel pack since it will drain the surrounding cells too.

>>2585524
>130c
What the fuck. Even at 1V drop, it's 50C/W, so you'd be seeing 2W = 2A minimum. I'm guessing it had some sort of semiconductor failure that made the effective voltage drop even higher, not that it was getting up to 130C from the get-go. Nobody would design a 1A rectifier to eat 2A of current, unless it was for a fraction of a second.

>>2585534
Comparing to other paid software with FOSS alternatives out there, KiCAD does a pretty good job. Like, look at photoshop alternatives, or video editing or music making software.

[Anonymous No.2585553]

>>2585547
>It's fucked
Thanks

[Anonymous No.2585566]

>>2585524
Yeah, gas furnace boards aren’t engineered very well, probably like led light bulbs—run a bunch of components to their max tolerance. My board lasted almost exactly 10 years… one of the resistors had a big brown burn mark from getting way too hot… probably re-soldered itself every day.
I couldn’t fix the board, furnace guy came over and said I had to get a new furnace. Only $5k later, I had one.

[Anonymous No.2585617]

>>2581966
Whats a good hot air station?

[Anonymous No.2585635]

>>2585445
>Looks good
Nice
>only how much power is the inverter rated at?
About 600w, 5a at 120v. The only thing connected to it will be a couple of 20w 0.17a pumps that run for ~6 hours a day, clicking on and off from a float valve. Thank you for the heads up about that though--if he wants to connect something else to the inverter in the future, I'll definitely keep those short term peaks in mind.
>Also use a heavy-duty crimp tool for the thick eye connectors
Sounds like you just saved me a lot of trouble fiddling around with the wrong tool as well. Many thanks, anon.

[Anonymous No.2585636]

>>2585635
>5a at 120v
But 50A at 12V.
>a couple of 20w 0.17a pumps
Check the inrush current, that or just measure the DC resistance through them and calculate it that way. Factor of 10 it to check you're still in business. Also the inverter better be pure-sine.

[Anonymous No.2585652]

>>2582464
seconded

can't talk about specifics else I get in serious trouble but there are autists here that have no idea that their no bullshit, practical advice actually helped me to fix some life saving equipment and probably prevented the death or injury of dozens of people last year. nobody will ever know.

[Anonymous No.2585657]

>>2585617
i have an old Aoyue I like. Chinese clone of a hakko

[Anonymous No.2585660]

>>2585652
Agree, this thread helped me making electronics for devices that drop explosions on russians heads pretty much everyday. I'm proud of this, will definitely tell my grandchildren how my work was on the frontpage of r/CombatFootage. Thanks Ohm.

[Anonymous No.2585707]

>>2585534
EasyEDA seems pretty good, especially if you source parts from LCSC. Is there anything you guys don't like about it? I know it stores all or part of the schematics online which sucks but there are probably ways to store your work locally or convert it to some other tool and EasyEDA can export to lots of different formats. For example, if you run an unregistered version of Altium and cannot search for parts online, you can import schematics and footprints from EasyEDA and create shlib and pcblib.

[Anonymous No.2585710]

>>2585534
>>2585470
no
its Kie-Kad

Dunno why no one ever talks about gEDA. its awesome

[Anonymous No.2585711]

>>2585710
You're all wrong. It's pronounced kick-odd.

[Anonymous No.2585748]

i want to build a battery powered text terminal and use a display that looks reasonably good and is visible in the dark while using the least amount of power. Monochrome is fine, though I'd prefer a few colors. Is OLED with a black background efficient? What are my other options?

I realize LCDs are very efficient, but I dont want the black background to glow at all. Want to be reminded of a CRT. Maybe a dual layer LCD in negative mode would be adequate.

[Anonymous No.2585750]

What's everything I need to know before I replace a mouse cable?
They don't make the model I use anymore, and it's the model I've used pretty much my entire life on my home PCs.
It's a USB cable, and it seems those are pretty much interchangeable, but I don't know anything about soldering, flux, or cables in general.

[Anonymous No.2585755]

>>2585750
take it apart, the cable probably has a connector.
my mouse cable went bad, I went into my mice collection, and some other mouse had the same connector.

[Anonymous No.2585756]

>>2585755
What does that mean? Does it make my life easier?

[Anonymous No.2585788]

>>2585756
>What does that mean?

it means google ''youtube repair mouse cable''
watch the first million videos on the subject before asking any more questions.

[Anonymous No.2585789]

>>2585788
What's "google"? Do I have to solder it?

[Anonymous No.2585840]

Did my first real electronics repair after reading the stickys book on electronic repair.
Mini pinball machine that had intermittent sound issues.
Replaced the portion of the amplifier circuit that was failing with an LM386 and bodged it back together.
Works great!

[Anonymous No.2585841]

>>2585840
Good job anon, I could not do anything similar.

[Anonymous No.2585847]

Is there something similar to the boat digital network NMEA or car CAN-BUS network but for the house? I don't want to deal with the wireless ceiling lamps lol, I want a wired digital bus line with digital switches and devices on it.

[Anonymous No.2585848]

>>2581966
Stupid question regarding Ohm's law:
Let's say I have a 5V power supply and a Cob-leb, which has a max rating of 300mA and 3.3V. I therefor need a current limiting resistor. Following Ohm's law this leads to:a 17 Ohm resistor. (5V/17Ohm=294mA).
Now calculating the voltage drop we get:
17*0.294=4.998V
This means, that there is no voltage drop over the led? How is it supposed to operate?
Running it through LT spice, there seems to be a voltage drop of around 4.1V over the resistor and 0.91V over the LED, which brings back my question, how it is supposed to light up, when 3.3V are necessary.

[Anonymous No.2585863]

>>2585756
yes it means you unplug your old cable and plug the new one it. This is assuming you have parts mice. You can get used mice really cheaply, so it may make sense for you to buy some

[Anonymous No.2585864]

>>2585636
>50A at 12V.
Yup, but that's just the rating for the inverter, I don't expect to be pulling anywhere near that many amps from the batteries, ergo the 20amp fuse and 25 gauge wire. I figure if the 120v pumps are pulling 20 watts at .17 amps, then the 12v batteries are supplying 20 watts at 1.7 amps. Is this an accurate understanding, or am I missing something else important?
>Check the inrush current
Wow, thanks for mentioning this, I would never have anticipated that AC motors could have such high currents at startup.
On the plus side, even if it pulls 20 times the steady state amperage it shouldn't be a problem, right? That'd be 3.4 amps (since only one pump would be running at a time) less than half of what the inverter's rated for. I'll definitely be checking DC resistance to calculate before wiring anything up, though.

[Anonymous No.2585876]

>>2585848
Current limiting resistor is an unfortunate name from literal perspective. It would maybe make more sense to call it voltage reducing resistor. The goal of this design problem is to get from 5V to 3.3V at some specified current. Your resistor needs to "shed" 5-3.3V=1.7V at 300mA and not whole 5V assuming you want to run your led at 3.3V and 300mA. So you have
>source: 5V -> resistor eats 1.7V -> LED eats the rest (3.3V at 300mA)
And you don't want to run your LED at max ratings because it will get very hot if you don't have good cooling and it will reduce its lifespan and decrease your lumen output.

BAT54 is a schottky diode and not LED so it will obviously give "wrong" results. You can't just use any diode model you find in the simulator. If you want results for LED you need to use a model of LED. You're making a "lemonade" from potatoes and expecting it to have lemon flavor.

[Anonymous No.2585898]

>>2585876
>Your resistor needs to "shed" 5-3.3V=1.7V at 300mA
That makes sense. So I got to calculate (V_supply-V_LED)/I_LEDmax to get the proper resistance value.
So (5V-3.3V)/03A=5.67 Ohms.
Though, now I don't understand how a potentiometer would allow the dimming. If I increase the resistance, then the pot would eat more then the 1.7V and therefore preventing the LED to light up at all, because it no longer gets its necessary 3.3V. But from stupid testing with the led and a 10kOhm pot I could dim it just fine.

> If you want results for LED you need to use a model of LED.
Fair point. Though I don't understand the parameters in LT Spice. Looking at Iave, the NSPW500BS would fit, but the voltage does not. Is Iave even equivalent to the forward current and Vbrkdn to the forward voltage? Is there an additional library to add generic 1W cob leds or simple 3 or 5mm leds?

[Anonymous No.2585919]

>>2585898
>it no longer gets its necessary 3.3V.

as you see from these sexy curves, the LED doesnt suddenly come on at 3.30V
3.3 is a nominal value only.
to confuse the noobs.

[Anonymous No.2585920]

>>2585898
>So I got to calculate (V_supply-V_LED)/I_LEDmax to get the proper resistance value
Yes, if you want to run it at maximum current which is fine if you have good cooling. If you really want to get proper resistance value you'll need to find current vs. voltage graph of your LED and determine resistance graphically like in pic related. But this is only for that one specific LED at one specific temperature. In real world LED curve will move with temperature and due to manufacturing tolerances so there isn't one exact resistor value that would be the correct answer. Ideally you would use LED driver instead as it would keep the current constant regardless of LED, temperature and manufacturing tolerances, as with resistor you're taking a guess.
>Is Iave even equivalent to the forward current and Vbrkdn to the forward voltage?
Those values are there just for information purposes and aren't actually used in the model. Just in the UI for quick reference. What matters are parameters for diode equation
>Is - reverse saturation current
>Rs - series resistance of LED
>N - ideality factor
Other parameters can be ignored for basic applications.
>Is there an additional library to add generic 1W cob leds or simple 3 or 5mm leds?
Probably not because parameters will vary too much between manufacturers and different types. It's best to just skip simulations all together and use datasheets or if you really want to simulate, measure your LEDs and then fit the diode curve to the data.

It's too easy to fall into math/modelling/simulation rabbit hole. You can spend days on this or just test it in real world with real devices because that's what matters in the end. All you really need to know is
>how much power do I want to push to LED
>read current and voltage from I(V) curve at that power
>determine resistor value either by reading it from graph or with before mentioned equation
>make sure resistor can handle power dissipated on it

[Anonymous No.2585942]

>>2585898
you don't need LTspice, this is a simple calculation. The resistor needs to drop 1.7V at, say, 200mA.
V=IR
P=IV
therefore the resistance is R=1.7V/0.2A=8.5ohm
and it needs to be rated for at least P=1.7V*0.2A=0.3 watts.

So get a 6-10ohm resistor rated for 1W or more and you're golden.

[Anonymous No.2585946]

my printer died and I'm troubleshooting the cause. I need the feedback of those more experienced than me.
tl;dr I don't know how to continue troubleshooting the circuit without replacing all of the broken components I've encountered, and checking to see if it's still busted (which if it is, will just blow the replaced components)
What is the smart way to go about troubleshooting a circuit without blowing every component until it works?

More info: I started with the power supply board. The fuse directly from mains didn't have continuity, so I replaced it. Obviously though, if the fuse blew, replacing it won't just fix what caused it to blow in the first place. Of course it instantly blew again when I plugged it in.

So, I'm trying to test other components on the board. I notice two of the diodes in the bridge rectifier have continuity in both directions. Now, I can replace these diodes (and the fuse again), but this has no guarantee of fixing the problem and I really don't expect it to.

There's a large multitap transformer on the board, and if I've learned anything at all from Big Clive, it's that these things are fairly common failure sources. I'm testing it now.

[Anonymous No.2585958]

>>2585710
>Dunno why no one ever talks about gEDA. its awesome
Everything I saw suggested it didn't really improve on KiCAD except in reliability and ability to be automated.

LibrePCB is also another one to consider.

>>2585847
Ethernet

>>2585864
>120v pumps are pulling 20 watts
Oh they're tiny. Yeah everything checks out fine.

>>2585920
>load line
I guess that kind of digram can give you interesting results for extreme situations, like if your dropper resistor is dropping only 1% the total voltage. The angle of the load line would be far steeper, so any shifts in the forward voltage as a function of temperature would naturally cause big shifts in current. And those temperature forward voltage changes are somewhat easily classified via the Shockley diode equation.

>>2585946
Consider bodging yourself a milliohm meter from an LM317, passives, and a 9V battery and your multimeter. The current doesn't have to be accurate, so long as it can be used for comparative measurements it can be very handy for tracing shorts. Dim bulb tester is also worth considering, but honestly I'd avoid it for anything with an SMPS. Those off-line switching converters have bootstrapping circuits that aren't meant to be run continuously, and at a lower voltage they could be running in a constant brownout mode. Thermal camera is also an idea.

Main suspects for a blown fuse aren't really the transformer, rather they're electrolytic capacitors, diodes, and power transistors. Specifically the main initial switching transistors that switch the transformer would be my first inspection. Look for shit-stains (traces of magic smoke), darkening or rippling of the solder-mask from heat, also cracked solder joints. If all the obvious parts test fine then just reflowing a bunch of solder joints on heavy/stressed components can be a good idea to eliminate cracked joints.

[Anonymous No.2585979]

>>2585958
No skidmarks on the board, or any other obvious signs of heat. There's one TO220 transistor and it seems to have passed the test (but I did test it in circuit). There's also a TO220 silicon double-diode. Both of them passed (but again, tested in circuit). The transformer also passed. I'll start testing the rest of the diodes on the board, and also the electrolytic capcitors.

Prior to dying completely, the printer would power cycle whenever I plugged a usb cable into its usb port, which made me think of some sort of short-circuit scenario. But, I didn't find anything of the sort when I cracked open the digital logic board and started measuring when the thing was still running.

I unplugged it, then started investigating the power supply board. Nothing looked wrong, so I closed the thing back up. When I went to plug it back in, it was dead dead. That's when I re-opened it and found the blown fuse.

I can't imagine it's just the two bridge rectifier diodes dying causing all the trouble. That feels like it's just a symptom of something else.

[Anonymous No.2585988]

>relay datasheet says coil can handle 3 volts DC, resulting in a coil resistance of 20 ohms
>ok cool
>apply 3 volts to relay coil, no resistor
>it begins to smoke
what the FUCK
god I hope I didn't just fuck this thing

[Anonymous No.2585989]

>>2585958
>Oh they're tiny
Lol, not wrong
>everything checks out fine.
Nice, thanks again for your help. I really appreciate it.

[Anonymous No.2586025]

>>2585979
replacing the rectifier diodes fixed it. Or at least, it powers on again. The power-cycling-when-USB-plugged-in issue is still there. I have no idea what's causing that and I don't think /ohm/ can help there.

[Anonymous No.2586029]

>>2585988
You did fuck it. 3V across 20Ω is only like half a watt so I wouldn't expect that to smoke, chances are you had a resistance much lower. Measure once, cut twice.

>>2586025
Gratz.
>power-cycling-when-USB-plugged-in
As in, the printer turns off then turns on again when you first plug USB in? Or it constantly does so? Assuming it's not a software/firmware issue, disconnecting the power line from the USB connector might stop it freaking out. Does it happen when you plug in a laptop that's not plugged in? If not it may be a grounding issue that would be solved with a USB opto-isolator.
If it is a software/firmware issue (e.g. only happens when it hears data from the USB end) it will probably be impossible to fix unless there's a driver/firmware update you can install. Plug it into a USB power bank and see if it happens then also.

[Anonymous No.2586030]

>>2586025
Can your printer print a test page by itself through its menu?

[Anonymous No.2586040]

>>2586029
With no USB cable attached to the printer, the printer operates as normal (except, it's a USB-only printer so it just sits there looking pretty).
Once I plug the USB cable into the USB port, it immediately begins to power on and off about once a second. It didn't use to do this.
Interestingly enough, it does in fact still do it with a USB powerbank.

>>2586030
it can print diagnostics and test patterns just fine with no USB cable plugged in

[Anonymous No.2586043]

>>2586029
With no USB cable attached to the printer, the printer operates as normal (except, it's a USB-only printer so it just sits there looking pretty).
Once I plug the USB cable into the USB port, it immediately begins to power on and off about once a second. It didn't use to do this. I've tested this with multiple USB cables, and it happens with all of them. It happens when it's plugged into my laptop (unplugged or charging), or my desktop on a separate house sub-circuit
Interestingly enough, it does in fact still do it with a USB powerbank.

>>2586030
it can print diagnostics and test patterns just fine with no USB cable plugged in

[Anonymous No.2586054]

>>2586043
>with no USB cable plugged in

i know some people are unaware their printers can be driven thru wifi or ethernet.

[Anonymous No.2586056]

>>2586054
this is a really basic usb-only monochromatic printer that I use to print datasheets and pirated textbooks, it has no wireless capabilities. Eventually I was going to stick a raspberry pi with CUPS to make it network-attached but I haven't gotten around to it

[Anonymous No.2586058]

>>2586043
>it can print diagnostics and test patterns just fine with no USB cable plugged in
Look for a voltage regulator on the logic board and test it, or inject 5V with an external PSU at the regulator output pin.

[Anonymous No.2586064]

Preparing my capstone project for next semester, and I really want to create a synth like pic related. Do you guys think this would be a good topic or is it too hobbyist? What can I add to spice it up?

[Anonymous No.2586073]

>>2586043
>Interestingly enough, it does in fact still do it with a USB powerbank.
My guess is it just doesn't like the 5V coming into its USB port. Try getting a USB extension cable and cutting its +5V wire to see if you get the same problem, I'd hope the circuitry on the other end doesn't rely on that external 5V for anything but you'd have to check the schematic to be sure.

>>2586064
I think the only requirement is that you make the circuit design itself innovative or unique in some way. I'd consider force sensing on each key so you can do continuous note modulation after pressing a key, as opposed to conventional key velocity detection just working as an initial thing. A fully-functional minimoog made entirely on a single PCB (with capacitive touch keyboard) would be pretty neat too, though maybe a bit simplistic as far as hardware design goes.

Also audio effects using PLLs is an under-explored realm I'm telling you.

[Anonymous No.2586153]

>>2585876
> current limiting resistor = unfortunate name
No it isn’t, it’s called that because the LED will “run away” and destroy itself unless you limit the current with a voltage higher than 3.3v. It’s not a resistor: it will drop 3.3v. Period. Just like a 1n4001 will drop .7v.

[Anonymous No.2586158]

>>2586153
>It’s not a resistor: it will drop 3.3v. Period.
Well that assumes the forward voltage remains constant. Which it doesn't, hence the confusion.

[Anonymous No.2586161]

>>2586073
I'm guessing vocoders and octave doublers, what else uses PLLs?

[Anonymous No.2586176]

>>2586161
>Additive synthesis
Have a base triangle wave, going into a bunch of different PLLs each with a different divider, and you adjust the amplitude of each output before summing them all together. Hard mode is coming up with a way to vary phase also (I figured it out using two comparators, an inverting amplifier, and some digital logic) but according to some youtube yobbo you can't hear the difference anyhow.

>Harmonic lock
Using a type-1 phase comparator (i.e. an XOR gate) you find that the wave will only lock onto frequencies within ~±1.4 times the neutral frequency, before it snaps to some integer or fractional harmonic of the original wave instead. My idea is adding an AGC/compressor to the input wave (e.g. a guitar output), schmitt triggering it, and feeding the result into a bunch of PLLs with different centre frequencies. Each PLL output goes through a VCA such that the VCA output amplitude is the same as the input signal amplitude. Then each VCA output has a potentiometer on it for adjusting amplitude before a summing amplifier that adds it to the original signal (i.e. pre-AGC). As you change notes, the PLLs will lock onto different harmonics. As you slide/bend a note, the harmonics will jump about. Might be worth adding a manual pedal for adjusting the centre frequencies of them all on the fly.

>Harmonic e-bow
As above, but instead of multiple PLLs feeding back into the guitar output, you just have one (maybe two) feeding a magnetic field directly into the strings. A lever on the e-bow would adjust the centre frequency.

PLL output wave-shaping filters/diodes are probably a good idea for all of the above.

[Anonymous No.2586179]

>>2586176
Oh here's the diagram I made. Totally not schizobabble I assure you. Each CD4046 would have a different integrating capacitor value to set the offset centre frequencies, and the same-sized potentiometer as a timing resistor. The passive filters are illustrative, and may be better replaced with multipole active filters. As for the VCAs, I'd probably use the diode-based one Moritz Klein used in his latest video on a compressor, it looks really elegant. Though OTAs are also an option. Moritz Klein's channel is a fucking goldmine.

[Anonymous No.2586186]

>>2586153
It's still confusing for beginners because they then assume that only resistor determines current in the circuit. You have to look at this from perspective of someone who has no previous exposure to this. It would be better to call it operating point resistance or something similar, because then they have to learn about diode I(V) curves. Just calling it current limiting resistor omits a lot of useful information, because it's a term used by people who already know why that resistor is there for in the first place.

[Anonymous No.2586207]

>>2581966
>just ordered a Spartan 6 FPGA dev board
What am I in for boys? I can program in C (STM32, AVR, 8051/8052, nRF52, ESP32, ESP8266), python, HTML+Javascript, Java, batch and shell scripts and wrote a few things back in VB6 days. I used to make redstone circuits in minecraft back in 2012, full adders, other shit etc.

[Anonymous No.2586276]

>>2586207
building logic by using lookup tables is what youre in for. hard mode: draw out the logic hardware and see how close you are to the software implementation

[Anonymous No.2586367]

>>2586186
>because then they have to learn about diode I(V) curves.
and get intimidated and drop the field forever lmao
between 1mA and 1A, diode voltages change by less than 10%, so ON-OFF model is leagues better for a beginner

[Anonymous No.2586425]

>>2585919
> graph showing white leds at 100 mA
Are these supposed to 20 mA LEDs?
As far as I know, from a semiconductor physics perspective, there are no white LEDs so I’m not sure what’s up with that.

>>2585876
> don't want to run your LED at max ratings
I remember TI said, decades ago, when the were making LEDs, that they were still testing them for longevity, and as near as the could tell, after 50 years they might get slightly dimmer.

Indeed, I have lots of LEDs that have been running for decades, and every LED ever made could theoretically still be in service today. The whole frequent failure aspect of LED light bulbs is completely engineered.

[Anonymous No.2586439]

>>2586158
> assuming voltage remains constant
We’re talking about more typical uses, I think. Most people are going to drive their circuit from some kind of regulated supply or equivalent (e.g battery)
Back in the day it was common to see LEDs, usually the original green ones, as low noise vRefs
Because of their astonishingly high voltage drop.

[Anonymous No.2586453]

>>2586186
>>2586367
> confusing and intimidating
I get that, but we might want to make this decade the one where we dispel the myth that “all loads are resistive in nature”

Even my kid took “batteries and bulbs” a few years ago in high school, just like I did. Trying to explain semiconductors was a nightmare.

I don’t see why current regulation is necessarily more difficult, conceptually, than voltage regulation.

Even the last household lighting generation—fluorescents—needed to be current regulated, I don’t see why we have to pretend that semiconductors don’t exist in modern times. Incandescent lightbulbs have I/V curves too, except the coefficient goes the other way.

[Anonymous No.2586457]

>>2586367
If you get scared by a curved line then electronics might not be the best hobby for you. Exponential functions are high-school level math and I wouldn't be surprised if kids already learn about them in elementary school nowadays (unless this has been replaced by some in depth course about gender theory or something).

[Anonymous No.2586462]

Why do you need drivers for LEDs?

[Anonymous No.2586467]

>>2586462
Either for efficiency if using switch mode drivers or stability to ensure constant current through the device and thus minimize light output variations.

[Anonymous No.2586482]

>>2586462
you dont

[Anonymous No.2586484]

>>2586457
You can do plenty of interesting projects without ever having to face the fact that diodes are logarithmic. It's a niche topic that can be safely ignored until it's needed.

[Anonymous No.2586490]

would this work correctly for data transfer and everything?

[Anonymous No.2586505]

>>2586484
But you will encounter it as soon as you read the LED datasheet. You will have to read the I(V) plot at some point, so why pretend that it doesn't exist? It's not that hard and LEDs aren't the only device with similar characteristics. What if you want to play with thermistors? There you have exponential again. Or even simple resistor capacitor circuit where you again have 1-exp(-t/RC) step response. Exponential function is literally everywhere, even in AC circuits you can use euler's formula to represent sines as exponentials. This then also expands to fourier and laplace transforms because you can represent arbitrary circuit response with exponential that has a complex argument which also happens to solve differential equations of circuits with reactive elements... Do you have ringing on your signal lines? Guess what: it decays exponentially. Want to know how current gets distributed due to skin effect at higher frequencies? Take a guess (hint: it's exponential). This thing is literally everywhere and the sooner you learn it the better.

[Anonymous No.2586508]

>>2586462
efficiency. you can drive an LED with a current limiting resistor or a linear regulator and generate waste heat... or you can use a driver, which is a current limited switching power supply.

[Anonymous No.2586520]

>>2586508
switching power supplies are cancer

[Anonymous No.2586521]

>>2586505
who tf reads LED datasheets wtflmao

[Anonymous No.2586522]

are there any budget (under $200) LCR meters that will do up to 100khz and have input protection so shit won't fry it I plug in a charged capacitor by mistake? DE-5000 looks great, aside from no input protection.

[Anonymous No.2586535]

Okay, it might sound stupid, but I have to ask and I would do almost anything to achieve desired result.

It's almost impossible hard to see the navigation display and my phones display in the direct sunlight on a bike even though phone has an AMOLED. I'd like to have a second screen, which duplicates my phone screen. The second screen should be in E-ink technology. How can I do that?

[Anonymous No.2586537]

>>2586535
Alternatively not the display clone, but Android Auto with E-ink.

[Anonymous No.2586566]

>>2586484
>oh I’ll just hook up this 3V LED to this 3V power supply
>hey how come it died?

>>2586490
Probably, I’d wrap it in some alfoil for shielding though.

>>2586521
>oh I’ll just hook up this LED to 50mA to make it brighter
>hey how come it died?

>>2586522
You could consider making your own.

>>2586535
Polarised filters for the screen or sunglasses/visor doesn’t do the trick?

[Anonymous No.2586601]

>>2586566
>>hey how come it died?
sounds like you didnt read the LED datasheet

[Anonymous No.2586678]

>>2586566
> Polarised filters for the screen or sunglasses/visor doesn’t do the trick?
No, here is an example of OLED with the filter. I found e-ink smartphones instead that cost $200-$400 and will run any navigation app I want with decent refresh rate.

[Anonymous No.2586681]

transisters... why are voltage-controlled resistors so simple in concept but so difficult to realize in practice

[Anonymous No.2586682]

>>2585847
SPI is 10m.

1-wire protocol can work up to 200m. But I don't know how well MCU to MCU 1-wire is going to hold up at that range.

You could use RS-232 at 5v (MCUs have built-in UARTs) and and some kind of line conditioning on Rx. But doing peer to peer networking over RS-232 is none-trivial.

[Anonymous No.2586696]

>>2586681
> vcr
A mosfet is probably a closer representative of a “voltage controlled resistor” assuming you meant bjt.

[Anonymous No.2586700]

>>2586535
E-ink displays are very limited, and slow.

Easiest way would be to get some low-ish end mcu that can drive your e-ink display, and write an app that screenshots every second and sends it to your mcu-eink combo once per second.
Of course, you want to downconvert the colors and resolution on your phone app before sending.
Also it will be easier if your phone is configured in accessible modes for visually impaired users.

[Anonymous No.2586706]

>>2585847
If you don’t want ethernet or rs232, there is something in between… both old school and robust, fast yet simple. Fairly long distance… hundreds of feet.
RS 422/RS 485.

[Anonymous No.2586709]

>>2586462
> why drivers for LEDs
Depends on what you mean

Lots of digital outputs don’t have enough current to drive a LED, and even if they did, they might only be able to drive one at a time. So usually your driver is a transistor or a chip like the ULN2003.

If you’re talking about lighting, it’s usually to control and limit the current (also called “ballasts”) or fancier dimming drivers that use pulse width modulation because they’re too lazy, stupid and/or ignorant to read the transistor datasheet’s current gain curve providing the output.

[Anonymous No.2586711]

>>2586678
Ah it's not just glare, it's full-on brightness output that's the limiting factor. Yeah e-ink is basically the only way around that. Or reflective LCD like a calculator, but I don't think anyone makes displays like that for phones and the like. Though you might be able to connect a generic display driver to one that can be run from a DVI port or whatever, then use USB-C adatpors.

[Anonymous No.2586716]

>>2586681
I've seen BJT's used, idk how accurate they are tho

[Anonymous No.2586747]

What do i use to cover about 50-100 hot LED on a single squere aluminium pcb?
Some kind of varnish? LEDs in question have a lenses already so i thought, if there is a proper varnish for the job, to somehow cover lenses and apply it.

[Anonymous No.2586752]

>>2586681
Because the main things that can be used to vary impedance as a function of voltage are semicondcutors, which by nature are rather nonlinear. Unless you want to use a triode. I'll run through a bunch of options I've found.
Resistances:
>JFET: reasonably linear, though their input voltage ranges are rather limited to preserve proper biasing, somewhat expensive, the control voltage is referenced to one/both legs of the JFET
>MOSFET: not linear at all, can only act as a resistor in one direction unless at very low amplitude, the control voltage is referenced to one leg of the MOSFET
>OTA: An operational transconductance amplifier can be used to make a virtual variable resistance that floats, rather linear and the control voltage can be dereferenced from the "resistor" terminals, but these circuits are complex and not cheap, and OTAs are barely sold anymore
>diode string: a diode string needs to be biased from either end with nontrivial external circuitry, and is effectively only usable as an impedance with one leg at ground potential, somewhat linear
>BJT gilbert cell: black magic, may or may not be usable as an impedance
>Vactrol: slow to respond and bulky, but simple, somewhat linear, and there's full isolation between control voltage and resistor terminals, not RoHS
>Photo-FET: a possibly obsolete, more expensive, monolithic, faster version of the vactrol
Capacitances:
>diode: not trivial to bias but not as difficult as a diode string resistance, is basically always a capacitance to ground
>MLCC: high-capacity dielectrics of modern ceramic capacitors display a significant voltage-dependant capacitance, arguably better than using a diode
Inductances:
>variably saturable inductor pair: idk i saw an image of this once, looks neat for easily tuning an oscillator or filter, and there's complete isolation between input and output, and you don't get input-output signal coupling

There's also servo versions of typical manual variable impedances.

[Anonymous No.2586760]

this is the funniest fucking thing i've seen today

[Anonymous No.2586770]

>>2581966
what awg wire should i buy

[Anonymous No.2586805]

>>2586770
3 strand solid braid where awg=3*pi*radius^2=14

[Anonymous No.2586820]

Why is Altium Designer so retarded? I tried a simple thing, put some parts on the PCB then moved the mouse and the entire fucking board scrolled out of the window and I can't find it anymore. I am sure there are some settings and hot keys to bring it back but why is it the default behavior ???

[Anonymous No.2586825]

>>2586820
lmao imagine being filtered by your mouse

[Anonymous No.2586864]

Why is a bridge rectifier better than alternating square waves for converting AC to DC?

[Anonymous No.2586881]

>>2586864
wtf r u talking about

[Anonymous No.2586896]

>>2586881
Well if you have two square waves which are offset by 180 degrees then you have a flat line in the voltage which is essentially dc no?

[Anonymous No.2586898]

>>2586896
Yes, but where will you get your square waves from? Only sines come out of the wall socket.

[Anonymous No.2586906]

Hello, I am retarded and am wondering if these two 220Ω rnet resistors are kill. I had to remove them and the headers were a bit wonky on one and askew on the other. Should I just use a new pair? Also, why aren't they available on ali?

[Anonymous No.2586908]

>>2586881
>>2586896

[Anonymous No.2586914]

>>2586898
Well you convert them to square waves.

[Anonymous No.2586920]

>>2586864
>>2586881
I think it means half bridge.
It’s because the full bridge fills in the “missing” alternating square wave, which is easier to smooth out with a capacitor.

Also, these are sine waves, which, again is smoother than a square wave resulting in less noise and spikes in the output.

[Anonymous No.2586923]

>>2586747
> cover hot leds
The only thing i can think of would be clear, pourable silicone.
You’ve got heat, silicone is good at that, and used to enphosphorize the LED.
Seems like you’re gonna have a lot of thickness as well.
Do a test run, silicone is weird stuff.

[Anonymous No.2586927]

>>2586176
Thanks for the ideas, I'm letting go of the idea of making a synth that sounds appealing and rather find ideas that make it more interesting on design level. thanks for the idea, I have also been looking at FM synths and wondering if there are resources into the creating them. It feels like a walled garden with how difficult and obscure they are

[Anonymous No.2586933]

>>2586906
Sometimes they are custom jobs from large mfg runs.
You can always make your own on a sip header and some discrete resistors.
Just test them with a $5 harbor freight ohm meter, or make one with a led, 5v usb power supply/9v battery and possibly another resistor.

[Anonymous No.2586971]

will tying a knot in a usb cable make any appreciable affect on noise

[Anonymous No.2586998]

am i doing this right? i don't really understand how the current flows through the FS312. looking at other schematics and the FS312 data sheet hasn't made it any clearer -- i can't tell if i'm missing a ground connection or if i have an erroneous ground connection.

basically i want to USB charge two 18650 batteries in parallel. while USB charging, cut the load. i want the 18650 batteries protected against over/under charge and over current.

[Anonymous No.2587001]

>>2586998
Erroneous, you've shorted the FETs together. Also there should be a connection to the bottom of the cells somewhere, since the FS312 probably doesn't carry the majority of the battery current; that's what the FETs are for.

[Anonymous No.2587005]

>>2587001
this is what the data sheet shows. the FETs are connected? but it doesn't make sense to me in any case.

[Anonymous No.2587020]

>>2586923
Looks almost as that i was looking for, thanks a lot.

[Anonymous No.2587045]

>>2587005
i think it's right, now that i found this. https://electronics.stackexchange.com/questions/446526/how-does-the-dw01a-prevent-overcharging-and-discharging

i'll have to test it with some current limiting resistors. i don't want to cause an 18650 meltdown.

[Anonymous No.2587053]

>>2587045
Looking good
Add a 1Meg pulldown to Q1 gate or it will start floating and turn off even when the charger isn't connected

[Anonymous No.2587055]

>>2586896
> have two .. waves which are offset by 180 degrees then you have ... dc
no, you have the exact opposite of DC

>>2586914
why would you convert them to square waves?

I think before this conversation can continue, you need to answer the following questions:

what are the min and max voltages (peak to peak ) of an AC wall outlet?
What are the min and max voltages of a rectified AC wall outlet?

[Anonymous No.2587056]

>>2587053
good suggestion, i forgot to consider floating gates. also my Q1 is backwards.

[Anonymous No.2587066]

>>2586770
stranded 24 for cables
solid 24 for breadboards
enameled 30 for bodges and winding toroids
enameled 24 for winding toroids

[Anonymous No.2587067]

>>2586914
You can also convert them to rectified half waves.

This discussion can continue ad nauseam.

[Anonymous No.2587092]

>>2587045
I meant you have a GND node on either side of the two FETs. That connects the two together, and means even if the FETs are turned off current will still flow from one side to the other. You need that wire connecting the FET source on the OD pin to the negative terminal of the cell.

Pic related is the correct way. Don't get confused by the BATT+ and BATT- terminology, those are the output terminals.

Also the DW01's minimum voltage cutoff is like 2.5V, the 2.9V on this FS312F-G is slightly better but it's still best not to rely on it.

[Anonymous No.2587101]

>>2587067
anon never said full wave or half wave. he said simply "bridge rectifier". you are bewilderingly interpreting:

"bridge rectifier" to mean "full wave bridge rectifier"
and
"alternating square wave" to mean "half wave bridge rectifier".

I have no idea what led you to this conclusion

[Anonymous No.2587106]

>>2587092
>Also the DW01's minimum voltage cutoff is like 2.5V, the 2.9V on this FS312F-G is slightly better but it's still best not to rely on it.

are there any cheap chinese parts that have a better termination voltage? i chose FS312 because it's better than DW01 but, like you said, it's not a whole lot better.

[Anonymous No.2587118]

>>2587106
If LCSC actually had a good parametric search tool I'd have found one years ago. But they don't. Maybe there's a good enough low-power RRIO CMOS comparator and low-power voltage reference out there to do the job. Or use a TI BMS IC like you should.

[Anonymous No.2587181]

>>2586971
if anything it will make it worse because a knot is a 1 turn coil

[Anonymous No.2587182]

>>2586971
USB is differential, digital and shielded. It is not affected by noise. Or rather, you would be hard pressed to generate enough EM noise to affect USB signal.

[Anonymous No.2587183]

>>2587182
youre making the bold assumption that all usb cables are shielded

[Anonymous No.2587185]

>>2587183
Fair enough.

[Anonymous No.2587225]

>>2587181
>a knot is a 1 turn coil
You get a clockwise turn of the ingoing current, and an anticlockwise turn of the outgoing current, so they cancel out. It's differential signalling.

[Anonymous No.2587245]

>>2586752
BJTs are used as variable impedances for controlling the pitch of an oscillator. The usable range is a few hundred millivolts. From my feeble understanding the response is exponential, so you can actually tune them to real pitches.

[Anonymous No.2587246]

>>2587245
Could you show a diagram? All I can imagine them is acting like diodes in a diode string, or as a variable current sink like in a variable integrator vco.

[Anonymous No.2587263]

>>2587246
Doesn't get any simpler than this. Send maybe 5V through a pot set up as a voltage divider into the base and there's your variable resistor. Tuning requires considerably more circuitry and you need a gain-matched pnp somewhere to counter the temperature component.

[Anonymous No.2587264]

>>2587263
don't forget the current limiting resistor to the base like I did!

[Anonymous No.2587279]

>>2587263
Yeah that's being used as a variable current sink. If you try to use that "voltage controlled resistor" as one of two resistors in a voltage divider for an audio signal, you'l find it distorts heavily. Plus it has bad crosstalk (e.g. a fast edge on the control voltage will make it through to the output), though that's a feature of all monopolar voltage controlled impedances to varying degrees.

[Anonymous No.2587295]

>>2587279
It works tho. You can plug a sequencer in, tune the offset and scale and it will work.

[Anonymous No.2587298]

>>2587295
It works for a VCO because distortion literally doesn't matter for a schmitt oscillator. It doesn't work as a generic voltage controlled impedance, as you'd want for a VCA or VCF.

[Anonymous No.2587398]

How to safely solder in a studio apartment? What us the safest flux + solder combo

[Anonymous No.2587410]

>>2587398
you're over thinking it. unless you're running a commercial PCB soldering operating out of your apartment, you'll be fine. you're exposed to more harmful stuff just sitting in traffic.

get a fan to blow the smoke out of your face if it bothers you.

[Anonymous No.2587425]

>>2587398
What's with you zoomers and obsession with safety? If a bit of smoke scares you picked a wrong hobby. Maybe try knitting instead? Or would that be also too unsafe, because needles are involved? Based on how you sound I wouldn't be surprised if you cook in a lab coat with full PPE and fill up 10 work permits before you even turn on the stove.

[Anonymous No.2587427]

>>2587425
>zoomer
I'm 34 and I'd like to live for another 40 years

[Anonymous No.2587437]

>>2587427
You should be worrying about other potential hazards way more than a little flux smoke. Unless you're working 40h/week in a factory soldering PCBs a little exposure won't do anything to you. You have way more chances that you will be severely injured or even die as a result of traffic accident and you don't obsess over safety every time you sit in a car.
>but muh long term effects
Unless you're some health expert vegan that watches out for everything you're literally consuming class 2A carcinogen every time you eat steak. Not to mention all the processed shit that goes along with it. Soldering fumes are nothing compared to what you're already exposed to on daily basis.

[Anonymous No.2587444]

>>2587045
USB 2 specification is 5 V @ 500 mA maximum. USB 3 specification is 5 V @ 1.5 A maximum. and obviously there's all the new power delivery standards that require ICs to do handshakes and other crazy shit to negotiate the proper voltages and current limits.

does USB 3 (5 V, 1.5 A) require a handshake or an IC?

or can i just swap out the USB connector on my schematic for a USB 3 connector, plug it into an off the shelf """USB 3""" charger, and have no issues negotiating and utilizing the full 1.5A (limited by the TP4056's 1A maximum)?

[Anonymous No.2587455]

>>2587444
okay never mind, i found this very helpful information.

https://lygte-info.dk/info/USBinfo%20UK.html

tying 200R between D+ and D- should allow for operation in DCP (dedicated charging port) up to 1.5A.

[Anonymous No.2587480]

>>2587437
>you don't obsess over safety every time you sit in a car.
But I do, I'm afraid to drive over 25mph and long distances.

[Anonymous No.2587501]

>>2587425
the smoke from flux causes cancer the same as cigarettes. your example of a needle prick is retarded, because that doesnt entail dying a slow painful death as your body rots away, and it is also easy to avoid pricking yourself with a needle, whereas without some sort of precaution you will inevitably inhale flux smoke while soldering.

you seem like the same type of person who smokes and claims some sort of superiority over people who vape

[Anonymous No.2587511]

>>2587501
You're talking like people are burning through spools of solder every day like some addicts or something. If you could actually manage to get cancer from soldering a few boards every few months there would be all sorts of warning labels and regulations before you could even purchase flux or solder. Flux fumes can be harmful if you inhale them 8h/day for 40 years, but no hobbyist is doing that.

[Anonymous No.2587589]

>>2587511
any amount of smoke elevates your risk of cancer. cancer is the number 2 leading cause of natural death. one way to avoid dying of cancer is by avoiding exposure to carcinogens

[Anonymous No.2587605]

>>2587589
If you go outside and breathe in air, especially in cities you would be at much greater risk due to car exhaust gasses, tire dust, chimney smoke... You're severely over-blowing risk of flux smoke and underestimating risk of cancers from food (guess what is number one leading cause of death) and other chemicals that you come into contact every day. You will get cancer sooner from sausages and ham than from flux fumes.

[Anonymous No.2587615]

>>2587398
Keep windows open while soldering.

[Anonymous No.2587617]

>>2587589
>cities
avoid

>car exhaust gasses, tire dust, chimney smoke...
avoid, avoid, avoid

>chemicals that you come into contact every day.
avoid

sausages and ham
avoid and avoid

[Anonymous No.2587621]

>>2587617
You can't avoid giant cancer ball in the sky and invisible high energy cosmic rays.

[Anonymous No.2587628]

Did you ever need to convert schematics made in LTSpice or any other spice to an EDA tool?

[Anonymous No.2587650]

>>2587589
number 1 leading cause of death is cringe

[Anonymous No.2587680]

>>2581966
How do you fix RAM? I've got a bunch of dead sticks, lastly I killed a pair of 2 8gigs. I don't read any short, I believe the little controller chip in the middle got fucked. These 2 ones I killed with a bad driver installed from Driver Booster. I got hit with a blue screen trying to restart then it stopped booting. Changed rams with a test stick and when I restarted again this got fucked too. The first 4 sticks I killed with another mobo which I bought broken trying to fix it, I found that it had around 20-30 ohms short on the ram VCC mosfets. But it had the same symptoms, ram were dying on restarts. Wth? Anyway, do you think the chip in pic related is at fault?

[Anonymous No.2587681]

>>2587617
I don't think you are making the point you seem to be trying to make.

[Anonymous No.2587685]

>>2587680
If it's just that 8-pin chip, then you could probably fix it with a temperature controlled hot air station and some foil tape and patience. If you can find a replacement that is, read the part number and see if it's sold on Digi-Key. But I don't have any idea if it's what caused the fault.

[Anonymous No.2587722]

Why are LEDs flickering? They should be supplied with constant DC power.

[Anonymous No.2587723]

>>2587722
Are they being overdriven? Is there any ripple on the DC rail?

[Anonymous No.2587726]

>>2587722
Maybe they're driven with PWM?

[Anonymous No.2587737]

>>2587722
> They should be supplied with constant DC power.
it is very common for them not to be

[Anonymous No.2587755]

>>2587681
you can die at 68 of cancer or you can die at 88 of heart failure

ill take the extra 20 years

[Anonymous No.2587759]

>>2587755
You're going to die in fewer than 10 years due to:
global thermonuclear war
more bioweapons/chem
starvation
exposure
dehydration
animal attack
accidental

Did you know every time a forest burns all of the animals and people downwind get cancer 100% of the time? Greta Toonberg told me in study hall.

[Anonymous No.2587774]

>>2587759
>global thermonuclear war
>more bioweapons/chem
>starvation
>exposure
>dehydration
>animal attack
>accidental

i avoid all these things

[Anonymous No.2587778]

>>2587774
lmao

[Anonymous No.2587793]

What the hell is
>pic related
?

I've had this in my junk drawer for years.
It has three legs (one of which appears to be grounded to the component's chassis) so I assume it's some sort of transistor, but the little window on it throws me off; I thought it might be a LCD but it doesn't look like any LCD I've ever seen.
The only markings on it are "200BP" and "314."

[Anonymous No.2587796]

>>2587793
IR pyroelectric sensor

[Anonymous No.2587799]

>>2587425
What's with Boomers and their obsession with ignoring safety, getting debilitating injuries and forcing everyone else to pay for their health care?

[Anonymous No.2587806]

>>2587796
Thanks!

[Anonymous No.2587824]

How good are these?

[Anonymous No.2587860]

>>2587722
inb4 breadboard

[Anonymous No.2587866]

>>2587722
E26 "LED" bulbs have shitty capacitive dropper PSUs.

[Anonymous No.2587879]

>>2587866
>E26 "LED" bulbs have shitty capacitive dropper PSUs.

so what?
you cant see 120Hz unless you shake your head like a dog drying itself off.
which i'm doing right now to prove the point to myself.

>>2587824
>How good are these?

they are good for one thing only.
which is protecting your bench from burn marks.
if you're a generalist and you're fabricating, testing, troubleshooting, repairing, then it just gets in the way.
what's best is nice clean table whose surface you can mar wo incurring the wrath of someone with objectophilia.

[Anonymous No.2587891]

>>2587879
Key point here is "shitty."

[Anonymous No.2587928]

>>2587879
>you cant see 120Hz unless you shake your head like a dog drying itself off.
wrong, all you have to do is be looking to the left of the light, then look to the right across the light. this is a common thing animals with movable eyeballs do.

>>2587866
not all of them. the nicer ones dont

[Anonymous No.2587933]

>>2587928
>the nicer ones dont
Yes. And they also don't flicker.

[Anonymous No.2587979]

i fucking hate gay hass BGAs, you can never see if it soldered correctly so all you can do is turn it on and hope it doesn't start to smoke

[Anonymous No.2587981]

Not sure if this is a good place to ask, but could you recommend any book that explain how electronical components work on the smallest level? Or maybe one of the book in the OP already does it?

[Anonymous No.2587990]

>>2587979
Reflow plate or oven is the way to go for them.

>>2587981
What do you mean by "smallest level"? Like the transistor level? Logic gate level? Block level? Ultimately there are a variety of levels of abstraction, each with different amounts of information for you depending on what you're doing. Usually the block diagram says most of what you need to know, but occasionally you'll need to see the actual logic pathway or even a transistor diagram in order to understand how a circuit will behave. Either way much of the rest of the datasheet is still required reading.

If you're after a fairly technical dive into components and how to construct circuits with them, then the Art of Electronics is probably worth a shot. If it's too dense then try one of the other books. You'll also greatly benefit from practicing them on a breadboard, even using a simulator like falstad is a lot better than nothing, though it can be a lot harder to know when something is wrong since parts don't start smoking in a simulator.

[Anonymous No.2587992]

>>2587981
Nobody actually knows. All current physics models are just approximation of reality.

[Anonymous No.2587993]

>>2587990
I meant the atomic level.

[Anonymous No.2588005]

>>2587993
The bulk semiconductor model is as deep as we usually ever go, like talking about how the body diode arises on a MOSFET from the PN junction between the body and drain, or how a diode's depletion region acts to block current in only one direction.
The atomic scale is literally useless if you want to use that information to make circuits, though I assume you're up for it anyhow. The atomic scale for semiconductors means you have to learn about how the phenomenon of conduction/valence bands of electrons arises in a bulk semiconductor, which isn't trivial at all. You basically have to do Monte Carlo simulations with the electron energy states to see what bands they form at different temperatures. Try a condensed matter physics textbook, but it's some the densest most incomprehensible quantum bullshit out there, and probably not immediately applicable to actual semiconductors (e.g. field effect and PN junctions). A specialty textbook on how semiconductor physics arises from the properties of single atoms (predicted, as opposed to the measured properties of bulk semiconductors) is going to be a uni textbook for someone who has completed 3rd year quantum and 3rd year condensed matter, so for an honours degree if not masters.

For a solely qualitative understanding there are a few less technical resources (like some of Alpha Phoenix's semiconductor crystal growing videos) but I've never seen one any video or popsci article even a hint of the transition between individual statistical properties to useful bulk properties. Even understanding resistors isn't trivial, the resistivity of a bulk solid falls out of the same kind of statistical mechanics that semiconductor physics does, though it's a lot less technical and could probably be conveyed qualitatively. Inductance and capacitance can basically be fully understood from Maxwell's Equations, though if you want to understand permeability/permittivity then that's an atomic level phenomenon a high-schooler could get.

[Anonymous No.2588015]

If i have small dc motor and i hold the shaft so it can't spin, will that damage it?

[Anonymous No.2588016]

>>2588015
Why don't you try and find out?

[Anonymous No.2588017]

>>2588016
because i don't want to break it, are you retarded or something?

[Anonymous No.2588019]

>>2588017
Then don't hold the shaft.

[Anonymous No.2588021]

>>2588019
thanks for nothing, i hope there are other people here smarter than you who can answer my question

[Anonymous No.2588032]

>>2588015
not as far as i know, but it might overheat faster

[Anonymous No.2588094]

>>2588015
It can only withstand a certain amount of mechanical resistance before the coil gets hot enough to burn the enamel away and short out. Stalling for a short time is fine, but if it stalls long enough it'll burn.

[Anonymous No.2588111]

I have a rgb led, common anode, rated current is 20mA, do i need to calculate the resistor for each color?

For example, for the red cathode, i need to put a 3.3 - 2 / 20mA resistor?

[Anonymous No.2588117]

>>2588111
>need to calculate the resistor for each color?

yes.
then you make small adjustments so they all have equal (apparent) brightness.

[Anonymous No.2588136]

>>2588111
no
i use 1k resistor globally for all my leds and colors
it works just fine

[Anonymous No.2588154]

>>2588136
even RGB LEDs?

[Anonymous No.2588190]

>>2588154
It will "work" but you won't get equal nor optimal brightness from from each colour.

[Anonymous No.2588223]

What is this orange copper crap on the battery?

[Anonymous No.2588226]

>>2588223
looks like it was stuck to copper foil

[Anonymous No.2588238]

>>2588117
should i take the typical voltage values from the datasheet? I'm going to use 3.3V to power it, but the Vf for blue/green is 2.9 min, 3.1 typ and 3.6 max.

[Anonymous No.2588253]

If a board says something like "10 analog inputs (16 bit ADC)", does that mean I have full 16 bit for every input or 16 bits shared between the inputs?

[Anonymous No.2588258]

>>2588111
>>2588238
If you feed 20mA into all three diodes at once, it will heat up more than a single LED in the same package, which may not be allowable. I'd check the datasheet for the maximum continuous power dissipation, and maybe rules on powering LEDs simultaneously.

I would use the typical values, though you may have to change the resistors anyhow, green and blue often look a lot brighter than red.

>>2588253
It just means there's one ADC connected by multiplexers. Actual multi-ADC chips are quite expensive, but do allow for simultaneous sampling.

[Anonymous No.2588267]

>>2588258
> It just means there's one ADC connected by multiplexers. Actual multi-ADC chips are quite expensive, but do allow for simultaneous sampling.
But this does not answer my question.
16 bits resolution means I'll get 0-65k values. I connect 9 axis gyro+accelerometer, magnetometer, barometer and thermometer. All together 12 readings. Does that mean I get 65k / 12 = 5.4k resolution for each?

[Anonymous No.2588268]

>>2588258
https://www.tme.eu/Document/f27dc74a4d1d935ac4d97f68b20521a4/OSTAMAA132A.pdf

> I'd check the datasheet for the maximum continuous power dissipation
Can it be the pulse forward current?

[Anonymous No.2588291]

>>2588267
No it connects a single 16-bit ADC to one pin at a time, so you get the full 2^16 = 65536 count on each value.

>>2588268
No, has to be continuous if you're having them be on for more than a second or so.

[Anonymous No.2588318]

>>2588291
then the datasheet says nothing

[Anonymous No.2588383]

>>2588318
Absolute Maximum Rating
If = 30mA / 30mA
Ifp = 100mA

I take that to mean your Max rating is 90mA and it can handle 100mA for 10ms.

[Anonymous No.2588395]

What is the cheapest good soldering station with hot air gun? I'm poor.

[Anonymous No.2588419]

>>2588383
I'll be making each for 20mA so i shouldn't have a any problem. Thanks.

[Anonymous No.2588435]

>>2587990
>Reflow plate or oven is the way to go for them.
Still doesn't tell you if any pins got bridged or got insufficient paste or whatever.

[Anonymous No.2588456]

>>2588419
Ok turns out you can’t even find a K/W rating on any thru-hole LED datasheet anyhow. Go with your gut, it it gets too hot then double the resistances.

>>2588435
Yeah not much you can do besides send it to someone for x-ray examination. Or shove it up your ass and get the TSA to do it for free.

[Anonymous No.2588476]

>>2588456
most basic bitch LED's have a max current of 20-30mA. They're dirt cheap so might as well just slowly turn up a pot until one of them pops.

[Anonymous No.2588555]

>>2587680
>>2587685
That chip is a 128 byte I2C EEPROM (probably 24C01) containing the specs of the ram stick so the computer knows how to set it up. All the necessary pins go out to the edge, but you have to read another one just like it to know what to write. Only 4 connections matter, but you can bodge them with tiny wires and use a larger chip.

[Anonymous No.2588556]

>>2587793
>>2587796
>>2587806
The thing you don't know (unless there's a part number you can find) is whether it's analog or serial digital. If it's really old it might be analog.

[Anonymous No.2588557]

>>2587879
>protecting your bench from burn marks
I use a cheap scrap piece of plywood, and that's on top of one of those green cutting board things that I found used. If it gets burnt it won't smell as bad as plastic or rubber.
Think of Big Clive's board only shittier.

[Anonymous No.2588631]

>>2588476
Yeah but when you have three LEDs in one package, can that package take the heat of running three LEDs all at 20mA? Or will the LED dies get above their maximum operating temperature?

[Anonymous No.2588634]

>>2588633
>>2588633
>>2588633
New Thread