/diy/ - Do It Yourself

Hi /ohm/,

I recently started getting into electronics more seriously. I understand the components and the general gist of a circuit, but whenever I see a circuit with a lot of nodes and parallel connections, for instance in BigClive videos, it takes me a lot of effort to understand how the current flows and why the circuit behaves the way it does.

When you guys see a circuit, how do you know what's going on? Do you have some tricks which help you get your head around it?

>>2020656
mostly i would say its where the positive and negative of the circuit are and path of least resistance but this isnt absolute and wont help in some cases with a oscillator. even simulators fail in some cases with these so then it comes down to experience but i would like to say that base line stuff without any funky oscilator are straight forward. positive and ground potential dictate direction with resistance . caps tend to be used for timing and dc blocking purposes. transistors open pathways. the combination allows for simple delay circuits and latches

more complicated things rely on potential changes and can also act as delay circuits and oscillators so then its a observation of the flow in its initial state vs over time. most talk about math

How the FUCK do you solder these little fuckers?
No matter what I do they're always slightly crooked.

>>2020679
I put them in the breadboard on the two sides of the pcb, one side which I'm soldering, and the other for even support.

Can some of you call me a retard so that I don't have impossibly high expectations of myself when I learn how to analyze circuits? All of this looks hard from the textbook.

>>2020679
I use one or more of these steps:
- If possible put then in a breadboard and then the board on top with some support on the other side to make it even
- I solder the two ends first and tweak the inclination by reflowing those pins, you have to be careful tho because this tends to melt the plastic quite easily
- Sometimes using some bluetac or play-dough in the center helps and combined with the second tip you can almost always have perfect alignment

>>2020746
It takes a lot of time to learn complex subjects, retard. You'll never know everything. Don't beat yourself up.

I am building an anemometer as part of a small home weather station, just trying to work around what would be the path of least resistance without losing accuracy for determining the number of rotations, and then working that back to speed.

Considering a small AC motor, where I work around the waves produced, but also looked at using reed switches and hall sensors as a possibility. Any ideas here? This seems to be a general problem about counting rotations.

>>2020757
Do you think you could gut an old mouse and use the scroll wheel sensor on a low-power/standby circuit similar to a tachometer?

>>2020757
You don't really need a motor unless you want to read the voltage levels it creates from spinning.
Ideally you'll have the part that spins and some sort of sensor to track how fast its spinning. A magnet+hall effect sensor can be used with a timer to calculate a full rotation.

>>2020768
Interesting, I forgot about tachometers, I do see some mentions of hall effect in them.

>>2020772
I'll double down on the hall for now then, and see what works, cheers!

The other idea was a light sensor through a slit in the bottom of the rotating shaft, but I have a feeling that has a tendency to need more fixing over use.

>>2020757
hot wire anemometer

>>2020782
this is the easiest anemometer you can make and one of the most precise. Seconding this

>>2020782
>>2020791
I considered this, but it's added complexity and visual appeal is something I want.

Is making a basic digital clock hard? Apparently, the final project of my lab is to make a digital clock.

>>2020820
Depends on what you're allowed to use

>>2020820
>buy real time IC
>program uC to display hour in a lcd or seven segment display and add basic set/reset/adjust functions
If you can order a pcb made then it is piss easy, if not then you will have a rats nest

Lads, I have a queston on TAOE exercise 3.4 a) (pic). Before going with Igate = 1A I want to 'redo' what the book did (Igate = 1mA implies 50us switching delay). With all that I get fucking 2us, far from 50 us so I don't know what the fuck should I do.

>>2020672
Wouldn't current spread throughout a circuit based on the resistance of the different paths?

How do you then approach this split? Do you try to find the resistance and voltage drop through each element?

>>2020847
you are looking at it wrong. There are very organic and intricate circuit where the voltage at every node matters (an example is the joule thief) and there are the most common "modular" circuits. You learn the little blocks and how they operate and build something functional out of them. (That is why block diagrams are useful).

>>2020656
After seeing so many circuits and the way components are used in them, I just recognize 'modules'. Like I'll spy an active component, say an op amp, and just look at what's around it. Sticking with the op amp example, you're typically always going to have the same style feedback network around it and filtering caps. You see a bunch of beefy transistors and a few inductors and capacitors near by? You've discovered the power supply module. It also helps my discovery process that almost all chips of a certain function typically have the same pinout, if not a very similar one which makes the path easy to follow without getting any kind of icky mathematics involved until you want to pin down actual performance characteristics. Hope that helps

>>2020853
designing things in the organic, integrated way has the advantage of (sometimes) using less parts, being faster or more efficient but generally is harder to understand and build. The modular approach is the bread and butter of design. It also fits neatly in the concept of a signal path and lends itself easily to analysis (specially filters and amps)

I made a relay NOT gate!
Could have made the wires shorter but this will do.

>>2020679
I align them with several stacks of perfboards and put them upside down

>>2020884
Looks cute, good stuff.

>>2020884
why tho

is using a single resistor in the common pin of a 7-segment display a good idea or should I use 1 resistor per led?

>>2020679
>How the FUCK do you solder these little fuckers?
>No matter what I do they're always slightly crooked.
I just re solder the first joint and push it flush against the board then it stays in place while I solder the rest.

what is this white fabric-like covering on the wires inside heating appliances? what purpose does it serve

>>2020953
It's fiberglass insulation for wires that get too hot for plastic insulation.

>>2020954
so it still allows heat to get through though, right?

>>2020842
Please, you're my only friends :(

>>2020957
It's usually on conductors near a heating element or an arc lamp. It just insulates the wire that delivers the current. Plastic insulation melts/rots/crumbles if the temp around it is hot all the time.

>>2020939
There is a YouTube channel called Clickety Clack.
It has been abandoned for 2+ years now and the owner (who has an active Reddit account that I stalked) does not seem to be planning to resume it.
If he doesn't do it, someone else has to.
I believe the world needs a good quality general-public oriented relay computing YouTube channel.
I also like digital electronics, I built a few redstone circuits in Minecraft when I was a teenager.
And I already invested almost 1000€ into this and won't hesitate to double or triple that.
Also I've given up on game development dreams so something has to replace that sunken ship.

>>2020972
Go for it buddy, don't stop because somebody else already did it, you'll find your own spin to put on things.

>Game dev
Every one I know who went down this road is now miserable, maybe you can dabble in restoring arcade machines.

>>2020842
Yeah I got 2µs too. Bring it up with the prof, or just leave it and assume the book is wrong.

>>2020972
Good luck anon!

>>2021002
>Every one I know who went down this road is now miserable
I still want to do so eventually. Because I want to make a game where you actually do engineering as a significant part of it. None of that subnautica-tier playing around. Designing working 3D models of engines, firearms, even circuits, that are simulated and characterised.
The physics would be very akin to real life, but with significant differences so people don't just copy things from this plane of existence. Say, having to make firearms from pellets that rapidly absorb gas when triggered.
Also there's floating islands, because lightning strikes this wacky dielectric in the ground. Lots of funky capacitors.

>>2020972
good luck bro. But a relay not gate is just a Nonimally closed relay I guess.

fuck weller rt tips are expensive
Why haven't the chinks cloned these yet?

>>2021048
Because they're too busy making money off T12 clones.

>>2021048
They don't know the secrets.
Weller tip = 5 years
chink tip = 5 minutes

>>2021044
And you should learn ladder

>>2021050
Even 5 minutes would last me years. And it'd cost $3 or so.

>>2020943
>should I use 1 resistor per led?

of course, otherwise your "1" will be much brighter than your "8" when the current thru each segment will be 3.5x less.

>>2021054
Have you ever used a real iron? Night & day difference.

>>2021067
No. I don't solder much

>>2020972
Make a relay-only 6502. This will take a long time, cost lots of money, and provide you with video material autists will watch religiously. To increase viewership, find hot g/f to be your "lady fractic" who occasionally helps, people will scan every video looking for a glimpse of her.

>>2021080
how do i find a hot gf? :(

>>2021080
Ah yes you see, although there's only one flaw in your plan, it's quite a major one...

>>2021092
Rent one from instagram or whatever the whores use these days.

>>2021097
but that would be a false love

If I've a weak PWM signal that's amplitude is between ground and 1V, and I set up a BJT so 1V will drive it into saturation, I assume I can easily convert it to a PWM signal with 5V amplitude assuming it's frequency isn't too high?
Also, what would be the best way to convert the 5V PWM signal to an analog voltage between 0 and 5V?

>>2021098
Who the hell said anything about love? lmao

>>2021099
>I assume I can easily convert it to a PWM signal with 5V amplitude assuming it's frequency isn't too high
Yes. With a simple BJT inverter topology. Though the duty-cycle will be inverted.
>the best way to convert the 5V PWM signal to an analog voltage between 0 and 5V
A low-pass filter, with a corner frequency significantly below that of the PWM frequency. If you need fast response rate and can't increase the PWM frequency that much, you may want more than a simple RC filter. Chaining 2-pole Sallen-Key filters together is an acceptable solution, keeps your output impedance low too.

>>2021104
>A low-pass filter, with a corner frequency significantly below that of the PWM frequency
So if I'm dealing with a 3KHz square wave, like on the order of hundreds of hertz?

>>2021114
Yeah, though it depends on what kind of ripple you’re fine with.

>>2021120
Cool.
I've basically a controller that sends a 1V PWM signal to an IGBT driver that changes it to a +/-15V signal to drive an IGBT. The current flowing in the circuit is then measured by a current sensor that feeds it back to the controller between 0 and 5V.
I'm basically creating a test scenario so the PWM is changed into an analog voltage between 0 and 5 and fed back to the controller so it thinks the current is increasing in line with the PWM signal it sends out.
I can always throw a small trimmer pot into the low pass filter to fine tune it.
Thanks for the help

I'm considering making this SMD instead of THT, can power supply filtering capacitors (like C8) be SMD?
It wouldn't be a sensible idea to make C22 (in the audio signal) a SMD cap right?

>>2020636
Asked on /g/ too, might get nore visibility here

I found an old HP 82913a monitor in my uni's tech recycling. From what I can gather it was coupled to HP 86 computers back in the mid 80s. The power/brightness was in the on position, so I'm prettt sure something is broken with it

Do you have any resources on how to fix it and perhaps get it working with a modern PC? I'd really like to tinker and get it working again

>>2021140
you could make C22 smd. Just use smd electrolytic for C8.

>>2021145
I have absolutely no experience with SMD and I was just basing my assumption off this post.

>>2021143
Go watch Arcade Jason's CRT repair videos.

>>2021143
It might just be the pot that went bad or got cold solder joints.

>>2021148
How would I go on testing this? I was thinking of buyinf a 75ohm-hdmi converter, but the monitor looks like one of those old black-green ones. I'n not sure it will display whatever W10 shows it

>>2021143
See if you have power and do a continuity check. I don't know anything about those and I generally am shit at electrical stuff, but I know that cathode ray stuff usually lasts a super long time. I would bet it's the transformer or a blown cap.

>>2021065
ok thanks. Is there a way to multiplex 4 displays with two pins? (without buying an expensive mux ICs)

>>2021152
>HP 82913a
If the inputs are screw terminals, connect a RF modulator with a 75-ohm adapter. If the input is an RCA connector, use a RCA patch cable between monitor and composite source.

>>2021146
Yeah, you're not gonna get a 1 microfarad NPO.
You could try an X7R cap or just stick with through hole for that and go with a film.
C8 you can get SMT aluminium caps, just make sure you make the pads stick out further than the component or they're a pain in the bollox to hand solder

>>2021152
http://www.hpmuseum.net/collection_document.php

>>2021154

a 74139, or CMOS equivalent, is pretty cheap.
or make your own from a bunch of gates.

>>2021152
>How would I go on testing this?

it appears to use regular old TV signals, in 2 colors. so, you can connect the video input to an NTSC source like a DVD player, a video card with NTSC or S-Video outputs (yellow RCA jack or 4 pin DIN), or an adapter, like the VGA-to-NTSC one in the pic.

>>2021158
>>2021153
>>2021148
Good news for the screen: It does turn on, although there's a noticeable really high pitched sound. I'll get some RCA cables and see if I can get anything to show on it

>>2021202
Something you could do right now (after discharging the tube) is check for bulging caps and cold solder joints, especially at the flyback transformer's pins.

>>2021162
Thanks anon! Will do.

>>2021100
your cynism will be your downfall

>>2021287
Ok, Socrates. Life is a downfall.

How long can you send 5v down a wire before voltage drop becomes an issue?

I want to measure resistance between two points in some soil but I don't want to have to have an MCU every 10 feet or whatever to monitor the entire garden, since that would just be retarded and costly for my size.

>>2021301
E = I*R
Multiply the current draw of the circuit by the resistance of the conductor. The gauge and type of wire are variables. Bigger conductor = less resistance.

>>2021301
>resistance between two points in some soil
Since that resistance will be so high, the wire resistance (and related V drop) would be negligible.
Electrical noise would be the main issue, I'd think

Fucking crystal pull oscillators.

I need to pull a crystal to the chroma subcarier of an NTSC signal. My circuit is basically this but with an analog variable capacitor and a phase comparator driven varactor diode.

I've got a lot of NTSC equipment and can subcarrier lock to about half of it. Everything else is outside the pull range of the crystal. I ordered every single 3.579545MHz crystal on Digikey hoping one would have a wide enough pull range, but they all only work with a subset of my test equipment.

tl/dr
How do I increase the pull range of a crystal? There must be some trick I'm missing.

Fun data sheet to read:
http://pdf.datasheetcatalog.com/datasheet/analogdevices/40671345AD724_b.pdf

I need help with a problem.
I have a heater that I a controlling with an arduino and a relay. However I would like to dynamically adjust the voltage to have more control of the total heat output via arduino.

Would anyone know a device that I could do this with? ~30 amps

I think it should be possible to controll a DC/DC buck converter with a PWM signal generated from the arduino but I don't know of a device that I could do this with. I've been looking and looking and I haven't found any digitally controllable buck converters

>>2021361
PID controller

>>2021361
>dynamically adjust the voltage to have more control of the total heat

you can just PWM the relay, which will shorten it's life, but still work for years if you do it slowly. heaters have a lot of thermal inertia which will smooth out the variations.
or replace relay by a solid state relay.

Anonymous 02/03/21(Wed)21:38:09 No.2021399▶
>>2021361
>dynamically adjust the voltage to have more control of the total heat

you can just PWM the relay, which will shorten its life, but still work for years if you do it slowly. heaters have a lot of thermal inertia which will smooth out the variations over seconds.
or replace relay by a solid state relay, and PWM it as fast as you like.

>>2021361
>dynamically adjust the voltage to have more control of the total heat

you can just PWM the relay, which will shorten its life, but still work for years if you do it slowly. heaters have a lot of thermal inertia which will smooth out the variations over seconds.
or replace relay by a solid state relay, and PWM it as fast as you like.

>https://www.solar-electric.com/learning-center/electrical-characteristics-awg-copper-wire.html/
>https://www.hflitzwire.com/frequency-as-per-awg-size/
Which one is right?
For 100 kHz one says I should use 38 AWG, and the other say I should use 28 AWG wire at minimum.
Are there several AWG standards?

>>2021417
Or just a MOSFET. Unless it’s 30A AC, in which case: what the fuck.

A dedicated PID controller will be the easiest method, but there’s nothing too wrong with implementing a PID algorithm on a micro, and adding a FET gate driver circuit/IC and thermocouple measuring IC and display/control hardware.

To be honest, I’m wanting to make an MCU dev-board with a gate driver, a couple of onboard op-amps, a nice ADC, maybe a small display and a rotary encoder too, in order to use it for PID stuff. DSP and motor driving too. MOSFETs of choice are to be soldered in by the user.
Could do all sorts of stuff with it, including a class-D amp, I think.

>>2021422
Fuck, I mean 26 instead of 28 AWG

>>2021422
>Are there several AWG standards?
No, but certain people of a yellow hue label shit wrong on purpose to swindle the gweilo.

>>2021146
that poster is a retard.
X5R and X7R ceramics do change over temperature but unless you're living in the arctic or Sahara you're fine. you can also design around it if you know what your operating temperature range is (add more caps).
They also have lower ESR so filtering can be better.

>>2021338
Personally, I'd use a clock generator like the SI5351 instead. That or pick a crystal that evenly divides into what you want. Could arguably use a 4046 or other PLL with one crystal and use some FFs/shift registers on both inputs of the phase comparator to divide and multiply to the right value, but that's just a more difficult way of doing what the 5351 already does.

>>2021301
>>2021336
Also, add a low pass filter to mitigate the intrerfearance.
Software wise, take samples over a long period of time, throw out the higs and lows, and average the rest

Post your soil measurements and specs plz.
Thought about doing this myself

>>2020656
kirchoff, learn it and understand
thevenin might help
understand those and you will be at peace with the electrons

>>2021569
don't forget millman's theorem
everyone forgets millman's theorem

>>2021301
twisted pair my nigga

>>2021645
TP is only good for signals or any kind of changing wave like AC. TP allows you to send differential signals down the pair so that the magnetic fields work in tandem.
Sending a DC voltage down a TP is pointless.

>>2021146
>Opinion post
That person is an idiot and factually wrong.

>>2020884
Nice, I assume you printed the case too?
What kind of inserts are you using for the screws to hold the top on?

>>2021718
>TP is only good for signals or any kind of changing wave like AC.
gee I wonder why they use them to connect digital stuff like micro controllers....
And the guy who asked the question, in industry lots of sensors use CURRENT signals because 10mA are 10mA no matter the distance. The current circuit is super simple to make.

What are my choices for micro soldering irons? Something with a short working distance because I have shaky hands.

>>2021734
not about iron but something that has helped me a lot was buying surgical pincers. They are usually stainless, dirt cheap and better than things for electronics.

i put a smoothing cap on a rectifier too high now i get 2-5 volts above what i'm supposed to get
does the choice of cap also depend upon the load, this is for a power supply btw so it won't be the same load everytime

>>2021771
does not make sense

>>2021776
converting 6volt ac to 6volt dc but cap value too high so i get around 8volts but that's just from the load of the multimeter
wouldn't the cap discharge much faster with higher load so i will get voltage than what i get now

>>2021778
Caps don't magically increase voltage. Attach a load and measure the output again.

>>2021722
Yes. Used square nuts.

>>2021778

when you use a proper cap, the DC voltage will be 1.41 times the rms voltage, so 8V is all proper and correct. as you add a bigger and bigger load, the ripple will cause the average DC voltage to appear to be lower, but the peaks will still be 8V.

so, it's a dumb idea to use a smaller cap, coz then you get a dirty ripply voltage. instead, you can drop the voltage with 3 series diodes to give you 6V.

Do you need (does a 50ohm cable exist? ) a 50ohm cable to connect a 50ohm coax (antenna) to my input pin? Or is a standard cable enough to connect my input pin to my antenna?

>>2021908

I need to make a proof of concept for a university dissertation which im assuming will be quite simple but i have never done anything electronics related so im here for help.
Pretty much i just need a little board that can receive a bluetooth signal from a mobile phone or a wifi connection and when it receives a ping / flag it will light up an LED.
I'm a software major and only doing this little addon as my supervisor really has a boner for "wearable tech".
Any idea what i should be buying?

>>2021908
Yeah, you need a 50 ohm cable or you'll have problems.

>>2021912
I know sma connectors but i have an esp and the WiFi is shit. So i thought that i can put my esp in a case and lay the cable outside the box. But the "normal" copper cable, not sure, have a different resistance. So does the signal even transmit to the esp, if i have normal copper in series with coax.
Or is it better to solder the coax direct to the pin.

>>2021936
The signal will be fucked up due to impedance mismatch and reflections. Try it and see for yourself.

>>2021945
Well, i tried,with no improvment. So i thought i soldered something wrong etc. But after checked everything, i doesnt found an error. The last error source was the coax series with normal copper.
Before destroy everything, i thought i ask diy, if they can confirm, that the copper in series with coax is shit. Like you guys did, so ty.

>>2021956
you need to match the impedance of everything when dealing with high frequency stuff and transmission lines

>>2021913
books

>>2020679
I may be dumb but I stick them onto the board with blue tack, do two or three pins, and then take the blue tack off before soldering the rest of the pins in.

>>2021913
esp32

>>2021913
Easy mode would be the arduino nano 33. It’s a super simple microcontroller with Bluetooth on board. It’s like $20. Just download the arduino IDE, check out the Bluetooth examples and TADA you’re done. If you need help, there are a brazilion instructional videos on YouTube.

so I'm fixing SONY walkman WM-DD10,
I got the mechanism moving, seems all fine, but it plays tapes super silent and polluted with some noise. Head is clean and the tape is fine.
I tries some switch combinations and it plays it slightly louder with dolby noise reduction on but it's still to silent for normal listening and once volume goes past 6 out of 10 on the potentiometer noise becomes LOUD distored buzzing. All components on the board look fairly OK, but I have not any reference to compare to board against and there seems to be one spot where a component could be but isn't, not sure if it's missing or the board was used for other walkmans with different components and it was never filled to begin with.
I'm pretty competent with mechanics but I'm completely green with electronics.
Mechanism and it's functions work, switches do their things and yet the sound is fucked.
what do?

>>2022009
Replace the volume potentiometer.

>>2022012
I think it's worth mentioning it's still smooth and doesn't crackle at all, it's just as if the zero was far below where it's supposed to be.
I don't have a replacement handy, I was thinking of first trying maybe a contact cleaner on it once I bough it this weekend.

>>2021908
receive only? just plug into input.
otherwise, a 9:1 balun is most commonly recommended (free space, ladder lines, and wire antennas tend to be 450ohm +/- 150ohms)

>>2022018
Contact cleaner is fine but those little pots wear out mechanically, especially after 35 years of use. You could remove it and test the resistance while you turn the pot to know for sure.

>>2021734
use your pinky+ring finger to anchor your hand
take a deep breath and exhale before s̶h̶o̶o̶t̶i̶n̶g̶ soldering

>>2022027
I'm already doing that

>>2022018
It's also possible that there are cold solder joints somewhere. Use magnification to examine the board.

>>2022018
>as if the zero was far below where it's supposed to be.
as in, negative resistance? negative voltage? Only time I've seen that crazy shit happen is when the ground trace gets broken/disconnected.

>>2022034
>>2022030
>>2022025
>>2022012
this is the board, and this is the spot where there seems to be something missing, but I can't find any reference that would tell me what was supposed to be here, if anything at all

>>2022047
It's hard to see the solder on the pads, but if you have a 220 uF through-hole cap, you can tack it on until you get a SMD replacement (if that's the culprit). The negative leg on the cap connects to the "AUDIO" label on the edge of the board.

>>2022055
>connects to the "AUDIO" label
*connects to the pad near the label

>>2022047

def looks like a cap was knocked off the PCB.

>>2021569
>>2021570
>>2020855
>>2020853
>>2020672

Thanks! Very clear

>>2022063
others of this size on this board seem to have L1 220 4v written on them, so it seems like that is the missing part
how does it play at all without it tho?

>>2022074
ok, now I've found it, seems the missing cap is the problem
I'm completely green with electronics, do I have to hunt down this exact cap or a modern replacement will suffice as long as it says 220 4v on it?

>>2022080
4V minimum, but the capacitance rating should be 220 uF. You can go higher on the voltage rating. It won't hurt your board.

>>2022080
>>2022085
and if you're ordering a SMD replacement, pay attention to the physical dimensions/footprint.

If i measure the ac voltage between the output of a 15v charger and the ground pin in the socket that its connected to it measures 77 V. Why is that?

>>2022130
Chargers output DC voltage, and you have a short between the AC input and DC output.

>>2022089
>>2022085
found exact same ones and ordered a couple
I suppose I'll fix the pads before they come in the mail, I wonder if I'll be able to pull it off with a regular non regulated iron

>>2022150
Sure you can. Practice on perfboard or old broken stuff first. Keep your tip clean and tinned and don't heat the pads any longer than it takes to melt the solder. You'll be fine.

>>2022150
>a regular non regulated iron

you've been brainwashed. temp-regulated irons are 100% a meme shared by those who know no better. regular irons are regulated by your eyes and hands in a feedback loop. when the solder melts, you remove the heat. that's regulation.

>>2022130
if the charger has no ground pin, what you're reading is a phantom voltage. if you place a 1M resistor across your meter, said voltage will disappear. coz it was never there to begin with.
if it does have a ground pin, then you're probably getting a tiny, harmless bit of current thru one of the noise-reducing caps.

>>2022158
>temp-regulated irons are 100% a meme shared by those who know no better. regular irons are regulated by your eyes and hands in a feedback loop. when the solder melts, you remove the heat. that's regulation.
I actually have no problem accepting this since this is how I worked so far, except I've never done SMDs

>>2022162
>I've never done SMDs

the peculiarity of SMDs isnt about heat but access. since they're meant to be soldered with hot air, there's often no good way to access the pins.

>>2022167
Lucky for him that SMD electrolytics are among the easier components to solder with an iron, not to mention that the old cap is already removed from the board. He's already half-way to completion.

>>2022172
well more like torn from the board
if I can fix the pads and preload with solder the one harder to access it should be pretty easy to attach new one
I just hope the actual pads aren't torn from the board and it was just a solder failure

>>2021969
Does an capacitor in series disturbs the signal? From the capacitor the wires are not 50ohm so... Futhermore is the signal with or without the dielectric better?

Not sure if this belongs here, /3DPG/ or /EMT, but here goes:
I'm working on a big CNC machine that's heavily dated, and we want to convert it for 3D-printing big stuff. The machine ran, sorta, a decade ago, using servo motors and Electrocraft BRU-200 drives. This gives me two options:
>Yank out the expensive servos and drives, throw in fat steppers, hook up to a 3D-printing motherboard, bob's your uncle
>Somehow find a way to get a 3D-printing board (RAMPS, Duet, whatever) to talk to the servo drives
What would you consider to be the most feasible option? We've gotten the bastard to move agian, but only barely and without the encoders. Is there any easy way to hook up ancient servo drives to modern 3DP boards?

>>2022177
There's always an option to bodge it by soldering jumper wires from the cap contacts to the nearest component in the trace path, if it comes to that You can keep it in place with a drop of hot glue or another temporary adhesive.

>>2022183
That's pretty retarded.
A big CNC has to move a ton of mass back and forth. It can't even go as fast as some cheap chink 3d printer.

>>2022180
Dude just match the fucking cables

>>2022183
>I want to take a good tool and turn it into a shit tool
How about turning it into a decent modern CNC?

>>2022191
>>2022196
Didn't say it was a CNC mill, just a CNC. Which it is. Was. May have been. I don't need to go fast with a huge nozzle anyways.

>>2022158
>never there to begin with

dunno about that. I just put the output connector on the wall socket ground pin (scandinavian "clip"type on the sides) and it arcs ever so slightly

>>2022200
and by ground i mean earth

>>2022089
The copper was delaminated, he's not going to be able to easily solder on a replacement of the same footprint.

>>2022177
>if I can fix the pads
Best shot would be adding something like copper tape to the board, then soldering jumper wires from the tape pieces to the traces going to the cap. But whatever you do, it won't be very mechanically stable, so you'll want to adhere the cap in place. I'd personally just glue a THT electrolytic cap there upside-down and bend the leads to the relevant traces. Then cover the whole assmebly with heat-shrink.

>>2022183
I'd see if you can get the servos running properly, its not like people don't diy CNC machines using servos. What sort of interface do the servo drives use? It's possible that they don't need any kind of extra board at all, and just connect directly to a master PC via ethernet or RS232 or something.

/ohm/, I am replacing fans on my 3d printer (Duplicator 6). The current fans are 24v 0.1 amp. Other folks have soldered together two Noctua 4010 fans (12v 0.05a).

I have a 12v 0.15a blower fan that I would prefer to use, but am guessing that splicing it together with a Noctua 4010 @ 12v 0.05a is going to cause some problems. Could anyone offer a little guidance? Thank you!

>>2022202
Servos use RS232 but aren't operating properly it seems. If we break into that system the encoder part seems to be gone, if you feed the a 0 to +10V signal they'll start moving like a stepper. Since my resident sparky is not gonna be able to finish the job, and I don't understand the system enough (maybe neither does he) I want to get the project back on track. Even if I could connect to a PC, it's doubtful the drives could understand .GCode suitable for a 3D-printer.

>>2022202
I don't think it's delaminated, I scratched it a little and there is shiny silver metal under the grime and oxidation
I just need to clean it in a non destructive fashion before tinning the pads - I don't have any tools for that so tips are welcome

>>2022200

presumably your Y cap is a bit leaky, then.

>>2022212
Just leave it as is. When you get your cap, put it on top of the solder and touch the iron to one side so it sinks into the solder, then the other side. Add more solder after if needed.

>>2022214
The connector to mains only has connection to live or neutral so i guess any cap wpuld be connected between them and to earth

>>2022217
>solder
>solder
>solder
lmao I need to slow down.

>>2022217
if I know anything is that you should never try to attach anything to a dirty pad

>>2022221
oh sorry i missed that

>>2022222

>>2022224
The pad is covered with solder. Use a little flux if you want it nice and shiny.
>>2022227
noice

>>2022204
Are the fans modulated or running full speed all the time? If the latter, wire two 12V fans in series. If the former, they'll make noise and the speed control won't work properly.

>>2022236

Thanks for asking - both are simple ground/vcc 2 cable fans, so their speed is controlled by the voltage that they receive. So, both run at 12v, but have different amperage (12v 0.05a and 12v 0.15a - original fan is 24v 0.1a)

>>2022210
>the encoder part seems to be gone
Fug. I'd still look into getting new servos though (or possibly repairing the existing ones), since doing so may be cheaper than getting steppers and their drivers.

>>2022212
Oh really? Maybe it was just a cold solder joint, would be the best case scenario if so. Try to reflow it with a tiny bit of solder (or just flux).

>>2022258
I was wrong. Don't wire them in series. Use a 24V fan.

>>2022236
>>2022277
well it might work if you put caps (and probably ~15V zener diodes) across them

>>2022222
based 2n2222 quints

>>2021724
>current signals
Interesting. That sounds alot more accurate than having to account for voltage drop, Is there a simple enough circuit I could wire up that could take in amps and spit out a voltage I could feed into one of the arduinos analog ports?

>>2021567
I'll be using my own electrodes out of a couple of finish nails and a 3d printed spacer, so I'm not sure if it'll translate well over to your setup.

Hey bros, why does this setup cost literal bank? My lab class just started today and wow, each of these equipment costs at least $1200 each, is there any way I can get them for at least half of the price each? I kind of want to have my own setup at home rather than risk my grandparents. I know that there is a power supply there but would I even be able to handle that with a regular outlet at home? Sorry if this is a dumb question...

>>2022513
Buy:
>$10 chink logic analyser
>$10 chink function generator
>$10 chink PSU adapter for RTX PSU
>$50 2nd hand CRT oscilloscope

There are a variety of choices for all of those, look about on the EEVblog channel for his reviews of cheap function generators and PSUs. A logic analyser isn't necessary, but helps a lot for doing digital electronics.

>>2022183
Is something wrong with the encoders?

I just fucking accidentally swallowed toothpaste by accident. I love you guys, I'll miss you guys, if there's a place in afterlife, I'll reserve a spot for my brothers.

>>2022513
sigilent function generator
sigilent multimeter, or get a nice fluke handheld.
rigol ds1054z or sigilent SDS1104X oscilliscope
bare minimum you can get that will be reliable.
there's always the chinkshit route but you're just going to end up buying something better in a year.

>>2022572
Better than accidentally putting your PCB-cleaning toothbrush in your mouth.

>>2022579
>bare minimum you can get that will be reliable.
Analog scopes are plenty reliable, wouldn't recommend anything over one or two hundred dollars to a newcomer either. An analog scope and accompanying cheapy logic analyser will cover 90% of use cases, so you'll probably only think about upgrading after 2-5 years. Chinkshit scopes are hot garbage, but chinkshit function generators aren't anything too bad.

>>2022432
>Is there a simple enough circuit I could wire up that could take in amps and spit out a voltage I could feed into one of the arduinos analog ports?
Yes, a resistor. V=IR

>>2022625
transconductance amplifiers are amazing!

hi guys I need some help

I'm trying to make an array of IR LEDs for a headtracker but when I hooked up 3x 1.5v rated IR diodes in tandem and ran 3x 1.5v AAA lithium batteries, they must've burnt out. I say this because I then tested a single such diode with a single AAA battery and it burnt out in about 5 seconds (and was very hot as it dimmed, etc).

What can I do? hook up 3 diodes and use only 2 batteries? For some reason these lithium batteries show up as producing 1.6v on my multimeter. Are all lithium batteries overrated? I don't have any resistors on hand, and don't know how to calculate what I'd need either.

>>2022432
Just look "opamp current sources" for converting to current and a resistor for converting to voltage.
https://en.wikipedia.org/wiki/Current_loop
https://www.predig.com/indicatorpage/back-basics-fundamentals-4-20-ma-current-loops
>>2022513
they can handle high frequency and were or are calibrated.

>>2022697
first you can stop being a sperg and then use a search engine of your choice to find solutions that already exist for your problem

>>2021801
Thats pretty cool. I use normal hexagon nuts and print out a spot for them to sit into. Its pretty snug, but i use a drop of glue to hold them in.
I might steal your concept.

>>2020636
>DIY
If you have ever woven core rope memory by hand (like I had to do as project/exercise during electrical engineering classes then you know what hell looks like.
Simply, don't.

>>2022513
>Lab class just started
Pick up one of these:
>https://store.digilentinc.com/analog-discovery-2-100msps-usb-oscilloscope-logic-analyzer-and-variable-power-supply/
Its all-in one and if you're just starting out in college, it will be within the power you'll need for a majority of stuff.
Its not the best, and does require a PC/Laptop to use it, but the cost vs what it can do is incredible.
It has extra boards you can connect to the header such as a 4x BNC board to connect oscilliscope probes and waveform generator output.
I use it a lot with audio design. You can set it up to generate a sweeping waveform from 1Hz to 20kHz on one of the waveform outputs for a signal to a crossover input. You can connect the output of the crossover to one of the scope inputs and using their software (Waveforms) to generate a frequency response graph.
Not trying to sound like a shill, I honestly like using mine and I feel it was more than worth the cost.

jerry rig in the hub of a computer fan and read the third wire for RPMs, 2 pulses per rotation. No need to reinvent the wheel when a common part can do it

>>2022513
used scope on craigslist or ebay. If you are starting out you are likely not going above 50Mhz.

Anyone?

>>2019926
>Anyone care to recommend me a crimping tool with replaceable dies for crimping ferrules and crimp connectors like pic related? BNC and those small header pins would be nice too

>Must be available from Digikey or Farnell, by the way. I've tried looking a little but it's just daunting as I barely know what I'm looking for and there are 17000 articles in Digikey's crimp tools category

I realize that "one crimp to rule them all" might be too much to ask for, if so I'd gladly take two-three separate tools
If you have any brands/systems you can recommend that'd be much appreciated

Budget 2-400€ ish

>>2020636
I have a pretty big issue. I was using a hot airgun to reflow a bios chip when it decided to drift away across the board. Luckily it only knocked 3 0603 resistors out of the way, but how do I resolder them on? I've never worked with electronics that small.

>>2022769
1. The fan speed on your gun is too high.
2. You should have used kapton tape or aluminum foil to shield the surrounding components.
3. Now you need magnification and fine-point tweezers to fix it.

>how do I resolder them on?
Magnification, tweezers, hot air station set on low fan speed.

>>2022818
AND USE FUCKING FLUX

>>2022829
He'll flux up the whole damn thing and blow it around with hurricane force winds.

>>2022765
non ratcheting: Engineer PA-09
ratcheting : IWISS IWS-3220M
Both are for 32-20 AWG pins

>>2020679
The tips by >>2020679 plus experience.
Soldering both ends and adjusting then soldering the rest is the basic procedure.
If you can just stab the headers into a breadboard and put your pcb on it, that's almost cheating territory.

>>2022834
>pins
wire

>>2020972
Christ. Get into FPGAs already. You should learn Verilog by yesterday.

>>2022834
Cheers, but can't find any of them on Digikey/Farnell so it doesn't help much

>>2021734
ts80p and pinecil.
The former's smaller. The latter's still small, but much more powerful.

>>2022844
This is the PA-09:
https://www.digikey.com/en/products/detail/adafruit-industries-llc/350/6026207
This is the other one basically:
https://www.digikey.com/en/products/detail/adafruit-industries-llc/1213/8258462

>>2022853
You're a scholar and a gentleman, anon
Sadly no dies for the ratcheting one, but I can live with that

https://datasheet.lcsc.com/szlcsc/1912111437_ON-Semiconductor-NCL30082BDR2G_C403847.pdf
>page 18
What does Nsp stand for?
How does this shit work without feedback from the secondaries? Is it assumming that current in the secondary will be given precisely in function of the primary current and the transformer windings? Shouldn't additional considerations be taken?

>>2022921
winding ratio.
> Is it assumming that current in the secondary will be given precisely in function of the primary current and the transformer windings?
no
>Shouldn't additional considerations be taken?
yes
Read the image subtitle. They obviously removed the clutter and left only the basic topology.

>>2022853
>Sadly no dies for the ratcheting one

did you see this shit? $355 for a replacement die.

https://www.digikey.com/en/products/detail/phoenix-contact/1886948/3605485

>>2022924
thanks m8, I was being a dumbass

>>2022697
LEDs
need
current limiting

isn't it supposed to be cringe to put the load/output on a BJT's emitter? Or is that normal for dual rail emitter followers?

>>2022957

every common topology has its advantages and disadvantages. the only cringe here is the shallowness of such a question.

>>2022957

every common topology has its advantages and disadvantages. the only cringe here is the shallowness of such a question. it's like those idiots who film themselves making fun of what people wear.

>>2022957
is there a reading comprehension problem in /ohm/? First this>>2022939
and now you? It is an emitter follower.

I just noticed this on my chink LCR. With inductors it shows a disproportionately large resistance value, it should be nearly 0. The inductance value is about right, I checked with my scope and signal generator.

Any clue as to why it does this? Can anyone point me to reliable firmware for this thing?

>>2022957
That is a common-collector amplifier circuit. Generally speaking, common-collector amplifiers are used for current amplifying circuits like buffers, but not voltage amplifying circuits. So they have their place, for sure.

>>2022995
>a disproportionately large resistance value
How large? What happens when you measure the inductor with your DMM? What happens when you measure a known resistor with the LCR meter?

>>2022957
>isn't it supposed to be cringe
JFC where do you get these ideas?

>probe slips
>short 40v to a 12v rail
rip 12v rail

>>2023000
>How large?
Like some hundred ohms. The odd thing is it measures resistance perfectly fine, just checker with a 1k resistor and it showed 1001 ohms. The multimeter reads 0 ohms on the inductor, since it's just a toroidal inductor with a thick copper wire.

>>2023015
it could be measuring impedance

>>2023017
Then it should show Z, not R. With smaller inductors it does fine. Idk, maybe it tries to measure resistance too quickly before the magnetic field has dissipated and returns the wrong value.

>>2023015
I was going to say that 100Ω is normal, but not for a coarse toroidal like that. If it measures resistance fine, then I guess it is a firmware issue. Have a look at the circuit still, maybe it's just not sensitive to really low resistances. With any luck there's info on different firmwares online.

>have guitar pedal with potmeters i want to replace with LDR+LED for cv input
>LDR resistance ranges start at 20k
is there also a component that works with the opposite of a ldr? with more light meaning more resistance and a base resistance around 0 or at least as low as possible? i need a range between 0 and 50k since that is what the pots use but ldrs in that range are rare so what do?

>>2023045
>more light meaning more resistance
Just tie the LED to the opposite rail instead. Or put an op-amp/transistor inverter in there.

Anyone ever use these bad boys?
I just ordered one. It doesn't have very good specs with regard to its various utilities, but it's open source and cheap and I want to see what I can do with it.
I'm thinking about building some sort of "GECK" like device in a briefcase with a raspberry pi, a couple screens, backup and error correcting storage, mouse/keyboard, a few different wifi and SDR antennas and one of these labrador boards.
It'd be like a post-apocalyptic hacker laptop-briefcase.
prepper engineering has always been one of my favorite subjects.
When the world falls apart, my kingdom will be enjoying homebrewed wine, functioning electronics, and the luxury of not dying from a small cut that got infected. Maybe if you peasants provide me with 75% of your grain I'll let you join me as we fry up some canned corned beef hash and read science textbook PDFs.

>>2023052
It looks cool, but that's about it. The specs are horrendous, there isn't much you can do with its capabilities.
Its only redeeming quality is that it is open source.

>>2023052
just buy a pickit 3 and several PICs from the cheapest garbage to high end ones.

>>2023052
>It'd be like a post-apocalyptic hacker laptop-briefcase.
That sounds very gay. No homo tho.

>>2023052
Christ. Get yourself some modern devboards.
attiny85 devboard digispark: https://www.aliexpress.com/item/32971847940.html
atmega328p devboard arduino uno r3: https://www.aliexpress.com/item/32680845470.html
atmega328p devboard arduino nano: https://www.aliexpress.com/item/32989224656.html
atmega32u4 devboard arduino pro micro: https://www.aliexpress.com/item/32833347030.html
atmega2560 devboard arduino mega 2560 r3: https://www.aliexpress.com/item/4000946054205.html
stm32 programmer/debugger st-link v2: https://www.aliexpress.com/item/32719963657.html
stm32 STM32F401CCU6/11CEU6 devboard: https://www.aliexpress.com/item/4000103610226.html
ch340g usb uart 3.3v/5v w/rtscts: https://www.aliexpress.com/item/33020575987.html
gd32v devboard longan nano: https://www.aliexpress.com/item/4000505297604.html
esp8266 devboard nodemcu v3: https://www.aliexpress.com/item/4000055280006.html
esp32-s2 devboard nanoESP32-S2: https://www.aliexpress.com/item/1005001403045642.html
esp32-s2 devboard ESP32-S2-WOOR V1.1: https://www.aliexpress.com/item/1005001562674526.html
ice40 up5k oshw FPGA devboard icesugar 1.5: https://www.aliexpress.com/item/4001201771358.html
ice40 up5k oshw FPGA devboard icesugar-nano: https://www.aliexpress.com/item/1005002073816863.html?
nanodla sigrok-compatible oshw logic analyzer: https://www.aliexpress.com/item/4000931572965.html
multimeter fluke 17b+: https://www.aliexpress.com/item/33049901676.html
lm2662 charge-pump voltage inverter: https://www.aliexpress.com/item/1005001639598862.html
txs0108e logic level translator 3.3v<>5v: https://www.aliexpress.com/item/4001278995191.html
ft232h usb jtag uart spi i2c gpio bridge: https://www.aliexpress.com/item/32903296960.html
cp2112 usb i2c bridge: https://www.aliexpress.com/item/1005001546685651.html
1602 16x2 lcd hd44780 w/i2c bridge: https://www.aliexpress.com/item/32763867041.html
ssd1331 96X64 oled spi: https://www.aliexpress.com/item/4000574446702.html

>>2023193
devboards are for cock suckers.

>>2023194
Interesting knowledge you have there, Anon.
I have to wonder where and how you've acquired it.

>>2023193
>devboards
That's not what I was posting about dude...

>>2023142
At this point in the game open source is the only way to go.
I'm just so fucking sick of getting tossed around by multinational conglomerates that shit out crappy products made by an army of H1b visa indians.

>>2023226
I do not know what you are on about. Scopes, meters and signal gens are usually pretty good. (I mean, you get what you pay for but still)

>>2022977
give me a break, I was falling asleep from exhaustion

Conjecture: MCUs that have dev-boards are easier to get started using.
Conjecture: micropython will never be good

>>2023045
idk, try this. LDR resistances go up as CV goes up right? how high?
You'll want that "220k" to be significantly higher than the lower bound, but significantly lower than the upper bound to get full range

>>2023314
listening to boomers talk about $100 programmer boards is physically painful

or like this, since apparently dark = 10meg, light = 20k.
whoops lmao

I'm trying to fix a game boy I bought that was broken. It's been fun exploring how to ID and fix shorts and shit.

I am confused by a readout I am getting. I'm comparing my broken board to one from a working machine. My harbor freight (tm) multimeter tells me that on one set of pins in the broken unit, there is continuity, but on the working unit, that same connection fluctuates wildly between various numbers on the screen, and sometimes settles on 0, but only occasionally, and not for long. What does this mean exactly? Does the disparity indicate a problem, or is it just indicating a small difference and not a significant factor?

>>2023342
One is the audio output, the other i presumably the negative battery terminal. Is the "fluctuation" happening while audio is playing?

>>2023347
No, it's happening when turned off. I'm in that state with both units because the broken one won't turn on at all. I thought that it was a problem with the power switch, so first I checked it and found that in the working one, there was continuity between the positive terminal and point 3 on the switch, and there wasn't any on the broken one. So, I went and soldered a wire between the two, thinking that might complete the circuit and resolve the problem. It was done, but it won't turn on still, so I'm not sure where to look next, and I'm just sort of trying to find spots where the connection behavior isn't matching up between the two systems.

>>2023193
Devboards are for blinkies.
Good job even looking up what the anon was posting about (hint, its not a fagboard)

>>2023226
You're not wrong, but do you have any specific examples?
I've picked up various engineering tools/hardware and the ones that gave me issues were cheap or clones.
The only large company that kinda fucked me over was MikroElectronica. They decided to stop updating their Pro compilers in favor of their Necto studio, which you have to buy new licenses for. That's software though.

>>2023332
*Laughs in $500 programmer/debugger/emulator* What are you, poor?
Just kidding, but seriously, shit gets retarded expensive. I'm no arduino fan, but it certainly drove down the cost of a lot of stuff.

>>2023364
dumdum
devboards are for development
you do shit on it due to its ease of use and debugging tools and then you program the end result on the mcu

>>2021418
>>2021381
Thank you guys.
I think i'll look into controlling a solid state relay with PWM. The relay I have now is for 20 amps and you can easily hear it switch and there is no way you could switch this thing more than once a second.
I'll have to look more into the PID controller.

>>2023332
If you’re paying more than ~2 times the price of the bare MCU for a dev-board, then you should just design and make your own dev-board. Should make one for every MCU you commonly use, for debugging and prototyping purposes, with perf-board if you’re in a pinch. Ensuring it works with external programming hardware, preferably with provisions for HV programming in case you need that reset pin.
Designing a general-purpose MCU dev-board should be a common entry/mid-level PCB designing project.

As for programmers themselves, any MCU family that REQUIRES an expensive programmer likely isn’t worth getting into, at least for a hobbyist. There might be some specialist applications where it is worth doing, but even most professional cases could probably just get away with general-purpose 32-bit RISCs like STM32s or equivalent. Cost optimising notwithstanding. Wouldn’t mind dropping up to $40 on a padauk programmer if it meant I could churn through 3c micros.

>>2020757
I'd just go with a reed switch and magnet. Reed switches are rated for millions of operations, and switch contacts are easy to interface.

Or you cold consider using a proximity sensor like these or similar:
http://www.ia.omron.com/products/family/459/feature.html
All you need is a piece of metal passing in front of it's sensing area every rotation.

Optical would work too, use an optocoupler, but if it's outside there's always a risk of it getting contaminated and not working reliably.

I used to work for a water quality sensor company and we did weather stations as well. The anemometers we used were off-the-shelf and were basically small generators, the output linearly proportional to the speed.

>>2020820
>Is making a basic digital clock hard?
Can be done with small-scale logic ICs like 74LS series TTL or 4000 series CMOS, or you can use a microcontroller.
Power it off the AC line and capture the 60Hz signal to divide down for your 1s pulse.

>>2020820
A fun way to go would be to get a time signal from somewhere. Simplest would be the AC mains, more complex would be a terrestrial radio time broadcast station, while the best would be a satnav-based system. GPS receivers can be bought for ~$15 and are somewhat easy to interface with an MCU, making them a reasonable choice.

>>2023472
>a reed switch and magnet
Why instead of a binary hall-sensor? Hall sensors are physically more robust, can be more sensitive, smaller, higher frequency, etc.
If he wants more adjustability he could use an analog hall-sensor plus a comparator and pot for tuning, but I doubt it would be necessary.

>>2022158
You sound like you've not worked with 0402's and 0201's all day long, therefore not knowing that you can delaminate traces with too hot an iron.
Don't even give me that "oh I IR reflow all that" or "oh, I use hot air for all that" because that's bullshit, hand rework is a Thing, and a hobbyist doesn't have thousands of dollars worth of soldering equipment.

>>2022432
>>2022625
>Is there a simple enough circuit I could wire up that could take in amps and spit out a voltage I could feed into one of the arduinos analog ports?

>>2023477
Did you not see where I said that reed switches are rated for millions of operations before wearing out? Note they use them in places like wheel rotation sensors for bicycle computers. They won't wear out for years and years, require no power (other than a pull-up resistor to interface to a digital input) are inexpensive and easy to use, no guesswork required, either closes reliably or it doesn't, and adjusting them is easy, most have a wide sensing range especially if you use a small rare-earth magnet.
Not everything has to be complicated.

>>2023477
>>2023488
First one I pulled up has a life expectancy of 100,000,000 operations at logic-level voltages. That's probably at least 10 years worth of windy days.

>>2022719
no thanks retard
>>2022950
thank you sir did me a needful

>>2023494
>12V/10mA
>greater life than at 10V/1mA

dude what

>>2023488
>>2023494
Yeah, I saw that. But I didn't mention number of cycles anywhere, it's not the only stat that matters. They're mechanical devices, therefore vibrations and thermal expansion and such have the potential to put physical wear on the parts. Especially for ones with a glass case. See:
>mechanical shock as the result of dropping the reed sensor typically from a distance of greater than 12" may change it's magnetic sensitivity and/or destroy the sensor
>it's

They're still larger and more difficult to mount in a case or on a PCB, though not by much. And hall sensors are still more sensitive and can react quicker. Plus they're like 5 times cheaper, which is the main reason for me.

>10 years worth of windy days
There are ~32M seconds per year. If it's spinning on average once per second, that's 3 years. If you're not a retard and use a multipole ring magnet for higher resolution, that's less than a year. Give or take an order of magnitude, considering the spin rate is proportional to the radius of the anemometer.

Hello anons. How might silicone or silicone casting liquids affect PCBs/PCBAs, solder, and other such components? I would like to produce a silicone cast against a few electronics parts, but I'm concerned about potential reaction-based damage that could arise from doing so. Any advice would be appreciated. For reference, I'd be using something like this (or better):
https://www.amazon.com/gp/product/B07XKBJWKH/
Alternatively, how can I make a silicone cast between electronic components safely?

>>2023556
PCBs are held together with thermoset resin, something akin to epoxy. Shouldn't be effected by all but the most viscous solvents. Same applies to plastic casings of ICs, though the same can't necessarily be said for heat-shrink-wrapped inductors, enamel insulation on magnet-wire, maybe the coatings on optical components, that sort of thing. The markings on glass diodes might also get wiped off. Silk-screen and solder mask also isn't necessarily safe from a curing polymer's solvents, though they aren't particularly vital for the operation of a potted PCB.
The glass fibres within the PCBs (FR4 at the least) are effectively chemically inert. The solder and copper and other metallic parts are chemically reactive, but shouldn't be too much of an issue with organic solvents. The exception I'd worry about is that silicone can be acidic when not cured, which might potentially erode some of the metals, though probably not appreciably.

The bigger worry is that the curing polymer will be poisoned by ingress of foreign chemicals leeching from the PCB, such that it doesn't cure properly or has altered properties. Could just spray-lacquer the whole thing beforehand if you're worried, assuming the lacquer will work. Are you potting a PCBA for ingress protection, or something else?
It's barely $10, I'd buy it and spot-test it. Also that manga sucks ass.

>>2022849
Neither seem to have decent working distance or even tips made for microsoldering.
Microsoldering doesn't mean small irons.

>>2023570
Thanks Anon. TIL.
Watching fosdem atm: https://app.element.io/#/room/#microkernel:fosdem.org

>>2023570
You said "micro soldering irons". How are we to know that you meant "micro-soldering irons" and not "micro soldering-irons"? For a short working distance, a T12 station is pretty decent at 45mm with my particular handle. IIRC its smallest common tip is the D12, a 1.2mm screwdriver tip, which is pretty reasonable for SMDs. Anything smaller is likely to have poor thermal conduct, though IIRC there's a really narrow bent conical tip out there. Can't see it being practical to have a tip any smaller than 1mm or so, unless you really worked to pump the heat into the tip. Go to pitches too tiny or QFNs, to say nothing of BGAs, you'll want hot air instead of an iron.

>>2023566
>The bigger worry is that the curing polymer will be poisoned by ingress of foreign chemicals leeching from the PCB, such that it doesn't cure properly or has altered properties.
>The exception I'd worry about is that silicone can be acidic when not cured, which might potentially erode some of the metals, though probably not appreciably.
Yes, this is what I was concerned about. This particular reaction and a potential few others have been on mind. I've heard of some dreadful creep occurring resulting from leech reactions against PCBs resulting from molding processes like these, but I wanted another opinion.

For reference, I'm trying to produce a silicone pad that would seat comfortably between an acrylic plate and PCB. I have CAD and plate design files for these, but no access to a laser cutter or 3D/resin printer to produce a negative for casting silicone against. The shape is fairly simple, so I could possibly produce something like a wooden negative according to a print of the plate, but there are many holes on it, so that would surely take a while to do safely and effectively. I'm unsure of the best course of action with these challenges in mind.
>Are you potting a PCBA for ingress protection, or something else?
Sound management, physical sturdiness improvements, and aesthetics. I don't need it for any other form of isolation.

I just thought about it, but what do you think of printing the plate frame to scale on paper and carving a silicone sheet to match? If so, do you have recommendations on such silicone, and how to make consistent, orderly trims?
>Also that manga sucks ass.
Yes, yes it does. I thought it would pick up, but never did. I stopped reading ages ago but kept that pic.

>>2023576
>Yes, this is what I was concerned about
Then I'd only do it with a conformal-coating or spray-lacquer, at least before doing rigorous testing.

>I'm trying to produce a silicone pad that would seat comfortably between an acrylic plate and PCB
Since there will be room between the two for ingress, without a solid bond or seal, you definitely don't want any potential nasties getting between the two.
>no access to a laser cutter or 3D/resin printer to produce a negative for casting silicone against
I'd just use some sort of modelling clay to make a negative from the PCB, then bake it into a solid negative. Then make a solid negative of the negative. Then silicone that.
>Sound management, physical sturdiness improvements, and aesthetics.
Fair enough.

>printing the plate frame to scale on paper and carving a silicone sheet to match?
Possible, if you're fine with some shitty foam-board then that would definitely be a cheap method. Same for neoprene. There should be various rubber sheets available if you look about too, see common ebay soldering mats for starters.

>Yes, yes it does
There are a few half-decent manga in a similar vein, like "Manuke FPS" and "The Villainess Will Crush Her Destruction End Through Midern Firepower". "Gunka no Baltzar" is great at not being an isekai.
There needs to be more engineering fiction, something that isn't just "hey look how clever this 100 year old invention is".

>>2023581
>There should be various rubber sheets available if you look about too, see common ebay soldering mats for starters
I took some initiative to hunt for some on Amazon and eBay. Found this 2mm sheet which could arrive reasonably soon and supply plenty of material for my project:
>https://www.ebay.com/itm/US-Stock-2mm-x-250mm-9-84-x-500mm-19-69-Silicone-Pad-Sheet-Rubber-Plate/391954816115
My only concern is regarding toughness. What value is enough or ideal for the qualities I'm seeking (as described earlier), while simultaneously being feasible to cut through? Furthermore, which tool that I'm likely to own would make this manual cutting and measuring process most accurate and efficient?

>>2023379
Tell that do all of the 'duino retards that program a devboard and slap it in a case as a "professional product"

>>2023456
>If you’re paying more than ~2 times the price of the bare MCU for a dev-board, then you should just design and make your own dev-board.
I'll interject that a good bit of dev boards have extra stuff soldered to them which can drive up the price. Although, those do tend to be more specific for their development scope, but you're not wrong.

>any MCU family that REQUIRES an expensive programmer likely isn’t worth getting into, at least for a hobbyist.
Again, totally agree, programmers should not be expensive at all.
Debuggers are where I would expect things to get pricey, based on performance. I have no issues paying a few hundred bucks for a debugger that can let me run a program, watch registers, pause, modify code on the fly, all real time.

>>2023546
>that's 3 years.
>that's less than a year.
Hell yeah anon, That's thinking like a business man. Make shit that dies out within a few years.
For real though, for detecting the rpm's of a prop like this, you want to eliminate physical switching if at all possible. Not to mention the added complexity of having to debounce the reed switch.

>>2020636
Do you guys have any good resources on creating your own solar panels?

>>2023616
It works. It sells. It's professional.

>>2023622
except it does not?

>>2023621
jerry elsworth made her own semiconductors. I guess you can make the copper oxide ones easily but they aren't worth anything. It is pretty much a futile attempt.

>>2023624
Why wouldn't it? Devboards are just glorified breakouts.

>wanted to get cool microchip free samples
>do not ship to brazil
please come to brazil (and kill me)

how to make mfd board project boxes if i have no laser cutter

>>2023680
buy boxes at the craft and art store and cut the holes.

>>2023456
>If you’re paying more than ~2 times the price of the bare MCU for a dev-board, then you should just design and make your own dev-board.
There's no such thing in that price range.
the purpose of a devboard is for rapid prototyping with a proven design that will work out of the box.
if that's too expensive for you, then you're not the target market. if you want to re-invent the wheel to save a few bucks, you're more than welcome to.
it also serves as a reference design where you can use the schematic for your custom design.

>>2023456
>As for programmers themselves, any MCU family that REQUIRES an expensive programmer likely isn’t worth getting into
>laughs in systemview

>>2023363
OK, so I took a break and came back, and decided that since I have a ton of game boys, maybe I should look at another working one. Turns out, I dunno if it matters really, but the revision of my working model was CGB-CPU-05 and my broken one was CGB-CPU-04. I found another working unit that was CPU-04 and found something funny; A spot where I got some odd resistance behavior on the working CPU-05 (working 1), which was + terminal to this little board at point 1, was fully continuous on working CPU-04 (working 2). I checked the same spot on the broken unit and found it was a dead connection. So, I wired a line directly from + terminal to the point 1, and wa la~, fucker turns on yet.

The only problem I'm having is, well, it doesn't turn off. The switch doesn't do anything, it's just always on once the batteries go in. I tried desoldering the connection I had made from + terminal to the switch and found no change (which at least means I can remove this wire) but it doesn't really answer why the switch isn't working. Anybody able to give me some advice? Pic is board as it is now, prob gonna desolder the loose wire next

>>2023765
pic related, sorry for double-posting, just very pleased as I've never fixed an electronic device like this before. Not over yet though, gotta figure out this power problem, and replace the speaker since it's very obviously ded

>>2022818
I was on a low fan speed, and I did use shielding, but the resistors were right next to the bios chip. I ordered some new resistors, hopefully I can salvage this board

>>2023768
You bypassed the switch (which seems bad) with a jumper. Install a new switch and remove the jumper.

>>2023769
Were you able to get a dump of the original BIOS? Are you going to install a socket for the new chip?

turns out the top side on a SOT-23 package is not what you would expect

>>2023693
i can't find them here why do you think i went this route

>>2023836

>>2023840
fuck is this homosexual garbage

>>2023817
This doesn't make sense anymore
If I swap the anode and cathode now it should be fine, but that would mean that every single pin is not correctly specified.
Maybe its a chinky tl431

>>2023841
Fuck you. Figure it out yourself, faggot.

>>2023846
stacking design is a representation of homosexual orgies

>>2023852
You seem preoccupied with faggot thoughts. Maybe you could stuff it in your mom's box and cope.

>>2023655
I'm surprised companies still give away samples. I remember being so poor that I thought that was reasonable way to get parts.

>>2023836
Huh that is weird. Usually there is always a store to sell this crap to housewives to make gift boxes and shit

>>2023863
>I'm surprised companies still give away samples. I remember being so poor
Remember? That is me lmao. If I can get two 10 $ micros for free and still do usefull shit with them for free I will. (or not in this case)

>>2023863
I remember saving up pocket money just to get the parts to make a shitty little parallel port programmer.

Now you can buy a complete dev board for 3$. I would have killed for something like AliExpress when I was a kid.

>>2023881
jej I did the same exact almost 20 years ago to program pics. Made my own pcbs with the UV and copper etchant.
At least now I can say I /diy/'d something.
Nowadays have zero interest in reinventing the wheel by handbuilding shit like that when I can focus more on the main goal rather than dabble in side quests.

>>2023556

maybe acrylic is more your style.

>>2023895
>cap fails
>throw away everything

>>2023895
>over heats and burns out

>>2023879
Most companies don't give samples to individuals because too many people exploited that. Before I got my current job I was dirt poor and trying to get into electronics and I tried to request samples so I could get parts, I found that most companies required you to put in information about the company you worked for with your sample request which effectively blocks regular /diy/ people. This was back in the early 2010s but I imagine it's still the same way today.

>>2023900
Microchip does. They said they give out to students and teacher and I have my university email. it was a shipping limitation because the brazilian mail system is hell. I know Coilcraft does too because they sent me a custom magnetic core once (I only paid shipping).

>>2023895
>a r*dditor's updoot finger begins twitching uncontrollably

>>2023902
Hmm, I think I tried companies like TI, Analog Devices, Linear Technologies, and Maxim and struck out everywhere. Maybe I'll try hitting up Microchip at some point, I'm out of uni but I think I still have an old university email.

>>2023900
The zoomer way to get free parts is to have a youtube channel and promise to feature their parts on your videos.

>>2023895
That is pure autism. Looks cool, but thats about it.

>>2023902
Hell yeah, I started back at college to get my masters and I'm already whoring for free pics.

>>2023913
e-e-njoy your functional mail system anon

>>2023621
What do you mean by that? Just assembling photovoltaic wafers in a frame? Or making your own semiconductors? If it's the latter, you'd be better off making a heat-engine-based energy collector.

>>2023895
>>2023911
I wonder if it's ever a good idea to make a circuit in the air like that, but instead of potting it you immerse it in liquid coolant.

>>2023954
3m makes special oil for this purpose. fluorinert i think it's called.

>>2020842
shouldn't Vdg be 50 V ? because you should look at the charge Q=Crss*50V and then the time constant let call it Tau = Q/Igate which would charge up the cap to 63%.
>3*Tau gets you to 95%
>5*Tau gets you to 99.%
So that would give you dt = 50µs
we should get TeX formatting like in /sci/

>>2023954
search bob pease hair ball

>>2023978
It's a MOSFET, shouldn't the majority of the gate capacitance be relative to the source? Now that I look at the datasheet, the gate-to-drain charge is significantly larger than the gate-to-source charge, which is apparently a miller effect thing. He definitely should be using 50V as the Vgd anyhow.

But even if that's the case, the context highly implies a constant "1mA", not some decreasing current exponential decay.

>>2023985
hair ball, not haiRY ball

>>2023900
I got alot of free shit from texas instruments that way when I was a broke teenager. They sent many, many samples to Popsicles LTD.

>>2022197
>go fast with a huge nozzle
That should be every man's goal

What kinda chip would I be looking at to drive 5 individual LEDs with one pin? I've got limited pins available but would definitely appreciate some status LEDs to make sure everything is still working at a glance.

>>2024051
serial in parallel out shift register.
74LS594

>>2024055
Thanks anon, but after seeing the price it's cheaper to just run another $4 nano. Why is that chip so pricey?

>>2024074
what? It should be way cheaper than a 4$ nano.
Check out another chip, maybe a 595 would be useful.

>>2024079
Ebay is usually my first goto and the lowest price I could find for a single one was $5 in china.

Anyways yeah the 595 is definitely cheaper. I just found a $4 listing for 5 of them that's based in the US.

Thanks anon.

>>2024074
are you sure? it costs literally cents

>>2024074
I misremembered the chip I used yesterday to do the same thing. It was the sn74hc595. It cost couple cents I guess. The 74LS594 seems a bit more expensive for some reason. But what you are looking for is that, a SIPO shift register

Looking at alternatives to Amazon for ordering components in small quantities for Canada, but I can't seem to find anything good rivaling their shipping costs.

>>2024085
it is always cheap when you steal your driver's salaries.
>>2024084
https://www.ebay.com/sch/i.html?_from=R40&_trksid=p2334524.m570.l1313&_nkw=sn74hc595&_sacat=0&LH_TitleDesc=0&_osacat=0&_odkw=74LS594

I'm building a dust blower using a San Ace b97 server fan I scavenged years ago, because I refuse to buy one. I have a question because I was curious and genuinely didn't know:

Does it make a difference where the switch is in the circuit?

>>2024092
>small battery shit that does not care about small signals, noise or other things
does not matter
>almost every application including line connected deviced and loads
bottom circuit

>>2024088
For sure, not trying to go with them, just curious if any other Canadians ITT had a recommendation.

>>2023786
I was removing the bios chip to add on a corebooted chip

>>2024120
Can't you flash it without removing it? I thought it was possible to use external hardware like a CH340 & IC clip or soldered leads. Too late now, but nice to know if you want to do it again in the future.

>>2024051
>5 individual LEDs with one pin?
Well a general-purpose solution would be to find some sort of demux/IO expander IC that uses a single-wire protocol of some sort. Like 1-wire, though it's specially designed to provide both data AND power. SDI-12 might also be good.
Even RS232 might work, if you're lenient and don't need any return data or status pins. Easiest way would be with an added low-spec 8-pin micro, such as an attiny13. But that feels like a Shenzhen IO-tier fix, it would be preferable to use a dedicated demux IC or a larger MCU in the first place.

>>2024055
Is that different from a 595? Either way he needs clock inputs, which need to be synchronised with the host. If he has two external clocks of the right frequencies, then great. If not, then he needs one or two pins from the MCU itself to do the timing. Arguably he could do some wacky shit like outputting manchester data from the spare pin, and using some circuitry to decode it into data and clk, or by using some external astable timing ICs and bit spacing/framing in order to get some asynch clocks, but both are clunky and increase part-count unnecessarily.

>>2024100
>centre negative circuit where GND/PE is the positive rail for galvanic corrosion purposes
top circuit

Did I get scammed by my college ohmbros? I've always wanted to delve into electronics so when the lab class came, I got really excited about it until I saw that this lab kit cost roughly $110. Is there anything I should buy more of? Or any measuring tools? I kind of want to have a mini-lab!!
PLEASE, WHY DOES IT ALWAYS ROTATE WHEN I POST THE IMAGE AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

>>2024156
>Is that different from a 595? Either way he needs clock inputs, which need to be synchronised with the host. If he has two external clocks of the right frequencies, then great. If not, then he needs one or two pins from the MCU itself to do the timing. Arguably he could do some wacky shit like outputting manchester data from the spare pin, and using some circuitry to decode it into data and clk, or by using some external astable timing ICs and bit spacing/framing in order to get some asynch clocks, but both are clunky and increase part-count unnecessarily.
1 clock for the input, another for the output registers and another for input.

>>2024165
what the fuck
stuff like this makes me want to become one of those student council suckers just so I can mass organize mouser orders.

>>2024197
I-I don't get it, is that a good thing or a bad thing...

>>2024199
if you find a aliexpress seller that has all of those items (or two) you can get that shit for 25$+shipping. Probably will come from the same factory or depot as your school

>>2024202
Fuck.

>>2024165
You could probably get all those components from alibay for $30-$50. Though quality breadboards are better than the chinky ones, and I'm not sure what that wire stripper is worth.
>1N4001
>not 1N4148 or better
>for small signal shit
I guess the whole thing isn't awful, but if you're willing to spend a few hours trawling the web, and a few weeks/months waiting for it to all arrive, I'd look into buying it elsewhere. There's also a bunch of stuff I'd consider important to have that I'd order from elsewhere anyhow. Speaking of which, I've got to make myself a list of SMDs to buy.

The phone stores direction as metadata, not as an actual transformed image. 4chins doesn't care about that metadata. That's why you should aways take landscape photos.

>>2024207
The breadbroad does look pretty quality actually.
>not 1N4148 or better
Should I go and buy a bunch of these 1N4148 things then? I actually don't know what any of these means, but with how scummy college can sometimes be, I don't think this is enough for 'room for error' and will be 'if you fuck up that's too bad' type of thing...

>>2024213
just look at this shit and the quantities it has
https://es.aliexpress.com/item/32633071171.html
The 1N4007 is practically identical to any 1N400X for your application.

>>2024213
If you're already getting the 1N4001 power diodes there's not much reason to go for some 1N4148 signal diodes, as at the low speeds you'll be caring about there won't be a difference. But if you do plan on building up a collection of parts, it isn't a bad idea. Along with general bags of assorted 1/4W resistors, ceramic capacitors, and electrolytic capacitors.

Cross posting because not sure which thread this goes in.

>>2024275
>>2024277

>Blinkie here.
>I'm trying to get a button to work. Wired the 5v pin to a 10k pull up resistor which was wired along with one lead to the button pin D2, the other side of the button wired to the GND pin.
>The button pin D2 is still acting as if it's floating, randomly fluctuating on and off. What the hell is going on?

>Just to be clear it's a CH340G arduino nano knock off and the pins are as follows,
>5v ---> 10k pull up resistor
>10k pull up resistor ---> pin D2
>One side of button ---> pin D2
>Other side of button ---> GND

>>2024213
if you plan on doing anything high speed, yes. 1N4001's have a capacitance of 25-50pF (actually pretty good as varicaps), 1N4148's are typically 0.9pF (max 4pF)
They're also smaller and prettier

>>2024283
Is the input active high or low?

>>2024283
So it looks like pic? With common ground and +5V rails between the arduino and the button+pullup?
If so, then the problem must be in code. Either you've set the wrong digital pin as the input, or you've not properly set it as an input in the first place.
You should have:
>before everything:
>int var = 0;
>in setup:
>pinMode(2, INPUT);
>in loop:
>var = digitalRead(2);

I'd also digitalWrite the button input value to the builtin LED, so you can see the change live.

Retard question, if I bought a replacement power cable for something that is 10 gauge because i'm an idiot to replace the original 16 gauge wires, will it make mustard gas if I use it anyway?

Mouser vs Digikey? Both seem to have comparable prices.
Or do you shop somewhere else?

>>2024131
Not with the chip I'm replacing, because the pins sit underneath the chip

>>2023954
People immerse PCs in mineral oil for cooling

>>2024357
I know people do that shit for PCs, but I'm specifically talking about making a circuit in mid-air for the purpose of letting as much fluid flow around it as possible.

>>2024349
I like Digikey's parametric search better than Mouser's so I tend to shop there more. I usually use Mouser when I can't get a part at Digikey though that's pretty rare.

>>2024359
High voltage transformers, the kinds on utility poles, are usually immersed in oil. Maybe it's done with power electronics in some particular applications as well. Probably not terribly common outside of HV distribution but it is done.

>>2024364
Oh yeah, they do it for a higher breakdown voltage.

>>2023575
BGA's yes you need a proper reflow station, but for ridiculous small pitch flat paks I've done just fine drag soldering with standard size but good quality iron and lots of good quality flux (I recommend mg chemicals gel in the syringe). Turn that thing up to max heat and the solder ball will follow the iron (it always is drawn towards the heat)

>>2023892
I made both a pic and an atmel programmer almost 20 years ago on perf board, parallel port controlled. PIC was to make PS1 modchip. Built them out of scrap parts and handful of digikey parts. One of them I subbed green LED's for zener diodes for a level converter? if I remember correctly. Maybe that was something else. Shit how did I get so old?

man i just spent like 3 hours trying to install or compile vesctool, still isn't working
gonna pay some fucker on /g/ to do it for me

>>2024419
Can't you just stick bga's in the oven?

>>2024469
Oven is better generally if you're building new board.
For rework a gun is more convenient.

>>2024345
>Retard question

you're not clear, and that should be a capital crime.
if you used 10AWG to replace 16AWG, that's fine.
if you used 16AWG to replace 10AWG, that's bad news.

>>2024345
>Retard question

you're not clear, and that should be a capital crime.
if you used 10AWG to replace 16AWG, that's fine.
if you used 16AWG to replace 10AWG, that's bad news.
and it's all so obvious.

>>2024367
and for heat management. substation transformers have pumps to circulate the oil. The oil is also a super cancer soup.

>>2024297
>>2024330
I have determined the 100 clicky button switches I bought are absolute fucking junk. Almost every single one of them don't pass the multimeter continuity check without an ungodly amount of force applied to them, even then it's 50/50

Didn't know it was possible to fuck up something so simple, Fucking chinks.

>>2024794
I bought a bunch of rubber dome tactile switches, they take less force to push.

>>2024809
I don't think you understand. The buttons will 'click' but there will still be no continuity. You have to apply enough force to snap a circuit board in half in order for them to actually switch.

>>2024816
yeah i got that, just "buying new switches of a different sort" is a good option for your situation in general, and "buying new switches that require less force and have a somewhat different actuation method" is a more likely way to get reliable switches.

Ok guys so I have a noob question, I am using a virtual ground and want to connect something to it, so I need a high reseistance to make as little current as possible to flow in the virtual ground node in order to not disturb it. My question is how high should that resistance be? Is 10 times larger than the resistances in the current ok? Should I go for 100?

r8 the AMS1117

>>2024794
>inb4 it's normally closed

>>2024904
1V2 input/output differential is pretty good vs. LM317, which requires 3 V.

New mosfets are not only superior, but cheaper than old models, is there any other component where this is also true?

>>2024922
You're comparing it to a ~50 year old chip, 1.2V isn't that phenomenal by today's standards. If using a P-channel analog MOSFET as the pass transistor, you could theoretically get arbitrarily low dropouts, though I can't speak for the ripple rejection.

Datasheet says to use a PNP transistor with a 180 beta, but datasheets for transistors don't list this value and trying to calculate by IC/IB yields very low values.
Are these transistors rare or something?

My chink T12 finally came, so to test it out I made an Arduino on a perfboard.
Any criticism/tips when it comes to soldering on a perfboard? I was designing it on the go, so I forgot about the crystal and VCC/GND header pins, will solder them on later. Also TX LED randomly disconnects, will need to reflow solder in it and some other places too.
The header pins are made a little narrow, so I can make a small shields for it with my smaller 2-sided perfboards.

>>2024971
are you stupid? of course they list beta. Or one of the other names (Hfe for exxample).
>>2024938
microcontrollers

>>2025012
You have a bad habit of asking people if they're stupid. I think you're stupid and gay. Prove me wrong.

>>2024794
it's fucked up on purpose. you bought the rejects that failed QC at the factory to be disposed of and wan pulled them from the trash to sell on ebay/aliexpress.
there's a reason shit costs more from catalog vendors. you know its correct because the supply chain is verified and controlled.
if you gamble to save pennies sometimes you loose.

>>2024938
everything technically.
things always become smaller/power efficient than the older models.

>>2025008
if you're buying chink stuff anyways, why not pick up a chink arduino clone for pennies and use that to make something new rather than make a half assed dev board based on the same architecture.

>>2025051
Obviously, however, when considering the common jellybean mosfets and its modern alternatives, the differente is so dramatic that I believe its worth mentioning.

>>2025053
I did that for fun, since I had a atmega328p laying around, I have a chink Arduino Mega I use for most of my projects.

>>2025008
>Any criticism/tips when it comes to soldering on a perfboard
Don’t. Manhattan prototyping is cleaner and easier to solder. Takes up more room though.

>>2025012
Technically, hFE and beta aren’t the same thing. Beta is the context-specific current gain, and changes with different circuits, while hFE is the maximum DC current gain.

Is this the optimal way to setup opto-isolated feedback for a low to high voltage boost converter?

All of the reference material for this online shows examples of offline converters (eg. mains to 5V) and it's a bit simpler since the LED can pull more current on the low voltage side and get a better CTR. I can't, at least not without using power resistors.

>>2025067
I'm dumb, forgot pic

>>2025070
I think you’d usually want to use a zener or tl431 or something on the HV side to get a relatively abrupt turn-on of the opto. Just throwing the opto’s output through an ADC is more susceptible to drifting, at least compared to a tl431 solution.

>>2025070
>>2025067
do not trust the CTR, unless you are transmitting a signal and need it to be exactly the same on the other side. As his dude said it is easier and more reliable to just set it to turn on and off.>>2025076

>>2021023
>I still want to do so eventually. Because I want to make a game where you actually do engineering as a significant part of it. None of that subnautica-tier playing around. Designing working 3D models of engines, firearms, even circuits, that are simulated and characterised.
That sounds super rad, hope you make it anon!

>>2025076
>>2025125
Any particular reason for using a TL431 over a Zener?

>>2025149
you can calibrate the 431 with a pot later

my neighbors gave me one of those bullshit "car jumpstarter battery packs". It has a very attractive lithium cell, and I want to tape the thing to the back of my EEEPc laptop since its battery is toast.

The owner's manual says it's rated at 12 volts out, 200-400 amps out. Does that mean that it will *always* try and output 200-400 amps, or just that if the load tries to source that much current, it can provide up to that current?

>>2025149
Within the TL431’s breakdown voltage, it can be used like an adjustable zener diode. You can extend this voltage limit arbitrarily with some more resistors and maybe also a transistor. But even more importantly, it has a much more ideal/sharp cutoff than a zener. Zener diodes have somewhat of a curve, their effective zener voltage depends on the current through them, and will drift with temperature. This is more of an issue with lower voltage zener diodes. The TL431 has an internal 2.5V voltage reference, connected to an internal op-amp/comparator, so it will always trigger much closer to 2.5V at its input.

>>2025145
Thanks, though it’s only a dream at this point. Maybe in 10 years once I’ve got two or three of my other dreams out of the way.

>>2025186
If you don’t know the basic characteristics of a battery and didn’t spend any time researching it yourself, then you should absolutely not be messing about with a lithium ion battery that could discharge itself in 30 seconds and explode in your face. They’re not to be trifled with if you aren’t willing to do your research.
Buy a cheapie replacement battery on eBay and keep that thing for your car, they work great when you’re in a pinch.

>>2025210
>They’re not to be trifled with if you aren’t willing to do your research.
That's what I'm doing right now, I'm trying to do my research. I know what I'm getting into, but my forte is digital signal processing so I'm perhaps not 100% up to speed with power electronics. And that's why I'm trying to take precautions. I'm trying to ascertain the feasibility of re-implementing this battery pack for my laptop.

>>2025214
Just fucking wire it straight up.

YOLO.

im gonna shitpost in this general until I make it. end goal is picking up a DIY modular synth hobby and continuously toying with it. right now I'm restarting chapter 1 of "Handmade Electronic Music". i got all the parts to make the LM386 amp awhile ago so hopefully they're still laying around.

>>2025214

your laptop prob needs 16-20V but your battery only provides 12V. so you need something like this. i found one for $2 at the thrift store, works fine.
also, check out the amp-hour rating of the battery, and the power consumption of your laptop, to see if it's worth using.

>>2025240
>LM386 amp

wtf for? you can use a million other amps with better performance at virtually no cost. like thrift store computer speakers for a couple of dollars. or your stereo, or TV.

save the construction for stuff you cant find everywhere already made.

>>2025214
>my forte is digital signal processing
Oh my bad, you’re less of a brainlet than I if nothing else. There should be a document in the OP that should get you up to speed on the particulars of lithium ion chemistry, though I’d read the Wikipedia article for “electrochemical cell” or “battery” or something first. TL;DR they’re voltage sources first, the voltage is the more independent variable while the current is the more dependant variable. I = V/R, though defining that R isn’t as trivial as it might, due to the internal resistance of the cells. Lithium ions in particular are very finicky in how they want to be charged and discharged, but thankfully there are a lot of solutions for doing this. Be they protection circuits packaged with the cells, BMS boards, or specialised battery chargers.

Problem is, a 12V lithium ion battery will be 3S (three 3.2-4.2V cells in series), while I believe laptops typically use 4S or 5S instead (hence 19-22V chargers). This could be a really big hurdle that requires bypassing the laptop’s BMS IC, adding your own BMS, and possibly modifying or replacing the built-in buck converter too, making it a huge pain in the arse.
Also the cells in it will be specialised for higher currents, and will have subpar energy density as a result, though I wager you don’t care too much about form-factor if you’re considering this.

As a part of your research, but also as looking into an alternative solution, look at diy laptop battery refurbishment. IIRC they just measure the cells, toss the shorty ones, and replace them with good ones, which can be pretty cost-effective. Might require a good PSU (or some clever DAC+MCU+buffer following a charge curve) to revive overdischarged cells though.

Using a voltage converter inline with it and powering the charge socket is a bit of a shitty solution that will lose a bit of efficiency, but it’s by far the simplest. Could maybe even keep its jump-starting capability while you do it

>>2025244
it's good to have a couple cheap amps handy. Also, to teach this noob a hard lesson about high gain amplifiers

>>2025248
>high gain amplifiers
LM386s aren’t really high-gain. They’re also kinda subpar, they’ve been around for yonks. There are likely some much more modern replacements, with nice benefits like “high input impedance” and “high stability” and “not demodulating AM broadcasts”.

Should I actually go about making a step down converter or just use an inefficient(but simpler) voltage regulator to step down the voltage from 12v to 5v to run a ~1.5w load?

I'm a little worried about the conversion efficiency because that load will be run off of a 7w solar panel and a 12v battery. I'm worried if there's multiple consecutive cloudy days my power supply will eventually go flat.

Somebody please ease my mind.

>>2025260
>not demodulating AM broadcasts”.
fuck off m8, touching random pins and getting chink/christcuck squabbles is based

>>2025282
To each his own. First time I ran into it, I was surprised at how easily it happened. I had an op-amp self-oscillating at a few hundred kHz (needed another cap or two), and the output of the circuit was noisy but definitely had non-pitch-shifted audio in it. Non-pitch-shifted implies something akin to an envelope detector, not a direct demodulator, but I had an oscillator running and no diodes near the audio output, so I don't really understand how I got what I got. It's certainly possible that some of the silicon in the op-amp was acting as a diode detector, but that still doesn't really explain why it only worked while it was oscillating. The frequency it was oscillating at was a bit low for AM radio (~400kHz) so maybe I'd made a superhet? But those are only better than a direct diode detector due to relying on a tuned amplifier, which I almost certainly didn't have, unless one of the two op-amps and its feedback loop's delay happens to work that way. Could be a real simple dip-8 op-amp AM radio circuit if I could get to the bottom of it, but it would be a real pain to engineer.
I'd rather make a CD4046 + NE612 direct demodulator, using an op-amp with a high gain-BW product on the input with no/minimal input filtration.

>>2025281
If it's a linear regulator you get 41% efficiency.

>>2025281
If you're bothering to have a solar panel, you likely want the efficiency of a switching converter. While building your own converter is feasible, if you're not planning on using a proper switching controller I'd highly recommend buying an eBay special buck module. 300mA is easily within a standard <$1 LM2596's range.

>>2025416
>>2025416
>>2025416
NEW THERAD

>>2025260
>>2025248
>>2025244
just to get my feet wet, it was the first project in the book. I don't know how to read schematics or solder or anything. my end goal is being able to make musical hardware, so it seemed like a good gentle start

>>2025281
>>2025396
So I went and bought a batch of these things.

https://www.ebay.com/itm/1x-10x-LM2596S-DC-DC-3A-Buck-Adjustable-Step-down-Power-Supply-Converter-Module/223524573272

My question is, is it possible to adjust these with a PWM signal from an arduino or nah? I noticed the little screw you're supposed to turn but I'm hoping I can use one as an adjustable benchtop unit along with that solar project of mine. Don't wanna have to turn the screw and fiddle around with getting the voltage spot on if I could just have an arduino do it for me.

>>2025955
It's not a bad project to start with, easy to troubleshoot if nothing else. There's a lot of audio circuits out there, both simple and complex, so there's a lot of learning to be done. You should first prioritise understanding basic circuit behaviour, meaning learning what components are and what equations govern them, and developing an intuition with regards to how they act in a circuit. For this, more ideal components will be better, so I'd definitely spend some time learning how resistors, capacitors, and op-amps work. Then you can go on to less ideal components, like transistors, diodes, and ICs like the LM386. Not that any of that matters much when copying a circuit diagram.
Once you have a reasonable understanding of the parts, you can start developing an understanding of different circuit blocks/topologies, like schmitt triggers, inverting amplifiers, common-emitter amplifiers, etc. With any luck, whatever book you're following has a proper progression of learning and experimenting. A breadboard is useful for testing purposes.

Made my first guitar effect a few months ago myself, using op-amps as my main building block. They're very versatile, so I recommend you emphasise learning them on your way to making neat audio circuits.

>>2026207
oh and next thread:
>>2025416