/diy/ - Do It Yourself

>>1905618
first for an /ohm/let breakfast

>>1905618
petition to add marcoreps and be eater into the recommended youtubers list

>>1905621
Eggcelent choice, sir.

man how the fuck do you lads get on with the maths, maths really is my weak point and its super depressing.
think i need to start from literally day one

>>1905683
which mafs giving you problems bruv?

>>1905683
only math i need everyday is frations and manipulating those with n ammounts of variables ....mich i can do on a casio fx991 pretty quickly ...the rest of high level maths you just need to understand to know what to plug into simulators

so which parts of mafs do you struggle with exactly ? ...keep in mind i was horrible at math in high school and when i was applying for college i decided to go to a tutor and try to iron out everything, it ended up stemming from the simplest shit i should have learned years ago but never really listened , so i went on khan academy and such and went from the stupidest shit like 2+2 all the way up to calculus and made a strong foundation for myself in problemsolving ...it took me 3 months of a few hours every day /other day to go through all of it but it was so worth it

Hello.
I'm trying to do something with an electron gun extracted from a CRT TV, but I can't do much unless I have actual schematics of the gun. Online I couldn't find anything and I even searched through 4 different search engines. Is there a way to get the schematics or is it possible they don't even exist to begin with?
The entire TV is a SINUDYNE model No. 14D00 and the electron tube is an ORION A34JXV70X73 (under this there is also the number "0204481").

>>1905683
you need to get good in complex integration, improper integral, series convergence etc. and clearly understand the differences between the fourier transform and the laplace transform and how the latter is a generalization of the former with the important difference that it is a complex function of a *complex* variable and how they help us convert complicated differential equations to simple algebraic expressions. and you need to have good knowledge of complex numbers in general. the euler's formula etc. or you could just blink leds.

>>1905712
Have you removed it from the tube?

>>1905724
Tragic how you need all this knowledge to be an electrical engineer but some retarded 90 iq code monkey can outearn you doing php webdev

>>1905712
First 'extract' the vacuum..

>>1905727
meh it's never been about the money for me, it's about the intellectual challenge

>>1905712
im the anon from the other thread ..i found a few things from google that might coincide with the parts you have

https://partstore.encompass.com/item/6197305/Toshiba/BZ614442K/

https://docs.google.com/viewer?url=http%3A%2F%2Fwww.dragonslairproject.com%2Ftech%2Fdocs%2Fsymphonic_st413e_st419e.pdf&pdf=true

>>1905727
you dont need to know all that to get a job as an electronics engineer ...you have to know that to be a good engineer that invents new things ...and those earn much more than code monkeys

>>1905727
yep and even if you land an EE job chances are you'll still be doing some kind of embedded development that doesnt require much of that knowledge so you will quickly forget most of it.
but the pay will be 30% less than what web monkeys make who didn't need to learn any of that complicated stuff in the first place

t. webdev monkey with an EE diploma

>>1905735
>invents new things
if you only knew how bad things really are

>>1905727
don't worry anon. they'll keep importing indians until webdev pay is on par with nurses and paralegals.

>>1905745
>importing
you mean outsourcing work to them. why import them when it is cheaper to hire them to do remote work. the technologies for that are getting better and faster especially now that everyone works from home. it is easy to outsource stuff but the quality isn't the best and there are various other hurdles such as the time difference.

>>1905751
>but the quality isn't the best
this is an understatement. my employer outsourced a huge chunk of important design work to an indian company with one american to supervise every team of ~50 indians. they had to completely scrap all of it 6 months later and redo it in house.

i had the opportunity to ask a VP if they'd learned their lesson. to paraphrase: "no, the labor costs are just too cheap for us to stop. we'll try again when everyone's forgotten this disaster."

>>1905727
i have been making electronic devices for 10 years now and most of the things that have fed me were stupid shit like blinkers and some MCU sesnor readings and such ...desu i forgot almost all math i learned in uni i still understand why and how everything works but i wouldnt be able to solve an integral by hand if my life depended on it

>>1905757
>i wouldnt be able to solve an integral by hand if my life depended on it
i've actually needed to do integrals at work and i can't even remember how to write one out in the first place. i just use numeric analysis.

>>1905760
i mean i know how to write them out since i know how to dump it in a calculator ..but doing it by hand would be tragic kek

>>1905757
>>1905760
When my dad was in med school apparently the way they figured out the area under a curve was to draw it out on graph paper, cut the area under the curve out, and weigh that, then compare it to the weight of a single square of the same graph paper.

>>1905754
>one american to supervise ~50 indians
Oh my, one tard wrangler for 5 of them might have been less of a disaster
But fuck me I'm definitely moving to a defence company since they have to do everything themselves, fuck working with pajeets again

>>1905757
>>1905760
yes numeric analysis is way to go. you can quickly code a simple integrating function. depends on how accurate you need it to be. it is just an area under the curve. so you define some arbitrary dx and thats your integration step, and then break the area under the curve into small rectangles or trapezoids, etc and calculate and sum their areas. there, done. you can even calculate a double integral boo sounds scary but all that is is just a nested loop. instead of areas you calculate volumes of every little 3d bar with the area being dx*dy and the height is the value of the function at (x,y). and add them together. done.

>>1905772
>But fuck me I'm definitely moving to a defence company since they have to do everything themselves, fuck working with pajeets again
That's a whole other fucking mess by itself. Defense contractor work is absolutely grim shit.

>>1905775
I've interned at one for my studies, everyone was super chill and they had the best canteen I've ever seen, too bad they were located in a city I absolute refuse to live in
I'm not american though

>>1905730
>>1905725
Everything has already been done.

>>1905734
Thank you, I saw that partstore website too but all I could find were unavailable parts of a seemingly identical TOSHIBA TV.
Regarding the service manual you found, apparently it only talks about everything outside the glass tube itself. I believe the glass tube is made by a different company while the rest of the TV is made by the brand (such as Toshiba or Sinudyne); in fact on the crt glass there was a different brand than on the plastic chassis of the TV. However I couldn't find any schematics for ORION glass tubes.

>>1905775
what do you mean? my limited experience with defense r&d work is "hey here's your task. we expect a final design in two years. have fun working on this cool shit!". meanwhile my civilian work has been almost schizophrenic with changing requirements and due dates and forcing you to cut corners and then asking you later why you cut them.

>>1905777
i was hoping pic rel would help you indentify some pins ...and the table under it ....also if you scroll down almost to the end you will see tables with different CRT tubes ...which most should be pin compatible ...you should be able to reverse enigineer something from all that ...i honestly dont care enough to take time to help you doing so because i have other things to do ...but it should be enough to be able to dig further

am i a retard for thinking Mims III, Getting Started in Electronics is really quite annoying to read?

>>1905773
desu 90% of things i do are eyeballing it ...regulators , shunts , capacitors , resistors , feedback circuits , voltage regulating , filter approximations ...everything you can take an arbitrary rule of thumb value and it will work kek ....alsmo most of the things you do are already invented and made in some form of IC ...most of my work in a private firm was taking a datasheet , basically copying the "this is how you connect to work" adding some basic protections circuits and publishing in a fancy box under our name...the sad part is its not even bad quality stuff ...since you can find for instance voltage regulators that can be fried ...you have to fuck up real bad to destroy some new IC-s ...even the fragile MCU and TTL level stuff has integrated protection everything because it costs almost nothing for the manufacturer to implement a resistor and diode here or there

>>1905781
No, it's honestly geared towards teens and people people who never went beyond high school. If you have more than the most cursory knowledge about electronics or physics it's infuriating.

>>1905781
its ok but the drawings and general layout make it confusing to read imo.
it could be a quarter of the size easily and still make sense

>>1905780
I saw that. I believe those are the connectors between the CRT CBA and the MAIN CBA but I don't think signals go directly from the connectors to the CRT socket.

Also I found Pic rel which is all the pins There are on my electron gun. all I have to do now is figure how it works...

>>1905564
I see. But I don't have any sort of embedded IDE at all, I just use Atom.

>>1905584
That's still comparatively expensive to a USB-UART dongle or a USBasp or whatever. I wonder why? Do they not support in-circuit-programming and therefore all need ZIF sockets?

>>1905778
Man, I wanna do defence contractor work too, but my country's military are too cheap for a single fighter jet or tank, so I doubt they do much r&d at all. Also I don't have a degree.

>>1905785
i find the analogy with software quite amusing. we do exact same things in software. in web development we call them "frameworks". those are based on other "libraries". so you look up some examples on stackoverflow how to connect them together and you are pretty much done. and those ICs are exactly that, "libraries". you customize them a little bit and your done. nobody does anything from scratch anymore. everything is "integrated". low level programming is as rare as designing with discrete (or less integrated) components. those days are gone. Embedded programmers that use high level HALs don't exactly know how their MCUs work at the register level. They just learn to call the right HAL functions with the right arguments. Thats almost no different than web development except you use plain C. But even that is changing.

Is there any good guide to figuring out what the critical electrical specs for any given component is (in particular, amplifiers and other linear shit)?

sad

>>1905770
Never thought of that, real clever.

>>1905773
I’m going to keep using euler’s method and there’s nothing you can do to stop me. Programmer time is worth more than processing time.

>>1905887
Learn how the components work, or how they’re characterised. An op-amp’s ideal equation is Vout = (V+ - V-)*G, I_in = 0. Your first variable that you might care about is that G, the DC gain. Then look deeper into how the component strays from those ideals, you’ll get it’s input leakage current, its gain bandwidth product, its power supply noise rejection, nonlinearity, slew rate, how close it will get to the rails, all of that.

behold, the Operational Transconductance Amplifier

>>1905964
>tfw brain too small to get the joke

>>1905887
Datasheet for any particular component is the canonical reference.
If the datasheet doesn't make sense, check the manufacturers website.
If a manufacturer creates certain components, you can find application notes and guides to the components, like here:
https://www.analog.com/en/products.html
Just keep digging through till you find guides.
They also have an education section which might be helpful.
Basically you need to research. The specification which might be critical to your application is something that I don't care about for my design. It depends on what you're doing.

>>1905967
>Operational Transconductance Amplifier
input voltage produces a current.

>>1905962
>I’m going to keep using euler’s method
Well thats about integrating 1st order ODEs. I thought we were talking about calculating definite integrals (area under the curve).

Are there any "kits" I should go for in order to start making electronics? I got the "Make: Electronics: Learning Through Discovery" book but I don't know what I should get along with it, buy a electronic component starter kit?

>>1905982
Can’t you always reformat a definite integral into a differential equation for use with a numerical method?

>>1906013
IIRC the MAKE electronics book(s) are hot garbage. Does it tell you to use a common-collector NPN for turning on an LED?

>>1906016
Not sure what you mean. I think we talk about different things. When you solve a DE, you need to find a function with the given derivative. So you use Euler or whatever to build the function point by point based on the values the derivative takes based on the initial condition. Lets say you need to solve y'=2x. You can use the Euler to reconstruct y=x^2 for y(0)=0. But when you calculate the definite integral you already have the function and you calculate the area under the curve. So you integrate y=x^2. But in the corresponding DE you integrate y=2x to get y=x^2. So firstly you are going in the opposite direction and secondly you dont care about the area so you take an indefinite integral, i.e. you need the function itself not the area.

>math shit
uh. i got a C in calc 3 and really didn't even need to take that for my major
am i still allowed ITT?

>>1906047
Oh I guess that it doesn’t apply if you don’t have some characteristic equation of whatever it is you’re modelling. I come from a pretty theoretical background. But in the case that you just want to find the area beneath a series of data points, isn’t that a trivial task to a computer? It will be constructed with a finite number of data points, for which you can draw trapezoids under. No need to split your data into dx-sized segments, it’s already in discrete segments due to being digitised. Heck, I’m 99% sure that there’s a really simple DSP algorithm that can integrate live, which could easily be modified slightly to allow for definite integration.

>>1906052
hehe you are fine. we rarely talk math here. /sci/ is a more suitable place for that but math major fags look down on EE and especially CS. but at least you can use lateX on /sci/. they tried to start an EE general but nobody posted there.

>>1906059
yeah it all depends on your task. if you are working with data points, maybe you'd need to interpolate in between to improve the accuracy i dunno. depends on the application. i was just thinking about a general task of a numeric integration: you have an analytic function (not a set of datapoints) and you need to calculate the definite integral on a given interval. but i guess it is more of a general math than an EE problem.

Please explain to me how the fuck transistors aren't fucking magic.

How do I account for voltage drift in hall effect sensors. The magnet is not stationary so I cant periodically zero it.

>>1905964
It's a transimpedance amplifier too!

>>1906077
Condensed matter physics practically is magic, at least to almost anyone who has ever looked into it. Even with a uni physics paper of it under my belt it's completely opaque to me. Depletion-mode FETs are almost explainable with knowledge of electrodynamics. If you want to actually understand what you're dealing with, stick to vacuum tubes.

>>1906092
Take AC readings instead? As in, replace the magnet with a solenoid that constantly shits out AC. Then you can slap a high-pass or band-pass filter on your output and completely ignore any drift. Same thing that IR remote controls do, they oscillate at 38kHz or something.
If you want to figure out the amplitude of the AC coming out of the high-pass filter, then you'll probably want to use an op-amp ideal rectifier to make an envelope follower.

Drift also comes from nearby objects of iron, nearby magnets, and how they interact with the earth's field, so there's no practical way to calculate it away.

>>1906136
>If you want to actually understand what you're dealing with, stick to vacuum tubes.
Um hello? Based department?

>>1906137
don't respond to me or my posts ever again

>>1905618
I have 2 devices feeding from the same power supply, one 12V and the other 5V. I learned the hard way that accidentally applying 12V on the 5V device will destroy it. Right now, I have the 12V device connected directly to a 12V battery along with a step down convertor, and the 5V devices connected to the output side of that convertor. It works, but there is no protection for the 5V circuits, which is risky since the devices' PCB's are close to one another. I was going to use pic related, a Zener diode overvoltage protection circuit, but was wondering if there were any IC voltage regulators that would work better.

is an oscilloscope just like a voltmeter that shows a graphical representation of changes with respect to time?

>>1906161
>wondering if there were any IC voltage regulators that would work better.

all regulators involve a voltage drop. older ones will drop like 2V, wheras newer models will drop maybe 0.2V. this is probably bad.
what i'd so is simply add a zener across the line i wanted to protect. a 5 or 10W zener will survive quite a lot of abuse, but a fuse inline will shorten the time that the zener is abused.

>>1906167
>a graphical representation of changes with respect to time?

exactly so. but, unlike a voltmeter, you can watch the voltage change over the course of microseconds, or even nanoseconds.

>>1906172
The project is a wearable device. I'd rather that the IC disconnects the load upon overvoltage, then reconnect it once the overvoltage goes away to avoid having to carry a pack of fuses and constantly disassembling the thing outside. The power convertor produces 5.4V and the 5V device can take down to 3.7V so this isn't super bad IMO.

>>1906167
An important thing about oscilloscopes is that you can see multiple traces at the same time. If you have an input signal and an output signal to some amplifier circuit, you can see what the gain and noise and other characteristics are live. If you have a digital circuit, you can see the propagation delay. If you have a counter, you can check that it's dividing your clock at the right rate. With a storage oscilloscope, you can see what the trace is doing before and after a single triggering event. This is handy for characterising something like switch bounce, or other transient characteristics that don't loop continually.

>>1906173
There exist resettable fuses. They're basically PTC thermistors, put too much current through them and they heat up and their resistance increases rapidly, saving the zener or MOV or whatever.

how is this happening
there is nothing connected to my collector but my 1M probe

>>1906202
oh and adding a schottky diode to turn it into a schottky transistor did nothing

>>1906191
>>1906173
>>1906172
>>1906161
Would a TVS diode like a BZW06-5V8 work?

>>1906206
Well yeah, pretty sure it's what they're made for. Probably something about uses in its datasheet if you're curious.

>>1906202
Increasing the size of my series resistor dropped that 7V to 6V, increasing the size of the pullup made it slightly better again. Would a cascode perform any better than just a single pullup?

https://www.youtube.com/watch?v=-f_h0ogRTjY

Damn, that's pretty neat.

>>1906212
Fuck this, I'm thinking I'll just use a TDA2822 to drive MOSFETs instead. Or use it to drive the TIP31 and TIP32 transistors.

>>1906173
So you already have a step down converter but you're afraid it's going to break randomly and put out too high voltage? Shouldn't you just replace it with one that won't just break for no reason?
>a pack of fuses
fuses don't protect against overvoltage

>>1906226
Ooh, the TDA's output is really good. Seems to transition in 1-2µs, which should be good enough for low-speed MOSFET switching. The TDA2822 circuit is really simple too, just a singe capacitor as my only passive. Pulldown on the input might be a good idea, but we'll see.

>>1906227
They do when you're using a MOV or zener or TVS or whatnot.

>>1906202
>how is this happening

that transistor must have a really really low beta. i suspect that's coz it's been abused, or is a fake.

>>1906242
Can't be abused, I just replaced it. Shouldn't be fake either since I bought it from a local overpriced retailer.

>>1905887
the go to are always max current(input and output) , max voltage and heat dissipation ...everything else depends which component it is ...if you have a specific question for a specific component then i can help ....but otherwise you gotta learn by yourself how a component works and where it can be used and go from there

>>1906013
yes ...the kit is called the electronics undergraduate bachelors degree ...it costs depending on the country you live in ...its generally not very standardised ...but you get everything you need in the kit already like books components and interactive tutoring

>>1906230
Ok my TDA-based circuit works really well, EXCEPT the output sits at about 5V when the input signal is a flat 0V. If I add a resistor from pin 5 (inverting input) to Vcc then it removes this issue, but instead I get an output voltage dipping to 7V at high duty-cycles. It's safer, but not perfect. I can use a larger capacitor on pin 5 than this 22µF to extend the time at which it behaves properly, but that's useless in the case that it's set to 100% duty-cycle. I'll do it anyway (got a 4.7mF cap lying about somewhere), probably doesn't matter since I doubt it will be set close to that duty-cycle anyhow.

should I focus on learning MATLAB, Octave, or Mathematica?

>>1906052
as an Electronics anything,,the math you need to know is the practical concepts for explanation of things that happen in electrical systems , and as a tool for understanding the physics behind it....being very succesful or well versed in solving textbbooks upon textbooks of pure math problems will help in some purely scientific and academic researches, but is otherwise absolutely unnescesary for almost any realworld application

every engineer has a working understanding of math ...but no engineer really likes math or is willing to phylosophise about it ...its just a tool

>>1906077
it cant because you apparently are not smart enough to understand it...there is an explanation in the previous thread that covered everything in most laymans terms possible...anything you dont understand about the explanation stems from not understanding the fundamentals of chemistry and physics needed

>>1906136
Even capacitors are tricky and require some college physics well above 101 to understand what actually happens when AC "passes" through a capacitor.

>>1906173
google mosfet overvoltage regulator

>>1906273
I once went on a futile mission to try to understand why exactly inductors and capacitors are linear devices. It obviously has to do with the derivative being linear but to understand why it is linear in general... Math fags will send you on a wild goose chase by simply giving you the definition of linearity (which is deceptively simple). It is a rabbit hole. I just need to accept it is linear.

>>1906283
i'm a mathlet. what do you mean by linear? just as in linear current changes with constant voltage (L), and linear voltage changes with constant current (C)? because that seems simple.

>>1906283
inductors and capacitors are not linear devices ...you use linear approximations to calculate them for each use case

there is a whole post doc course about non linear analysis becuase its so complicated so we make due with easier but more tedious approximations

>>1906286
For example voltage on an inductor Vl= LdI/dt. How can you tell this expression is linear? For example if your signal V(t) is some random function how do you know there will be no harmonics in the output signal? We know that if your input a sin, the output will be a time shifted sin, with no additional harmonics. How do you prove that?

>>1906288
What makes them non-linear?

>>1906286
no...you have something called taylor series linear approximations which takes a high level polynomial and derivates it in a sequence .....basically any curve can be considered linear (as in a line on a graph)if you zoom into it enough.... so you take some kind of curve(high level poly), choose a point around which you want to approximate and use the taylor series to get a line that approximates moving left or right on that line ...thats is a ver very basic explanation how you analyse inductors and capacitors or any power accumulator and their interactions in a circuit

>>1906293
>What makes them non-linear?

because their response function in laplace space is non trivial

>>1906293
Oh forgot to mention (and I believe we talked about it here) of course you can argue that the derivative of sin is cos which is a time shifted sin, so there you go, done. But then what happens if your exciting signal is something other than sin? The usual answer to that is any signal can be represented a sum of sin/cos (the FT) and then you apply the superposition. Well but why can you do that? Because the derivative is a linear operation! Back to square one.

>>1906296
Why though? You wouldn't even need lalplace transform for a simple case of a simple DE charging/discharging capacitor, for example. Where is the complication coming from? Some tricky physical properties that are neglected in common analysis?

capacitor go zzzt

>>1906274
>it cant because you apparently are not smart enough to understand it

>the virgin MOSFET
>the chad BJT

>>1906202
base resistor too dam low
5mA going thru the base is way too much, make Rb 10k

>>1906300
>Where is the complication coming from? Some tricky physical properties that are neglected in common analysis?
yes

for general analysis of circuits where you deal with mostly theoretical simulations you take an inductor and capacitor and consider it a one quadrant element...in this case you take it as either an impedance (when calculating resistance) or curren or voltage source(when doing kirchoff/thevenin)

but in advanced circuit analysis like system analysis you look at them as four quadrant elemnts that are used t sink and source currents and are affected by frequency all together including their innec resistances and capacitances...they become energy accumulators and energy sources with complex simulations and are non linear.....thats literally the reason why laplace functions exist and are used in this way ...because they are easire to use when claculating non linear devices by transforming them in a linear maths space and then transforming them back

>>1906314
im sorry if it does not make a lot of sense ...my english is not really good enough to translat from my mother language

>>1906302
if you go around claiming MaGIC HRKHRK instead of saying what exactly you dont understand so someone might help you then you are smooth brained and ngmi

>>1906313
its not too low ...if anything a highr base current will make it sink more current ...and i higly doubt 5mA is too much for that trans

>>1906261
>if you have a specific question for a specific component then i can help ....but otherwise you gotta learn by yourself how a component works and where it can be used and go from there
I'm the guy with the Apple II disk drive that's become obsessed with reverse engineering the MC3470, but I can't figure out how to even begin speccing out the analog components based on the datasheet for it or the rip-off HA16631.

>>1906202
>how is this happening

you are using a BJT to do a MOSFET-s job

your transisotr has a small B factor and is probably fried

>>1906322
>small B factor
oof, you hate to see it happen

>>1906304

>>1906321
you will be lucky to reverse engineer a pcb full of discreete components let alone an IC with appareently no prior knowledge but oh well ...i recommend you either find someone to do it for you , find a replacement part off of another old piece of hardware...or put it off for a year or two and learn the basics in the meantime

>>1906325
more BJT were destroyed with static than mosfets in my hands but funny meme nontheless

>>1906330
do you live in a cold climate? i've never killed anything with esd

>>1906325
this is based (or "gated" to you MOSlets ITT) but needs more text

>>1906330
I'm not doing this to fix something, I'm doing it for the challenge.

>>1906314
>..thats literally the reason why laplace functions exist and are used in this way ...because they are easire to use when claculating non linear devices by transforming them in a linear maths space and then transforming them back
I understand the concept of the laplace transform and I don't think its primary purpose is to deal with non-linear devices. At least at the undergrad level the laplace transform is used with linear devices since the primary study focus is LTI systems. In that context, the LT is about various tricky signals (dirac delta, step functions etc) fed into linear devices.
LT converts the time domain to the frequency domain. And the transform itself is again linear. I never thought about it in the context of non-linear devices and I don't really understand what that means. Are you talking about nonlinear (partial) differential equations?

does anyone recognize this chip

>>1906355
just google the first line:
https://www.ti.com/lit/ds/symlink/ucc28019a.pdf

>>1906358
I did but I kept on getting no good results. Thanks.

>>1906368
I just googled "28019a" and it was my first result

>>1904437
>>1904396
I got a new rectifier, connected ac to opposite corners and dc to + and -.
Same thing, 26mV DC on the + and -.

>>1906398
is the resistor getting warm

>>1906399
Stays stone cold

>>1906398

divide and conquer: test xformer all by itself. add rectifier only and measure everything (each diode and output voltage), if ok add caps (they are in parallel, not series, so your drawing is wrong, along with everything else in your life). finally add a load, and measure voltage across, and current thru it.

also, describe voltage rating on caps.

you could also undo the series x-former connection, and use only half the secondary so you get a more usable 24V instead of 48.

anyone got any blueprints or schematics on a super magnet>?
I'm looking to apply a magnetic field with a strength up to about a foot away.

>>1906429
we did it reddit, undoing the tap shows 16v dc after the rectifier (the transformer's 2x15v).

>>1906434
magnetic fields are really bad at traveling any appreciable distance through air. consider a different solution.

>>1906441
question is that how did i fuck the center tap

>>1906470

maybe you got the phasing wrong. see the gray dots on Na and Nb? if you reverse one secondary, you get zero volts, or close (the difference instead of the addition).

>>1906476

forgot pic

My cassette deck plays way too fast. After exploring all the options everyone says it's either a busted motor or a faulting 7812 voltage regulator somewhere. I soldered shit before but I never messed with figuring out the if stuff is getting the correct voltage. I'm gunna see if I can figure out if the motor is getting too much voltage. I just ordered a multimeter. Does anyone have any tips or advice?

>>1906479
>Does anyone have any tips or advice?

i once had that. fixed it by wiring a power pot in series with the motor. then adjusted it while playing a voice recording i was familiar with. (dont use music). if i recall, it was 100 ohms at 10watts, not as big as the pic.

>>1906483
It already has pots to regulate the speed. It's just that I set them to the absolutely minimum they can go to and it's still too fast. I just have no idea how to narrow down whether or not it's the voltage regulators going bad, other than testing the multimeter on the regulators themselves.

>>1906476
Great, ended up getting 37V out of it with a 220ohm load

>>1906277
So long as you understand the quantitative part behind a capacitor (I = C*dV/dt) you'll be doing fine. That's worth more than a fundamental electrodynamic understanding for almost any user, though frequency-dependant complex impedance of the dielectric might be required in some instances. Throw in some hydraulic analogy for a bit of intuition.

>>1906320
No he's saying that the resistor is too low, not the current.

>>1906322
The datasheet shows base currents up to 50mA, and the hfe is 75-375. The whole reason I'm using BJTs is because they switch faster than common MOSFETs, it's a MOSFET driver after all. Though I've never really worked with small-signal FETs much before.

>>1906465
doesn't answer my question

>>1906682
ok

>>1906682
it does ...it says its a futile idea and should not be pursued ...are you dumb or being thick on purpose ?

also you didnt specify what you consider a strong magnetic field nor did you specify what that field should do

>>1906532
lower the voltage by at least a few volts ...a voltage fluctuation will kill your LM317 ...or find a reg with higher input voltage

>>1906756
Im using LM317HV, 47V going into it.
The parallel 4700uf filters are storing a fuckton of energy though, maybe i should figure something else for them

Does anyone here have a TS80 or TS80P iron? Struggling with my TS80, not sure how to fix my problems. Even if I tin my tip, calibrate the unit, and let it warm up, it sucks at heating solder and maintaining a tin. I can't heat up anything at 250C. What should only require 300C pushes me to do 340-360C. I'm using eutectic 63/37 solder. What could I be doing wrong?

>>1906754
its cool lol, im not gonna waste my time playing around with a couple of faggots who don't even know how to answer my question

>>1906787
bye

>>1906787
have sex

today was my first Digital Circuits lab and we had to install this "Vivado" software to emulate an FPGA
working with the software was like pulling fingernails with rusty pliers
please tell me there's actual better software used in the industry
please tell me FPGAs and DSP with FPGAs isn't as arduous as this

>>1906787
ah I see, you must be the "Ideas" guy

>>1905618
Retard question:
What would I call a PCB that's blank?
When I look up blank PCB I keep getting either normal PCBs with nothing on them, or full sheets.


I just need some cheap as piss non conduction shit to use as a hard board holder for something else,

>>1906800
>>1906801
>>1906851
let's count em out now.
1
2
3
faggots that cant help me

>>1906886
like you just want a board made of the laminate with no copper cladding?

>>1906894
imagine being able to count to 3 but unable to understand magnets without help

>>1906886
>non conduction shit

perfboard: but it's like 10x less strong
ABS: like the back of an old keyboard, or the lid of an old scanner
acrylic: see through, but it cracks and is thicker
other plastics: like dollar-store platters and plates, some are incredibly strong
wood: cheap but thick
PCB: use acid to remove the copper

For SMD work on electronics, what is a preferable solder diameter between 0.5mm, 0.6mm, and 0.8mm (0.020", 0.025", and 0.031")? Any recommended 63/37 spools in these gauges? Used to 0.5mm which seems fine, but my spool is trash quality. Would like to buy a 1lb spool and be at peace.

>>1906848
>please tell me FPGAs and DSP with FPGAs isn't as arduous as this
why do you think they pay so well?

>>1906787
You can use mathematics to calculate the number of amp-turns required for a given magnetic field strength at a certain distance and given dimensions. 1 tesla is considered strong for a magnetic field, so you'll definitely want something no weaker than 0.1T. But magnetic field strength falls away with the inverse square law at short distances and the inverse cube law at long distances, so the magnetic field strength inside the solenoid will need to be much stronger than that 1ft away.

Here's a link that should work:
https://physics.stackexchange.com/questions/164797/how-to-get-the-magnetic-field-strength-in-space-near-a-solenoid
Just plug in the values, looks to be a relatively straightforward integral, trivial if you do it numerically.

>>1906886
FR4?

>>1906932
I'd go for smallest.

>finally finish troubleshooting broken parts and resoldering other parts
>assemble everything back together again
>hook up the battery
>turn it on
>sparks
>the battery was backwards
FUCK

>>1906950
>I'd go for smallest.
How do these two items look?
https://www.amazon.com/gp/product/B00068IJWC
https://www.amazon.com/gp/product/B008ZIV85A
Earliest shipping for free I can get is next Thursday though (same with every solder spool). Should I even bother grabbing more flux if I have a bunch of Chinese counterfeit RMA-223?

>>1906932

if you wanna put a dab on a single pin of an SMD chip, you need 0.5mm solder. but for most other stuff, 0.8mm is better coz you dont have to shovel an inch of it to fill up the joint.

>>1906971
I'm doing small diode pins and temp-sensitive LEDs at the moment. I'm comfortable with the 0.5mm I'm using now (it's just low quality), but I don't recall if I've used 0.8mm or not. Still not worth it? I don't mind "shoveling an inch" if the stuff at least works. I'm more unsure if I need more flux or not, since I just have counterfeit RMA-223. I have family that has spare flux if needed as well, so I guess I can use that too. Just trying to be wise with my spending.

>>1906971

proper flux works in almost-magical fashion to remove solder blobs, and to direct solder to where it should go. if you're happy with the magical properties of your current product then no need to change, otherwise get the brand name stuff. personally, i hardly ever need it. the flux inside the solder is good enough 98% of the time.

>>1906269
octave is basically matlab, but I think learning in matlab is more useful because you'll probably use it in work.
I've only seen one professor use Mathematica and it was to show some integral or smtn i dont remember. Dont know if its really useful for ee

>>1906077
Had electronics 1 last year and it really opened my eyes regarding transistors. We covered A LOT of theory about semiconductors in gemeral and transistor theory just falls into place when you start to get it.
I never got zener diodes though. Quantum tunnelling still messes me up

>>1906986
>Quantum tunnelling still messes me up
the way I learned tunneling was, for a sine wave particle in a box, the wave starts at 0, ends at 0, and that's it.
But for a cosine-in-a-box, it starts at 1 and ends at 1. So when the wave crashes into the side of the box, it has to exponentially decay from 1 to 0.

Am I right in guessing the distance tunnelled is how they set the voltage drop?

>>1906764
high input caps especially after a gretz is prefferable and should be left right were they are since they cant hurt

>>1906787
are yo uthe " igot a great idea just need to get it patented" guy from another thrread ? ...go fuck yourself and learn how to use google to find literature about the topic if you are so smart

>>1906886
fr4 laminate is the "plastic" part of the board on which you add copper traces and such ...a fr4 board with a copper layer over it that you etch yourself is called a photoresist fr4 board ....the board that has holes with copper around them is called a prototyping perfboard

>>1906848
universities usually use software that is either free...they got in a academic licensing or is something the prof personally preffers for one reason or the other ...so yes there are other software out there and you have to find it yourself, i suggest going to eevblog or somewhere where you know people work in the industry and ask them what they use professionally and learn that

>>1906951
i always check twice for polarity even on boards that are labeled by the traces ...and always check all capacitors before ever plugging in

>>1906269
you should focus on learning math and electronics and use whatever tool you end up thrown into ...businesses use whatever they get for cheapest and most of the time changes from one license to the next ...all of those work on the same principle and getting into any of them takes about a week to a month at most so its basically unimportant whic you choose

Is 70% isopropyl alcohol insufficient for cleaning off flux residue? I used some counterfeit/Chinese Amtech RMA-223 flux recently, which left a lot of residue that has been difficult to get off of my PCBs. All I've had is 70% isopropyl alcohol. I don't think finding 91% rubbing alcohol would be easy right now considering current events. Advice?

>>1906986
most of the time when people say omg this is magic how the hell does it work i can never wrap my head around it ...are people that never went to school and learned what they need , or are either people that went to school and never listened to the classes...from the 120 people in my major only those that were either stupid or didnt do the work and were drunk half of the time or were sleeping through it the other half of the time didnt pass the classes and were bitching about "this is too hard"

If I complete The Art of Electronics, will I be an electrical-engineer-level electrical engineer?

>>1907030
>Is 70% isopropyl alcohol insufficient for cleaning off flux residue

it depends on the flux ...but it should work well enough ...do you have a medium hard bristle brush ? ...you can use an oil paintings hard brush clipped short and rub on the contacts where you want to clean it with alchohol, or buy the same thing for a bit more money as an electric brush thing .....also i use contact spray for cleaning my boards ..it is a stronger cleaning solution and drys almost as fast as isopropyl

>>1907032
not really...you might understand electronics on a basic to intermediate level tho ...but just understanding electronic devices on a basic level doesnt make you an engineer ...it makes you a techncian

>>1907032
just knowing how a circuit works does not make an engineer ...here are the definitions of both stages of an engineering degree from my faculty


>bachelors degree of electronics engineering (real engineering degree)
The study aims to train students to apply basic and specialist knowledge in electrical engineering, identify, design and solve problems in practice, apply other acquired knowledge in engineering, mathematics and computer science, use modern engineering tools, understand teamwork and effective communication, understand ethics and ethical responsibilities, understanding the impact of engineering solutions on society and the environment. The graduate student of this study must be able to engage in continuing education and professional development, and visit a broader education (knowledge of the topic outside the technique).

>masters degree in electronics engineering
students acquire the necessary narrow-specialist knowledge in the above areas and are thus trained to perform the most complex engineering tasks based on a scientific approach to problem solving. New specialist knowledge in electrical engineering and the ability to apply it are acquired, as well as knowledge and application of other specialist knowledge in engineering, mathematics and computer science. Students acquire the ability of continuing education and self-education, the ability of independent research, discovering new knowledge, preparing and performing experiments, and interpreting data. The study acquires the knowledge and competencies needed to design new systems, components or processes, and to act effectively in the role of team leader.

>>1907034
>do you have a medium hard bristle brush ?
I can find one. I was trying to use cotton swabs and thick tissue material soaked in the 70% isopropyl. Results were so-so. I found my fingers did a better job in some cases, though that is of course not the most desirable. My concern is small amounts of of residue getting under conponents like diodes. It took a long time, but I did manage to get them vaguely clean.
>contact spray
Do you have any you recommend for PCBs? How about this?:
https://www.amazon.com/CRC-05103-Electronic-Cleaner-11/dp/B000BXOGNI
Really cheap, good reviews.

>>1907041
Did you try acetone?
Just need to be careful around any plastic parts from connectors and shit.

>>1907041
in europe pic rel is the golden standard for cleaning electronics and contacts
https://sa.rsdelivers.com/product/crc/70013-ah/crc-kontakt-60-400-ml-electrical-contact-cleaner/8232662

also get one of these to clean boards .the horse hair brush is the most effective in my experience and can be DIY-ed from arts and craft shops just get a high quality horse hair brush and shorten the bristles
https://sra-solder.com/cleaners/cleaning-brushes

>>1907043
I didn't try acetone. I'm unsure if I want to grab anything too heavy like that. Whatever makes the job easier without costing a fortune is in my interest. Just don't know how much residue can be left behind safely.
>>1907045
Thank you for the notes. I'll see about modifying a brush or two I own for next time.

>>1907039
How would this not be easily acquired by texts?

>>1907048
also get a no clean flux/solder ..its not that much more expensive and its not corrosive ...you should still clean it a bit since it stays sticky but its not gonna need scrubbing since it wont eat away at anything ...also i dont know why people drill on about cleaning flux awaay , its not as destructive as it was 50 years ago and i personally rarely clean it off and have had devices for 10 years working in industrial situations without failures due to corrosion caused by flux, just dont use the " looks like shoe polish tin can" one that stings your eyes and nose lmao

Has anyone in here made enclosures for their electronics out of Polycarbonate? I know acrylic is common for the sake of lasercutting, but I don't want to spend much, and was reading some instructables on bending polycarbonate (and it seems promising). I'd like to hear about any gotchas or encouragements about how surprisingly easy it was.

>>1907049
it can and it is most of the time, but he asked for one specific book ...also universities have laboratory practice that allows you to use instruments and devices that you most likely dont have at home and allow you to have concise and accurate representations of real world problems

but otherwise yes , you can absolutely get all the knowledge nescessary to be at an engineer level by yourself , and i put the full definition so you can see what exactly you need if you want to .....but at that point why not just go to college and actually get the official degree ..also just an fyi it will take you much more time by yourself to get to that level (thats why schools exist and professors teach and tutor)

>>1907052
i have tried doing enclosures by printing pcb boards and joining them as a box ...or buying a half open meta/plastic enclosure and making the front and back panels from pcbs (usb cutouts for instance or holes ...and making graphics with silkscreen)

there are some plastics that bend well with hot wire formation also

>>1907051
>also get a no clean flux/solder
Is this fine?: https://www.radioshack.com/products/rosin-soldering-paste-flux
Labeled as "non-spill paste". It's the stuff my father's used. I've got some of an older portion of it that I borrowed to try tinning my tip with yesterday, and it seemed to work rather well for me. Thanks again, Anon.

>>1907039
Yeah, my university had lots of nice words about how edumacated we would be by the end, but then it was just dicking around in netbeans for 4 years.

>>1907058
i ususally use this stuff
https://uk.rs-online.com/web/p/flux-removers/1462863/

but i do smd and it makes it easier to dose up when you put it in a syringe holder and glob it on ....but yes any new flux should work for general purpose stuff

>>1907063
a lot of my peers also talked about learning nothing in college because they didnt even try or didnt do anything outside of the university building ...i used the college environment to find appropriate literature and asked any and all professors if i can try and join some of their researches or at least see what they do...i joined some clubs that were active in trying to do university level projects like solar panel maximum power tracking simulator , or doing ECU for formula student cars ...also i used the connections i got from all that and got a nice internship that ended up hiring me and teaching me the rest too untill i decided to get a loan and oppened my own business....i started from a family in which i was the first with a college degree with not a lot of money (before someone says im privileged) you get what youput in from collegew and is probably the most satisfying part of the education journey because its the only point at which your effort actually ammounts to something (unlike elementary/middle/highschool)

i respect choosing to not give a lot of personal drive and effort in something like going to university but your personal prefferences and choices might end up discouraging to other people and perpetuate the notion that college is a waste of time and money and nothing good comes out of it (again i respect the choice of not going to college because it is absolutely not nescessary and you might have your priorities elswhere or might be interested in something else )

>>1906951
Welp fix it again.

>>1907071
My god this. I didn’t learn this lesson too well in undergrad, and it was only when I hit law school that I really learned the extreme importance of making good use of your time at uni.

Even then I regret not making better use of it.

>>1907065
Thanks again, Anon. It helps to get some straight answers on this stuff. Wish me luck on my SMD journey!

>>1907073
our family was not wealthy so we didnt have much toys so i usually tinkered since i was a kid and read any book i could find ...which was usually encyclopedias or technical books ...when i first got internet at age 14 a whole new world oppened and i could expand on what ui learned and tinkered some more, it helped me that i learned and liked to learn by myself and get more knowledge in my own personal time because i had a passion and interst in electronics ...i personally never understood people going into unversity , doing nothing and then blaming the system for their own failures...some of my friends including myself trhived in the academic environment because we were used to doing things "academically" anyways ....lower levels of schooling were horrible for me and i struggeld really hard with motivation because it wass all so stiffling...at uni you have the standardised test and everything because that the fastest way of grading , but thats only the smallest parts of the whole pool of opportunities

>>1907071
I mean, that all sounds like shit you could do without paying thousands of dollars per semester to go to worthless classes and live in a 50sqft apartment. I did some interesting projects (to me at least) back in my college days too, but none of it was "college work".

>>1906897
Correct, just blank nothing
>>1906931
Fantastic list, thank you very much
>>1906950
>>1907023
And now I know what I was originally thinking of is called Fr4, fantastic.


Thanks diy, got to love it when you actually get not only your question answered but even better info than you wanted.

>>1906787
>>1906682
>>1906465
>>1906434
9/10 chance you not only have 0 experience with anything remotely on the topic, but will never and have never build anything.

This reeks of reddit tier highschool ideas that never turn into anything physical.

>>1907111
>that all sounds like shit you could do
Most people can't. The people who go to college and later claim they could've done it themselves are usually ignoring the the intellectual development they underwent and studying skills they acquired in college, instead selfishly attributing everything to their innate intelligence and self-discipline.

Can you use hot glue to cover up some 30 amp hot leads so you don't kill yourself?

>>1907142
dunno mate, let us know if it works

Does anyone have experience with the fedevel courses on high speed PCB layouts?
I would like to try it but I dont have Altium. The 15 day trial version wouldnt be enough and I dont know how secure cracks really are.
Stuff like Catia V6 and the newest solidworks communicate with the cloud and dasault system's servers.
Not sure if the newer Altium programs do the same.

>>1907172
>I dont know how secure cracks really are.
i don't run my pirated altium in a vm, but i do run other software suites (like my CAD) in a vm from companies that are known to be litigious. you could always do that.

>>1907115
FR = fiberglass resin
4 = thickness

>>1907175
are you using vpns on your vm?
also which altium version are you using?

>>1907183
zero network connection on the vm, just a shared folder. as i understand it shared folders are a vulnerability that someone could try to intentionally exploit, but i don't give a shit about russians, i'm worried about the software phoning home.

>version
17.0

>>1907184
ok so I'm guessing there is no internet connection required when installing Altium 17.0?

>>1907111
not necessarily. Being able to mooch time off of professional equipment, not to mention find peers/mentors who know they're shit is not trivial
also, some fields of STEM are less amenable to self study than electronics. I would have either died or achieved nothing years ago without proper lab guidance

>>1907188
my cracked copy did not. i'd be suspicious of any cracked software that does need an internet connection.

>>1907142
if the voltage is low enough so it does not conduct through the glue then yes ...but i would need to know the voltage to give you an estimae ...otherwise dont do hot glue if possible , two component epoxy is better and more temperature stable

>>1907178
fantastic, my stupid ass was under the impression that it was silicone.

>>1907193
35V 30-40amp
>, two component epoxy is better and more temperature stable
Nice, i'll just do that then, thank you

>>1907200
35V isn't high voltage and epoxy is overkill.

>>1907197
Now you know. Silicone is rubber, silicon is a crystalline mineral. Silicon is used to make integrated circuits which sit atop fiberglass resin boards with copper adhered to it.

>>1907200
>>1907193
yeah then hot glue is ok ...just dont let it get too warm so it doesnt melt away

>>1907205
Jesus actual Christ I thought Silicone was made up of silicon.

I thought it was like silver paste where it's 99% silver and 1% something to make it flow or whatever, I always just assumed that silicone was 99% silicon with some flow shit added.

>>1907227
>I thought Silicone was made up of silicon.
It is, but silicone is processed into a semi-solid state (gel, adhesive, rubber, etc).

How the FUCK do you desolder stuff that uses rohs solder, I'm trying to clean up an old mechanical keyboard because it's filthy and the stainless plate rusted so I need to desolder all the switches

The problem is this fucking lead free solder they used, I have to set my soldering iron to 480c and even then it barely gets it flowing after a while, I can take some away using solder wick but as I do that it stops flowing eventually, I had to literally switch to a conical tip and jam it into the hole to get it to flow

I managed to desolder 6 switches by pulling them from the other side while heating them but it's a pain in the ass, it took me over 2 hours and I mangled the plastic on the switches + burned off part of the pads on the pcb, there's gotta be an easier way?

>>1907227
>Jesus actual Christ I thought Silicone was made up of silicon.
I have a fun story for you.

When I was 19, I took a course over the summer at uni. A friend I had from a college job was down there too and we'd hang out on the weekends while he put a high-end sound system in his car, which involved putting a 0 AWG wire through his firewall. To seal it against the elements, we got silicone sealant (like the clear caulk you use for fishtanks). I didn't know this about my buddy but he was one of those guys that tastes stuff. I caught him taking a taste of the silicone sealant and was like, "What the fuck are you doing man?" "What it's just silicon, like sand." "Dude it's sili*cone*." "Oh. That's different?"

Smart guy otherwise.

>>1907240
drop the money on a vacuum desoldering pump. i'm an absolute fucking newbie to desoldering and fucking destroyed several pads on a pcb trying to use one of those sucker bulbs and irons. picked up an actual desoldering pump and haven't damaged anything since.

got my DS1054Z today lads

>>1907240
>desoldering
just throw it out breh. ain't worth the effort

>>1907258
>ain't worth the effort
yes, you must consume the new product

>>1907262
>there's gotta be an easier
trust me, I used to be like you. When the plastic starts melting and the pins get misaligned, it's fucked.
judging by your post, it's not just one pin - it's tens of pins. That will take genuine talent and expertise to pull off without frying your board.
It'd dead, jim. If you can get a new board for anything less than $40, it'll be a no-brainer. Desoldering is really, really fucking hard. But if you're adamant about continuing - use flux, a heat sink, and apply a tiny blob of fresh solder to get it flowing. And practice on scrap before you fuck your precious keyboard

>>1907268
He could use a blowtorch on the board and smack it against the edge of a table to get it out.

>>1907270
>holding it like a pencil
I did that accidentally when I was like 12. Grabbed the iron without thinking like a pencil. I still have a scar on my middle finger.

>>1907262
you add regular or low melt point solder to it... and most likely you ll still have to pull it from other side. just dumb it if you not enjoy it, time and effort is more valuable that 100 a buck keyboard

rolling for a project

>>1907296
Hmm what would "automatic etch-a-sketch controller" be?
I'm from EU so I don't really know what etch-a-sketch is

>>1907253
don't be a lazy shit like me and put off cracking it for years

>>1907300
An etch-a-sketch is a toy that draws a line on a surface. You control where the "nib" is by turning two dials, one to control the X position and one to control the Y position.

>>1907300
If it makes it any easier to understand, think of how an X-Y table on a milling machine works. An etch-a-sketch does virtually the same thing but to achieve a goal that's orders of magnitude less interesting.

I'm a complete beginner and I've just bought my arduino, all these 'rolls for a project' possible with an arduino?
Also, roll!

>>1907315
Linear and switching PSU probably not. Or at least I don't think it'd be easier or more functional.

Is it unsafe to stick my oscilloscope probe into a mains outlet?

I bought a 5 m strip of addressable leds on ebay and also a bluetooth controller and i am trying to figure out if one of them is bad as i am getting no lights. I have the wires attached as in the picture, 5 wires, a red and black, and then a separate group of three white wires which go to the controllers molex which makes them white, green, and red. I just want to know if this at least looks right before i start trying to get the controller or light seller to see about replacing pieces. Most i can get is like two random lights to turn on and then they usually fade. Power supply is 5V 2.5A as the led lights wanted 5V and the controller wants 5-12 V.

>>1907030
It takes some practice to get good at cleaning stuff. You might be smearing isopropyl alcohol around the board which would then just evaporate and leave all the flux behind. You have to soak up the flux with your swab. I found it works better to rub more aggressively on the areas that are fluxy. At work we have foam tip swabs which I like better than the cotton swabs.

https://harmonycr.com/swabs-applicators/puritan-foam-swab-cylindrical-tip-4-polypropylene-shaft/1804-pf?sku=P1804-PF&gclid=EAIaIQobChMIseWhuI_i6wIVUr7ACh2hnwPfEAQYCCABEgJcJ_D_BwE

I know FPGAs and DSP go together like bread and butter, but what do DSP engineers actually implement on FPGAs for DSP

>>1907178
This is incorrect. FR stands for “fibre reinforced”, it doesn’t specify fibreglass. FR-2 is paper fibre reinforced resin, which is why it’s brown.

Sorry for the basic babby question. I'm looking to power a small series of 2 flickering yellow LEDs.

Right now I have a simple 2 AA holder, but I quickly realized they don't have enough power to run both (needs 4v, only supplying 3. they don't even light up.)

My plan is to either use a 3AA holder to bump me up to 4.5v, or a 9v battery.

I did the math (google) and figured out that with 3 AAA batteries, and 2 led lights in series i'll have .5v left over, each LED uses 20mA, so I should need a 25ohm resistor.

Obviously I would want the 300ohm resistor if I go with 9v, but is there a point to even putting on a 25ohm resistor in this circuit? the device I'm making contains lots of plastic and wood, so I'm just worried about it burning up or something.

Here's my shit drawing.

RF anons: does this look like I/Q data?
I am reverse engineering a device and these pairs of 32 bit words are sampled and used to do a FFT to calculate the bandpower in a given frequency range
Does that sound plausible?

>>1907360
>is there a point to even putting on a 25ohm resistor in this circuit?
Yea, Small changes in LED voltage can lead to large changes in current because they're not linear like resistors. Even if they survived with no resistor, the brightness would be sensitive to small changes in battery voltage

To run from 2 AA's You could put the LED's in parallel, each with their own resistor. Downside is 40mA draw instead of 20mA

>>1907391
>To run from 2 AA's You could put the LED's in parallel, each with their own resistor. Downside is 40mA draw instead of 20mA

That would just mean lower battery life, right? I mean is there a practical reason not to do it in parallel instead of series? I could just put 150ohm resistors on each LED

>>1907399
>That would just mean lower battery life, right?
Right

>practical reason not to do it in parallel instead of series?
Yea, as you said, lower battery life. Also you need a resistor for each LED instead of just one

>150ohm resistors on each LED
Resistor Voltage = 3V-2.1V = 0.9V
R= V/I = 0.9V/.02A = 45ohms

>>1907240
Tin your iron and blob up a bit of melted solder on the tip of your iron before you go to heat the joint. Add leaded solder to the joint to help it flow a little better. Use a solder sucker for big through hole components. Wicks are only good after you've taken care of the big stuff. Dunk your solder wick in some flux. Put flux on the joint. More flux makes everything easier.

>>1907240
I tend to just melt solder and blow it off with a straw with the end pinched

>>1907366
Well, a lot of the half-words are FFFF or 0000, does that make sense? I’m not very familiar with I and Q signals, might have better luck with that on the ham radio thread.
But I do have experience with graphing shit on python. I’d suggest writing code that converts whatever input wave into I and Q signals, and then into words. Maybe use audio files as the data that you modulate to create the source signal, plus some noise.
The other method is to take your existing data and write a code to turn it back into what it should have looked like before being turned into I and Q signals.

>>1907411
Huh, neat trick.

>>1907332
you can stick the probe anywhere you want as long as it's under the voltage rating that's probably marked on the front of the scope. you *can't* connect the ground clip anywhere though, since it's tied to mains earth through the scope.

>>1907406
Thanks for the help. I am not very good at math. Forgot to remove the voltage drop on the LED.

>>1907434
>Well, a lot of the half-words are FFFF or 0000, does that make sense?
If they are signed 32 bit ints, that means they are close to 0. e.g. +10 is 0x0000000a and -10 is 0xfffffff6
Note that the ffffs and 0000s are all in the even columns (most significant bits), not the odd columns.
Unless you are massively oversampling, plotting might not be terribly helpful since it might look like noise until it is properly decoded.

>>1907434
>>1907475
Yeah I think I’m going to need to get an SDR and maybe a car FM transmitter to compare readings in order to debug it

>>1907528
Not sure if you need to, but 32 bit ints are gonna be hard to get from any SSR I know of. Or any ADC, for that matter. Could crush down the bandwidth and over sample with some noise if you want, but even that’s only good to an extra bit or two.

I find it pretty hard to believe that anything analog could actually output 32 bits of resolution, so I’m guessing it’s just an oversized comms stream from a lower resolution ADC.

I’ve never worked with GNUradio before, but I’ve heard it’s the only real way to mess with I and Q signals from an SDR.

Wait a sec. Why are these diodes facing that direction if the signal comes from the read head? Don't diodes block in that direction? Or am I retarded?

>>1907555
it's an AC signal, the diodes will conduct for the negative half rather than the positive half that you're used to seeing. i don't know shit about read heads, just going by the schematic.

>>1907550
It ANDs them with 0x1fffff later on so I think they are actually only 20 bits each something

how to keep project budgets under control ._.

Two parter question here,

How do I source parts that are not made for retail? I am looking at a Winstar OLED display specifically bc it uses their own WS0010 controller that can switch between alphanumeric and graphic mode addressing, but I reached out to the company and they don't sell to individuals

And what is the likelihood, if I get my hands on their display, that I would be able to modify it to suit my needs? To give some detail, the WS0010 can only support up to 100x16 pixels but they have a 20 character, 4 line display (each character is 5x8) so on this display, there must be TWO WS0010s, although they don't seem to say there's two controller ICs. My guess is that they are in a master-slave configuration so that the first controller automatically passes data to the second controller when its buffers are full. However, I'd like to be able to address each chip individually so that I can make pretty pictures in a 100x32 display. Do you guys think I can somehow "un-engineer" this logic and gain access to the second controller's chip select pin?

>>1907590
>How do I source parts that are not made for retail?
Very often the same way you get other electronic components: Through a reseller of components. The reason they don't sell to individuals is that you're only buying one. They want to sell a hundred or a thousand at a time. And then places like Mouser take that stock and sell them to people that need one or two.

Of course if you can't find someone that carries it you're probably SOL. Sometimes you might be able to talk to the sales department at a reseller and they might be able to order it if they think it'll sell for them.

>>1907341
Bumping this xause fpga is awesome
absolutely loved labs where we learned vhdl

>>1907573
>20 bits each
You mean 12? Even 20b is pretty high, while 12b is rather common. 8b is even more common for lower-end SDRs.

>>1907590
Look for a datasheet of the WS0010. You might not need the other controller's CS pin if you can use the existing master-slave setup to feed data through the master into the slave. Not like the existing arrangement shows the same information on both screens or anything, lots of lower-level display divers like that are designed (like one of those meme programmable RGB LEDs) to be daisy-chained.

>>1907595
Makes sense, must be why the few I've seen on reseller sites charge you up the ass. Unfortunately there aren't a lot of green on black OLEDs with dual mode driver ICs around

>>1907600
Yeah I figured the same but it looks like based on how the graphic mode addressing works, the Y address indicates the current line and is only controlled by a single bit. Then the X address is controlled counting up from 0x80, and at each X address the data bits determine the vertical pixels which turn on. Hopefully I didn't explain that like dogshit, but the 20x4 display will most likely work fine with text but I'll be limited to the top half for any graphics. For what it's worth I'm just trying to build an over-engineered, fallout-inspired Geiger counter so its not like I NEED this display, I just think it'd be dope

>>1907590
>How do I source parts that are not made for retail?

one tried-and-true method is to pretend you're a manufacturer and ask for engineering samples. if you actually have a job and can get deliveries at work, this makes it easier.

>>1907576
>how to keep project budgets under control ._.

instead of overtime pay, get your daughter to blow the engineers.

>>1907629
could just do it myself but i don't work with engineers

>>1907555
So uh. I just tried to scope this at pins 1 and 16 (non-inverting input and non-inverting output for the diff amp). It didn't look like there was any 100x amplification.

But the drive boots fine. All I can figure is I must've fucked up probing somehow.

>>1907633
That or maybe I just don't understand wtf the preamp does?

>>1907611
Sounds neat to do. You're going to use a microcontroller, right? Now that I'm more MCU savvy maybe I should get back to my SBM-20. Got a few cheap 7-segs and 74HC595s lying about that would work for the counter side. Ideally I'd measure the length of each pulse, since I think you can calculate the energy of each incident particle like that, instead of just counts per second. That way I could swap my display from Bq to Gy or whatever other units. Both "per second" units and cumulative units.

>>1907555
That's a bridge rectifier, sort of. It only works because of the centre-tap pulling the sense-current to GND. Because it's a differential amp it results in a normal AC output like what you'd see before the rectifier, I think.

>>1907634
>That or maybe I just don't understand wtf the preamp does?

it has a differential output so presumably you'd measure the signal across 16 and 17. (if both your scope and the equipment are grounded to earth ground, you'd be shorting out one output pin, so make sure one of them isnt.)

Just built my first guitar pesal, bros. I used someone else's PCB design which used three different PCBs (one for the effect, one for the I/O, and one for the foot switch) and that design was AIDS. It would be so much easier with one PCB, I need to learn to design that

Wondering if anyone can help me out with VFDs and motors. I got a nice piece of industrial machinery with 220v 3000w motor. I want to run this machine at home on 110v 1000w because wiring, which is still plenty for my needs. Do I need to buy both a new VFD and a new motor?

>>1907550
Anything can put out 32 bits if you just sign bit extend whatever you have. You could even make it 64 or 128 bits!

>>1907757
> Do I need to buy both a new VFD and a new motor?

if you dont mind a bit of noise, you can hook up a universal motor, like from a vacuum cleaner (costs $6 at the thrift store), and use a simple motor control which is essentially a more powerful version of a light dimmer.

>>1907794
It's a CNC machine so unfortunately the motor needs to be a very expensive waterproof and precise one with a toolholder at the end.

>>1907811
the wattage requirement is probably justified. if you're american i would think you could run it off 220v split phase that every house is already wired with for the stove, but i'm not an electrician.

Whats up gamers.
Ive been trying to fix this old AM radio but i cant figure out whats wrong. When i plug it in it works fairly well, but as soon as i connect the case to ground with the dedicated cable theres a small explosion and a sharp burning smell even before i flip the on-switch. I opened the case and found this grey unconnected cable coming from the rectifier. I also found a resistor that seemed burnt and cracking, but when i measured it and got 1.3 mega ohm so i dont think its broken.
Any clue what this grey cable might be used for, or even what the connectors of the tube might do? The rectifier is a Philips miniwatt AZ1.
Ive been looking for circuit diagrams or similar for it, but i cant understand them since i have no experience with vacuum tubes, and it seems a bit more complicated than modern electronics. I _think_ the grey cable might have came from one of the pins to another vacuum tube, a philips miniwatt EL3N, but i cant be sure.
Pictured is the underside of the whole transformer/rectifier assembly. In the top left connected to a capacitor and the two resistors is the grey cable i was talking about.
Appreciate any help.

Been working on a tube amp kit for the past few weeks, I managed to get it put together in less than a week but the left output isn't working. I've been trying my best and I've gotten some of the test voltages ironed out, but I'm still getting about 3-4 volts high on test points 18-19. Odd points are left side, even right. Any ideas on what the problem is?

>>1907851
Here's the top of the main pcb.

>>1907852
Here's the backside.

>>1907854
And here's the schematic.

>>1907856
Sorry, meant 3-4 volts high on 19, 20, 23, and 24.

I bought an exhaust fan only to find out it apparently takes AC power.
I've never heard of an AC exhaust fan, does this just have a rectifier in it or did I fuck up?

>>1907865
>I've never heard of an AC exhaust fan,
????

>>1907240
I was able to finally desolder all those switches by adding extra leaded solder like someone suggested and then pulling from the other side, there's a trick for these switches since they have two latches, you need to press down on the top latch and pull it away from the pcb, then it starts flowing and you pull it away a little, afterwards push the bottom latch in which is kept in place by a little mechanism and re apply heat to both pins this time pulling from left and right with your plier, this way I can remove the switches without damaging them

The PCB looks really fucked though, it's corroded and I can't seem to remove it with isopropyl alcohol, pic related

Is it over?

>>1907666
>(if both your scope and the equipment are grounded to earth ground, you'd be shorting out one output pin, so make sure one of them isnt.)
well they're definitely both grounded. hmm.
instead of hacking something together to make this safe, could I just probe both pins with reference to ground and use one of the math functions in my scope to achieve the same effect?

>>1907878
Flux is good at cleaning corrosion. You can kind of gently rub it with your iron. You should be able to get the corrosion off then evaluate if there enough trace left connect to, otherwise youd have to run bodge wires.

>>1907821
The motor doesn't need to be that big, there are CNC machines with smaller motors. Houses with gas stoves don't have 220v.

>>1907910
>could I just probe both pins with reference to ground and use one of the math functions in my scope to achieve the same effect?

i dont see how that's doable.

you can easily unground both items using two of these, or breaking off the ground lug on a dollar-store extension.

>>1907910
>could I just probe both pins with reference to ground and use one of the math functions in my scope to achieve the same effect?

i dont see how that's doable.

you can easily unground both items using two of these, or breaking off the ground lug on a dollar-store extension.

>>1907878
>The PCB looks really fucked though,

looks dont affect functionality.

>>1907940
>Houses with gas stoves don't have 220v.

pretty much every house in N America has it. heating, some A/Cs, and dryers need it as well as stoves.

>>1907945
>pretty much every house in N America has it. heating, some A/Cs, and dryers need it as well as stoves.
It's just simply not true of older houses.

>>1907945
>i dont see how that's doable.
>you can easily unground both items using two of these, or breaking off the ground lug on a dollar-store extension.
Yeah I was just hoping I could take one pin and subtract it from the other or something and it’d work lol. Guess not eh?

I probably have one of those shits in the basement. Should I put the computer or the scope on it? Or does it not matter?

>>1908008
>Should I put the computer or the scope on it? Or does it not matter?
Oh you said both. Pardon my reading comprehension lol. It’s a football gameday and I’m already half in the bag.

I'm simulating a circuit in SimScape and I'm not sure what's going on here why is the 3ohm resistor showing that strange, wobbly current?

>>1908009
>Oh you said both

doesnt have to be both, but doing both decreases chances of mishaps. for example, i have my function generator sitting on my scope, so it's making metal-to-metal contact. so even if i ungrounded the scope, it would still likely be grounded thru the generator. similarly you may have externals like printers, cassette decks, etc, connected to the computer which forces it to be grounded thru them.

Is there any downside to a 2-pin switch instead of a 9-pin? Why do they make them with 9 pins? I'm looking at guitar pedal layouts and everything is so much easier with 2-pin switches, so I don't understand why I'm seeing 9-pin in most of the pedals I look inside.

>>1907851
If you have a scope, you'll want to probe the signal all the way from the left input to the left output and figure out where it isn't working. Should be relatively easy to follow its path with the schematic. Comparing it to the right channel, to ensure it's getting amplified or attenuated at the right stages. I'm betting it's a poorly seated/crimped connector or a cracked solder joint.
If you don't have a scope, I can only suggest to send an AC signal that your DMM can measure without issue, a 60Hz sine is a good bet.
If that doesn't work or you don't have either, you could solder together a little op-amp buffer amplifier + speaker that you touch to parts of the circuit to listen how they sound.

>>1907865
Induction motor.

>>1908163
9-pin switches are used for switching a stereo signal two directions, as well as a third signal of some kind like power. In one position the power is on and the signal goes through the guitar effect, in the other position the power is off and the signal goes right through to the other end. Might be a mono signal and they're doing extra shit like grounding the input to the amp or detaching the output from the pass-through signal or whatever.

I'd still lean towards a 2-pin switch and do everything else with analog switch ICs, because it's cheaper that way. Recently bought a CD4053 for this purpose. Could even use an extra cheap tactile switch alongside a T-flip-flop of some variety. Does make your PCB bigger and requires SOME quiescent current, which you have to design around.

And what in sam hell are those power resistors for?

>>1908182
Thank you, that's perfect information. Yeah I also want to stick to 2-pin when I start designing my own stuff.

That's a fuzz pedal with the giant power resistors, no idea why they're there though.
https://www.youtube.com/watch?v=6pmaHj2lK9Q

>>1908199
>>1908182
That pedal sounds fantastic.
But reading the description they talk about how it uses "...a highly curated cache of rare and irreplaceable components including paper-in-oil capacitors and NOS half-watt carbon composition resistors from Jamie Stillman’s personal stash..."
so that's probably why the resistors you're referring to are used -- hipster bait.

>>1906848
I've heard a lot of good things about using Vivado from the command line using TCL scripts in non-project mode
You can integrate Vivado into other IDEs this way, so you don't have to deal with Xilinx's IDE as long as your not flashing a bit file.

>>1908224
Yup its all marketing. A lot of analog pedal shit is snakeoil. I want to learn MCUs like ARM Cortex for weird shit

>>1908250
Imagine a DSP guitar pedal. Just plug in a USB cable, program it, and let it do literally anything you want. Pretty sure a common bluepill would be more than sufficient processor-wise, though maybe with external ADC and DAC. And external RAM for delay/echo.

I'm looking at stereo class-d amplifier modules/boards, and almost all of them seem to have differential outputs (and hence have 2 output filters per audio channel). If I wanted to power speakers (or some other load) that has a common ground, that would require commoning two of the 4 outputs together. Anyone know if this can be done?

Just looking at the tpa3118d2 that I saw on ali as one of the only listings with only two inductors on the board. But the datasheet makes no mention of single-ended outputs:
https://www.ti.com/lit/ds/symlink/tpa3118d2.pdf

When getting wire, do you get gauges by amps, watts, or volts?
To be safe and not have things light on fire and all

>>1908396
Better question I should ask, 42 amps, 40 volts.
What gauge wire do I need?

>>1908411
https://grealpha.com/resources/dc-load-wiring-calculator/calc/suitable-wire-gauge/

>>1908287
Well digital is really prevalent starting about 5 years ago. The Spin FV-1 is really popular, but now ARM Cortex seems to be growing in popular. Yeah it is sweet to just write code and send it via USB

>>1908396
For almost all use, voltage rating isn't important. The extremely thin layer of enamel insulation on "magnet wire" is sufficient for building transformers for mains electricity, let alone the 0.5mm thick PVC layer commonly found on hookup wire.
Power doesn't really mean anything to the wire, at least not the power of the device being supplied by the wire in question.

The main limit on wire is a thermal one, whether the flow of current through it is sufficient to cause it to heat up beyond some safety limit. Naturally, wires conduct and radiate heat away from themselves into their surroundings, so at steady-state, there is an equilibrium of electrical power in equalling radiative power out. Because wires are intended to not have a voltage drop significant to the rest of the device across them, we talk about the current through them. A wire has a certain cross-sectional-area in copper, which is proportional to its resistance per unit length. Double the resistance, and for the same current the voltage will double (V = IR). Double the voltage, and the power will double (P = IV). Hence, resistance is inversely proportional to how much power it wastes; it goes up by the square of the radius of the conductor.
But thermal radiation and conduction are not proportional to cross-sectional area, but rather to circumference. The total thermal loss from a given length of wire is proportional to its temperature multiplied by its outer area. This scales by the radius, but also note that it's not the radius of the conductor, but the radius of the insulation.
The ultimate limiting factor is the temperature you want the wire to get up to, or more importantly, the insulation. PVC insulation is something I wouldn't mechanically trust much above 100C, and I wouldn't touch it above 60C. Hence, you plug that temperature and current into a somewhat complicated equation, and out drops the radius of wire.

In practice, just use a lookup table. I use the wikipedia page for AWG.

Anyone has experience with chinkshit rotary encoders?

Bastards won't give datasheets, this is what I found, do I need pull ups for this?

I don't have oscilloscope.

>>1908487
>do I need pull ups for this?

yeah, says so right there, clear as day, ''bring a pull, can be directly output pulse''.

>>1908489
Actually, alongside some info I've read, I just realized it means the opposite. Though using a pull up is fine too.

Thanks anyway.

>>1908487
Only with the bog-standard mechanical 20-detent encoders. No clue about magnetic encoders, I assume they use binary hall-sensors? If that's the case, then the output should already be buffered to some extent, provided there's seperate lines for Vcc and GND.

>>1908456
>Double the voltage, and the power will double (P = IV)
Double the voltage, and the power will quadruple. Current itself is proportional to voltage, so the V term actually appears twice on the right side of that equation. P=V^2/R, to substitute the relation from the other equation. But within a thermal constraint due to resistive heating, then yes, doubling the voltage will double the deliverable power.

>>1908528
>the V term actually appears twice on the right side of that equation
Not through a wire, where as I stated:
>wires are intended to not have a voltage drop significant to the rest of the device across them
If we were talking about a single resistive electrical load then you'd be correct, but I was referring to increasing the resistance of a wire where that resistance is insignificantly small compared to that of the load it's in series with. Current in such a situation is effectively a constant, decreasing minimally with a doubling wire resistance.

stupid question (just got into electronics for 2 months)
why GaN give a benefit of smaller transformer?

>>1908582
i dont know who this gan character is but the smaller transformer needs less food and thus never asks for a pay rate increase because his household budget is likewise smaller. he needs no big screen tv. he is content with his crappy low flow toilet because it can still flush his small poops

>>1908582
>GaN

the band Guns and Noses?

>>1908489
>>1908503
The Chinese says 特色功能,自带上拉,可真接输出方波. I don't know what the hell that means.

>>1907945
>i dont see how that's doable.
https://youtu.be/a28lVMsWFEY suggests it works if you don't want to use a differential probe, which I gather is exactly what we're talking about here.

>>1908597
>>1908648
This is how you treat a new guy??

>>1908649
>特色功能,自带上拉,可真接输出方波.

you can type chinese but cant read it? and dont know about google translate? which gives ''Special function, with pull-up, can be connected to output square wave.''

>>1908649
>特色功能,自带上拉,可真接输出方波.

you can type chinese but cant read it? and dont know about google translate? which gives ''Special function, with pull-up, can be connected to output square wave.''


>>1908660
>This is how you treat a new guy??

yes, coz idiots like this just waste your time. you have to pester them before they'll give you enough information to make sense of their gibberish.

>>1908664
>you can type chinese but cant read it? and dont know about google translate? which gives ''Special function, with pull-up, can be connected to output square wave.''
Yeah I put it in Google Translate. I didn't trust the results. I last took a Chinese course over 15 years ago, and haven't been able to speak it in over 10 years. At my peak I probably would've used a dictionary.

>>1908456
a rule of thumb for magnetic wire for transformers is about 500-700 mils per amp. except it is not a very handy unit to deal with. so roughly speaking if you are ok with high temperatures, you can even go as low as 300 mils per A. or as high as 1000. but thats for transformers. if it is about connectors, then 300 mpA is probably perfectly fine.

>>1908287
every multieffect pedal from since 2000 has a DSP ...and most modern ones are programmable

>>1908660
what do you mean ? ...the anwser was perfectly written as a response to the question asked

stupid questions that are stupidly worded give stupid anwsers that are stupidly worded...basically we anwsered your question and gave you good advice so i would say we treat newcomers very well

>>1908938
Well, shows how much I know.

I'm having some hard times with soldering. I watch videos like this one, for example (see link https://youtu.be/MOywVfRrYTg?t=252)), and it seems like it's trivial to get solder to flow and stick to things. I feel like, whenever I do soldering, the solder never sticks to my iron, it never flows when I touch it to things, and I never get easy or quality results. I can put the iron on top of solder point for 5-10 seconds before it starts to flow. I feel like I must be doing something very wrong, but I don't know where to begin.

Is there like, a good website, or book, or video or something, to demonstrate all the details of soldering 101, even down to the most basic elements of how to pick and care for equipment and how to do every little thing? I'd find somebody locally to help me, but I don't know that anything is going on presently. Guess I just feel like I'm really failing hard, and I'm hoping some resource can rescue me from my horribly lacking skills.

>AP62200 buck converter datasheet
>.08V - 7V output voltage
>need to get a weird, specific output voltage
>check the formula to determine R1 for a given R2 and Vout
>formula is impossible to use

Is Diodes Incorporated retarded or did I miss something here?

>>1908991
That sounds like heat is flowing to the solder joint too slowly. This can be for two or three main reasons. The first is that your tip isn’t hot enough, pretty self explanatory. If it’s showing what you think to be a sufficiently hot temperature, then it’s possible that the thermal sensor is inaccurate. The second is a short-term one, that the thermal conductivity of your tip is too low. This could either be from poor conductivity from element to tip, indicating a poor internal construction, or from tip to PCB, indicating improper tip geometry for the task. Personally I find conical tips awful for most work. Higher surface area from tip to solder and solder to board means higher thermal conductivity, which is why it’s important to keep a small amount of unoxidised solder on the tip itself. A similar problem is insufficient heat capacity of the top and element, where the tip cools down too quickly after making contact with the workpiece. The third issue is a long-term one, power. If you’ve only got a 20W iron, then after depleting the tip’s heat capacity you’ll only be able to push a certain amount of heat into the workpiece, which naturally dissipates it into its surroundings, limiting the temperature you can get it to.

>>1908991
what iron do you use?

>>1908996
It’s correct, you proved it yourself. The Vfb = Vout(R2/(R1+R2)) identity from the voltage divider is the same as the formula you were given, just in a different form. By plugging one into the other you made a tautology.

>>1909017
T12 STM32 OLED chink station

>>1908996
>>1909020
In addition, with just that one equation, you’re missing a piece of information; you have an extra degree of freedom. I suspect Vfb will self-regulate to a constant value due to being in a negative feedback loop, maybe 2.5V or some other common reference voltage. So you should look in the data sheet for some other constraint on Vfb that isn’t that equation.

>>1909017
>>1909020
Wait I thought that was in response to me (>>1909001), I'm not the anon with the problem.

I'm currently choosing my courses for the upcoming semester and I'm trying to choose between Control Systems or Circuit Theory for my project. I'm definitely doing something practical (not just theory), so I thought about picking a system to automate for CS. If I choose CT I would probably be able to make a audio amp as a project cause we covered filters, amps and other circuits from that subject.

Which choice is more practical, and do you have some suggestions on maybe other projects I could pick?

>>1909128
Correction - I probably can't do amps from CT because we only did filters and analysis, not synthesis.

>>1908996
you are retarded i think ...all you had to do is plug in a set R2 , your output voltage , and the Vfeb is the feedback voltage of the internal votlage refference which yo ualso plug in

>>1909128
pick the one that is easier to pass so you have more time for the harder classes you cant choose, college is a min-maxing situation
and do your projects on your own time and ask a prof for help if you need it..no professor i ever spoke to refused helping me with my own projects because most of the time they like to help students that want to learn

>>1909001
I'm not familiar, how would I really tell if the iron was getting hot enough? Maybe I should get an infrared thermometer? There's a chance I have issues but I'm not sure. Mine just has a dial that can set heat but doesn't read it out.

>>1909017
Here is the one. I bought it a long time ago now. I have noticed that it seems to only work on occasion, even when I use new tips. I don't know what the cause is of that, whether bad iron, poor tip maintenance, or something else.

>>1909206
And here's a picture of what mine looks like. I've seen this happen on an older iron I had too. Something about it seems seriously wrong, but like I said before, I'm totally clueless in this world. My one success story is fixing a vacuum cleaner switch that wouldn't turn on, though it was a shitty solder job the project still worked. Still, I feel like I must be doing something really wrong, cuz my soldering experience just feels inferior to everyone else's.

>>1909209
Hard to tell, except that it is a rather thin pointy tip and at 40W it might not get hot enough. Try a wider/bigger tip.

>>1909209
For example, I have an older Weller WES51 station which is something like 50-60W and I didn't have much success with those pointy tips. Even with moderately wide tips I sometimes have to crank the temp almost all the way up (close to 800F/400C). And also make sure you use good flux. Try some liquid flux. There is this brown stuff I buy on ebay, it is very sticky and requires cleaning but it is very potent. You may need to buy a set of tips like these and try different ones.

>>1909211
>>1909214
My concern was with doing small electronics, that I would mess things up with a bigger tip. I can go with that though, I think there are some around here from a pack I bought long ago. I'll try turning up the heat more too.

Another thing that comes to mind, I noticed that often my iron either didn't pull solder at all, or it pulls it instantly but doesn't get coated, the solder just turns into a hot fluid ball and falls right off the tip. I tried putting flux on (mine is like a thick light gold colored gel that smokes in response to the hot solder tip) but it didn't help matters much. Is there a reason the solder doesn't stick to the iron tip when I try to tin it?

Also, should you apply flux only to the place you want solder to flow? Or should you apply it to the tip directly? I'm not sure which is the proper option.

>>1909255
>doesn't get coated
Thats bad, thats what you should fix first. Try cleaning and re-tinning it. You should absolutely make sure the tip is tinned. It should tin easily.

>apply flux only to the place you want solder to flow
Yes. Only apply it to the tip when re-tinning it. Which you shouldn't need to do frequently.

i need an LCD/OLED of a very particular size range (on the order of 1x4"). digikey has what i want but all options are comically overpriced. aliexpress is impossible to search. are there any other sites i should take a look at?

>>1909287
digikey/mouser/rs/farnell/tme/ but the prices will most likely be about the same

otherwise try banggood or other chink sites just learn how to search properly...but if its not a mainstream clone like for arduinos or rasbpis then you wont have much luck in the price department anyways

also protip when searching for shit on chink sites try typing the names into google translate to chinese then back to english again ....they use translate to do all the english names and they are sometimes called differently there

>>1909209
thats a too small tip for anything bigger than SMD soldering on non ground plane pins ....you have to take into consideration that the smaller the tip it will lose temperature much faster , and if its a slow heater or not very aggresive heater regulator it might take a long time to get to the desired temperature because the tip sinks heat faster than it can generate it...thats why most high end soledring irons have a overboost heater for temperature regulation , and these old wellers are most likely termomagnetic heaters

also wellers with removable tips can have big problems with corrosion buildup on the part where it connects to the heater element/connection so it doesnt get enough current (less power) and it does not have an efficient thermal connection (thats why most high end soldering irons have integrated cartriges instead of tips)

basically always use the biggest tip you can keep the tip and board clean , use flux and good solder (rolls can get up to 100$ but they last for tens of years so are really worth it)
every time before soldering clean the tip on a brass wirebrush or watered sponge and try to tin the tip with solder , if it does not want to tin nicely then something is wrong (bad tip, low temp or solder with higher temp ratings)

Capacitor across the momentary ON contacts of a no-voltage-release safety switch failed short-circuit, causing a table saw in an empty workshop to start up and run unattended for at least a week.
Already fixed by deleto/yeeto the cap, which as far as I can tell was only in there because enviropussies are afraid of contact arc noise.
Just about fucking done with this fucking clown world where a SAFETY STOP SWITCH fails into the *ON* position and still gets UL approval. Fucking rooms full of machines here that can turn a man into a picasso painting, now we find out this shit could start up at any time and all the warning you'd get is the muffled popcorn-pop of a polystyrene capacitor spooging it's molten guts out into an IP67 switch box.
I mad.

>>1908067
>markers enabled to highlight data
doesnt that put on markers (circles) on the activated channel?

>>1907832
if it stink up even before turning on the switch then you have a short on the power input side and should check that first (connectors switches , loose wires) second thing if one of those two resistors is showing 1.2MOhms then its broken since it says on it 5000Ohms

trace alll the cables where they go and draw a schematic based on the wiring , it will be much easier o troubleshoot for you or anypone that wants to look at it to help you

>>1907851
pretty much everything looks good ....you should try cheking again if all the caps and diodes are in their right polarisation .....also check those jumpers if they are on proper headers

if that turns up nothing then rote back the voltage paths to their regulators and see if they are not broken for some reason (overheating during soldering can destroy IC-s or just pulling out legs while bending)

>>1909331
most industrial power tools i used have two switches in series for the reason if one fails ...maybe try wiring up a master ON switch...or add a power on rotary switch in the main electrical box with your tools and turn it off every time the workshop is left unattended like most professional workshops in urop

>>1909206
if all else fails you can always ge this

So my scope just came with the regular hook tips (removable of course). I'd like to probe shit smaller than these hooks can grab. What the hell am I looking for? Everything I find online looks like it's for hooking up to a multimeter or has wires going to a BNC or banana plugs on the end, not for slipping over the end of a probe.

>>1909359
pic rel

or just take off the hooks and probe witth the tip inside by hand

uh lads
my floppy drive magically started writing successfully this afternoon
here's probing across the two read/write head terminals during a write operation... at least I'm pretty sure.

I'm actually kind of upset that I didn't manage to probe it while it wasn't working. I don't know what changed or why it's suddenly working now. All I did was probe it a few times in a couple of places. Could that possibly have fixed anything?

>>1909397

maybe if some of the solder joints were cracked

>>1909209

your station is perfectly adequate at 40W (i mostly use one at 25W), and your tip is first rate, so that leaves your solder as the likely problem. it absolutely must be Sn63Pb36 rosin core solder from a name brand. search this thread for examples using keyword ''Kester''.

>>1909359
>What the hell am I looking for?

you can get some of these and wire them to alligator clips using thin flexible wire. this is much more workable than attaching it to the end of the probe coz the probe cable tends to be inflexible and pulls the clips away from the work. (and dont worry about having a couple of inches of un-shielded cable unless you're approaching the gigahertz range)

>>1909209

your station is perfectly adequate at 40W (i mostly use one at 25W), and your tip is first rate, so that leaves your solder as the likely problem. it absolutely must be Sn63Pb36 rosin core solder from a name brand. search the previous thread for examples using keyword ''Kester''.

>I have noticed that it seems to only work on occasion.

that could mean an intermittent connection. if you have a watt-meter like a KillaWatt, you can see if it stops pulling current at times.

>>1909420
That’s all I can think of either. And it works without being probed so it’s not like it’s probe loading or something equally strange.

Reckon I should melt and resolder the board? Or just leave well enough alone?

>>1909509
it could be loading it depending on the probes and input impedance of your scope ...but i doubt it

you can try refreshing the joints if you can by taking as much off as you can and resoldering the pins with frsh solder... but try not to burn any IC-s

>>1905618
>>I have junk, what do?
>Shitcan it
Hello, I'm a /g/eentoman with some old electronics. I'm new to this. Is it worth scraping for parts like microcontrollers, displays and stuff? Also, is a cheap multimeter enough for simple projects, or should I spend a little more for anything decent?

Fellas.... why is this shit so expensive?

>>1909627
I find it really fun to salvage parts
my multimeter is a POS it doesnt even measure capacitance, I'm getting another one. I'd say its one of the essentials that is best to just get a decent one up front

>>1909627
>microcontrollers
no. you won't find good hobbyist microcontrollers in most shit.
>displays
depends. if they're LCDs then trash them. you'll never figure out which controller chip is in them, and half the time you'd need an FFC adapter to break them out. if they're VFDs or other retro tech and don't have a bunch of gay hardwired characters like "DVD" then they could be fun to repurpose.

the electronics hobby lends itself to hoarding. you really have to make an effort to toss shit you'll never use, or it'll pile up forever.

>>1909652
how about cathode ray tubes?

here's a 10,000 lumen COB being strobed at 10hz. it was physically painful.

the COB is going to be bolted to an aluminum slug and attached to my shotgun along with the custom controller pictured and a 4S lithium pack. i posted the controller last yearish but didn't have an adequate supply to drive it fully until now.

>>1909646
>>1909652
Yeah I think I'll try salvaging a few of them, just to see how I handle the task and what I can get. I read online that older electronics are better to salvage, since nowadays everything is surface mounted and small.
>or it'll pile up forever
That's what is happening. I think I'll throw all away. I also saw how cheap electronics are online
>my multimeter is a POS it doesnt even measure capacitance, I'm getting another one. I'd say its one of the essentials that is best to just get a decent one up front
I'll get something in the middle term then.
Ty for the tips

>>1909670
>strobe on shotgun
as long as you don't write a bunch of /pol/ shit on your guns in white marker go ahead

>>1909682
no officer i would never do that haha

>>1909667
If this is a serious comment, then maybe. They’re a real pain to drive, worse than VFDs, but there are a lot of ICs and entire boards dedicated to that purpose, so in reality it’s not that tough to draw a raster image if you’ve got a proper power supply. Vector images on the other hand, now that’s a fun project.

But that’s just the electrical side, you’ll need some way of mounting the thing, meaning relatively proficient case making of some sort is a requirement. The high-voltage is also somewhat prohibitive towards hobbyist testing equipment.

I'm about to spend $300 on chinkshit from ebay
any objections? missing anything?

>>1909767
why so much on shipping? is that just a leaf fee?

>>1909768
holy shit that was a quick response
No the seller said they will combine shipping

>>1909736
I'm honestly about half serious. My bro does maintenance for this business that runs group homes for the retarded and he said he pulls CRT TVs now and again and they usually just get junked.

>>1909768
5 dorra shipping is not much on 300 dollars of goods

>>1909767
don't buy an LED cube with a premade base. That's all the pain of soldering 512 LEDs to each other and none of the fun of designing and programming the interface

bold of you to assume I can program the interface

>>1909783
after looking into it, maybe you are right
I would rather use my own controller to code or at least upload animations to.
Was not the highlight of the cart, just something cool I saw from the seller
thx

>>1909770
>CRT TVs
Those CRTs have the tri-color electron guns and phosphor inside them, making them semi-useless for vector displays. Might still be interesting though.
If you want to use them for raster displays, go ahead I guess but LCDs are cheap and ubiquitous these days so I wouldn't really bother, unless you specifically want to make/alter images with an analog circuit.

Not sure if he's still around, but there was one anon who made a thread here last year or so about his synth setup, where he made visual synthesisers. Basically trying to make cool visual patterns using analog electronics to feed into CRTs. Guy had like a dozen CRTs, and like 10 times even more racks and shit to mount his complicated circuitry, really cool but a full-on rabbit-hole.

>>1909670
make an incapacitator

Hello anons. Would anyone be willing to assist me with selecting some lipo batteries on Digikey? I intend to use them with my Nice!Nano MCUs in a split keyboard project. These units run at 3.3V. Here's the site page about them:
https://docs.nicekeyboards.com/#/nice!nano/
I also have 32x per-key SMD LEDs and 1x OLED displays on each side, so power draw is relatively high. As such, I want to utilize a reasonably high-capacity battery with minimal height impact. I'm having some difficulty ascertaining which ones have sufficient protections when reading spec sheets, so that is my primary concern. I've trimmed it down to five options here (the link is a bit large due to adding search filters):
https://www.digikey.com/products/en/battery-products/batteries-rechargeable-secondary/91?k=&pkeyword=&sv=0&pv33=302251&pv33=48429&pv33=47336&pv33=50934&pv33=51897&pv33=60350&pv33=62747&pv33=123430&pv33=136026&pv33=201772&pv33=210701&pv33=266441&pv46=33533&pv46=33872&pv46=42662&pv46=47660&pv46=469280&pv46=50288&pv46=52343&pv46=53790&pv46=54019&pv46=54516&pv46=56642&pv46=57331&pv46=58332&pv46=59159&pv46=123637&pv46=429844&pv46=124802&pv46=129309&pv46=130583&pv46=173518&sf=0&FV=412%7C365310%2Cmu3.7V%7C2079%2C-8%7C91&quantity=&ColumnSort=46&page=1&pageSize=100

This one in particular seems nice, at only 4.0mm in height (though it is mislabeled on the site, and actually rated for only 1,050mAh):
https://www.digikey.com/product-detail/en/globtek-inc/BL1200P4054481S1PCAC/1939-BL1200P4054481S1PCAC-ND/11201333
Not sure if the yellow wire (sense?) will complicate anything either. Any input or advice would be appreciated.

>>1909767
>a literal dozen power supply modules/chips
>nothing interesting to power with them
Do you have a plan here or just winging it?
Also,
>SMD voltage regulator

>>1909627
>Is it worth scraping for parts like microcontrollers
absolutely not, most mcus and processors cost under a dollar, and those that are expensive are usually always overkill for anything especially if you are noob and just beginning
>displays
the same as before , the things that are easy to controll cost very little and are not worth taking from scrap because they might not work properly anyways, and the more expensive stuff require special controllers and firmware

scrapping parts is useful when you are looking for specific parts that you know are not available anywhere elselike retro ROMS, and different types of IC-s that dont have a modern equivalent and you have an old design you want to make or fix...or if you are able to find very high power components that are massive and cost a lot just by their own self, or some very rare capacitors/resistors with specific values...most of the time the most valuable parts of scrap electronics is the box it comes in , potentiometer knobs attached to them, some types of connectors and switches and most of the time transformers and inductors the rest is almost always found either for cheap or is not usable otherwise and not worth hoarding

>>1909627
> is a cheap multimeter enough for simple projects
if by cheap you mean 50& to 100$ then yes , just make sure its a true RMS meter everything else is a meme

get a cheap second hand osciloscope and a cheap lab voltage/current source with your multimeter and you are set for 90% of things as a noob

>>1909980
I am making a diy bench power supply.
the ac/dc converter will power it. I have the wires and cords already.
I am also making a solder extractor, I have the fans coming in the mail and the carbon filters.
>smd regulator
all the seller offered. I think I have the th ordered already from another seller but I want to have it all come in the mail at once and it's only a dollar

>>1909646
the filter or the solder vapor sucker ? ...i bouhgt a chink filer for a dollar and cut out an L shape in an old PSU case with the fan attached and connected that to a 5v regulator

>>1910019
i just spent 6 dollars on 5 13x13 filters

>>1909767
almost everything is premade , 90% of it is power sources, the rest is geeky gadets to give your 10 year old nephew ...whats the point of all that ?

also those fuses are expensive for 10 pcs ...and those pots are not precision if you get 10 for 7$

desu i would rather you gave me the 300 so i can actually buy components and an extra soldering tip i have been looking at

>>1909785
whats the point of having the led cube if you cant even say you made it

>>1910022
yeah ? so whats the expensive part ...that means you probably got real carbon filters

>>1910026
I'm a beginner cut me some slack
I like the tree
it's 5pc not 10
>>1910028
see >>1909797
>>1910029
i'm very stingy. It just seemed like they are carbon sponges, they should be 50 cents each max

>>1910033
thats desu a good price anyways ...real activated carbon filters with a bit of thickness cost two to three times waht you paid but w/e

>I like the tree
fair enough ...but you asked for objections on a mongolian basket weaving forum

>it's 5pc not 10
proper precision pots with multiple turns are usually over 10$ a piece ....but thats if you really need real precision , for what you have in mind a regular carbon pot is good enough anyways

>>1910036
what words are changed with desu so i can stop using them

and why the fuck is that done automatically weebs should go kill themselves for being a burden

>>1910036
maybe I'll get some fuses at a hardware store instead
but thanks for the tips, I have regular pots already
I just need chink precision

>>1910037
t*b*h

If I where to replace this with a double sweep 10K B potentiometer I'd have to solder both 1 & 3 to the center pins and then all the black ones to both pin 1's and lastly 5 & 7 to both pin 3 individually?

Does IEC62133 certification for lipo batteries cover the following?:
>Anti-shortcircuiting
>Anti-overcharging
>Anti-overcurrenting
>Anti-overloading

>>1910037
baka desu senpai
baka desu senpai
baka desu senpai

>>1910056
>>/lgbt/

>>1910042

what a silly silly question. you're essentially asking if you should wire the thing according to the diagram. obvious answer is: yes, wire it like the diagram tells you to.

also, the term ''double sweep'' isnt standard terminology. if it means ''ganged'' then you're good.

This is the dumbest question ever but I have to ask it. I'm trying to get an EE degree because I'm interested in the field and I want to be able to build shit like my dad and brother, I'm taking both Calculus 3 and Physics 2, which course do I prioritize more? I'm also taking Electrical Circuits and Logic Design as well but those require less time than the first two I've listed. My goal is to build a temperature-controlled fan or a small swarm bots before the start of the next summer.

>>1910073
>which course do I prioritize more?
whichever one you can't get test answers for

>>1910075
>whichever one you can't get test answers for
Calculus 3 it is, isn't there a large focus on Physics in EE in general? Though Physics is based on math so I can understand what you mean.

Any good senior design project recommendations? Apparently a voice-changer is not enough to be considered one so I'm looking for ideas.

is it safe to store lithium batteries in a sealed ammo can or do you need vent holes or some shit?

I don't know if it's a meme or not just trying to be safe

I wanted to build a mobile phone jammer based on the github rolling site but it's illegal, wtf??

brainlet here, trying to make an analogue type soldering station for the hakko t12 iron, would this circuit work?

>>1910073
Calculus 3 is probably useless for most of the stuff you'd ever need in practical EE. It is supposed to prepare you for more advanced classes such as EM fields/maxwell equations. But those go way beyond both Calc3 and Physics-2. So it all depends if you will be required to take advanced physics or EE specific classes such as theory of the EM field. And you won't need any of that to be able to build a temperature controlled fan.

>>1910143
So I don't need to know calculus 3 and physics 2 like the back of my hand?

>>1910157

one in a million will need it. for everyone else, this is just a hurdle to prove to potential employers that you have the intelligence or the tenacity to succeed at difficult tasks. (or that you can cheat, blackmail, or bend over to get what you want)

>>1910163
>one in a million will need it.
Like one in a million electrical engineers? If so, thank god. I'll still try my best but thank you regardless.

>>1910157
you don't learn that stuff to know it long term. you learn it to pass the test and forget it. and you also learn how to learn. nothing new there. so from that perspective you absolute need to do your best. but if you are asking if you will ever need that knowledge at work that then answer is very unlikely. and unused knowledge goes to waste. you will forget most of this stuff quickly. but if you put some effort into it at school, you should be able to retain at least the gist of it, enough to be able to refresh your memory and at least figure out what to look for in a very unlikely case you will ever need this stuff.

>>1910168

yeah, those who end up at theoretical research centers, or some in chip manufacturers.

>>1910181
>>1910181
>>1910181
New Thread!

So I got some RGB 4 pin leds one is Common Anode and the Other is Common Cathode. Besides polarity, is their advantages to using one over the other?

>>1910045
https://www.iecee.org/dyn/www/f?p=106:49:0::::FSP_STD_ID:6515

>>1910076
at my college physics was extremely basic , and all the "physics" you need will be as part of the introduction to electrnics

>>1910105
i soldered with a clothes iron in a pinch ..so yes most likely you will be able to solder with it if it works

>>1910157
no

Anyone ever make an anlaogue slide rule?

>inb4 just use a calculator bro